None:\\n pass\",\n \"def main(self):\\n self.log = logging.getLogger()\\n\\n opts = self.options\\n args = self.args\\n if opts.debug:\\n self.log.setLevel(logging.DEBUG)\\n else:\\n self.log.setLevel(logging.WARNING)\\n if len(args) < 2 :\\n self.log.error(\\\"Not enough arguments\\\")\\n sys.exit(1)\\n else :\\n project = args[0].upper()\\n version = args[1]\\n\\n # Connect to the ConfDB to update the platform\\n self.mConfDB = SoftConfDB()\\n\\n buildtool = \\\"cmake\\\"\\n if self.options.usecmt:\\n if self.options.usecmake:\\n raise Exception(\\\"Please specify one build system only\\\")\\n buildtool = \\\"cmt\\\"\\n \\n if self.options.remove:\\n self.mConfDB.unsetBuildTool(project, version)\\n else:\\n self.mConfDB.setBuildTool(project, version, buildtool)\",\n \"def main():\\n try:\\n return run_sysca(sys.argv[1:])\\n except InvalidCertificate as ex:\\n die(str(ex))\",\n \"def run():\\n print(\\\"clewsy CLEWs Model Building Script.\\\")\\n print(\\\"When using clewsy please reference:\\\")\\n print(\\\"T. Niet and A. Shivakumar (2020): clewsy: Script for building CLEWs models.\\\")\\n main(sys.argv[1:])\",\n \"def generate(env) -> None:\\n c_file, cxx_file = SCons.Tool.createCFileBuilders(env)\\n\\n # C\\n c_file.add_action(\\\".l\\\", LexAction)\\n c_file.add_emitter(\\\".l\\\", lexEmitter)\\n\\n c_file.add_action(\\\".lex\\\", LexAction)\\n c_file.add_emitter(\\\".lex\\\", lexEmitter)\\n\\n # Objective-C\\n cxx_file.add_action(\\\".lm\\\", LexAction)\\n cxx_file.add_emitter(\\\".lm\\\", lexEmitter)\\n\\n # C++\\n cxx_file.add_action(\\\".ll\\\", LexAction)\\n cxx_file.add_emitter(\\\".ll\\\", lexEmitter)\\n\\n if sys.platform == 'win32':\\n # ignore the return, all we need is for the path to be added\\n _ = get_lex_path(env, append_paths=True)\\n\\n env.SetDefault(\\n LEX=env.Detect(BINS),\\n LEXFLAGS=CLVar(\\\"\\\"),\\n LEX_HEADER_FILE=\\\"\\\",\\n LEX_TABLES_FILE=\\\"\\\",\\n )\\n\\n if sys.platform == 'win32':\\n env.SetDefault(LEXUNISTD=CLVar(\\\"\\\"))\\n env[\\\"LEXCOM\\\"] = \\\"$LEX $LEXUNISTD $LEXFLAGS $_LEX_HEADER $_LEX_TABLES -t $SOURCES > $TARGET\\\"\\n else:\\n env[\\\"LEXCOM\\\"] = \\\"$LEX $LEXFLAGS $_LEX_HEADER $_LEX_TABLES -t $SOURCES > $TARGET\\\"\\n\\n env['_LEX_HEADER'] = '${LEX_HEADER_FILE and \\\"--header-file=\\\" + str(LEX_HEADER_FILE)}'\\n env['_LEX_TABLES'] = '${LEX_TABLES_FILE and \\\"--tables-file=\\\" + str(LEX_TABLES_FILE)}'\",\n \"def main():\\n \\n usage = \\\"\\\"\\\"usage: %prog [target]\\nTargets :\\n\\tall\\t\\tCompile all projects in csunplugged directory\\n\\tindex\\t\\tCreate the index page\\n\\tdirectory\\tCompile single project in csunplugged directory\\n \\\"\\\"\\\"\\n parser = optparse.OptionParser(usage)\\n \\n (options, args) = parser.parse_args()\\n if not (len(args) == 1) :\\n parser.error('incorrect number of arguments')\\n \\n if args[0]=='all' :\\n build_all()\\n \\n elif args[0]=='index' :\\n make_index_page()\\n \\n else :\\n jsFile = args[0] + '.js'\\n jsFilePath = os.path.join(csunpluggedPath, args[0], jsFile)\\n if not os.path.exists(jsFilePath) :\\n print 'Can not found ' + jsFilePath+'\\\\n' \\n else :\\n \\tbuild(args[0])\",\n \"def main():\\n CLI_APP.run()\",\n \"def build(self):\\n env = ConfigureEnvironment(self.deps_cpp_info, self.settings)\\n\\n set_path_command = \\\"\\\"\\n # Download nasm as build tool. This should go to source()\\n if self.options.SSE == True:\\n if self.settings.os == \\\"Linux\\\":\\n # TODO: We should build nasm from source then.\\n self.options.SSE = False # Or is removing here better? I'm not familiar with python..\\n else:\\n nasm_version = \\\"2.12.02\\\"\\n nasm_os_url_id = \\\"\\\" #nasm url identifier\\n if self.settings.os == \\\"Windows\\\":\\n if self.settings.arch == \\\"x86\\\":\\n nasm_os_url_id = \\\"win32\\\"\\n else:\\n nasm_os_url_id = \\\"win64\\\" \\n elif self.settings.os == \\\"Macos\\\":\\n nasm_os_url_id = \\\"macosx\\\"\\n nasm_folder_name = \\\"nasm-%s-%s\\\" % (nasm_version, nasm_os_url_id)\\n nasm_zip_name = \\\"%s.zip\\\" % nasm_folder_name\\n download(\\\"http://www.nasm.us/pub/nasm/releasebuilds/%s/%s/%s\\\" % (nasm_version, nasm_os_url_id, nasm_zip_name), nasm_zip_name)\\n self.output.warn(\\\"Downloading nasm: http://www.nasm.us/pub/nasm/releasebuilds/%s/%s/%s\\\" % (nasm_version, nasm_os_url_id, nasm_zip_name))\\n unzip(nasm_zip_name)\\n os.unlink(nasm_zip_name)\\n nasm_path = os.path.join(os.getcwd(), nasm_folder_name)\\n\\n #env.environ[\\\"PATH\\\"] += os.pathsep + nasm_path #its probably as easy as this, but i cant append to the path self.run operates in.\\n if self.settings.os == \\\"Windows\\\":\\n set_path_command = \\\"set \\\\\\\"PATH=%s\\\\\\\" &&\\\" % os.environ[\\\"PATH\\\"]\\n else:\\n set_path_command = \\\"PATH=\\\\\\\"%s\\\\\\\" &&\\\" % os.environ[\\\"PATH\\\"]\\n\\n if self.settings.os == \\\"Linux\\\" or self.settings.os == \\\"Macos\\\":\\n if self.options.fPIC:\\n env_line = env.command_line.replace('CFLAGS=\\\"', 'CFLAGS=\\\"-fPIC ')\\n else:\\n env_line = env.command_line\\n self.run(\\\"cd %s && autoreconf -fiv\\\" % self.ZIP_FOLDER_NAME)\\n config_options = \\\"\\\"\\n if self.settings.arch == \\\"x86\\\":\\n if self.settings.os == \\\"Linux\\\":\\n config_options = \\\"--host i686-pc-linux-gnu CFLAGS='-O3 -m32' LDFLAGS=-m32\\\"\\n else:\\n config_options = \\\"--host i686-apple-darwin CFLAGS='-O3 -m32' LDFLAGS=-m32\\\"\\n\\n if self.settings.os == \\\"Macos\\\":\\n old_str = '-install_name \\\\$rpath/\\\\$soname'\\n new_str = '-install_name \\\\$soname'\\n replace_in_file(\\\"./%s/configure\\\" % self.ZIP_FOLDER_NAME, old_str, new_str)\\n\\n self.run(\\\"cd %s && %s ./configure %s\\\" % (self.ZIP_FOLDER_NAME, env_line, config_options))\\n self.run(\\\"cd %s && %s make\\\" % (self.ZIP_FOLDER_NAME, env_line))\\n else: # We should (for simplicity) always use cmake shouldnt we?\\n conan_magic_lines = '''project(libjpeg-turbo)\\n cmake_minimum_required(VERSION 3.0)\\n include(../conanbuildinfo.cmake)\\n CONAN_BASIC_SETUP()\\n '''\\n replace_in_file(\\\"%s/CMakeLists.txt\\\" % self.ZIP_FOLDER_NAME, \\\"cmake_minimum_required(VERSION 2.8.8)\\\", conan_magic_lines)\\n replace_in_file(\\\"%s/CMakeLists.txt\\\" % self.ZIP_FOLDER_NAME, \\\"project(libjpeg-turbo C)\\\", \\\"\\\")\\n \\n cmake = CMake(self.settings)\\n builddir = os.path.join(self.ZIP_FOLDER_NAME, \\\"_build\\\")\\n\\n if os.path.exists(builddir):\\n shutil.rmtree(builddir) # We need to remove this folder first for windows\\n os.makedirs(builddir)\\n\\n cmake_options = []\\n if self.options.shared == True:\\n cmake_options += [\\\"-DENABLE_STATIC=0\\\"]\\n else:\\n cmake_options = [\\\"-DENABLE_SHARED=0\\\"]\\n cmake_options += [\\\"-DWITH_SIMD=%s\\\" % \\\"1\\\" if self.options.SSE else \\\"0\\\"]\\n\\n # why this comment: \\\"Don't change runtime, conan will take care of\\\"? conan_basic_setup() runs before this cmake option replaces MT with MD again\\n cmake_options += [\\\"-DWITH_CRT_DLL=%s\\\" % \\\"1\\\" if self.settings.compiler.runtime == \\\"MD\\\" or self.settings.compiler.runtime == \\\"MDd\\\" else \\\"0\\\"]\\n\\n self.run('%s cd %s && cmake .. %s %s' % (set_path_command, builddir, cmake.command_line, \\\" \\\".join(cmake_options)))\\n self.run(\\\"%s cd %s && cmake --build . %s\\\" % (set_path_command, builddir, cmake.build_config))\",\n \"def main() -> None:\\n args = docopt.docopt(__doc__, version=__version__)\\n # Set up logging\\n log_level = args[\\\"--log-level\\\"]\\n try:\\n logging.basicConfig(\\n format=\\\"%(asctime)-15s %(levelname)s %(message)s\\\", level=log_level.upper()\\n )\\n except ValueError:\\n logging.critical(\\n '\\\"%s\\\" is not a valid logging level. Possible values '\\n \\\"are debug, info, warning, and error.\\\",\\n log_level,\\n )\\n sys.exit(1)\\n\\n # see if the user is providing any external hash blob data\\n hashblob = None\\n if args[\\\"--stdin\\\"] is True:\\n logging.debug(\\\"Reading hashes from stdin\\\")\\n hashblob = sys.stdin.read()\\n elif args[\\\"--file\\\"] is not None:\\n logging.debug(\\\"Reading hashes from %s\\\", args[\\\"--file\\\"])\\n with open(args[\\\"--file\\\"]) as f:\\n hashblob = f.read()\\n\\n exit_code = ioc_scanner.ioc_search(hashblob, args[\\\"--target\\\"])\\n\\n # Stop logging and clean up\\n logging.shutdown()\\n\\n if exit_code:\\n sys.exit(exit_code)\",\n \"def main():\\n\\n args = parse_arguments()\\n show_parameters(args)\\n\\n bundleDefinitions = get_bundle_definitions(args.bfile)\\n show_bundle_definitions(bundleDefinitions)\\n\\n check_definition_integrity(bundleDefinitions)\\n\\n bundlesForConfig = determine_bundles_for_config(args.config, bundleDefinitions)\\n show_bundles_for_config(args.config, bundlesForConfig)\\n\\n output_result(bundlesForConfig, args.ofile)\\n\\n sys.exit(0)\",\n \"def main():\\n sdoc_application = SDocApplication()\\n sdoc_application.run()\",\n \"def main(argv: Sequence[Text]) -> None:\\n\\n\\n print(\\\"TODO\\\")\",\n \"def DistEntry():\\n flags.StartMain(main)\",\n \"def main(self):\\n\\n try:\\n if os.getuid():\\n print(\\\"**** Error: This script (%s) must be run as root!\\\" % sys.argv[0])\\n raise RuntimeError\\n\\n exit_status = self.EXIT_STATUS_SUCCESS\\n self._parse_options()\\n self._init_logger()\\n logger = self._logger\\n\\n # TODO: Switch to spts' shared onject API!\\n if self._pexpect_available:\\n self.tool_prompt='spt> \\\\? .*\\\\n'\\n #self.tool_prompt='spt> ? .*\\\\r\\\\n'\\n self.tool_start='enable=pipes'\\n self.tool_prompt_index = TOOL_PROMPT_INDEX\\n self.tool_unique_index = TOOL_UNIQUE_INDEX\\n self.tool_status_index = TOOL_STATUS_INDEX\\n self.start_tool(tool=self.tool, start_cmd=self.tool_start)\\n\\n self.get_devices()\\n self.get_drive_information()\\n self.get_enclosure_information()\\n self.get_drive_enclosure_information()\\n\\n if self._number_drives:\\n self.show_device_information()\\n else:\\n self._logger.warning(\\\"No SCSI disk drives found or none match your criteria!\\\")\\n\\n except KeyboardInterrupt:\\n pass\\n\\n except IOError as e:\\n if e.errno == errno.EPIPE:\\n pass\\n\\n except RuntimeError:\\n exit_status = self.EXIT_STATUS_RUNTIME_ERROR\\n if self._debug:\\n logger.debug(\\\"Exiting with status code {0}, RuntimeError\\\".format(exit_status))\\n\\n except TimeoutExpiredError:\\n exit_status = self.EXIT_STATUS_TIMEOUT_EXPIRED\\n if self._debug:\\n logger.debug(\\\"Exiting with status code {0}, TimeoutExpiredError\\\".format(exit_status))\\n\\n if self.tc:\\n self.tc.terminate()\\n sys.exit(exit_status)\",\n \"def main(args):\\n utils.check_setuptools_version()\\n opts = parse_args(args)\\n make_sanity_checks(opts)\\n opts = get_default_opts(opts['project'], **opts)\\n create_project(opts)\\n if opts['update'] and not opts['force']:\\n note = \\\"Update accomplished!\\\\n\\\" \\\\\\n \\\"Please check if your setup.cfg still complies with:\\\\n\\\" \\\\\\n \\\"http://pyscaffold.readthedocs.org/en/v{}/configuration.html\\\"\\n print(note.format(pyscaffold.__version__))\",\n \"def main(self):\\n self.log = logging.getLogger()\\n opts = self.options\\n args = self.args\\n\\n if len(args) < 2 :\\n self.log.error(\\\"Not enough arguments\\\")\\n sys.exit(1)\\n else :\\n project = args[0]\\n version = args[1]\\n\\n # Creating the object to import dependencies\\n self.mAppImporter = AppImporter(not opts.norelease)\\n\\n self.log.warning(\\\"Checking SoftConfDB for %s %s\\\" % (project, version))\\n self.mAppImporter.processProjectVersion(project, version)\",\n \"def cli(self, env):\\n raise NotImplementedError\",\n \"def main():\\n pass\",\n \"def main() -> None:\\n init(args=sys.argv[1:])\",\n \"def main(self):\\r\\n pass\",\n \"def main():\\n setup(**setup_params)\",\n \"def setup():\\n\\n generators = {\\\"man\\\": gen_manpage, \\\"cpl\\\": gen_completions}\\n\\n prsr = argparse.ArgumentParser(\\n description=\\\"xNVMe CLI Bash-completions and man page generator\\\",\\n formatter_class=argparse.ArgumentDefaultsHelpFormatter,\\n )\\n prsr.add_argument(\\n \\\"generator\\\",\\n help=\\\"Generator to run\\\",\\n default=sorted(generators.keys())[0],\\n choices=sorted(generators.keys()),\\n )\\n prsr.add_argument(\\n \\\"--tools\\\",\\n nargs=\\\"*\\\",\\n help=\\\"Name of tools to generate bash-completions for\\\",\\n )\\n prsr.add_argument(\\n \\\"--output\\\",\\n help=\\\"Path to directory in which to emit completion scripts\\\",\\n default=os.sep.join([\\\".\\\"]),\\n )\\n prsr.add_argument(\\n \\\"--log-level\\\",\\n help=\\\"log-devel\\\",\\n default=\\\"INFO\\\",\\n choices=[\\\"DEBUG\\\", \\\"INFO\\\", \\\"WARNING\\\", \\\"ERROR\\\", \\\"CRITICAL\\\"],\\n )\\n\\n args = prsr.parse_args()\\n args.output = expand_path(args.output)\\n args.gen = generators[args.generator]\\n\\n if not args.tools:\\n args.tools = find_binaries()\\n\\n logging.basicConfig(\\n format=\\\"%(asctime)s %(message)s\\\",\\n level=getattr(logging, args.log_level.upper(), None),\\n )\\n\\n return args\",\n \"def main() -> None:\",\n \"def main() -> None:\",\n \"def main() -> None:\",\n \"def main() -> None:\",\n \"def main():\\n return\",\n \"def main():\\n if Args['clean']:\\n print(f\\\"Clear the temporary output directory {Args['tmpdir']}\\\",\\n file=sys.stderr)\\n subprocess.run(f\\\"rm -r {Args['tmpdir']}\\\", shell=True)\\n sys.exit(0)\\n\\n # Configure snakemake execution\\n print(\\\"Configure Snakemake for execution.\\\")\\n os.makedirs(Args['tmpdir'], exist_ok=True)\\n config.argparse_to_json(Args)\\n if Args['snakemakedir'] is None:\\n Args['snakemakedir'] = os.path.dirname(os.path.realpath(__file__)) + \\\"/snakemake\\\"\\n if Args['clusterconfig_template'] is None:\\n Args['clusterconfig_template'] = Args['snakemakedir'] + \\\"/cluster_config_template.json\\\"\\n config.argparse_to_clusterconfig(Args,\\n Args['clusterconfig_template'])\\n os.makedirs(Args['tmpdir'] + \\\"/logs\\\", exist_ok=True)\\n if not Args['local']:\\n os.makedirs(Args['tmpdir'] + \\\"/cluster_logs\\\", exist_ok=True)\\n Args['cluster_cmd'] = f\\\"--cluster-config {Args['tmpdir']}/snakemake_cluster.json --cluster '{Args['cluster_cmd']}'\\\"\\n Args['cluster_cmd'] += f\\\" --resources cores={Args['max_resources']}\\\"\\n Args['cluster_cmd'] += f\\\" {Args['snakemake_args']}\\\"\\n print(f\\\"Cluster command: {Args['cluster_cmd']}\\\", file=sys.stderr)\\n else:\\n Args['cluster_cmd'] = ''\\n\\n # Extract list of species from MetaPhlAn database pickle\\n if (not os.path.isfile(Args['tmpdir'] + \\\"/repgenomes_urls.txt\\\") or\\n Args['force']):\\n print(\\\"1. Extract all species from MetaPhlAn database and prepare URLS \\\"\\n \\\"for download from NCBI.\\\", file=sys.stderr)\\n if os.path.isdir(Args['databasedir']):\\n db_versions = [re.search(r'mpa_v([0-9]+)_.+.pkl',\\n os.path.basename(db)).group(1)\\n for db in glob(f\\\"{Args['databasedir']}/*.pkl\\\")]\\n db_versions.sort()\\n if Args['metaphlanversion'] == 'latest':\\n db_version = db_versions[-1]\\n elif Args['metaphlanversion'].replace(\\\"v\\\", \\\"\\\") in db_versions:\\n db_version = Args['metaphlanversion'].replace(\\\"v\\\", \\\"\\\")\\n else:\\n print(f\\\"The database version {Args['metaphlanversion']} is not \\\"\\n f\\\"present in the database directory {Args['databasedir']}. \\\"\\n \\\"The following database versions are currently available: \\\"\\n f\\\"{', '.join(['v' + db for db in db_versions])}. Either \\\"\\n \\\"pick from the available or download the database using \\\"\\n \\\"MetaPhlAn.\\\",\\n file=sys.stderr)\\n sys.exit(1)\\n species_cont = species.Species(db_version, Args['databasedir'])\\n print(\\\"\\\\tExtract the strain information from the MetaPhlAn database\\\",\\n file=sys.stderr)\\n species_cont.extract_strains()\\n else:\\n print(f'The directory {Args[\\\"databasedir\\\"]} does not exist. Specify a '\\n 'valid directory that contains the MetaPhlAn databases.',\\n file=sys.stderr)\\n sys.exit(1)\\n\\n # Download the overview of RefSeq genomes and join information with genomes\\n print(\\\"\\\\tDownload the GCA assembly summary from NCBI\\\\n\\\",\\n file=sys.stderr)\\n subprocess.run(f'cd {Args[\\\"tmpdir\\\"]} && wget -N -nH '\\n '--user-agent=Mozilla/5.0 --relative -r --no-parent '\\n '--reject \\\"index.html*\\\" --cut-dirs=2 -e robots=off '\\n f'{Args[\\\"genbankurl\\\"]}', shell=True)\\n print(\\\"\\\\tJoin the GCA assembly summary information with the MetaPhlAn \\\"\\n \\\"database information\\\", file=sys.stderr)\\n species_cont.join_genbank(Args['tmpdir'] + \\\"/\\\" +\\n os.path.basename(Args['genbankurl']))\\n print(f\\\"\\\\tFetch information for missing genomes from NCBI Assembly directly.\\\",\\n file=sys.stderr)\\n species_cont.get_missing_information()\\n print(\\\"\\\\tDetermine the representative genomes of \\\"\\n f\\\"{species_cont.genomes.shape[0]} genomes present in the database\\\",\\n file=sys.stderr)\\n if Args['taxnames'] is None:\\n taxnames = []\\n else:\\n if os.path.isfile(Args['taxnames']):\\n taxnames = [line.rstrip()\\n for line in open(Args['taxnames'], 'rt')]\\n else:\\n print(f\\\"The species list file {Args['taxnames']} does not exist. \\\"\\n \\\"Specify the correct path to file.\\\", file=sys.stderr)\\n sys.exit(1)\\n species_cont.subset_taxa(taxnames)\\n species_cont.determine_representative_genomes()\\n print(f\\\"\\\\tIdentified {len(species_cont.representative_genomes)} genomes.\\\\n\\\"\\n \\\"\\\\tPrepare URL list for download of genomes from NCBI.\\\", file=sys.stderr)\\n species_cont.write_url_list(Args['tmpdir'] + \\\"/repgenomes_urls.txt\\\")\\n print(\\\"Write table with information to genomes used in phylogenetic \\\"\\n f\\\"analysis to {Args['tmpdir'] + '/genomes.tsv'}.\\\", file=sys.stderr)\\n species_cont.genomes_set.loc[species_cont.genomes_set['GCAid']\\n .isin(species_cont.representative_genomes)] \\\\\\n .to_csv(Args['tmpdir'] + \\\"/genomes.tsv\\\", sep=\\\"\\\\t\\\", index=False)\\n\\n if Args['stop_at'] == 'download_representative_genomes':\\n print(f\\\"Terminate at checkpoint {Args['stop_at']}.\\\", file=sys.stderr)\\n sys.exit(0)\\n elif (not os.path.isfile(Args['tmpdir'] + \\\"/done/download_representative_genomes\\\") or\\n Args['force']):\\n print(\\\"\\\\nDownload the representative genomes from NCBI\\\\n\\\", file=sys.stderr)\\n subprocess.run(f\\\"snakemake -s {Args['snakemakedir']}/download_genomes.Snakefile \\\"\\n f\\\"--configfile {Args['tmpdir']}/snakemake_config.json \\\"\\n f\\\"{Args['cluster_cmd']} \\\"\\n \\\"--restart-times 5 -k \\\"\\n f\\\"-j {Args['nproc']}\\\", shell=True,\\n stderr=open(Args['tmpdir'] +\\n \\\"/logs/snakemake-download_representative_genomes.log\\\",\\n \\\"at\\\"))\\n if not os.path.isfile(Args['tmpdir'] +\\n \\\"/done/download_representative_genomes\\\"):\\n errormessages.print_errormessage(\\\"download_representative_genomes\\\",\\n Args['tmpdir'])\\n sys.exit(1)\\n\\n if Args['stop_at'] == 'install_marker_database':\\n print(f\\\"Terminate at checkpoint {Args['stop_at']}.\\\", file=sys.stderr)\\n sys.exit(0)\\n elif (not os.path.isfile(Args['tmpdir'] + \\\"/done/install_marker_database\\\") or\\n Args['force']):\\n print(\\\"Prepare marker gene database of Segata et al. (2013) for tree \\\"\\n \\\"building\\\", file=sys.stderr)\\n database.write_superconfig_aa(Args['tmpdir'])\\n phylophlan_config = database.load_check_config(Args['tmpdir'],\\n Args['nproc'])\\n database.install_marker_database(Args['tmpdir'], phylophlan_config)\\n Path(Args['tmpdir'] + '/done/install_marker_database').touch(exist_ok=True)\\n\\n if Args['stop_at'] == 'fake_proteomes':\\n print(f\\\"Terminate at checkpoint {Args['stop_at']}.\\\", file=sys.stderr)\\n sys.exit(0)\\n elif (not os.path.isfile(Args['tmpdir'] + \\\"/done/fake_proteomes\\\") or\\n Args['force']):\\n print(\\\"Uncompress FastA files downloaded from NCBI, align against the \\\"\\n \\\"protein marker database of Segata et al. (2013) using DIAMOND \\\"\\n \\\"blastx, identify and extract marker genes, and translate into \\\"\\n \\\"amino acid sequences\\\", file=sys.stderr)\\n subprocess.run(f\\\"snakemake -s {Args['snakemakedir']}/fake_proteomes.Snakefile \\\"\\n f\\\"--configfile {Args['tmpdir']}/snakemake_config.json \\\"\\n f\\\"{Args['cluster_cmd']} \\\"\\n \\\"--restart-times 5 \\\"\\n f\\\"-j {Args['nproc']}\\\", shell=True,\\n stderr=open(Args['tmpdir'] +\\n \\\"/logs/snakemake-fake_proteomes.log\\\", \\\"at\\\"))\\n if not os.path.isfile(Args['tmpdir'] +\\n \\\"/done/fake_proteomes\\\"):\\n errormessages.print_errormessage(\\\"fake_proteomes\\\", Args['tmpdir'])\\n sys.exit(1)\\n\\n if Args['stop_at'] == 'protein_markers':\\n print(f\\\"Terminate at checkpoint {Args['stop_at']}.\\\", file=sys.stderr)\\n sys.exit(0)\\n elif (not os.path.isfile(Args['tmpdir'] + \\\"/done/protein_markers\\\") or\\n Args['force']):\\n print(\\\"Clean the fake proteomes, align against the protein marker \\\"\\n \\\"database of Segata et al. (2013) using DIAMOND blastp, and \\\"\\n \\\"identify and extract marker gene sequences.\\\", file=sys.stderr)\\n subprocess.run(f\\\"snakemake -s {Args['snakemakedir']}/protein_markers.Snakefile \\\"\\n f\\\"--configfile {Args['tmpdir']}/snakemake_config.json \\\"\\n f\\\"{Args['cluster_cmd']} \\\"\\n \\\"--restart-times 5 \\\"\\n f\\\"-j {Args['nproc']}\\\", shell=True,\\n stderr=open(Args['tmpdir'] +\\n \\\"/logs/snakemake-protein_markers.log\\\", \\\"at\\\"))\\n if not os.path.isfile(Args['tmpdir'] +\\n \\\"/done/protein_markers\\\"):\\n errormessages.print_errormessage(\\\"protein_markers\\\", Args['tmpdir'])\\n sys.exit(1)\\n\\n if Args['stop_at'] == 'alignment':\\n print(f\\\"Terminate at checkpoint {Args['stop_at']}.\\\", file=sys.stderr)\\n sys.exit(0)\\n elif (not os.path.isfile(Args['tmpdir'] + \\\"/done/alignment\\\") or\\n Args['force']):\\n print(\\\"Identify the marker genes that are present at least in four \\\"\\n \\\"genomes and make alignments, trim non-variant sites and remove \\\"\\n \\\"samples consisting of >= 90% gaps\\\", file=sys.stderr)\\n subprocess.run(f\\\"snakemake -s {Args['snakemakedir']}/alignment.Snakefile \\\"\\n f\\\"--configfile {Args['tmpdir']}/snakemake_config.json \\\"\\n f\\\"{Args['cluster_cmd']} \\\"\\n \\\"--restart-times 5 \\\"\\n f\\\"-j {Args['nproc']}\\\", shell=True,\\n stderr=open(Args['tmpdir'] +\\n \\\"/logs/snakemake-alignment.log\\\", \\\"at\\\"))\\n if not os.path.isfile(Args['tmpdir'] +\\n \\\"/done/alignment\\\"):\\n errormessages.print_errormessage(\\\"alignment\\\", Args['tmpdir'])\\n sys.exit(1)\\n\\n if Args['stop_at'] == 'tree':\\n print(f\\\"Terminate at checkpoint {Args['stop_at']}.\\\", file=sys.stderr)\\n sys.exit(0)\\n else:\\n if (not os.path.isfile(Args['tmpdir'] + \\\"/done/tree\\\") or\\n Args['force']):\\n print(\\\"Concatenate all markers into one alignment, build tree using \\\"\\n \\\"FastTree, and refining using RAxML\\\", file=sys.stderr)\\n subprocess.run(f\\\"snakemake -s {Args['snakemakedir']}/tree.Snakefile \\\"\\n f\\\"--configfile {Args['tmpdir']}/snakemake_config.json \\\"\\n f\\\"{Args['cluster_cmd']} \\\"\\n \\\"--restart-times 5 \\\"\\n f\\\"-j {Args['nproc']}\\\", shell=True,\\n stderr=open(Args['tmpdir'] +\\n \\\"/logs/snakemake-tree.log\\\", \\\"at\\\"))\\n if not os.path.isfile(Args['tmpdir'] + \\\"/done/tree\\\"):\\n errormessages.print_errormessage(\\\"tree\\\", Args['tmpdir'])\\n sys.exit(1)\\n\\n if (not os.path.isfile(Args['output']) or Args['force']):\\n print(\\\"Annotate the tree with taxonomic information and write to output \\\"\\n \\\"file.\\\", file=sys.stderr)\\n tree_annot_df = pd.read_csv(Args['tmpdir'] + \\\"/genomes.tsv\\\",\\n sep=\\\"\\\\t\\\")[['GCAid', 'label']] \\\\\\n .set_index(['GCAid'])\\n\\n treefn = Args['tmpdir'] + \\\"/MetaPhlAn3tree.RAxML.tre\\\"\\n tre = Tree(open(treefn, \\\"rt\\\").readline())\\n for l in tre.iter_leaves():\\n l.name = tree_annot_df.loc[l.name.replace(\\\".faa\\\", \\\"\\\"), 'label']\\n tre.write(outfile=Args['output'])\\n else:\\n print(f\\\"The tree output file {Args['output']} already exists and the \\\"\\n \\\"option '--force' has not been enabled to re-run it. Activate \\\"\\n \\\"this option to re-run all steps or remove the output file prior \\\"\\n \\\"to running the script.\\\", file=sys.stderr)\\n sys.exit(1)\",\n \"def __init__(self, *args, **kws):\\n from PyInstaller.config import CONF\\n Target.__init__(self)\\n self.strip_binaries = kws.get('strip', False)\\n self.upx_exclude = kws.get(\\\"upx_exclude\\\", [])\\n self.console = True\\n self.target_arch = None\\n self.codesign_identity = None\\n self.entitlements_file = None\\n\\n if CONF['hasUPX']:\\n self.upx_binaries = kws.get('upx', False)\\n else:\\n self.upx_binaries = False\\n\\n self.name = kws.get('name')\\n # Old .spec format included in 'name' the path where to collect files for the created app. app. New format\\n # includes only directory name.\\n #\\n # The 'name' directory is created in DISTPATH and necessary files are then collected to this directory.\\n self.name = os.path.join(CONF['distpath'], os.path.basename(self.name))\\n\\n self.toc = TOC()\\n for arg in args:\\n if isinstance(arg, TOC):\\n self.toc.extend(arg)\\n elif isinstance(arg, Target):\\n self.toc.append((os.path.basename(arg.name), arg.name, arg.typ))\\n if isinstance(arg, EXE):\\n self.console = arg.console\\n self.target_arch = arg.target_arch\\n self.codesign_identity = arg.codesign_identity\\n self.entitlements_file = arg.entitlements_file\\n for tocnm, fnm, typ in arg.toc:\\n if tocnm == os.path.basename(arg.name) + \\\".manifest\\\":\\n self.toc.append((tocnm, fnm, typ))\\n if not arg.append_pkg:\\n self.toc.append((os.path.basename(arg.pkgname), arg.pkgname, 'PKG'))\\n self.toc.extend(arg.dependencies)\\n else:\\n self.toc.extend(arg)\\n self.__postinit__()\",\n \"def main():\\n\\n BASIC.run(PROGRAM)\",\n \"def main() -> None:\\n # The first thing to do is get the lines of the PyFlex file we are given.\\n parser = Parser(filename=sys.argv[1])\\n parsed_data = parser.ParseFile()\\n\\n # Upon retrieving the Parsed Data, assign the parsed data to the\\n # Symbol Table.\\n SymbolTable.RULESET = parsed_data['ruleset']\\n SymbolTable.INSTRUCTIONS = parsed_data['instructions']\\n SymbolTable.CODE = parsed_data['code']\\n # SymbolTable.PrintTable()\\n\\n # Using the Generator backend, we can build the generated script\\n generator = Generator()\\n generator.GenerateNewScript()\\n\\n autopep8.fix_file(filename=generator.file_main)\\n\\n print(\\\"Generated Script can be found in {}\\\".format(generator.file_main))\",\n \"def generate(env):\\n # NOTE: SCons requires the use of this name, which fails gpylint.\\n\\n # Add pseudo-builders to set up the project and solution builders. These\\n # need to be available on all platforms so that SConscripts which reference\\n # them will work.\\n env.AddMethod(ComponentVSDirProject)\\n env.AddMethod(ComponentVSProject)\\n env.AddMethod(ComponentVSSolution)\\n env.AddMethod(ComponentVSSourceProject)\\n\\n # If not on Windows, don't do anything else\\n if sys.platform not in ('win32', 'cygwin'):\\n return\\n\\n # Include tools we need\\n env.Tool('gather_inputs')\\n\\n env.SetDefault(\\n COMPONENT_VS_PROJECT_DIR='$COMPONENT_VS_SOLUTION_DIR/projects',\\n COMPONENT_VS_PROJECT_SCRIPT_NAME='hammer.bat',\\n COMPONENT_VS_PROJECT_SCRIPT_PATH=(\\n '$$(ProjectDir)/$VS_PROJECT_TO_MAIN_DIR/'\\n '$COMPONENT_VS_PROJECT_SCRIPT_NAME'),\\n COMPONENT_VS_PROJECT_SUFFIX='.vcproj',\\n COMPONENT_VS_SOLUTION_DIR='$DESTINATION_ROOT/solution',\\n COMPONENT_VS_SOLUTION_SUFFIX='.sln',\\n COMPONENT_VS_ENABLED_PROJECTS=['$COMPONENT_VS_SCONS_DEFAULT_TARGETS'],\\n COMPONENT_VS_SOURCE_SUFFIXES=['$CPPSUFFIXES', '.rc', '.scons'],\\n COMPONENT_VS_SOURCE_FOLDERS=[('source', '$MAIN_DIR')], )\\n\\n AddTargetGroup('all_solutions', 'solutions can be built')\\n\\n # Add builders\\n vcprojaction = SCons.Script.Action(ComponentVSProjectBuilder,\\n varlist=[\\n 'COMPONENT_VS_PROJECT_SCRIPT_PATH',\\n 'TARGET_NAME',\\n 'TARGET_PATH',\\n ])\\n vcprojbuilder = SCons.Script.Builder(\\n action=vcprojaction,\\n suffix='$COMPONENT_VS_PROJECT_SUFFIX')\\n\\n source_vcproj_action = SCons.Script.Action(\\n ComponentVSSourceProjectBuilder,\\n varlist=[\\n 'COMPONENT_VS_SOURCE_FOLDERS',\\n 'COMPONENT_VS_SOURCE_SUFFIXES',\\n 'COMPONENT_VS_SOURCE_TARGETS',\\n ])\\n source_vcproj_builder = SCons.Script.Builder(\\n action=source_vcproj_action,\\n suffix='$COMPONENT_VS_PROJECT_SUFFIX')\\n\\n dir_vcproj_action = SCons.Script.Action(\\n ComponentVSDirProjectBuilder,\\n varlist=[\\n 'COMPONENT_VS_SOURCE_FOLDERS',\\n 'COMPONENT_VS_SOURCE_SUFFIXES',\\n 'PROJECT_SOURCES',\\n ])\\n dir_vcproj_builder = SCons.Script.Builder(\\n action=dir_vcproj_action,\\n suffix='$COMPONENT_VS_PROJECT_SUFFIX')\\n\\n slnaction = SCons.Script.Action(ComponentVSSolutionBuilder,\\n varlist=[\\n 'COMPONENT_VS_ENABLED_PROJECTS',\\n 'SOLUTION_FOLDERS',\\n 'SOLUTION_PROJECTS',\\n 'SOLUTION_TARGETS',\\n ])\\n slnbuilder = SCons.Script.Builder(\\n action=slnaction,\\n suffix='$COMPONENT_VS_SOLUTION_SUFFIX')\\n\\n env.Append(BUILDERS={\\n 'ComponentVSDirProjectBuilder': dir_vcproj_builder,\\n 'ComponentVSProjectBuilder': vcprojbuilder,\\n 'ComponentVSSolutionBuilder': slnbuilder,\\n 'ComponentVSSourceProjectBuilder': source_vcproj_builder,\\n })\",\n \"def cli(ctx):\",\n \"def cli(ctx):\",\n \"def main(_config, _run):\\n sacred.commands.print_config(_run)\\n dump_config_and_makefile()\\n prepare_and_train()\",\n \"def main():\\n\\n # pylint: disable=import-outside-toplevel\\n\\n import sys\\n cmd = ToyMaker()\\n sys.exit(cmd.main())\",\n \"def run() -> ():\\n if len(sys.argv) > 1:\\n show_help()\\n errs = get_cargo_input()\\n main_stack = make_proj_stack(errs)\\n while len(main_stack) > 0:\\n file_stack = make_file_stack(main_stack)\\n overwrite(file_stack)\\n\\n # FIXME\",\n \"def main():\",\n \"def main():\",\n \"def main():\",\n \"def main():\",\n \"def main():\",\n \"def main():\",\n \"def main():\",\n \"def main():\",\n \"def main():\",\n \"def main():\",\n \"def main():\",\n \"def main():\",\n \"def main():\",\n \"def main():\",\n \"def main():\",\n \"def main():\",\n \"def main():\",\n \"def main():\",\n \"def main():\",\n \"def main():\",\n \"def main():\",\n \"def main():\",\n \"def scons_write(self, target, source, env):\\n self.write()\",\n \"def main(source):\\n pass\",\n \"def main():\\n start_time = time()\\n\\n # x.py help <cmd> ...\\n if len(sys.argv) > 1 and sys.argv[1] == 'help':\\n sys.argv = [sys.argv[0], '-h'] + sys.argv[2:]\\n\\n help_triggered = (\\n '-h' in sys.argv) or ('--help' in sys.argv) or (len(sys.argv) == 1)\\n try:\\n bootstrap(help_triggered)\\n if not help_triggered:\\n print(\\\"Build completed successfully in {}\\\".format(\\n format_build_time(time() - start_time)))\\n except (SystemExit, KeyboardInterrupt) as error:\\n if hasattr(error, 'code') and isinstance(error.code, int):\\n exit_code = error.code\\n else:\\n exit_code = 1\\n print(error)\\n if not help_triggered:\\n print(\\\"Build completed unsuccessfully in {}\\\".format(\\n format_build_time(time() - start_time)))\\n sys.exit(exit_code)\",\n \"def Application(self, app, **kargs):\\n app_builddir = \\\"%s/%s/%s\\\" % (self.builddir, self.name, app)\\n\\n # Export env and args to the app SConstruct/home/benno/local//lib/scons-0.96/SCons/Node/FS.py\\n env = self\\n Export(\\\"env\\\")\\n args = kargs\\n Export(\\\"args\\\")\\n\\n # Call the apps SConstruct file to build it\\n self.apps[app] = self.SConscript(app, build_dir=app_builddir, duplicate=0)\\n return self.apps[app]\",\n \"def main():\\n\\n parser = argparse.ArgumentParser(description=_(\\\"Deploy and setup developers environment easily on ubuntu\\\"),\\n epilog=dedent(_(\\\"\\\"\\\"\\\\\\n Note that you can also configure different debug logs behaviors using LOG_CFG\\n pointing to a log yaml profile.\\n \\\"\\\"\\\")))\\n parser.add_argument(\\\"-v\\\", \\\"--verbose\\\", action=\\\"count\\\", default=0, help=_(\\\"increase output verbosity (2 levels)\\\"))\\n args = parser.parse_args()\\n\\n # setup logging level if set by the command line\\n if args.verbose == 1:\\n _setup_logging(level=logging.INFO)\\n elif args.verbose > 1:\\n _setup_logging(level=logging.DEBUG)\\n else:\\n _setup_logging()\\n\\n DownloadCenter([\\\"http://victorlin.me/posts/2012/08/26/good-logging-practice-in-python\\\"], lambda x: print(x))\",\n \"def main() -> None:\\n return\",\n \"def main(cls):\\n raise NotImplementedError\",\n \"def main():\\n licensify(_parse_args())\",\n \"def Main(parsed_args):\\n logging.basicConfig(\\n format='[[%(filename)s %(funcName)s]] - %(message)s',\\n level=(logging.ERROR if os.environ.get('CI') else logging.INFO))\\n EnsureNodeJsIsInstalled()\\n CheckPrereqs()\\n InstallNodeDependencies()\\n SetupOutDir(out_dir='dist')\\n GenValidatorProtoascii(out_dir='dist')\\n GenValidatorPb2Py(out_dir='dist')\\n GenValidatorProtoGeneratedJs(out_dir='dist')\\n GenValidatorGeneratedJs(out_dir='dist')\\n CompileValidatorMinified(out_dir='dist')\\n RunSmokeTest(out_dir='dist')\\n RunIndexTest()\\n CompileValidatorTestMinified(out_dir='dist')\\n CompileHtmlparserTestMinified(out_dir='dist')\\n CompileParseCssTestMinified(out_dir='dist')\\n CompileParseUrlTestMinified(out_dir='dist')\\n CompileAmp4AdsParseCssTestMinified(out_dir='dist')\\n CompileKeyframesParseCssTestMinified(out_dir='dist')\\n CompileParseSrcsetTestMinified(out_dir='dist')\\n GenerateTestRunner(out_dir='dist')\\n RunTests(update_tests=parsed_args.update_tests, out_dir='dist')\",\n \"def cli(args): # noqa; pylint: disable=unused-argument\",\n \"def setup(self):\\n super(__class__, self).setup()\\n # construct command line call\\n setup_script = '%s/tfMRI.py' % \\\\\\n os.environ['ABCDTASKPREPDIR']\\n arg1 = self.kwargs['path']\\n arg2 = self.kwargs['sourcedata_root']\\n arg3 = self.kwargs['subject']\\n arg4 = self.kwargs['session']\\n anat_metadata = self.config.get_bids('t1w_metadata')\\n # get make/software information\\n make = anat_metadata['Manufacturer']\\n if make == 'GE':\\n reg = re.compile(r'.*(DV2[56]).*')\\n software_version = reg.match(anat_metadata[\\n 'SoftwareVersions']).group(1)\\n else:\\n software_version = 'NA'\\n cmd = ' '.join((setup_script, arg1, arg2, arg3, arg4, make,\\n software_version))\\n print(cmd)\\n\\n log_dir = self._get_log_dir()\\n out_log = os.path.join(log_dir, self.__class__.__name__ + '_setup.out')\\n err_log = os.path.join(log_dir, self.__class__.__name__ + '_setup.err')\\n result = self.call(cmd, out_log, err_log)\",\n \"def main(cmd_line_args):\\n if cmd_line_args.clean:\\n clean_wrapper(cmd_line_args)\\n if cmd_line_args.nobuild == False:\\n build_wrapper(cmd_line_args)\",\n \"def main():\\n if len(sys.argv) != 2:\\n print('USAGE:\\\\tpython3 inoti_make.py <CONFIG_FILE>')\\n exit(1)\\n config = parse(sys.argv[1])\\n listen_folders(config)\",\n \"def cli():\\n pass\",\n \"def main():\\n logger.info(\\\"Running main function...\\\")\\n args = get_args()\\n # rose suite default location\\n if args.suite_dir:\\n default_suite = args.suite_dir\\n rose_config_template = os.path.join(default_suite, \\\"rose-suite.conf\\\")\\n\\n # get command line arguments\\n recipe_files = args.recipe_files\\n config_file = args.config_file\\n main_dir = args.main_dir\\n log_level = args.log_level\\n\\n # setup rose suite\\n run_rose = _setup_work(rose_config_template, recipe_files,\\n config_file, main_dir, default_suite, log_level)\\n\\n # submit to cylc\\n if not args.no_submit:\\n _run_suite(run_rose)\",\n \"def main():\\n with open('config.json') as config_file:\\n configs = json.load(config_file)\\n\\n jar_list = utilities.upload_jars(configs)\\n utilities.sign_jars(configs)\\n\\n artifact_folder = utilities.prepare_artifacts(configs, jar_list)\\n\\n repo_id = utilities.create_staging_repo(configs)\\n utilities.deploy_to_staging_repo(configs, artifact_folder, repo_id)\\n utilities.close_staging_repo(configs, repo_id)\",\n \"def main():\\n # Parse arguments.\\n parser = argparse.ArgumentParser(description='Console for cluster test.')\\n parser.add_argument('-c', dest='command_config_file',\\n default='command_config',\\n help='set command config file path (default=command_config)')\\n args = parser.parse_args()\\n\\n console = Console()\\n console.config(args.command_config_file)\\n console.init_proxies()\\n console.run()\",\n \"def main():\\n logging.basicConfig(\\n level=logging.DEBUG, format=\\\"%(levelname)s: %(message)s\\\")\\n root.add_command(all_)\\n root.add_command(build)\\n root.add_command(cleanup)\\n root.add_command(clitest)\\n root.add_command(run)\\n root.add_command(push)\\n root.add_command(unittest)\\n root.add_command(list_images)\\n root.add_command(list_stages)\\n root()\",\n \"def main():\\n parser = optparse.OptionParser(usage='%prog [options]', version=\\\"0.0.1\\\")\\n parser.add_option('--settings', \\\\\\n help='Python path to settings module. If this isn\\\\'t provided, the DJANGO_SETTINGS_MODULE enviroment variable will be used.')\\n\\n parser.add_option('-v', '--verbose', action='store_true', dest='verbose', \\\\\\n default=False, help='Verbose output.')\\n options = parser.parse_args()[0]\\n if options.settings:\\n os.environ[\\\"DJANGO_SETTINGS_MODULE\\\"] = options.settings\\n else:\\n os.environ[\\\"DJANGO_SETTINGS_MODULE\\\"] = \\\"settings\\\"\\n\\n probe_all()\",\n \"def main():\\n # This have specific paths to prevent abitrary binaries from being\\n # executed. The \\\"gsi\\\"* utilities are configured to use either grid proxies\\n # or ssh, automatically.\\n remoteLoginCmd = \\\"/usr/bin/gsissh\\\"\\n remoteCopyCmd = \\\"/usr/bin/gsiscp\\\"\\n\\n UNKNOWN_PLATFORM_EXIT_CODE = 10\\n MISSING_PBS_CONFIG_EXIT_CODE = 20\\n\\n p = AllocatorParser(sys.argv[0])\\n platform = p.getPlatform()\\n\\n creator = Allocator(platform, p.getArgs(), \\\"$HOME/.lsst/condor-info.py\\\")\\n\\n platformPkgDir = lsst.utils.getPackageDir(\\\"ctrl_platform_\\\"+platform)\\n configName = os.path.join(platformPkgDir, \\\"etc\\\", \\\"config\\\", \\\"pbsConfig.py\\\")\\n execConfigName = os.path.join(platformPkgDir, \\\"etc\\\", \\\"config\\\", \\\"execConfig.py\\\")\\n\\n creator.load(execConfigName)\\n\\n creator.loadPbs(configName)\\n\\n verbose = creator.isVerbose()\\n \\n pbsName = os.path.join(platformPkgDir, \\\"etc\\\", \\\"templates\\\", \\\"generic.pbs.template\\\")\\n generatedPbsFile = creator.createPbsFile(pbsName)\\n\\n condorFile = os.path.join(platformPkgDir, \\\"etc\\\", \\\"templates\\\", \\\"glidein_condor_config.template\\\")\\n generatedCondorConfigFile = creator.createCondorConfigFile(condorFile)\\n\\n scratchDirParam = creator.getScratchDirectory()\\n template = Template(scratchDirParam)\\n scratchDir = template.substitute(USER_HOME=creator.getUserHome())\\n userName = creator.getUserName()\\n \\n hostName = creator.getHostName()\\n\\n utilityPath = creator.getUtilityPath()\\n\\n #\\n # execute copy of PBS file to XSEDE node\\n #\\n cmd = \\\"%s %s %s@%s:%s/%s\\\" % (remoteCopyCmd, generatedPbsFile, userName, hostName, scratchDir, os.path.basename(generatedPbsFile))\\n if verbose:\\n print cmd\\n exitCode = runCommand(cmd, verbose)\\n if exitCode != 0:\\n print \\\"error running %s to %s.\\\" % (remoteCopyCmd, hostName)\\n sys.exit(exitCode)\\n\\n #\\n # execute copy of Condor config file to XSEDE node\\n #\\n cmd = \\\"%s %s %s@%s:%s/%s\\\" % (remoteCopyCmd, generatedCondorConfigFile, userName, hostName, scratchDir, os.path.basename(generatedCondorConfigFile))\\n if verbose:\\n print cmd\\n exitCode = runCommand(cmd, verbose)\\n if exitCode != 0:\\n print \\\"error running %s to %s.\\\" % (remoteCopyCmd, hostName)\\n sys.exit(exitCode)\\n\\n #\\n # execute qsub command on XSEDE node to perform Condor glide-in\\n #\\n cmd = \\\"%s %s@%s %s/qsub %s/%s\\\" % (remoteLoginCmd, userName, hostName, utilityPath, scratchDir, os.path.basename(generatedPbsFile))\\n if verbose:\\n print cmd\\n exitCode = runCommand(cmd, verbose)\\n if exitCode != 0:\\n print \\\"error running %s to %s.\\\" % (remoteLoginCmd, hostName)\\n sys.exit(exitCode)\\n\\n nodes = creator.getNodes()\\n slots = creator.getSlots()\\n wallClock = creator.getWallClock()\\n nodeString = \\\"\\\"\\n if int(nodes) > 1:\\n nodeString = \\\"s\\\"\\n print \\\"%s node%s will be allocated on %s with %s slots per node and maximum time limit of %s\\\" % (nodes, nodeString, platform, slots, wallClock)\\n print \\\"Node set name:\\\"\\n print creator.getNodeSetName()\\n sys.exit(0)\",\n \"def cli(ctx):\\n pass\",\n \"def cli(ctx):\\n pass\",\n \"def setup(self, args={}):\\n\\n return Status.RUN\",\n \"def entry() -> None:\\n import argparse\\n import configparser\\n import multiprocessing\\n import sys\\n import fab\\n\\n logger = logging.getLogger('fab')\\n logger.addHandler(logging.StreamHandler(sys.stderr))\\n\\n description = 'Flexible build system for scientific software.'\\n\\n parser = argparse.ArgumentParser(add_help=False,\\n description=description)\\n # We add our own help so as to capture as many permutations of how people\\n # might ask for help. The default only looks for a subset.\\n parser.add_argument('-h', '-help', '--help', action='help',\\n help='Print this help and exit')\\n parser.add_argument('-V', '--version', action='version',\\n version=fab.__version__,\\n help='Print version identifier and exit')\\n parser.add_argument('-v', '--verbose', action='count', default=0,\\n help='Increase verbosity (may be specified once '\\n 'for moderate and twice for debug verbosity)')\\n parser.add_argument('-w', '--workspace', metavar='PATH', type=Path,\\n default=Path.cwd() / 'working',\\n help='Directory for working files.')\\n parser.add_argument('--nprocs', action='store', type=int, default=2,\\n choices=range(2, multiprocessing.cpu_count()),\\n help='Provide number of processors available for use,'\\n 'default is 2 if not set.')\\n parser.add_argument('source', type=Path,\\n help='The path of the source tree to build')\\n parser.add_argument('conf_file', type=Path, default='config.ini',\\n help='The path of the configuration file')\\n arguments = parser.parse_args()\\n\\n verbosity_levels = [logging.WARNING, logging.INFO, logging.DEBUG]\\n verbosity = min(arguments.verbose, 2)\\n logger.setLevel(verbosity_levels[verbosity])\\n\\n config = configparser.ConfigParser(allow_no_value=True)\\n configfile = arguments.conf_file\\n config.read(configfile)\\n settings = config['settings']\\n flags = config['flags']\\n\\n # If not provided, name the exec after the target\\n if settings['exec-name'] == '':\\n settings['exec-name'] = settings['target']\\n\\n application = Fab(arguments.workspace,\\n settings['target'],\\n settings['exec-name'],\\n flags['fpp-flags'],\\n flags['fc-flags'],\\n flags['ld-flags'],\\n arguments.nprocs)\\n application.run(arguments.source)\",\n \"def main():\\n log = get_logger(\\\"bakery.bake.main\\\")\\n config = configure()\\n\\n bake_instructions = PREAMBLE + open(config.bakefile).read()\\n\\n try:\\n exec(bake_instructions, globals())\\n log.info(fg.bright.green(\\\"BUILD SUCCEEDED\\\"))\\n except Exception as e:\\n build.critical = True\\n log.error(str(e))\\n if DEBUG:\\n log.exception(\\\"Exception details follow.\\\")\\n log.info(fg.bright.red(\\\"BUILD FAILED\\\"))\\n\\n if build.build_count == 0 and not build.critical:\\n log.warn(\\n ansilog.Node.list(\\n \\\"Nothing was built, did you forget to decorate your module with \\\",\\n fg.bright.yellow(\\\"@build\\\"),\\n \\\"?\\\",\\n )\\n )\\n\\n sys.exit(0)\",\n \"def main(args=None):\\n if args is None:\\n args = sys.argv[1:]\\n\\n print('This is the main function of pycharm')\\n\\n \\\"\\\"\\\"Read the args\\\"\\\"\\\"\\n if len(args) > 0:\\n for i in args:\\n print(i)\\n\\n if args[0] == 'create_setup':\\n remove = False\\n if len(args) > 2:\\n if args[2] == '-removeold':\\n remove = True\\n\\n from pymake.tools.create_setup import create_setup\\n create_setup(args[1], remove)\\n else:\\n print('No parameters passed')\",\n \"def main(args=None):\\n pass\",\n \"def main(self):\",\n \"def initialise(self, args, environ):\",\n \"def main(args):\\n\\n tools = []\\n for tool in args.tools: # Parse tools, their subs and args\\n logging.info(\\\"Parsing tool: %r\\\", tool)\\n\\n tsig = parse_tool_sig(tool)\\n if not tsig or not tsig.get(\\\"snames\\\", None):\\n logging.error(\\\"failed parsing snames from tool: '%s'\\\", tool)\\n continue\\n\\n tools.append(tsig)\\n\\n return args.gen(args, tools)\",\n \"def run(self):\\n self.archive_bash_inits()\\n self.create_paths()\\n self.copy_files()\\n self.make_git_config()\\n self.ensure_bash_history()\",\n \"def setup():\\n\\tglobal config_parser, config_file\\n\\tglobal prefix\\n\\n\\tif os.path.islink(sys.argv[0]):\\n\\t\\tlink = os.readlink(sys.argv[0])\\n\\n\\t\\tif not os.path.isabs(link):\\n\\t\\t\\tlink = os.path.join(os.path.dirname(sys.argv[0]), link)\\n\\n\\t\\tprefix = os.path.dirname(os.path.abspath(link))\\n\\telse:\\n\\t\\tprefix = os.path.dirname(os.path.abspath(sys.argv[0]))\\n\\n\\tconfig_parser = ConfigParser.ConfigParser()\\n\\tset_defaults()\\n\\n\\tconfig_file = os.path.join (xdg_config_home, \\\"sushi\\\", \\\"nigiri\\\")\\n\\n\\tif not check_config_file(config_file):\\n\\t\\tprint \\\"Config file creation failed. Aborting.\\\"\\n\\t\\treturn\\n\\n\\tread_config_file()\"\n]","isConversational":false},"negative_scores":{"kind":"list like","value":["0.6464546","0.6412459","0.6104805","0.6104805","0.60741913","0.6071515","0.5993396","0.59785044","0.59724134","0.59724134","0.59724104","0.59639555","0.59407467","0.5926329","0.5874594","0.58656955","0.5856603","0.58456635","0.5837887","0.5832198","0.58242744","0.5819705","0.5816474","0.5812681","0.58060247","0.57843465","0.5778255","0.5777902","0.5771409","0.5757208","0.573817","0.57302827","0.57265884","0.57265884","0.57265884","0.57265884","0.57239914","0.5716788","0.5711425","0.57045263","0.56916535","0.5689738","0.5681911","0.5681911","0.5681617","0.56756335","0.5673323","0.5672927","0.5672927","0.5672927","0.5672927","0.5672927","0.5672927","0.5672927","0.5672927","0.5672927","0.5672927","0.5672927","0.5672927","0.5672927","0.5672927","0.5672927","0.5672927","0.5672927","0.5672927","0.5672927","0.5672927","0.5672927","0.5672927","0.5663778","0.56513345","0.5649717","0.5644603","0.5643543","0.563129","0.56220126","0.5620307","0.5616807","0.56131345","0.5605021","0.56008995","0.55956817","0.55935884","0.5582362","0.5580359","0.5578551","0.55784225","0.5576641","0.5571094","0.5568178","0.5568178","0.5557815","0.55560523","0.55549717","0.5551391","0.5550394","0.55492043","0.55463076","0.5540798","0.5539586","0.55384046"],"string":"[\n \"0.6464546\",\n \"0.6412459\",\n \"0.6104805\",\n \"0.6104805\",\n \"0.60741913\",\n \"0.6071515\",\n \"0.5993396\",\n \"0.59785044\",\n \"0.59724134\",\n \"0.59724134\",\n \"0.59724104\",\n \"0.59639555\",\n \"0.59407467\",\n \"0.5926329\",\n \"0.5874594\",\n \"0.58656955\",\n \"0.5856603\",\n \"0.58456635\",\n \"0.5837887\",\n \"0.5832198\",\n \"0.58242744\",\n \"0.5819705\",\n \"0.5816474\",\n \"0.5812681\",\n \"0.58060247\",\n \"0.57843465\",\n \"0.5778255\",\n \"0.5777902\",\n \"0.5771409\",\n \"0.5757208\",\n \"0.573817\",\n \"0.57302827\",\n \"0.57265884\",\n \"0.57265884\",\n \"0.57265884\",\n \"0.57265884\",\n \"0.57239914\",\n \"0.5716788\",\n \"0.5711425\",\n \"0.57045263\",\n \"0.56916535\",\n \"0.5689738\",\n \"0.5681911\",\n \"0.5681911\",\n \"0.5681617\",\n \"0.56756335\",\n \"0.5673323\",\n \"0.5672927\",\n \"0.5672927\",\n \"0.5672927\",\n \"0.5672927\",\n \"0.5672927\",\n \"0.5672927\",\n \"0.5672927\",\n \"0.5672927\",\n \"0.5672927\",\n \"0.5672927\",\n \"0.5672927\",\n \"0.5672927\",\n \"0.5672927\",\n \"0.5672927\",\n \"0.5672927\",\n \"0.5672927\",\n \"0.5672927\",\n \"0.5672927\",\n \"0.5672927\",\n \"0.5672927\",\n \"0.5672927\",\n \"0.5672927\",\n \"0.5663778\",\n \"0.56513345\",\n \"0.5649717\",\n \"0.5644603\",\n \"0.5643543\",\n \"0.563129\",\n \"0.56220126\",\n \"0.5620307\",\n \"0.5616807\",\n \"0.56131345\",\n \"0.5605021\",\n \"0.56008995\",\n \"0.55956817\",\n \"0.55935884\",\n \"0.5582362\",\n \"0.5580359\",\n \"0.5578551\",\n \"0.55784225\",\n \"0.5576641\",\n \"0.5571094\",\n \"0.5568178\",\n \"0.5568178\",\n \"0.5557815\",\n \"0.55560523\",\n \"0.55549717\",\n \"0.5551391\",\n \"0.5550394\",\n \"0.55492043\",\n \"0.55463076\",\n \"0.5540798\",\n \"0.5539586\",\n \"0.55384046\"\n]","isConversational":false},"document_score":{"kind":"string","value":"0.0"},"document_rank":{"kind":"string","value":"-1"}}},{"rowIdx":93278,"cells":{"query":{"kind":"string","value":"Extract tags and mentions from a string. Returns a tuple with a set of tags and mentioned users, if they exist."},"document":{"kind":"string","value":"def extract_tags_and_mentions(s: str) -> typing.Tuple[set, set]:\n if not s:\n return set(), set()\n\n # Build a set after stripping the # or @ sign before every item found with\n # the regular expression.\n tags = set(t for t in find_hashtags_re.findall(s))\n mentions = set(m[1:] for m in find_mentions_re.findall(s))\n mentioned_user = set(\n User.objects.get(username=username)\n for username in mentions\n if User.objects.filter(username=username).exists()\n )\n\n # mentions = set(m for m in mentions)\n\n return tags, mentioned_user"},"metadata":{"kind":"string","value":"{\n \"objective\": {\n \"self\": [],\n \"paired\": [],\n \"triplet\": [\n [\n \"query\",\n \"document\",\n \"negatives\"\n ]\n ]\n }\n}"},"negatives":{"kind":"list like","value":["def get_tagged_user(line, unique_users):\n tagged_user = None\n\n for user in unique_users:\n \n tagged_user = re.search(f\"@{user}\\s*\", line)\n \n if tagged_user != None:\n tagged_user = tagged_user.group(0).replace(\"@\", \"\").strip()\n line = line.replace(f\"@{user} \", \"\")\n break\n \n return (tagged_user, line)","def tags(string):\n\treturn set([x.strip() for x in string.split(',')])","def _parse_text_as_direct_mention(\n message_text: str,\n ) -> tuple[Optional[str], Optional[str]]:\n matches = re.search(MENTION_REGEX, message_text)\n # the first group contains the username, the second group contains the remaining message\n if matches:\n return matches.group(1), matches.group(2).strip()\n return None, None","def findTags(user_input, tagged_text):\n result = []\n for item in tagged_text:\n for w in user_input:\n if w[WORD] == item[WORD]:\n tup = (w[WORD], item[TAG])\n result.append(tup)\n continue\n\n return result","def parse_direct_mention(message_text):\n matches = re.search(MENTION_REGEX, message_text)\n print (matches)\n # the first group contains the username, the second group contains the remaining message\n return (matches.group(1), matches.group(2).strip()) if matches else (None, None)","def parse_direct_mention(message_text):\n # print(message_text)\n matches = re.search(MENTION_REGEX, message_text)\n # the first group contains the username, the second group contains the remaining message\n return (matches.group(1), matches.group(2).strip()) if matches else (None, None)","def parse_direct_mention(message_text):\r\n matches = re.search(MENTION_REGEX, message_text)\r\n # the first group contains the username, the second group contains the remaining message\r\n return (matches.group(1), matches.group(2).strip()) if matches else (None, None)","def suggest_tags(data):\r\n tag_set = set()\r\n if not data:\r\n return tag_set\r\n\r\n # The string might be a url that needs some cleanup before we parse for\r\n # suggestions\r\n parsed = urlparse(data)\r\n\r\n # Check if title is url. If title is not a string, url, and title will be\r\n # the same so no need to consider tags from url.\r\n if parsed.hostname:\r\n tag_set.update(_generate_nouns_from_url(parsed.path))\r\n else:\r\n # If the title is not a url extract nouns from title and the url.\r\n tag_set.update(_generate_nouns_from_url(data))\r\n\r\n return tag_set","def parse_direct_mention(message_text):\n matches = re.search(MENTION_REGEX, message_text)\n # the first group contains the username, the second group contains the remaining message\n return (matches.group(1), matches.group(2).strip()) if matches else (None, None)","def parse_direct_mention(message_text):\n matches = re.search(MENTION_REGEX, message_text)\n # the first group contains the username, the second group contains the remaining message\n return (matches.group(1), matches.group(2).strip()) if matches else (None, None)","def parse_direct_mention(message_text):\n matches = re.search(MENTION_REGEX, message_text)\n # the first group contains the username, the second group contains the remaining message\n return (matches.group(1), matches.group(2).strip()) if matches else (None, None)","def parse_direct_mention(message_text):\n matches = re.search(MENTION_REGEX, message_text)\n # the first group contains the username, the second group contains the remaining message\n return (matches.group(1), matches.group(2).strip()) if matches else (None, None)","def parse_direct_mention(message_text):\n matches = re.search(MENTION_REGEX, message_text)\n # the first group contains the username, the second group contains the remaining message\n return (matches.group(1), matches.group(2).strip()) if matches else (None, None)","def parse_direct_mention(message_text):\n matches = re.search(MENTION_REGEX, message_text)\n # 1st gourp has username, 2nd group called remaining message\n return (matches.group(1), matches.group(2).strip()) if matches else (None, None)","def parse_direct_mention(message_text):\n matches = re.search(MENTION_REGEX, message_text)\n # the first group contains the username, the second group contains\n # the remaining message\n return (matches.group(1),\n matches.group(2).strip()) if matches else (None, None)","def parse_direct_mention(message_text):\n matches = re.search(MENTION_REGEX, message_text)\n # the first group contains the username, the second group\n # contains the remaining message\n return (matches.group(1),\n matches.group(2).strip()) if matches else (None, None)","def getTagList(tags):\n tags = tags[1:len(tags)-1]\n return tags.split('><')","def parse_direct_mention(self, message_text):\n matches = re.search(self.MENTION_REGEX, message_text)\n # the first group contains the username, the second group contains the remaining message\n return (matches.group(1), matches.group(2).strip()) if matches else (None, None)","def parse_direct_mention(self, message_text):\n matches = re.search(self.MENTION_REGEX, message_text)\n # the first group contains the username, the second group contains the remaining message\n return (matches.group(1), matches.group(2).strip()) if matches else (None, None)","def parse_direct_mention(self,message_text):\n matches = re.search(self.MENTION_REGEX, message_text)\n # the first group contains the username, the second group contains the remaining message\n return (matches.group(1), matches.group(2).strip()) if matches else (None, None)","def parse_direct_mention(message_text):\n\tmatches = re.search(MENTION_REGEX, message_text)\n\t# the first group contains the username, the second group contains the remaining message\n\treturn (matches.group(1), matches.group(2).strip()) if matches else (None, None)","def parse_direct_mention(message_text):\n matches = re.search(MENTION_REGEX, message_text)\n # the first group contains the username, the second group contains the remaining message\n return (matches.group(1), matches.group(2).strip()) if matches else (None, message_text.strip())","def parseFriends(line):\n #parts = re.split('\\t', line)\n parts = line.split('\\t')\n #parts[1] = re.split(',', parts[1])\n parts[1] = set(parts[1].split(','))\n return parts[0], parts[1]","def parse_tags(tagstring):\n if not tagstring:\n return []\n\n tagstring = force_str(tagstring)\n\n words = []\n buffer = []\n # Defer splitting of non-quoted sections until we know if there are\n # any unquoted commas.\n to_be_split = []\n i = iter(tagstring)\n try:\n while True:\n c = six.next(i)\n if c == '\"':\n if buffer:\n to_be_split.append(''.join(buffer))\n buffer = []\n c = six.next(i)\n while c != '\"':\n buffer.append(c)\n c = six.next(i)\n if buffer:\n word = ''.join(buffer).strip()\n if word:\n words.append(word)\n buffer = []\n else:\n buffer.append(c)\n except StopIteration:\n # If we were parsing an open quote which was never closed treat\n # the buffer as unquoted.\n if buffer:\n to_be_split.append(''.join(buffer))\n if to_be_split:\n for chunk in to_be_split:\n words.extend(split_strip(chunk, settings.TAGGIT_SELECTIZE['DELIMITER']))\n words = list(set(words))\n words.sort()\n return words","def _parse_tags(tags: str):\n return dict(item.split(\":\") for item in shlex.split(tags)) # type: ignore","def parse(text):\n # Remove hashtag and trailing whitespaces\n clean_text = reHASH.sub(\"\", text).strip()\n # We discard those entries with URLs inside\n if reURL.search(clean_text):\n raise\n # Also discard those that contain @user forms\n if '@' in clean_text:\n raise\n parts = clean_text.split(':')\n if len(parts) < 2:\n raise\n term = parts[0].strip().lower()\n meaning = \":\".join(parts[1:]).strip()\n return term, meaning","def get_tags(html):\n\ttitle = re.findall('\"title\":\"(.*?)\",', html)[0]\n\ttitle = codecs.getdecoder(\"unicode_escape\")(title)[0]\n\n\tartist = re.findall('\"username\":\"(.*?)\",', html)[0]\n\tartist = codecs.getdecoder(\"unicode_escape\")(artist)[0]\n\n\tgenre = re.findall('\"genre\":\"(.*?)\",', html)[0]\n\tgenre = codecs.getdecoder(\"unicode_escape\")(genre)[0]\n\n\treturn title, artist, genre","def separate_hastags_mentions_urls(tweet):\n \n text = tweet.lower()\n hashtag_list = re.findall(\"#([a-zA-Z0-9_]{1,50})\", text)\n \n text = re.sub(r'http\\S+', '', text)\n clean_tweet = re.sub(\"@[A-Za-z0-9_]+\",\"\", text)\n clean_tweet = re.sub(\"#[A-Za-z0-9_]+\",\"\", clean_tweet)\n \n return clean_tweet, hashtag_list","def extract(msg):\n\n rgx = url_regex()\n found = re.findall(rgx, msg)\n links_set = Set()\n for link in found:\n # dunno if I like this. don't need seperate groups to make this work.\n links_set.add(link[0].strip())\n return list(links_set)","def get_usernames_from_tag(tag):\n try:\n tag_url = 'https://www.instagram.com/explore/tags/{0}/?__a=1'.format(tag)\n tag_res = requests.get(tag_url).json()\n\n nodes = tag_res['tag']['media']['nodes']\n\n new_urls = [node['code'] for node in nodes]\n new_req = [requests.get('https://www.instagram.com/p/{0}/?__a=1'.format(url)).json() for url in new_urls]\n\n usernames = [req['graphql']['shortcode_media']['owner']['username'] for req in new_req]\n\n return usernames\n\n except Exception as err:\n print('{0} - This account was not classified...'.format(str(err)))\n return []","def parse_tag(self, tag):\n \n mytag = \"latest\"\n mydigest = None\n\n regex = \"([\\w\\d\\.\\-]+)@?([\\w\\d\\.\\-]*)$\"\n\n regex_matched = re.match(regex, tag)\n mytag = regex_matched.group(1)\n mydigest = regex_matched.group(2)\n \n if regex_matched is None:\n mytag = \"latest\"\n\n return (mytag, mydigest)","def parse_output(slack_rtm_output): # pragma: no cover\n output_list = slack_rtm_output\n if output_list and len(output_list) > 0:\n for output in output_list:\n if output and 'text' in output and AT_BOT in output['text']:\n # return text after the @ mention, whitespace removed\n return output['text'].split(AT_BOT)[1].strip().lower(), \\\n output['channel'], output['user']\n return None, None, None","def getHashtagsAndMentions(tweets):\n hashtags = Counter()\n mentions = Counter()\n plain = Counter()\n\n pattern = re.compile(r\"[^#@\\w'-]+\")\n\n for t in tweets:\n words = pattern.split(t.message)\n for word in words:\n # Ignore null strings caused by split characters at the end of a\n # message and remove standalone hyphens.\n if word and not word.startswith(\"-\"):\n # Increment count for the word in the Counter.\n if word.startswith(\"#\"):\n hashtags.update({word: 1})\n elif word.startswith(\"@\"):\n mentions.update({word: 1})\n else:\n # TODO: apply nltk.corpus.stopwords.words() here,\n # across languages. Consider that the stopwords cut off\n # before apostrophe, therefore check if the word\n # starts with the stopword.\n plain.update({word: 1})\n\n return hashtags, mentions, plain","def tag(s, tokenize=True, encoding=\"utf-8\"):\n tags = []\n for sentence in parse(s, tokenize, True, False, False, False, encoding).split():\n for token in sentence:\n tags.append((token[0], token[1]))\n return tags","def parse_slack_output(slack_rtm_output):\n output_list = slack_rtm_output\n if output_list and len(output_list) > 0:\n for output in output_list:\n if output and 'text' in output and AT_BOT in output['text']:\n # return text after the @ mention, whitespace removed\n s = output['text'].split(AT_BOT)[1].strip().lower()\n exclude = set(string.punctuation)\n s = ''.join(ch for ch in s if ch not in exclude)\n return output['channel'], s, output['user'],\n return None, None, None","def parse_tags(s: str) -> List[str]:\n tags = []\n buf = []\n in_quoted = None\n\n for c in s:\n if in_quoted:\n if c == in_quoted:\n in_quoted = None\n else:\n buf.append(c)\n elif c == '\"' or c == '\\'':\n in_quoted = c\n elif c == ',':\n if buf:\n tag = ''.join(buf).strip()\n if tag:\n tags.append(tag)\n buf.clear()\n else:\n buf.append(c)\n\n if buf:\n tag = ''.join(buf).strip()\n if tag:\n tags.append(tag)\n\n return tags","def _get_tags(tag_string):\n print \"inside get_tags, string is: \", tag_string\n tag_objects_list = []\n # remove all whitespaces\n tag_string_cleaned = tag_string.replace(\" \", \"\")\n tokens = tag_string_cleaned.split(',')\n for tok in tokens:\n try:\n tag_object = Tag.objects.get(tag_name=tok)\n except Tag.DoesNotExist:\n tag_object = Tag(tag_name=tok)\n tag_object.save()\n if tag_object not in tag_objects_list:\n tag_objects_list.append(tag_object)\n return tag_objects_list","def parse(owners_lines):\n results = []\n for owner_line in owners_lines:\n # strip the leading @ for github users/teams\n if owner_line.startswith('@'):\n owner_line = owner_line[1:]\n owner_line = owner_line.strip()\n if not owner_line:\n continue\n if owner_line.startswith('#'):\n continue\n if ' ' in owner_line:\n result = tuple(owner_line.split(' ', 1))\n else:\n result = (owner_line, None)\n results.append(result)\n return results","def tweet_parser(tweet):\n tweet = tweet.encode()\n\n match = re.match(\n r'^@\\w{11} (@(?P<opponent>[^\\s]+) )?#(?P<game>\\w+)( (?P<move>[^\\s]+))?( (?P<message>.*))?$',\n tweet\n )\n\n if match:\n return match.groupdict()\n else:\n raise ValueError(\"Tweet not formatted correctly.\")","def parse_tags(self, in_str):\n self.projects = []\n self.contexts = []\n for match in TAG_REGEX.finditer(in_str):\n tag_str = match.group(1)\n if tag_str[0] == \"+\":\n self.projects.append(tag_str[1:])\n elif tag_str[0] == \"@\":\n self.contexts.append(tag_str[1:])\n else:\n err_str = \"Unable to parse tag: %s\" % tag_str\n raise ValueError(err_str)\n words = re.sub(TAG_REGEX, \"\", in_str).split()\n return \" \".join(words)","def tags(self):\n return tuple([x.strip() for x in self._dict.get('tags').split(',')])","def parse_etags(etag_str):\n if etag_str.strip() == \"*\":\n return [\"*\"]\n else:\n # Parse each ETag individually, and return any that are valid.\n etag_matches = (ETAG_MATCH.match(etag.strip()) for etag in etag_str.split(\",\"))\n return [match[1] for match in etag_matches if match]","def twitter_text(\n self,\n text: str,\n urls: List[Dict[str, str]],\n user_mentions: List[Dict[str, Any]],\n media: List[Dict[str, Any]],\n ) -> Element:\n\n to_process: List[Dict[str, Any]] = []\n # Build dicts for URLs\n for url_data in urls:\n to_process.extend(\n {\n \"type\": \"url\",\n \"start\": match.start(),\n \"end\": match.end(),\n \"url\": url_data[\"url\"],\n \"text\": url_data[\"expanded_url\"],\n }\n for match in re.finditer(re.escape(url_data[\"url\"]), text, re.IGNORECASE)\n )\n # Build dicts for mentions\n for user_mention in user_mentions:\n screen_name = user_mention[\"screen_name\"]\n mention_string = \"@\" + screen_name\n to_process.extend(\n {\n \"type\": \"mention\",\n \"start\": match.start(),\n \"end\": match.end(),\n \"url\": \"https://twitter.com/\" + urllib.parse.quote(screen_name),\n \"text\": mention_string,\n }\n for match in re.finditer(re.escape(mention_string), text, re.IGNORECASE)\n )\n # Build dicts for media\n for media_item in media:\n short_url = media_item[\"url\"]\n expanded_url = media_item[\"expanded_url\"]\n to_process.extend(\n {\n \"type\": \"media\",\n \"start\": match.start(),\n \"end\": match.end(),\n \"url\": short_url,\n \"text\": expanded_url,\n }\n for match in re.finditer(re.escape(short_url), text, re.IGNORECASE)\n )\n # Build dicts for emojis\n for match in POSSIBLE_EMOJI_RE.finditer(text):\n orig_syntax = match.group(\"syntax\")\n codepoint = emoji_to_hex_codepoint(unqualify_emoji(orig_syntax))\n if codepoint in codepoint_to_name:\n display_string = \":\" + codepoint_to_name[codepoint] + \":\"\n to_process.append(\n {\n \"type\": \"emoji\",\n \"start\": match.start(),\n \"end\": match.end(),\n \"codepoint\": codepoint,\n \"title\": display_string,\n }\n )\n\n to_process.sort(key=lambda x: x[\"start\"])\n p = current_node = Element(\"p\")\n\n def set_text(text: str) -> None:\n \"\"\"\n Helper to set the text or the tail of the current_node\n \"\"\"\n if current_node == p:\n current_node.text = text\n else:\n current_node.tail = text\n\n db_data: Optional[DbData] = self.zmd.zulip_db_data\n current_index = 0\n for item in to_process:\n # The text we want to link starts in already linked text skip it\n if item[\"start\"] < current_index:\n continue\n # Add text from the end of last link to the start of the current\n # link\n set_text(text[current_index : item[\"start\"]])\n current_index = item[\"end\"]\n if item[\"type\"] != \"emoji\":\n elem = url_to_a(db_data, item[\"url\"], item[\"text\"])\n assert isinstance(elem, Element)\n else:\n elem = make_emoji(item[\"codepoint\"], item[\"title\"])\n current_node = elem\n p.append(elem)\n\n # Add any unused text\n set_text(text[current_index:])\n return p","def _parse_mercurial_author(data, id_gen):\n angled = ur'(?P<author>.+?) <(?P<email>.+?)>'\n paren = ur'(?P<email>.+?) \\((?P<author>.+?)\\)'\n simple = ur'(?P<author>[^,]+)'\n author_list = []\n for regex in (angled, paren, simple):\n # Watch out for commas separating multiple names.\n regex += u'(,\\s*)?'\n for match in re.finditer(regex, data):\n # Watch out for suffixes like 'Jr.' when they are comma-separated\n # from the name and thus cause issues when *all* names are only\n # separated by commas.\n match_dict = match.groupdict()\n author = match_dict['author']\n if not author.partition(' ')[1] and author.endswith('.'):\n prev_author = author_list.pop()\n author = ', '.join([prev_author, author])\n if u'email' not in match_dict:\n email = ''\n else:\n email = match_dict['email']\n author_list.append((author, email))\n else:\n # If authors were found then stop searching as only expect one\n # style of author citation.\n if author_list:\n break\n author = Author(author_list[0])\n user = author.first_last\n email = author.email\n uid = id_gen[user]\n return (uid, user, email)","def extract_hashtag(text):\n if text is not None:\n text = text.replace('\\n', ' ').replace('\\r', '')\n text = text.split(\" \")\n text = [word for word in text if \"#\" in word]\n if len(text) == 0:\n text = [\"no tags\"]\n else:\n text = [\"no tags\"]\n return text","async def tag_sanskrit(sanskrit_string: str):\n result = search_utils.tag_sanskrit(sanskrit_string).replace(\"\\n\",\" # \")\n return {\"tagged\": result}","def parse_slack_output(slack_rtm_output):\r\n output_list = slack_rtm_output\r\n if output_list and len(output_list) > 0:\r\n for output in output_list:\r\n if output and 'text' in output and output['text'].startswith(AT_BOT):\r\n # return text after the @ mention, whitespace removed\r\n return output['text'].split(AT_BOT)[1].strip().lower(), \\\r\n output['channel'], output['user']\r\n return None, None, None","def detag_string(self, string):\r\n counter = itertools.count(0)\r\n count = lambda m: '<%s>' % counter.next()\r\n tags = self.tag_pattern.findall(string)\r\n tags = [''.join(tag) for tag in tags]\r\n (new, nfound) = self.tag_pattern.subn(count, string)\r\n if len(tags) != nfound:\r\n raise Exception('tags dont match:' + string)\r\n return (new, tags)","def author_name(text):\n tag = text.split()\n\n \"\"\"\n We take the beginning of the text since the\n author name will likely be there\n \"\"\"\n\n tag = tag[:100]\n author = []\n\n current_tag = 0\n \"\"\"\n We go through each word until we find the first instance\n of the word 'by' or 'author', which should mean the author\n will be written right after that.\n We save the first word after 'by' or 'author' since it should\n be the authors first name\n \"\"\"\n\n for word in tag:\n if (word.lower() == ('by') or\n word.lower() == ('author') or\n word.lower() == ('author:')):\n\n author.append(tag[current_tag+1].decode(encoding='UTF8',\n errors='ignore'))\n current_tag += 1\n tag = tag[current_tag+1:]\n break\n current_tag += 1\n\n \"\"\"\n We go through each word after the first name of the author\n until we find a word that is not capitalized. We assume that\n it marks the end of the author name.\n We then return a list of the author's name split up.\n \"\"\"\n current_tag = 0\n for word in tag:\n if tag[current_tag].lower() == 'this':\n break\n if tag[current_tag].istitle():\n author.append(tag[current_tag].decode(encoding='UTF8',\n errors='ignore'))\n current_tag += 1\n\n return author","def read(string):\n\treturn (re.finditer('(?<=\\[)[a-z]+(?=\\])', string), re.finditer('(?<=\\])[a-z]+|[a-z]+(?=\\[)', string))","def parse_tag_input(input):\r\n if not input:\r\n return []\r\n\r\n input = force_unicode(input)\r\n\r\n # Special case - if there are no commas or double quotes in the\r\n # input, we don't *do* a recall... I mean, we know we only need to\r\n # split on spaces.\r\n if u',' not in input and u'\"' not in input:\r\n words = list(set(split_strip(input, u' ')))\r\n words.sort()\r\n return words\r\n\r\n words = []\r\n buffer = []\r\n # Defer splitting of non-quoted sections until we know if there are\r\n # any unquoted commas.\r\n to_be_split = []\r\n saw_loose_comma = False\r\n open_quote = False\r\n i = iter(input)\r\n try:\r\n while 1:\r\n c = i.next()\r\n if c == u'\"':\r\n if buffer:\r\n to_be_split.append(u''.join(buffer))\r\n buffer = []\r\n # Find the matching quote\r\n open_quote = True\r\n c = i.next()\r\n while c != u'\"':\r\n buffer.append(c)\r\n c = i.next()\r\n if buffer:\r\n word = u''.join(buffer).strip()\r\n if word:\r\n words.append(word)\r\n buffer = []\r\n open_quote = False\r\n else:\r\n if not saw_loose_comma and c == u',':\r\n saw_loose_comma = True\r\n buffer.append(c)\r\n except StopIteration:\r\n # If we were parsing an open quote which was never closed treat\r\n # the buffer as unquoted.\r\n if buffer:\r\n if open_quote and u',' in buffer:\r\n saw_loose_comma = True\r\n to_be_split.append(u''.join(buffer))\r\n if to_be_split:\r\n if saw_loose_comma:\r\n delimiter = u','\r\n else:\r\n delimiter = u' '\r\n for chunk in to_be_split:\r\n words.extend(split_strip(chunk, delimiter))\r\n words = list(set(words))\r\n words.sort()\r\n return words","def parse_slack_output(slack_rtm_output):\n output_list = slack_rtm_output\n if output_list and len(output_list) > 0:\n print 'output list is: ', output_list\n for output in output_list:\n \n if (output and 'text' in output and AT_BOT in output['text'] \n and output['channel'].startswith('D')\n ):\n # return text after the @ mention, whitespace removed\n return output['text'].split(AT_BOT)[1].strip().lower(), \\\n output['channel'], output['user']\n\n elif (output and 'text' in output and output['channel'].startswith('D')\n and output['user'] != BOT_ID):\n return output['text'], output['channel'], output['user']\n\n return None, None, None","def get_mentions_from_body_and_issue_comments(pull_request_body, pull_request_comments):\n unique_mentions = set()\n unique_mentions.update(github_mention_regex.findall(pull_request_body))\n for comment in pull_request_comments:\n unique_mentions.update(github_mention_regex.findall(comment.body))\n\n return unique_mentions","def pos_tag(self,sentence):\n tagged = self.brill_tagger.tag(sentence.split())\n tagged_sentence = \" \".join([nltk.tag.tuple2str(tok) for tok in tagged])\n print tagged_sentence\n\n tag_list = [(each.split(\"/\")[0],each.split(\"/\")[1]) for each in tagged_sentence.split()]\n return tag_list","def parse(self, string, root=None):\n\n\t\tphrases = []\n\n\t\tmeta = self.meta.search(string)\n\n\t\twhile meta:\n\n\t\t\t# Save some function calls\n\t\t\tpos = meta.start()\n\n\t\t\tif meta.group() == \"<\":\n\t\t\t\tstring, child, meta = self.open_phrase(string, pos)\n\n\t\t\t\tif child and root:\n\t\t\t\t\troot.nested.append(child)\n\t\t\t\telif child:\n\t\t\t\t\tphrases.append(child)\n\n\t\t\t\t# else it was escaped (+ new meta)\n\t\t\t\tcontinue\n\n\t\t\telif root:\n\n\t\t\t\tif meta.group() == \"(\":\n\t\t\t\t\tmeta = self.meta.search(string, pos + 1)\n\t\t\t\t\tif meta.group() == \")\":\n\t\t\t\t\t\tstring, root, meta = self.handle_arguments(string,\n\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t root,\n\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t pos,\n\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t meta.start())\n\t\t\t\t\t\tcontinue\n\n\t\t\t\telif meta.group() == \">\":\n\t\t\t\t\tstring, phrase, meta = self.close_phrase(string,\n\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t root,\n\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t pos)\n\t\t\t\t\tif phrase:\n\t\t\t\t\t\treturn string, phrase\n\n\t\t\t\t\t# else was escaped (+ new meta)\n\t\t\t\t\tcontinue\n\n\t\t\tstring, meta = self.escape_meta(string, pos)\n\n\t\tif not root:\n\t\t\treturn string, phrases\n\n\t\t# If this is not the first stack-depth the function should\n\t\t# have returned upon finding a closing tag,\n\t\t# i.e. we should never have gotten here.\n\t\tword = re.search(r\"([\\w\\s]+)(?![\\d]*>[\\w\\s]+>)\", string)\n\n\t\twhat = \"No closing tag found for opening tag\"\n\n\t\tif word:\n\t\t\twhat += \" after expression '{0}'\".format(word.group())\n\n\t\traise errors.ParseError(what + \"!\")","def extract_mentions_or_lists_with_indices(self, transform = lambda x: x):\r\n if not REGEXEN['at_signs'].search(self.text):\r\n return []\r\n\r\n possible_entries = []\r\n for match in REGEXEN['valid_mention_or_list'].finditer(self.text):\r\n try:\r\n after = self.text[match.end()]\r\n except IndexError:\r\n # the mention was the last character in the string\r\n after = None\r\n if after and REGEXEN['end_mention_match'].match(after) or match.groups()[2].find('http') == 0:\r\n continue\r\n possible_entries.append({\r\n 'screen_name': transform(match.groups()[2]),\r\n 'list_slug': match.groups()[3] or '',\r\n 'indices': [match.start() + len(match.groups()[0]), match.end()]\r\n })\r\n\r\n return possible_entries","def get_mentions(texts, word):\r\n mentions = {}\r\n for text, label, text_id in texts:\r\n if word in text.lower():\r\n if label not in mentions:\r\n mentions[label] = [text_id]\r\n else:\r\n if text_id not in mentions[label]:\r\n mentions[label].append(text_id)\r\n return mentions","def parse_tags(tags: str = None) -> Iterable[str]:\n if not tags:\n return []\n return [x.strip() for x in tags.split(\",\")]","def _find_subtype_mentions(self, message):\n subtype_mention_regex = '<@[a-zA-Z0-9]+|'\n user_mentions = re.findall(subtype_mention_regex, message['text'])\n user_ids = [mention.strip('@<>') for mention in user_mentions]\n return [user_id for user_id in user_ids if user_id != self.poolbot.bot_id]","def parse(self, text):\n assert isinstance(text, str)\n if text.strip() == '':\n return [], []\n\n output = self._annotate(text, properties={\n \"annotators\": \"tokenize,ssplit,pos\",\n \"coref.md.type\": \"dep\",\n \"coref.mode\": \"statistical\"\n })\n\n words = []\n postags = []\n\n for sentence in output['sentences']:\n for token in sentence['tokens']:\n word = token['word']\n pos = token['pos']\n word = re.sub(r'\\s', '', word)\n words.append(word)\n postags.append(pos)\n return words, postags","def parse_atwho(my_text):\n notification_list = []\n \n #markdown_link_regex = re.compile(\"\\[.*?\\] \\((.*?) \\\".*?\\\"\\)\", re.IGNORECASE) # need to test this.\n markdown_link_regex = re.compile(\"\\[([^\\]]+)\\]\\(([^)\\\"]+)(?: \\\\\\\"([^\\\\\\\"]+)\\\\\\\")?\\)\", re.IGNORECASE)\n direct_link_regex = re.compile(\"data-notify=\\\\\\\"([^\\\\\\\"]+)\\\\\\\"\", re.IGNORECASE)\n link_list = [i[2] for i in markdown_link_regex.findall(my_text)]\n link_list += [i for i in direct_link_regex.findall(my_text)]\n\n for i in link_list:\n _type, _id = i.split(\"#\", 1)\n if _type == \"user\":\n notification_list.append(User.objects.get(id=_id))\n elif _type == \"subdept\":\n notification_list.append(Subdept.objects.get(id=_id))\n elif _type == \"dept\":\n notification_list.append(Dept.objects.get(id=_id))\n elif _type == \"page\":\n notification_list.append(Page.objects.get(id=_id))\n\n return my_text, notification_list","def completion(s):\n if len(s) >= 1 and s[0] == 'h':\n return ('hello', 'hello there')\n return None","def resolve_function(cls, phrase_mention: List[Tuple[Tuple[str, str], str, str, Tuple[str, str]]]) -> List[str]:\n\n # Iterate over each component in `phrase_mention`, and extract possible user groups if the first POS tag is a\n # verb\n res = [cls.get_user_group(*str_split) for pos, _, _, str_split in phrase_mention if \"verb\" in pos[0]]\n\n # Return non-empty string user groups\n return [r for r in res if r != \"\"]","def split_hashtags(hashtag):\n matches = re.findall('.+?(?:(?<=[a-z])(?=[A-Z])|(?<=[A-Z])(?=[A-Z][a-z])|$)', hashtag)\n return matches","def _pairs(content: str) -> Generator[Tuple[Any, Any], None, None]:\n if not content.strip():\n return\n\n for pair in content.split(\",\"):\n\n x, *y = pair.split(\":\", maxsplit=1)\n if len(y) == 0:\n raise ValueError(f\"Cannot parse pair {pair}\")\n\n yield x.strip(), y[0].strip()","def parse_slack_output(slack_rtm_output, bot):\n output_list = slack_rtm_output\n if output_list and len(output_list) > 0:\n for output in output_list:\n at_bot = \"<@%s>\" % bot['bot_id']\n if output and 'text' in output and at_bot in output['text'] and 'channel' in output:\n # return text after the @ mention, whitespace removed\n return output['text'].split(at_bot)[1].strip(), \\\n output['channel']\n return None, None","def linkify_tags_and_mentions(value):\n value = find_hashtags_re.sub(tag_match_to_url, sanitize(value))\n value = find_mentions_re.sub(mention_match_to_url, value)\n # value = link_tags_parse(value)\n return mark_safe(value)","def get_tags_for_tags_and_malware_family_fields(tags: Optional[list], is_malware_family=False):\n if not tags:\n return None\n results = []\n for item in tags:\n results.append(item.get('tag_name'))\n results.append(item.get('public_tag_name'))\n for alias in item.get('aliases', []):\n results.append(alias)\n if not is_malware_family:\n for group in item.get('tagGroups', [{}]):\n results.append(group.get('tag_group_name'))\n # Returns a list without duplicates and empty elements\n return list(set(filter(None, results)))","def get_author(html_soup):\n auth_text = html_soup.find('div', attrs = {\"class\" : \"author\"}).text.split(\"|\")\n for i in auth_text:\n if 'By' in i:\n auth_text_split = i.split()\n auth_text_split = auth_text_split[auth_text_split.index('By')+1:auth_text_split.index('By')+3]\n auth_text_split = ' '.join(auth_text_split)\n return(auth_text_split)","def _parse_tags (tag, multi_kind_dataset ='bagoue'): \r\n tag = str(tag); t = tag.strip().split() \r\n \r\n if len(t) ==1 : \r\n if t[0].lower() not in _DTAGS: \r\n tag = multi_kind_dataset +' ' + t[0]\r\n \r\n warn(f\"Fetching {multi_kind_dataset.title()!r} data without\"\r\n \" explicitly prefixing the kind of data with the area\"\r\n \" name will raise an error. In future, the argument\"\r\n f\" should be '{tag}' instead.\", FutureWarning \r\n )\r\n elif len(t) >1 : \r\n # only the multi kind dataset is allowed \r\n # to contain two words for fetching data \r\n if t[0].lower() !=multi_kind_dataset: \r\n tag = t[0].lower() # skip the second word \r\n return tag","def hashtag_seperator(s: List[tuple]) -> List[str]:\n result = []\n for tups in s:\n text = tups[0]\n result = result + extract_hashtags(text)\n return result","def _extract_tags(html):\n tags = re.findall(r'<[^>]+>', html)\n\n return tags","def ig_users_in(body):\n import re\n\n try:\n user_re = Formatter.USER_TAG_REGEX\n except AttributeError:\n # escape special characters so we can format the user-tag into a\n # regex pattern\n escaped_user_tag = re.sub(\n # match any '[', ']', '(', or ')'\n r'([\\[\\]\\(\\)])',\n # escape the matched character\n r'\\\\\\1',\n Formatter.USER_TAG\n )\n pattern = escaped_user_tag.format(\n user_raw='({0})'.format(instagram.USERNAME_PTN)\n )\n user_re = re.compile(pattern)\n Formatter.USER_TAG_REGEX = user_re\n\n return user_re.findall(body)","def bio_tag_to_entity_mentions(bio_tags):\n within_entity_mention = False\n mentions = []\n for i, tag in enumerate(bio_tags):\n if not within_entity_mention:\n # start of entity mention\n if 'B-' in tag:\n start = i\n within_entity_mention = True\n else:\n if tag == 'O':\n end = i\n mentions.append({\n 'start' : start,\n 'end' : end,\n 'labels' : []\n })\n within_entity_mention = False\n # two entities side by side without O tag\n if 'B-' in tag:\n end = i\n mentions.append({\n 'start' : start,\n 'end' : end,\n 'labels' : []\n })\n start = i\n if within_entity_mention:\n end = len(bio_tags)\n mentions.append({\n 'start' : start,\n 'end' : end,\n 'labels' : []\n })\n return mentions","def get_speakers(words):\n speakers = []\n utterances = \" \".join(words).split(\"<|endoftext|>\")[:-1]\n for u in utterances:\n s = u.split(\":\")[0].strip() # select names according to \":\"\n if s:\n speakers.append(s)\n return list(set(speakers))","def gen_sent_coresponding_tags(self, sentence: str,\n tagged_tokens: TaggedTokens,\n unigrams: Unigrams,\n tag_sets: TagSets,\n syl_stats: List = None) -> Tuple:\n\n sentence_tokens = sentence.split()\n sentence_tags = tagged_tokens.get_sentence_tags(sentence)\n new_sentence = []\n new_sentence_meta = []\n syl_stats = syl_stats or [None] * len(sentence_tokens)\n\n for index, (tag, source) in enumerate(sentence_tags):\n n_syl = syl_stats[index]\n\n try:\n new_token = self.rand_token_tag_unigrams(tag, unigrams,\n tag_sets,\n n_syllables=n_syl)\n\n new_sentence.append(new_token)\n new_sentence_meta.append(source)\n\n except (TagDoesNotExist, NoContinuation):\n new_sentence.append(sentence_tokens[index])\n new_sentence_meta.append('not_replaced')\n\n return new_sentence, new_sentence_meta","def parse_slack_output(slack_rtm_output):\n output_list = slack_rtm_output\n if output_list and len(output_list) > 0:\n for output in output_list:\n if output and 'text' in output and AT_BOT in output['text']:\n # return text after the @ mention, whitespace removed\n return output['text'].split(AT_BOT)[1].strip(), \\\n output['channel']\n return None, None","def parse_tweet(line):\n # The following regex just strips of an URL (not just http), any punctuations,\n # or Any non alphanumeric characters\n # http://goo.gl/J8ZxDT\n text = re.sub(\"(@[A-Za-z0-9]+)|([^0-9A-Za-z \\t])|(\\w+:\\/\\/\\S+)\",\" \",json.loads(line[1])[\"value\"]).strip()\n # remove terms <= 2 characters\n text = ' '.join(filter(lambda x: len(x) > 2, text.split(\" \")))\n\n return (line[0], text)","def extract_entities(event):\n # TODO The text should probably already be tagged and tokenized before this step\n tree = ne_chunk(event.pos_tagged)\n entities = set([])\n\n people = tree.subtrees(lambda x: x.node == \"PERSON\")\n for person in people:\n entities.add(\" \".join([leaf[0] for leaf in person.leaves()]))\n\n places = tree.subtrees(lambda x: x.node == \"GPE\")\n for place in places:\n entities.add(\" \".join([leaf[0] for leaf in place.leaves()]))\n\n organizations = tree.subtrees(lambda x: x.node == \"ORGANIZATION\")\n for org in organizations:\n entities.add(\" \".join([leaf[0] for leaf in org.leaves()]))\n \n return entities","def extract_metadata(info_str):\n info = [i.strip() for i in info_str.split(';')\n if i.strip().startswith('gene_id')\n or i.strip().startswith('gene_type')]\n assert len(info) == 2, '{0}'.format(info)\n gene_id, gene_type = extract_gene_data(info)\n return gene_id, gene_type","def test_returns_nouns_for_string(self):\r\n test_value = 'google drives autonomous cars'\r\n self.assertEqual(\r\n set([u'cars', u'autonomous']),\r\n suggest_tags(test_value))","def splitTags(user_input):\n \n elements = []\n if ',' in user_input:\n elements = user_input.split(',')\n elif ' ' in user_input:\n elements = user_input.split(' ')\n else:\n elements.append(user_input)\n\n tags = []\n for element in elements:\n element = element.strip(' \\t\\n\\r').lower()\n if(len(element) == 0): continue\n if element not in tags:\n tags.append(element)\n return tags","def findall(self, string: str) -> List[str]:\r\n \r\n string = list(string)\r\n matches = []\r\n\r\n while len(string) > 0:\r\n tokens = self.tokens.copy()\r\n match = []\r\n\r\n while len(tokens) > 0:\r\n token = _advance(tokens)\r\n char = _advance(string)\r\n\r\n if token is None:\r\n break\r\n if char is None and token.type != TokenType.OPTIONAL:\r\n match = []\r\n break\r\n\r\n value = token.match(tokens, string, char)\r\n if value is None:\r\n match = []\r\n break\r\n match.extend(value)\r\n \r\n if len(match) > 0:\r\n matches.append(''.join(match))\r\n\r\n return matches","def parse(string):\n doc = nlp(string)\n return [str(n) for n in doc.noun_chunks]","def preprocess_tweet(tweet):\n\n\n clean_tweet, hashtags = separate_hastags_mentions_urls(tweet)\n clean_tweet = remove_emoji_punc(clean_tweet)\n return clean_tweet, hashtags","def parseMuttAddressbook(addressbook):\n\n alias1 = re.compile('(^alias\\s*)(\\S*\\s*)(\".*\"\\s*)(<)(.*)(>$)')\n alias2 = re.compile(\"(^alias\\s*)(\\S*\\s*)(\\S*)(\\s*\\(.*$)\")\n\n addresses = set()\n for line in open(addressbook).readlines():\n alias1match = alias1.match(line)\n alias2match = alias2.match(line)\n if alias1match is not None:\n address = alias1.match(line).group(5)\n addresses.add(address)\n elif alias2match is not None:\n address = alias2.match(line).group(3)\n addresses.add(address)\n\n return addresses","def detect_author(user_to_tweets: Dict[str, List[tuple]], tweet_text: str) -> \\\n str:\n acc = []\n \n for keys in user_to_tweets:\n author_hashes = hashtag_seperator(user_to_tweets[keys])\n text_hashes = extract_hashtags(tweet_text)\n if set(text_hashes).issubset(author_hashes):\n acc.append(keys)\n if len(acc) == 1:\n return acc[0]\n return 'unknown'","def _collect_user_data(utt_sets: Collection[Collection[Utterance]]) -> Tuple[Dict[str, Dict[Hashable, str]], Dict[str, Dict[Hashable, bool]]]:\n # Collect USER data and metadata\n # all_users_data = defaultdict(lambda: defaultdict(set))\n all_users_meta = defaultdict(lambda: defaultdict(str))\n all_users_meta_conflict = defaultdict(lambda: defaultdict(bool))\n for utt_set in utt_sets:\n for utt in utt_set:\n for meta_key, meta_val in utt.user.meta.items():\n curr = all_users_meta[utt.user][meta_key]\n if curr != meta_val:\n if curr != \"\":\n all_users_meta_conflict[utt.user][meta_key] = True\n all_users_meta[utt.user][meta_key] = meta_val\n\n return all_users_meta, all_users_meta_conflict","def extract_mentions(text: str) -> List[str]:\n final = []\n result = list(text.split())\n \n for word in result: \n if word[0] == MENTION_SYMBOL:\n final.append(alnum_prefix(word[1:]))\n if alnum_prefix(word[1:]) == '':\n final.remove(alnum_prefix(word[1:]))\n return final","def parse_slack_output(slack_rtm_output):\n output_list = slack_rtm_output\n if output_list and len(output_list) > 0:\n for output in output_list:\n if output and 'text' in output and AT_BOT in output['text']:\n # return text after the @ mention, whitespace removed\n return output['text'].split(AT_BOT)[1].strip().lower(), \\\n output['channel']\n return None, None","def parse(s, tokenize=True, tags=True, chunks=True, relations=False, lemmata=False, encoding=\"utf-8\", **kwargs):\n if tokenize:\n s = _tokenize(s)\n if isinstance(s, (list, tuple)):\n s = [isinstance(s, basestring) and s.split(\" \") or s for s in s]\n if isinstance(s, basestring):\n s = [s.split(\" \") for s in s.split(\"\\n\")]\n # Reuse the English parser:\n kwargs.update({\n \"lemmata\": False,\n \"light\": False,\n \"lexicon\": LEXICON,\n \"language\": \"es\",\n \"default\": \"NC\",\n \"map\": kwargs.get(\"tagset\", \"\") != PAROLE and parole2penntreebank or None,\n })\n s = _en_parse(s, False, tags, chunks, relations, **kwargs)\n # Use pattern.es.inflect for lemmatization:\n if lemmata:\n p = [find_lemmata(sentence) for sentence in s.split()]\n p = \"\\n\".join([\" \".join([\"/\".join(token) for token in sentence]) for sentence in p])\n s = TaggedString(p, tags=s.tags+[\"lemma\"], language=\"es\")\n return s","def speaking_pairs(self, selector: Optional[Callable[[User, User], bool]]=None,\n user_names_only: bool=False) -> Set[Tuple]:\n pairs = set()\n for u2 in self.utterances.values():\n if u2.user is not None and u2.reply_to is not None and u2.reply_to in self.utterances:\n u1 = self.utterances[u2.reply_to]\n if u1.user is not None:\n if selector is None or selector(u2.user, u1.user):\n pairs.add((u2.user.name, u1.user.name) if\n user_names_only else (u2.user, u1.user))\n return pairs","def _parse_tags(self):\n tokens = self.tags_str[1:].split(\";\")\n self._tags = {\n k.strip(): v\n for token in tokens\n for k, v in [token.split(\"=\")]\n }","def tags_in_string(msg):\r\n def is_linguistic_tag(tag):\r\n \"\"\"Is this tag one that can change with the language?\"\"\"\r\n if tag.startswith(\"&\"):\r\n return True\r\n if any(x in tag for x in [\"<abbr>\", \"<abbr \", \"</abbr>\"]):\r\n return True\r\n return False\r\n\r\n __, tags = Converter().detag_string(msg)\r\n return set(t for t in tags if not is_linguistic_tag(t))","def tokens(text) -> Tuple[Word]:\n return tuple(re.findall('[a-z]+', text.lower()))","def tags_from_csv_field(tag_string):\n split_string = tag_string.split()\n out_list = []\n for tag in split_string:\n out_list.append(clean_tag(tag))\n\n return out_list","def parse_slack_output(ARGS, slack_rtm_output):\n output_list = slack_rtm_output\n if output_list and len(output_list) > 0:\n for output in output_list:\n if output and 'text' in output and ARGS.atbot in output['text']:\n # return text after the @ mention, whitespace removed\n return output['text'].split(ARGS.atbot)[1].strip().lower(), \\\n output['channel']\n return None, None","def get_user_by_email(tx: Transaction, user_email: str) -> BoltStatementResult:\n\n # NOTE: tag labels are hardcoded here. If they change in tags csv, must be changed here.\n query = f\"\"\"\n MATCH (user:Person {{email: '{user_email}'}})\n OPTIONAL MATCH (user)-->(skill_tag:Tag:CanHelpWithTag)\n OPTIONAL MATCH (user)-->(passion_tag:Tag:PassionsTag)\n RETURN user, COLLECT(DISTINCT skill_tag.name) AS help_others, COLLECT(DISTINCT passion_tag.name) AS passions\"\"\"\n return tx.run(query)","def parse_person(person):\n parsed = person_parser(person)\n if not parsed:\n parsed = person_parser_only_name(person)\n name = parsed.group('name')\n email = None\n else:\n name = parsed.group('name')\n email = parsed.group('email')\n\n return name, email","def get_tags(x):\n poster = x['_source']['user']['id']\n for user in x['_source']['users_in_photo']:\n if user['user']['id'] != poster:\n yield (poster, user['user']['id'])"],"string":"[\n \"def get_tagged_user(line, unique_users):\\n tagged_user = None\\n\\n for user in unique_users:\\n \\n tagged_user = re.search(f\\\"@{user}\\\\s*\\\", line)\\n \\n if tagged_user != None:\\n tagged_user = tagged_user.group(0).replace(\\\"@\\\", \\\"\\\").strip()\\n line = line.replace(f\\\"@{user} \\\", \\\"\\\")\\n break\\n \\n return (tagged_user, line)\",\n \"def tags(string):\\n\\treturn set([x.strip() for x in string.split(',')])\",\n \"def _parse_text_as_direct_mention(\\n message_text: str,\\n ) -> tuple[Optional[str], Optional[str]]:\\n matches = re.search(MENTION_REGEX, message_text)\\n # the first group contains the username, the second group contains the remaining message\\n if matches:\\n return matches.group(1), matches.group(2).strip()\\n return None, None\",\n \"def findTags(user_input, tagged_text):\\n result = []\\n for item in tagged_text:\\n for w in user_input:\\n if w[WORD] == item[WORD]:\\n tup = (w[WORD], item[TAG])\\n result.append(tup)\\n continue\\n\\n return result\",\n \"def parse_direct_mention(message_text):\\n matches = re.search(MENTION_REGEX, message_text)\\n print (matches)\\n # the first group contains the username, the second group contains the remaining message\\n return (matches.group(1), matches.group(2).strip()) if matches else (None, None)\",\n \"def parse_direct_mention(message_text):\\n # print(message_text)\\n matches = re.search(MENTION_REGEX, message_text)\\n # the first group contains the username, the second group contains the remaining message\\n return (matches.group(1), matches.group(2).strip()) if matches else (None, None)\",\n \"def parse_direct_mention(message_text):\\r\\n matches = re.search(MENTION_REGEX, message_text)\\r\\n # the first group contains the username, the second group contains the remaining message\\r\\n return (matches.group(1), matches.group(2).strip()) if matches else (None, None)\",\n \"def suggest_tags(data):\\r\\n tag_set = set()\\r\\n if not data:\\r\\n return tag_set\\r\\n\\r\\n # The string might be a url that needs some cleanup before we parse for\\r\\n # suggestions\\r\\n parsed = urlparse(data)\\r\\n\\r\\n # Check if title is url. If title is not a string, url, and title will be\\r\\n # the same so no need to consider tags from url.\\r\\n if parsed.hostname:\\r\\n tag_set.update(_generate_nouns_from_url(parsed.path))\\r\\n else:\\r\\n # If the title is not a url extract nouns from title and the url.\\r\\n tag_set.update(_generate_nouns_from_url(data))\\r\\n\\r\\n return tag_set\",\n \"def parse_direct_mention(message_text):\\n matches = re.search(MENTION_REGEX, message_text)\\n # the first group contains the username, the second group contains the remaining message\\n return (matches.group(1), matches.group(2).strip()) if matches else (None, None)\",\n \"def parse_direct_mention(message_text):\\n matches = re.search(MENTION_REGEX, message_text)\\n # the first group contains the username, the second group contains the remaining message\\n return (matches.group(1), matches.group(2).strip()) if matches else (None, None)\",\n \"def parse_direct_mention(message_text):\\n matches = re.search(MENTION_REGEX, message_text)\\n # the first group contains the username, the second group contains the remaining message\\n return (matches.group(1), matches.group(2).strip()) if matches else (None, None)\",\n \"def parse_direct_mention(message_text):\\n matches = re.search(MENTION_REGEX, message_text)\\n # the first group contains the username, the second group contains the remaining message\\n return (matches.group(1), matches.group(2).strip()) if matches else (None, None)\",\n \"def parse_direct_mention(message_text):\\n matches = re.search(MENTION_REGEX, message_text)\\n # the first group contains the username, the second group contains the remaining message\\n return (matches.group(1), matches.group(2).strip()) if matches else (None, None)\",\n \"def parse_direct_mention(message_text):\\n matches = re.search(MENTION_REGEX, message_text)\\n # 1st gourp has username, 2nd group called remaining message\\n return (matches.group(1), matches.group(2).strip()) if matches else (None, None)\",\n \"def parse_direct_mention(message_text):\\n matches = re.search(MENTION_REGEX, message_text)\\n # the first group contains the username, the second group contains\\n # the remaining message\\n return (matches.group(1),\\n matches.group(2).strip()) if matches else (None, None)\",\n \"def parse_direct_mention(message_text):\\n matches = re.search(MENTION_REGEX, message_text)\\n # the first group contains the username, the second group\\n # contains the remaining message\\n return (matches.group(1),\\n matches.group(2).strip()) if matches else (None, None)\",\n \"def getTagList(tags):\\n tags = tags[1:len(tags)-1]\\n return tags.split('><')\",\n \"def parse_direct_mention(self, message_text):\\n matches = re.search(self.MENTION_REGEX, message_text)\\n # the first group contains the username, the second group contains the remaining message\\n return (matches.group(1), matches.group(2).strip()) if matches else (None, None)\",\n \"def parse_direct_mention(self, message_text):\\n matches = re.search(self.MENTION_REGEX, message_text)\\n # the first group contains the username, the second group contains the remaining message\\n return (matches.group(1), matches.group(2).strip()) if matches else (None, None)\",\n \"def parse_direct_mention(self,message_text):\\n matches = re.search(self.MENTION_REGEX, message_text)\\n # the first group contains the username, the second group contains the remaining message\\n return (matches.group(1), matches.group(2).strip()) if matches else (None, None)\",\n \"def parse_direct_mention(message_text):\\n\\tmatches = re.search(MENTION_REGEX, message_text)\\n\\t# the first group contains the username, the second group contains the remaining message\\n\\treturn (matches.group(1), matches.group(2).strip()) if matches else (None, None)\",\n \"def parse_direct_mention(message_text):\\n matches = re.search(MENTION_REGEX, message_text)\\n # the first group contains the username, the second group contains the remaining message\\n return (matches.group(1), matches.group(2).strip()) if matches else (None, message_text.strip())\",\n \"def parseFriends(line):\\n #parts = re.split('\\\\t', line)\\n parts = line.split('\\\\t')\\n #parts[1] = re.split(',', parts[1])\\n parts[1] = set(parts[1].split(','))\\n return parts[0], parts[1]\",\n \"def parse_tags(tagstring):\\n if not tagstring:\\n return []\\n\\n tagstring = force_str(tagstring)\\n\\n words = []\\n buffer = []\\n # Defer splitting of non-quoted sections until we know if there are\\n # any unquoted commas.\\n to_be_split = []\\n i = iter(tagstring)\\n try:\\n while True:\\n c = six.next(i)\\n if c == '\\\"':\\n if buffer:\\n to_be_split.append(''.join(buffer))\\n buffer = []\\n c = six.next(i)\\n while c != '\\\"':\\n buffer.append(c)\\n c = six.next(i)\\n if buffer:\\n word = ''.join(buffer).strip()\\n if word:\\n words.append(word)\\n buffer = []\\n else:\\n buffer.append(c)\\n except StopIteration:\\n # If we were parsing an open quote which was never closed treat\\n # the buffer as unquoted.\\n if buffer:\\n to_be_split.append(''.join(buffer))\\n if to_be_split:\\n for chunk in to_be_split:\\n words.extend(split_strip(chunk, settings.TAGGIT_SELECTIZE['DELIMITER']))\\n words = list(set(words))\\n words.sort()\\n return words\",\n \"def _parse_tags(tags: str):\\n return dict(item.split(\\\":\\\") for item in shlex.split(tags)) # type: ignore\",\n \"def parse(text):\\n # Remove hashtag and trailing whitespaces\\n clean_text = reHASH.sub(\\\"\\\", text).strip()\\n # We discard those entries with URLs inside\\n if reURL.search(clean_text):\\n raise\\n # Also discard those that contain @user forms\\n if '@' in clean_text:\\n raise\\n parts = clean_text.split(':')\\n if len(parts) < 2:\\n raise\\n term = parts[0].strip().lower()\\n meaning = \\\":\\\".join(parts[1:]).strip()\\n return term, meaning\",\n \"def get_tags(html):\\n\\ttitle = re.findall('\\\"title\\\":\\\"(.*?)\\\",', html)[0]\\n\\ttitle = codecs.getdecoder(\\\"unicode_escape\\\")(title)[0]\\n\\n\\tartist = re.findall('\\\"username\\\":\\\"(.*?)\\\",', html)[0]\\n\\tartist = codecs.getdecoder(\\\"unicode_escape\\\")(artist)[0]\\n\\n\\tgenre = re.findall('\\\"genre\\\":\\\"(.*?)\\\",', html)[0]\\n\\tgenre = codecs.getdecoder(\\\"unicode_escape\\\")(genre)[0]\\n\\n\\treturn title, artist, genre\",\n \"def separate_hastags_mentions_urls(tweet):\\n \\n text = tweet.lower()\\n hashtag_list = re.findall(\\\"#([a-zA-Z0-9_]{1,50})\\\", text)\\n \\n text = re.sub(r'http\\\\S+', '', text)\\n clean_tweet = re.sub(\\\"@[A-Za-z0-9_]+\\\",\\\"\\\", text)\\n clean_tweet = re.sub(\\\"#[A-Za-z0-9_]+\\\",\\\"\\\", clean_tweet)\\n \\n return clean_tweet, hashtag_list\",\n \"def extract(msg):\\n\\n rgx = url_regex()\\n found = re.findall(rgx, msg)\\n links_set = Set()\\n for link in found:\\n # dunno if I like this. don't need seperate groups to make this work.\\n links_set.add(link[0].strip())\\n return list(links_set)\",\n \"def get_usernames_from_tag(tag):\\n try:\\n tag_url = 'https://www.instagram.com/explore/tags/{0}/?__a=1'.format(tag)\\n tag_res = requests.get(tag_url).json()\\n\\n nodes = tag_res['tag']['media']['nodes']\\n\\n new_urls = [node['code'] for node in nodes]\\n new_req = [requests.get('https://www.instagram.com/p/{0}/?__a=1'.format(url)).json() for url in new_urls]\\n\\n usernames = [req['graphql']['shortcode_media']['owner']['username'] for req in new_req]\\n\\n return usernames\\n\\n except Exception as err:\\n print('{0} - This account was not classified...'.format(str(err)))\\n return []\",\n \"def parse_tag(self, tag):\\n \\n mytag = \\\"latest\\\"\\n mydigest = None\\n\\n regex = \\\"([\\\\w\\\\d\\\\.\\\\-]+)@?([\\\\w\\\\d\\\\.\\\\-]*)$\\\"\\n\\n regex_matched = re.match(regex, tag)\\n mytag = regex_matched.group(1)\\n mydigest = regex_matched.group(2)\\n \\n if regex_matched is None:\\n mytag = \\\"latest\\\"\\n\\n return (mytag, mydigest)\",\n \"def parse_output(slack_rtm_output): # pragma: no cover\\n output_list = slack_rtm_output\\n if output_list and len(output_list) > 0:\\n for output in output_list:\\n if output and 'text' in output and AT_BOT in output['text']:\\n # return text after the @ mention, whitespace removed\\n return output['text'].split(AT_BOT)[1].strip().lower(), \\\\\\n output['channel'], output['user']\\n return None, None, None\",\n \"def getHashtagsAndMentions(tweets):\\n hashtags = Counter()\\n mentions = Counter()\\n plain = Counter()\\n\\n pattern = re.compile(r\\\"[^#@\\\\w'-]+\\\")\\n\\n for t in tweets:\\n words = pattern.split(t.message)\\n for word in words:\\n # Ignore null strings caused by split characters at the end of a\\n # message and remove standalone hyphens.\\n if word and not word.startswith(\\\"-\\\"):\\n # Increment count for the word in the Counter.\\n if word.startswith(\\\"#\\\"):\\n hashtags.update({word: 1})\\n elif word.startswith(\\\"@\\\"):\\n mentions.update({word: 1})\\n else:\\n # TODO: apply nltk.corpus.stopwords.words() here,\\n # across languages. Consider that the stopwords cut off\\n # before apostrophe, therefore check if the word\\n # starts with the stopword.\\n plain.update({word: 1})\\n\\n return hashtags, mentions, plain\",\n \"def tag(s, tokenize=True, encoding=\\\"utf-8\\\"):\\n tags = []\\n for sentence in parse(s, tokenize, True, False, False, False, encoding).split():\\n for token in sentence:\\n tags.append((token[0], token[1]))\\n return tags\",\n \"def parse_slack_output(slack_rtm_output):\\n output_list = slack_rtm_output\\n if output_list and len(output_list) > 0:\\n for output in output_list:\\n if output and 'text' in output and AT_BOT in output['text']:\\n # return text after the @ mention, whitespace removed\\n s = output['text'].split(AT_BOT)[1].strip().lower()\\n exclude = set(string.punctuation)\\n s = ''.join(ch for ch in s if ch not in exclude)\\n return output['channel'], s, output['user'],\\n return None, None, None\",\n \"def parse_tags(s: str) -> List[str]:\\n tags = []\\n buf = []\\n in_quoted = None\\n\\n for c in s:\\n if in_quoted:\\n if c == in_quoted:\\n in_quoted = None\\n else:\\n buf.append(c)\\n elif c == '\\\"' or c == '\\\\'':\\n in_quoted = c\\n elif c == ',':\\n if buf:\\n tag = ''.join(buf).strip()\\n if tag:\\n tags.append(tag)\\n buf.clear()\\n else:\\n buf.append(c)\\n\\n if buf:\\n tag = ''.join(buf).strip()\\n if tag:\\n tags.append(tag)\\n\\n return tags\",\n \"def _get_tags(tag_string):\\n print \\\"inside get_tags, string is: \\\", tag_string\\n tag_objects_list = []\\n # remove all whitespaces\\n tag_string_cleaned = tag_string.replace(\\\" \\\", \\\"\\\")\\n tokens = tag_string_cleaned.split(',')\\n for tok in tokens:\\n try:\\n tag_object = Tag.objects.get(tag_name=tok)\\n except Tag.DoesNotExist:\\n tag_object = Tag(tag_name=tok)\\n tag_object.save()\\n if tag_object not in tag_objects_list:\\n tag_objects_list.append(tag_object)\\n return tag_objects_list\",\n \"def parse(owners_lines):\\n results = []\\n for owner_line in owners_lines:\\n # strip the leading @ for github users/teams\\n if owner_line.startswith('@'):\\n owner_line = owner_line[1:]\\n owner_line = owner_line.strip()\\n if not owner_line:\\n continue\\n if owner_line.startswith('#'):\\n continue\\n if ' ' in owner_line:\\n result = tuple(owner_line.split(' ', 1))\\n else:\\n result = (owner_line, None)\\n results.append(result)\\n return results\",\n \"def tweet_parser(tweet):\\n tweet = tweet.encode()\\n\\n match = re.match(\\n r'^@\\\\w{11} (@(?P<opponent>[^\\\\s]+) )?#(?P<game>\\\\w+)( (?P<move>[^\\\\s]+))?( (?P<message>.*))?$',\\n tweet\\n )\\n\\n if match:\\n return match.groupdict()\\n else:\\n raise ValueError(\\\"Tweet not formatted correctly.\\\")\",\n \"def parse_tags(self, in_str):\\n self.projects = []\\n self.contexts = []\\n for match in TAG_REGEX.finditer(in_str):\\n tag_str = match.group(1)\\n if tag_str[0] == \\\"+\\\":\\n self.projects.append(tag_str[1:])\\n elif tag_str[0] == \\\"@\\\":\\n self.contexts.append(tag_str[1:])\\n else:\\n err_str = \\\"Unable to parse tag: %s\\\" % tag_str\\n raise ValueError(err_str)\\n words = re.sub(TAG_REGEX, \\\"\\\", in_str).split()\\n return \\\" \\\".join(words)\",\n \"def tags(self):\\n return tuple([x.strip() for x in self._dict.get('tags').split(',')])\",\n \"def parse_etags(etag_str):\\n if etag_str.strip() == \\\"*\\\":\\n return [\\\"*\\\"]\\n else:\\n # Parse each ETag individually, and return any that are valid.\\n etag_matches = (ETAG_MATCH.match(etag.strip()) for etag in etag_str.split(\\\",\\\"))\\n return [match[1] for match in etag_matches if match]\",\n \"def twitter_text(\\n self,\\n text: str,\\n urls: List[Dict[str, str]],\\n user_mentions: List[Dict[str, Any]],\\n media: List[Dict[str, Any]],\\n ) -> Element:\\n\\n to_process: List[Dict[str, Any]] = []\\n # Build dicts for URLs\\n for url_data in urls:\\n to_process.extend(\\n {\\n \\\"type\\\": \\\"url\\\",\\n \\\"start\\\": match.start(),\\n \\\"end\\\": match.end(),\\n \\\"url\\\": url_data[\\\"url\\\"],\\n \\\"text\\\": url_data[\\\"expanded_url\\\"],\\n }\\n for match in re.finditer(re.escape(url_data[\\\"url\\\"]), text, re.IGNORECASE)\\n )\\n # Build dicts for mentions\\n for user_mention in user_mentions:\\n screen_name = user_mention[\\\"screen_name\\\"]\\n mention_string = \\\"@\\\" + screen_name\\n to_process.extend(\\n {\\n \\\"type\\\": \\\"mention\\\",\\n \\\"start\\\": match.start(),\\n \\\"end\\\": match.end(),\\n \\\"url\\\": \\\"https://twitter.com/\\\" + urllib.parse.quote(screen_name),\\n \\\"text\\\": mention_string,\\n }\\n for match in re.finditer(re.escape(mention_string), text, re.IGNORECASE)\\n )\\n # Build dicts for media\\n for media_item in media:\\n short_url = media_item[\\\"url\\\"]\\n expanded_url = media_item[\\\"expanded_url\\\"]\\n to_process.extend(\\n {\\n \\\"type\\\": \\\"media\\\",\\n \\\"start\\\": match.start(),\\n \\\"end\\\": match.end(),\\n \\\"url\\\": short_url,\\n \\\"text\\\": expanded_url,\\n }\\n for match in re.finditer(re.escape(short_url), text, re.IGNORECASE)\\n )\\n # Build dicts for emojis\\n for match in POSSIBLE_EMOJI_RE.finditer(text):\\n orig_syntax = match.group(\\\"syntax\\\")\\n codepoint = emoji_to_hex_codepoint(unqualify_emoji(orig_syntax))\\n if codepoint in codepoint_to_name:\\n display_string = \\\":\\\" + codepoint_to_name[codepoint] + \\\":\\\"\\n to_process.append(\\n {\\n \\\"type\\\": \\\"emoji\\\",\\n \\\"start\\\": match.start(),\\n \\\"end\\\": match.end(),\\n \\\"codepoint\\\": codepoint,\\n \\\"title\\\": display_string,\\n }\\n )\\n\\n to_process.sort(key=lambda x: x[\\\"start\\\"])\\n p = current_node = Element(\\\"p\\\")\\n\\n def set_text(text: str) -> None:\\n \\\"\\\"\\\"\\n Helper to set the text or the tail of the current_node\\n \\\"\\\"\\\"\\n if current_node == p:\\n current_node.text = text\\n else:\\n current_node.tail = text\\n\\n db_data: Optional[DbData] = self.zmd.zulip_db_data\\n current_index = 0\\n for item in to_process:\\n # The text we want to link starts in already linked text skip it\\n if item[\\\"start\\\"] < current_index:\\n continue\\n # Add text from the end of last link to the start of the current\\n # link\\n set_text(text[current_index : item[\\\"start\\\"]])\\n current_index = item[\\\"end\\\"]\\n if item[\\\"type\\\"] != \\\"emoji\\\":\\n elem = url_to_a(db_data, item[\\\"url\\\"], item[\\\"text\\\"])\\n assert isinstance(elem, Element)\\n else:\\n elem = make_emoji(item[\\\"codepoint\\\"], item[\\\"title\\\"])\\n current_node = elem\\n p.append(elem)\\n\\n # Add any unused text\\n set_text(text[current_index:])\\n return p\",\n \"def _parse_mercurial_author(data, id_gen):\\n angled = ur'(?P<author>.+?) <(?P<email>.+?)>'\\n paren = ur'(?P<email>.+?) \\\\((?P<author>.+?)\\\\)'\\n simple = ur'(?P<author>[^,]+)'\\n author_list = []\\n for regex in (angled, paren, simple):\\n # Watch out for commas separating multiple names.\\n regex += u'(,\\\\s*)?'\\n for match in re.finditer(regex, data):\\n # Watch out for suffixes like 'Jr.' when they are comma-separated\\n # from the name and thus cause issues when *all* names are only\\n # separated by commas.\\n match_dict = match.groupdict()\\n author = match_dict['author']\\n if not author.partition(' ')[1] and author.endswith('.'):\\n prev_author = author_list.pop()\\n author = ', '.join([prev_author, author])\\n if u'email' not in match_dict:\\n email = ''\\n else:\\n email = match_dict['email']\\n author_list.append((author, email))\\n else:\\n # If authors were found then stop searching as only expect one\\n # style of author citation.\\n if author_list:\\n break\\n author = Author(author_list[0])\\n user = author.first_last\\n email = author.email\\n uid = id_gen[user]\\n return (uid, user, email)\",\n \"def extract_hashtag(text):\\n if text is not None:\\n text = text.replace('\\\\n', ' ').replace('\\\\r', '')\\n text = text.split(\\\" \\\")\\n text = [word for word in text if \\\"#\\\" in word]\\n if len(text) == 0:\\n text = [\\\"no tags\\\"]\\n else:\\n text = [\\\"no tags\\\"]\\n return text\",\n \"async def tag_sanskrit(sanskrit_string: str):\\n result = search_utils.tag_sanskrit(sanskrit_string).replace(\\\"\\\\n\\\",\\\" # \\\")\\n return {\\\"tagged\\\": result}\",\n \"def parse_slack_output(slack_rtm_output):\\r\\n output_list = slack_rtm_output\\r\\n if output_list and len(output_list) > 0:\\r\\n for output in output_list:\\r\\n if output and 'text' in output and output['text'].startswith(AT_BOT):\\r\\n # return text after the @ mention, whitespace removed\\r\\n return output['text'].split(AT_BOT)[1].strip().lower(), \\\\\\r\\n output['channel'], output['user']\\r\\n return None, None, None\",\n \"def detag_string(self, string):\\r\\n counter = itertools.count(0)\\r\\n count = lambda m: '<%s>' % counter.next()\\r\\n tags = self.tag_pattern.findall(string)\\r\\n tags = [''.join(tag) for tag in tags]\\r\\n (new, nfound) = self.tag_pattern.subn(count, string)\\r\\n if len(tags) != nfound:\\r\\n raise Exception('tags dont match:' + string)\\r\\n return (new, tags)\",\n \"def author_name(text):\\n tag = text.split()\\n\\n \\\"\\\"\\\"\\n We take the beginning of the text since the\\n author name will likely be there\\n \\\"\\\"\\\"\\n\\n tag = tag[:100]\\n author = []\\n\\n current_tag = 0\\n \\\"\\\"\\\"\\n We go through each word until we find the first instance\\n of the word 'by' or 'author', which should mean the author\\n will be written right after that.\\n We save the first word after 'by' or 'author' since it should\\n be the authors first name\\n \\\"\\\"\\\"\\n\\n for word in tag:\\n if (word.lower() == ('by') or\\n word.lower() == ('author') or\\n word.lower() == ('author:')):\\n\\n author.append(tag[current_tag+1].decode(encoding='UTF8',\\n errors='ignore'))\\n current_tag += 1\\n tag = tag[current_tag+1:]\\n break\\n current_tag += 1\\n\\n \\\"\\\"\\\"\\n We go through each word after the first name of the author\\n until we find a word that is not capitalized. We assume that\\n it marks the end of the author name.\\n We then return a list of the author's name split up.\\n \\\"\\\"\\\"\\n current_tag = 0\\n for word in tag:\\n if tag[current_tag].lower() == 'this':\\n break\\n if tag[current_tag].istitle():\\n author.append(tag[current_tag].decode(encoding='UTF8',\\n errors='ignore'))\\n current_tag += 1\\n\\n return author\",\n \"def read(string):\\n\\treturn (re.finditer('(?<=\\\\[)[a-z]+(?=\\\\])', string), re.finditer('(?<=\\\\])[a-z]+|[a-z]+(?=\\\\[)', string))\",\n \"def parse_tag_input(input):\\r\\n if not input:\\r\\n return []\\r\\n\\r\\n input = force_unicode(input)\\r\\n\\r\\n # Special case - if there are no commas or double quotes in the\\r\\n # input, we don't *do* a recall... I mean, we know we only need to\\r\\n # split on spaces.\\r\\n if u',' not in input and u'\\\"' not in input:\\r\\n words = list(set(split_strip(input, u' ')))\\r\\n words.sort()\\r\\n return words\\r\\n\\r\\n words = []\\r\\n buffer = []\\r\\n # Defer splitting of non-quoted sections until we know if there are\\r\\n # any unquoted commas.\\r\\n to_be_split = []\\r\\n saw_loose_comma = False\\r\\n open_quote = False\\r\\n i = iter(input)\\r\\n try:\\r\\n while 1:\\r\\n c = i.next()\\r\\n if c == u'\\\"':\\r\\n if buffer:\\r\\n to_be_split.append(u''.join(buffer))\\r\\n buffer = []\\r\\n # Find the matching quote\\r\\n open_quote = True\\r\\n c = i.next()\\r\\n while c != u'\\\"':\\r\\n buffer.append(c)\\r\\n c = i.next()\\r\\n if buffer:\\r\\n word = u''.join(buffer).strip()\\r\\n if word:\\r\\n words.append(word)\\r\\n buffer = []\\r\\n open_quote = False\\r\\n else:\\r\\n if not saw_loose_comma and c == u',':\\r\\n saw_loose_comma = True\\r\\n buffer.append(c)\\r\\n except StopIteration:\\r\\n # If we were parsing an open quote which was never closed treat\\r\\n # the buffer as unquoted.\\r\\n if buffer:\\r\\n if open_quote and u',' in buffer:\\r\\n saw_loose_comma = True\\r\\n to_be_split.append(u''.join(buffer))\\r\\n if to_be_split:\\r\\n if saw_loose_comma:\\r\\n delimiter = u','\\r\\n else:\\r\\n delimiter = u' '\\r\\n for chunk in to_be_split:\\r\\n words.extend(split_strip(chunk, delimiter))\\r\\n words = list(set(words))\\r\\n words.sort()\\r\\n return words\",\n \"def parse_slack_output(slack_rtm_output):\\n output_list = slack_rtm_output\\n if output_list and len(output_list) > 0:\\n print 'output list is: ', output_list\\n for output in output_list:\\n \\n if (output and 'text' in output and AT_BOT in output['text'] \\n and output['channel'].startswith('D')\\n ):\\n # return text after the @ mention, whitespace removed\\n return output['text'].split(AT_BOT)[1].strip().lower(), \\\\\\n output['channel'], output['user']\\n\\n elif (output and 'text' in output and output['channel'].startswith('D')\\n and output['user'] != BOT_ID):\\n return output['text'], output['channel'], output['user']\\n\\n return None, None, None\",\n \"def get_mentions_from_body_and_issue_comments(pull_request_body, pull_request_comments):\\n unique_mentions = set()\\n unique_mentions.update(github_mention_regex.findall(pull_request_body))\\n for comment in pull_request_comments:\\n unique_mentions.update(github_mention_regex.findall(comment.body))\\n\\n return unique_mentions\",\n \"def pos_tag(self,sentence):\\n tagged = self.brill_tagger.tag(sentence.split())\\n tagged_sentence = \\\" \\\".join([nltk.tag.tuple2str(tok) for tok in tagged])\\n print tagged_sentence\\n\\n tag_list = [(each.split(\\\"/\\\")[0],each.split(\\\"/\\\")[1]) for each in tagged_sentence.split()]\\n return tag_list\",\n \"def parse(self, string, root=None):\\n\\n\\t\\tphrases = []\\n\\n\\t\\tmeta = self.meta.search(string)\\n\\n\\t\\twhile meta:\\n\\n\\t\\t\\t# Save some function calls\\n\\t\\t\\tpos = meta.start()\\n\\n\\t\\t\\tif meta.group() == \\\"<\\\":\\n\\t\\t\\t\\tstring, child, meta = self.open_phrase(string, pos)\\n\\n\\t\\t\\t\\tif child and root:\\n\\t\\t\\t\\t\\troot.nested.append(child)\\n\\t\\t\\t\\telif child:\\n\\t\\t\\t\\t\\tphrases.append(child)\\n\\n\\t\\t\\t\\t# else it was escaped (+ new meta)\\n\\t\\t\\t\\tcontinue\\n\\n\\t\\t\\telif root:\\n\\n\\t\\t\\t\\tif meta.group() == \\\"(\\\":\\n\\t\\t\\t\\t\\tmeta = self.meta.search(string, pos + 1)\\n\\t\\t\\t\\t\\tif meta.group() == \\\")\\\":\\n\\t\\t\\t\\t\\t\\tstring, root, meta = self.handle_arguments(string,\\n\\t\\t\\t\\t\\t\\t\\t\\t\\t\\t\\t\\t\\t\\t\\t\\t root,\\n\\t\\t\\t\\t\\t\\t\\t\\t\\t\\t\\t\\t\\t\\t\\t\\t pos,\\n\\t\\t\\t\\t\\t\\t\\t\\t\\t\\t\\t\\t\\t\\t\\t\\t meta.start())\\n\\t\\t\\t\\t\\t\\tcontinue\\n\\n\\t\\t\\t\\telif meta.group() == \\\">\\\":\\n\\t\\t\\t\\t\\tstring, phrase, meta = self.close_phrase(string,\\n\\t\\t\\t\\t\\t\\t\\t\\t\\t\\t\\t\\t\\t\\t\\t root,\\n\\t\\t\\t\\t\\t\\t\\t\\t\\t\\t\\t\\t\\t\\t\\t pos)\\n\\t\\t\\t\\t\\tif phrase:\\n\\t\\t\\t\\t\\t\\treturn string, phrase\\n\\n\\t\\t\\t\\t\\t# else was escaped (+ new meta)\\n\\t\\t\\t\\t\\tcontinue\\n\\n\\t\\t\\tstring, meta = self.escape_meta(string, pos)\\n\\n\\t\\tif not root:\\n\\t\\t\\treturn string, phrases\\n\\n\\t\\t# If this is not the first stack-depth the function should\\n\\t\\t# have returned upon finding a closing tag,\\n\\t\\t# i.e. we should never have gotten here.\\n\\t\\tword = re.search(r\\\"([\\\\w\\\\s]+)(?![\\\\d]*>[\\\\w\\\\s]+>)\\\", string)\\n\\n\\t\\twhat = \\\"No closing tag found for opening tag\\\"\\n\\n\\t\\tif word:\\n\\t\\t\\twhat += \\\" after expression '{0}'\\\".format(word.group())\\n\\n\\t\\traise errors.ParseError(what + \\\"!\\\")\",\n \"def extract_mentions_or_lists_with_indices(self, transform = lambda x: x):\\r\\n if not REGEXEN['at_signs'].search(self.text):\\r\\n return []\\r\\n\\r\\n possible_entries = []\\r\\n for match in REGEXEN['valid_mention_or_list'].finditer(self.text):\\r\\n try:\\r\\n after = self.text[match.end()]\\r\\n except IndexError:\\r\\n # the mention was the last character in the string\\r\\n after = None\\r\\n if after and REGEXEN['end_mention_match'].match(after) or match.groups()[2].find('http') == 0:\\r\\n continue\\r\\n possible_entries.append({\\r\\n 'screen_name': transform(match.groups()[2]),\\r\\n 'list_slug': match.groups()[3] or '',\\r\\n 'indices': [match.start() + len(match.groups()[0]), match.end()]\\r\\n })\\r\\n\\r\\n return possible_entries\",\n \"def get_mentions(texts, word):\\r\\n mentions = {}\\r\\n for text, label, text_id in texts:\\r\\n if word in text.lower():\\r\\n if label not in mentions:\\r\\n mentions[label] = [text_id]\\r\\n else:\\r\\n if text_id not in mentions[label]:\\r\\n mentions[label].append(text_id)\\r\\n return mentions\",\n \"def parse_tags(tags: str = None) -> Iterable[str]:\\n if not tags:\\n return []\\n return [x.strip() for x in tags.split(\\\",\\\")]\",\n \"def _find_subtype_mentions(self, message):\\n subtype_mention_regex = '<@[a-zA-Z0-9]+|'\\n user_mentions = re.findall(subtype_mention_regex, message['text'])\\n user_ids = [mention.strip('@<>') for mention in user_mentions]\\n return [user_id for user_id in user_ids if user_id != self.poolbot.bot_id]\",\n \"def parse(self, text):\\n assert isinstance(text, str)\\n if text.strip() == '':\\n return [], []\\n\\n output = self._annotate(text, properties={\\n \\\"annotators\\\": \\\"tokenize,ssplit,pos\\\",\\n \\\"coref.md.type\\\": \\\"dep\\\",\\n \\\"coref.mode\\\": \\\"statistical\\\"\\n })\\n\\n words = []\\n postags = []\\n\\n for sentence in output['sentences']:\\n for token in sentence['tokens']:\\n word = token['word']\\n pos = token['pos']\\n word = re.sub(r'\\\\s', '', word)\\n words.append(word)\\n postags.append(pos)\\n return words, postags\",\n \"def parse_atwho(my_text):\\n notification_list = []\\n \\n #markdown_link_regex = re.compile(\\\"\\\\[.*?\\\\] \\\\((.*?) \\\\\\\".*?\\\\\\\"\\\\)\\\", re.IGNORECASE) # need to test this.\\n markdown_link_regex = re.compile(\\\"\\\\[([^\\\\]]+)\\\\]\\\\(([^)\\\\\\\"]+)(?: \\\\\\\\\\\\\\\"([^\\\\\\\\\\\\\\\"]+)\\\\\\\\\\\\\\\")?\\\\)\\\", re.IGNORECASE)\\n direct_link_regex = re.compile(\\\"data-notify=\\\\\\\\\\\\\\\"([^\\\\\\\\\\\\\\\"]+)\\\\\\\\\\\\\\\"\\\", re.IGNORECASE)\\n link_list = [i[2] for i in markdown_link_regex.findall(my_text)]\\n link_list += [i for i in direct_link_regex.findall(my_text)]\\n\\n for i in link_list:\\n _type, _id = i.split(\\\"#\\\", 1)\\n if _type == \\\"user\\\":\\n notification_list.append(User.objects.get(id=_id))\\n elif _type == \\\"subdept\\\":\\n notification_list.append(Subdept.objects.get(id=_id))\\n elif _type == \\\"dept\\\":\\n notification_list.append(Dept.objects.get(id=_id))\\n elif _type == \\\"page\\\":\\n notification_list.append(Page.objects.get(id=_id))\\n\\n return my_text, notification_list\",\n \"def completion(s):\\n if len(s) >= 1 and s[0] == 'h':\\n return ('hello', 'hello there')\\n return None\",\n \"def resolve_function(cls, phrase_mention: List[Tuple[Tuple[str, str], str, str, Tuple[str, str]]]) -> List[str]:\\n\\n # Iterate over each component in `phrase_mention`, and extract possible user groups if the first POS tag is a\\n # verb\\n res = [cls.get_user_group(*str_split) for pos, _, _, str_split in phrase_mention if \\\"verb\\\" in pos[0]]\\n\\n # Return non-empty string user groups\\n return [r for r in res if r != \\\"\\\"]\",\n \"def split_hashtags(hashtag):\\n matches = re.findall('.+?(?:(?<=[a-z])(?=[A-Z])|(?<=[A-Z])(?=[A-Z][a-z])|$)', hashtag)\\n return matches\",\n \"def _pairs(content: str) -> Generator[Tuple[Any, Any], None, None]:\\n if not content.strip():\\n return\\n\\n for pair in content.split(\\\",\\\"):\\n\\n x, *y = pair.split(\\\":\\\", maxsplit=1)\\n if len(y) == 0:\\n raise ValueError(f\\\"Cannot parse pair {pair}\\\")\\n\\n yield x.strip(), y[0].strip()\",\n \"def parse_slack_output(slack_rtm_output, bot):\\n output_list = slack_rtm_output\\n if output_list and len(output_list) > 0:\\n for output in output_list:\\n at_bot = \\\"<@%s>\\\" % bot['bot_id']\\n if output and 'text' in output and at_bot in output['text'] and 'channel' in output:\\n # return text after the @ mention, whitespace removed\\n return output['text'].split(at_bot)[1].strip(), \\\\\\n output['channel']\\n return None, None\",\n \"def linkify_tags_and_mentions(value):\\n value = find_hashtags_re.sub(tag_match_to_url, sanitize(value))\\n value = find_mentions_re.sub(mention_match_to_url, value)\\n # value = link_tags_parse(value)\\n return mark_safe(value)\",\n \"def get_tags_for_tags_and_malware_family_fields(tags: Optional[list], is_malware_family=False):\\n if not tags:\\n return None\\n results = []\\n for item in tags:\\n results.append(item.get('tag_name'))\\n results.append(item.get('public_tag_name'))\\n for alias in item.get('aliases', []):\\n results.append(alias)\\n if not is_malware_family:\\n for group in item.get('tagGroups', [{}]):\\n results.append(group.get('tag_group_name'))\\n # Returns a list without duplicates and empty elements\\n return list(set(filter(None, results)))\",\n \"def get_author(html_soup):\\n auth_text = html_soup.find('div', attrs = {\\\"class\\\" : \\\"author\\\"}).text.split(\\\"|\\\")\\n for i in auth_text:\\n if 'By' in i:\\n auth_text_split = i.split()\\n auth_text_split = auth_text_split[auth_text_split.index('By')+1:auth_text_split.index('By')+3]\\n auth_text_split = ' '.join(auth_text_split)\\n return(auth_text_split)\",\n \"def _parse_tags (tag, multi_kind_dataset ='bagoue'): \\r\\n tag = str(tag); t = tag.strip().split() \\r\\n \\r\\n if len(t) ==1 : \\r\\n if t[0].lower() not in _DTAGS: \\r\\n tag = multi_kind_dataset +' ' + t[0]\\r\\n \\r\\n warn(f\\\"Fetching {multi_kind_dataset.title()!r} data without\\\"\\r\\n \\\" explicitly prefixing the kind of data with the area\\\"\\r\\n \\\" name will raise an error. In future, the argument\\\"\\r\\n f\\\" should be '{tag}' instead.\\\", FutureWarning \\r\\n )\\r\\n elif len(t) >1 : \\r\\n # only the multi kind dataset is allowed \\r\\n # to contain two words for fetching data \\r\\n if t[0].lower() !=multi_kind_dataset: \\r\\n tag = t[0].lower() # skip the second word \\r\\n return tag\",\n \"def hashtag_seperator(s: List[tuple]) -> List[str]:\\n result = []\\n for tups in s:\\n text = tups[0]\\n result = result + extract_hashtags(text)\\n return result\",\n \"def _extract_tags(html):\\n tags = re.findall(r'<[^>]+>', html)\\n\\n return tags\",\n \"def ig_users_in(body):\\n import re\\n\\n try:\\n user_re = Formatter.USER_TAG_REGEX\\n except AttributeError:\\n # escape special characters so we can format the user-tag into a\\n # regex pattern\\n escaped_user_tag = re.sub(\\n # match any '[', ']', '(', or ')'\\n r'([\\\\[\\\\]\\\\(\\\\)])',\\n # escape the matched character\\n r'\\\\\\\\\\\\1',\\n Formatter.USER_TAG\\n )\\n pattern = escaped_user_tag.format(\\n user_raw='({0})'.format(instagram.USERNAME_PTN)\\n )\\n user_re = re.compile(pattern)\\n Formatter.USER_TAG_REGEX = user_re\\n\\n return user_re.findall(body)\",\n \"def bio_tag_to_entity_mentions(bio_tags):\\n within_entity_mention = False\\n mentions = []\\n for i, tag in enumerate(bio_tags):\\n if not within_entity_mention:\\n # start of entity mention\\n if 'B-' in tag:\\n start = i\\n within_entity_mention = True\\n else:\\n if tag == 'O':\\n end = i\\n mentions.append({\\n 'start' : start,\\n 'end' : end,\\n 'labels' : []\\n })\\n within_entity_mention = False\\n # two entities side by side without O tag\\n if 'B-' in tag:\\n end = i\\n mentions.append({\\n 'start' : start,\\n 'end' : end,\\n 'labels' : []\\n })\\n start = i\\n if within_entity_mention:\\n end = len(bio_tags)\\n mentions.append({\\n 'start' : start,\\n 'end' : end,\\n 'labels' : []\\n })\\n return mentions\",\n \"def get_speakers(words):\\n speakers = []\\n utterances = \\\" \\\".join(words).split(\\\"<|endoftext|>\\\")[:-1]\\n for u in utterances:\\n s = u.split(\\\":\\\")[0].strip() # select names according to \\\":\\\"\\n if s:\\n speakers.append(s)\\n return list(set(speakers))\",\n \"def gen_sent_coresponding_tags(self, sentence: str,\\n tagged_tokens: TaggedTokens,\\n unigrams: Unigrams,\\n tag_sets: TagSets,\\n syl_stats: List = None) -> Tuple:\\n\\n sentence_tokens = sentence.split()\\n sentence_tags = tagged_tokens.get_sentence_tags(sentence)\\n new_sentence = []\\n new_sentence_meta = []\\n syl_stats = syl_stats or [None] * len(sentence_tokens)\\n\\n for index, (tag, source) in enumerate(sentence_tags):\\n n_syl = syl_stats[index]\\n\\n try:\\n new_token = self.rand_token_tag_unigrams(tag, unigrams,\\n tag_sets,\\n n_syllables=n_syl)\\n\\n new_sentence.append(new_token)\\n new_sentence_meta.append(source)\\n\\n except (TagDoesNotExist, NoContinuation):\\n new_sentence.append(sentence_tokens[index])\\n new_sentence_meta.append('not_replaced')\\n\\n return new_sentence, new_sentence_meta\",\n \"def parse_slack_output(slack_rtm_output):\\n output_list = slack_rtm_output\\n if output_list and len(output_list) > 0:\\n for output in output_list:\\n if output and 'text' in output and AT_BOT in output['text']:\\n # return text after the @ mention, whitespace removed\\n return output['text'].split(AT_BOT)[1].strip(), \\\\\\n output['channel']\\n return None, None\",\n \"def parse_tweet(line):\\n # The following regex just strips of an URL (not just http), any punctuations,\\n # or Any non alphanumeric characters\\n # http://goo.gl/J8ZxDT\\n text = re.sub(\\\"(@[A-Za-z0-9]+)|([^0-9A-Za-z \\\\t])|(\\\\w+:\\\\/\\\\/\\\\S+)\\\",\\\" \\\",json.loads(line[1])[\\\"value\\\"]).strip()\\n # remove terms <= 2 characters\\n text = ' '.join(filter(lambda x: len(x) > 2, text.split(\\\" \\\")))\\n\\n return (line[0], text)\",\n \"def extract_entities(event):\\n # TODO The text should probably already be tagged and tokenized before this step\\n tree = ne_chunk(event.pos_tagged)\\n entities = set([])\\n\\n people = tree.subtrees(lambda x: x.node == \\\"PERSON\\\")\\n for person in people:\\n entities.add(\\\" \\\".join([leaf[0] for leaf in person.leaves()]))\\n\\n places = tree.subtrees(lambda x: x.node == \\\"GPE\\\")\\n for place in places:\\n entities.add(\\\" \\\".join([leaf[0] for leaf in place.leaves()]))\\n\\n organizations = tree.subtrees(lambda x: x.node == \\\"ORGANIZATION\\\")\\n for org in organizations:\\n entities.add(\\\" \\\".join([leaf[0] for leaf in org.leaves()]))\\n \\n return entities\",\n \"def extract_metadata(info_str):\\n info = [i.strip() for i in info_str.split(';')\\n if i.strip().startswith('gene_id')\\n or i.strip().startswith('gene_type')]\\n assert len(info) == 2, '{0}'.format(info)\\n gene_id, gene_type = extract_gene_data(info)\\n return gene_id, gene_type\",\n \"def test_returns_nouns_for_string(self):\\r\\n test_value = 'google drives autonomous cars'\\r\\n self.assertEqual(\\r\\n set([u'cars', u'autonomous']),\\r\\n suggest_tags(test_value))\",\n \"def splitTags(user_input):\\n \\n elements = []\\n if ',' in user_input:\\n elements = user_input.split(',')\\n elif ' ' in user_input:\\n elements = user_input.split(' ')\\n else:\\n elements.append(user_input)\\n\\n tags = []\\n for element in elements:\\n element = element.strip(' \\\\t\\\\n\\\\r').lower()\\n if(len(element) == 0): continue\\n if element not in tags:\\n tags.append(element)\\n return tags\",\n \"def findall(self, string: str) -> List[str]:\\r\\n \\r\\n string = list(string)\\r\\n matches = []\\r\\n\\r\\n while len(string) > 0:\\r\\n tokens = self.tokens.copy()\\r\\n match = []\\r\\n\\r\\n while len(tokens) > 0:\\r\\n token = _advance(tokens)\\r\\n char = _advance(string)\\r\\n\\r\\n if token is None:\\r\\n break\\r\\n if char is None and token.type != TokenType.OPTIONAL:\\r\\n match = []\\r\\n break\\r\\n\\r\\n value = token.match(tokens, string, char)\\r\\n if value is None:\\r\\n match = []\\r\\n break\\r\\n match.extend(value)\\r\\n \\r\\n if len(match) > 0:\\r\\n matches.append(''.join(match))\\r\\n\\r\\n return matches\",\n \"def parse(string):\\n doc = nlp(string)\\n return [str(n) for n in doc.noun_chunks]\",\n \"def preprocess_tweet(tweet):\\n\\n\\n clean_tweet, hashtags = separate_hastags_mentions_urls(tweet)\\n clean_tweet = remove_emoji_punc(clean_tweet)\\n return clean_tweet, hashtags\",\n \"def parseMuttAddressbook(addressbook):\\n\\n alias1 = re.compile('(^alias\\\\s*)(\\\\S*\\\\s*)(\\\".*\\\"\\\\s*)(<)(.*)(>$)')\\n alias2 = re.compile(\\\"(^alias\\\\s*)(\\\\S*\\\\s*)(\\\\S*)(\\\\s*\\\\(.*$)\\\")\\n\\n addresses = set()\\n for line in open(addressbook).readlines():\\n alias1match = alias1.match(line)\\n alias2match = alias2.match(line)\\n if alias1match is not None:\\n address = alias1.match(line).group(5)\\n addresses.add(address)\\n elif alias2match is not None:\\n address = alias2.match(line).group(3)\\n addresses.add(address)\\n\\n return addresses\",\n \"def detect_author(user_to_tweets: Dict[str, List[tuple]], tweet_text: str) -> \\\\\\n str:\\n acc = []\\n \\n for keys in user_to_tweets:\\n author_hashes = hashtag_seperator(user_to_tweets[keys])\\n text_hashes = extract_hashtags(tweet_text)\\n if set(text_hashes).issubset(author_hashes):\\n acc.append(keys)\\n if len(acc) == 1:\\n return acc[0]\\n return 'unknown'\",\n \"def _collect_user_data(utt_sets: Collection[Collection[Utterance]]) -> Tuple[Dict[str, Dict[Hashable, str]], Dict[str, Dict[Hashable, bool]]]:\\n # Collect USER data and metadata\\n # all_users_data = defaultdict(lambda: defaultdict(set))\\n all_users_meta = defaultdict(lambda: defaultdict(str))\\n all_users_meta_conflict = defaultdict(lambda: defaultdict(bool))\\n for utt_set in utt_sets:\\n for utt in utt_set:\\n for meta_key, meta_val in utt.user.meta.items():\\n curr = all_users_meta[utt.user][meta_key]\\n if curr != meta_val:\\n if curr != \\\"\\\":\\n all_users_meta_conflict[utt.user][meta_key] = True\\n all_users_meta[utt.user][meta_key] = meta_val\\n\\n return all_users_meta, all_users_meta_conflict\",\n \"def extract_mentions(text: str) -> List[str]:\\n final = []\\n result = list(text.split())\\n \\n for word in result: \\n if word[0] == MENTION_SYMBOL:\\n final.append(alnum_prefix(word[1:]))\\n if alnum_prefix(word[1:]) == '':\\n final.remove(alnum_prefix(word[1:]))\\n return final\",\n \"def parse_slack_output(slack_rtm_output):\\n output_list = slack_rtm_output\\n if output_list and len(output_list) > 0:\\n for output in output_list:\\n if output and 'text' in output and AT_BOT in output['text']:\\n # return text after the @ mention, whitespace removed\\n return output['text'].split(AT_BOT)[1].strip().lower(), \\\\\\n output['channel']\\n return None, None\",\n \"def parse(s, tokenize=True, tags=True, chunks=True, relations=False, lemmata=False, encoding=\\\"utf-8\\\", **kwargs):\\n if tokenize:\\n s = _tokenize(s)\\n if isinstance(s, (list, tuple)):\\n s = [isinstance(s, basestring) and s.split(\\\" \\\") or s for s in s]\\n if isinstance(s, basestring):\\n s = [s.split(\\\" \\\") for s in s.split(\\\"\\\\n\\\")]\\n # Reuse the English parser:\\n kwargs.update({\\n \\\"lemmata\\\": False,\\n \\\"light\\\": False,\\n \\\"lexicon\\\": LEXICON,\\n \\\"language\\\": \\\"es\\\",\\n \\\"default\\\": \\\"NC\\\",\\n \\\"map\\\": kwargs.get(\\\"tagset\\\", \\\"\\\") != PAROLE and parole2penntreebank or None,\\n })\\n s = _en_parse(s, False, tags, chunks, relations, **kwargs)\\n # Use pattern.es.inflect for lemmatization:\\n if lemmata:\\n p = [find_lemmata(sentence) for sentence in s.split()]\\n p = \\\"\\\\n\\\".join([\\\" \\\".join([\\\"/\\\".join(token) for token in sentence]) for sentence in p])\\n s = TaggedString(p, tags=s.tags+[\\\"lemma\\\"], language=\\\"es\\\")\\n return s\",\n \"def speaking_pairs(self, selector: Optional[Callable[[User, User], bool]]=None,\\n user_names_only: bool=False) -> Set[Tuple]:\\n pairs = set()\\n for u2 in self.utterances.values():\\n if u2.user is not None and u2.reply_to is not None and u2.reply_to in self.utterances:\\n u1 = self.utterances[u2.reply_to]\\n if u1.user is not None:\\n if selector is None or selector(u2.user, u1.user):\\n pairs.add((u2.user.name, u1.user.name) if\\n user_names_only else (u2.user, u1.user))\\n return pairs\",\n \"def _parse_tags(self):\\n tokens = self.tags_str[1:].split(\\\";\\\")\\n self._tags = {\\n k.strip(): v\\n for token in tokens\\n for k, v in [token.split(\\\"=\\\")]\\n }\",\n \"def tags_in_string(msg):\\r\\n def is_linguistic_tag(tag):\\r\\n \\\"\\\"\\\"Is this tag one that can change with the language?\\\"\\\"\\\"\\r\\n if tag.startswith(\\\"&\\\"):\\r\\n return True\\r\\n if any(x in tag for x in [\\\"<abbr>\\\", \\\"<abbr \\\", \\\"</abbr>\\\"]):\\r\\n return True\\r\\n return False\\r\\n\\r\\n __, tags = Converter().detag_string(msg)\\r\\n return set(t for t in tags if not is_linguistic_tag(t))\",\n \"def tokens(text) -> Tuple[Word]:\\n return tuple(re.findall('[a-z]+', text.lower()))\",\n \"def tags_from_csv_field(tag_string):\\n split_string = tag_string.split()\\n out_list = []\\n for tag in split_string:\\n out_list.append(clean_tag(tag))\\n\\n return out_list\",\n \"def parse_slack_output(ARGS, slack_rtm_output):\\n output_list = slack_rtm_output\\n if output_list and len(output_list) > 0:\\n for output in output_list:\\n if output and 'text' in output and ARGS.atbot in output['text']:\\n # return text after the @ mention, whitespace removed\\n return output['text'].split(ARGS.atbot)[1].strip().lower(), \\\\\\n output['channel']\\n return None, None\",\n \"def get_user_by_email(tx: Transaction, user_email: str) -> BoltStatementResult:\\n\\n # NOTE: tag labels are hardcoded here. If they change in tags csv, must be changed here.\\n query = f\\\"\\\"\\\"\\n MATCH (user:Person {{email: '{user_email}'}})\\n OPTIONAL MATCH (user)-->(skill_tag:Tag:CanHelpWithTag)\\n OPTIONAL MATCH (user)-->(passion_tag:Tag:PassionsTag)\\n RETURN user, COLLECT(DISTINCT skill_tag.name) AS help_others, COLLECT(DISTINCT passion_tag.name) AS passions\\\"\\\"\\\"\\n return tx.run(query)\",\n \"def parse_person(person):\\n parsed = person_parser(person)\\n if not parsed:\\n parsed = person_parser_only_name(person)\\n name = parsed.group('name')\\n email = None\\n else:\\n name = parsed.group('name')\\n email = parsed.group('email')\\n\\n return name, email\",\n \"def get_tags(x):\\n poster = x['_source']['user']['id']\\n for user in x['_source']['users_in_photo']:\\n if user['user']['id'] != poster:\\n yield (poster, user['user']['id'])\"\n]","isConversational":false},"negative_scores":{"kind":"list like","value":["0.5984887","0.5866291","0.5816416","0.57498","0.57427794","0.56972706","0.5674774","0.5634183","0.5608683","0.5608683","0.5608683","0.5608683","0.5608683","0.56065565","0.56022114","0.55869055","0.55849683","0.5581752","0.5581752","0.5576589","0.5564691","0.5544455","0.5471089","0.5328544","0.5284833","0.5251685","0.5229503","0.52277327","0.5227208","0.515475","0.5141442","0.51396847","0.51209074","0.50916505","0.5091154","0.5068153","0.50406426","0.5002337","0.4991865","0.49857992","0.49806345","0.49737516","0.49589115","0.4956012","0.4950601","0.49287212","0.4927015","0.48997182","0.4894269","0.4888491","0.4869861","0.48611107","0.48459375","0.48374268","0.4835791","0.48224494","0.48135632","0.47998577","0.47934458","0.47887626","0.47775355","0.47629088","0.4761683","0.4747417","0.4734717","0.4732464","0.47312027","0.4722337","0.47149828","0.47126278","0.46805844","0.46785206","0.4675827","0.46747628","0.4648504","0.46458644","0.4642029","0.46388945","0.46355167","0.46277887","0.4625336","0.46234548","0.46226332","0.46125147","0.4602479","0.45985332","0.45947742","0.45936242","0.45890263","0.45851415","0.45807093","0.4578852","0.4574745","0.45636302","0.45607746","0.45563284","0.45541897","0.45438915","0.45366102","0.45311308"],"string":"[\n \"0.5984887\",\n \"0.5866291\",\n \"0.5816416\",\n \"0.57498\",\n \"0.57427794\",\n \"0.56972706\",\n \"0.5674774\",\n \"0.5634183\",\n \"0.5608683\",\n \"0.5608683\",\n \"0.5608683\",\n \"0.5608683\",\n \"0.5608683\",\n \"0.56065565\",\n \"0.56022114\",\n \"0.55869055\",\n \"0.55849683\",\n \"0.5581752\",\n \"0.5581752\",\n \"0.5576589\",\n \"0.5564691\",\n \"0.5544455\",\n \"0.5471089\",\n \"0.5328544\",\n \"0.5284833\",\n \"0.5251685\",\n \"0.5229503\",\n \"0.52277327\",\n \"0.5227208\",\n \"0.515475\",\n \"0.5141442\",\n \"0.51396847\",\n \"0.51209074\",\n \"0.50916505\",\n \"0.5091154\",\n \"0.5068153\",\n \"0.50406426\",\n \"0.5002337\",\n \"0.4991865\",\n \"0.49857992\",\n \"0.49806345\",\n \"0.49737516\",\n \"0.49589115\",\n \"0.4956012\",\n \"0.4950601\",\n \"0.49287212\",\n \"0.4927015\",\n \"0.48997182\",\n \"0.4894269\",\n \"0.4888491\",\n \"0.4869861\",\n \"0.48611107\",\n \"0.48459375\",\n \"0.48374268\",\n \"0.4835791\",\n \"0.48224494\",\n \"0.48135632\",\n \"0.47998577\",\n \"0.47934458\",\n \"0.47887626\",\n \"0.47775355\",\n \"0.47629088\",\n \"0.4761683\",\n \"0.4747417\",\n \"0.4734717\",\n \"0.4732464\",\n \"0.47312027\",\n \"0.4722337\",\n \"0.47149828\",\n \"0.47126278\",\n \"0.46805844\",\n \"0.46785206\",\n \"0.4675827\",\n \"0.46747628\",\n \"0.4648504\",\n \"0.46458644\",\n \"0.4642029\",\n \"0.46388945\",\n \"0.46355167\",\n \"0.46277887\",\n \"0.4625336\",\n \"0.46234548\",\n \"0.46226332\",\n \"0.46125147\",\n \"0.4602479\",\n \"0.45985332\",\n \"0.45947742\",\n \"0.45936242\",\n \"0.45890263\",\n \"0.45851415\",\n \"0.45807093\",\n \"0.4578852\",\n \"0.4574745\",\n \"0.45636302\",\n \"0.45607746\",\n \"0.45563284\",\n \"0.45541897\",\n \"0.45438915\",\n \"0.45366102\",\n \"0.45311308\"\n]","isConversational":false},"document_score":{"kind":"string","value":"0.8448166"},"document_rank":{"kind":"string","value":"0"}}},{"rowIdx":93279,"cells":{"query":{"kind":"string","value":"On post(type) creation, set hash_id."},"document":{"kind":"string","value":"def on_created_post(sender, instance, created, **kwargs):\n if not created:\n return\n if not issubclass(sender, dillo.models.posts.Post):\n return\n instance.hash_id = instance.id\n instance.save()\n log.debug('Set user %s as follower of own post %i' % (instance.user, instance.id))\n follow(instance.user, instance, actor_only=False)"},"metadata":{"kind":"string","value":"{\n \"objective\": {\n \"self\": [],\n \"paired\": [],\n \"triplet\": [\n [\n \"query\",\n \"document\",\n \"negatives\"\n ]\n ]\n }\n}"},"negatives":{"kind":"list like","value":["def handler(event, _context):\n model = PostModel()\n post_id = model.create(**json.loads(event['body']))\n return dump_result({'post_id': post_id}, status_code=201)","def save(self, *args, **kwargs):\n if not self.pkhash:\n self.pkhash = compute_hash(self.script)\n super(DataOpener, self).save(*args, **kwargs)","def hash_id(self, hash_id):\n\n self._hash_id = hash_id","def set_last_post(self, post):\n with open(self.cache_filename,'w') as cache_file:\n json.dump(post['id'], cache_file)","def setHash(self):\n chash_string = str(self.code) + str(\"CAMPAIGN\") + str(self.created_at)\n chash = hashlib.sha1()\n chash.update(chash_string)\n \n self.chash = chash.hexdigest()\n self.save()","def _pre_put_hook(self) -> None:\n super()._pre_put_hook()\n self.email_hash = self._generate_hash(\n self.recipient_id, self.subject, self.html_body)","def post(self):\n blob_key = self.request.get(\"blobkey\")\n\n database_creation.run(blob_key)","def on_create(self):\n self.contentnode.make_content_id_unique()","def save_post(self, post):\n return self.collection.insert_one(post.serialize())","def post_create(self, state):\n\n self.id = self.get_flags_from_list(self.id)\n self.flags = self.get_flags_from_list(self.flags)","def post(self, post):\n\n self._post = post","def post(self, post):\n\n self._post = post","def _post(self, data):\n new_ticket_id = DB_TICKET_TABLE.insert(data)\n return new_ticket_id","def set_hash(self, hash_name, data):\n self.hashes[hash_name] = data","def hash_card_id(sender: Card.__class__, **kwargs):\n new_card = kwargs['instance']\n new_card.hash_id = sha1(str(new_card.card_id).encode('utf-8')).hexdigest()","def hash(self, hash):\n\n self._hash = hash","def hash(self, hash):\n\n self._hash = hash","def on_create(self, payload):\n pass","def store(self, hash, original_url):\n self.r.set(hash, original_url)","def hash_save(db: Redis[bytes], hash: hash_t) -> None:\n db.zadd(HASH_INDEX, {str(hash): \"0\"})","def post_save(cls: any, sender: any, document: Document, **kwargs: dict) -> None:\n is_new = kwargs.get('created')\n \n if not is_new:\n return\n\n password = document.password.encode('utf-8')\n hashed_password = hashpw(password, gensalt())\n document.password = hashed_password.decode('utf-8')\n document.save()","def set_primary_object_hash(self, hsh):\n self.hash = hsh","def add_to_db(self, data_base, image_path):\n image_data = get_binary(image_path)\n post_data = self.get_post(dictionary=True)\n post_data.pop('id')\n post_data.pop('image_id')\n cursor = data_base.cursor(dictionary=True)\n cursor.execute(\"INSERT INTO image (data) VALUES (%s)\", (image_data,))\n cursor.execute(\"SELECT id FROM image ORDER BY id DESC LIMIT 1;\")\n post_data['image_id'] = cursor.fetchone()['id']\n cursor.execute(f\"INSERT INTO post ({', '.join(post_data.keys())}) VALUES {tuple(post_data.values())}\")\n cursor.execute(\"SELECT id FROM post ORDER BY id DESC LIMIT 1;\")\n post_id = cursor.fetchone()['id']\n if self.commit_to_db:\n data_base.commit()\n cursor.close()\n return post_id","def post_database_node_create(self, resource_dict):\n pass","def post(self, record_type, record_id, record, metadata):\n \n if not self.cache.get(record_type, None):\n self.cache[record_type] = {}\n\n if not self.cache[record_type].get(record_id, None):\n self.cache[record_type][record_id] = {}\n\n \n self.cache[record_type][record_id]['record'] = record\n self.cache[record_type][record_id]['metadata'] = metadata\n\n \n d = Date()\n self.cache[record_type][record_id]['last_updated'] = d.now()\n\n # Check space, remove old items if not enough space","def post_analytics_node_create(self, resource_dict):\n pass","def _post(self, data):\n new_comment_id = DB_COMMENT_TABLE.insert(data)\n return new_comment_id","def create_definition(self, block_type, slug=None):\n prefix = \"d\"\n if slug:\n prefix += \"_\" + slug\n def_id = self._next_id(prefix)\n self._definitions[def_id] = block_type\n return def_id","def __init__(self):\n self.id = uuid.uuid4().hex","def on_update(self):\n self.contentnode.make_content_id_unique()","def _post(self, data):\n new_user_id = DB_USER_TABLE.insert(data)\n return new_user_id","def save(self):\n\n data = super().save('name, type', self.name, self.type)\n\n self.id = data.get('id')\n return data","def create(self, validated_data):\n \"\"\" Create post with a location \"\"\"\n location_data = validated_data.pop('location')\n\n # create a new one or get a old for reference\n this_location = Location.objects.get_or_create(\n **location_data\n )\n\n # pop the photo url's data\n photo_data = validated_data.pop('photo')\n\n # must pop the tags data before it would used to create a post \n tags_data = validated_data.pop('tag')\n # create a instance of this post\n this_post = Post.objects.create(\n location = this_location[0],\n **validated_data)\n\n \"\"\"Associate tag's informatiion to post\"\"\"\n for tag in tags_data:\n this_tag = Tag.objects.get_or_create(name = tag.get('name'))\n print(tag.get('name'))\n print(this_tag)\n # attach this tag to this photos_datapost \n this_post.tag.add(this_tag[0])\n\n \"\"\"Associate the photo url \"\"\"\n for photo in photo_data:\n this_post.photo.create(name = photo.get('name'))\n # return the created post \n this_post.save()\n return this_post","def create(self, identity, record=None, data=None, **kwargs):\n self.set_slug(record, data[\"slug\"])","def save(self):\n if not self.id:\n self.id = uuid4()\n DataStore.add_instance(self)","def post(self, post_id=None):\n\n if post_id:\n abort(400)\n else:\n args = parsers.post_post_parser.parse_args(strict=True)\n\n new_post = Post(args['title'])\n new_post.text = args['text']\n # new_post.user = user\n\n if args['tags']:\n for item in args['tags']:\n tag = Tag.query.filter_by(name=item).first()\n # If the tag already exist, append.\n if tag:\n new_post.tags.append(tag)\n # If the tag not exist, create the new one.\n # Will be write into DB with session do.\n else:\n new_tag = Tag(item)\n new_post.tags.append(new_tag)\n db.session.add(new_post)\n db.session.commit()\n return (new_post.id, 201)","def postTweet(self, userId: int, tweetId: int) -> None:\n self.cnt += 1\n self.posts[userId].append((self.cnt, tweetId))","def newPost(self, useRawHTML):\n print\n content, publish = self._fillPost(useRawHTML)\n\n # Upload to server\n try :\n postid = self.server.metaWeblog.newPost(\n self.blogid, self.username, self.password,\n content, publish\n )\n except xmlrpclib.Fault as fault:\n display_XMLRPC_errors(\"post the new entry\", fault)\n import pdb\n pdb.set_trace()\n else :\n self._setCategorie(postid)\n print \"New post created with ID =\", postid","def _hash_and_save(self, node):\n validate_is_bin_node(node)\n\n node_hash = keccak(node)\n self.db[node_hash] = node\n return node_hash","def post(self, body):\n return self.objects.create(body)","def addPost(postCreatedEvent):\n query = generatePostSQLQuery(postCreatedEvent)\n with conn.cursor() as cur:\n cur.execute(query)\n conn.commit()","def postTweet(self, userId: int, tweetId: int) -> None:\n ts = time.time()\n self.posts[userId].append((ts, tweetId))","def insert_new_post(post_arg_set):\n api, post_data, acct_data, page_id, config = post_arg_set\n\n try:\n post_id = post_data['id'] if post_data.has_key('id') else None\n\n except Exception as e:\n log.error( e )\n\n else:\n\n # parse date\n if post_data.has_key('created_time') and post_data['created_time'] is not None: \n dt = datetime.strptime(post_data['created_time'], FB_DATE_FORMAT)\n date_time = tz_adj(dt, config)\n time_bucket = round_datetime(date_time, config)\n raw_timestamp = int(date_time.strftime(\"%s\"))\n \n else:\n time_bucket = None\n raw_timestamp = None\n \n # extract message so we can find links within the msg if not in url\n article_urls = [get_fb_link(post_data, config, unshorten=True)]\n message = post_data['message'].encode('utf-8') if post_data.has_key('message') else None\n message_urls = get_message_urls(article_urls, message, config)\n\n # detect article links, unshorten and parse\n article_urls = [\n parse_url(unshorten_link(url, config)) \\\n for url in article_urls + message_urls\n if url is not None\n ]\n\n article_urls = [url for url in article_urls if is_article(url, config)]\n\n if article_urls:\n for article_url in set(article_urls):\n\n # sluggify url\n article_slug = sluggify(article_url)\n\n # format data\n post_value = {\n 'article_slug': article_slug,\n 'article_url': article_url,\n 'time_bucket': time_bucket,\n 'fb_post_created': raw_timestamp,\n 'raw_timestamp': raw_timestamp,\n 'fb_raw_link' : get_fb_link(post_data, config=config),\n 'fb_page_id': page_id,\n 'fb_post_id': post_id,\n 'fb_page_likes': acct_data['likes'] if acct_data.has_key('likes') else None,\n 'fb_page_talking_about': acct_data['talking_about_count'] if acct_data.has_key('talking_about_count') else None,\n 'fb_type': post_data['type'] if post_data.has_key('type') else None,\n 'fb_status_type': post_data['status_type'] if post_data.has_key('status_type') else None,\n 'fb_message': message\n }\n \n # always insert insights data\n if is_insights(page_id, config):\n \n log.info( \"INSIGHTS\\tAdding data from %s re: %s\" % (page_id, article_slug) )\n\n # fetch data\n insights_value = get_insights_data(api, page_id, post_id)\n\n # create datasource name\n data_source = \"facebook_insights_%s\" % page_id \n \n # upsert url\n upsert_url(article_url, article_slug, data_source, config)\n\n # insert id\n db.sadd('facebook_post_ids', post_id)\n\n # format time bucket\n current_time_bucket = gen_time_bucket(config)\n insights_value['time_bucket'] = current_time_bucket\n post_value.pop('time_bucket', None)\n \n value = json.dumps({\n data_source : dict(post_value.items() + insights_value.items())\n })\n\n # upload data to redis\n db.zadd(article_slug, current_time_bucket, value) \n \n # only insert new posts\n if not db.sismember('facebook_post_ids', post_id):\n \n log.info( \"FACEBOOK\\tNew post %s\\t%s\" % (post_id, article_url) )\n \n # insert id\n db.sadd('facebook_post_ids', post_id) \n \n # upsert url\n data_source = \"facebook_%s\" % page_id\n upsert_url(article_url, article_slug, data_source, config)\n\n value = json.dumps( {data_source : post_value} )\n\n\n # upload data to redis\n db.zadd(article_slug, time_bucket, value)","def create_note_meta_data(sender, instance, created, **kwargs):\n if created:\n NoteMetaData.objects.create(note=instance)","def create_db_post(entry, keys, like):\n h = get_hash(entry['link'])\n collection = pos if like else neg\n return collection.update(\n {'hash': h},\n {\n 'link': entry['link'],\n 'title': entry['title'],\n 'published': '',\n 'content': \" \".join(keys),\n 'hash': h,\n 'read': False\n }, upsert=True\n )","def add_post(self, post: Post) -> None:\n self.post_process.append(post)","def save(self, data):\n data['id'] = self.id\n\n self.db.append(data)","def instance(data):\n return Fieldsha1(data)","def post_kw_view_func(kw, obj, obj_type, obj_id, **kwargs):\n kw = default_view_kw(kw, obj.thread, obj_type, obj_id, **kwargs)\n kw['thread_id'] = obj.thread_id\n kw['_anchor'] = u'post_{:d}'.format(obj.id)\n return kw","def process_IN_CREATE(self, event):","def __init__(self, post):\n self.id = int(post[0])\n self.type_id = int(post[1])\n try:\n self.parent_id = int(post[2])\n except:\n self.parent_id = None\n try:\n self.accepted_answer_id = int(post[3])\n except:\n self.accepted_answer_id = None\n self.creation_date = post[4]\n try:\n self.score = post[5]\n except:\n self.score = None\n try:\n self.view_count = post[6]\n except:\n self.view_count = 0\n self.body = post[7]\n try:\n self.owner_user_id = int(post[8])\n except:\n self.owner_user_id = None \n try:\n self.last_editor_user_id = int(post[9])\n except:\n self.last_editor_user_id = None\n try:\n self.last_editor_display_name = post[10]\n except:\n self.last_editor_display_name = None\n self.last_activity_date = post[11]\n self.last_edit_date = post[12]\n self.community_owned_date = post[13]\n self.closed_date = post[14]\n try:\n self.title = post[15]\n except:\n self.title = None\n try:\n self.tags = post[16]\n except:\n self.tags = None\n try:\n self.answer_count = int(post[17])\n except:\n self.answer_count = 0\n try:\n self.comment_count = int(post[18])\n except:\n self.comment_count = 0\n try:\n self.favorite_count = int(post[19])\n except:\n self.favorite_count = 0","def save_current_post(entry):\n return current.insert_one(entry).inserted_id","def get_post(self):\n\t\tself.post = graph.get_object(POST_ID)","def save(self,**kwargs):\n if not self.id:\n self.created_at = datetime.now()\n super(TestUrl, self).save(**kwargs)","def postTweet(self, userId: int, tweetId: int) -> None:\n self.users.add(userId)\n self.user_post[userId].append((tweetId, self.time))\n self.time += 1","def set_new_hash(court_id, new_hash):\n seals_data[court_id]['hash'] = new_hash","def postTweet(self, userId, tweetId):\n if not self.user_pool[userId].user_id:\n user = UserInfo()\n user.user_id = userId\n self.user_pool[userId] = user\n self.time += 1\n tw_info = (tweetId, self.time) # 保存一个和时间戳\n self.twitter_pool[userId].append(tw_info)","def __init__(self):\n self.id = id(self)","def save(self, *args, **kwargs):\n if not self.post_id:\n self.created = timezone.now()\n dup = StoryPost.objects.filter(title=self.title)\n if len(dup) > 0:\n # objects with the same slug exist -> duplicate!\n nos = str(len(dup))\n # append number of duplicates as modifier\n self.slug = slugify(self.title[:49 - len(dup)] + '-' + nos)\n else:\n self.slug = slugify(self.title[:50])\n self.modified = timezone.now()\n\n return super(StoryPost, self).save(*args, **kwargs)","def add(self, obj):\n ID = id(obj)\n self.pDict[ID] = obj\n return ID","def save(self, *args, **kwargs):\n\n if not self.trackerid:\n self.trackerid = generate_trackerid()\n super(Profile, self).save(*args, **kwargs)","def store(self, dikt):\n if self.key is not None and self.body is not None:\n dikt[self.key] = self\n elif self.key in dikt:\n del dikt[self.key]","def setTypeId(self, x, y, z, type_):\n state = self.world.getBlockAt(x, y, z).getState()\n state.setTypeId(type_)\n self.list_.add(state)","def set_hasher(self, hasher_type):\n if hasher_type == 'sha256':\n self.hasher = hashlib.sha256()\n self.hasher_size = 64\n elif hasher_type == 'md5':\n self.hasher = hashlib.md5()\n self.hasher_size = 32\n else:\n raise ValueError(\n \"Unknown hasher_type for checksum: {}\".format(hasher_type))","def create_definition(self, block_type, slug=None):\n raise NotImplementedError()","def __hash__(self):\n return self['id'].__hash__()","def hash_code(sender: Key.__class__, **kwargs):\n new_key = kwargs['instance']\n new_key.code = randint(MIN_CODE, MAX_CODE)\n new_key.hash_code = sha1(str(new_key.code).encode('utf-8')).hexdigest()","def create_post():\n\n #Get prompt id\n prompt_id = request.form.get('prompt_id')\n\n # Get post text\n post_text = request.form.get('user_post')\n\n # Create post timestamp\n created_at = datetime.now()\n user_facing_date = created_at.strftime(\"%B %d, %Y\")\n\n # Save post and related data to database\n post = crud.create_post(session['user_id'], prompt_id, post_text, session['lat'], session['lng'], session['user_facing_location'], created_at)\n\n return render_template('post_data.html', post=post, user_facing_date=user_facing_date)","def save(self, *args, **kwargs):\r\n\r\n if not self.trackerid:\r\n self.trackerid = generate_trackerid()\r\n super(Profile, self).save(*args, **kwargs)","def save(self, **kwargs):\n if not self.id and not self.slug:\n self.set_unique_slug()\n\n super().save(**kwargs)","def post(self, args):\n\n try:\n base64.b64decode(\n args[\"cap\"]\n ) # decode to ensure valid base64 (but don't use it)\n except:\n return {\"status\": \"not valid base64\"}, 400\n\n got_hash = Hash.query.filter_by(cap=args[\"cap\"]).first()\n\n if got_hash is None:\n got_hash = Hash(args[\"cap\"])\n\n db.session.add(got_hash)\n db.session.commit()\n\n return (\n {\"status\": \"success\", \"body\": {\"hash\": hash_schema.dump(got_hash)}},\n 200,\n ) # make new hash\n elif len(got_hash.jobs) != 0:\n return (\n {\"status\": \"completed\", \"body\": {\"hash\": hash_schema.dump(got_hash)}},\n 200,\n ) # if hash has finished jobs\n else:\n return (\n {\"status\": \"queued\", \"body\": {\"hash\": hash_schema.dump(got_hash)}},\n 202,\n ) # if hash is already queued","def handle_new_post(post_data, user_agent, remote_addr):\n \n for required in POST_REQUIRED_PARAMS:\n if required not in post_data:\n return None, None\n\n try:\n value = int(string_from_interwebs(post_data.getfirst(\"code\", \"\")))\n except ValueError:\n return None, None\n \n if value != 98098098098:\n return None, None\n\n # not yet safe to use.\n location = post_data.getfirst(\"location\", \"\")\n tags = string_from_interwebs(post_data.getfirst(\"tags\")) \n author = post_data.getfirst(\"author\")\n \n split_tags = [string_from_interwebs(tag).strip().lower() for tag in tags.split(\",\")] # temporary\n \n if len(split_tags) > 3:\n return None, None\n \n author_id = string_from_interwebs(author).strip()\n \n with Connection('localhost', 27017) as connection:\n reply_to = string_from_interwebs(post_data.getfirst(\"reply_to\"))\n \n if not verify_author(author_id, connection):\n return None, None\n\n if not verify_post(reply_to, connection):\n return None, None\n\n # if reply then it's verified.\n # XXX: I need to make a standard object structure for this, so that I don't \n # have to update separate things.\n\n post = {\"viewed\" : 0,\n \"comments\" : 0,\n \"flagged\" : 0,\n \"disliked\" : 0,\n \"enjoyed\" : 0,\n \"num_replies\" : 0,\n \"num_reposts\" : 0,\n \"content-type\" : \"image\", # need to pull this from the mime lookup\n \"file\" : \"placeholder\",\n \"user_agent\" : user_agent,\n \"remote_addr\" : remote_addr,\n \"created\" : datetime.utcnow(),\n \"location\" : string_from_interwebs(location).strip(),\n \"author\" : ObjectId(author_id),\n \"reply_to\" : ObjectId(reply_to),\n \"tags\" : split_tags}\n\n update_post(reply_to, connection)\n\n return post_data.getfirst(\"data\"), post","def save(self, *args, **kwargs):\n created = not self.id\n if not self.digest:\n self.digest = self.get_digest()\n if not self.cover_img:\n self.cover_img = getattr(settings, 'WEIXIN_DEFAULT_COVER')\n self.fileid = getattr(settings, 'WEIXIN_DEFAULT_COVER_ID')\n super(WeixinMp, self).save(*args, **kwargs)\n\n if self.sync and created:\n from sae.taskqueue import add_task\n task_link = reverse('weixinmp.upload', args=(self.pk,))\n add_task('weixin', task_link, delay=300)\n task_link = reverse('weixinmp.image_collect', args=(self.pk,))\n add_task('weixin', task_link, delay=180)\n\n if created:\n event = Event(user=self.user, content_object=self)\n event.save()","def save(self, *args, **kwargs):\n if not self.id:\n self.created = timezone.now()\n\n self.modified = timezone.now()\n\n return super(CodeSnippet, self).save(*args, **kwargs)","def hash_lock_id(sender: Lock.__class__, **kwargs):\n new_lock = kwargs['instance']\n new_lock.hash_id = sha1(str(new_lock.uuid).encode('utf-8')).hexdigest()","def post(self, uid):\n json_data = request.get_json()\n b = event_builders.FbEventBuilder()\n b.build_with_fb_dict(json_data)\n e: event_models.SocialEvent = b.export_as_class(event_models.SocialEvent)\n\n # Note that e.firestore_ref will not be set by create()\n ref: DocumentReference = EventDao().create_fb_event(e)\n e.set_firestore_ref(ref)\n dict_view = e.to_dict_view()\n dict_view[\"eventId\"] = ref.id\n\n # TODO: add error handling\n UserDao().add_user_event_dict(uid, dict_view[\"fbEventId\"], dict_view)\n\n return {\n \"id\": e.get_firestore_ref().id\n }","def _set_id(self):\n raise NotImplementedError()","def create(self):\n ...","def post_routing_instance_create(self, resource_dict):\n pass","def save(self, *args, **kwargs):\n if not self.identifier:\n if self.event is None:\n self.identifier = self.email.lower()\n else:\n self.identifier = \"%s@%s.event.pretix\" % (self.username.lower(), self.event.id)\n if not self.pk:\n self.identifier = self.identifier.lower()\n super().save(*args, **kwargs)","def __init__(self, memo, tags=()):\n\n self.memo = memo\n self.tags = tags\n self.creation_date = datetime.date.today()\n global last_id\n last_id += 1\n self.id = last_id","def post(self, copy_id):\n checkCopyValidity(copy_id)\n note_body = request.get_json()\n new_note = models.Notes()\n new_note.parse_body(note_body)\n new_note.copy_id = copy_id\n db.session.add(new_note)\n db.session.commit()\n return 'A note \\\"{}\\\" has been added to book copy of {}'.format(new_note.note, copy_id), 201","def _set_unique_id(self, json_request):\n values = []\n for field in value_fields:\n value = json_request.get(field, '')\n values.append(quote(self.fully_decode_uri(value), safe=''))\n if len(values) == 1:\n self.unique_id = values[0]\n elif len(values) == 2:\n self.unique_id = self.build_summary(values[0], values[1])\n elif len(values) == 1:\n self.unique_id = self.build_summary(values[0], values[1], values[2])","def __addPost(self, post, is_question = False):\n try:\n post_id = post.find('a',id=re.compile('postcount\\d+')) \n unique_key = post_id['id'].replace('postcount','')\n log.info(unique_key)\n if checkSessionInfo(self.genre, self.session_info_out, unique_key,\\\n self.task.instance_data.get('update'),parent_list\\\n = [self.task.instance_data['uri']]):\n log.info(self.log_msg('Session info returns True'))\n return False\n page = self.__getData(post, is_question)\n except:\n log.info(post)\n log.exception(self.log_msg('Cannot add the post for the url %s'%\\\n self.currenturi))\n return False\n try:\n page['uri'] = self.currenturi + '#' + stripHtml(post_id.renderContents()) \n except:\n log.info(self.log_msg('Cannot find the uri'))\n page['uri'] = self.currenturi\n try:\n result=updateSessionInfo(self.genre, self.session_info_out, unique_key, \\\n get_hash( page ),'Review', self.task.instance_data.get('update'),\\\n parent_list=[self.task.instance_data['uri']])\n if not result['updated']:\n log.info(self.log_msg('Update session info returns False'))\n return True\n page['parent_path'] = page['path'] = [self.task.instance_data['uri']]\n page['path'].append(unique_key)\n #page['uri'] = unique_key\n page['uri_domain'] = urlparse.urlparse(page['uri'])[1]\n page.update(self.__task_elements_dict)\n self.pages.append(page)\n log.info(self.log_msg('Post Added'))\n return True\n except:\n log.exception(self.log_msg('Error while adding session info'))\n return False","def handle_post(cls, **kwargs):\n raise NotImplementedError","def calculate_hash_id(self):\n return get_md5_hash(f'{self.type}{self.get_primary_id()}')","def __init__(self, rpc, post):\n super(PostModule, self).__init__(rpc, 'post', post)\n self._action = self._info.get('default_action', \"\")","def _calculate_hash(self, entry):\n entry.pop('id', None)\n return hashlib.sha224(json.dumps(\n entry, cls=DjangoJSONEncoder).encode('utf-8')).hexdigest()","def __hash__(self):\n return hash(self.id)","def __hash__(self):\n return hash(self.id)","def __hash__(self):\n return hash(self.id)","def __hash__(self):\n return hash(self.id)","def post_to_activity(self, post):\n id = None\n if post.get('id'):\n # strip USERID_ prefix if it's there\n post['id'] = post['id'].split('_', 1)[-1]\n id = post['id']\n\n obj = self.post_to_object(post)\n activity = {\n 'verb': VERBS.get(post.get('type', obj.get('objectType')), 'post'),\n 'published': obj.get('published'),\n 'updated': obj.get('updated'),\n 'id': self.tag_uri(id) if id else None,\n 'url': self.post_url(post),\n 'actor': obj.get('author'),\n 'object': obj,\n }\n\n application = post.get('application')\n if application:\n activity['generator'] = {\n 'displayName': application.get('name'),\n 'id': self.tag_uri(application.get('id')),\n }\n return self.postprocess_activity(activity)","def regist_hash(cobj, hash, handler, dir):\n pass","def perform_create(self,serializer):\n serializer.save(user_id=self.request.user)","def insert_hash(self, title, artist, song_id:int, fingerprint: str, offset:int):\n fingerprint = Fingerprints(song_id=song_id, song_title=title, artist=artist, hash=fingerprint, offset=offset)\n fingerprint.save()","def post(self, request, *args, **kwargs):\n return super().create(request, *args, **kwargs)","def write_post(id, args):\n graph = facebook.GraphAPI(id) \n graph.put_object(parent_object='me', connection_name = 'feed', **args)","def _set_hash(self, *args):\n if self.df is None:\n raise ValueError(\"df is None\")\n self._hash_df = get_df_hash(self.df)\n self._hash_args = \", \".join([str(arg) for arg in args])\n self._hash=\"{klass:}({df_hash:}, {args_hash:})\".format(\n klass = self.__class__.__name__,\n df_hash = self._hash_df,\n args_hash = self._hash_args,\n )","def post_provider_attachment_create(self, resource_dict):\n pass"],"string":"[\n \"def handler(event, _context):\\n model = PostModel()\\n post_id = model.create(**json.loads(event['body']))\\n return dump_result({'post_id': post_id}, status_code=201)\",\n \"def save(self, *args, **kwargs):\\n if not self.pkhash:\\n self.pkhash = compute_hash(self.script)\\n super(DataOpener, self).save(*args, **kwargs)\",\n \"def hash_id(self, hash_id):\\n\\n self._hash_id = hash_id\",\n \"def set_last_post(self, post):\\n with open(self.cache_filename,'w') as cache_file:\\n json.dump(post['id'], cache_file)\",\n \"def setHash(self):\\n chash_string = str(self.code) + str(\\\"CAMPAIGN\\\") + str(self.created_at)\\n chash = hashlib.sha1()\\n chash.update(chash_string)\\n \\n self.chash = chash.hexdigest()\\n self.save()\",\n \"def _pre_put_hook(self) -> None:\\n super()._pre_put_hook()\\n self.email_hash = self._generate_hash(\\n self.recipient_id, self.subject, self.html_body)\",\n \"def post(self):\\n blob_key = self.request.get(\\\"blobkey\\\")\\n\\n database_creation.run(blob_key)\",\n \"def on_create(self):\\n self.contentnode.make_content_id_unique()\",\n \"def save_post(self, post):\\n return self.collection.insert_one(post.serialize())\",\n \"def post_create(self, state):\\n\\n self.id = self.get_flags_from_list(self.id)\\n self.flags = self.get_flags_from_list(self.flags)\",\n \"def post(self, post):\\n\\n self._post = post\",\n \"def post(self, post):\\n\\n self._post = post\",\n \"def _post(self, data):\\n new_ticket_id = DB_TICKET_TABLE.insert(data)\\n return new_ticket_id\",\n \"def set_hash(self, hash_name, data):\\n self.hashes[hash_name] = data\",\n \"def hash_card_id(sender: Card.__class__, **kwargs):\\n new_card = kwargs['instance']\\n new_card.hash_id = sha1(str(new_card.card_id).encode('utf-8')).hexdigest()\",\n \"def hash(self, hash):\\n\\n self._hash = hash\",\n \"def hash(self, hash):\\n\\n self._hash = hash\",\n \"def on_create(self, payload):\\n pass\",\n \"def store(self, hash, original_url):\\n self.r.set(hash, original_url)\",\n \"def hash_save(db: Redis[bytes], hash: hash_t) -> None:\\n db.zadd(HASH_INDEX, {str(hash): \\\"0\\\"})\",\n \"def post_save(cls: any, sender: any, document: Document, **kwargs: dict) -> None:\\n is_new = kwargs.get('created')\\n \\n if not is_new:\\n return\\n\\n password = document.password.encode('utf-8')\\n hashed_password = hashpw(password, gensalt())\\n document.password = hashed_password.decode('utf-8')\\n document.save()\",\n \"def set_primary_object_hash(self, hsh):\\n self.hash = hsh\",\n \"def add_to_db(self, data_base, image_path):\\n image_data = get_binary(image_path)\\n post_data = self.get_post(dictionary=True)\\n post_data.pop('id')\\n post_data.pop('image_id')\\n cursor = data_base.cursor(dictionary=True)\\n cursor.execute(\\\"INSERT INTO image (data) VALUES (%s)\\\", (image_data,))\\n cursor.execute(\\\"SELECT id FROM image ORDER BY id DESC LIMIT 1;\\\")\\n post_data['image_id'] = cursor.fetchone()['id']\\n cursor.execute(f\\\"INSERT INTO post ({', '.join(post_data.keys())}) VALUES {tuple(post_data.values())}\\\")\\n cursor.execute(\\\"SELECT id FROM post ORDER BY id DESC LIMIT 1;\\\")\\n post_id = cursor.fetchone()['id']\\n if self.commit_to_db:\\n data_base.commit()\\n cursor.close()\\n return post_id\",\n \"def post_database_node_create(self, resource_dict):\\n pass\",\n \"def post(self, record_type, record_id, record, metadata):\\n \\n if not self.cache.get(record_type, None):\\n self.cache[record_type] = {}\\n\\n if not self.cache[record_type].get(record_id, None):\\n self.cache[record_type][record_id] = {}\\n\\n \\n self.cache[record_type][record_id]['record'] = record\\n self.cache[record_type][record_id]['metadata'] = metadata\\n\\n \\n d = Date()\\n self.cache[record_type][record_id]['last_updated'] = d.now()\\n\\n # Check space, remove old items if not enough space\",\n \"def post_analytics_node_create(self, resource_dict):\\n pass\",\n \"def _post(self, data):\\n new_comment_id = DB_COMMENT_TABLE.insert(data)\\n return new_comment_id\",\n \"def create_definition(self, block_type, slug=None):\\n prefix = \\\"d\\\"\\n if slug:\\n prefix += \\\"_\\\" + slug\\n def_id = self._next_id(prefix)\\n self._definitions[def_id] = block_type\\n return def_id\",\n \"def __init__(self):\\n self.id = uuid.uuid4().hex\",\n \"def on_update(self):\\n self.contentnode.make_content_id_unique()\",\n \"def _post(self, data):\\n new_user_id = DB_USER_TABLE.insert(data)\\n return new_user_id\",\n \"def save(self):\\n\\n data = super().save('name, type', self.name, self.type)\\n\\n self.id = data.get('id')\\n return data\",\n \"def create(self, validated_data):\\n \\\"\\\"\\\" Create post with a location \\\"\\\"\\\"\\n location_data = validated_data.pop('location')\\n\\n # create a new one or get a old for reference\\n this_location = Location.objects.get_or_create(\\n **location_data\\n )\\n\\n # pop the photo url's data\\n photo_data = validated_data.pop('photo')\\n\\n # must pop the tags data before it would used to create a post \\n tags_data = validated_data.pop('tag')\\n # create a instance of this post\\n this_post = Post.objects.create(\\n location = this_location[0],\\n **validated_data)\\n\\n \\\"\\\"\\\"Associate tag's informatiion to post\\\"\\\"\\\"\\n for tag in tags_data:\\n this_tag = Tag.objects.get_or_create(name = tag.get('name'))\\n print(tag.get('name'))\\n print(this_tag)\\n # attach this tag to this photos_datapost \\n this_post.tag.add(this_tag[0])\\n\\n \\\"\\\"\\\"Associate the photo url \\\"\\\"\\\"\\n for photo in photo_data:\\n this_post.photo.create(name = photo.get('name'))\\n # return the created post \\n this_post.save()\\n return this_post\",\n \"def create(self, identity, record=None, data=None, **kwargs):\\n self.set_slug(record, data[\\\"slug\\\"])\",\n \"def save(self):\\n if not self.id:\\n self.id = uuid4()\\n DataStore.add_instance(self)\",\n \"def post(self, post_id=None):\\n\\n if post_id:\\n abort(400)\\n else:\\n args = parsers.post_post_parser.parse_args(strict=True)\\n\\n new_post = Post(args['title'])\\n new_post.text = args['text']\\n # new_post.user = user\\n\\n if args['tags']:\\n for item in args['tags']:\\n tag = Tag.query.filter_by(name=item).first()\\n # If the tag already exist, append.\\n if tag:\\n new_post.tags.append(tag)\\n # If the tag not exist, create the new one.\\n # Will be write into DB with session do.\\n else:\\n new_tag = Tag(item)\\n new_post.tags.append(new_tag)\\n db.session.add(new_post)\\n db.session.commit()\\n return (new_post.id, 201)\",\n \"def postTweet(self, userId: int, tweetId: int) -> None:\\n self.cnt += 1\\n self.posts[userId].append((self.cnt, tweetId))\",\n \"def newPost(self, useRawHTML):\\n print\\n content, publish = self._fillPost(useRawHTML)\\n\\n # Upload to server\\n try :\\n postid = self.server.metaWeblog.newPost(\\n self.blogid, self.username, self.password,\\n content, publish\\n )\\n except xmlrpclib.Fault as fault:\\n display_XMLRPC_errors(\\\"post the new entry\\\", fault)\\n import pdb\\n pdb.set_trace()\\n else :\\n self._setCategorie(postid)\\n print \\\"New post created with ID =\\\", postid\",\n \"def _hash_and_save(self, node):\\n validate_is_bin_node(node)\\n\\n node_hash = keccak(node)\\n self.db[node_hash] = node\\n return node_hash\",\n \"def post(self, body):\\n return self.objects.create(body)\",\n \"def addPost(postCreatedEvent):\\n query = generatePostSQLQuery(postCreatedEvent)\\n with conn.cursor() as cur:\\n cur.execute(query)\\n conn.commit()\",\n \"def postTweet(self, userId: int, tweetId: int) -> None:\\n ts = time.time()\\n self.posts[userId].append((ts, tweetId))\",\n \"def insert_new_post(post_arg_set):\\n api, post_data, acct_data, page_id, config = post_arg_set\\n\\n try:\\n post_id = post_data['id'] if post_data.has_key('id') else None\\n\\n except Exception as e:\\n log.error( e )\\n\\n else:\\n\\n # parse date\\n if post_data.has_key('created_time') and post_data['created_time'] is not None: \\n dt = datetime.strptime(post_data['created_time'], FB_DATE_FORMAT)\\n date_time = tz_adj(dt, config)\\n time_bucket = round_datetime(date_time, config)\\n raw_timestamp = int(date_time.strftime(\\\"%s\\\"))\\n \\n else:\\n time_bucket = None\\n raw_timestamp = None\\n \\n # extract message so we can find links within the msg if not in url\\n article_urls = [get_fb_link(post_data, config, unshorten=True)]\\n message = post_data['message'].encode('utf-8') if post_data.has_key('message') else None\\n message_urls = get_message_urls(article_urls, message, config)\\n\\n # detect article links, unshorten and parse\\n article_urls = [\\n parse_url(unshorten_link(url, config)) \\\\\\n for url in article_urls + message_urls\\n if url is not None\\n ]\\n\\n article_urls = [url for url in article_urls if is_article(url, config)]\\n\\n if article_urls:\\n for article_url in set(article_urls):\\n\\n # sluggify url\\n article_slug = sluggify(article_url)\\n\\n # format data\\n post_value = {\\n 'article_slug': article_slug,\\n 'article_url': article_url,\\n 'time_bucket': time_bucket,\\n 'fb_post_created': raw_timestamp,\\n 'raw_timestamp': raw_timestamp,\\n 'fb_raw_link' : get_fb_link(post_data, config=config),\\n 'fb_page_id': page_id,\\n 'fb_post_id': post_id,\\n 'fb_page_likes': acct_data['likes'] if acct_data.has_key('likes') else None,\\n 'fb_page_talking_about': acct_data['talking_about_count'] if acct_data.has_key('talking_about_count') else None,\\n 'fb_type': post_data['type'] if post_data.has_key('type') else None,\\n 'fb_status_type': post_data['status_type'] if post_data.has_key('status_type') else None,\\n 'fb_message': message\\n }\\n \\n # always insert insights data\\n if is_insights(page_id, config):\\n \\n log.info( \\\"INSIGHTS\\\\tAdding data from %s re: %s\\\" % (page_id, article_slug) )\\n\\n # fetch data\\n insights_value = get_insights_data(api, page_id, post_id)\\n\\n # create datasource name\\n data_source = \\\"facebook_insights_%s\\\" % page_id \\n \\n # upsert url\\n upsert_url(article_url, article_slug, data_source, config)\\n\\n # insert id\\n db.sadd('facebook_post_ids', post_id)\\n\\n # format time bucket\\n current_time_bucket = gen_time_bucket(config)\\n insights_value['time_bucket'] = current_time_bucket\\n post_value.pop('time_bucket', None)\\n \\n value = json.dumps({\\n data_source : dict(post_value.items() + insights_value.items())\\n })\\n\\n # upload data to redis\\n db.zadd(article_slug, current_time_bucket, value) \\n \\n # only insert new posts\\n if not db.sismember('facebook_post_ids', post_id):\\n \\n log.info( \\\"FACEBOOK\\\\tNew post %s\\\\t%s\\\" % (post_id, article_url) )\\n \\n # insert id\\n db.sadd('facebook_post_ids', post_id) \\n \\n # upsert url\\n data_source = \\\"facebook_%s\\\" % page_id\\n upsert_url(article_url, article_slug, data_source, config)\\n\\n value = json.dumps( {data_source : post_value} )\\n\\n\\n # upload data to redis\\n db.zadd(article_slug, time_bucket, value)\",\n \"def create_note_meta_data(sender, instance, created, **kwargs):\\n if created:\\n NoteMetaData.objects.create(note=instance)\",\n \"def create_db_post(entry, keys, like):\\n h = get_hash(entry['link'])\\n collection = pos if like else neg\\n return collection.update(\\n {'hash': h},\\n {\\n 'link': entry['link'],\\n 'title': entry['title'],\\n 'published': '',\\n 'content': \\\" \\\".join(keys),\\n 'hash': h,\\n 'read': False\\n }, upsert=True\\n )\",\n \"def add_post(self, post: Post) -> None:\\n self.post_process.append(post)\",\n \"def save(self, data):\\n data['id'] = self.id\\n\\n self.db.append(data)\",\n \"def instance(data):\\n return Fieldsha1(data)\",\n \"def post_kw_view_func(kw, obj, obj_type, obj_id, **kwargs):\\n kw = default_view_kw(kw, obj.thread, obj_type, obj_id, **kwargs)\\n kw['thread_id'] = obj.thread_id\\n kw['_anchor'] = u'post_{:d}'.format(obj.id)\\n return kw\",\n \"def process_IN_CREATE(self, event):\",\n \"def __init__(self, post):\\n self.id = int(post[0])\\n self.type_id = int(post[1])\\n try:\\n self.parent_id = int(post[2])\\n except:\\n self.parent_id = None\\n try:\\n self.accepted_answer_id = int(post[3])\\n except:\\n self.accepted_answer_id = None\\n self.creation_date = post[4]\\n try:\\n self.score = post[5]\\n except:\\n self.score = None\\n try:\\n self.view_count = post[6]\\n except:\\n self.view_count = 0\\n self.body = post[7]\\n try:\\n self.owner_user_id = int(post[8])\\n except:\\n self.owner_user_id = None \\n try:\\n self.last_editor_user_id = int(post[9])\\n except:\\n self.last_editor_user_id = None\\n try:\\n self.last_editor_display_name = post[10]\\n except:\\n self.last_editor_display_name = None\\n self.last_activity_date = post[11]\\n self.last_edit_date = post[12]\\n self.community_owned_date = post[13]\\n self.closed_date = post[14]\\n try:\\n self.title = post[15]\\n except:\\n self.title = None\\n try:\\n self.tags = post[16]\\n except:\\n self.tags = None\\n try:\\n self.answer_count = int(post[17])\\n except:\\n self.answer_count = 0\\n try:\\n self.comment_count = int(post[18])\\n except:\\n self.comment_count = 0\\n try:\\n self.favorite_count = int(post[19])\\n except:\\n self.favorite_count = 0\",\n \"def save_current_post(entry):\\n return current.insert_one(entry).inserted_id\",\n \"def get_post(self):\\n\\t\\tself.post = graph.get_object(POST_ID)\",\n \"def save(self,**kwargs):\\n if not self.id:\\n self.created_at = datetime.now()\\n super(TestUrl, self).save(**kwargs)\",\n \"def postTweet(self, userId: int, tweetId: int) -> None:\\n self.users.add(userId)\\n self.user_post[userId].append((tweetId, self.time))\\n self.time += 1\",\n \"def set_new_hash(court_id, new_hash):\\n seals_data[court_id]['hash'] = new_hash\",\n \"def postTweet(self, userId, tweetId):\\n if not self.user_pool[userId].user_id:\\n user = UserInfo()\\n user.user_id = userId\\n self.user_pool[userId] = user\\n self.time += 1\\n tw_info = (tweetId, self.time) # 保存一个和时间戳\\n self.twitter_pool[userId].append(tw_info)\",\n \"def __init__(self):\\n self.id = id(self)\",\n \"def save(self, *args, **kwargs):\\n if not self.post_id:\\n self.created = timezone.now()\\n dup = StoryPost.objects.filter(title=self.title)\\n if len(dup) > 0:\\n # objects with the same slug exist -> duplicate!\\n nos = str(len(dup))\\n # append number of duplicates as modifier\\n self.slug = slugify(self.title[:49 - len(dup)] + '-' + nos)\\n else:\\n self.slug = slugify(self.title[:50])\\n self.modified = timezone.now()\\n\\n return super(StoryPost, self).save(*args, **kwargs)\",\n \"def add(self, obj):\\n ID = id(obj)\\n self.pDict[ID] = obj\\n return ID\",\n \"def save(self, *args, **kwargs):\\n\\n if not self.trackerid:\\n self.trackerid = generate_trackerid()\\n super(Profile, self).save(*args, **kwargs)\",\n \"def store(self, dikt):\\n if self.key is not None and self.body is not None:\\n dikt[self.key] = self\\n elif self.key in dikt:\\n del dikt[self.key]\",\n \"def setTypeId(self, x, y, z, type_):\\n state = self.world.getBlockAt(x, y, z).getState()\\n state.setTypeId(type_)\\n self.list_.add(state)\",\n \"def set_hasher(self, hasher_type):\\n if hasher_type == 'sha256':\\n self.hasher = hashlib.sha256()\\n self.hasher_size = 64\\n elif hasher_type == 'md5':\\n self.hasher = hashlib.md5()\\n self.hasher_size = 32\\n else:\\n raise ValueError(\\n \\\"Unknown hasher_type for checksum: {}\\\".format(hasher_type))\",\n \"def create_definition(self, block_type, slug=None):\\n raise NotImplementedError()\",\n \"def __hash__(self):\\n return self['id'].__hash__()\",\n \"def hash_code(sender: Key.__class__, **kwargs):\\n new_key = kwargs['instance']\\n new_key.code = randint(MIN_CODE, MAX_CODE)\\n new_key.hash_code = sha1(str(new_key.code).encode('utf-8')).hexdigest()\",\n \"def create_post():\\n\\n #Get prompt id\\n prompt_id = request.form.get('prompt_id')\\n\\n # Get post text\\n post_text = request.form.get('user_post')\\n\\n # Create post timestamp\\n created_at = datetime.now()\\n user_facing_date = created_at.strftime(\\\"%B %d, %Y\\\")\\n\\n # Save post and related data to database\\n post = crud.create_post(session['user_id'], prompt_id, post_text, session['lat'], session['lng'], session['user_facing_location'], created_at)\\n\\n return render_template('post_data.html', post=post, user_facing_date=user_facing_date)\",\n \"def save(self, *args, **kwargs):\\r\\n\\r\\n if not self.trackerid:\\r\\n self.trackerid = generate_trackerid()\\r\\n super(Profile, self).save(*args, **kwargs)\",\n \"def save(self, **kwargs):\\n if not self.id and not self.slug:\\n self.set_unique_slug()\\n\\n super().save(**kwargs)\",\n \"def post(self, args):\\n\\n try:\\n base64.b64decode(\\n args[\\\"cap\\\"]\\n ) # decode to ensure valid base64 (but don't use it)\\n except:\\n return {\\\"status\\\": \\\"not valid base64\\\"}, 400\\n\\n got_hash = Hash.query.filter_by(cap=args[\\\"cap\\\"]).first()\\n\\n if got_hash is None:\\n got_hash = Hash(args[\\\"cap\\\"])\\n\\n db.session.add(got_hash)\\n db.session.commit()\\n\\n return (\\n {\\\"status\\\": \\\"success\\\", \\\"body\\\": {\\\"hash\\\": hash_schema.dump(got_hash)}},\\n 200,\\n ) # make new hash\\n elif len(got_hash.jobs) != 0:\\n return (\\n {\\\"status\\\": \\\"completed\\\", \\\"body\\\": {\\\"hash\\\": hash_schema.dump(got_hash)}},\\n 200,\\n ) # if hash has finished jobs\\n else:\\n return (\\n {\\\"status\\\": \\\"queued\\\", \\\"body\\\": {\\\"hash\\\": hash_schema.dump(got_hash)}},\\n 202,\\n ) # if hash is already queued\",\n \"def handle_new_post(post_data, user_agent, remote_addr):\\n \\n for required in POST_REQUIRED_PARAMS:\\n if required not in post_data:\\n return None, None\\n\\n try:\\n value = int(string_from_interwebs(post_data.getfirst(\\\"code\\\", \\\"\\\")))\\n except ValueError:\\n return None, None\\n \\n if value != 98098098098:\\n return None, None\\n\\n # not yet safe to use.\\n location = post_data.getfirst(\\\"location\\\", \\\"\\\")\\n tags = string_from_interwebs(post_data.getfirst(\\\"tags\\\")) \\n author = post_data.getfirst(\\\"author\\\")\\n \\n split_tags = [string_from_interwebs(tag).strip().lower() for tag in tags.split(\\\",\\\")] # temporary\\n \\n if len(split_tags) > 3:\\n return None, None\\n \\n author_id = string_from_interwebs(author).strip()\\n \\n with Connection('localhost', 27017) as connection:\\n reply_to = string_from_interwebs(post_data.getfirst(\\\"reply_to\\\"))\\n \\n if not verify_author(author_id, connection):\\n return None, None\\n\\n if not verify_post(reply_to, connection):\\n return None, None\\n\\n # if reply then it's verified.\\n # XXX: I need to make a standard object structure for this, so that I don't \\n # have to update separate things.\\n\\n post = {\\\"viewed\\\" : 0,\\n \\\"comments\\\" : 0,\\n \\\"flagged\\\" : 0,\\n \\\"disliked\\\" : 0,\\n \\\"enjoyed\\\" : 0,\\n \\\"num_replies\\\" : 0,\\n \\\"num_reposts\\\" : 0,\\n \\\"content-type\\\" : \\\"image\\\", # need to pull this from the mime lookup\\n \\\"file\\\" : \\\"placeholder\\\",\\n \\\"user_agent\\\" : user_agent,\\n \\\"remote_addr\\\" : remote_addr,\\n \\\"created\\\" : datetime.utcnow(),\\n \\\"location\\\" : string_from_interwebs(location).strip(),\\n \\\"author\\\" : ObjectId(author_id),\\n \\\"reply_to\\\" : ObjectId(reply_to),\\n \\\"tags\\\" : split_tags}\\n\\n update_post(reply_to, connection)\\n\\n return post_data.getfirst(\\\"data\\\"), post\",\n \"def save(self, *args, **kwargs):\\n created = not self.id\\n if not self.digest:\\n self.digest = self.get_digest()\\n if not self.cover_img:\\n self.cover_img = getattr(settings, 'WEIXIN_DEFAULT_COVER')\\n self.fileid = getattr(settings, 'WEIXIN_DEFAULT_COVER_ID')\\n super(WeixinMp, self).save(*args, **kwargs)\\n\\n if self.sync and created:\\n from sae.taskqueue import add_task\\n task_link = reverse('weixinmp.upload', args=(self.pk,))\\n add_task('weixin', task_link, delay=300)\\n task_link = reverse('weixinmp.image_collect', args=(self.pk,))\\n add_task('weixin', task_link, delay=180)\\n\\n if created:\\n event = Event(user=self.user, content_object=self)\\n event.save()\",\n \"def save(self, *args, **kwargs):\\n if not self.id:\\n self.created = timezone.now()\\n\\n self.modified = timezone.now()\\n\\n return super(CodeSnippet, self).save(*args, **kwargs)\",\n \"def hash_lock_id(sender: Lock.__class__, **kwargs):\\n new_lock = kwargs['instance']\\n new_lock.hash_id = sha1(str(new_lock.uuid).encode('utf-8')).hexdigest()\",\n \"def post(self, uid):\\n json_data = request.get_json()\\n b = event_builders.FbEventBuilder()\\n b.build_with_fb_dict(json_data)\\n e: event_models.SocialEvent = b.export_as_class(event_models.SocialEvent)\\n\\n # Note that e.firestore_ref will not be set by create()\\n ref: DocumentReference = EventDao().create_fb_event(e)\\n e.set_firestore_ref(ref)\\n dict_view = e.to_dict_view()\\n dict_view[\\\"eventId\\\"] = ref.id\\n\\n # TODO: add error handling\\n UserDao().add_user_event_dict(uid, dict_view[\\\"fbEventId\\\"], dict_view)\\n\\n return {\\n \\\"id\\\": e.get_firestore_ref().id\\n }\",\n \"def _set_id(self):\\n raise NotImplementedError()\",\n \"def create(self):\\n ...\",\n \"def post_routing_instance_create(self, resource_dict):\\n pass\",\n \"def save(self, *args, **kwargs):\\n if not self.identifier:\\n if self.event is None:\\n self.identifier = self.email.lower()\\n else:\\n self.identifier = \\\"%s@%s.event.pretix\\\" % (self.username.lower(), self.event.id)\\n if not self.pk:\\n self.identifier = self.identifier.lower()\\n super().save(*args, **kwargs)\",\n \"def __init__(self, memo, tags=()):\\n\\n self.memo = memo\\n self.tags = tags\\n self.creation_date = datetime.date.today()\\n global last_id\\n last_id += 1\\n self.id = last_id\",\n \"def post(self, copy_id):\\n checkCopyValidity(copy_id)\\n note_body = request.get_json()\\n new_note = models.Notes()\\n new_note.parse_body(note_body)\\n new_note.copy_id = copy_id\\n db.session.add(new_note)\\n db.session.commit()\\n return 'A note \\\\\\\"{}\\\\\\\" has been added to book copy of {}'.format(new_note.note, copy_id), 201\",\n \"def _set_unique_id(self, json_request):\\n values = []\\n for field in value_fields:\\n value = json_request.get(field, '')\\n values.append(quote(self.fully_decode_uri(value), safe=''))\\n if len(values) == 1:\\n self.unique_id = values[0]\\n elif len(values) == 2:\\n self.unique_id = self.build_summary(values[0], values[1])\\n elif len(values) == 1:\\n self.unique_id = self.build_summary(values[0], values[1], values[2])\",\n \"def __addPost(self, post, is_question = False):\\n try:\\n post_id = post.find('a',id=re.compile('postcount\\\\d+')) \\n unique_key = post_id['id'].replace('postcount','')\\n log.info(unique_key)\\n if checkSessionInfo(self.genre, self.session_info_out, unique_key,\\\\\\n self.task.instance_data.get('update'),parent_list\\\\\\n = [self.task.instance_data['uri']]):\\n log.info(self.log_msg('Session info returns True'))\\n return False\\n page = self.__getData(post, is_question)\\n except:\\n log.info(post)\\n log.exception(self.log_msg('Cannot add the post for the url %s'%\\\\\\n self.currenturi))\\n return False\\n try:\\n page['uri'] = self.currenturi + '#' + stripHtml(post_id.renderContents()) \\n except:\\n log.info(self.log_msg('Cannot find the uri'))\\n page['uri'] = self.currenturi\\n try:\\n result=updateSessionInfo(self.genre, self.session_info_out, unique_key, \\\\\\n get_hash( page ),'Review', self.task.instance_data.get('update'),\\\\\\n parent_list=[self.task.instance_data['uri']])\\n if not result['updated']:\\n log.info(self.log_msg('Update session info returns False'))\\n return True\\n page['parent_path'] = page['path'] = [self.task.instance_data['uri']]\\n page['path'].append(unique_key)\\n #page['uri'] = unique_key\\n page['uri_domain'] = urlparse.urlparse(page['uri'])[1]\\n page.update(self.__task_elements_dict)\\n self.pages.append(page)\\n log.info(self.log_msg('Post Added'))\\n return True\\n except:\\n log.exception(self.log_msg('Error while adding session info'))\\n return False\",\n \"def handle_post(cls, **kwargs):\\n raise NotImplementedError\",\n \"def calculate_hash_id(self):\\n return get_md5_hash(f'{self.type}{self.get_primary_id()}')\",\n \"def __init__(self, rpc, post):\\n super(PostModule, self).__init__(rpc, 'post', post)\\n self._action = self._info.get('default_action', \\\"\\\")\",\n \"def _calculate_hash(self, entry):\\n entry.pop('id', None)\\n return hashlib.sha224(json.dumps(\\n entry, cls=DjangoJSONEncoder).encode('utf-8')).hexdigest()\",\n \"def __hash__(self):\\n return hash(self.id)\",\n \"def __hash__(self):\\n return hash(self.id)\",\n \"def __hash__(self):\\n return hash(self.id)\",\n \"def __hash__(self):\\n return hash(self.id)\",\n \"def post_to_activity(self, post):\\n id = None\\n if post.get('id'):\\n # strip USERID_ prefix if it's there\\n post['id'] = post['id'].split('_', 1)[-1]\\n id = post['id']\\n\\n obj = self.post_to_object(post)\\n activity = {\\n 'verb': VERBS.get(post.get('type', obj.get('objectType')), 'post'),\\n 'published': obj.get('published'),\\n 'updated': obj.get('updated'),\\n 'id': self.tag_uri(id) if id else None,\\n 'url': self.post_url(post),\\n 'actor': obj.get('author'),\\n 'object': obj,\\n }\\n\\n application = post.get('application')\\n if application:\\n activity['generator'] = {\\n 'displayName': application.get('name'),\\n 'id': self.tag_uri(application.get('id')),\\n }\\n return self.postprocess_activity(activity)\",\n \"def regist_hash(cobj, hash, handler, dir):\\n pass\",\n \"def perform_create(self,serializer):\\n serializer.save(user_id=self.request.user)\",\n \"def insert_hash(self, title, artist, song_id:int, fingerprint: str, offset:int):\\n fingerprint = Fingerprints(song_id=song_id, song_title=title, artist=artist, hash=fingerprint, offset=offset)\\n fingerprint.save()\",\n \"def post(self, request, *args, **kwargs):\\n return super().create(request, *args, **kwargs)\",\n \"def write_post(id, args):\\n graph = facebook.GraphAPI(id) \\n graph.put_object(parent_object='me', connection_name = 'feed', **args)\",\n \"def _set_hash(self, *args):\\n if self.df is None:\\n raise ValueError(\\\"df is None\\\")\\n self._hash_df = get_df_hash(self.df)\\n self._hash_args = \\\", \\\".join([str(arg) for arg in args])\\n self._hash=\\\"{klass:}({df_hash:}, {args_hash:})\\\".format(\\n klass = self.__class__.__name__,\\n df_hash = self._hash_df,\\n args_hash = self._hash_args,\\n )\",\n \"def post_provider_attachment_create(self, resource_dict):\\n pass\"\n]","isConversational":false},"negative_scores":{"kind":"list like","value":["0.6146022","0.60286593","0.5837189","0.5832979","0.5738559","0.57207257","0.56405866","0.5623517","0.5464062","0.54599774","0.5413562","0.5413562","0.53702575","0.5368633","0.53681165","0.5332242","0.5332242","0.5330535","0.5305785","0.5294041","0.529331","0.52832794","0.5248875","0.52467287","0.52222544","0.5185685","0.51674175","0.51457155","0.51391906","0.5106024","0.5101177","0.50934684","0.5090562","0.50765276","0.50763553","0.5057839","0.5046716","0.5041662","0.5041131","0.5032813","0.50327146","0.50270694","0.49972203","0.49948514","0.49880946","0.49825674","0.49766868","0.4973952","0.4962082","0.49562517","0.49529055","0.495021","0.4949068","0.49490646","0.49488392","0.4941289","0.4939031","0.49218783","0.49138296","0.49109253","0.491014","0.49097142","0.49085152","0.49062434","0.49024478","0.48995078","0.4889808","0.48880997","0.48804957","0.48771536","0.48712695","0.48581666","0.48576194","0.48453423","0.48438168","0.4842885","0.48379752","0.48371","0.48369873","0.48269248","0.48245844","0.48232567","0.4815294","0.48120382","0.48071998","0.4799367","0.4792241","0.47890267","0.47770268","0.47770268","0.47770268","0.47770268","0.4774468","0.4773196","0.4769521","0.47687274","0.4767418","0.4755435","0.4754653","0.4753641"],"string":"[\n \"0.6146022\",\n \"0.60286593\",\n \"0.5837189\",\n \"0.5832979\",\n \"0.5738559\",\n \"0.57207257\",\n \"0.56405866\",\n \"0.5623517\",\n \"0.5464062\",\n \"0.54599774\",\n \"0.5413562\",\n \"0.5413562\",\n \"0.53702575\",\n \"0.5368633\",\n \"0.53681165\",\n \"0.5332242\",\n \"0.5332242\",\n \"0.5330535\",\n \"0.5305785\",\n \"0.5294041\",\n \"0.529331\",\n \"0.52832794\",\n \"0.5248875\",\n \"0.52467287\",\n \"0.52222544\",\n \"0.5185685\",\n \"0.51674175\",\n \"0.51457155\",\n \"0.51391906\",\n \"0.5106024\",\n \"0.5101177\",\n \"0.50934684\",\n \"0.5090562\",\n \"0.50765276\",\n \"0.50763553\",\n \"0.5057839\",\n \"0.5046716\",\n \"0.5041662\",\n \"0.5041131\",\n \"0.5032813\",\n \"0.50327146\",\n \"0.50270694\",\n \"0.49972203\",\n \"0.49948514\",\n \"0.49880946\",\n \"0.49825674\",\n \"0.49766868\",\n \"0.4973952\",\n \"0.4962082\",\n \"0.49562517\",\n \"0.49529055\",\n \"0.495021\",\n \"0.4949068\",\n \"0.49490646\",\n \"0.49488392\",\n \"0.4941289\",\n \"0.4939031\",\n \"0.49218783\",\n \"0.49138296\",\n \"0.49109253\",\n \"0.491014\",\n \"0.49097142\",\n \"0.49085152\",\n \"0.49062434\",\n \"0.49024478\",\n \"0.48995078\",\n \"0.4889808\",\n \"0.48880997\",\n \"0.48804957\",\n \"0.48771536\",\n \"0.48712695\",\n \"0.48581666\",\n \"0.48576194\",\n \"0.48453423\",\n \"0.48438168\",\n \"0.4842885\",\n \"0.48379752\",\n \"0.48371\",\n \"0.48369873\",\n \"0.48269248\",\n \"0.48245844\",\n \"0.48232567\",\n \"0.4815294\",\n \"0.48120382\",\n \"0.48071998\",\n \"0.4799367\",\n \"0.4792241\",\n \"0.47890267\",\n \"0.47770268\",\n \"0.47770268\",\n \"0.47770268\",\n \"0.47770268\",\n \"0.4774468\",\n \"0.4773196\",\n \"0.4769521\",\n \"0.47687274\",\n \"0.4767418\",\n \"0.4755435\",\n \"0.4754653\",\n \"0.4753641\"\n]","isConversational":false},"document_score":{"kind":"string","value":"0.57402194"},"document_rank":{"kind":"string","value":"4"}}},{"rowIdx":93280,"cells":{"query":{"kind":"string","value":"Assign Tags and Mentions to a Post(type) by parsing the description."},"document":{"kind":"string","value":"def on_saved_post(sender, instance, created, **kwargs):\n # If the post was just created, stop here since we have no tags yet\n if created:\n return\n if not issubclass(sender, dillo.models.posts.Post):\n return\n # Extract tags and mentions from text and assign them to the Post\n tags, mentions = extract_tags_and_mentions(instance.title)\n instance.tags.set(*tags)\n # Delete all existing mentions\n dillo.models.mixins.Mentions.objects.filter(\n content_type_id=instance.content_type_id, object_id=instance.id\n ).delete()\n for mentioned_user in mentions:\n log.debug('Mentioning user %s in Post %i' % (mentioned_user, instance.id))\n dillo.models.mixins.Mentions.objects.create(\n user=mentioned_user,\n content_object=instance,\n )"},"metadata":{"kind":"string","value":"{\n \"objective\": {\n \"self\": [],\n \"paired\": [],\n \"triplet\": [\n [\n \"query\",\n \"document\",\n \"negatives\"\n ]\n ]\n }\n}"},"negatives":{"kind":"list like","value":["def add_meta(self, post, *args, **kwargs):\n\t\tsuper(Command, self).add_meta(post, *args, **kwargs)\n\t\tpost.gen_description = False\n\t\tpost.description = description_from_content(post)\n\t\tpost.save()","def post(self, post_id=None):\n\n if post_id:\n abort(400)\n else:\n args = parsers.post_post_parser.parse_args(strict=True)\n\n new_post = Post(args['title'])\n new_post.text = args['text']\n # new_post.user = user\n\n if args['tags']:\n for item in args['tags']:\n tag = Tag.query.filter_by(name=item).first()\n # If the tag already exist, append.\n if tag:\n new_post.tags.append(tag)\n # If the tag not exist, create the new one.\n # Will be write into DB with session do.\n else:\n new_tag = Tag(item)\n new_post.tags.append(new_tag)\n db.session.add(new_post)\n db.session.commit()\n return (new_post.id, 201)","def preprocess_post(self, post):\n # tokenize, clean, & tag part-of-speech for all words\n if self.document_level == 'postwise':\n\n doc_text = all_comments_from_post(post)\n # leave early if there's nothing there\n if doc_text == '':\n return []\n\n tokens = nltk.word_tokenize(doc_text)\n # TODO: skip this if there's no POS filtering args!\n tagged = nltk.pos_tag(tokens)\n\n # filter out most invalid words with valid_word()\n processed_document = []\n for word, pos_tag in tagged:\n if self.valid_word(word, pos_tag):\n cleaned_word = self.clean_word(word)\n # things like digits and other junk become empty string,\n # so exclude them from final document\n if cleaned_word:\n processed_document.append(cleaned_word)\n # finally, update the post\n post['postwise'] = {'tokens': processed_document, 'text': doc_text}\n self.postman.posts_write.update_one({'_id':post['_id']}, {'$set':post}, upsert=True)\n else:\n raise NotImplementedError('document_level: \"%s\"' % self.document_level)\n\n return processed_document","def setPostUp(self, post):\n # type: (tp.Union[str, tp.List[str]])->None\n if isinstance(post, list):\n self._ifAttributes['post-up'] = post\n elif isinstance(post, str):\n self._ifAttributes['post-up'] = [post]\n else:\n raise ValueError(\"Invalid value type {0}, expected str or List[str]\".format(type(post)))","def refactor_post(self,post_name):\n for name in list(self.rules):\n related_post = \"{}.post.{}\".format(name,post_name)\n if related_post in self.rules:\n parts = [self.MakeSymbolName(x) for x in [post_name, related_post]]\n self.rules[name] = self.MakeChoice([self.MakeSeq(parts)])","def parse_post_text(formatted_content):\n post = {}\n # Parse Mod comments and remove them from the text.\n potential_comments = re.finditer(\"\\[.+?\\]\", formatted_content, re.DOTALL)\n comments = []\n for comment_match in potential_comments:\n comment = comment_match.group()\n mod = re.search(r\"\\-\\s?Mod\\.\\s?(?P<mod>\\w+\\b)\", comment)\n if mod:\n comments.append({\n \"comment\" : comment,\n \"mod\" : mod.group(\"mod\")\n })\n post[\"modComments\"] = comments\n \n # Comments are removed from the post test so that\n # links, reports, etc. mentioned by mods are not extracted.\n no_comment_txt = formatted_content\n for comment in comments:\n no_comment_txt = no_comment_txt.replace(comment[\"comment\"], \"\")\n \n metadata, header_end = parse_post_metadata(no_comment_txt)\n post.update(metadata)\n \n sections = re.split(r\"^[\\*#]{3,}\\s*$\", no_comment_txt[header_end:], flags=re.M)\n articles = []\n \n # Some posts have articles which are parsed into multiple sections:\n # Ex: http://www.promedmail.org/direct.php?id=2194235\n # The section parsing code tries to recombine these by concatenating\n # unrecognized sections onto the previous sections if they form an article.\n # article_start_idx keeps track of the first section in the article.\n article_start_idx = None\n \n for idx, section in enumerate(sections):\n section = section.strip()\n article = parse_article_text(section, post_date=post['promedDate'])\n # Check if the section contains an actual article by seeing which\n # properties could be parsed.\n if article.get('source') or article.get('date'):\n articles.append(article)\n article_start_idx = idx\n else:\n # When a section cannot be parsed as an article the following code\n # tries to determine what it is. If the type cannot be determined\n # an error or warning is thrown.\n # These warnings can be used to find sections which are not being\n # correctly parsed.\n # Posts with known issues:\n # http://www.promedmail.org/direct.php?id=19990512.0773\n if re.search(r\"Visit ProMED-mail\\'s web site at|\"\n r\"Please support (the \\d{4}\\s)?ProMED\\-mail|\"\n r\"Donate to ProMED\\-mail. Details available at|\"\n r\"ProMED\\-mail makes every effort to verify the reports|\"\n r\"PROMED\\-MAIL FREQUENTLY ASKED QUESTIONS|\"\n r\"Become a ProMED\\-mail Premium Subscriber|\"\n r\"A ProMED\\-mail post\",\n section, re.I):\n # boilerplate promed notice section\n pass\n elif re.search(r\"In this (update|post(ing)?)\", section):\n # table of contents section\n pass\n elif re.search(r\"Cases in various countries\", section):\n # This type of post typically has links to several articles\n # with single sentence summaries.\n # Ex: http://www.promedmail.org/direct.php?id=20131125.2073661\n pass\n elif section == \"\":\n # empty section\n pass\n elif idx == 0 and section.count(\"\\n\") < 2:\n # probably the article title\n pass\n else:\n if article_start_idx != None:\n article = parse_article_text(\n \"\\n#####\\n\".join(\n sections[article_start_idx:idx]).strip(),\n post_date=post['promedDate'])\n assert article.get('source') or article.get('date')\n articles[-1] = article\n continue\n else:\n print \"Unexpected Section (%s):\" % post['archiveNumber'], [section[0:50] + \"...\"]\n article_start_idx = None\n post['articles'] = articles\n return post","def add_post(self, post: Post) -> None:\n self.post_process.append(post)","def post_to_activity(self, post):\n id = None\n if post.get('id'):\n # strip USERID_ prefix if it's there\n post['id'] = post['id'].split('_', 1)[-1]\n id = post['id']\n\n obj = self.post_to_object(post)\n activity = {\n 'verb': VERBS.get(post.get('type', obj.get('objectType')), 'post'),\n 'published': obj.get('published'),\n 'updated': obj.get('updated'),\n 'id': self.tag_uri(id) if id else None,\n 'url': self.post_url(post),\n 'actor': obj.get('author'),\n 'object': obj,\n }\n\n application = post.get('application')\n if application:\n activity['generator'] = {\n 'displayName': application.get('name'),\n 'id': self.tag_uri(application.get('id')),\n }\n return self.postprocess_activity(activity)","def update_post_format(post):\n\n post_dict = {\n \"title\": post[1],\n \"genre\": get_genre(post[0]),\n \"content\": post[2],\n \"repeater_link\": get_links(post[3], post[4]),\n }\n \n return post_dict","def _fillPost(self, useRawHTML, old_data=None):\n # Initialize empty dictionnary ct (aka content)\n # to be sent through self.server.metaWeblog.newPost()\n ct = {}\n\n # if no old_data, create a fake one\n if old_data == None:\n old_data = { 'title': None\n , 'mt_keywords': None\n , 'formatted_text': BALISE\n , 'mt_excerpt': None\n , 'description': None}\n\n def updateField(prompt, string=None):\n if (string == None) or (string == \"\") :\n return raw_input(prompt)\n else :\n r = raw_input(prompt + \" [default:\" + string + \"]\\n\")\n if r == \"\" :\n return string\n else :\n return r\n\n # Now get information\n ct['title'] = updateField( \"Title?\\n\", old_data['title'])\n ct['mt_keywords'] = updateField(\n \"Tags? (comma separated lists)?\\n\",\n old_data['mt_keywords'])\n\n # Categories are not included in the struct \"ct\"\n # see _setCategorie()\n\n # Get excerpt/content\n # Method0: external XML file\n if useRawHTML:\n with open( useRawHTML, 'rb') as f:\n doc = xml.dom.minidom.parse(f)\n # Parse our XHTML file\n text = doc.getElementsByTagName(\"body\")[0].toxml()\n #text = text.decode() # convert bytes to string\n text = text.replace(\"<body>\", \"\").replace(\"</body>\", \"\")\n ct['mt_excerpt'], ct['description'] = split_excerpt( text)\n\n # Method1: custom editor\n elif self.editor :\n prev_data = old_data['formatted_text']\n data = self._externalEditor( wrap_with_template(prev_data) )\n ct['mt_excerpt'], ct['description'] = split_excerpt( data)\n\n # Method2: input\n else :\n ct['mt_excerpt'] = updateField(\n \"Excerpt? (beware of xHTML tags !)\\n\",\n old_data['mt_excerpt'])\n ct['description'] = updateField(\n \"Main content? (beware of xHTML tags !)\\n\",\n old_data['description'])\n\n # Process the rest of the attributes (comments, pings, ...)\n def set_boolean( prompt, default):\n if default == True:\n return raw_input(prompt + \"[Y|n]\") != \"n\"\n else:\n return raw_input(prompt + \"[y|N]\") != \"y\"\n\n ct['mt_allow_comments'] = set_boolean( \"Allow comments ?\"\n , self.auto_comments)\n ct['mt_allow_pings'] = set_boolean( \"Allow pings ?\"\n , self.auto_pings)\n publish = set_boolean( \"Publish ?\" , self.auto_publish)\n\n return ct, publish","def post(self, post):\n\n self._post = post","def post(self, post):\n\n self._post = post","def _populate_posts(self, channel, url):\n import feedparser\n\n Post = get_model('articles', 'Post')\n Image = get_model('images', 'Image')\n\n parser = feedparser.parse(url)\n\n for entry in parser['entries']:\n # Some entries are incomplete and have only the title, need to\n # ignore these entries.\n if not entry.get('summary'):\n continue\n\n # The title may have only 140 characters\n title = self._truncate_string(entry['title'], 140)\n slug = slugify(title)\n headline = entry['summary']\n\n # Some entries do not have the 'content' field, in this case we\n # get the 'summary' field instead.\n if entry.get('content'):\n content = entry['content'][0]['value']\n else:\n content = entry['summary']\n\n # When we find a entry that already is registered we don't need\n # continue because the following registries already be registered.\n exists = Post.objects.filter(slug=slug).count()\n if exists:\n break\n\n # Check if has some image in the post content.\n # NOTE: For the best user experience we use only the posts that\n # have images.\n image_url = self._get_image_url_in_content(content)\n if image_url:\n main_image = Image.objects.create(\n title=title,\n slug=slug,\n archive_link=image_url,\n published=True,\n user=self._user\n )\n # Generate the 'short_title' based on 'content'\n short_title = re.sub('<[^<]+?>', '', content).encode('utf-8')\n short_title = self._truncate_string(short_title.strip(), 140)\n\n post = Post.objects.create(\n title=title,\n short_title=short_title,\n slug=slug,\n headline=headline,\n content=content,\n channel=channel,\n main_image=main_image,\n show_on_root_channel=True,\n published=True,\n hat='',\n user=self._user\n )","def insert_new_post(post_arg_set):\n api, post_data, acct_data, page_id, config = post_arg_set\n\n try:\n post_id = post_data['id'] if post_data.has_key('id') else None\n\n except Exception as e:\n log.error( e )\n\n else:\n\n # parse date\n if post_data.has_key('created_time') and post_data['created_time'] is not None: \n dt = datetime.strptime(post_data['created_time'], FB_DATE_FORMAT)\n date_time = tz_adj(dt, config)\n time_bucket = round_datetime(date_time, config)\n raw_timestamp = int(date_time.strftime(\"%s\"))\n \n else:\n time_bucket = None\n raw_timestamp = None\n \n # extract message so we can find links within the msg if not in url\n article_urls = [get_fb_link(post_data, config, unshorten=True)]\n message = post_data['message'].encode('utf-8') if post_data.has_key('message') else None\n message_urls = get_message_urls(article_urls, message, config)\n\n # detect article links, unshorten and parse\n article_urls = [\n parse_url(unshorten_link(url, config)) \\\n for url in article_urls + message_urls\n if url is not None\n ]\n\n article_urls = [url for url in article_urls if is_article(url, config)]\n\n if article_urls:\n for article_url in set(article_urls):\n\n # sluggify url\n article_slug = sluggify(article_url)\n\n # format data\n post_value = {\n 'article_slug': article_slug,\n 'article_url': article_url,\n 'time_bucket': time_bucket,\n 'fb_post_created': raw_timestamp,\n 'raw_timestamp': raw_timestamp,\n 'fb_raw_link' : get_fb_link(post_data, config=config),\n 'fb_page_id': page_id,\n 'fb_post_id': post_id,\n 'fb_page_likes': acct_data['likes'] if acct_data.has_key('likes') else None,\n 'fb_page_talking_about': acct_data['talking_about_count'] if acct_data.has_key('talking_about_count') else None,\n 'fb_type': post_data['type'] if post_data.has_key('type') else None,\n 'fb_status_type': post_data['status_type'] if post_data.has_key('status_type') else None,\n 'fb_message': message\n }\n \n # always insert insights data\n if is_insights(page_id, config):\n \n log.info( \"INSIGHTS\\tAdding data from %s re: %s\" % (page_id, article_slug) )\n\n # fetch data\n insights_value = get_insights_data(api, page_id, post_id)\n\n # create datasource name\n data_source = \"facebook_insights_%s\" % page_id \n \n # upsert url\n upsert_url(article_url, article_slug, data_source, config)\n\n # insert id\n db.sadd('facebook_post_ids', post_id)\n\n # format time bucket\n current_time_bucket = gen_time_bucket(config)\n insights_value['time_bucket'] = current_time_bucket\n post_value.pop('time_bucket', None)\n \n value = json.dumps({\n data_source : dict(post_value.items() + insights_value.items())\n })\n\n # upload data to redis\n db.zadd(article_slug, current_time_bucket, value) \n \n # only insert new posts\n if not db.sismember('facebook_post_ids', post_id):\n \n log.info( \"FACEBOOK\\tNew post %s\\t%s\" % (post_id, article_url) )\n \n # insert id\n db.sadd('facebook_post_ids', post_id) \n \n # upsert url\n data_source = \"facebook_%s\" % page_id\n upsert_url(article_url, article_slug, data_source, config)\n\n value = json.dumps( {data_source : post_value} )\n\n\n # upload data to redis\n db.zadd(article_slug, time_bucket, value)","def Translate(self, infile, outfile, mapfile):\r\n # Create the top-level feed object\r\n feed = []\r\n comments = []\r\n\r\n # Calculate the last updated time by inspecting all of the posts\r\n last_updated = 0\r\n\r\n # These three variables keep the state as we parse the file\r\n post_entry = {} # The current post atom.Entry to populate\r\n comment_entry = {} # The current comment atom.Entry to populate\r\n last_entry = None # The previous post atom.Entry if exists\r\n tag_name = None # The current name of multi-line values\r\n tag_contents = '' # The contents of multi-line values\r\n\r\n # Loop through the text lines looking for key/value pairs\r\n split_re = re.compile('^[A-Z ]+:')\r\n for line in infile:\r\n\r\n # Remove whitespace\r\n line = line.strip()\r\n\r\n # Check for the post ending token\r\n if line == '-' * 8 and tag_name != 'BODY':\r\n if post_entry:\r\n # Add the post to our feed\r\n sys.stderr.write(\"Adding post %s\\n\" % post_entry['title'])\r\n self.add_to_user_map(post_entry.get('author'), post_entry.get('authorEmail'))\r\n feed.insert(0, post_entry)\r\n last_entry = post_entry\r\n\r\n # Reset the state variables\r\n post_entry = {}\r\n comment_entry = {}\r\n tag_name = None\r\n tag_contents = ''\r\n continue\r\n\r\n # Check for the tag ending separator\r\n elif line == '-' * 5:\r\n # Get the contents of the body and set the entry contents\r\n if tag_name == 'BODY':\r\n post_entry['description'] = self._Encode(tag_contents)\r\n\r\n # This is the start of the COMMENT section. Create a new entry for\r\n # the comment and add a link to the original post.\r\n elif tag_name == 'COMMENT':\r\n comment_entry['body'] = self._Encode(tag_contents)\r\n post_entry.setdefault('comments', []).append(comment_entry)\r\n self.add_to_user_map(comment_entry.get('author'), comment_entry.get('authorEmail'))\r\n comment_entry = {}\r\n\r\n # Get the contents of the extended body\r\n elif tag_name == 'EXTENDED BODY':\r\n if post_entry:\r\n post_entry['mt_text_more'] = self._Encode(tag_contents)\r\n elif last_entry:\r\n last_entry['mt_text_more'] = self._Encode(tag_contents)\r\n\r\n # Convert any keywords (comma separated values) into Blogger labels\r\n elif tag_name == 'KEYWORDS':\r\n post_entry['mt_keywords'] = tag_contents\r\n\r\n # Reset the current tag and its contents\r\n tag_name = None\r\n tag_contents = ''\r\n continue\r\n\r\n # Split the line into key/value pairs\r\n key = line\r\n value = ''\r\n if split_re.match(line):\r\n elems = line.split(':')\r\n key = elems[0]\r\n if len(elems) > 1:\r\n value = ':'.join(elems[1:]).strip()\r\n\r\n # The author key indicates the start of a post as well as the author of\r\n # the post entry or comment\r\n if key == 'AUTHOR':\r\n # Create a new entry \r\n entry = {}\r\n\r\n # Add the author's name\r\n author_name = self._Encode(value)\r\n if not author_name:\r\n author_name = 'Anonymous'\r\n entry['author'] = author_name\r\n\r\n # Add the appropriate kind, either a post or a comment\r\n if tag_name == 'COMMENT':\r\n entry['postid'] = post_entry['postid']\r\n comment_entry = entry\r\n else:\r\n entry['postid'] = 'post-' + self._GetNextId()\r\n post_entry = entry\r\n\r\n # The title only applies to new posts\r\n elif key == 'TITLE' and tag_name != 'PING':\r\n post_entry['title'] = self._Encode(value)\r\n\r\n # If the status is a draft, mark it as so in the entry. If the status\r\n # is 'Published' there's nothing to do here\r\n elif key == 'STATUS':\r\n post_entry['status'] = value\r\n\r\n # Turn categories into labels\r\n elif key == 'CATEGORY':\r\n post_entry.setdefault('category', []).append(value)\r\n\r\n # Convert the date and specify it as the published/updated time\r\n elif key == 'DATE' and tag_name != 'PING':\r\n entry = post_entry\r\n if tag_name == 'COMMENT':\r\n entry = comment_entry\r\n entry['dateCreated'] = value\r\n\r\n # Check to see if this was the last post published (so far)\r\n # seconds = time.mktime(time_val)\r\n # last_updated = max(seconds, last_updated)\r\n\r\n # Convert all tags into Blogger labels\r\n elif key == 'TAGS':\r\n post_entry.setdefault('tags', []).append(value)\r\n\r\n # Update the author's email if it is present and not empty\r\n elif tag_name == 'COMMENT' and key == 'EMAIL':\r\n comment_entry['authorEmail'] = value\r\n\r\n # Update the author's URI if it is present and not empty\r\n elif tag_name == 'COMMENT' and key == 'URL':\r\n comment_entry['authorUrl'] = value\r\n\r\n # If any of these keys are used, they contain information beyond this key\r\n # on following lines\r\n elif key in ('COMMENT', 'BODY', 'EXTENDED BODY', 'EXCERPT', 'KEYWORDS', 'PING'):\r\n tag_name = key\r\n\r\n # These lines can be safely ignored\r\n elif key in ('BASENAME', 'ALLOW COMMENTS', 'CONVERT BREAKS', \r\n 'ALLOW PINGS', 'PRIMARY CATEGORY', 'IP', 'URL', 'EMAIL'):\r\n continue\r\n\r\n # If the line is empty and we're processing the body, add a line break\r\n elif (tag_name == 'BODY' or tag_name == 'EXTENDED BODY' or tag_name == 'COMMENT') and len(line) == 0:\r\n tag_contents += '\\n'\r\n\r\n # This would be a line of content beyond a key/value pair\r\n elif len(key) != 0:\r\n tag_contents += line + '\\n'\r\n\r\n\r\n # Update the feed with the last updated time\r\n # feed.updated = atom.Updated(self._ToBlogTime(time.gmtime(last_updated)))\r\n\r\n # Serialize the feed object\r\n yaml.dump(feed, outfile, Dumper=yaml.CDumper)\r\n \r\n # Write out the user map\r\n user_map_dict = {}\r\n for name, email in self.user_map:\r\n user_map_dict[name] = email\r\n yaml.dump(user_map_dict, mapfile, Dumper=yaml.CDumper)","def parse_post_content(self, response):\n post = Post()\n post['title'] = response.xpath('//h2/a/text()')[0].extract()\n post['image_url'] = response.xpath(\"//div[@class='cont group']//img/@src\")[0].extract()\n yield post","def proto_post(self, proto_category):\n return baker.make(\n Post,\n content=(\n \"Aliquip excepteur qui mollit labore nulla et culpa \"\n \"minim et commodo reprehenderit consequat sint.\"\n ),\n categories=proto_category,\n make_m2m=True,\n _create_files=True,\n )","def update_post(prev_data, data, db_conn):\n\n schema = get_post_schema(data)\n post_kind = prev_data['kind']\n if post_kind is 'post' or post_kind is 'proposal':\n data = pick(data, ('body',))\n elif post_kind is 'vote':\n data = pick(data, ('body', 'response',))\n data, errors = update_document(schema, prev_data, data, db_conn)\n if not errors:\n add_post_to_es(data, db_conn)\n return data, errors","def proto_post(self, proto_category):\n return baker.make(\n Post,\n content=\"Aute non ex nostrud amet ipsum.\",\n categories=proto_category,\n make_m2m=True,\n _create_files=True,\n )","def proto_post(self, proto_category):\n return baker.make(\n Post,\n content=\"Aute non ex nostrud amet ipsum.\",\n categories=proto_category,\n make_m2m=True,\n _create_files=True,\n )","def parse_description(self, in_str):\n in_str = self.parse_tags(in_str)\n self.description = self.parse_metadata(in_str)","def process_posts(app, doctree):\n env = app.builder.env\n if not hasattr(env, \"ablog_posts\"):\n env.ablog_posts = {}\n post_nodes = list(doctree.findall(PostNode))\n if not post_nodes:\n return\n post_date_format = app.config[\"post_date_format\"]\n should_auto_orphan = app.config[\"post_auto_orphan\"]\n docname = env.docname\n if should_auto_orphan:\n # mark the post as 'orphan' so that\n # \"document isn't included in any toctree\" warning is not issued\n # We do not simply assign to should_auto_orphan because if auto-orphan\n # is false, we still want to respect the per-post :rst:dir`orphan` setting\n app.env.metadata[docname][\"orphan\"] = True\n blog = Blog(app)\n auto_excerpt = blog.post_auto_excerpt\n multi_post = len(post_nodes) > 1 or blog.post_always_section\n for order, node in enumerate(post_nodes, start=1):\n if node[\"excerpt\"] is None:\n node[\"excerpt\"] = auto_excerpt\n if multi_post:\n # section title, and first few paragraphs of the section of post\n # are used when there are more than 1 posts\n section = node\n while True:\n if isinstance(section, nodes.section):\n break\n section = node.parent\n else:\n section = doctree\n # get updates here, in the section that post belongs to\n # Might there be orphan updates?\n update_dates = _get_update_dates(section, docname, post_date_format)\n # Making sure that post has a title because all post titles\n # are needed when resolving post lists in documents\n title = node[\"title\"] or _get_section_title(section)\n # creating a summary here, before references are resolved\n excerpt = []\n if node.children:\n if node[\"exclude\"]:\n node.replace_self([])\n else:\n node.replace_self(node.children)\n for child in node.children:\n excerpt.append(child.deepcopy())\n elif node[\"excerpt\"]:\n count = 0\n for nod in section.findall(nodes.paragraph):\n excerpt.append(nod.deepcopy())\n count += 1\n if count >= (node[\"excerpt\"] or 0):\n break\n node.replace_self([])\n else:\n node.replace_self([])\n nimg = node[\"image\"] or blog.post_auto_image\n if nimg:\n for img, nod in enumerate(section.findall(nodes.image), start=1):\n if img == nimg:\n excerpt.append(nod.deepcopy())\n break\n date = node[\"date\"]\n if date:\n try:\n date = datetime.strptime(date, post_date_format)\n except ValueError:\n if date_parser:\n try:\n date = date_parser(date)\n except ValueError:\n raise ValueError(\"invalid post date in: \" + docname)\n else:\n raise ValueError(\n f\"invalid post date ({date}) in \" + docname + f\". Expected format: {post_date_format}\"\n )\n else:\n date = None\n # if docname ends with `index` use folder name to reference the document\n # a potential problem here is that there may be files/folders with the\n # same name, so issuing a warning when that's the case may be a good idea\n folder, label = os.path.split(docname)\n if label == \"index\":\n folder, label = os.path.split(folder)\n if not label:\n label = slugify(title)\n section_name = \"\"\n if multi_post and section.parent is not doctree:\n section_name = section.attributes[\"ids\"][0]\n label += \"-\" + section_name\n else:\n # create a reference for the post\n # if it is posting the document\n # ! this does not work for sections\n app.env.domains[\"std\"].data[\"labels\"][label] = (docname, label, title)\n app.env.domains[\"std\"].data[\"anonlabels\"][label] = (docname, label)\n if section.parent is doctree:\n section_copy = section[0].deepcopy()\n else:\n section_copy = section.deepcopy()\n # multiple posting may result having post nodes\n for nn in section_copy.findall(PostNode):\n if nn[\"exclude\"]:\n nn.replace_self([])\n else:\n nn.replace_self(node.children)\n postinfo = {\n \"docname\": docname,\n \"section\": section_name,\n \"order\": order,\n \"date\": date,\n \"update\": max(update_dates + [date]),\n \"title\": title,\n \"excerpt\": excerpt,\n \"tags\": node[\"tags\"],\n \"author\": node[\"author\"],\n \"category\": node[\"category\"],\n \"location\": node[\"location\"],\n \"language\": node[\"language\"],\n \"redirect\": node[\"redirect\"],\n \"nocomments\": node[\"nocomments\"],\n \"image\": node[\"image\"],\n \"exclude\": node[\"exclude\"],\n \"external_link\": node[\"external_link\"],\n \"doctree\": section_copy,\n }\n if docname not in env.ablog_posts:\n env.ablog_posts[docname] = []\n env.ablog_posts[docname].append(postinfo)\n # instantiate catalogs and collections here\n # so that references are created and no warnings are issued\n if app.builder.format == \"html\":\n stdlabel = env.domains[\"std\"].data[\"labels\"] # NOQA\n else:\n if hasattr(env, \"intersphinx_inventory\"):\n stdlabel = env.intersphinx_inventory.setdefault(\"std:label\", {}) # NOQA\n baseurl = getattr(env.config, \"blog_baseurl\").rstrip(\"/\") + \"/\" # NOQA\n project, version = env.config.project, str(env.config.version) # NOQA\n for key in [\"tags\", \"author\", \"category\", \"location\", \"language\"]:\n catalog = blog.catalogs[key]\n for label in postinfo[key]:\n coll = catalog[label] # NOQA\n if postinfo[\"date\"]:\n coll = blog.archive[postinfo[\"date\"].year] # NOQA","def run_get_post(m):\n\n doc = get_doc(m)\n assert doc is not None\n\n wp = get_wp(m)\n\n post = find_post(wp, doc.identifier)\n\n if post:\n post.content = \"…content elided…\"\n from pprint import pprint\n pprint(post.struct)\n return\n else:\n warn(f\"Didn't find post for identifier {doc.identifier}\")\n return","def parse_post_metadata(post_text):\n result = {}\n \n header_end = 0\n \n promed_date_match = re.search(\n r\"Published Date:\\s(?P<date>.*)\", post_text)\n result[\"promedDate\"] = parse_promed_pub_datetime(\n promed_date_match.group(\"date\"))\n \n archive_match = re.search(r\"Archive Number: (?P<num>.*)\", post_text)\n result[\"archiveNumber\"] = archive_match.group(\"num\")\n header_end = archive_match.end()\n \n subject = re.search(r\"Subject:\\s(?P<subject>.*)\", post_text).group(\"subject\")\n result[\"subject\"] = parse_subject_line(subject)\n result[\"subject\"][\"raw\"] = subject\n \n # This will not find all linked reports.\n # Some older posts refrence posts using different indexes I do not know\n # how to interpret.\n # Example: http://promedmail.org/direct.php?id=2194235\n result[\"linkedReports\"] = [\n report_id for report_id in re.findall(r\"\\d{8}\\.\\d+\", post_text)]\n \n # Most links will be article source urls or links to promed.\n result[\"links\"] = list(set(\n re.findall(r\"http\\S+[^(\\.\\])(\\.\\)>\\s]\", post_text)))\n result[\"links\"].sort()\n \n communicated_match = re.search(communicated_by_regex, post_text, re.M)\n if communicated_match:\n result[\"communicatedBy\"] = communicated_match.group(\"communicated_by\")\n return result, header_end","def register_post_parser(self, fct, cfg, ctx):\n self.post_parsers.append((fct, cfg, ctx))","def __init__(self, post):\n self.id = int(post[0])\n self.type_id = int(post[1])\n try:\n self.parent_id = int(post[2])\n except:\n self.parent_id = None\n try:\n self.accepted_answer_id = int(post[3])\n except:\n self.accepted_answer_id = None\n self.creation_date = post[4]\n try:\n self.score = post[5]\n except:\n self.score = None\n try:\n self.view_count = post[6]\n except:\n self.view_count = 0\n self.body = post[7]\n try:\n self.owner_user_id = int(post[8])\n except:\n self.owner_user_id = None \n try:\n self.last_editor_user_id = int(post[9])\n except:\n self.last_editor_user_id = None\n try:\n self.last_editor_display_name = post[10]\n except:\n self.last_editor_display_name = None\n self.last_activity_date = post[11]\n self.last_edit_date = post[12]\n self.community_owned_date = post[13]\n self.closed_date = post[14]\n try:\n self.title = post[15]\n except:\n self.title = None\n try:\n self.tags = post[16]\n except:\n self.tags = None\n try:\n self.answer_count = int(post[17])\n except:\n self.answer_count = 0\n try:\n self.comment_count = int(post[18])\n except:\n self.comment_count = 0\n try:\n self.favorite_count = int(post[19])\n except:\n self.favorite_count = 0","def save(self, *args, **kwargs):\n\n if len(self.text) > 97:\n self.short_description = self.text[:97] + '...'\n else:\n self.short_description = self.text\n\n super(Post, self).save(*args, **kwargs)","def parse_posts(self):\n logger.info(\"Parsing posts\")\n\n self.df.title = self.df.title.str.strip()\n\n spam_companies = [\"Indeed Prime\"]\n self.df = self.df[~self.df[\"company\"].isin(spam_companies)]\n self.df = self.df.dropna(subset=[\"company\"])\n self.df = self.df.drop_duplicates(subset=[\"company\", \"date_posted\", \"title\"])","def parse_article(self, response):\n\n raw_post = response.css(\"div.blog.post > div.inner > div.row > article\")\n\n post_loader = ItemLoader(item=BlogPostItem(), selector=raw_post)\n post_loader.default_output_processor = TakeFirst()\n\n post_title = raw_post.css(\"div#postcontent > h1::text\").extract_first()\n post_loader.add_value(\"title\", post_title)\n post_loader.add_value(\"url\", response.request.url)\n\n post_text_selector = raw_post.css(\"div#postcontent > div#mypost\")\n post_text = post_text_selector.xpath('string(.)').extract_first()\n post_loader.add_value(\"content\", post_text[:160])\n\n pub_date_text = raw_post.css(\"div#postcontent > div.no-mobile > div.posttag.right.nomobile > span::text\").extract_first()\n pub_date = parse_date(pub_date_text)\n post_loader.add_value(\"publication_date\", pub_date)\n\n initial_author_list = raw_post.css(\n \"div#postcontent > div.no-mobile > div.postauthor > span > a.goauthor > span::text\").extract()\n author_list = [name.strip() for name in initial_author_list]\n post_authors = \"::\".join(author_list)\n post_loader.add_value(\"author\", post_authors)\n\n post_tags = raw_post.css(\"div#postcontent > a.tag.secondary::attr(title)\").extract()\n post_tags_str = \"::\".join(post_tags)\n post_loader.add_value(\"tags\", post_tags_str)\n\n return post_loader.load_item()","def parse_post(post):\n # Ignore posts less than 24 hours old\n if time.time() - post.created < 60 * 60 * 24:\n logger.debug('Ignoring post (too new)')\n return False\n\n # Add post to database\n postid_db = db.insert('Posts',\n (None,\n post.id,\n post.title,\n post.url,\n post.selftext,\n post.author,\n post.permalink,\n post.subreddit,\n post.num_comments,\n post.upvotes,\n post.downvotes,\n post.score,\n post.created_utc,\n int(post.is_self),\n int(post.over_18)))\n # If post already exists, we've already indexed it; skip!\n if postid_db == -1:\n logger.debug('Ignoring post (already indexed)')\n return False\n # Write post to DB so we don't hit it again\n\n # NOTE: postid_db is the ID of the post in the database; NOT on reddit\n\n # Check for self-post\n if post.selftext != '':\n urls = get_links_from_body(post.selftext)\n for url in urls:\n parse_url(url, postid=postid_db)\n else:\n # Attempt to retrieve hash(es) from link\n parse_url(post.url, postid=postid_db)\n\n # Iterate over top-level comments\n if post.num_comments > 0:\n reddit.fetch_comments(post)\n for comment in post.comments:\n parse_comment(comment, postid_db)","def newPost(self, useRawHTML):\n print\n content, publish = self._fillPost(useRawHTML)\n\n # Upload to server\n try :\n postid = self.server.metaWeblog.newPost(\n self.blogid, self.username, self.password,\n content, publish\n )\n except xmlrpclib.Fault as fault:\n display_XMLRPC_errors(\"post the new entry\", fault)\n import pdb\n pdb.set_trace()\n else :\n self._setCategorie(postid)\n print \"New post created with ID =\", postid","def inputPostFormatter(post):\n\n post[\"ttl\"] = from_decimal(post[\"ttl\"])\n post[\"workToProve\"] = from_decimal(post.get(\"workToProve\", 0))\n post[\"priority\"] = from_decimal(post[\"priority\"])\n\n if not is_array(post.get(\"topics\")):\n post[\"topics\"] = [post[\"topics\"]] if post.get(\"topics\") else []\n\n post[\"topics\"] = [topic if is_0x_prefixed(topic) else encode_hex(topic)\n for topic in post[\"topics\"]]\n\n return post","def post_process_post(self, post):\r\n post.article = self.rewrite_ob_urls(post.article)\r\n post._commit()\r\n \r\n comments = Comment._query(Comment.c.link_id == post._id, data = True)\r\n for comment in comments:\r\n comment.body = self.rewrite_ob_urls(comment.body)\r\n comment._commit()","def _parse_yaml_description(self, meta: Mapping):\n with contextlib.suppress(Exception):\n yaml_md = yaml.safe_load(meta[\"description\"])\n # YAML could be anything: plain string, list, …\n if isinstance(yaml_md, dict):\n meta.pop(\"description\")\n meta.update(yaml_md)","def change_post(mfp_form, request):\n article_id = int(mfp_form.cleaned_data['id'])\n newpost = APost(article_id)\n if newpost.exist:\n newpost.article.title = mfp_form.cleaned_data['title']\n # 缩略名\n newpost.article.short_title = mfp_form.cleaned_data['short_title']\n newpost.article.cover = mfp_form.cleaned_data['cover_url']\n newpost.article.introduction = mfp_form.cleaned_data['introduction']\n newpost.article.content = mfp_form.cleaned_data['content']\n # id为2是已发布的文章，默认为已发布，后面再改\n newpost.article.status = Status.objects.get(id=2)\n tagids = mfp_form.cleaned_data['tags']\n if len(tagids) != 0:\n for tagid in tagids:\n tagid = int(tagid)\n tag = Tags.objects.get(id=tagid)\n newpost.article.tags.add(tag)\n threadtypeid = int(mfp_form.cleaned_data['threadtypeid'])\n newpost.article.threadtypeid = ThreadTypes.objects.get(id=threadtypeid)\n if mfp_form.cleaned_data['commentnotshow'] != '':\n newpost.article.comment_status = False\n else:\n newpost.article.comment_status = True\n return newpost.article\n else:\n return False","def get_post(self):\n\t\tself.post = graph.get_object(POST_ID)","def tumblrPost(body,tags=None,title=None):\t\n\tglobal settings\n\tpost_address = \"http://api.tumblr.com/v2/blog/\"+settings['blog']+\"/post\"\n\tdata = {\"type\":\"text\",\"state\":settings['state'],\"body\" : body, \"oauth_token\":settings['oauth_token']}\n\tif tags != None:\n\t\tdata['tags'] = str(tags)\n\tif title != None:\n\t\tdata['title'] = str(title)\n\tconsumer = oauth.Consumer(settings['consumer_key'], settings['consumer_secret'])\n\ttoken = oauth.Token(settings['oauth_token'], settings['oauth_token_secret'])\n\tclient = oauth.Client(consumer,token)\n\tresp, content = client.request(post_address, \"POST\", urllib.urlencode(data))\n\tjs = json.loads(content)\n \tif js['meta']['status'] != 201: #something wrong\n\t\tprint js['meta']['msg']\n\t\tprint js","def _update_post_node(node, options, arguments):\n node[\"date\"] = arguments[0] if arguments else None\n node[\"tags\"] = options.get(\"tags\", [])\n node[\"author\"] = options.get(\"author\", [])\n node[\"category\"] = options.get(\"category\", [])\n node[\"location\"] = options.get(\"location\", [])\n node[\"language\"] = options.get(\"language\", [])\n node[\"redirect\"] = options.get(\"redirect\", [])\n node[\"title\"] = options.get(\"title\", None)\n node[\"image\"] = options.get(\"image\", None)\n node[\"excerpt\"] = options.get(\"excerpt\", None)\n node[\"exclude\"] = \"exclude\" in options\n node[\"nocomments\"] = \"nocomments\" in options\n node[\"external_link\"] = options.get(\"external_link\", [])\n return node","def parse_post(post: Dict[str, Any],\n image_retriever: str = \"pretrained\",\n image_basedir: Optional[str] = \"documentIntent_emnlp19/resnet18_feat\") -> Dict[str, Any]:\n id = post['id']\n label_intent = post['intent']\n label_semiotic = post['semiotic']\n label_contextual = post['contextual']\n caption = post['caption']\n\n if image_retriever == \"url\":\n image = post['url']\n raise NotImplementedError(\"Currently cannot download an image from {}\".format(image))\n elif image_retriever == \"pretrained\":\n image_path = Path(image_basedir) / \"{}.npy\".format(id)\n image = np.load(image_path)\n elif image_retriever == \"ignored\":\n image = None\n else:\n raise NotImplementedError(\"image_retriever method doesn't exist\")\n\n output_dict = {\n 'id': id,\n 'label': {\n 'intent': label_intent,\n 'semiotic': label_semiotic,\n 'contextual': label_contextual,\n },\n 'caption': caption,\n 'image': image,\n }\n\n return output_dict","def add_post_to_es(post, db_conn):\n\n from database.topic import get_topic, deliver_topic\n from database.user import get_user, deliver_user\n\n data = json_prep(deliver_post(post))\n topic = get_topic({'id': post['topic_id']}, db_conn)\n if topic:\n data['topic'] = json_prep(deliver_topic(topic))\n user = get_user({'id': post['user_id']}, db_conn)\n if user:\n data['user'] = json_prep(deliver_user(user))\n\n return es.index(\n index='entity',\n doc_type='post',\n body=data,\n id=post['id'],\n )","def add_post_into_ds(post, ds_group, ds_item_role):\n post.ds_group = ds_group\n post.ds_item_role = ds_item_role\n post.is_in_ds = True\n post.save()","def handle_meta(self, tag, attrs):\n ad = {}\n for tup in attrs:\n ad[tup[0]] = tup[1]\n if 'name' in ad.keys() \\\n and 'keywords' == ad['name'] \\\n and 'content' in ad.keys():\n self.filetype = ad['content']\n if 'name' in ad.keys() \\\n and 'description' == ad['name']:\n self.description = 'present'\n if 'charset' in ad.keys():\n self.charset = 'present'","def remotePostCreate(host, post):\n post = post.get('posts')[0]\n author = remoteAddAuthor(post.get('author'))\n title = post.get('title')\n description = post.get('description')\n contentType = post.get('contentType')\n content = post.get('content')\n published = utc.localize(datetime.strptime(post.get('published'), '%Y-%m-%dT%H:%M:%S.%fZ'))\n visibility = post.get('visibility')\n unlisted = post.get('unlisted')\n id = post.get('id')\n origin = post.get('origin')\n count = post.get('count')\n comments = remoteCommentList(post)\n source = \"{}/api/posts/{}\".format(DOMAIN, post.get('id'))\n\n post_dict = {'author': author, 'title': title, 'description': description,\n 'contentType': contentType, 'content': content, 'published': published,\n 'visibility': visibility, 'unlisted': unlisted, 'id': id,\n 'comments': comments, 'origin': origin, 'count': count,\n 'source': source}\n return post_dict","def new_post(mkp_form, request):\n newpost = Posts()\n newpost.init()\n newpost.authorid = int(request.user.id)\n newpost.title = mkp_form.cleaned_data['title']\n newpost.name = mkp_form.cleaned_data['short_title'] # 缩略名\n newpost.cover = mkp_form.cleaned_data['cover_url']\n newpost.introduction = mkp_form.cleaned_data['introduction']\n newpost.content = js_resize_img(mkp_form.cleaned_data['content'])\n newpost.status = Status.objects.get(id=2) # id为2是已发布的文章，默认为已发布，后面再改\n tagids = mkp_form.cleaned_data['tags']\n if len(tagids) != 0:\n for tagid in tagids:\n tagid = int(tagid)\n tag = Tags.objects.get(id=tagid)\n newpost.tags.add(tag)\n threadtypeid = mkp_form.cleaned_data['threadtypeid']\n newpost.threadtypeid = ThreadTypes.objects.get(id=threadtypeid)\n if mkp_form.cleaned_data['commentnotshow'] != '':\n newpost.comment_status = False\n else:\n newpost.comment_status = True\n return newpost","def add(self, posts):\n for post in posts:\n self._feed.add(FeedEntry(\n summary=post.summary,\n title=post.title,\n title_type='html',\n url=post.url,\n updated=post.date,\n ))","def update_post(request, pk):\n data = JSONParser().parse(request)\n\n if request.method == 'PATCH':\n if data['user_id']:\n user = User.objects.get(id = data['user_id'])\n else:\n user = request.user\n try:\n postObject = Post.objects.filter(id=pk, user = user).first()\n except Post.DoesNotExist:\n content = {'user forbidden': 'you should be user of the requested post.'}\n return Response(content, status=status.HTTP_403_FORBIDDEN)\n\n\n postObject.content = data['content']\n postObject.tags.clear()\n\n for tag in data[\"tags\"]:\n if len(tag)>0:\n if tag['name'] == '':\n continue\n try:\n tagObject = Tag.objects.get(wikidataID=tag['wikidataID'])\n except ObjectDoesNotExist:\n tagObject = Tag.objects.create(wikidataID=tag['wikidataID'], name=tag['name'])\n except MultipleObjectsReturned:\n return HttpResponse(\"Multiple tags exist for.\" + tag + \" Invalid State.\")\n\n unique_hidden_tags = list(set(tag['hidden_tags']))\n if unique_hidden_tags:\n tagObject.hidden_tags = unique_hidden_tags\n\n tagObject.save()\n postObject.tags.add(tagObject)\n postObject.save()\n return Response(status=status.HTTP_200_OK)","def on_frontmatter_loaded(self, source_file, frontmatter):\n if not self._is_post(frontmatter):\n return\n self._validate_post(source_file, frontmatter)\n post = BlogPost(\n date=frontmatter['date'],\n source_file=source_file,\n summary=frontmatter.get('summary', ''),\n title=frontmatter['title'],\n route=self._resolver.as_route(source_file),\n url=self._resolver.as_url(source_file),\n posts=self.posts,\n )\n frontmatter['post'] = post\n if post != self.posts.get(source_file):\n self.posts[source_file] = post\n self._should_generate = True","def markdown_post(post):\n post['entry'] = markdown(post['entry'].replace(\"\\n\",\" \\n\"), output=\"html5\")\n return post","async def upload_new_post(self, ctx: commands.Context, safety: str, *tags):\n safety = safety.lower()\n if safety not in ['safe', 'sketchy', 'unsafe']:\n raise ValueError(f\"Must specify safety: safe, sketchy, or unsafe\")\n anon = False\n urls = []\n sources = []\n tags = list(tags)\n if 'anonymous' in tags:\n anon = True\n tags.remove('anonymous')\n if 'anon' in tags:\n anon = True\n tags.remove('anon')\n for tag in tags:\n if '://' in tag:\n url = tag.strip('<>')\n urls.append(url)\n tags.remove(tag)\n # Checks to see if any of the tags being uploaded has implied tags that should also be applied to the post.\n for tag in tags:\n tagDL = await self.api_get(ctx, f\"/tag/{tag}\"}\n if 'name' in tagDL and tagDL['name'] == \"TagNotFoundError\":\n print(\"New Tag\")\n else:\n for implied_tag in tagDL['implications']:\n tags.append(implied_tag['names'][0])","def post_to_object(self, post, remove_id_prefix=False):\n id = post.get('id')\n if not id:\n return {}\n\n post_type = post.get('type')\n status_type = post.get('status_type')\n url = self.post_url(post)\n picture = post.get('picture')\n display_name = None\n message = (post.get('message') or post.get('story') or\n post.get('description') or post.get('name'))\n\n data = post.get('data', {})\n for field in ('object', 'song'):\n obj = data.get(field)\n if obj:\n id = obj.get('id')\n post_type = obj.get('type')\n url = obj.get('url')\n display_name = obj.get('title')\n\n object_type = OBJECT_TYPES.get(post_type)\n author = self.user_to_actor(post.get('from'))\n link = post.get('link', '')\n\n if link.startswith('/gifts/'):\n object_type = 'product'\n if not object_type:\n if picture and not message:\n object_type = 'image'\n else:\n object_type = 'note'\n\n obj = {\n 'id': self.tag_uri(str(id)),\n 'objectType': object_type,\n 'published': util.maybe_iso8601_to_rfc3339(post.get('created_time')),\n 'updated': util.maybe_iso8601_to_rfc3339(post.get('updated_time')),\n 'author': author,\n 'content': message,\n # FB post ids are of the form USERID_POSTID\n 'url': url,\n 'image': {'url': picture},\n 'displayName': display_name,\n 'fb_object_id': post.get('object_id'),\n }\n\n privacy = post.get('privacy', {})\n if isinstance(privacy, dict):\n privacy = privacy.get('value')\n if privacy is not None:\n # privacy value '' means it doesn't have an explicit audience set, so i\n # *think* it inherits from its parent. TODO: use that value as opposed to\n # defaulting to public.\n public = privacy.lower() in ('', 'everyone', 'open')\n obj['to'] = [{'objectType': 'group',\n 'alias': '@public' if public else '@private'}]\n\n # tags and likes\n tags = itertools.chain(post.get('to', {}).get('data', []),\n post.get('with_tags', {}).get('data', []),\n *post.get('message_tags', {}).values())\n obj['tags'] = [self.postprocess_object({\n 'objectType': OBJECT_TYPES.get(t.get('type'), 'person'),\n 'id': self.tag_uri(t.get('id')),\n 'url': self.object_url(t.get('id')),\n 'displayName': t.get('name'),\n 'startIndex': t.get('offset'),\n 'length': t.get('length'),\n }) for t in tags]\n\n obj['tags'] += [self.postprocess_object({\n 'id': self.tag_uri('%s_liked_by_%s' % (id, like.get('id'))),\n 'url': url,\n 'objectType': 'activity',\n 'verb': 'like',\n 'object': {'url': url},\n 'author': self.user_to_actor(like),\n 'content': 'likes this.',\n }) for like in post.get('likes', {}).get('data', [])]\n\n # \"See Original\" links\n post_actions = post.get('actions',[])\n see_orig_actions = (act for act in post_actions\n if act.get('name', '').lower() in SEE_ORIGINAL_ACTIONS)\n obj['tags'] += [self.postprocess_object({\n 'objectType': 'article',\n 'url': act.get('link'),\n 'displayName': act.get('name')\n }) for act in see_orig_actions]\n\n # is there an attachment? prefer to represent it as a picture (ie image\n # object), but if not, fall back to a link.\n att = {\n 'url': link if link else url,\n 'image': {'url': picture},\n 'displayName': post.get('name'),\n 'summary': post.get('caption'),\n 'content': post.get('description'),\n }\n\n if (picture and picture.endswith('_s.jpg') and\n (post_type == 'photo' or status_type == 'added_photos')):\n # a picture the user posted. get a larger size.\n att.update({\n 'objectType': 'image',\n 'image': {'url': picture[:-6] + '_o.jpg'},\n })\n obj['attachments'] = [att]\n elif link and not link.startswith('/gifts/'):\n att['objectType'] = 'article'\n obj['attachments'] = [att]\n\n # location\n place = post.get('place')\n if place:\n id = place.get('id')\n obj['location'] = {\n 'displayName': place.get('name'),\n 'id': id,\n 'url': self.object_url(id),\n }\n location = place.get('location', None)\n if isinstance(location, dict):\n lat = location.get('latitude')\n lon = location.get('longitude')\n if lat and lon:\n obj['location'].update({\n 'latitude': lat,\n 'longitude': lon,\n # ISO 6709 location string. details: http://en.wikipedia.org/wiki/ISO_6709\n 'position': '%+f%+f/' % (lat, lon),\n })\n elif 'location' in post:\n obj['location'] = {'displayName': post['location']}\n\n # comments go in the replies field, according to the \"Responses for\n # Activity Streams\" extension spec:\n # http://activitystrea.ms/specs/json/replies/1.0/\n comments = post.get('comments', {}).get('data')\n if comments:\n items = [self.comment_to_object(c) for c in comments]\n obj['replies'] = {\n 'items': items,\n 'totalItems': len(items),\n }\n\n return self.postprocess_object(obj)","def save_post(self, post):\n return self.collection.insert_one(post.serialize())","def mutate(self, info, **arguments):\n post_id = from_global_id(arguments.pop('id'))[1]\n post = PostModel.objects.get(pk=post_id)\n\n if not post:\n print(\"=\" * 100)\n print(\"schema.py@41\")\n print(\"Erro while updating post\")\n print(\"=\" * 100)\n return UpdatePostMutation(success=False)\n\n for k, v in arguments.items():\n setattr(post, k, v)\n post.save()\n return UpdatePostMutation(success=True, post=post)","def json_to_post(self, post, tag):\r\n # Get media type (image/video)\r\n mediaType = self.get_value(post, (\"images\", 0, \"type\"))\r\n if mediaType is None:\r\n mediaType = self.get_value(post, (\"type\", ))\r\n mediaType = self.get_media_type(mediaType)\r\n\r\n imageCount = self.get_value(post, ('images_count',))\r\n if imageCount is None:\r\n imageCount = 1\r\n \r\n # Only want 1 image/video\r\n if imageCount > 1:\r\n return None\r\n \r\n # Get media url and size\r\n if mediaType == MediaType.IMAGE:\r\n media = self.get_value(post, ('images', 0, 'id', ))\r\n size = self.get_value(post, ('size', ))\r\n if size is None:\r\n size = self.get_value(post, ('images', 0, 'size', ))\r\n if media is None:\r\n media = self.get_value(post, ('link', ))\r\n elif mediaType == MediaType.VIDEO:\r\n media = self.get_value(post, ('images', 0, 'id', ))\r\n size = self.get_value(post, ('mp4_size', ))\r\n if size is None:\r\n size = self.get_value(post, ('images', 0, 'mp4_size', ))\r\n if media is None:\r\n media = self.get_value(post, ('mp4', ))\r\n else:\r\n self.logger.log(logger.LogLevel.WARNING, 'mediaType is not Image or Video')\r\n return None\r\n\r\n #check if image/video is over max size\r\n if mediaType == MediaType.IMAGE:\r\n if size >= self.setting.maxImageSize:\r\n return None\r\n elif mediaType == MediaType.VIDEO:\r\n if size >= self.setting.maxVideoSize:\r\n return None\r\n\r\n postId = self.get_value(post, (\"id\", ))\r\n title = self.get_value(post, (\"title\", ))\r\n views = self.get_value(post, (\"views\", ))\r\n ups = self.get_value(post, (\"ups\", ))\r\n downs = self.get_value(post, (\"downs\", ))\r\n return Post(postId, title, mediaType.value, media, size, tag, views, ups, downs)","def setPostDown(self, post):\n # type: (tp.Union[str, tp.List[str]])->None\n if isinstance(post, list):\n self._ifAttributes['post-down'] = post\n elif isinstance(post, str):\n self._ifAttributes['post-down'] = [post]\n else:\n raise ValueError(\"Invalid value type {0}, expected str or List[str]\".format(type(post)))","def handle_new_post(post_data, user_agent, remote_addr):\n \n for required in POST_REQUIRED_PARAMS:\n if required not in post_data:\n return None, None\n\n try:\n value = int(string_from_interwebs(post_data.getfirst(\"code\", \"\")))\n except ValueError:\n return None, None\n \n if value != 98098098098:\n return None, None\n\n # not yet safe to use.\n location = post_data.getfirst(\"location\", \"\")\n tags = string_from_interwebs(post_data.getfirst(\"tags\")) \n author = post_data.getfirst(\"author\")\n \n split_tags = [string_from_interwebs(tag).strip().lower() for tag in tags.split(\",\")] # temporary\n \n if len(split_tags) > 3:\n return None, None\n \n author_id = string_from_interwebs(author).strip()\n \n with Connection('localhost', 27017) as connection:\n reply_to = string_from_interwebs(post_data.getfirst(\"reply_to\"))\n \n if not verify_author(author_id, connection):\n return None, None\n\n if not verify_post(reply_to, connection):\n return None, None\n\n # if reply then it's verified.\n # XXX: I need to make a standard object structure for this, so that I don't \n # have to update separate things.\n\n post = {\"viewed\" : 0,\n \"comments\" : 0,\n \"flagged\" : 0,\n \"disliked\" : 0,\n \"enjoyed\" : 0,\n \"num_replies\" : 0,\n \"num_reposts\" : 0,\n \"content-type\" : \"image\", # need to pull this from the mime lookup\n \"file\" : \"placeholder\",\n \"user_agent\" : user_agent,\n \"remote_addr\" : remote_addr,\n \"created\" : datetime.utcnow(),\n \"location\" : string_from_interwebs(location).strip(),\n \"author\" : ObjectId(author_id),\n \"reply_to\" : ObjectId(reply_to),\n \"tags\" : split_tags}\n\n update_post(reply_to, connection)\n\n return post_data.getfirst(\"data\"), post","def PostData(self, postdata):\n if len(postdata) > 0:\n for key in postdata:\n if postdata[key] == \"%TARGET%\":\n postdata[key] = self._target\n self._postdata = postdata\n else:\n self._postdata = None","def add_post(content):\n db = psycopg2.connect(database=DBNAME)\n c = db.cursor()\n cleaned = bleach.clean(content, strip = True)\n c.execute(\"insert into posts values(%s)\", (cleaned,))\n db.commit()\n db.close()","def process_postlist(app, doctree, docname):\n blog = Blog(app)\n if not blog:\n register_posts(app)\n for node in doctree.findall(PostList):\n colls = []\n for cat in [\"tags\", \"author\", \"category\", \"location\", \"language\"]:\n for coll in node[cat]:\n if coll in blog.catalogs[cat].collections:\n colls.append(blog.catalogs[cat].collections[coll])\n if colls:\n posts = set(blog.posts)\n for coll in colls:\n posts = posts & set(coll)\n posts = list(posts)\n posts.sort(reverse=True)\n posts = posts[: node.attributes[\"length\"]]\n else:\n posts = list(blog.recent(node.attributes[\"length\"], docname, **node.attributes))\n if node.attributes[\"sort\"]:\n posts.sort() # in reverse chronological order, so no reverse=True\n fmts = list(Formatter().parse(node.attributes[\"format\"]))\n not_in = {\"date\", \"title\", \"author\", \"location\", \"language\", \"category\", \"tags\", None}\n for text, key, __, __ in fmts:\n if key not in not_in:\n raise KeyError(f\"{key} is not recognized in postlist format\")\n excerpts = node.attributes[\"excerpts\"]\n expand = node.attributes[\"expand\"]\n date_format = node.attributes[\"date\"] or _(blog.post_date_format_short)\n bl = nodes.bullet_list()\n bl.attributes[\"classes\"].append(\"postlist-style-\" + node[\"list-style\"])\n bl.attributes[\"classes\"].append(\"postlist\")\n for post in posts:\n bli = nodes.list_item()\n bli.attributes[\"classes\"].append(\"ablog-post\")\n bl.append(bli)\n par = nodes.paragraph()\n bli.append(par)\n for text, key, __, __ in fmts:\n if text:\n par.append(nodes.Text(text))\n if key is None:\n continue\n if key == \"date\":\n par.append(nodes.Text(post.date.strftime(date_format)))\n else:\n if key == \"title\":\n items = [post]\n else:\n items = getattr(post, key)\n\n for i, item in enumerate(items, start=1):\n if key == \"title\":\n ref = nodes.reference()\n if item.options.get(\"external_link\"):\n ref[\"refuri\"] = post.options.get(\"external_link\")\n else:\n ref[\"refuri\"] = app.builder.get_relative_uri(docname, item.docname)\n ref[\"internal\"] = True\n ref[\"ids\"] = []\n ref[\"backrefs\"] = []\n ref[\"dupnames\"] = []\n ref[\"classes\"] = []\n ref[\"names\"] = []\n ref.append(nodes.Text(str(item)))\n par.attributes[\"classes\"].append(\"ablog-post-title\")\n else:\n ref = _missing_reference(app, item.xref, docname)\n par.append(ref)\n if i < len(items):\n par.append(nodes.Text(\", \"))\n if excerpts and post.excerpt:\n for enode in post.excerpt:\n enode = enode.deepcopy()\n enode.attributes[\"classes\"].append(\"ablog-post-excerpt\")\n revise_pending_xrefs(enode, docname)\n app.env.resolve_references(enode, docname, app.builder)\n enode.parent = bli.parent\n bli.append(enode)\n if expand:\n ref = app.builder.get_relative_uri(docname, post.docname)\n enode = nodes.paragraph()\n enode.attributes[\"classes\"].append(\"ablog-post-expand\")\n refnode = nodes.reference(\"\", \"\", internal=True, refuri=ref)\n innernode = nodes.emphasis(text=expand)\n refnode.append(innernode)\n enode.append(refnode)\n bli.append(enode)\n node.replace_self(bl)","def parse_post_values(self): \n self.parse_values(sys.stdin.read())","def __init__(self, title, text, tags, created, modified, modifier):\n self.title = title\n self.text = text\n self.tags = tags\n self.created = created\n self.modified = modified\n self.modifier = modifier","def process_posts_and_pages(*, posts, pages, settings):\n for post in posts:\n post.test_attr = 'post'\n for page in pages:\n page.test_attr = 'page'\n return {'posts': posts, 'pages': pages}","def analyse_posts(self, event):\n\n # Iterate over each post listed in \"deployed\"\n for post in event[\"deployed\"]:\n title = post.title()\n\n # Don't send for drafts\n if post.is_draft or post.post_status != \"published\":\n self.logger.info(\n \"Skipping Draft Post {0} with status {1}\".format(\n title, post.post_status\n )\n )\n continue\n\n # Extract some details\n link = post.permalink(absolute=True)\n text = post.text()\n self.logger.info(\"Processing {0}\".format(link))\n\n # Calculate and retrieve the state-key for this URL\n key = \"webmention-info-{0}\".format(link)\n observed_links = self.site.state.get(key)\n\n if not observed_links:\n observed_links = []\n\n # Extract links from within the rendered page\n links = self.extract_links(text)\n\n # Set up a requests session so that HTTP keep-alives can be used where possible\n # this reduces connection overhead etc\n session = requests.session()\n\n # Set the user-agent for all requests in this session\n session.headers.update({\"User-Agent\": \"Nikola SSG Webmention plugin\"})\n\n # Send mentions for each\n for dest in links:\n\n # See whether a webmention's already been sent for this page and url\n # means we won't reping links every time a post is updated\n if dest in observed_links:\n continue\n\n sent, has_mentions = self.send_webmention(link, dest, session)\n\n # We want to cache two categories of link\n #\n # Has webmentions, sent successfully\n # Does not have webmentions\n\n if sent or not has_mentions:\n observed_links.append(dest)\n\n # Now that all links have been processed, save the state\n self.site.state.set(key, observed_links)","def post(self):\n title = self.request.get(\"title\")\n body = self.request.get(\"body\")\n\n if title and body:\n\n # create a new Post object and store it in the database\n post = Post(\n title=title,\n body=body\n )\n post.put()\n\n # get the id of the new post, so we can render the post's page (via the permalink)\n id = post.key().id()\n self.redirect(\"/blog/%s\" % id)\n else:\n error = \"we need both a title and a body!\"\n #self.render_form(title, body, error)\n self.render(\"newpost.html\", title, body, error)","def load_post(path: Path) -> LocalPost:\n md_path = path / \"post.md\"\n if not md_path.exists() or not md_path.is_file():\n raise ValueError(f\"Could not find a post.md file in {path.absolute()}\")\n meta_path = path / \"post-meta.json\"\n if not meta_path.exists() or not meta_path.is_file():\n raise ValueError(f\"Could not find a post-meta.json file in {path.absolute()}\")\n try:\n post_meta = PostMeta.parse_from_file(meta_path)\n except ValueError as e:\n raise e\n return LocalPost(path, post_meta)","def createPost(request):\n\n #save the organization's post\n if request.method == 'POST':\n form = PostForm(request.user, request.POST, request.FILES)\n if form.is_valid():\n filterList = ['everyone', 'black', 'hispanic', 'female', 'lgbt', 'immigrants', 'disabled', 'poor'] \n newpost = form.save()\n\n #Add tags to the object only if in the filterlist\n tags = form.cleaned_data.get('tags')\n tags = [tag.lower() for tag in tags if tag.lower() in filterList]\n\n newpost.tags.add(*tags)\n messages.success(request, 'You have successful created the post')\n form = PostForm(request.user)\n context = {'form':form} \n return render(request, 'create_post.html', context=context)\n \n #form to fill out for the post\n form = PostForm(request.user)\n context = {'form':form} \n return render(request, 'create_post.html', context=context)","def process_post(new_post, cfg):\n id_already_handled_in_db = i18n['debug']['id_already_handled_in_db']\n discovered_submit_title = i18n['posts']['discovered_submit_title']\n rules_comment = i18n['posts']['rules_comment']\n yt_already_has_transcripts = i18n['posts']['yt_already_has_transcripts']\n\n if new_post['subreddit'] in cfg.upvote_filter_subs:\n # ignore posts if they don't meet the threshold for karma and the sub\n # is in our list of upvoted filtered ones\n if new_post['ups'] < cfg.upvote_filter_subs[new_post['subreddit']]:\n return\n\n if not is_valid(new_post['name'], cfg):\n logging.debug(id_already_handled_in_db.format(new_post['name']))\n return\n\n if new_post['archived']:\n return\n\n if new_post['author'] is None:\n # we don't want to handle deleted posts, that's just silly\n return\n\n logging.info(\n f'Posting call for transcription on ID {new_post[\"name\"]} posted by '\n f'{new_post[\"author\"]}'\n )\n\n if new_post['domain'] in cfg.image_domains:\n content_type = 'image'\n content_format = cfg.image_formatting\n\n elif new_post['domain'] in cfg.audio_domains:\n content_type = 'audio'\n content_format = cfg.audio_formatting\n\n elif new_post['domain'] in cfg.video_domains:\n if 'youtu' in new_post['domain']:\n if not valid_youtube_video(new_post['url']):\n add_complete_post_id(new_post['name'], cfg)\n return\n if get_yt_transcript(new_post['url']):\n np = cfg.r.submission(id=new_post['name'])\n np.reply(_(\n yt_already_has_transcripts\n ))\n add_complete_post_id(new_post['name'], cfg)\n logging.info(\n f'Found YouTube video, {get_yt_video_id(new_post[\"url\"])},'\n f' with good transcripts.'\n )\n return\n content_type = 'video'\n content_format = cfg.video_formatting\n else:\n # This means we pulled from a subreddit bypassing the filters.\n content_type = 'Other'\n content_format = cfg.other_formatting\n\n # Truncate a post title if it exceeds 250 characters, so the added\n # formatting still fits in Reddit's 300 char limit for post titles\n post_title = new_post['title']\n max_title_length = 250\n if len(post_title) > max_title_length:\n post_title = post_title[:max_title_length - 3] + '...'\n\n # noinspection PyBroadException\n try:\n result = cfg.tor.submit(\n title=discovered_submit_title.format(\n sub=new_post['subreddit'],\n type=content_type.title(),\n title=post_title\n ),\n url=reddit_url.format(new_post['permalink'])\n )\n result.reply(\n _(\n rules_comment.format(\n post_type=content_type,\n formatting=content_format,\n header=cfg.header\n )\n )\n )\n flair_post(result, flair.unclaimed)\n\n add_complete_post_id(new_post['name'], cfg)\n cfg.redis.incr('total_posted', amount=1)\n\n if cfg.OCR and content_type == 'image':\n # hook for OCR bot; in order to avoid race conditions, we add the\n # key / value pair that the bot isn't looking for before adding\n # to the set that it's monitoring.\n cfg.redis.set(new_post['name'], result.fullname)\n cfg.redis.rpush('ocr_ids', new_post['name'])\n\n cfg.redis.incr('total_new', amount=1)\n\n # The only errors that happen here are on Reddit's side -- pretty much\n # exclusively 503s and 403s that arbitrarily resolve themselves. A missed\n # post or two is not the end of the world.\n except Exception as e:\n logging.error(\n f'{e} - unable to post content.\\nID: {new_post[\"name\"]}\\n '\n f'Title: {new_post[\"title\"]}\\n Subreddit: '\n f'{new_post[\"subreddit\"]}'\n )","def cleanup_post(post):\n \n post_data = post\n post_data[\"id\"] = str(post[\"_id\"])\n post_data[\"author\"] = str(post[\"author\"])\n post_data[\"created\"] = str(post[\"created\"].ctime())\n del post_data[\"_id\"]\n \n if \"reply_to\" in post:\n post_data[\"reply_to\"] = str(post[\"reply_to\"])\n\n if \"repost_of\" in post:\n post_data[\"repost_of\"] = str(post[\"repost_of\"])\n\n return post_data","def parse(self, content):\n pass","def xmlpost_to_dict(post):\n\n tree = ET.parse(post)\n root = tree.getroot()\n msg = root.find('message')\n\n post_data = {}\n\n board_id = msg.find('board_id')\n post_data['board_id'] = int(board_id.text)\n\n root_post = msg.find('root').attrib['href']\n post_data['root_post'] = root_post.split('/')[-1]\n\n kudos = msg.find('kudos')\n count = kudos.find('count')\n post_data['kudos_count'] = int(count.text)\n\n edit_author_id = msg.find('last_edit_author').attrib['href']\n post_data['edit_author_id'] = int(edit_author_id.split('/')[-1])\n\n post_time = msg.find('post_time')\n post_data['post_time'] = post_time.text\n\n last_edit_time = msg.find('last_edit_time')\n post_data['last_edit_time'] = last_edit_time.text\n\n body = msg.find('body')\n post_data['body'] = body.text\n\n thread = msg.find('thread').attrib['href']\n post_data['thread'] = int(thread.split('/')[-1])\n\n board = msg.find('board').attrib['href']\n post_data['board'] = board.split('/')[-1]\n\n try:\n parent_post = msg.find('parent').attrib['href']\n post_data['parent_post'] = int(parent_post.split('/')[-1])\n except KeyError:\n post_data['parent_post'] = None\n\n views = msg.find('views')\n post_data['views'] = int(views.find('count').text)\n\n subject = msg.find('subject')\n post_data['subject'] = subject.text\n\n post_id = msg.find('id')\n post_data['post_id'] = int(post_id.text)\n\n author_id = msg.find('author').attrib['href']\n post_data['author_id'] = int(author_id.split('/')[-1])\n\n return post_data","def parse_article_text(article_text, post_date=datetime.datetime.now()):\n result = {}\n\n metadata_start = 0\n main_content_start = 0\n main_content_end = len(article_text)\n \n article_date_match = re.search(r\"^Date:\\s(?P<date>[^\\(\\[\\n]+)\", article_text, re.M)\n if article_date_match:\n # There may be more than one source date in summary articles.\n # Example: http://promedmail.org/direct.php?id=1073176\n # Summary articles are not a focus so currently only the first date\n # is recorded.\n source_date = parse_datetime(\n article_date_match.group(\"date\")\n )\n\n if source_date:\n result[\"date\"] = datetime_to_utc(source_date)\n metadata_start = min(article_date_match.start(), metadata_start)\n main_content_start = max(article_date_match.end(), main_content_start)\n # The year is checked to avoid typos like 200_ that throw\n # the date off by a large factor.\n # Example: http://www.promedmail.org/direct.php?id=45850 (article 2)\n if result[\"date\"].year < 1900:\n result[\"date\"] = None\n # Some articles have timestamps that are incorrectly parsed.\n # Current examples:\n # http://www.promedmail.org/direct.php?id=43918\n # http://www.promedmail.org/direct.php?id=2200173\n # Some of these incorrect timestamps can be removed by verifying that\n # they preceed the time of the posting. A day of slop time is allowed\n # to account for variations due to incorrect timezones.\n elif result[\"date\"] > post_date + datetime.timedelta(1):\n result[\"date\"] = None\n else:\n result[\"date\"] = None\n \n source_match = re.search(r\"Source:\\s(?P<name>[^\\[\\n]+)\" +\\\n r\"(\\s(?P<edits>\\[.*))?\" +\\\n r\"\\n\" +\\\n r\"(?P<url>http.+)?\", article_text)\n if source_match:\n result[\"source\"] = source_match.groupdict()\n metadata_start = min(source_match.start(), metadata_start)\n main_content_start = max(source_match.end(), main_content_start)\n \n heading_match = re.search(r\"^(?P<idx>\\[\\d\\]\\s)?\" +\\\n r\"(?P<heading>\\S+.*)\\n\",\n article_text[0:metadata_start], re.M)\n if heading_match:\n result[\"heading\"] = heading_match.group(\"heading\")\n \n communicated_match = re.search(communicated_by_regex, article_text, re.M)\n if communicated_match:\n result[\"communicatedBy\"] = communicated_match.group(\"communicated_by\")\n main_content_end = min(communicated_match.start(), main_content_end)\n \n result[\"content\"] = article_text[main_content_start:main_content_end].strip()\n return result","def postCreate(post):\n post_list = list()\n comments = commentList(post)\n comment_url = \"{}/api/posts/{}/comments\".format(DOMAIN, post.id)\n visible_to = list()\n visible = post.visibleTo.all()\n if visible:\n for author in visible:\n auth = \"{}/api/author/{}\".format(DOMAIN, author.id)\n visible_to.append(auth)\n\n # visible_to = list(post.visibleTo)\n post_dict = {'author': addAuthor(post.author), 'title': post.title, 'description': post.description,\n 'contentType': post.contentType, 'content': post.content, 'published': post.published,\n 'visibility': post.visibility, 'visibleTo': visible_to, 'unlisted': post.unlisted, 'id': post.id,\n 'comments': comments[:5], 'next': comment_url, 'count': len(comments),\n 'source': \"{}/api/posts/{}\".format(DOMAIN, post.id),\n 'origin': \"{}/api/posts/{}\".format(DOMAIN, post.id)}\n post_list.append(post_dict)\n return post_list","def addPost(self,text,id,url,date):\n self.topComments.append(Post(text,id,url,date))\n return None","def mutate(self, info, **arguments):\n title = arguments.get('title')\n text = arguments.get('text')\n\n # Get author id from `relay` id and find author on database\n author_id = from_global_id(arguments.get('author_id'))[1]\n author = UserModel.objects.get(pk=author_id)\n\n # Get post id from `relay` id and find it on database\n post_id = from_global_id(arguments.pop('id'))[1]\n post = PostModel.objects.get(pk=post_id)\n\n if not author:\n print(\"=\" * 100)\n print(\"schema.py@41\")\n print(\"We should return an error here\")\n print(\"=\" * 100)\n return CreatePostMutation(success=False)\n\n post = PostModel.objects.create(title=title, author=author, text=text)\n return CreatePostMutation(success=True, post=post)","def set_date_post(self, date_post):\n if check_data_exist(date_post) is True:\n self.date_post = date_post.text","def wp2fields(xml, wp_custpost=False):\r\n\r\n items = get_items(xml)\r\n for item in items:\r\n\r\n if item.find('status').string == \"publish\":\r\n\r\n try:\r\n # Use HTMLParser due to issues with BeautifulSoup 3\r\n title = HTMLParser().unescape(item.title.contents[0])\r\n except IndexError:\r\n title = 'No title [%s]' % item.find('post_name').string\r\n logger.warning('Post \"%s\" is lacking a proper title' % title)\r\n\r\n filename = item.find('post_name').string\r\n post_id = item.find('post_id').string\r\n filename = get_filename(filename, post_id)\r\n\r\n content = item.find('encoded').string\r\n raw_date = item.find('post_date').string\r\n date_object = time.strptime(raw_date, \"%Y-%m-%d %H:%M:%S\")\r\n date = time.strftime(\"%Y-%m-%d %H:%M\", date_object)\r\n author = item.find('creator').string\r\n\r\n categories = [cat.string for cat in item.findAll('category', {'domain' : 'category'})]\r\n # caturl = [cat['nicename'] for cat in item.find(domain='category')]\r\n\r\n tags = [tag.string for tag in item.findAll('category', {'domain' : 'post_tag'})]\r\n\r\n kind = 'article'\r\n post_type = item.find('post_type').string\r\n if post_type == 'page':\r\n kind = 'page'\r\n elif wp_custpost:\r\n if post_type == 'post':\r\n pass\r\n # Old behaviour was to name everything not a page as an article.\r\n # Theoretically all attachments have status == inherit so\r\n # no attachments should be here. But this statement is to\r\n # maintain existing behaviour in case that doesn't hold true.\r\n elif post_type == 'attachment':\r\n pass\r\n else:\r\n kind = post_type\r\n yield (title, content, filename, date, author, categories, tags,\r\n kind, \"wp-html\")","def post(self):\n args = parser.parse_args()","def newPost(self, postLink, zserverBlogEntry): #$NON-NLS-1$\r\n atomEntry = self.createNewBlogEntry()\r\n self._populateAtomEntry(atomEntry, zserverBlogEntry)\r\n # publish entry\r\n atomRespEntry = self.createAtomEntry(postLink, atomEntry)\r\n return atomRespEntry","def __init__(self, rpc, post):\n super(PostModule, self).__init__(rpc, 'post', post)\n self._action = self._info.get('default_action', \"\")","def create(self, validated_data):\n \"\"\" Create post with a location \"\"\"\n location_data = validated_data.pop('location')\n\n # create a new one or get a old for reference\n this_location = Location.objects.get_or_create(\n **location_data\n )\n\n # pop the photo url's data\n photo_data = validated_data.pop('photo')\n\n # must pop the tags data before it would used to create a post \n tags_data = validated_data.pop('tag')\n # create a instance of this post\n this_post = Post.objects.create(\n location = this_location[0],\n **validated_data)\n\n \"\"\"Associate tag's informatiion to post\"\"\"\n for tag in tags_data:\n this_tag = Tag.objects.get_or_create(name = tag.get('name'))\n print(tag.get('name'))\n print(this_tag)\n # attach this tag to this photos_datapost \n this_post.tag.add(this_tag[0])\n\n \"\"\"Associate the photo url \"\"\"\n for photo in photo_data:\n this_post.photo.create(name = photo.get('name'))\n # return the created post \n this_post.save()\n return this_post","def create(cls, headline, text, blog):\n post = cls()\n post.headline = headline\n post.text = text\n post.blog = blog\n post.posted_date = timezone.now()\n try:\n post.save()\n return post\n except(ValueError, IntegrityError, OperationalError):\n return None","def add_post(content):\n db = psycopg2.connect(\"dbname=forum\")\n c = db.cursor()\n content = bleach.clean(content)\n c.execute(\"insert into posts values (%s)\", (content,))\n db.commit()\n db.close()\n # POSTS.append((content, datetime.datetime.now()))","def adapt_tweet(feedpost):\n tweet = feedpost['title']\n for action in (_make_links, _clean_name, urlize):\n tweet = action(tweet)\n feedpost['title'] = _get_tweet_number(feedpost['link'])\n feedpost['body'] = u'<p>%s</p>' % tweet\n return feedpost","def each_comment_from_post(post):\n # first yield the post text body, if any\n if post['text']:\n yield post['text']\n # then yield each comment\n for comment in post['comments']:\n yield comment['text']","def post_create(faker_obj, profile_obj, tag_list, num=3):\n for i in range(num):\n obj = faker_obj\n title = obj.sentence(nb_words=random.randint(5, 10))\n author = User.objects.get(id=profile_obj)\n body = \" \".join(obj.paragraphs(nb=random.randint(8, 20)))\n status = \"published\"\n post = Post.objects.create(title=title, author=author, body=body, status=status)\n post.tags.add(\", \".join(random.sample(tag_list, 1)))\n print(\n \"Created post title:'{}' for user '{}'\".format(post.title, author.username)\n )\n create_comment_list(obj, post)","def setUp(self):\n self.xpaths = [\"./author/text()\",\n \"./title/text()\",\n \"./time/text()\",\n \"./number/text()\",\n \"./tags/tag/text()\",\n \" \"]\n self.meta = {\"Time\": \"2012-02-23\",\n \"Author\": \"rachel\",\n \"Title\": \"Post 1\",\n \"Number\": \"1\",\n \"Tag\": [\"Tag 0\", \"Tag 3\"]}\n self.sentence_contents = \"This is the text of post 1. I love clouds.\"\n super(PostTests, self).setUp(\n \"tests/data/articles/\", \"structure.json\", \"post1.xml\")","def _update_post_entry(self, target, source):\n target.post_id = target.post_id if source.id==source.id else source.id\n # target.author = target.author if source.author==source.author else source.author\n target.author_premium = target.author_premium if source.author_premium==source.author_premium else source.author_premium\n target.subreddit_subscribers = target.subreddit_subscribers if source.subreddit_subscribers==source.subreddit_subscribers else source.subreddit_subscribers\n target.title = target.title if source.title==source.title else source.title\n target.downs = target.downs if source.downs==source.downs else source.downs\n target.ups = target.ups if source.ups==source.ups else source.ups\n target.selftext = target.selftext if source.selftext==source.selftext else source.selftext\n target.num_comments = target.num_comments if source.num_comments==source.num_comments else source.num_comments\n target.total_awards_received = target.total_awards_received if source.total_awards_received==source.total_awards_received else source.total_awards_received\n target.view_count = target.view_count if source.view_count==source.view_count else source.view_count\n # target.permalink = target.permalink if source.permalink==source.permalink else source.permalink\n # target.url = target.url if source.url==source.url else source.url\n # target.created = target.created if source.created==source.created else source.created\n # target.created_utc = target.created_utc if source.created_utc==source.created_utc else source.created_utc\n\n return target","def parse_text(self, page):\n text = page.find(self.tag_prefix + self.revision_tag).find(self.tag_prefix + self.text_tag).text\n title = page.find(self.tag_prefix + self.title_tag).text\n categories = []\n #\n text = self.parse_archivo(text)\n text = self.parse_foto(text)\n text = self.parse_by_line(text)\n text = self.parse_link(text)\n text = self.parse_url(text)\n text = self.parse_fecha(text)\n text = self.parse_bracketed_word(text)\n #\n if text:\n categories = re.findall(self.category_finder_regex, text)\n #\n text = self.parse_category(text)\n text = self.parse_other_language(text)\n text = self.parse_table_regex(text)\n text = self.parse_ver_fuente(text)\n text = self.remove_extra_text(text)\n text = self.remove_extra_characters(text)\n\n categorias = []\n for cat in categories:\n categorias.append(cat[6])\n\n if text:\n if 'REDIRECT' in text or 'redirect' in text:\n return None\n\n return Article(title=title, content=text, categories=categorias)","def get_post_attrs(post_content, **kwargs):\n ref_filters = kwargs[\"ref_filters\"]\n # Attributes without a recognizable ID will be appended to \"misc\".\n post_attrs = {\"misc\": []}\n for attr in get_attrs(post_content):\n # Add to post_attrs !! BY REFERENCE !!.\n parse_attrs(post_attrs, attr, ref_filters)\n\n return post_attrs","def _parse(self):\n soup = BS(self._current_html, 'lxml')\n for item in soup.select('div.c'):\n temp = {}\n # main content\n ctt = item.select('span.ctt')\n if not ctt:\n continue\n weibo_body = item.select('div')\n if len(weibo_body) > 1:\n temp['content'] = weibo_body[0].text\n btn_group = weibo_body[1].text\n else:\n temp['content'] = weibo_body[0].select('span.ctt')[0].text\n btn_group = weibo_body[0].text\n temp['is_repost'] = True if REPO_TEST_PATTERN.match(\n temp['content']) else False\n try:\n temp['like_num'] = LIKE_NUM_PATTERN.findall(btn_group)[0]\n temp['cmt_num'] = COMMENT_NUM_PATTERN.findall(btn_group)[0]\n temp['repo_num'] = REPO_NUM_PATTERN.findall(btn_group)[0]\n except Exception:\n pass\n cmt = item.select('.cmt')\n # visibility\n if cmt:\n try:\n temp['visibility'] = VISIBILITY_PATTERN.findall(\n cmt[0].text)[0]\n except Exception:\n pass\n\n # img in main content\n img = item.select('div a img')\n img_src = img[0].attrs['src'] if img else None\n temp['img_src'] = img_src\n LOGGER.debug('img_src: {}'.format(img_src))\n # time & source device\n ct = item.select('span.ct')\n if ct:\n ct = ct[0]\n text = ct.text\n reg_result = TIME_PATTERN.findall(text)[0]\n\n temp['time'] = ar(\n '{}年{}'.format(self._current_year, reg_result[0]),\n DATE_FMTS[0]\n ).naive if reg_result[0] else ar(\n reg_result[1], DATE_FMTS[1]\n ).naive\n temp['source'] = SOURCE_DEVICE_PATTERN.findall(text)[0]\n self._post_item = Post(**temp)\n self._attachment_item = Attachment(\n uri=img_src, post=self._post_item)\n self._store()","def process_entry(entry, blog, START):\n try:\n when = entry['updated_parsed']\n except KeyError:\n try:\n when = entry['published_parsed']\n except KeyError:\n return # Ignore undateable posts\n\n if when:\n when = pytz.timezone('UTC').localize(datetime.fromtimestamp(time.mktime(when)))\n else:\n # print blog, entry\n return\n\n if when < START:\n return\n\n title = entry.get('title', \"Null\")\n\n try:\n author = entry['author']\n except KeyError:\n try:\n author = ', '.join(a['name'] for a in entry.get('authors', []))\n except KeyError:\n author = 'Anonymous'\n\n link = entry['link']\n\n try:\n body = entry['content'][0]['value'].replace(\"h1>\",\"b>\").replace(\"h2>\",\"b>\")\n except KeyError:\n body = entry['summary']\n\n return Post(when, blog, title, author, link, body)","def parse_posts(posts_dict):\n return posts_dict['posts']","def insert_post(data, db_conn):\n\n schema = get_post_schema(data)\n data, errors = insert_document(schema, data, db_conn)\n if not errors:\n add_post_to_es(data, db_conn)\n return data, errors","def test_post(self):\n self.assertEqual(self.post.title, \"Ugandan anthem\")\n self.assertEqual(self.post.body, \"George William Kakoma\")\n self.assertEqual(str(self.post), \"Ugandan anthem - George William Kakoma\")","def post(self):\n subject = self.request.get('subject')\n content = self.request.get('content')\n\n # if user enter good subject and content, redirect them to new post page\n if subject and content:\n p = Post(parent = blog_key(), subject = subject, content = content)\n p.put() # store the post element into database\n self.redirect('/blog/%s' % str(p.key().id()))\n # otherwise, render an error page \n else:\n error = \"subject and content, please!\"\n self.render(\"newpost.html\", subject=subject, content=content, error=error)","def add_to_postre(self, post_course):\n self.postre[post_course.get_course_id()] = post_course","def test_parse_post_code_field(self):\n fields = {'Post code': {'offset': 171,\n 'length': 4}}\n p = top.Parser(fields=fields)\n received = p.parse_line(self._line)\n expected = {'Post code': '2048'}\n msg = 'Post code field parse incorrect'\n self.assertEqual(received, expected, msg)","def editPost(self, id, useRawHTML):\n old_data = self._extractPost(id)\n print\n content, publish = self._fillPost(useRawHTML, old_data)\n\n # Upload to server\n try :\n self.server.metaWeblog.editPost(\n id, self.username, self.password,\n content, publish\n )\n if raw_input(\"Change category ?[y|N] \") == \"y\" :\n self._setCategorie(id)\n except xmlrpclib.Fault as fault:\n display_XMLRPC_errors(\"edit entry\", fault)","def sortByPostText(self,post_text,desc):\n \n return self.__sortByPostText(post_text, \"\", desc)","def proceed(self, source):\n\n for line in source:\n line = line.rstrip()\n if istag(line) and isinstance(maketag(line), Time): # here, handle Time tags\n self.time_annotate_date(maketag(line)) # annotate Date tag and queue Time tag\n if not default_options.mdjpp_time_propagate:\n continue\n # not a tag!\n if not istag(line) or isinstance(maketag(line), Time):\n if self.metastable_tag: # and len(line):\n # open metastable tag\n self.open_tag(self.metastable_tag)\n self.metastable_tag = None\n # print line\n if self.printable():\n self.printme_text(line)\n # tag!\n else:\n # make regular tag\n this_tag = maketag(line)\n # no metastable tag\n if not self.metastable_tag:\n # make new metastable tag\n self.metastable_tag = this_tag\n # does new tag close last tag?\n self.close_tag(this_tag)\n # metastable tag present\n else:\n # is metastable tag or this tag of Date type?\n if isinstance(self.metastable_tag, Date) or isinstance(this_tag, Date):\n # open metastable tag\n self.open_tag(self.metastable_tag)\n # recreate metastable tag with new tag\n self.metastable_tag = this_tag\n # does new tag close last tag?\n self.close_tag(this_tag)\n else:\n # add new tag to metastable tag\n self.metastable_tag.append(this_tag)","def parse_post_category(advertise: Dict[str, Any]) -> Optional[List[str]]:\n\n noise_terms: List[str] = [\"voltar\"]\n\n if \"post_category\" in advertise.keys():\n tmp: List[str] = re.split('[ \\t]{2,}', advertise[\"post_category\"])\n tmp = Parser.remove_noise_terms(tmp, noise_terms)\n return tmp"],"string":"[\n \"def add_meta(self, post, *args, **kwargs):\\n\\t\\tsuper(Command, self).add_meta(post, *args, **kwargs)\\n\\t\\tpost.gen_description = False\\n\\t\\tpost.description = description_from_content(post)\\n\\t\\tpost.save()\",\n \"def post(self, post_id=None):\\n\\n if post_id:\\n abort(400)\\n else:\\n args = parsers.post_post_parser.parse_args(strict=True)\\n\\n new_post = Post(args['title'])\\n new_post.text = args['text']\\n # new_post.user = user\\n\\n if args['tags']:\\n for item in args['tags']:\\n tag = Tag.query.filter_by(name=item).first()\\n # If the tag already exist, append.\\n if tag:\\n new_post.tags.append(tag)\\n # If the tag not exist, create the new one.\\n # Will be write into DB with session do.\\n else:\\n new_tag = Tag(item)\\n new_post.tags.append(new_tag)\\n db.session.add(new_post)\\n db.session.commit()\\n return (new_post.id, 201)\",\n \"def preprocess_post(self, post):\\n # tokenize, clean, & tag part-of-speech for all words\\n if self.document_level == 'postwise':\\n\\n doc_text = all_comments_from_post(post)\\n # leave early if there's nothing there\\n if doc_text == '':\\n return []\\n\\n tokens = nltk.word_tokenize(doc_text)\\n # TODO: skip this if there's no POS filtering args!\\n tagged = nltk.pos_tag(tokens)\\n\\n # filter out most invalid words with valid_word()\\n processed_document = []\\n for word, pos_tag in tagged:\\n if self.valid_word(word, pos_tag):\\n cleaned_word = self.clean_word(word)\\n # things like digits and other junk become empty string,\\n # so exclude them from final document\\n if cleaned_word:\\n processed_document.append(cleaned_word)\\n # finally, update the post\\n post['postwise'] = {'tokens': processed_document, 'text': doc_text}\\n self.postman.posts_write.update_one({'_id':post['_id']}, {'$set':post}, upsert=True)\\n else:\\n raise NotImplementedError('document_level: \\\"%s\\\"' % self.document_level)\\n\\n return processed_document\",\n \"def setPostUp(self, post):\\n # type: (tp.Union[str, tp.List[str]])->None\\n if isinstance(post, list):\\n self._ifAttributes['post-up'] = post\\n elif isinstance(post, str):\\n self._ifAttributes['post-up'] = [post]\\n else:\\n raise ValueError(\\\"Invalid value type {0}, expected str or List[str]\\\".format(type(post)))\",\n \"def refactor_post(self,post_name):\\n for name in list(self.rules):\\n related_post = \\\"{}.post.{}\\\".format(name,post_name)\\n if related_post in self.rules:\\n parts = [self.MakeSymbolName(x) for x in [post_name, related_post]]\\n self.rules[name] = self.MakeChoice([self.MakeSeq(parts)])\",\n \"def parse_post_text(formatted_content):\\n post = {}\\n # Parse Mod comments and remove them from the text.\\n potential_comments = re.finditer(\\\"\\\\[.+?\\\\]\\\", formatted_content, re.DOTALL)\\n comments = []\\n for comment_match in potential_comments:\\n comment = comment_match.group()\\n mod = re.search(r\\\"\\\\-\\\\s?Mod\\\\.\\\\s?(?P<mod>\\\\w+\\\\b)\\\", comment)\\n if mod:\\n comments.append({\\n \\\"comment\\\" : comment,\\n \\\"mod\\\" : mod.group(\\\"mod\\\")\\n })\\n post[\\\"modComments\\\"] = comments\\n \\n # Comments are removed from the post test so that\\n # links, reports, etc. mentioned by mods are not extracted.\\n no_comment_txt = formatted_content\\n for comment in comments:\\n no_comment_txt = no_comment_txt.replace(comment[\\\"comment\\\"], \\\"\\\")\\n \\n metadata, header_end = parse_post_metadata(no_comment_txt)\\n post.update(metadata)\\n \\n sections = re.split(r\\\"^[\\\\*#]{3,}\\\\s*$\\\", no_comment_txt[header_end:], flags=re.M)\\n articles = []\\n \\n # Some posts have articles which are parsed into multiple sections:\\n # Ex: http://www.promedmail.org/direct.php?id=2194235\\n # The section parsing code tries to recombine these by concatenating\\n # unrecognized sections onto the previous sections if they form an article.\\n # article_start_idx keeps track of the first section in the article.\\n article_start_idx = None\\n \\n for idx, section in enumerate(sections):\\n section = section.strip()\\n article = parse_article_text(section, post_date=post['promedDate'])\\n # Check if the section contains an actual article by seeing which\\n # properties could be parsed.\\n if article.get('source') or article.get('date'):\\n articles.append(article)\\n article_start_idx = idx\\n else:\\n # When a section cannot be parsed as an article the following code\\n # tries to determine what it is. If the type cannot be determined\\n # an error or warning is thrown.\\n # These warnings can be used to find sections which are not being\\n # correctly parsed.\\n # Posts with known issues:\\n # http://www.promedmail.org/direct.php?id=19990512.0773\\n if re.search(r\\\"Visit ProMED-mail\\\\'s web site at|\\\"\\n r\\\"Please support (the \\\\d{4}\\\\s)?ProMED\\\\-mail|\\\"\\n r\\\"Donate to ProMED\\\\-mail. Details available at|\\\"\\n r\\\"ProMED\\\\-mail makes every effort to verify the reports|\\\"\\n r\\\"PROMED\\\\-MAIL FREQUENTLY ASKED QUESTIONS|\\\"\\n r\\\"Become a ProMED\\\\-mail Premium Subscriber|\\\"\\n r\\\"A ProMED\\\\-mail post\\\",\\n section, re.I):\\n # boilerplate promed notice section\\n pass\\n elif re.search(r\\\"In this (update|post(ing)?)\\\", section):\\n # table of contents section\\n pass\\n elif re.search(r\\\"Cases in various countries\\\", section):\\n # This type of post typically has links to several articles\\n # with single sentence summaries.\\n # Ex: http://www.promedmail.org/direct.php?id=20131125.2073661\\n pass\\n elif section == \\\"\\\":\\n # empty section\\n pass\\n elif idx == 0 and section.count(\\\"\\\\n\\\") < 2:\\n # probably the article title\\n pass\\n else:\\n if article_start_idx != None:\\n article = parse_article_text(\\n \\\"\\\\n#####\\\\n\\\".join(\\n sections[article_start_idx:idx]).strip(),\\n post_date=post['promedDate'])\\n assert article.get('source') or article.get('date')\\n articles[-1] = article\\n continue\\n else:\\n print \\\"Unexpected Section (%s):\\\" % post['archiveNumber'], [section[0:50] + \\\"...\\\"]\\n article_start_idx = None\\n post['articles'] = articles\\n return post\",\n \"def add_post(self, post: Post) -> None:\\n self.post_process.append(post)\",\n \"def post_to_activity(self, post):\\n id = None\\n if post.get('id'):\\n # strip USERID_ prefix if it's there\\n post['id'] = post['id'].split('_', 1)[-1]\\n id = post['id']\\n\\n obj = self.post_to_object(post)\\n activity = {\\n 'verb': VERBS.get(post.get('type', obj.get('objectType')), 'post'),\\n 'published': obj.get('published'),\\n 'updated': obj.get('updated'),\\n 'id': self.tag_uri(id) if id else None,\\n 'url': self.post_url(post),\\n 'actor': obj.get('author'),\\n 'object': obj,\\n }\\n\\n application = post.get('application')\\n if application:\\n activity['generator'] = {\\n 'displayName': application.get('name'),\\n 'id': self.tag_uri(application.get('id')),\\n }\\n return self.postprocess_activity(activity)\",\n \"def update_post_format(post):\\n\\n post_dict = {\\n \\\"title\\\": post[1],\\n \\\"genre\\\": get_genre(post[0]),\\n \\\"content\\\": post[2],\\n \\\"repeater_link\\\": get_links(post[3], post[4]),\\n }\\n \\n return post_dict\",\n \"def _fillPost(self, useRawHTML, old_data=None):\\n # Initialize empty dictionnary ct (aka content)\\n # to be sent through self.server.metaWeblog.newPost()\\n ct = {}\\n\\n # if no old_data, create a fake one\\n if old_data == None:\\n old_data = { 'title': None\\n , 'mt_keywords': None\\n , 'formatted_text': BALISE\\n , 'mt_excerpt': None\\n , 'description': None}\\n\\n def updateField(prompt, string=None):\\n if (string == None) or (string == \\\"\\\") :\\n return raw_input(prompt)\\n else :\\n r = raw_input(prompt + \\\" [default:\\\" + string + \\\"]\\\\n\\\")\\n if r == \\\"\\\" :\\n return string\\n else :\\n return r\\n\\n # Now get information\\n ct['title'] = updateField( \\\"Title?\\\\n\\\", old_data['title'])\\n ct['mt_keywords'] = updateField(\\n \\\"Tags? (comma separated lists)?\\\\n\\\",\\n old_data['mt_keywords'])\\n\\n # Categories are not included in the struct \\\"ct\\\"\\n # see _setCategorie()\\n\\n # Get excerpt/content\\n # Method0: external XML file\\n if useRawHTML:\\n with open( useRawHTML, 'rb') as f:\\n doc = xml.dom.minidom.parse(f)\\n # Parse our XHTML file\\n text = doc.getElementsByTagName(\\\"body\\\")[0].toxml()\\n #text = text.decode() # convert bytes to string\\n text = text.replace(\\\"<body>\\\", \\\"\\\").replace(\\\"</body>\\\", \\\"\\\")\\n ct['mt_excerpt'], ct['description'] = split_excerpt( text)\\n\\n # Method1: custom editor\\n elif self.editor :\\n prev_data = old_data['formatted_text']\\n data = self._externalEditor( wrap_with_template(prev_data) )\\n ct['mt_excerpt'], ct['description'] = split_excerpt( data)\\n\\n # Method2: input\\n else :\\n ct['mt_excerpt'] = updateField(\\n \\\"Excerpt? (beware of xHTML tags !)\\\\n\\\",\\n old_data['mt_excerpt'])\\n ct['description'] = updateField(\\n \\\"Main content? (beware of xHTML tags !)\\\\n\\\",\\n old_data['description'])\\n\\n # Process the rest of the attributes (comments, pings, ...)\\n def set_boolean( prompt, default):\\n if default == True:\\n return raw_input(prompt + \\\"[Y|n]\\\") != \\\"n\\\"\\n else:\\n return raw_input(prompt + \\\"[y|N]\\\") != \\\"y\\\"\\n\\n ct['mt_allow_comments'] = set_boolean( \\\"Allow comments ?\\\"\\n , self.auto_comments)\\n ct['mt_allow_pings'] = set_boolean( \\\"Allow pings ?\\\"\\n , self.auto_pings)\\n publish = set_boolean( \\\"Publish ?\\\" , self.auto_publish)\\n\\n return ct, publish\",\n \"def post(self, post):\\n\\n self._post = post\",\n \"def post(self, post):\\n\\n self._post = post\",\n \"def _populate_posts(self, channel, url):\\n import feedparser\\n\\n Post = get_model('articles', 'Post')\\n Image = get_model('images', 'Image')\\n\\n parser = feedparser.parse(url)\\n\\n for entry in parser['entries']:\\n # Some entries are incomplete and have only the title, need to\\n # ignore these entries.\\n if not entry.get('summary'):\\n continue\\n\\n # The title may have only 140 characters\\n title = self._truncate_string(entry['title'], 140)\\n slug = slugify(title)\\n headline = entry['summary']\\n\\n # Some entries do not have the 'content' field, in this case we\\n # get the 'summary' field instead.\\n if entry.get('content'):\\n content = entry['content'][0]['value']\\n else:\\n content = entry['summary']\\n\\n # When we find a entry that already is registered we don't need\\n # continue because the following registries already be registered.\\n exists = Post.objects.filter(slug=slug).count()\\n if exists:\\n break\\n\\n # Check if has some image in the post content.\\n # NOTE: For the best user experience we use only the posts that\\n # have images.\\n image_url = self._get_image_url_in_content(content)\\n if image_url:\\n main_image = Image.objects.create(\\n title=title,\\n slug=slug,\\n archive_link=image_url,\\n published=True,\\n user=self._user\\n )\\n # Generate the 'short_title' based on 'content'\\n short_title = re.sub('<[^<]+?>', '', content).encode('utf-8')\\n short_title = self._truncate_string(short_title.strip(), 140)\\n\\n post = Post.objects.create(\\n title=title,\\n short_title=short_title,\\n slug=slug,\\n headline=headline,\\n content=content,\\n channel=channel,\\n main_image=main_image,\\n show_on_root_channel=True,\\n published=True,\\n hat='',\\n user=self._user\\n )\",\n \"def insert_new_post(post_arg_set):\\n api, post_data, acct_data, page_id, config = post_arg_set\\n\\n try:\\n post_id = post_data['id'] if post_data.has_key('id') else None\\n\\n except Exception as e:\\n log.error( e )\\n\\n else:\\n\\n # parse date\\n if post_data.has_key('created_time') and post_data['created_time'] is not None: \\n dt = datetime.strptime(post_data['created_time'], FB_DATE_FORMAT)\\n date_time = tz_adj(dt, config)\\n time_bucket = round_datetime(date_time, config)\\n raw_timestamp = int(date_time.strftime(\\\"%s\\\"))\\n \\n else:\\n time_bucket = None\\n raw_timestamp = None\\n \\n # extract message so we can find links within the msg if not in url\\n article_urls = [get_fb_link(post_data, config, unshorten=True)]\\n message = post_data['message'].encode('utf-8') if post_data.has_key('message') else None\\n message_urls = get_message_urls(article_urls, message, config)\\n\\n # detect article links, unshorten and parse\\n article_urls = [\\n parse_url(unshorten_link(url, config)) \\\\\\n for url in article_urls + message_urls\\n if url is not None\\n ]\\n\\n article_urls = [url for url in article_urls if is_article(url, config)]\\n\\n if article_urls:\\n for article_url in set(article_urls):\\n\\n # sluggify url\\n article_slug = sluggify(article_url)\\n\\n # format data\\n post_value = {\\n 'article_slug': article_slug,\\n 'article_url': article_url,\\n 'time_bucket': time_bucket,\\n 'fb_post_created': raw_timestamp,\\n 'raw_timestamp': raw_timestamp,\\n 'fb_raw_link' : get_fb_link(post_data, config=config),\\n 'fb_page_id': page_id,\\n 'fb_post_id': post_id,\\n 'fb_page_likes': acct_data['likes'] if acct_data.has_key('likes') else None,\\n 'fb_page_talking_about': acct_data['talking_about_count'] if acct_data.has_key('talking_about_count') else None,\\n 'fb_type': post_data['type'] if post_data.has_key('type') else None,\\n 'fb_status_type': post_data['status_type'] if post_data.has_key('status_type') else None,\\n 'fb_message': message\\n }\\n \\n # always insert insights data\\n if is_insights(page_id, config):\\n \\n log.info( \\\"INSIGHTS\\\\tAdding data from %s re: %s\\\" % (page_id, article_slug) )\\n\\n # fetch data\\n insights_value = get_insights_data(api, page_id, post_id)\\n\\n # create datasource name\\n data_source = \\\"facebook_insights_%s\\\" % page_id \\n \\n # upsert url\\n upsert_url(article_url, article_slug, data_source, config)\\n\\n # insert id\\n db.sadd('facebook_post_ids', post_id)\\n\\n # format time bucket\\n current_time_bucket = gen_time_bucket(config)\\n insights_value['time_bucket'] = current_time_bucket\\n post_value.pop('time_bucket', None)\\n \\n value = json.dumps({\\n data_source : dict(post_value.items() + insights_value.items())\\n })\\n\\n # upload data to redis\\n db.zadd(article_slug, current_time_bucket, value) \\n \\n # only insert new posts\\n if not db.sismember('facebook_post_ids', post_id):\\n \\n log.info( \\\"FACEBOOK\\\\tNew post %s\\\\t%s\\\" % (post_id, article_url) )\\n \\n # insert id\\n db.sadd('facebook_post_ids', post_id) \\n \\n # upsert url\\n data_source = \\\"facebook_%s\\\" % page_id\\n upsert_url(article_url, article_slug, data_source, config)\\n\\n value = json.dumps( {data_source : post_value} )\\n\\n\\n # upload data to redis\\n db.zadd(article_slug, time_bucket, value)\",\n \"def Translate(self, infile, outfile, mapfile):\\r\\n # Create the top-level feed object\\r\\n feed = []\\r\\n comments = []\\r\\n\\r\\n # Calculate the last updated time by inspecting all of the posts\\r\\n last_updated = 0\\r\\n\\r\\n # These three variables keep the state as we parse the file\\r\\n post_entry = {} # The current post atom.Entry to populate\\r\\n comment_entry = {} # The current comment atom.Entry to populate\\r\\n last_entry = None # The previous post atom.Entry if exists\\r\\n tag_name = None # The current name of multi-line values\\r\\n tag_contents = '' # The contents of multi-line values\\r\\n\\r\\n # Loop through the text lines looking for key/value pairs\\r\\n split_re = re.compile('^[A-Z ]+:')\\r\\n for line in infile:\\r\\n\\r\\n # Remove whitespace\\r\\n line = line.strip()\\r\\n\\r\\n # Check for the post ending token\\r\\n if line == '-' * 8 and tag_name != 'BODY':\\r\\n if post_entry:\\r\\n # Add the post to our feed\\r\\n sys.stderr.write(\\\"Adding post %s\\\\n\\\" % post_entry['title'])\\r\\n self.add_to_user_map(post_entry.get('author'), post_entry.get('authorEmail'))\\r\\n feed.insert(0, post_entry)\\r\\n last_entry = post_entry\\r\\n\\r\\n # Reset the state variables\\r\\n post_entry = {}\\r\\n comment_entry = {}\\r\\n tag_name = None\\r\\n tag_contents = ''\\r\\n continue\\r\\n\\r\\n # Check for the tag ending separator\\r\\n elif line == '-' * 5:\\r\\n # Get the contents of the body and set the entry contents\\r\\n if tag_name == 'BODY':\\r\\n post_entry['description'] = self._Encode(tag_contents)\\r\\n\\r\\n # This is the start of the COMMENT section. Create a new entry for\\r\\n # the comment and add a link to the original post.\\r\\n elif tag_name == 'COMMENT':\\r\\n comment_entry['body'] = self._Encode(tag_contents)\\r\\n post_entry.setdefault('comments', []).append(comment_entry)\\r\\n self.add_to_user_map(comment_entry.get('author'), comment_entry.get('authorEmail'))\\r\\n comment_entry = {}\\r\\n\\r\\n # Get the contents of the extended body\\r\\n elif tag_name == 'EXTENDED BODY':\\r\\n if post_entry:\\r\\n post_entry['mt_text_more'] = self._Encode(tag_contents)\\r\\n elif last_entry:\\r\\n last_entry['mt_text_more'] = self._Encode(tag_contents)\\r\\n\\r\\n # Convert any keywords (comma separated values) into Blogger labels\\r\\n elif tag_name == 'KEYWORDS':\\r\\n post_entry['mt_keywords'] = tag_contents\\r\\n\\r\\n # Reset the current tag and its contents\\r\\n tag_name = None\\r\\n tag_contents = ''\\r\\n continue\\r\\n\\r\\n # Split the line into key/value pairs\\r\\n key = line\\r\\n value = ''\\r\\n if split_re.match(line):\\r\\n elems = line.split(':')\\r\\n key = elems[0]\\r\\n if len(elems) > 1:\\r\\n value = ':'.join(elems[1:]).strip()\\r\\n\\r\\n # The author key indicates the start of a post as well as the author of\\r\\n # the post entry or comment\\r\\n if key == 'AUTHOR':\\r\\n # Create a new entry \\r\\n entry = {}\\r\\n\\r\\n # Add the author's name\\r\\n author_name = self._Encode(value)\\r\\n if not author_name:\\r\\n author_name = 'Anonymous'\\r\\n entry['author'] = author_name\\r\\n\\r\\n # Add the appropriate kind, either a post or a comment\\r\\n if tag_name == 'COMMENT':\\r\\n entry['postid'] = post_entry['postid']\\r\\n comment_entry = entry\\r\\n else:\\r\\n entry['postid'] = 'post-' + self._GetNextId()\\r\\n post_entry = entry\\r\\n\\r\\n # The title only applies to new posts\\r\\n elif key == 'TITLE' and tag_name != 'PING':\\r\\n post_entry['title'] = self._Encode(value)\\r\\n\\r\\n # If the status is a draft, mark it as so in the entry. If the status\\r\\n # is 'Published' there's nothing to do here\\r\\n elif key == 'STATUS':\\r\\n post_entry['status'] = value\\r\\n\\r\\n # Turn categories into labels\\r\\n elif key == 'CATEGORY':\\r\\n post_entry.setdefault('category', []).append(value)\\r\\n\\r\\n # Convert the date and specify it as the published/updated time\\r\\n elif key == 'DATE' and tag_name != 'PING':\\r\\n entry = post_entry\\r\\n if tag_name == 'COMMENT':\\r\\n entry = comment_entry\\r\\n entry['dateCreated'] = value\\r\\n\\r\\n # Check to see if this was the last post published (so far)\\r\\n # seconds = time.mktime(time_val)\\r\\n # last_updated = max(seconds, last_updated)\\r\\n\\r\\n # Convert all tags into Blogger labels\\r\\n elif key == 'TAGS':\\r\\n post_entry.setdefault('tags', []).append(value)\\r\\n\\r\\n # Update the author's email if it is present and not empty\\r\\n elif tag_name == 'COMMENT' and key == 'EMAIL':\\r\\n comment_entry['authorEmail'] = value\\r\\n\\r\\n # Update the author's URI if it is present and not empty\\r\\n elif tag_name == 'COMMENT' and key == 'URL':\\r\\n comment_entry['authorUrl'] = value\\r\\n\\r\\n # If any of these keys are used, they contain information beyond this key\\r\\n # on following lines\\r\\n elif key in ('COMMENT', 'BODY', 'EXTENDED BODY', 'EXCERPT', 'KEYWORDS', 'PING'):\\r\\n tag_name = key\\r\\n\\r\\n # These lines can be safely ignored\\r\\n elif key in ('BASENAME', 'ALLOW COMMENTS', 'CONVERT BREAKS', \\r\\n 'ALLOW PINGS', 'PRIMARY CATEGORY', 'IP', 'URL', 'EMAIL'):\\r\\n continue\\r\\n\\r\\n # If the line is empty and we're processing the body, add a line break\\r\\n elif (tag_name == 'BODY' or tag_name == 'EXTENDED BODY' or tag_name == 'COMMENT') and len(line) == 0:\\r\\n tag_contents += '\\\\n'\\r\\n\\r\\n # This would be a line of content beyond a key/value pair\\r\\n elif len(key) != 0:\\r\\n tag_contents += line + '\\\\n'\\r\\n\\r\\n\\r\\n # Update the feed with the last updated time\\r\\n # feed.updated = atom.Updated(self._ToBlogTime(time.gmtime(last_updated)))\\r\\n\\r\\n # Serialize the feed object\\r\\n yaml.dump(feed, outfile, Dumper=yaml.CDumper)\\r\\n \\r\\n # Write out the user map\\r\\n user_map_dict = {}\\r\\n for name, email in self.user_map:\\r\\n user_map_dict[name] = email\\r\\n yaml.dump(user_map_dict, mapfile, Dumper=yaml.CDumper)\",\n \"def parse_post_content(self, response):\\n post = Post()\\n post['title'] = response.xpath('//h2/a/text()')[0].extract()\\n post['image_url'] = response.xpath(\\\"//div[@class='cont group']//img/@src\\\")[0].extract()\\n yield post\",\n \"def proto_post(self, proto_category):\\n return baker.make(\\n Post,\\n content=(\\n \\\"Aliquip excepteur qui mollit labore nulla et culpa \\\"\\n \\\"minim et commodo reprehenderit consequat sint.\\\"\\n ),\\n categories=proto_category,\\n make_m2m=True,\\n _create_files=True,\\n )\",\n \"def update_post(prev_data, data, db_conn):\\n\\n schema = get_post_schema(data)\\n post_kind = prev_data['kind']\\n if post_kind is 'post' or post_kind is 'proposal':\\n data = pick(data, ('body',))\\n elif post_kind is 'vote':\\n data = pick(data, ('body', 'response',))\\n data, errors = update_document(schema, prev_data, data, db_conn)\\n if not errors:\\n add_post_to_es(data, db_conn)\\n return data, errors\",\n \"def proto_post(self, proto_category):\\n return baker.make(\\n Post,\\n content=\\\"Aute non ex nostrud amet ipsum.\\\",\\n categories=proto_category,\\n make_m2m=True,\\n _create_files=True,\\n )\",\n \"def proto_post(self, proto_category):\\n return baker.make(\\n Post,\\n content=\\\"Aute non ex nostrud amet ipsum.\\\",\\n categories=proto_category,\\n make_m2m=True,\\n _create_files=True,\\n )\",\n \"def parse_description(self, in_str):\\n in_str = self.parse_tags(in_str)\\n self.description = self.parse_metadata(in_str)\",\n \"def process_posts(app, doctree):\\n env = app.builder.env\\n if not hasattr(env, \\\"ablog_posts\\\"):\\n env.ablog_posts = {}\\n post_nodes = list(doctree.findall(PostNode))\\n if not post_nodes:\\n return\\n post_date_format = app.config[\\\"post_date_format\\\"]\\n should_auto_orphan = app.config[\\\"post_auto_orphan\\\"]\\n docname = env.docname\\n if should_auto_orphan:\\n # mark the post as 'orphan' so that\\n # \\\"document isn't included in any toctree\\\" warning is not issued\\n # We do not simply assign to should_auto_orphan because if auto-orphan\\n # is false, we still want to respect the per-post :rst:dir`orphan` setting\\n app.env.metadata[docname][\\\"orphan\\\"] = True\\n blog = Blog(app)\\n auto_excerpt = blog.post_auto_excerpt\\n multi_post = len(post_nodes) > 1 or blog.post_always_section\\n for order, node in enumerate(post_nodes, start=1):\\n if node[\\\"excerpt\\\"] is None:\\n node[\\\"excerpt\\\"] = auto_excerpt\\n if multi_post:\\n # section title, and first few paragraphs of the section of post\\n # are used when there are more than 1 posts\\n section = node\\n while True:\\n if isinstance(section, nodes.section):\\n break\\n section = node.parent\\n else:\\n section = doctree\\n # get updates here, in the section that post belongs to\\n # Might there be orphan updates?\\n update_dates = _get_update_dates(section, docname, post_date_format)\\n # Making sure that post has a title because all post titles\\n # are needed when resolving post lists in documents\\n title = node[\\\"title\\\"] or _get_section_title(section)\\n # creating a summary here, before references are resolved\\n excerpt = []\\n if node.children:\\n if node[\\\"exclude\\\"]:\\n node.replace_self([])\\n else:\\n node.replace_self(node.children)\\n for child in node.children:\\n excerpt.append(child.deepcopy())\\n elif node[\\\"excerpt\\\"]:\\n count = 0\\n for nod in section.findall(nodes.paragraph):\\n excerpt.append(nod.deepcopy())\\n count += 1\\n if count >= (node[\\\"excerpt\\\"] or 0):\\n break\\n node.replace_self([])\\n else:\\n node.replace_self([])\\n nimg = node[\\\"image\\\"] or blog.post_auto_image\\n if nimg:\\n for img, nod in enumerate(section.findall(nodes.image), start=1):\\n if img == nimg:\\n excerpt.append(nod.deepcopy())\\n break\\n date = node[\\\"date\\\"]\\n if date:\\n try:\\n date = datetime.strptime(date, post_date_format)\\n except ValueError:\\n if date_parser:\\n try:\\n date = date_parser(date)\\n except ValueError:\\n raise ValueError(\\\"invalid post date in: \\\" + docname)\\n else:\\n raise ValueError(\\n f\\\"invalid post date ({date}) in \\\" + docname + f\\\". Expected format: {post_date_format}\\\"\\n )\\n else:\\n date = None\\n # if docname ends with `index` use folder name to reference the document\\n # a potential problem here is that there may be files/folders with the\\n # same name, so issuing a warning when that's the case may be a good idea\\n folder, label = os.path.split(docname)\\n if label == \\\"index\\\":\\n folder, label = os.path.split(folder)\\n if not label:\\n label = slugify(title)\\n section_name = \\\"\\\"\\n if multi_post and section.parent is not doctree:\\n section_name = section.attributes[\\\"ids\\\"][0]\\n label += \\\"-\\\" + section_name\\n else:\\n # create a reference for the post\\n # if it is posting the document\\n # ! this does not work for sections\\n app.env.domains[\\\"std\\\"].data[\\\"labels\\\"][label] = (docname, label, title)\\n app.env.domains[\\\"std\\\"].data[\\\"anonlabels\\\"][label] = (docname, label)\\n if section.parent is doctree:\\n section_copy = section[0].deepcopy()\\n else:\\n section_copy = section.deepcopy()\\n # multiple posting may result having post nodes\\n for nn in section_copy.findall(PostNode):\\n if nn[\\\"exclude\\\"]:\\n nn.replace_self([])\\n else:\\n nn.replace_self(node.children)\\n postinfo = {\\n \\\"docname\\\": docname,\\n \\\"section\\\": section_name,\\n \\\"order\\\": order,\\n \\\"date\\\": date,\\n \\\"update\\\": max(update_dates + [date]),\\n \\\"title\\\": title,\\n \\\"excerpt\\\": excerpt,\\n \\\"tags\\\": node[\\\"tags\\\"],\\n \\\"author\\\": node[\\\"author\\\"],\\n \\\"category\\\": node[\\\"category\\\"],\\n \\\"location\\\": node[\\\"location\\\"],\\n \\\"language\\\": node[\\\"language\\\"],\\n \\\"redirect\\\": node[\\\"redirect\\\"],\\n \\\"nocomments\\\": node[\\\"nocomments\\\"],\\n \\\"image\\\": node[\\\"image\\\"],\\n \\\"exclude\\\": node[\\\"exclude\\\"],\\n \\\"external_link\\\": node[\\\"external_link\\\"],\\n \\\"doctree\\\": section_copy,\\n }\\n if docname not in env.ablog_posts:\\n env.ablog_posts[docname] = []\\n env.ablog_posts[docname].append(postinfo)\\n # instantiate catalogs and collections here\\n # so that references are created and no warnings are issued\\n if app.builder.format == \\\"html\\\":\\n stdlabel = env.domains[\\\"std\\\"].data[\\\"labels\\\"] # NOQA\\n else:\\n if hasattr(env, \\\"intersphinx_inventory\\\"):\\n stdlabel = env.intersphinx_inventory.setdefault(\\\"std:label\\\", {}) # NOQA\\n baseurl = getattr(env.config, \\\"blog_baseurl\\\").rstrip(\\\"/\\\") + \\\"/\\\" # NOQA\\n project, version = env.config.project, str(env.config.version) # NOQA\\n for key in [\\\"tags\\\", \\\"author\\\", \\\"category\\\", \\\"location\\\", \\\"language\\\"]:\\n catalog = blog.catalogs[key]\\n for label in postinfo[key]:\\n coll = catalog[label] # NOQA\\n if postinfo[\\\"date\\\"]:\\n coll = blog.archive[postinfo[\\\"date\\\"].year] # NOQA\",\n \"def run_get_post(m):\\n\\n doc = get_doc(m)\\n assert doc is not None\\n\\n wp = get_wp(m)\\n\\n post = find_post(wp, doc.identifier)\\n\\n if post:\\n post.content = \\\"…content elided…\\\"\\n from pprint import pprint\\n pprint(post.struct)\\n return\\n else:\\n warn(f\\\"Didn't find post for identifier {doc.identifier}\\\")\\n return\",\n \"def parse_post_metadata(post_text):\\n result = {}\\n \\n header_end = 0\\n \\n promed_date_match = re.search(\\n r\\\"Published Date:\\\\s(?P<date>.*)\\\", post_text)\\n result[\\\"promedDate\\\"] = parse_promed_pub_datetime(\\n promed_date_match.group(\\\"date\\\"))\\n \\n archive_match = re.search(r\\\"Archive Number: (?P<num>.*)\\\", post_text)\\n result[\\\"archiveNumber\\\"] = archive_match.group(\\\"num\\\")\\n header_end = archive_match.end()\\n \\n subject = re.search(r\\\"Subject:\\\\s(?P<subject>.*)\\\", post_text).group(\\\"subject\\\")\\n result[\\\"subject\\\"] = parse_subject_line(subject)\\n result[\\\"subject\\\"][\\\"raw\\\"] = subject\\n \\n # This will not find all linked reports.\\n # Some older posts refrence posts using different indexes I do not know\\n # how to interpret.\\n # Example: http://promedmail.org/direct.php?id=2194235\\n result[\\\"linkedReports\\\"] = [\\n report_id for report_id in re.findall(r\\\"\\\\d{8}\\\\.\\\\d+\\\", post_text)]\\n \\n # Most links will be article source urls or links to promed.\\n result[\\\"links\\\"] = list(set(\\n re.findall(r\\\"http\\\\S+[^(\\\\.\\\\])(\\\\.\\\\)>\\\\s]\\\", post_text)))\\n result[\\\"links\\\"].sort()\\n \\n communicated_match = re.search(communicated_by_regex, post_text, re.M)\\n if communicated_match:\\n result[\\\"communicatedBy\\\"] = communicated_match.group(\\\"communicated_by\\\")\\n return result, header_end\",\n \"def register_post_parser(self, fct, cfg, ctx):\\n self.post_parsers.append((fct, cfg, ctx))\",\n \"def __init__(self, post):\\n self.id = int(post[0])\\n self.type_id = int(post[1])\\n try:\\n self.parent_id = int(post[2])\\n except:\\n self.parent_id = None\\n try:\\n self.accepted_answer_id = int(post[3])\\n except:\\n self.accepted_answer_id = None\\n self.creation_date = post[4]\\n try:\\n self.score = post[5]\\n except:\\n self.score = None\\n try:\\n self.view_count = post[6]\\n except:\\n self.view_count = 0\\n self.body = post[7]\\n try:\\n self.owner_user_id = int(post[8])\\n except:\\n self.owner_user_id = None \\n try:\\n self.last_editor_user_id = int(post[9])\\n except:\\n self.last_editor_user_id = None\\n try:\\n self.last_editor_display_name = post[10]\\n except:\\n self.last_editor_display_name = None\\n self.last_activity_date = post[11]\\n self.last_edit_date = post[12]\\n self.community_owned_date = post[13]\\n self.closed_date = post[14]\\n try:\\n self.title = post[15]\\n except:\\n self.title = None\\n try:\\n self.tags = post[16]\\n except:\\n self.tags = None\\n try:\\n self.answer_count = int(post[17])\\n except:\\n self.answer_count = 0\\n try:\\n self.comment_count = int(post[18])\\n except:\\n self.comment_count = 0\\n try:\\n self.favorite_count = int(post[19])\\n except:\\n self.favorite_count = 0\",\n \"def save(self, *args, **kwargs):\\n\\n if len(self.text) > 97:\\n self.short_description = self.text[:97] + '...'\\n else:\\n self.short_description = self.text\\n\\n super(Post, self).save(*args, **kwargs)\",\n \"def parse_posts(self):\\n logger.info(\\\"Parsing posts\\\")\\n\\n self.df.title = self.df.title.str.strip()\\n\\n spam_companies = [\\\"Indeed Prime\\\"]\\n self.df = self.df[~self.df[\\\"company\\\"].isin(spam_companies)]\\n self.df = self.df.dropna(subset=[\\\"company\\\"])\\n self.df = self.df.drop_duplicates(subset=[\\\"company\\\", \\\"date_posted\\\", \\\"title\\\"])\",\n \"def parse_article(self, response):\\n\\n raw_post = response.css(\\\"div.blog.post > div.inner > div.row > article\\\")\\n\\n post_loader = ItemLoader(item=BlogPostItem(), selector=raw_post)\\n post_loader.default_output_processor = TakeFirst()\\n\\n post_title = raw_post.css(\\\"div#postcontent > h1::text\\\").extract_first()\\n post_loader.add_value(\\\"title\\\", post_title)\\n post_loader.add_value(\\\"url\\\", response.request.url)\\n\\n post_text_selector = raw_post.css(\\\"div#postcontent > div#mypost\\\")\\n post_text = post_text_selector.xpath('string(.)').extract_first()\\n post_loader.add_value(\\\"content\\\", post_text[:160])\\n\\n pub_date_text = raw_post.css(\\\"div#postcontent > div.no-mobile > div.posttag.right.nomobile > span::text\\\").extract_first()\\n pub_date = parse_date(pub_date_text)\\n post_loader.add_value(\\\"publication_date\\\", pub_date)\\n\\n initial_author_list = raw_post.css(\\n \\\"div#postcontent > div.no-mobile > div.postauthor > span > a.goauthor > span::text\\\").extract()\\n author_list = [name.strip() for name in initial_author_list]\\n post_authors = \\\"::\\\".join(author_list)\\n post_loader.add_value(\\\"author\\\", post_authors)\\n\\n post_tags = raw_post.css(\\\"div#postcontent > a.tag.secondary::attr(title)\\\").extract()\\n post_tags_str = \\\"::\\\".join(post_tags)\\n post_loader.add_value(\\\"tags\\\", post_tags_str)\\n\\n return post_loader.load_item()\",\n \"def parse_post(post):\\n # Ignore posts less than 24 hours old\\n if time.time() - post.created < 60 * 60 * 24:\\n logger.debug('Ignoring post (too new)')\\n return False\\n\\n # Add post to database\\n postid_db = db.insert('Posts',\\n (None,\\n post.id,\\n post.title,\\n post.url,\\n post.selftext,\\n post.author,\\n post.permalink,\\n post.subreddit,\\n post.num_comments,\\n post.upvotes,\\n post.downvotes,\\n post.score,\\n post.created_utc,\\n int(post.is_self),\\n int(post.over_18)))\\n # If post already exists, we've already indexed it; skip!\\n if postid_db == -1:\\n logger.debug('Ignoring post (already indexed)')\\n return False\\n # Write post to DB so we don't hit it again\\n\\n # NOTE: postid_db is the ID of the post in the database; NOT on reddit\\n\\n # Check for self-post\\n if post.selftext != '':\\n urls = get_links_from_body(post.selftext)\\n for url in urls:\\n parse_url(url, postid=postid_db)\\n else:\\n # Attempt to retrieve hash(es) from link\\n parse_url(post.url, postid=postid_db)\\n\\n # Iterate over top-level comments\\n if post.num_comments > 0:\\n reddit.fetch_comments(post)\\n for comment in post.comments:\\n parse_comment(comment, postid_db)\",\n \"def newPost(self, useRawHTML):\\n print\\n content, publish = self._fillPost(useRawHTML)\\n\\n # Upload to server\\n try :\\n postid = self.server.metaWeblog.newPost(\\n self.blogid, self.username, self.password,\\n content, publish\\n )\\n except xmlrpclib.Fault as fault:\\n display_XMLRPC_errors(\\\"post the new entry\\\", fault)\\n import pdb\\n pdb.set_trace()\\n else :\\n self._setCategorie(postid)\\n print \\\"New post created with ID =\\\", postid\",\n \"def inputPostFormatter(post):\\n\\n post[\\\"ttl\\\"] = from_decimal(post[\\\"ttl\\\"])\\n post[\\\"workToProve\\\"] = from_decimal(post.get(\\\"workToProve\\\", 0))\\n post[\\\"priority\\\"] = from_decimal(post[\\\"priority\\\"])\\n\\n if not is_array(post.get(\\\"topics\\\")):\\n post[\\\"topics\\\"] = [post[\\\"topics\\\"]] if post.get(\\\"topics\\\") else []\\n\\n post[\\\"topics\\\"] = [topic if is_0x_prefixed(topic) else encode_hex(topic)\\n for topic in post[\\\"topics\\\"]]\\n\\n return post\",\n \"def post_process_post(self, post):\\r\\n post.article = self.rewrite_ob_urls(post.article)\\r\\n post._commit()\\r\\n \\r\\n comments = Comment._query(Comment.c.link_id == post._id, data = True)\\r\\n for comment in comments:\\r\\n comment.body = self.rewrite_ob_urls(comment.body)\\r\\n comment._commit()\",\n \"def _parse_yaml_description(self, meta: Mapping):\\n with contextlib.suppress(Exception):\\n yaml_md = yaml.safe_load(meta[\\\"description\\\"])\\n # YAML could be anything: plain string, list, …\\n if isinstance(yaml_md, dict):\\n meta.pop(\\\"description\\\")\\n meta.update(yaml_md)\",\n \"def change_post(mfp_form, request):\\n article_id = int(mfp_form.cleaned_data['id'])\\n newpost = APost(article_id)\\n if newpost.exist:\\n newpost.article.title = mfp_form.cleaned_data['title']\\n # 缩略名\\n newpost.article.short_title = mfp_form.cleaned_data['short_title']\\n newpost.article.cover = mfp_form.cleaned_data['cover_url']\\n newpost.article.introduction = mfp_form.cleaned_data['introduction']\\n newpost.article.content = mfp_form.cleaned_data['content']\\n # id为2是已发布的文章，默认为已发布，后面再改\\n newpost.article.status = Status.objects.get(id=2)\\n tagids = mfp_form.cleaned_data['tags']\\n if len(tagids) != 0:\\n for tagid in tagids:\\n tagid = int(tagid)\\n tag = Tags.objects.get(id=tagid)\\n newpost.article.tags.add(tag)\\n threadtypeid = int(mfp_form.cleaned_data['threadtypeid'])\\n newpost.article.threadtypeid = ThreadTypes.objects.get(id=threadtypeid)\\n if mfp_form.cleaned_data['commentnotshow'] != '':\\n newpost.article.comment_status = False\\n else:\\n newpost.article.comment_status = True\\n return newpost.article\\n else:\\n return False\",\n \"def get_post(self):\\n\\t\\tself.post = graph.get_object(POST_ID)\",\n \"def tumblrPost(body,tags=None,title=None):\\t\\n\\tglobal settings\\n\\tpost_address = \\\"http://api.tumblr.com/v2/blog/\\\"+settings['blog']+\\\"/post\\\"\\n\\tdata = {\\\"type\\\":\\\"text\\\",\\\"state\\\":settings['state'],\\\"body\\\" : body, \\\"oauth_token\\\":settings['oauth_token']}\\n\\tif tags != None:\\n\\t\\tdata['tags'] = str(tags)\\n\\tif title != None:\\n\\t\\tdata['title'] = str(title)\\n\\tconsumer = oauth.Consumer(settings['consumer_key'], settings['consumer_secret'])\\n\\ttoken = oauth.Token(settings['oauth_token'], settings['oauth_token_secret'])\\n\\tclient = oauth.Client(consumer,token)\\n\\tresp, content = client.request(post_address, \\\"POST\\\", urllib.urlencode(data))\\n\\tjs = json.loads(content)\\n \\tif js['meta']['status'] != 201: #something wrong\\n\\t\\tprint js['meta']['msg']\\n\\t\\tprint js\",\n \"def _update_post_node(node, options, arguments):\\n node[\\\"date\\\"] = arguments[0] if arguments else None\\n node[\\\"tags\\\"] = options.get(\\\"tags\\\", [])\\n node[\\\"author\\\"] = options.get(\\\"author\\\", [])\\n node[\\\"category\\\"] = options.get(\\\"category\\\", [])\\n node[\\\"location\\\"] = options.get(\\\"location\\\", [])\\n node[\\\"language\\\"] = options.get(\\\"language\\\", [])\\n node[\\\"redirect\\\"] = options.get(\\\"redirect\\\", [])\\n node[\\\"title\\\"] = options.get(\\\"title\\\", None)\\n node[\\\"image\\\"] = options.get(\\\"image\\\", None)\\n node[\\\"excerpt\\\"] = options.get(\\\"excerpt\\\", None)\\n node[\\\"exclude\\\"] = \\\"exclude\\\" in options\\n node[\\\"nocomments\\\"] = \\\"nocomments\\\" in options\\n node[\\\"external_link\\\"] = options.get(\\\"external_link\\\", [])\\n return node\",\n \"def parse_post(post: Dict[str, Any],\\n image_retriever: str = \\\"pretrained\\\",\\n image_basedir: Optional[str] = \\\"documentIntent_emnlp19/resnet18_feat\\\") -> Dict[str, Any]:\\n id = post['id']\\n label_intent = post['intent']\\n label_semiotic = post['semiotic']\\n label_contextual = post['contextual']\\n caption = post['caption']\\n\\n if image_retriever == \\\"url\\\":\\n image = post['url']\\n raise NotImplementedError(\\\"Currently cannot download an image from {}\\\".format(image))\\n elif image_retriever == \\\"pretrained\\\":\\n image_path = Path(image_basedir) / \\\"{}.npy\\\".format(id)\\n image = np.load(image_path)\\n elif image_retriever == \\\"ignored\\\":\\n image = None\\n else:\\n raise NotImplementedError(\\\"image_retriever method doesn't exist\\\")\\n\\n output_dict = {\\n 'id': id,\\n 'label': {\\n 'intent': label_intent,\\n 'semiotic': label_semiotic,\\n 'contextual': label_contextual,\\n },\\n 'caption': caption,\\n 'image': image,\\n }\\n\\n return output_dict\",\n \"def add_post_to_es(post, db_conn):\\n\\n from database.topic import get_topic, deliver_topic\\n from database.user import get_user, deliver_user\\n\\n data = json_prep(deliver_post(post))\\n topic = get_topic({'id': post['topic_id']}, db_conn)\\n if topic:\\n data['topic'] = json_prep(deliver_topic(topic))\\n user = get_user({'id': post['user_id']}, db_conn)\\n if user:\\n data['user'] = json_prep(deliver_user(user))\\n\\n return es.index(\\n index='entity',\\n doc_type='post',\\n body=data,\\n id=post['id'],\\n )\",\n \"def add_post_into_ds(post, ds_group, ds_item_role):\\n post.ds_group = ds_group\\n post.ds_item_role = ds_item_role\\n post.is_in_ds = True\\n post.save()\",\n \"def handle_meta(self, tag, attrs):\\n ad = {}\\n for tup in attrs:\\n ad[tup[0]] = tup[1]\\n if 'name' in ad.keys() \\\\\\n and 'keywords' == ad['name'] \\\\\\n and 'content' in ad.keys():\\n self.filetype = ad['content']\\n if 'name' in ad.keys() \\\\\\n and 'description' == ad['name']:\\n self.description = 'present'\\n if 'charset' in ad.keys():\\n self.charset = 'present'\",\n \"def remotePostCreate(host, post):\\n post = post.get('posts')[0]\\n author = remoteAddAuthor(post.get('author'))\\n title = post.get('title')\\n description = post.get('description')\\n contentType = post.get('contentType')\\n content = post.get('content')\\n published = utc.localize(datetime.strptime(post.get('published'), '%Y-%m-%dT%H:%M:%S.%fZ'))\\n visibility = post.get('visibility')\\n unlisted = post.get('unlisted')\\n id = post.get('id')\\n origin = post.get('origin')\\n count = post.get('count')\\n comments = remoteCommentList(post)\\n source = \\\"{}/api/posts/{}\\\".format(DOMAIN, post.get('id'))\\n\\n post_dict = {'author': author, 'title': title, 'description': description,\\n 'contentType': contentType, 'content': content, 'published': published,\\n 'visibility': visibility, 'unlisted': unlisted, 'id': id,\\n 'comments': comments, 'origin': origin, 'count': count,\\n 'source': source}\\n return post_dict\",\n \"def new_post(mkp_form, request):\\n newpost = Posts()\\n newpost.init()\\n newpost.authorid = int(request.user.id)\\n newpost.title = mkp_form.cleaned_data['title']\\n newpost.name = mkp_form.cleaned_data['short_title'] # 缩略名\\n newpost.cover = mkp_form.cleaned_data['cover_url']\\n newpost.introduction = mkp_form.cleaned_data['introduction']\\n newpost.content = js_resize_img(mkp_form.cleaned_data['content'])\\n newpost.status = Status.objects.get(id=2) # id为2是已发布的文章，默认为已发布，后面再改\\n tagids = mkp_form.cleaned_data['tags']\\n if len(tagids) != 0:\\n for tagid in tagids:\\n tagid = int(tagid)\\n tag = Tags.objects.get(id=tagid)\\n newpost.tags.add(tag)\\n threadtypeid = mkp_form.cleaned_data['threadtypeid']\\n newpost.threadtypeid = ThreadTypes.objects.get(id=threadtypeid)\\n if mkp_form.cleaned_data['commentnotshow'] != '':\\n newpost.comment_status = False\\n else:\\n newpost.comment_status = True\\n return newpost\",\n \"def add(self, posts):\\n for post in posts:\\n self._feed.add(FeedEntry(\\n summary=post.summary,\\n title=post.title,\\n title_type='html',\\n url=post.url,\\n updated=post.date,\\n ))\",\n \"def update_post(request, pk):\\n data = JSONParser().parse(request)\\n\\n if request.method == 'PATCH':\\n if data['user_id']:\\n user = User.objects.get(id = data['user_id'])\\n else:\\n user = request.user\\n try:\\n postObject = Post.objects.filter(id=pk, user = user).first()\\n except Post.DoesNotExist:\\n content = {'user forbidden': 'you should be user of the requested post.'}\\n return Response(content, status=status.HTTP_403_FORBIDDEN)\\n\\n\\n postObject.content = data['content']\\n postObject.tags.clear()\\n\\n for tag in data[\\\"tags\\\"]:\\n if len(tag)>0:\\n if tag['name'] == '':\\n continue\\n try:\\n tagObject = Tag.objects.get(wikidataID=tag['wikidataID'])\\n except ObjectDoesNotExist:\\n tagObject = Tag.objects.create(wikidataID=tag['wikidataID'], name=tag['name'])\\n except MultipleObjectsReturned:\\n return HttpResponse(\\\"Multiple tags exist for.\\\" + tag + \\\" Invalid State.\\\")\\n\\n unique_hidden_tags = list(set(tag['hidden_tags']))\\n if unique_hidden_tags:\\n tagObject.hidden_tags = unique_hidden_tags\\n\\n tagObject.save()\\n postObject.tags.add(tagObject)\\n postObject.save()\\n return Response(status=status.HTTP_200_OK)\",\n \"def on_frontmatter_loaded(self, source_file, frontmatter):\\n if not self._is_post(frontmatter):\\n return\\n self._validate_post(source_file, frontmatter)\\n post = BlogPost(\\n date=frontmatter['date'],\\n source_file=source_file,\\n summary=frontmatter.get('summary', ''),\\n title=frontmatter['title'],\\n route=self._resolver.as_route(source_file),\\n url=self._resolver.as_url(source_file),\\n posts=self.posts,\\n )\\n frontmatter['post'] = post\\n if post != self.posts.get(source_file):\\n self.posts[source_file] = post\\n self._should_generate = True\",\n \"def markdown_post(post):\\n post['entry'] = markdown(post['entry'].replace(\\\"\\\\n\\\",\\\" \\\\n\\\"), output=\\\"html5\\\")\\n return post\",\n \"async def upload_new_post(self, ctx: commands.Context, safety: str, *tags):\\n safety = safety.lower()\\n if safety not in ['safe', 'sketchy', 'unsafe']:\\n raise ValueError(f\\\"Must specify safety: safe, sketchy, or unsafe\\\")\\n anon = False\\n urls = []\\n sources = []\\n tags = list(tags)\\n if 'anonymous' in tags:\\n anon = True\\n tags.remove('anonymous')\\n if 'anon' in tags:\\n anon = True\\n tags.remove('anon')\\n for tag in tags:\\n if '://' in tag:\\n url = tag.strip('<>')\\n urls.append(url)\\n tags.remove(tag)\\n # Checks to see if any of the tags being uploaded has implied tags that should also be applied to the post.\\n for tag in tags:\\n tagDL = await self.api_get(ctx, f\\\"/tag/{tag}\\\"}\\n if 'name' in tagDL and tagDL['name'] == \\\"TagNotFoundError\\\":\\n print(\\\"New Tag\\\")\\n else:\\n for implied_tag in tagDL['implications']:\\n tags.append(implied_tag['names'][0])\",\n \"def post_to_object(self, post, remove_id_prefix=False):\\n id = post.get('id')\\n if not id:\\n return {}\\n\\n post_type = post.get('type')\\n status_type = post.get('status_type')\\n url = self.post_url(post)\\n picture = post.get('picture')\\n display_name = None\\n message = (post.get('message') or post.get('story') or\\n post.get('description') or post.get('name'))\\n\\n data = post.get('data', {})\\n for field in ('object', 'song'):\\n obj = data.get(field)\\n if obj:\\n id = obj.get('id')\\n post_type = obj.get('type')\\n url = obj.get('url')\\n display_name = obj.get('title')\\n\\n object_type = OBJECT_TYPES.get(post_type)\\n author = self.user_to_actor(post.get('from'))\\n link = post.get('link', '')\\n\\n if link.startswith('/gifts/'):\\n object_type = 'product'\\n if not object_type:\\n if picture and not message:\\n object_type = 'image'\\n else:\\n object_type = 'note'\\n\\n obj = {\\n 'id': self.tag_uri(str(id)),\\n 'objectType': object_type,\\n 'published': util.maybe_iso8601_to_rfc3339(post.get('created_time')),\\n 'updated': util.maybe_iso8601_to_rfc3339(post.get('updated_time')),\\n 'author': author,\\n 'content': message,\\n # FB post ids are of the form USERID_POSTID\\n 'url': url,\\n 'image': {'url': picture},\\n 'displayName': display_name,\\n 'fb_object_id': post.get('object_id'),\\n }\\n\\n privacy = post.get('privacy', {})\\n if isinstance(privacy, dict):\\n privacy = privacy.get('value')\\n if privacy is not None:\\n # privacy value '' means it doesn't have an explicit audience set, so i\\n # *think* it inherits from its parent. TODO: use that value as opposed to\\n # defaulting to public.\\n public = privacy.lower() in ('', 'everyone', 'open')\\n obj['to'] = [{'objectType': 'group',\\n 'alias': '@public' if public else '@private'}]\\n\\n # tags and likes\\n tags = itertools.chain(post.get('to', {}).get('data', []),\\n post.get('with_tags', {}).get('data', []),\\n *post.get('message_tags', {}).values())\\n obj['tags'] = [self.postprocess_object({\\n 'objectType': OBJECT_TYPES.get(t.get('type'), 'person'),\\n 'id': self.tag_uri(t.get('id')),\\n 'url': self.object_url(t.get('id')),\\n 'displayName': t.get('name'),\\n 'startIndex': t.get('offset'),\\n 'length': t.get('length'),\\n }) for t in tags]\\n\\n obj['tags'] += [self.postprocess_object({\\n 'id': self.tag_uri('%s_liked_by_%s' % (id, like.get('id'))),\\n 'url': url,\\n 'objectType': 'activity',\\n 'verb': 'like',\\n 'object': {'url': url},\\n 'author': self.user_to_actor(like),\\n 'content': 'likes this.',\\n }) for like in post.get('likes', {}).get('data', [])]\\n\\n # \\\"See Original\\\" links\\n post_actions = post.get('actions',[])\\n see_orig_actions = (act for act in post_actions\\n if act.get('name', '').lower() in SEE_ORIGINAL_ACTIONS)\\n obj['tags'] += [self.postprocess_object({\\n 'objectType': 'article',\\n 'url': act.get('link'),\\n 'displayName': act.get('name')\\n }) for act in see_orig_actions]\\n\\n # is there an attachment? prefer to represent it as a picture (ie image\\n # object), but if not, fall back to a link.\\n att = {\\n 'url': link if link else url,\\n 'image': {'url': picture},\\n 'displayName': post.get('name'),\\n 'summary': post.get('caption'),\\n 'content': post.get('description'),\\n }\\n\\n if (picture and picture.endswith('_s.jpg') and\\n (post_type == 'photo' or status_type == 'added_photos')):\\n # a picture the user posted. get a larger size.\\n att.update({\\n 'objectType': 'image',\\n 'image': {'url': picture[:-6] + '_o.jpg'},\\n })\\n obj['attachments'] = [att]\\n elif link and not link.startswith('/gifts/'):\\n att['objectType'] = 'article'\\n obj['attachments'] = [att]\\n\\n # location\\n place = post.get('place')\\n if place:\\n id = place.get('id')\\n obj['location'] = {\\n 'displayName': place.get('name'),\\n 'id': id,\\n 'url': self.object_url(id),\\n }\\n location = place.get('location', None)\\n if isinstance(location, dict):\\n lat = location.get('latitude')\\n lon = location.get('longitude')\\n if lat and lon:\\n obj['location'].update({\\n 'latitude': lat,\\n 'longitude': lon,\\n # ISO 6709 location string. details: http://en.wikipedia.org/wiki/ISO_6709\\n 'position': '%+f%+f/' % (lat, lon),\\n })\\n elif 'location' in post:\\n obj['location'] = {'displayName': post['location']}\\n\\n # comments go in the replies field, according to the \\\"Responses for\\n # Activity Streams\\\" extension spec:\\n # http://activitystrea.ms/specs/json/replies/1.0/\\n comments = post.get('comments', {}).get('data')\\n if comments:\\n items = [self.comment_to_object(c) for c in comments]\\n obj['replies'] = {\\n 'items': items,\\n 'totalItems': len(items),\\n }\\n\\n return self.postprocess_object(obj)\",\n \"def save_post(self, post):\\n return self.collection.insert_one(post.serialize())\",\n \"def mutate(self, info, **arguments):\\n post_id = from_global_id(arguments.pop('id'))[1]\\n post = PostModel.objects.get(pk=post_id)\\n\\n if not post:\\n print(\\\"=\\\" * 100)\\n print(\\\"schema.py@41\\\")\\n print(\\\"Erro while updating post\\\")\\n print(\\\"=\\\" * 100)\\n return UpdatePostMutation(success=False)\\n\\n for k, v in arguments.items():\\n setattr(post, k, v)\\n post.save()\\n return UpdatePostMutation(success=True, post=post)\",\n \"def json_to_post(self, post, tag):\\r\\n # Get media type (image/video)\\r\\n mediaType = self.get_value(post, (\\\"images\\\", 0, \\\"type\\\"))\\r\\n if mediaType is None:\\r\\n mediaType = self.get_value(post, (\\\"type\\\", ))\\r\\n mediaType = self.get_media_type(mediaType)\\r\\n\\r\\n imageCount = self.get_value(post, ('images_count',))\\r\\n if imageCount is None:\\r\\n imageCount = 1\\r\\n \\r\\n # Only want 1 image/video\\r\\n if imageCount > 1:\\r\\n return None\\r\\n \\r\\n # Get media url and size\\r\\n if mediaType == MediaType.IMAGE:\\r\\n media = self.get_value(post, ('images', 0, 'id', ))\\r\\n size = self.get_value(post, ('size', ))\\r\\n if size is None:\\r\\n size = self.get_value(post, ('images', 0, 'size', ))\\r\\n if media is None:\\r\\n media = self.get_value(post, ('link', ))\\r\\n elif mediaType == MediaType.VIDEO:\\r\\n media = self.get_value(post, ('images', 0, 'id', ))\\r\\n size = self.get_value(post, ('mp4_size', ))\\r\\n if size is None:\\r\\n size = self.get_value(post, ('images', 0, 'mp4_size', ))\\r\\n if media is None:\\r\\n media = self.get_value(post, ('mp4', ))\\r\\n else:\\r\\n self.logger.log(logger.LogLevel.WARNING, 'mediaType is not Image or Video')\\r\\n return None\\r\\n\\r\\n #check if image/video is over max size\\r\\n if mediaType == MediaType.IMAGE:\\r\\n if size >= self.setting.maxImageSize:\\r\\n return None\\r\\n elif mediaType == MediaType.VIDEO:\\r\\n if size >= self.setting.maxVideoSize:\\r\\n return None\\r\\n\\r\\n postId = self.get_value(post, (\\\"id\\\", ))\\r\\n title = self.get_value(post, (\\\"title\\\", ))\\r\\n views = self.get_value(post, (\\\"views\\\", ))\\r\\n ups = self.get_value(post, (\\\"ups\\\", ))\\r\\n downs = self.get_value(post, (\\\"downs\\\", ))\\r\\n return Post(postId, title, mediaType.value, media, size, tag, views, ups, downs)\",\n \"def setPostDown(self, post):\\n # type: (tp.Union[str, tp.List[str]])->None\\n if isinstance(post, list):\\n self._ifAttributes['post-down'] = post\\n elif isinstance(post, str):\\n self._ifAttributes['post-down'] = [post]\\n else:\\n raise ValueError(\\\"Invalid value type {0}, expected str or List[str]\\\".format(type(post)))\",\n \"def handle_new_post(post_data, user_agent, remote_addr):\\n \\n for required in POST_REQUIRED_PARAMS:\\n if required not in post_data:\\n return None, None\\n\\n try:\\n value = int(string_from_interwebs(post_data.getfirst(\\\"code\\\", \\\"\\\")))\\n except ValueError:\\n return None, None\\n \\n if value != 98098098098:\\n return None, None\\n\\n # not yet safe to use.\\n location = post_data.getfirst(\\\"location\\\", \\\"\\\")\\n tags = string_from_interwebs(post_data.getfirst(\\\"tags\\\")) \\n author = post_data.getfirst(\\\"author\\\")\\n \\n split_tags = [string_from_interwebs(tag).strip().lower() for tag in tags.split(\\\",\\\")] # temporary\\n \\n if len(split_tags) > 3:\\n return None, None\\n \\n author_id = string_from_interwebs(author).strip()\\n \\n with Connection('localhost', 27017) as connection:\\n reply_to = string_from_interwebs(post_data.getfirst(\\\"reply_to\\\"))\\n \\n if not verify_author(author_id, connection):\\n return None, None\\n\\n if not verify_post(reply_to, connection):\\n return None, None\\n\\n # if reply then it's verified.\\n # XXX: I need to make a standard object structure for this, so that I don't \\n # have to update separate things.\\n\\n post = {\\\"viewed\\\" : 0,\\n \\\"comments\\\" : 0,\\n \\\"flagged\\\" : 0,\\n \\\"disliked\\\" : 0,\\n \\\"enjoyed\\\" : 0,\\n \\\"num_replies\\\" : 0,\\n \\\"num_reposts\\\" : 0,\\n \\\"content-type\\\" : \\\"image\\\", # need to pull this from the mime lookup\\n \\\"file\\\" : \\\"placeholder\\\",\\n \\\"user_agent\\\" : user_agent,\\n \\\"remote_addr\\\" : remote_addr,\\n \\\"created\\\" : datetime.utcnow(),\\n \\\"location\\\" : string_from_interwebs(location).strip(),\\n \\\"author\\\" : ObjectId(author_id),\\n \\\"reply_to\\\" : ObjectId(reply_to),\\n \\\"tags\\\" : split_tags}\\n\\n update_post(reply_to, connection)\\n\\n return post_data.getfirst(\\\"data\\\"), post\",\n \"def PostData(self, postdata):\\n if len(postdata) > 0:\\n for key in postdata:\\n if postdata[key] == \\\"%TARGET%\\\":\\n postdata[key] = self._target\\n self._postdata = postdata\\n else:\\n self._postdata = None\",\n \"def add_post(content):\\n db = psycopg2.connect(database=DBNAME)\\n c = db.cursor()\\n cleaned = bleach.clean(content, strip = True)\\n c.execute(\\\"insert into posts values(%s)\\\", (cleaned,))\\n db.commit()\\n db.close()\",\n \"def process_postlist(app, doctree, docname):\\n blog = Blog(app)\\n if not blog:\\n register_posts(app)\\n for node in doctree.findall(PostList):\\n colls = []\\n for cat in [\\\"tags\\\", \\\"author\\\", \\\"category\\\", \\\"location\\\", \\\"language\\\"]:\\n for coll in node[cat]:\\n if coll in blog.catalogs[cat].collections:\\n colls.append(blog.catalogs[cat].collections[coll])\\n if colls:\\n posts = set(blog.posts)\\n for coll in colls:\\n posts = posts & set(coll)\\n posts = list(posts)\\n posts.sort(reverse=True)\\n posts = posts[: node.attributes[\\\"length\\\"]]\\n else:\\n posts = list(blog.recent(node.attributes[\\\"length\\\"], docname, **node.attributes))\\n if node.attributes[\\\"sort\\\"]:\\n posts.sort() # in reverse chronological order, so no reverse=True\\n fmts = list(Formatter().parse(node.attributes[\\\"format\\\"]))\\n not_in = {\\\"date\\\", \\\"title\\\", \\\"author\\\", \\\"location\\\", \\\"language\\\", \\\"category\\\", \\\"tags\\\", None}\\n for text, key, __, __ in fmts:\\n if key not in not_in:\\n raise KeyError(f\\\"{key} is not recognized in postlist format\\\")\\n excerpts = node.attributes[\\\"excerpts\\\"]\\n expand = node.attributes[\\\"expand\\\"]\\n date_format = node.attributes[\\\"date\\\"] or _(blog.post_date_format_short)\\n bl = nodes.bullet_list()\\n bl.attributes[\\\"classes\\\"].append(\\\"postlist-style-\\\" + node[\\\"list-style\\\"])\\n bl.attributes[\\\"classes\\\"].append(\\\"postlist\\\")\\n for post in posts:\\n bli = nodes.list_item()\\n bli.attributes[\\\"classes\\\"].append(\\\"ablog-post\\\")\\n bl.append(bli)\\n par = nodes.paragraph()\\n bli.append(par)\\n for text, key, __, __ in fmts:\\n if text:\\n par.append(nodes.Text(text))\\n if key is None:\\n continue\\n if key == \\\"date\\\":\\n par.append(nodes.Text(post.date.strftime(date_format)))\\n else:\\n if key == \\\"title\\\":\\n items = [post]\\n else:\\n items = getattr(post, key)\\n\\n for i, item in enumerate(items, start=1):\\n if key == \\\"title\\\":\\n ref = nodes.reference()\\n if item.options.get(\\\"external_link\\\"):\\n ref[\\\"refuri\\\"] = post.options.get(\\\"external_link\\\")\\n else:\\n ref[\\\"refuri\\\"] = app.builder.get_relative_uri(docname, item.docname)\\n ref[\\\"internal\\\"] = True\\n ref[\\\"ids\\\"] = []\\n ref[\\\"backrefs\\\"] = []\\n ref[\\\"dupnames\\\"] = []\\n ref[\\\"classes\\\"] = []\\n ref[\\\"names\\\"] = []\\n ref.append(nodes.Text(str(item)))\\n par.attributes[\\\"classes\\\"].append(\\\"ablog-post-title\\\")\\n else:\\n ref = _missing_reference(app, item.xref, docname)\\n par.append(ref)\\n if i < len(items):\\n par.append(nodes.Text(\\\", \\\"))\\n if excerpts and post.excerpt:\\n for enode in post.excerpt:\\n enode = enode.deepcopy()\\n enode.attributes[\\\"classes\\\"].append(\\\"ablog-post-excerpt\\\")\\n revise_pending_xrefs(enode, docname)\\n app.env.resolve_references(enode, docname, app.builder)\\n enode.parent = bli.parent\\n bli.append(enode)\\n if expand:\\n ref = app.builder.get_relative_uri(docname, post.docname)\\n enode = nodes.paragraph()\\n enode.attributes[\\\"classes\\\"].append(\\\"ablog-post-expand\\\")\\n refnode = nodes.reference(\\\"\\\", \\\"\\\", internal=True, refuri=ref)\\n innernode = nodes.emphasis(text=expand)\\n refnode.append(innernode)\\n enode.append(refnode)\\n bli.append(enode)\\n node.replace_self(bl)\",\n \"def parse_post_values(self): \\n self.parse_values(sys.stdin.read())\",\n \"def __init__(self, title, text, tags, created, modified, modifier):\\n self.title = title\\n self.text = text\\n self.tags = tags\\n self.created = created\\n self.modified = modified\\n self.modifier = modifier\",\n \"def process_posts_and_pages(*, posts, pages, settings):\\n for post in posts:\\n post.test_attr = 'post'\\n for page in pages:\\n page.test_attr = 'page'\\n return {'posts': posts, 'pages': pages}\",\n \"def analyse_posts(self, event):\\n\\n # Iterate over each post listed in \\\"deployed\\\"\\n for post in event[\\\"deployed\\\"]:\\n title = post.title()\\n\\n # Don't send for drafts\\n if post.is_draft or post.post_status != \\\"published\\\":\\n self.logger.info(\\n \\\"Skipping Draft Post {0} with status {1}\\\".format(\\n title, post.post_status\\n )\\n )\\n continue\\n\\n # Extract some details\\n link = post.permalink(absolute=True)\\n text = post.text()\\n self.logger.info(\\\"Processing {0}\\\".format(link))\\n\\n # Calculate and retrieve the state-key for this URL\\n key = \\\"webmention-info-{0}\\\".format(link)\\n observed_links = self.site.state.get(key)\\n\\n if not observed_links:\\n observed_links = []\\n\\n # Extract links from within the rendered page\\n links = self.extract_links(text)\\n\\n # Set up a requests session so that HTTP keep-alives can be used where possible\\n # this reduces connection overhead etc\\n session = requests.session()\\n\\n # Set the user-agent for all requests in this session\\n session.headers.update({\\\"User-Agent\\\": \\\"Nikola SSG Webmention plugin\\\"})\\n\\n # Send mentions for each\\n for dest in links:\\n\\n # See whether a webmention's already been sent for this page and url\\n # means we won't reping links every time a post is updated\\n if dest in observed_links:\\n continue\\n\\n sent, has_mentions = self.send_webmention(link, dest, session)\\n\\n # We want to cache two categories of link\\n #\\n # Has webmentions, sent successfully\\n # Does not have webmentions\\n\\n if sent or not has_mentions:\\n observed_links.append(dest)\\n\\n # Now that all links have been processed, save the state\\n self.site.state.set(key, observed_links)\",\n \"def post(self):\\n title = self.request.get(\\\"title\\\")\\n body = self.request.get(\\\"body\\\")\\n\\n if title and body:\\n\\n # create a new Post object and store it in the database\\n post = Post(\\n title=title,\\n body=body\\n )\\n post.put()\\n\\n # get the id of the new post, so we can render the post's page (via the permalink)\\n id = post.key().id()\\n self.redirect(\\\"/blog/%s\\\" % id)\\n else:\\n error = \\\"we need both a title and a body!\\\"\\n #self.render_form(title, body, error)\\n self.render(\\\"newpost.html\\\", title, body, error)\",\n \"def load_post(path: Path) -> LocalPost:\\n md_path = path / \\\"post.md\\\"\\n if not md_path.exists() or not md_path.is_file():\\n raise ValueError(f\\\"Could not find a post.md file in {path.absolute()}\\\")\\n meta_path = path / \\\"post-meta.json\\\"\\n if not meta_path.exists() or not meta_path.is_file():\\n raise ValueError(f\\\"Could not find a post-meta.json file in {path.absolute()}\\\")\\n try:\\n post_meta = PostMeta.parse_from_file(meta_path)\\n except ValueError as e:\\n raise e\\n return LocalPost(path, post_meta)\",\n \"def createPost(request):\\n\\n #save the organization's post\\n if request.method == 'POST':\\n form = PostForm(request.user, request.POST, request.FILES)\\n if form.is_valid():\\n filterList = ['everyone', 'black', 'hispanic', 'female', 'lgbt', 'immigrants', 'disabled', 'poor'] \\n newpost = form.save()\\n\\n #Add tags to the object only if in the filterlist\\n tags = form.cleaned_data.get('tags')\\n tags = [tag.lower() for tag in tags if tag.lower() in filterList]\\n\\n newpost.tags.add(*tags)\\n messages.success(request, 'You have successful created the post')\\n form = PostForm(request.user)\\n context = {'form':form} \\n return render(request, 'create_post.html', context=context)\\n \\n #form to fill out for the post\\n form = PostForm(request.user)\\n context = {'form':form} \\n return render(request, 'create_post.html', context=context)\",\n \"def process_post(new_post, cfg):\\n id_already_handled_in_db = i18n['debug']['id_already_handled_in_db']\\n discovered_submit_title = i18n['posts']['discovered_submit_title']\\n rules_comment = i18n['posts']['rules_comment']\\n yt_already_has_transcripts = i18n['posts']['yt_already_has_transcripts']\\n\\n if new_post['subreddit'] in cfg.upvote_filter_subs:\\n # ignore posts if they don't meet the threshold for karma and the sub\\n # is in our list of upvoted filtered ones\\n if new_post['ups'] < cfg.upvote_filter_subs[new_post['subreddit']]:\\n return\\n\\n if not is_valid(new_post['name'], cfg):\\n logging.debug(id_already_handled_in_db.format(new_post['name']))\\n return\\n\\n if new_post['archived']:\\n return\\n\\n if new_post['author'] is None:\\n # we don't want to handle deleted posts, that's just silly\\n return\\n\\n logging.info(\\n f'Posting call for transcription on ID {new_post[\\\"name\\\"]} posted by '\\n f'{new_post[\\\"author\\\"]}'\\n )\\n\\n if new_post['domain'] in cfg.image_domains:\\n content_type = 'image'\\n content_format = cfg.image_formatting\\n\\n elif new_post['domain'] in cfg.audio_domains:\\n content_type = 'audio'\\n content_format = cfg.audio_formatting\\n\\n elif new_post['domain'] in cfg.video_domains:\\n if 'youtu' in new_post['domain']:\\n if not valid_youtube_video(new_post['url']):\\n add_complete_post_id(new_post['name'], cfg)\\n return\\n if get_yt_transcript(new_post['url']):\\n np = cfg.r.submission(id=new_post['name'])\\n np.reply(_(\\n yt_already_has_transcripts\\n ))\\n add_complete_post_id(new_post['name'], cfg)\\n logging.info(\\n f'Found YouTube video, {get_yt_video_id(new_post[\\\"url\\\"])},'\\n f' with good transcripts.'\\n )\\n return\\n content_type = 'video'\\n content_format = cfg.video_formatting\\n else:\\n # This means we pulled from a subreddit bypassing the filters.\\n content_type = 'Other'\\n content_format = cfg.other_formatting\\n\\n # Truncate a post title if it exceeds 250 characters, so the added\\n # formatting still fits in Reddit's 300 char limit for post titles\\n post_title = new_post['title']\\n max_title_length = 250\\n if len(post_title) > max_title_length:\\n post_title = post_title[:max_title_length - 3] + '...'\\n\\n # noinspection PyBroadException\\n try:\\n result = cfg.tor.submit(\\n title=discovered_submit_title.format(\\n sub=new_post['subreddit'],\\n type=content_type.title(),\\n title=post_title\\n ),\\n url=reddit_url.format(new_post['permalink'])\\n )\\n result.reply(\\n _(\\n rules_comment.format(\\n post_type=content_type,\\n formatting=content_format,\\n header=cfg.header\\n )\\n )\\n )\\n flair_post(result, flair.unclaimed)\\n\\n add_complete_post_id(new_post['name'], cfg)\\n cfg.redis.incr('total_posted', amount=1)\\n\\n if cfg.OCR and content_type == 'image':\\n # hook for OCR bot; in order to avoid race conditions, we add the\\n # key / value pair that the bot isn't looking for before adding\\n # to the set that it's monitoring.\\n cfg.redis.set(new_post['name'], result.fullname)\\n cfg.redis.rpush('ocr_ids', new_post['name'])\\n\\n cfg.redis.incr('total_new', amount=1)\\n\\n # The only errors that happen here are on Reddit's side -- pretty much\\n # exclusively 503s and 403s that arbitrarily resolve themselves. A missed\\n # post or two is not the end of the world.\\n except Exception as e:\\n logging.error(\\n f'{e} - unable to post content.\\\\nID: {new_post[\\\"name\\\"]}\\\\n '\\n f'Title: {new_post[\\\"title\\\"]}\\\\n Subreddit: '\\n f'{new_post[\\\"subreddit\\\"]}'\\n )\",\n \"def cleanup_post(post):\\n \\n post_data = post\\n post_data[\\\"id\\\"] = str(post[\\\"_id\\\"])\\n post_data[\\\"author\\\"] = str(post[\\\"author\\\"])\\n post_data[\\\"created\\\"] = str(post[\\\"created\\\"].ctime())\\n del post_data[\\\"_id\\\"]\\n \\n if \\\"reply_to\\\" in post:\\n post_data[\\\"reply_to\\\"] = str(post[\\\"reply_to\\\"])\\n\\n if \\\"repost_of\\\" in post:\\n post_data[\\\"repost_of\\\"] = str(post[\\\"repost_of\\\"])\\n\\n return post_data\",\n \"def parse(self, content):\\n pass\",\n \"def xmlpost_to_dict(post):\\n\\n tree = ET.parse(post)\\n root = tree.getroot()\\n msg = root.find('message')\\n\\n post_data = {}\\n\\n board_id = msg.find('board_id')\\n post_data['board_id'] = int(board_id.text)\\n\\n root_post = msg.find('root').attrib['href']\\n post_data['root_post'] = root_post.split('/')[-1]\\n\\n kudos = msg.find('kudos')\\n count = kudos.find('count')\\n post_data['kudos_count'] = int(count.text)\\n\\n edit_author_id = msg.find('last_edit_author').attrib['href']\\n post_data['edit_author_id'] = int(edit_author_id.split('/')[-1])\\n\\n post_time = msg.find('post_time')\\n post_data['post_time'] = post_time.text\\n\\n last_edit_time = msg.find('last_edit_time')\\n post_data['last_edit_time'] = last_edit_time.text\\n\\n body = msg.find('body')\\n post_data['body'] = body.text\\n\\n thread = msg.find('thread').attrib['href']\\n post_data['thread'] = int(thread.split('/')[-1])\\n\\n board = msg.find('board').attrib['href']\\n post_data['board'] = board.split('/')[-1]\\n\\n try:\\n parent_post = msg.find('parent').attrib['href']\\n post_data['parent_post'] = int(parent_post.split('/')[-1])\\n except KeyError:\\n post_data['parent_post'] = None\\n\\n views = msg.find('views')\\n post_data['views'] = int(views.find('count').text)\\n\\n subject = msg.find('subject')\\n post_data['subject'] = subject.text\\n\\n post_id = msg.find('id')\\n post_data['post_id'] = int(post_id.text)\\n\\n author_id = msg.find('author').attrib['href']\\n post_data['author_id'] = int(author_id.split('/')[-1])\\n\\n return post_data\",\n \"def parse_article_text(article_text, post_date=datetime.datetime.now()):\\n result = {}\\n\\n metadata_start = 0\\n main_content_start = 0\\n main_content_end = len(article_text)\\n \\n article_date_match = re.search(r\\\"^Date:\\\\s(?P<date>[^\\\\(\\\\[\\\\n]+)\\\", article_text, re.M)\\n if article_date_match:\\n # There may be more than one source date in summary articles.\\n # Example: http://promedmail.org/direct.php?id=1073176\\n # Summary articles are not a focus so currently only the first date\\n # is recorded.\\n source_date = parse_datetime(\\n article_date_match.group(\\\"date\\\")\\n )\\n\\n if source_date:\\n result[\\\"date\\\"] = datetime_to_utc(source_date)\\n metadata_start = min(article_date_match.start(), metadata_start)\\n main_content_start = max(article_date_match.end(), main_content_start)\\n # The year is checked to avoid typos like 200_ that throw\\n # the date off by a large factor.\\n # Example: http://www.promedmail.org/direct.php?id=45850 (article 2)\\n if result[\\\"date\\\"].year < 1900:\\n result[\\\"date\\\"] = None\\n # Some articles have timestamps that are incorrectly parsed.\\n # Current examples:\\n # http://www.promedmail.org/direct.php?id=43918\\n # http://www.promedmail.org/direct.php?id=2200173\\n # Some of these incorrect timestamps can be removed by verifying that\\n # they preceed the time of the posting. A day of slop time is allowed\\n # to account for variations due to incorrect timezones.\\n elif result[\\\"date\\\"] > post_date + datetime.timedelta(1):\\n result[\\\"date\\\"] = None\\n else:\\n result[\\\"date\\\"] = None\\n \\n source_match = re.search(r\\\"Source:\\\\s(?P<name>[^\\\\[\\\\n]+)\\\" +\\\\\\n r\\\"(\\\\s(?P<edits>\\\\[.*))?\\\" +\\\\\\n r\\\"\\\\n\\\" +\\\\\\n r\\\"(?P<url>http.+)?\\\", article_text)\\n if source_match:\\n result[\\\"source\\\"] = source_match.groupdict()\\n metadata_start = min(source_match.start(), metadata_start)\\n main_content_start = max(source_match.end(), main_content_start)\\n \\n heading_match = re.search(r\\\"^(?P<idx>\\\\[\\\\d\\\\]\\\\s)?\\\" +\\\\\\n r\\\"(?P<heading>\\\\S+.*)\\\\n\\\",\\n article_text[0:metadata_start], re.M)\\n if heading_match:\\n result[\\\"heading\\\"] = heading_match.group(\\\"heading\\\")\\n \\n communicated_match = re.search(communicated_by_regex, article_text, re.M)\\n if communicated_match:\\n result[\\\"communicatedBy\\\"] = communicated_match.group(\\\"communicated_by\\\")\\n main_content_end = min(communicated_match.start(), main_content_end)\\n \\n result[\\\"content\\\"] = article_text[main_content_start:main_content_end].strip()\\n return result\",\n \"def postCreate(post):\\n post_list = list()\\n comments = commentList(post)\\n comment_url = \\\"{}/api/posts/{}/comments\\\".format(DOMAIN, post.id)\\n visible_to = list()\\n visible = post.visibleTo.all()\\n if visible:\\n for author in visible:\\n auth = \\\"{}/api/author/{}\\\".format(DOMAIN, author.id)\\n visible_to.append(auth)\\n\\n # visible_to = list(post.visibleTo)\\n post_dict = {'author': addAuthor(post.author), 'title': post.title, 'description': post.description,\\n 'contentType': post.contentType, 'content': post.content, 'published': post.published,\\n 'visibility': post.visibility, 'visibleTo': visible_to, 'unlisted': post.unlisted, 'id': post.id,\\n 'comments': comments[:5], 'next': comment_url, 'count': len(comments),\\n 'source': \\\"{}/api/posts/{}\\\".format(DOMAIN, post.id),\\n 'origin': \\\"{}/api/posts/{}\\\".format(DOMAIN, post.id)}\\n post_list.append(post_dict)\\n return post_list\",\n \"def addPost(self,text,id,url,date):\\n self.topComments.append(Post(text,id,url,date))\\n return None\",\n \"def mutate(self, info, **arguments):\\n title = arguments.get('title')\\n text = arguments.get('text')\\n\\n # Get author id from `relay` id and find author on database\\n author_id = from_global_id(arguments.get('author_id'))[1]\\n author = UserModel.objects.get(pk=author_id)\\n\\n # Get post id from `relay` id and find it on database\\n post_id = from_global_id(arguments.pop('id'))[1]\\n post = PostModel.objects.get(pk=post_id)\\n\\n if not author:\\n print(\\\"=\\\" * 100)\\n print(\\\"schema.py@41\\\")\\n print(\\\"We should return an error here\\\")\\n print(\\\"=\\\" * 100)\\n return CreatePostMutation(success=False)\\n\\n post = PostModel.objects.create(title=title, author=author, text=text)\\n return CreatePostMutation(success=True, post=post)\",\n \"def set_date_post(self, date_post):\\n if check_data_exist(date_post) is True:\\n self.date_post = date_post.text\",\n \"def wp2fields(xml, wp_custpost=False):\\r\\n\\r\\n items = get_items(xml)\\r\\n for item in items:\\r\\n\\r\\n if item.find('status').string == \\\"publish\\\":\\r\\n\\r\\n try:\\r\\n # Use HTMLParser due to issues with BeautifulSoup 3\\r\\n title = HTMLParser().unescape(item.title.contents[0])\\r\\n except IndexError:\\r\\n title = 'No title [%s]' % item.find('post_name').string\\r\\n logger.warning('Post \\\"%s\\\" is lacking a proper title' % title)\\r\\n\\r\\n filename = item.find('post_name').string\\r\\n post_id = item.find('post_id').string\\r\\n filename = get_filename(filename, post_id)\\r\\n\\r\\n content = item.find('encoded').string\\r\\n raw_date = item.find('post_date').string\\r\\n date_object = time.strptime(raw_date, \\\"%Y-%m-%d %H:%M:%S\\\")\\r\\n date = time.strftime(\\\"%Y-%m-%d %H:%M\\\", date_object)\\r\\n author = item.find('creator').string\\r\\n\\r\\n categories = [cat.string for cat in item.findAll('category', {'domain' : 'category'})]\\r\\n # caturl = [cat['nicename'] for cat in item.find(domain='category')]\\r\\n\\r\\n tags = [tag.string for tag in item.findAll('category', {'domain' : 'post_tag'})]\\r\\n\\r\\n kind = 'article'\\r\\n post_type = item.find('post_type').string\\r\\n if post_type == 'page':\\r\\n kind = 'page'\\r\\n elif wp_custpost:\\r\\n if post_type == 'post':\\r\\n pass\\r\\n # Old behaviour was to name everything not a page as an article.\\r\\n # Theoretically all attachments have status == inherit so\\r\\n # no attachments should be here. But this statement is to\\r\\n # maintain existing behaviour in case that doesn't hold true.\\r\\n elif post_type == 'attachment':\\r\\n pass\\r\\n else:\\r\\n kind = post_type\\r\\n yield (title, content, filename, date, author, categories, tags,\\r\\n kind, \\\"wp-html\\\")\",\n \"def post(self):\\n args = parser.parse_args()\",\n \"def newPost(self, postLink, zserverBlogEntry): #$NON-NLS-1$\\r\\n atomEntry = self.createNewBlogEntry()\\r\\n self._populateAtomEntry(atomEntry, zserverBlogEntry)\\r\\n # publish entry\\r\\n atomRespEntry = self.createAtomEntry(postLink, atomEntry)\\r\\n return atomRespEntry\",\n \"def __init__(self, rpc, post):\\n super(PostModule, self).__init__(rpc, 'post', post)\\n self._action = self._info.get('default_action', \\\"\\\")\",\n \"def create(self, validated_data):\\n \\\"\\\"\\\" Create post with a location \\\"\\\"\\\"\\n location_data = validated_data.pop('location')\\n\\n # create a new one or get a old for reference\\n this_location = Location.objects.get_or_create(\\n **location_data\\n )\\n\\n # pop the photo url's data\\n photo_data = validated_data.pop('photo')\\n\\n # must pop the tags data before it would used to create a post \\n tags_data = validated_data.pop('tag')\\n # create a instance of this post\\n this_post = Post.objects.create(\\n location = this_location[0],\\n **validated_data)\\n\\n \\\"\\\"\\\"Associate tag's informatiion to post\\\"\\\"\\\"\\n for tag in tags_data:\\n this_tag = Tag.objects.get_or_create(name = tag.get('name'))\\n print(tag.get('name'))\\n print(this_tag)\\n # attach this tag to this photos_datapost \\n this_post.tag.add(this_tag[0])\\n\\n \\\"\\\"\\\"Associate the photo url \\\"\\\"\\\"\\n for photo in photo_data:\\n this_post.photo.create(name = photo.get('name'))\\n # return the created post \\n this_post.save()\\n return this_post\",\n \"def create(cls, headline, text, blog):\\n post = cls()\\n post.headline = headline\\n post.text = text\\n post.blog = blog\\n post.posted_date = timezone.now()\\n try:\\n post.save()\\n return post\\n except(ValueError, IntegrityError, OperationalError):\\n return None\",\n \"def add_post(content):\\n db = psycopg2.connect(\\\"dbname=forum\\\")\\n c = db.cursor()\\n content = bleach.clean(content)\\n c.execute(\\\"insert into posts values (%s)\\\", (content,))\\n db.commit()\\n db.close()\\n # POSTS.append((content, datetime.datetime.now()))\",\n \"def adapt_tweet(feedpost):\\n tweet = feedpost['title']\\n for action in (_make_links, _clean_name, urlize):\\n tweet = action(tweet)\\n feedpost['title'] = _get_tweet_number(feedpost['link'])\\n feedpost['body'] = u'<p>%s</p>' % tweet\\n return feedpost\",\n \"def each_comment_from_post(post):\\n # first yield the post text body, if any\\n if post['text']:\\n yield post['text']\\n # then yield each comment\\n for comment in post['comments']:\\n yield comment['text']\",\n \"def post_create(faker_obj, profile_obj, tag_list, num=3):\\n for i in range(num):\\n obj = faker_obj\\n title = obj.sentence(nb_words=random.randint(5, 10))\\n author = User.objects.get(id=profile_obj)\\n body = \\\" \\\".join(obj.paragraphs(nb=random.randint(8, 20)))\\n status = \\\"published\\\"\\n post = Post.objects.create(title=title, author=author, body=body, status=status)\\n post.tags.add(\\\", \\\".join(random.sample(tag_list, 1)))\\n print(\\n \\\"Created post title:'{}' for user '{}'\\\".format(post.title, author.username)\\n )\\n create_comment_list(obj, post)\",\n \"def setUp(self):\\n self.xpaths = [\\\"./author/text()\\\",\\n \\\"./title/text()\\\",\\n \\\"./time/text()\\\",\\n \\\"./number/text()\\\",\\n \\\"./tags/tag/text()\\\",\\n \\\" \\\"]\\n self.meta = {\\\"Time\\\": \\\"2012-02-23\\\",\\n \\\"Author\\\": \\\"rachel\\\",\\n \\\"Title\\\": \\\"Post 1\\\",\\n \\\"Number\\\": \\\"1\\\",\\n \\\"Tag\\\": [\\\"Tag 0\\\", \\\"Tag 3\\\"]}\\n self.sentence_contents = \\\"This is the text of post 1. I love clouds.\\\"\\n super(PostTests, self).setUp(\\n \\\"tests/data/articles/\\\", \\\"structure.json\\\", \\\"post1.xml\\\")\",\n \"def _update_post_entry(self, target, source):\\n target.post_id = target.post_id if source.id==source.id else source.id\\n # target.author = target.author if source.author==source.author else source.author\\n target.author_premium = target.author_premium if source.author_premium==source.author_premium else source.author_premium\\n target.subreddit_subscribers = target.subreddit_subscribers if source.subreddit_subscribers==source.subreddit_subscribers else source.subreddit_subscribers\\n target.title = target.title if source.title==source.title else source.title\\n target.downs = target.downs if source.downs==source.downs else source.downs\\n target.ups = target.ups if source.ups==source.ups else source.ups\\n target.selftext = target.selftext if source.selftext==source.selftext else source.selftext\\n target.num_comments = target.num_comments if source.num_comments==source.num_comments else source.num_comments\\n target.total_awards_received = target.total_awards_received if source.total_awards_received==source.total_awards_received else source.total_awards_received\\n target.view_count = target.view_count if source.view_count==source.view_count else source.view_count\\n # target.permalink = target.permalink if source.permalink==source.permalink else source.permalink\\n # target.url = target.url if source.url==source.url else source.url\\n # target.created = target.created if source.created==source.created else source.created\\n # target.created_utc = target.created_utc if source.created_utc==source.created_utc else source.created_utc\\n\\n return target\",\n \"def parse_text(self, page):\\n text = page.find(self.tag_prefix + self.revision_tag).find(self.tag_prefix + self.text_tag).text\\n title = page.find(self.tag_prefix + self.title_tag).text\\n categories = []\\n #\\n text = self.parse_archivo(text)\\n text = self.parse_foto(text)\\n text = self.parse_by_line(text)\\n text = self.parse_link(text)\\n text = self.parse_url(text)\\n text = self.parse_fecha(text)\\n text = self.parse_bracketed_word(text)\\n #\\n if text:\\n categories = re.findall(self.category_finder_regex, text)\\n #\\n text = self.parse_category(text)\\n text = self.parse_other_language(text)\\n text = self.parse_table_regex(text)\\n text = self.parse_ver_fuente(text)\\n text = self.remove_extra_text(text)\\n text = self.remove_extra_characters(text)\\n\\n categorias = []\\n for cat in categories:\\n categorias.append(cat[6])\\n\\n if text:\\n if 'REDIRECT' in text or 'redirect' in text:\\n return None\\n\\n return Article(title=title, content=text, categories=categorias)\",\n \"def get_post_attrs(post_content, **kwargs):\\n ref_filters = kwargs[\\\"ref_filters\\\"]\\n # Attributes without a recognizable ID will be appended to \\\"misc\\\".\\n post_attrs = {\\\"misc\\\": []}\\n for attr in get_attrs(post_content):\\n # Add to post_attrs !! BY REFERENCE !!.\\n parse_attrs(post_attrs, attr, ref_filters)\\n\\n return post_attrs\",\n \"def _parse(self):\\n soup = BS(self._current_html, 'lxml')\\n for item in soup.select('div.c'):\\n temp = {}\\n # main content\\n ctt = item.select('span.ctt')\\n if not ctt:\\n continue\\n weibo_body = item.select('div')\\n if len(weibo_body) > 1:\\n temp['content'] = weibo_body[0].text\\n btn_group = weibo_body[1].text\\n else:\\n temp['content'] = weibo_body[0].select('span.ctt')[0].text\\n btn_group = weibo_body[0].text\\n temp['is_repost'] = True if REPO_TEST_PATTERN.match(\\n temp['content']) else False\\n try:\\n temp['like_num'] = LIKE_NUM_PATTERN.findall(btn_group)[0]\\n temp['cmt_num'] = COMMENT_NUM_PATTERN.findall(btn_group)[0]\\n temp['repo_num'] = REPO_NUM_PATTERN.findall(btn_group)[0]\\n except Exception:\\n pass\\n cmt = item.select('.cmt')\\n # visibility\\n if cmt:\\n try:\\n temp['visibility'] = VISIBILITY_PATTERN.findall(\\n cmt[0].text)[0]\\n except Exception:\\n pass\\n\\n # img in main content\\n img = item.select('div a img')\\n img_src = img[0].attrs['src'] if img else None\\n temp['img_src'] = img_src\\n LOGGER.debug('img_src: {}'.format(img_src))\\n # time & source device\\n ct = item.select('span.ct')\\n if ct:\\n ct = ct[0]\\n text = ct.text\\n reg_result = TIME_PATTERN.findall(text)[0]\\n\\n temp['time'] = ar(\\n '{}年{}'.format(self._current_year, reg_result[0]),\\n DATE_FMTS[0]\\n ).naive if reg_result[0] else ar(\\n reg_result[1], DATE_FMTS[1]\\n ).naive\\n temp['source'] = SOURCE_DEVICE_PATTERN.findall(text)[0]\\n self._post_item = Post(**temp)\\n self._attachment_item = Attachment(\\n uri=img_src, post=self._post_item)\\n self._store()\",\n \"def process_entry(entry, blog, START):\\n try:\\n when = entry['updated_parsed']\\n except KeyError:\\n try:\\n when = entry['published_parsed']\\n except KeyError:\\n return # Ignore undateable posts\\n\\n if when:\\n when = pytz.timezone('UTC').localize(datetime.fromtimestamp(time.mktime(when)))\\n else:\\n # print blog, entry\\n return\\n\\n if when < START:\\n return\\n\\n title = entry.get('title', \\\"Null\\\")\\n\\n try:\\n author = entry['author']\\n except KeyError:\\n try:\\n author = ', '.join(a['name'] for a in entry.get('authors', []))\\n except KeyError:\\n author = 'Anonymous'\\n\\n link = entry['link']\\n\\n try:\\n body = entry['content'][0]['value'].replace(\\\"h1>\\\",\\\"b>\\\").replace(\\\"h2>\\\",\\\"b>\\\")\\n except KeyError:\\n body = entry['summary']\\n\\n return Post(when, blog, title, author, link, body)\",\n \"def parse_posts(posts_dict):\\n return posts_dict['posts']\",\n \"def insert_post(data, db_conn):\\n\\n schema = get_post_schema(data)\\n data, errors = insert_document(schema, data, db_conn)\\n if not errors:\\n add_post_to_es(data, db_conn)\\n return data, errors\",\n \"def test_post(self):\\n self.assertEqual(self.post.title, \\\"Ugandan anthem\\\")\\n self.assertEqual(self.post.body, \\\"George William Kakoma\\\")\\n self.assertEqual(str(self.post), \\\"Ugandan anthem - George William Kakoma\\\")\",\n \"def post(self):\\n subject = self.request.get('subject')\\n content = self.request.get('content')\\n\\n # if user enter good subject and content, redirect them to new post page\\n if subject and content:\\n p = Post(parent = blog_key(), subject = subject, content = content)\\n p.put() # store the post element into database\\n self.redirect('/blog/%s' % str(p.key().id()))\\n # otherwise, render an error page \\n else:\\n error = \\\"subject and content, please!\\\"\\n self.render(\\\"newpost.html\\\", subject=subject, content=content, error=error)\",\n \"def add_to_postre(self, post_course):\\n self.postre[post_course.get_course_id()] = post_course\",\n \"def test_parse_post_code_field(self):\\n fields = {'Post code': {'offset': 171,\\n 'length': 4}}\\n p = top.Parser(fields=fields)\\n received = p.parse_line(self._line)\\n expected = {'Post code': '2048'}\\n msg = 'Post code field parse incorrect'\\n self.assertEqual(received, expected, msg)\",\n \"def editPost(self, id, useRawHTML):\\n old_data = self._extractPost(id)\\n print\\n content, publish = self._fillPost(useRawHTML, old_data)\\n\\n # Upload to server\\n try :\\n self.server.metaWeblog.editPost(\\n id, self.username, self.password,\\n content, publish\\n )\\n if raw_input(\\\"Change category ?[y|N] \\\") == \\\"y\\\" :\\n self._setCategorie(id)\\n except xmlrpclib.Fault as fault:\\n display_XMLRPC_errors(\\\"edit entry\\\", fault)\",\n \"def sortByPostText(self,post_text,desc):\\n \\n return self.__sortByPostText(post_text, \\\"\\\", desc)\",\n \"def proceed(self, source):\\n\\n for line in source:\\n line = line.rstrip()\\n if istag(line) and isinstance(maketag(line), Time): # here, handle Time tags\\n self.time_annotate_date(maketag(line)) # annotate Date tag and queue Time tag\\n if not default_options.mdjpp_time_propagate:\\n continue\\n # not a tag!\\n if not istag(line) or isinstance(maketag(line), Time):\\n if self.metastable_tag: # and len(line):\\n # open metastable tag\\n self.open_tag(self.metastable_tag)\\n self.metastable_tag = None\\n # print line\\n if self.printable():\\n self.printme_text(line)\\n # tag!\\n else:\\n # make regular tag\\n this_tag = maketag(line)\\n # no metastable tag\\n if not self.metastable_tag:\\n # make new metastable tag\\n self.metastable_tag = this_tag\\n # does new tag close last tag?\\n self.close_tag(this_tag)\\n # metastable tag present\\n else:\\n # is metastable tag or this tag of Date type?\\n if isinstance(self.metastable_tag, Date) or isinstance(this_tag, Date):\\n # open metastable tag\\n self.open_tag(self.metastable_tag)\\n # recreate metastable tag with new tag\\n self.metastable_tag = this_tag\\n # does new tag close last tag?\\n self.close_tag(this_tag)\\n else:\\n # add new tag to metastable tag\\n self.metastable_tag.append(this_tag)\",\n \"def parse_post_category(advertise: Dict[str, Any]) -> Optional[List[str]]:\\n\\n noise_terms: List[str] = [\\\"voltar\\\"]\\n\\n if \\\"post_category\\\" in advertise.keys():\\n tmp: List[str] = re.split('[ \\\\t]{2,}', advertise[\\\"post_category\\\"])\\n tmp = Parser.remove_noise_terms(tmp, noise_terms)\\n return tmp\"\n]","isConversational":false},"negative_scores":{"kind":"list like","value":["0.612241","0.59482175","0.5902697","0.5901005","0.5845779","0.58072627","0.5710131","0.56681836","0.5645754","0.5642677","0.5604725","0.5604725","0.55815965","0.55642414","0.54975563","0.54782796","0.5466466","0.54545844","0.54401195","0.54401195","0.541922","0.5409034","0.5400706","0.5390442","0.538884","0.53849745","0.53497034","0.5344574","0.5329334","0.53209454","0.53066385","0.5302247","0.5282716","0.5240877","0.52024925","0.520177","0.51928824","0.5187756","0.5172042","0.51632726","0.51106536","0.51057374","0.50753975","0.50450104","0.5029565","0.50281745","0.5028038","0.50277686","0.50206065","0.50187635","0.50064784","0.50050604","0.49763513","0.4973024","0.4969637","0.49462712","0.49454296","0.49374026","0.49302164","0.49204236","0.49051136","0.48870695","0.4882631","0.48819044","0.48751384","0.4869544","0.4867964","0.48675692","0.4864146","0.4852124","0.4842459","0.48399118","0.4828747","0.4824032","0.4807159","0.4785618","0.47825232","0.4771394","0.47635496","0.47609276","0.47497216","0.47360307","0.4734091","0.47188875","0.47126693","0.47065228","0.4706451","0.47057736","0.47005826","0.4692655","0.4684948","0.46840125","0.4677141","0.46709916","0.46611884","0.46605882","0.4652629","0.46437302","0.46371007","0.46330455"],"string":"[\n \"0.612241\",\n \"0.59482175\",\n \"0.5902697\",\n \"0.5901005\",\n \"0.5845779\",\n \"0.58072627\",\n \"0.5710131\",\n \"0.56681836\",\n \"0.5645754\",\n \"0.5642677\",\n \"0.5604725\",\n \"0.5604725\",\n \"0.55815965\",\n \"0.55642414\",\n \"0.54975563\",\n \"0.54782796\",\n \"0.5466466\",\n \"0.54545844\",\n \"0.54401195\",\n \"0.54401195\",\n \"0.541922\",\n \"0.5409034\",\n \"0.5400706\",\n \"0.5390442\",\n \"0.538884\",\n \"0.53849745\",\n \"0.53497034\",\n \"0.5344574\",\n \"0.5329334\",\n \"0.53209454\",\n \"0.53066385\",\n \"0.5302247\",\n \"0.5282716\",\n \"0.5240877\",\n \"0.52024925\",\n \"0.520177\",\n \"0.51928824\",\n \"0.5187756\",\n \"0.5172042\",\n \"0.51632726\",\n \"0.51106536\",\n \"0.51057374\",\n \"0.50753975\",\n \"0.50450104\",\n \"0.5029565\",\n \"0.50281745\",\n \"0.5028038\",\n \"0.50277686\",\n \"0.50206065\",\n \"0.50187635\",\n \"0.50064784\",\n \"0.50050604\",\n \"0.49763513\",\n \"0.4973024\",\n \"0.4969637\",\n \"0.49462712\",\n \"0.49454296\",\n \"0.49374026\",\n \"0.49302164\",\n \"0.49204236\",\n \"0.49051136\",\n \"0.48870695\",\n \"0.4882631\",\n \"0.48819044\",\n \"0.48751384\",\n \"0.4869544\",\n \"0.4867964\",\n \"0.48675692\",\n \"0.4864146\",\n \"0.4852124\",\n \"0.4842459\",\n \"0.48399118\",\n \"0.4828747\",\n \"0.4824032\",\n \"0.4807159\",\n \"0.4785618\",\n \"0.47825232\",\n \"0.4771394\",\n \"0.47635496\",\n \"0.47609276\",\n \"0.47497216\",\n \"0.47360307\",\n \"0.4734091\",\n \"0.47188875\",\n \"0.47126693\",\n \"0.47065228\",\n \"0.4706451\",\n \"0.47057736\",\n \"0.47005826\",\n \"0.4692655\",\n \"0.4684948\",\n \"0.46840125\",\n \"0.4677141\",\n \"0.46709916\",\n \"0.46611884\",\n \"0.46605882\",\n \"0.4652629\",\n \"0.46437302\",\n \"0.46371007\",\n \"0.46330455\"\n]","isConversational":false},"document_score":{"kind":"string","value":"0.6026173"},"document_rank":{"kind":"string","value":"1"}}},{"rowIdx":93281,"cells":{"query":{"kind":"string","value":"Create activity for published Post."},"document":{"kind":"string","value":"def on_post_published(sender, instance: dillo.models.posts.Post, **kwargs):\n # If activity was already created (the post was already published),\n # skip this signal to prevent the activity from appearing twice in the feeds.\n if models_actstream.Action.objects.filter(\n actor_object_id=instance.user.id, action_object_object_id=instance.id, verb='posted'\n ).exists():\n log.info('Content already published, skipping notification generation')\n return\n action.send(instance.user, verb='posted', action_object=instance)"},"metadata":{"kind":"string","value":"{\n \"objective\": {\n \"self\": [],\n \"paired\": [],\n \"triplet\": [\n [\n \"query\",\n \"document\",\n \"negatives\"\n ]\n ]\n }\n}"},"negatives":{"kind":"list like","value":["def post_to_activity(self, post):\n id = None\n if post.get('id'):\n # strip USERID_ prefix if it's there\n post['id'] = post['id'].split('_', 1)[-1]\n id = post['id']\n\n obj = self.post_to_object(post)\n activity = {\n 'verb': VERBS.get(post.get('type', obj.get('objectType')), 'post'),\n 'published': obj.get('published'),\n 'updated': obj.get('updated'),\n 'id': self.tag_uri(id) if id else None,\n 'url': self.post_url(post),\n 'actor': obj.get('author'),\n 'object': obj,\n }\n\n application = post.get('application')\n if application:\n activity['generator'] = {\n 'displayName': application.get('name'),\n 'id': self.tag_uri(application.get('id')),\n }\n return self.postprocess_activity(activity)","def newPost(self, postLink, zserverBlogEntry): #$NON-NLS-1$\r\n atomEntry = self.createNewBlogEntry()\r\n self._populateAtomEntry(atomEntry, zserverBlogEntry)\r\n # publish entry\r\n atomRespEntry = self.createAtomEntry(postLink, atomEntry)\r\n return atomRespEntry","def create_a_post():\n subj = create_subject()\n post = Post.create(subject=subj, title=\"A great title\", body=\"Just a great day!\")\n post.save()\n return post","def post(self):\n data = request.json\n\n now = datetime.utcnow().isoformat()\n data['created_time'] = now\n\n # Our type is already registered within the DB, so generate a\n # model object that looks like what we'll be interacting with\n try:\n activity = Activity(data)\n except KeyError:\n raise BadRequest(\"payload validation failed: {}\".format(data))\n\n activity.save()\n log.debug(\"Wrote activity: \" + str(activity._to_dict()))\n return activity._to_dict(), 201","def create_post(category, author, name, content, status):\n return Post.objects.create(category=category, author=author, name=name, content=content, status=status)","def postCreate(post):\n post_list = list()\n comments = commentList(post)\n comment_url = \"{}/api/posts/{}/comments\".format(DOMAIN, post.id)\n visible_to = list()\n visible = post.visibleTo.all()\n if visible:\n for author in visible:\n auth = \"{}/api/author/{}\".format(DOMAIN, author.id)\n visible_to.append(auth)\n\n # visible_to = list(post.visibleTo)\n post_dict = {'author': addAuthor(post.author), 'title': post.title, 'description': post.description,\n 'contentType': post.contentType, 'content': post.content, 'published': post.published,\n 'visibility': post.visibility, 'visibleTo': visible_to, 'unlisted': post.unlisted, 'id': post.id,\n 'comments': comments[:5], 'next': comment_url, 'count': len(comments),\n 'source': \"{}/api/posts/{}\".format(DOMAIN, post.id),\n 'origin': \"{}/api/posts/{}\".format(DOMAIN, post.id)}\n post_list.append(post_dict)\n return post_list","def create_post():\n\n #Get prompt id\n prompt_id = request.form.get('prompt_id')\n\n # Get post text\n post_text = request.form.get('user_post')\n\n # Create post timestamp\n created_at = datetime.now()\n user_facing_date = created_at.strftime(\"%B %d, %Y\")\n\n # Save post and related data to database\n post = crud.create_post(session['user_id'], prompt_id, post_text, session['lat'], session['lng'], session['user_facing_location'], created_at)\n\n return render_template('post_data.html', post=post, user_facing_date=user_facing_date)","def post_publish(request, pk):\n post = get_object_or_404(Post, pk=pk)\n post.publish()\n return redirect('post_detail', pk=pk)","def post_activities():\n pass","def publish(self):\n self.published_date = timezone.now()\n self.save()","def publish(self):\n self.published_date = timezone.now()\n self.save()","def publish(self):\n self.published_date = timezone.now\n self.save()","def remotePostCreate(host, post):\n post = post.get('posts')[0]\n author = remoteAddAuthor(post.get('author'))\n title = post.get('title')\n description = post.get('description')\n contentType = post.get('contentType')\n content = post.get('content')\n published = utc.localize(datetime.strptime(post.get('published'), '%Y-%m-%dT%H:%M:%S.%fZ'))\n visibility = post.get('visibility')\n unlisted = post.get('unlisted')\n id = post.get('id')\n origin = post.get('origin')\n count = post.get('count')\n comments = remoteCommentList(post)\n source = \"{}/api/posts/{}\".format(DOMAIN, post.get('id'))\n\n post_dict = {'author': author, 'title': title, 'description': description,\n 'contentType': contentType, 'content': content, 'published': published,\n 'visibility': visibility, 'unlisted': unlisted, 'id': id,\n 'comments': comments, 'origin': origin, 'count': count,\n 'source': source}\n return post_dict","def new_post(self, content):\n return self.proxy.wp.newPost(self.blog_id, self.username, self.password,\n content)","def on_created_action(sender, instance: models_actstream.Action, created, **kwargs):\n if not created:\n return\n\n # If content_type is Post or Profile\n if (\n instance.action_object_content_type\n == ContentType.objects.get_for_model(dillo.models.posts.Post)\n and instance.verb == 'posted'\n ) or (\n instance.action_object_content_type\n == ContentType.objects.get_for_model(dillo.models.profiles.Profile)\n and instance.verb == 'updated their reel'\n ):\n dillo.tasks.feeds.establish_time_proximity(instance)\n\n log.debug('Creating background fanout operation for action %i' % instance.id)\n dillo.tasks.feeds.activity_fanout_to_feeds(instance.id)","def insert_new_post(post_arg_set):\n api, post_data, acct_data, page_id, config = post_arg_set\n\n try:\n post_id = post_data['id'] if post_data.has_key('id') else None\n\n except Exception as e:\n log.error( e )\n\n else:\n\n # parse date\n if post_data.has_key('created_time') and post_data['created_time'] is not None: \n dt = datetime.strptime(post_data['created_time'], FB_DATE_FORMAT)\n date_time = tz_adj(dt, config)\n time_bucket = round_datetime(date_time, config)\n raw_timestamp = int(date_time.strftime(\"%s\"))\n \n else:\n time_bucket = None\n raw_timestamp = None\n \n # extract message so we can find links within the msg if not in url\n article_urls = [get_fb_link(post_data, config, unshorten=True)]\n message = post_data['message'].encode('utf-8') if post_data.has_key('message') else None\n message_urls = get_message_urls(article_urls, message, config)\n\n # detect article links, unshorten and parse\n article_urls = [\n parse_url(unshorten_link(url, config)) \\\n for url in article_urls + message_urls\n if url is not None\n ]\n\n article_urls = [url for url in article_urls if is_article(url, config)]\n\n if article_urls:\n for article_url in set(article_urls):\n\n # sluggify url\n article_slug = sluggify(article_url)\n\n # format data\n post_value = {\n 'article_slug': article_slug,\n 'article_url': article_url,\n 'time_bucket': time_bucket,\n 'fb_post_created': raw_timestamp,\n 'raw_timestamp': raw_timestamp,\n 'fb_raw_link' : get_fb_link(post_data, config=config),\n 'fb_page_id': page_id,\n 'fb_post_id': post_id,\n 'fb_page_likes': acct_data['likes'] if acct_data.has_key('likes') else None,\n 'fb_page_talking_about': acct_data['talking_about_count'] if acct_data.has_key('talking_about_count') else None,\n 'fb_type': post_data['type'] if post_data.has_key('type') else None,\n 'fb_status_type': post_data['status_type'] if post_data.has_key('status_type') else None,\n 'fb_message': message\n }\n \n # always insert insights data\n if is_insights(page_id, config):\n \n log.info( \"INSIGHTS\\tAdding data from %s re: %s\" % (page_id, article_slug) )\n\n # fetch data\n insights_value = get_insights_data(api, page_id, post_id)\n\n # create datasource name\n data_source = \"facebook_insights_%s\" % page_id \n \n # upsert url\n upsert_url(article_url, article_slug, data_source, config)\n\n # insert id\n db.sadd('facebook_post_ids', post_id)\n\n # format time bucket\n current_time_bucket = gen_time_bucket(config)\n insights_value['time_bucket'] = current_time_bucket\n post_value.pop('time_bucket', None)\n \n value = json.dumps({\n data_source : dict(post_value.items() + insights_value.items())\n })\n\n # upload data to redis\n db.zadd(article_slug, current_time_bucket, value) \n \n # only insert new posts\n if not db.sismember('facebook_post_ids', post_id):\n \n log.info( \"FACEBOOK\\tNew post %s\\t%s\" % (post_id, article_url) )\n \n # insert id\n db.sadd('facebook_post_ids', post_id) \n \n # upsert url\n data_source = \"facebook_%s\" % page_id\n upsert_url(article_url, article_slug, data_source, config)\n\n value = json.dumps( {data_source : post_value} )\n\n\n # upload data to redis\n db.zadd(article_slug, time_bucket, value)","def proto_post(self, proto_category):\n return baker.make(\n Post,\n content=(\n \"Aliquip excepteur qui mollit labore nulla et culpa \"\n \"minim et commodo reprehenderit consequat sint.\"\n ),\n categories=proto_category,\n make_m2m=True,\n _create_files=True,\n )","def post_to_writeas(token, publication, post_title, post_date, post_content):\n writeas_post_url = 'https://write.as/api/collections/{}/posts'.format(\n publication)\n writeas_auth_header = {\n 'Authorization': token,\n 'Content-Type': 'application/json'\n }\n payload = {\n 'title': post_title,\n 'created': post_date,\n 'body': post_content\n }\n r = requests.post(\n writeas_post_url,\n headers=writeas_auth_header,\n json=payload)\n\n if r.status_code == 201:\n result = r.json()\n post_id = result['data']['id']\n print('Post {} successfully created!'.format(post_id))\n else:\n print(\"Publishing post FAILED. Response: {}\".format(r.text))\n writeas_logout(writeas_auth_token)\n sys.exit(1)","async def create_post(self, community: Community, post_id) -> w_Post:\n post_url = self._api_communities_url + str(community.id) + '/posts/' + str(post_id)\n async with self.web_session.get(post_url, headers=self._headers) as resp:\n if self.check_status(resp.status, post_url):\n data = await resp.json()\n return (create_post_objects([data], community, new=True))[0]","def topic_posting_new(self, request):\n col = ObjectPostings(self.cdb_object_id, {})\n return posting_new(col, request)","def publish(self):\n self.published = True\n self.save()# pylint: disable=no-member","def post(self):\n subject = self.request.get('subject')\n post_content = self.request.get('post_content')\n submit = self.request.get('submit')\n cancel = self.request.get('cancel')\n user = self.get_active_user()\n created_by = int(user.key().id())\n post_id = self.request.get('post_id')\n\n if not user:\n self.redirect('/login')\n if post_id:\n post = Posts.get_by_id(int(post_id))\n else:\n post = None\n\n if cancel == \"cancel\":\n self.redirect('/%s' % str(post.key().id()))\n return\n if (post and post.submitter_id == user.key().id()) or not post:\n if submit == \"submit\" and subject and post_content:\n if post:\n post.subject = subject\n post.content = post_content\n post.put()\n else:\n post = Posts(subject=subject,\n content=post_content,\n submitter_id=created_by)\n post.put()\n self.redirect('/%s' % str(post.key().id()))\n else:\n self.render_newpage(user=user,\n subject=subject,\n post_content=post_content,\n error=\"\"\"Please provide both a subject and a\n post!\"\"\")\n else:\n self.redirect('/login')","def create_post_with_comment(pauthor, cauthor, visibility, ptext, ctext):\n\n post = Post.objects.create(content = ptext, author = pauthor, visibility=visibility)\n comment = Comment.objects.create(comment = ctext, post = post, author = cauthor)\n return (post, comment)","def test_auto_set_published_on_create(self):\n title = ('Transportation Challenges Limit Education Choices for '\n 'Denver Parents')\n summary = \"\"\"\n Many families in the Denver metro area use public\n transportation instead of a school bus because for them, a\n quality education is worth hours of daily commuting. Colorado's\n school choice program is meant to foster educational equity,\n but the families who benefit most are those who have time and\n money to travel. Low-income families are often left in a lurch.\n \"\"\"\n byline = \"Mile High Connects\"\n story = create_story(title=title, summary=summary, byline=byline,\n status='published')\n self.assertNowish(story.published)","def test_auto_set_published_on_create(self):\n title = ('Transportation Challenges Limit Education Choices for '\n 'Denver Parents')\n summary = \"\"\"\n Many families in the Denver metro area use public\n transportation instead of a school bus because for them, a\n quality education is worth hours of daily commuting. Colorado's\n school choice program is meant to foster educational equity,\n but the families who benefit most are those who have time and\n money to travel. Low-income families are often left in a lurch.\n \"\"\"\n byline = \"Mile High Connects\"\n story = create_story(title=title, summary=summary, byline=byline,\n status='published')\n self.assertNowish(story.published)","def proto_post(self, proto_category):\n return baker.make(\n Post,\n content=\"Aute non ex nostrud amet ipsum.\",\n categories=proto_category,\n make_m2m=True,\n _create_files=True,\n )","def proto_post(self, proto_category):\n return baker.make(\n Post,\n content=\"Aute non ex nostrud amet ipsum.\",\n categories=proto_category,\n make_m2m=True,\n _create_files=True,\n )","def test_post_creation_success(self):\n url = reverse('post-list', args=[self.topic1.url_name])\n payload = {\n 'author': self.user1.id,\n 'title': 'Creating a post',\n 'content': 'Rich content 4',\n }\n self.client.credentials(\n HTTP_AUTHORIZATION = 'Token ' + self.user1.auth_token.key\n )\n response = self.client.post(url, payload)\n\n self.assertEqual(response.status_code, status.HTTP_201_CREATED)\n new_post = Post.objects.filter(\n author=self.user1,\n title=payload.get('title'),\n content=payload.get('content'),\n topic=self.topic1\n )\n self.assertTrue(new_post.exists())","def write_post(id, args):\n graph = facebook.GraphAPI(id) \n graph.put_object(parent_object='me', connection_name = 'feed', **args)","def handler(event, _context):\n model = PostModel()\n post_id = model.create(**json.loads(event['body']))\n return dump_result({'post_id': post_id}, status_code=201)","def test_show_post_view_with_a_published_post(self):\n category = create_category('Category 1')\n author = create_author('Author 1')\n published_post = create_post(category=category, author=author, name='Published Post',\n content='Lorem ipsum dolor sit amet, consectetur adipiscing elit. Praesent sollicitudin.',\n status='Published')\n url = reverse('blog.post', args=(published_post.id,))\n create_question(question_text=\"Past question.\", days=-30)\n response = self.client.get(url)\n self.assertContains(response, published_post.content)","def post(self):\n\n subject = self.request.get('subject')\n content = self.request.get('content')\n\n have_errors = False\n\n if not subject:\n error_subject = \"Please write down the subject\"\n have_errors = True\n if not content:\n error_content = \"Content is required\"\n have_errors = True\n\n if have_errors:\n self.render(\"newpost.html\",\n subject=subject,\n content=content,\n error_subject=error_subject,\n error_content=error_content,\n user=self.user)\n else:\n post = Post(parent=blog_key(),\n subject=subject,\n content=content,\n user=self.user)\n post.put()\n self.redirect('/blog/%s' % str(post.key().id()))","def create(self, request, *args, **kwargs):\n serializer = self.get_serializer(data=request.data)\n\n if serializer.is_valid():\n data = serializer.data\n Article.objects.create_article(url=data['url'],\n authors=data['authors'],\n publish_time=data['publish_time'],\n title_image=data['images'])\n return Response({'message': 'News story added.'})\n else:\n return Response({'error': serializer.errors})","def new_post(mkp_form, request):\n newpost = Posts()\n newpost.init()\n newpost.authorid = int(request.user.id)\n newpost.title = mkp_form.cleaned_data['title']\n newpost.name = mkp_form.cleaned_data['short_title'] # 缩略名\n newpost.cover = mkp_form.cleaned_data['cover_url']\n newpost.introduction = mkp_form.cleaned_data['introduction']\n newpost.content = js_resize_img(mkp_form.cleaned_data['content'])\n newpost.status = Status.objects.get(id=2) # id为2是已发布的文章，默认为已发布，后面再改\n tagids = mkp_form.cleaned_data['tags']\n if len(tagids) != 0:\n for tagid in tagids:\n tagid = int(tagid)\n tag = Tags.objects.get(id=tagid)\n newpost.tags.add(tag)\n threadtypeid = mkp_form.cleaned_data['threadtypeid']\n newpost.threadtypeid = ThreadTypes.objects.get(id=threadtypeid)\n if mkp_form.cleaned_data['commentnotshow'] != '':\n newpost.comment_status = False\n else:\n newpost.comment_status = True\n return newpost","def post(self):\n title = self.request.get(\"title\")\n body = self.request.get(\"body\")\n\n if title and body:\n\n # create a new Post object and store it in the database\n post = Post(\n title=title,\n body=body\n )\n post.put()\n\n # get the id of the new post, so we can render the post's page (via the permalink)\n id = post.key().id()\n self.redirect(\"/blog/%s\" % id)\n else:\n error = \"we need both a title and a body!\"\n #self.render_form(title, body, error)\n self.render(\"newpost.html\", title, body, error)","def newPost(self, useRawHTML):\n print\n content, publish = self._fillPost(useRawHTML)\n\n # Upload to server\n try :\n postid = self.server.metaWeblog.newPost(\n self.blogid, self.username, self.password,\n content, publish\n )\n except xmlrpclib.Fault as fault:\n display_XMLRPC_errors(\"post the new entry\", fault)\n import pdb\n pdb.set_trace()\n else :\n self._setCategorie(postid)\n print \"New post created with ID =\", postid","def test_create_post(self):\n with self.client:\n result = self.client.post('/users/spike-test/posts', data={\n \"post-title\": \"autotest title\",\n \"post-content\": \"autotest content for blog post testing\"\n }, follow_redirects=True)\n self.assertEqual(result.status_code, 200)\n self.assertIn(b'autotest title', result.data)","def test_home_view_with_a_published_post(self):\n category = create_category('Category 1')\n author = create_author('Author 1')\n create_post(category=category, author=author, name='Published Post', content='Lorem ipsum dolor sit amet, consectetur adipiscing elit. Praesent sollicitudin.', status='Published')\n create_question(question_text=\"Past question.\", days=-30)\n response = self.client.get(reverse('blog.home'))\n self.assertQuerysetEqual(\n response.context['posts'],\n ['<Post: Published Post>']\n )","def post(self):\n subject = self.request.get('subject')\n content = self.request.get('content')\n\n # if user enter good subject and content, redirect them to new post page\n if subject and content:\n p = Post(parent = blog_key(), subject = subject, content = content)\n p.put() # store the post element into database\n self.redirect('/blog/%s' % str(p.key().id()))\n # otherwise, render an error page \n else:\n error = \"subject and content, please!\"\n self.render(\"newpost.html\", subject=subject, content=content, error=error)","def create(cls, headline, text, blog):\n post = cls()\n post.headline = headline\n post.text = text\n post.blog = blog\n post.posted_date = timezone.now()\n try:\n post.save()\n return post\n except(ValueError, IntegrityError, OperationalError):\n return None","def new(request):\n\n form = PostForm(request.POST or None)\n\n if request.method == \"POST\":\n if form.is_valid():\n instance = form.save(commit=False)\n instance.user = request.user\n instance.save()\n return HttpResponseRedirect(reverse('posts:toggle_publish', args=(instance.id,)))\n\n context = {'form':form}\n\n return render(request, 'posts/new.html', context)","def make_post(request):\n if request.user.is_authenticated() and request.POST:\n member = Member.objects.get(user=request.user)\n thread_id = request.POST.get('thread_id', -1)\n content = request.POST.get('content', -1)\n if thread_id != -1 and content != -1 and member:\n post = Post()\n post.author = member\n post.thread = Thread.objects.get(pk=thread_id)\n post.content = content\n post.save()\n return HttpResponse(200)\n else:\n return server_error(request)\n else:\n return server_error(request)","def post(self):\n\n title = self.request.get(\"title\")\n blogPost = self.request.get(\"blogPost\")\n author = self.request.cookies.get('name')\n\n if title and blogPost:\n\n bp = Blogposts(parent=blog_key(), title=title,\n blogPost=blogPost, author=check_secure_val(author))\n\n bp.put()\n\n self.redirect('/%s' % str(bp.key.integer_id()))\n else:\n error = \"Please submit both a title and a blogpost!\"\n self.render(\"newpost.html\", title=title,\n blogPost=blogPost, error=error)","def post(self):\n data = request.json\n return create_new_blog(data=data)","def create_activity(self, created_user, source, action,\n privacy=Privacy.PRIVATE, **kwargs):\n Activity = get_activity_model()\n return Activity.objects.create(\n about=self,\n action=action,\n created_user=created_user,\n source=source,\n privacy=privacy,\n **kwargs\n )","def _create(self, obj, preview=None, include_link=False):\n # TODO: validation, error handling\n assert preview in (False, True)\n type = obj.get('objectType')\n verb = obj.get('verb')\n base_id, base_url = self.base_object(obj, verb=verb)\n if base_id and not base_url:\n base_url = self.object_url(base_id)\n\n content = self._content_for_create(obj)\n if not content:\n if type == 'activity':\n content = verb\n else:\n return source.creation_result(\n abort=False, # keep looking for things to post\n error_plain='No content text found.',\n error_html='No content text found.')\n\n url = obj.get('url')\n if include_link and url:\n content += '\\n\\n(%s)' % url\n preview_content = util.linkify(content)\n msg_data = {'message': content.encode('utf-8')}\n if appengine_config.DEBUG:\n msg_data['privacy'] = json.dumps({'value': 'SELF'})\n msg_data = urllib.urlencode(msg_data)\n\n if type == 'comment':\n if not base_url:\n return source.creation_result(\n abort=True,\n error_plain='Could not find a Facebook status to reply to.',\n error_html='Could not find a Facebook status to <a href=\"http://indiewebcamp.com/comment\">reply to</a>. '\n 'Check that your post has an <a href=\"http://indiewebcamp.com/comment\">in-reply-to</a> '\n 'link a Facebook URL or to an original post that publishes a '\n '<a href=\"http://indiewebcamp.com/rel-syndication\">rel-syndication</a> link to Facebook.')\n\n if preview:\n return source.creation_result(\n 'will <span class=\"verb\">comment</span> <em>%s</em> on '\n '<a href=\"%s\">this post</a>:\\n%s' %\n (preview_content, base_url, EMBED_POST % base_url))\n else:\n resp = json.loads(self.urlopen(API_COMMENTS_URL % base_id,\n data=msg_data).read())\n resp.update({'url': self.comment_url(base_id, resp['id']),\n 'type': 'comment'})\n\n elif type == 'activity' and verb == 'like':\n if not base_url:\n return source.creation_result(\n abort=True,\n error_plain='Could not find a Facebook status to like.',\n error_html='Could not find a Facebook status to <a href=\"http://indiewebcamp.com/favorite\">like</a>. '\n 'Check that your post has an <a href=\"http://indiewebcamp.com/favorite\">like-of</a> '\n 'link a Facebook URL or to an original post that publishes a '\n '<a href=\"http://indiewebcamp.com/rel-syndication\">rel-syndication</a> link to Facebook.')\n\n if preview:\n return source.creation_result(\n 'will <span class=\"verb\">like</span> <a href=\"%s\">this post</a>:\\n%s' %\n (base_url, EMBED_POST % base_url))\n else:\n resp = json.loads(self.urlopen(API_LIKES_URL % base_id, data='').read())\n assert resp == True, resp\n resp = {'type': 'like'}\n\n elif type == 'activity' and verb in RSVP_ENDPOINTS:\n if not base_url:\n return source.creation_result(\n abort=True,\n error_plain=\"This looks like an RSVP, but it's missing an \"\n \"in-reply-to link to the Facebook event.\",\n error_html=\"This looks like an <a href='http://indiewebcamp.com/rsvp'>RSVP</a>, \"\n \"but it's missing an <a href='http://indiewebcamp.com/comment'>in-reply-to</a> \"\n \"link to the Facebook event.\")\n\n # TODO: event invites\n if preview:\n assert verb.startswith('rsvp-')\n return source.creation_result(\n 'will <span class=\"verb\">RSVP %s</span> to '\n '<a href=\"%s\">this event</a>.<br />' % (verb[5:], base_url))\n else:\n resp = json.loads(self.urlopen(RSVP_ENDPOINTS[verb] % base_id, data='').read())\n assert resp == True, resp\n resp = {'type': 'rsvp'}\n\n elif type in ('note', 'article'):\n if preview:\n return source.creation_result(\n 'will <span class=\"verb\">post</span>:<br /><br />'\n '<em>%s</em><br />' % preview_content)\n else:\n resp = json.loads(self.urlopen(API_FEED_URL, data=msg_data).read())\n resp.update({'url': self.post_url(resp), 'type': 'post'})\n\n elif type == 'activity' and verb == 'share':\n return source.creation_result(\n abort=True,\n error_plain='Cannot publish shares on Facebook.',\n error_html='Cannot publish <a href=\"https://www.facebook.com/help/163779957017799\">shares</a> '\n 'on Facebook. This limitation is imposed by the '\n '<a href=\"https://developers.facebook.com/docs/graph-api/reference/v2.0/object/sharedposts/#publish\">Facebook Graph API</a>.')\n\n else:\n return source.creation_result(\n abort=False,\n error_plain='Cannot publish type=%s, verb=%s to Facebook' % (type, verb),\n error_html='Cannot publish type=%s, verb=%s to Facebook' % (type, verb))\n\n if 'url' not in resp:\n resp['url'] = base_url\n return source.creation_result(resp)","def post():\n\n form = forms.PostForm()\n if form.validate_on_submit():\n models.Post.create(title=form.title.data,\n date=form.date.data,\n time_spent=form.time_spent.data,\n details=form.details.data,\n remember=form.remember.data)\n return redirect(url_for('index'))\n return render_template('new.html', form=form)","def post(self, post_id=None):\n\n if post_id:\n abort(400)\n else:\n args = parsers.post_post_parser.parse_args(strict=True)\n\n new_post = Post(args['title'])\n new_post.text = args['text']\n # new_post.user = user\n\n if args['tags']:\n for item in args['tags']:\n tag = Tag.query.filter_by(name=item).first()\n # If the tag already exist, append.\n if tag:\n new_post.tags.append(tag)\n # If the tag not exist, create the new one.\n # Will be write into DB with session do.\n else:\n new_tag = Tag(item)\n new_post.tags.append(new_tag)\n db.session.add(new_post)\n db.session.commit()\n return (new_post.id, 201)","def post(self):\n current_user = self.authenticate_user()\n\n if not current_user:\n self.redirect(\"/login\")\n else:\n content = self.request.get(\"content\")\n title = self.request.get(\"subject\")\n\n if not content or not title:\n self.render_front(title, content, \"We need both a title and content\")\n else:\n post = Post(title=title, content=content, user=current_user.key)\n post.put()\n\n current_user.posts.append(post.key)\n current_user.put()\n\n self.redirect(\"/post/\" + str(post.key.id()))","def create_action(instance, verb, user):\n return instance.activities.create(action=verb, owner=user)","def create_activity(request: Request, activity_type: str, msg_context: dict, object_id: UUID, user: User):\n dbsession = Session.object_session(user)\n\n stream = Stream.get_or_create_user_stream(user)\n\n a = Activity()\n a.object_id = object_id\n a.activity_type = activity_type\n a.msg_context = msg_context\n\n stream.activities.append(a)\n dbsession.flush()\n\n return a","def post(activity, create, extracted, **kwargs):\n if activity.verb in [\n models.ReferralActivityVerb.ASSIGNED,\n models.ReferralActivityVerb.UNASSIGNED,\n ]:\n # pylint: disable=attribute-defined-outside-init\n activity.item_content_object = UserFactory()\n elif activity.verb in [\n models.ReferralActivityVerb.DRAFT_ANSWERED,\n models.ReferralActivityVerb.ANSWERED,\n ]:\n # pylint: disable=attribute-defined-outside-init\n activity.item_content_object = ReferralAnswerFactory(\n referral=activity.referral\n )\n elif activity.verb in [\n models.ReferralActivityVerb.VALIDATED,\n models.ReferralActivityVerb.VALIDATION_DENIED,\n ]:\n # pylint: disable=attribute-defined-outside-init\n activity.item_content_object = ReferralAnswerValidationResponseFactory()\n elif activity.verb == models.ReferralActivityVerb.VALIDATION_REQUESTED:\n # pylint: disable=attribute-defined-outside-init\n activity.item_content_object = ReferralAnswerValidationRequestFactory()\n elif activity.verb == models.ReferralActivityVerb.CREATED:\n pass\n else:\n raise Exception(f\"Incorrect activity verb {activity.verb}\")","def create_multiple_posts(author, num, ptext = TEXT, visibility = ACL_DEFAULT):\n posts = []\n\n for i in range(num):\n posts.append(Post.objects.create(content = ptext, author = author, visibility=visibility))\n\n return posts","def get_post(self):\n\t\tself.post = graph.get_object(POST_ID)","def addPost(postCreatedEvent):\n query = generatePostSQLQuery(postCreatedEvent)\n with conn.cursor() as cur:\n cur.execute(query)\n conn.commit()","def perform_create(self, serializer):\n lookup_url_kwarg = self.lookup_url_kwarg or self.lookup_field\n lookup = self.kwargs.get(lookup_url_kwarg, None)\n \n if lookup is not None:\n filter_kwargs = {self.lookup_field: lookup}\n video = get_object_or_404(Video, **filter_kwargs)\n clip = serializer.save(video=video, owner=self.request.user) \n \n # register this new activity\n activity_send(self.request.user, verb='add', object=clip, target=video)\n else:\n raise Http404","def insert_post(shard, **kwargs):\n # Create the posting and insert it.\n post_id = kwargs.pop('post_id', None)\n if not post_id:\n post_id = models.human_uuid()\n\n new_topic = kwargs.get('new_topic', None)\n\n kwargs['post_time'] = datetime.datetime.now()\n\n post_key = ndb.Key(models.Post._get_kind(), post_id)\n post = models.Post(\n key=post_key,\n **kwargs)\n\n @ndb.tasklet\n def txn():\n if (yield post_key.get_async(use_memcache=False, use_cache=False)):\n logging.warning('Post already exists for shard=%r, post_id=%r',\n shard, post_id)\n raise ndb.Rollback()\n\n yield post.put_async(use_memcache=False, use_cache=False)\n\n # Pull task that indicates the post to apply. This must encode the\n # new_topic data for this post so the apply_posts() function doesn't\n # need the models.Post entity in order to make progress.\n enqueue_post_task(shard, [post_id], new_topic=new_topic)\n\n # Notify all users of the post.\n futures = []\n futures.append(ndb.transaction_async(txn))\n futures.append(notify_posts(shard, [post]))\n\n # Set the dirty bit for this shard. This causes apply_posts to run a\n # second time if the Post transaction above completed while apply_posts\n # was already in flight.\n dirty_bit(shard, set=True)\n\n # Enqueue an apply task to sequence and notify the new post.\n futures.append(enqueue_apply_task(shard, post_id=post_id))\n\n # Wait on futures in case they raise errors.\n ndb.Future.wait_all(futures)\n\n return post_key","def post(title, drafts=False):\n # Parsing options\n if drafts:\n dest = _drafts_dest\n else:\n dest = _posts_dest\n\n # File name\n date = get_date()\n name = sanitize(title)\n fname = \"{}-{}{}\".format(date, name, _post_ext)\n\n # Front Matter\n front_matter = []\n front_matter.append('---')\n front_matter.append('layout: post')\n front_matter.append('title: {}'.format(title))\n front_matter.append('description: \"\"')\n front_matter.append('tags: \"\"')\n front_matter.append('comments: true')\n front_matter.append('permalink: \"\"')\n front_matter.append('sitemap:\\n lastmod: {}'.format(date))\n front_matter.append('---')\n\n # Create post file and write Front Matter\n print(\"\\nCreating new post '{}' in {}\\n\".format(fname, dest))\n try:\n f = open(dest + fname, 'w')\n except Exception as e:\n print(\"* [Error] occured: {}\\n\".format(e))\n else:\n f.write('\\n'.join(front_matter))\n f.close()\n print(\"* Done.\\n\")","def post(self, post):\n\n self._post = post","def post(self, post):\n\n self._post = post","def post(self):\n data = request.json\n create_entry(data)\n return None, 201","def create(self, request, *args, **kwargs):\n response = super(ReleaseViewSet, self).create(request, *args, **kwargs)\n if response.status_code == status.HTTP_201_CREATED:\n signals.release_post_update.send(sender=self.object.__class__,\n release=self.object,\n request=request)\n return response","def create_post(request):\n if request.method == 'POST':\n title = request.POST['title']\n content = request.POST['content']\n user_id = request.POST['author_id']\n category = request.POST['category']\n\n slug = \"-\".join(list(map(lambda word: word.lower(), title.split())))\n author = User.objects.get(id=int(user_id))\n\n # save info in models\n post = Post()\n post.author = author\n post.category = category\n post.title = title\n post.content = content\n post.slug = slug\n post.save()\n return redirect('post')\n\n return render(request, 'posts/create_post.html')","def post_create(request):\n\tform = PostForm(request.POST or None, request.FILES or None)\n\tif request.POST:\n\t\tif form.is_valid():\n\t\t\tinstance = form.save(commit=False)\n\t\t\tinstance.user = request.user\n\t\t\tinstance.save()\n\t\t\tmessages.success(request, \"Post created!\")\n\t\t\treturn HttpResponseRedirect(instance.get_absolute_url())\n\t\telse:\n\t\t\tmessages.error(request, \"Sorry! Something went wrong.\", extra_tags=\"\")\n\tcontext = {\n\t\t'title': \"Create Post\",\n\t\t'form' : form,\n\t}\n\treturn render(request, 'post/create.html', context)","def PublishIt(name, path, comments, task=os.getenv('TASK'), status=\"WORK IN PROGRESS\"):\n\n db = get_connection()\n\n PubCollections = db['submissions']\n\n # creation of the dailies submission entry\n publishDict = dict()\n publishDict['date'] = now\n publishDict['type'] = \"publish\"\n publishDict['user_name'] = main_user\n publishDict['task'] = task\n publishDict['status'] = status\n publishDict['asset'] = name\n publishDict['path'] = path\n publishDict['comment'] = comments\n PubCollections.save(publishDict)\n notifications.push_notifications({\"name\": main_user, \"email\": os.getenv('USER_EMAIL')}, users_list, \"publish\", shot, now)","def save(self, *args, **kwargs):\n if not self.id:\n self.created = timezone.now()\n return super(UserActivity, self).save(*args, **kwargs)","def generate_presigned_post(Bucket=None, Key=None, Fields=None, Conditions=None, ExpiresIn=None):\n pass","def create_post(bid):\n form = PostForm(request.form)\n if request.method == 'POST':\n if form.validate():\n DB.session.add(\n Post(\n bid,\n current_user.uid,\n form.name.data,\n form.desc.data))\n DB.session.commit()\n flash('Post ({}) successfully created!'.format(form.name.data))\n else:\n flash(constants.DEFAULT_SUBMISSION_ERR)\n return redirect(request.referrer)","def test_create_post():\n post = Post(\"Test\", \"Test Content\")\n\n assert post.title == \"Test\"\n assert post.content == \"Test Content\"","def publish():\n pass","def save_post(self, post):\n return self.collection.insert_one(post.serialize())","def should_publish(self, create, extracted, **kwargs):\n if create and extracted:\n return True\n return False","def community_post_create_view(request):\n task = \"Create New\"\n form = AddEditPostForm() # An unbound form\n\n if request.method == 'POST': # If the form has been submitted...\n form = AddEditPostForm(request.POST, request.FILES) # A form bound to the POST data\n if form.is_valid(): # All validation rules pass\n post = form.save(commit=False) # Create a new object from the form, but don't save it to the database\n post.author = request.user # Set the author to the current user\n post.save() # Save the object to the database\n slug_str = \"%s %s\" % (post.title, post.date_posted) # Create a slug from the title and date\n post.slug = slugify(slug_str) # Create the slug\n post.save() # Save the object to the database\n return redirect('community-home') # Redirect to the home page\n\n context = { # Pass the variables to the template\n 'task': task,\n 'form': form,\n }\n return render(request,\n 'pages/patient-community/community-create-update-post.html',\n context) # render the patient community create post page","def announce_board_posting_created(event: BoardPostingCreated) -> None:\n posting_creator_screen_name = get_screen_name_or_fallback(\n event.posting_creator_screen_name\n )\n board_label = _get_board_label(event.topic_id)\n\n if event.topic_muted:\n return\n\n text = (\n f'{posting_creator_screen_name} hat im {board_label} '\n f'auf das Thema \"{event.topic_title}\" geantwortet: {event.url}'\n )\n\n send_message(CHANNEL_ORGA_LOG, text)\n send_message(CHANNEL_PUBLIC, text)","def publish_release(ctx):\n rel = _get_release()\n rel.update_release(rel.title, rel.raw_data[\"body\"], draft=False)","def create_post(request):\n\n # modified from: http://django-angular.readthedocs.org/en/latest/angular-model-form.html\n\n # get data\n in_data = getRequestData(request)\n\n try:\n # save in database\n # note that in_data.mytitle throws an error while in_data.get('mytitle') works smoothly\n post = Thread(pub_date = datetime.datetime.now(pytz.timezone('US/Eastern')), username = in_data.get('myusername'), title = in_data.get('mytitle'), description = in_data.get('mydescription'))\n post.save()\n except:\n return HttpResponseBadRequest('Error saving to database!')\n\n return JsonResponse(in_data)","def toggle_publish(request,id):\n \n instance = get_object_or_404(Post, id=id)\n\n if request.method==\"POST\":\n instance.published = not instance.published\n instance.save()\n\n t, created = LastUpdate.objects.get_or_create(id=1)\n\n t.updated = instance.updated\n t.save()\n\n cache.clear()\n return redirect('posts:home')\n\n context = {'post':instance}\n\n return render(request, 'posts/publish.html', context)","def create(self, *args, **kwargs):\n pass","def post(self, user):\n subject = self.request.get(\"subject\")\n content = self.request.get(\"content\")\n\n if subject and content:\n post = Post(subject=subject,\n content=content,\n author=self.user)\n post.put()\n return self.redirect(\"/%s\" % post.key().id())\n else:\n\n error = \"subject and content, please!\"\n return self.render(\"newpost.html\",\n subject=subject,\n content=content,\n error=error)","def create_url_post(post_id, config):\n text = \"https://graph.facebook.com/\"\n text += config[\"graph_api_version\"]\n text += \"/\" + post_id + \"?fields=\"\n text += \"id, message, admin_creator, backdated_time,\"\n text += \"caption,coordinates,created_time,description,\"\n text += \"feed_targeting,from,event,icon,is_popular,link,\"\n text += \"message_tags,name,object_id,type,shares,\"\n text += \"story,source,properties,place,target,targeting,\"\n text += \"status_type,comments.fields(comment_count,from,id,like_count,\"\n text += \"permalink_url,created_time,message,comments.fields(comment_count,\"\n text += \"from,id,like_count,created_time,message,\"\n text += \"reactions.fields(id,name,type,username,profile_type)),\"\n text += \"reactions.fields(id,name,type,username,profile_type)),\"\n text += \"reactions.fields(id,name,type,username,profile_type), sharedposts\"\n text += \"&access_token=\" + config[\"access_token\"]\n\n return text","def create_post(self: User, content: str, is_public: bool, circles: List[Circle], reshareable: bool,\n reshared_from: Optional[Post], media_list: List[Media], mentioned_users: List[User],\n is_update_avatar: bool) \\\n -> Union[Post, bool]:\n if not content and not media_list:\n # a post has to have either content or media\n return False\n\n new_post = Post()\n new_post.eid = make_uuid()\n new_post.author = self.id\n if content:\n new_post.content = bleach.clean(content)\n new_post.is_public = is_public\n new_post.circles = circles\n new_post.media_list = media_list\n new_post.is_update_avatar = is_update_avatar\n\n if reshared_from and not reshareable:\n # if resharing from a post, this post must also be reshareable, otherwise it's logically wrong\n return False\n\n if reshared_from:\n if media_list:\n # when resharing, only allow content (text), e.g. no media\n return False\n\n if reshared_from.reshared_from:\n # if reshared_from itself is a reshared post, reshare reshared_from's original post\n # reshared_from.reshared_from is LazyReference so need to retrieve the full post\n reshared_from = get_in_post_cache(reshared_from.reshared_from.id)\n\n # same explanation for context_home_or_profile=False\n if not sees_post(self, reshared_from, context_home_or_profile=False):\n return False\n\n if not reshared_from.reshareable:\n return False\n\n new_post.reshared_from = reshared_from\n\n new_post.reshareable = reshareable\n new_post.save()\n\n if reshared_from:\n create_notification(\n self,\n notifying_href=new_post.make_href(),\n notifying_summary=new_post.content,\n notifying_action=NotifyingAction.Reshare,\n notified_href=reshared_from.make_href(),\n notified_summary=reshared_from.content,\n owner=reshared_from.author\n )\n # only cache reshared post\n set_in_post_cache(reshared_from)\n\n mention(\n self,\n notified_href=new_post.make_href(),\n notified_summary=new_post.content,\n mentioned_users=mentioned_users\n )\n\n return new_post","def post(self):\n post_id = self.request.get('post_id')\n post = Post.get_by_id(int(post_id), parent=blog_key())\n content = self.request.get('comment')\n\n if content:\n comment = Comment(parent=comment_key(),\n content=content,\n user=self.user,\n post=post)\n comment.put()\n\n time.sleep(0.1)\n self.redirect('/blog/%s' % str(post.key().id()))","def post(self, req):\n error_messages = []\n success_message = ''\n\n # Creamos owner y se lo pasamos al form con un objeto pre-establecido\n post_with_owner = Post()\n post_with_owner.owner = req.user\n post_with_owner.blog = Blog.objects.filter(owner=req.user)[0]\n\n form = PostCreateForm(req.POST, instance=post_with_owner)\n if form.is_valid():\n\n new_post = form.save()\n form = PostCreateForm()\n success_message = u'Post guardado con éxito! '\n success_message += u'<a href=\"{0}\">'.format(reverse('post_detail', args=[req.user.username, new_post.pk]))\n success_message += u'(ver post)</a>'\n else:\n error_messages.append(u'Formulario incompleto.')\n\n context = {\n 'form': form,\n 'success_message': success_message\n }\n return render(req, 'posts/new_post.html', context)","def for_publishing_only(self):\n self.token['type'] = 'publish'\n\n return self","def write_page():\n #1480631468\n page_id = request.args['page_id']\n post_time = request.args['time']\n #bday=2016-11-10\n args={}\n if post_time is not None:\n pattern = '%Y-%m-%d'\n epoch_time=int(time.mktime(time.strptime(post_time, pattern)))\n print_me(epoch_time)\n \n args['scheduled_publish_time']=epoch_time\n\n args['message'] = request.args['text']\n args['published']= request.args['p_type'] == 'Published'\n\n \"\"\"\n demo purposes\"\n \"\"\"\n #args['scheduled_publish_time']=\"1480631468\"\n #args['published']=\"0\"\n write_post(page_id, args)\n return render_template('write_done.html')","def posts_create(request):\n if request.method == 'POST':\n form = PostForm(request.POST, request.FILES)\n if form.is_valid():\n form.save()\n return redirect('feed')\n\n else:\n form = PostForm()\n\n return render(\n request=request,\n template_name='posts/new.html',\n context={\n 'form': form,\n 'user': request.user,\n 'perfil': request.user.perfil\n }\n )","def post(self):\n user_id = get_jwt_identity()\n user = find_user(user_id)\n args = post_parser.parse_args()\n\n # check circles\n circles = []\n for circle_id in args['circle_ids']:\n found_circle = find_circle(user, circle_id)\n if not found_circle:\n return {'msg': f'Circle {circle_id} is not found'}, 404\n circles.append(found_circle)\n\n # check reshare\n reshared_from = args['reshared_from']\n reshared_from_post = None\n if reshared_from:\n reshared_from_post = dangerously_get_post(reshared_from)\n if not reshared_from_post:\n return {\"msg\": f\"Post {reshared_from} is not found\"}, 404\n\n # check media\n media_object_names = args['media_object_names']\n if reshared_from and media_object_names:\n return {'msg': \"Reshared post is not allowed to have media\"}, 400\n\n post = create_post(\n user,\n content=args['content'],\n is_public=args['is_public'],\n circles=circles,\n reshareable=args['reshareable'],\n reshared_from=reshared_from_post,\n media_list=check_media_object_names(media_object_names, MaxPostMediaCount),\n mentioned_users=check_mentioned_user_ids(args['mentioned_user_ids']),\n is_update_avatar=False\n )\n if not post:\n return {\"msg\": f\"Not allowed to reshare post {reshared_from}\"}, 403\n return post, 201","def create_link_page_post(self, page_id, link=None, message=None, picture=None,\n thumbnail=None, name=None, caption=None,\n description=None, published=None, call_to_action=None, batch=False):\n page_access_token = self.get_page_access_token(page_id)\n if 'error' in page_access_token:\n return page_access_token\n if 'access_token' not in page_access_token:\n raise AdsAPIError('Could not get page access token. (Do you have manage pages permission?)')\n path = '%s/feed' % page_id\n args = {\n 'access_token': page_access_token['access_token'],\n }\n files = {}\n if link is not None:\n args['link'] = link\n if message is not None:\n args['message'] = message\n if picture is not None:\n args['picture'] = picture\n if thumbnail is not None:\n files['thumbnail'] = thumbnail\n if published is not None:\n args['published'] = published\n if name is not None:\n args['name'] = name\n if caption is not None:\n args['caption'] = caption\n if description is not None:\n args['description'] = description\n if call_to_action is not None:\n args['call_to_action'] = json.dumps(call_to_action)\n return self.make_request(path, 'POST', args, files, batch=batch)","def post(request):\n if request.method == \"POST\":\n post = Post()\n post.content = request.POST['content']\n post.author = request.user\n post.save()\n return HttpResponseRedirect(reverse(\"index\"))","def create_post():\r\n\r\n # Check for and reject empty username or whinge\r\n if not request.values.get(\"username\") or not request.values.get(\"whinge\"):\r\n print(\"Ignoring request to with empty username or whinge\")\r\n else:\r\n # Form data ok; add to DB\r\n con = get_db()\r\n con.execute(\"INSERT INTO posts (submitter,content,ts) VALUES (?,?,?);\",\r\n (\r\n request.values.get(\"username\"), # form field username -> DB column submitter\r\n request.values.get(\"whinge\"), # form field whinge -> DB column content\r\n time.time()\r\n )\r\n )\r\n con.commit()\r\n con.close()\r\n \r\n # TODO: Handle possibility of failed INSERT\r\n\r\n # Send them back to the main page\r\n return redirect(url_for(\"display_top\"))","def post(self):\n post_title = self.request.get(\"post_title\")\n post_content = self.request.get(\"post_content\")\n param_list = dict(post_title=post_title, post_content=post_content)\n any_error = False\n\n if not post_title:\n param_list['title_error'] = \"Title is missing!\"\n any_error = True\n if not post_content:\n param_list['content_error'] = \"Content is missing!\"\n any_error = True\n\n if any_error:\n self.render(\"blog/addpost.html\", **param_list)\n else:\n p = Post.add_post(post_title, post_content, self.user)\n self.redirect('/blog/%s' % str(p.key().id()))","def test_publish(self):\n spiderman = User.objects.create(username=\"spiderman\", password=\"mary_jane\")\n BlogEntry.objects.create(title=\"Spam Blog\", text=\"This is BlogSpam\", created_by=spiderman)","def add_post(self, post: Post) -> None:\n self.post_process.append(post)","def create_task(author, title, text, **kwargs):\n mc = MathContent(text=text)\n mc.save()\n task = Task(author=author, name=title, content=mc, **kwargs)\n task.save()\n return task","def post(self):\n post_id = int(self.request.get('post_id'))\n post = Posts.get_by_id(post_id)\n comment = self.request.get('comment')\n submitter_id = self.get_active_user().key().id()\n\n if submitter_id:\n comment = Comments(post_id=post_id, content=comment,\n submitter_id=submitter_id)\n comment.put()\n self.redirect('/%s' % str(post.key().id()))\n else:\n self.error(403)","def addPost(self,text,id,url,date):\n self.topComments.append(Post(text,id,url,date))\n return None","def create_post(session: Session) -> Post:\n try:\n admin_user = session.query(User).filter(User.username == \"toddthebod\").first()\n post = Post(\n author_id=admin_user.id,\n slug=\"fake-post-slug\",\n title=\"Fake Post Title\",\n summary=\"A fake post to have some fake comments.\",\n feature_image=\"https://hackersandslackers-cdn.storage.googleapis.com/2021/01/logo-smaller@2x.png\",\n body=\"Cheese slices monterey jack cauliflower cheese dolcelatte cheese and wine fromage frais rubber cheese gouda. Rubber cheese cheese and wine cheeseburger cheesy grin paneer paneer taleggio caerphilly. Edam mozzarella.\",\n )\n session.add(admin_user) # Add the user\n session.commit() # Commit the change\n LOGGER.success(f\"Created post {post} published by user {admin_user}\")\n return post\n except IntegrityError as e:\n LOGGER.error(e.orig)","def post(self, request, *args, **kwargs):\n return self.create(request, *args, **kwargs)","def post(self, request, *args, **kwargs):\n return self.create(request, *args, **kwargs)","def post(self, request, *args, **kwargs):\n return self.create(request, *args, **kwargs)"],"string":"[\n \"def post_to_activity(self, post):\\n id = None\\n if post.get('id'):\\n # strip USERID_ prefix if it's there\\n post['id'] = post['id'].split('_', 1)[-1]\\n id = post['id']\\n\\n obj = self.post_to_object(post)\\n activity = {\\n 'verb': VERBS.get(post.get('type', obj.get('objectType')), 'post'),\\n 'published': obj.get('published'),\\n 'updated': obj.get('updated'),\\n 'id': self.tag_uri(id) if id else None,\\n 'url': self.post_url(post),\\n 'actor': obj.get('author'),\\n 'object': obj,\\n }\\n\\n application = post.get('application')\\n if application:\\n activity['generator'] = {\\n 'displayName': application.get('name'),\\n 'id': self.tag_uri(application.get('id')),\\n }\\n return self.postprocess_activity(activity)\",\n \"def newPost(self, postLink, zserverBlogEntry): #$NON-NLS-1$\\r\\n atomEntry = self.createNewBlogEntry()\\r\\n self._populateAtomEntry(atomEntry, zserverBlogEntry)\\r\\n # publish entry\\r\\n atomRespEntry = self.createAtomEntry(postLink, atomEntry)\\r\\n return atomRespEntry\",\n \"def create_a_post():\\n subj = create_subject()\\n post = Post.create(subject=subj, title=\\\"A great title\\\", body=\\\"Just a great day!\\\")\\n post.save()\\n return post\",\n \"def post(self):\\n data = request.json\\n\\n now = datetime.utcnow().isoformat()\\n data['created_time'] = now\\n\\n # Our type is already registered within the DB, so generate a\\n # model object that looks like what we'll be interacting with\\n try:\\n activity = Activity(data)\\n except KeyError:\\n raise BadRequest(\\\"payload validation failed: {}\\\".format(data))\\n\\n activity.save()\\n log.debug(\\\"Wrote activity: \\\" + str(activity._to_dict()))\\n return activity._to_dict(), 201\",\n \"def create_post(category, author, name, content, status):\\n return Post.objects.create(category=category, author=author, name=name, content=content, status=status)\",\n \"def postCreate(post):\\n post_list = list()\\n comments = commentList(post)\\n comment_url = \\\"{}/api/posts/{}/comments\\\".format(DOMAIN, post.id)\\n visible_to = list()\\n visible = post.visibleTo.all()\\n if visible:\\n for author in visible:\\n auth = \\\"{}/api/author/{}\\\".format(DOMAIN, author.id)\\n visible_to.append(auth)\\n\\n # visible_to = list(post.visibleTo)\\n post_dict = {'author': addAuthor(post.author), 'title': post.title, 'description': post.description,\\n 'contentType': post.contentType, 'content': post.content, 'published': post.published,\\n 'visibility': post.visibility, 'visibleTo': visible_to, 'unlisted': post.unlisted, 'id': post.id,\\n 'comments': comments[:5], 'next': comment_url, 'count': len(comments),\\n 'source': \\\"{}/api/posts/{}\\\".format(DOMAIN, post.id),\\n 'origin': \\\"{}/api/posts/{}\\\".format(DOMAIN, post.id)}\\n post_list.append(post_dict)\\n return post_list\",\n \"def create_post():\\n\\n #Get prompt id\\n prompt_id = request.form.get('prompt_id')\\n\\n # Get post text\\n post_text = request.form.get('user_post')\\n\\n # Create post timestamp\\n created_at = datetime.now()\\n user_facing_date = created_at.strftime(\\\"%B %d, %Y\\\")\\n\\n # Save post and related data to database\\n post = crud.create_post(session['user_id'], prompt_id, post_text, session['lat'], session['lng'], session['user_facing_location'], created_at)\\n\\n return render_template('post_data.html', post=post, user_facing_date=user_facing_date)\",\n \"def post_publish(request, pk):\\n post = get_object_or_404(Post, pk=pk)\\n post.publish()\\n return redirect('post_detail', pk=pk)\",\n \"def post_activities():\\n pass\",\n \"def publish(self):\\n self.published_date = timezone.now()\\n self.save()\",\n \"def publish(self):\\n self.published_date = timezone.now()\\n self.save()\",\n \"def publish(self):\\n self.published_date = timezone.now\\n self.save()\",\n \"def remotePostCreate(host, post):\\n post = post.get('posts')[0]\\n author = remoteAddAuthor(post.get('author'))\\n title = post.get('title')\\n description = post.get('description')\\n contentType = post.get('contentType')\\n content = post.get('content')\\n published = utc.localize(datetime.strptime(post.get('published'), '%Y-%m-%dT%H:%M:%S.%fZ'))\\n visibility = post.get('visibility')\\n unlisted = post.get('unlisted')\\n id = post.get('id')\\n origin = post.get('origin')\\n count = post.get('count')\\n comments = remoteCommentList(post)\\n source = \\\"{}/api/posts/{}\\\".format(DOMAIN, post.get('id'))\\n\\n post_dict = {'author': author, 'title': title, 'description': description,\\n 'contentType': contentType, 'content': content, 'published': published,\\n 'visibility': visibility, 'unlisted': unlisted, 'id': id,\\n 'comments': comments, 'origin': origin, 'count': count,\\n 'source': source}\\n return post_dict\",\n \"def new_post(self, content):\\n return self.proxy.wp.newPost(self.blog_id, self.username, self.password,\\n content)\",\n \"def on_created_action(sender, instance: models_actstream.Action, created, **kwargs):\\n if not created:\\n return\\n\\n # If content_type is Post or Profile\\n if (\\n instance.action_object_content_type\\n == ContentType.objects.get_for_model(dillo.models.posts.Post)\\n and instance.verb == 'posted'\\n ) or (\\n instance.action_object_content_type\\n == ContentType.objects.get_for_model(dillo.models.profiles.Profile)\\n and instance.verb == 'updated their reel'\\n ):\\n dillo.tasks.feeds.establish_time_proximity(instance)\\n\\n log.debug('Creating background fanout operation for action %i' % instance.id)\\n dillo.tasks.feeds.activity_fanout_to_feeds(instance.id)\",\n \"def insert_new_post(post_arg_set):\\n api, post_data, acct_data, page_id, config = post_arg_set\\n\\n try:\\n post_id = post_data['id'] if post_data.has_key('id') else None\\n\\n except Exception as e:\\n log.error( e )\\n\\n else:\\n\\n # parse date\\n if post_data.has_key('created_time') and post_data['created_time'] is not None: \\n dt = datetime.strptime(post_data['created_time'], FB_DATE_FORMAT)\\n date_time = tz_adj(dt, config)\\n time_bucket = round_datetime(date_time, config)\\n raw_timestamp = int(date_time.strftime(\\\"%s\\\"))\\n \\n else:\\n time_bucket = None\\n raw_timestamp = None\\n \\n # extract message so we can find links within the msg if not in url\\n article_urls = [get_fb_link(post_data, config, unshorten=True)]\\n message = post_data['message'].encode('utf-8') if post_data.has_key('message') else None\\n message_urls = get_message_urls(article_urls, message, config)\\n\\n # detect article links, unshorten and parse\\n article_urls = [\\n parse_url(unshorten_link(url, config)) \\\\\\n for url in article_urls + message_urls\\n if url is not None\\n ]\\n\\n article_urls = [url for url in article_urls if is_article(url, config)]\\n\\n if article_urls:\\n for article_url in set(article_urls):\\n\\n # sluggify url\\n article_slug = sluggify(article_url)\\n\\n # format data\\n post_value = {\\n 'article_slug': article_slug,\\n 'article_url': article_url,\\n 'time_bucket': time_bucket,\\n 'fb_post_created': raw_timestamp,\\n 'raw_timestamp': raw_timestamp,\\n 'fb_raw_link' : get_fb_link(post_data, config=config),\\n 'fb_page_id': page_id,\\n 'fb_post_id': post_id,\\n 'fb_page_likes': acct_data['likes'] if acct_data.has_key('likes') else None,\\n 'fb_page_talking_about': acct_data['talking_about_count'] if acct_data.has_key('talking_about_count') else None,\\n 'fb_type': post_data['type'] if post_data.has_key('type') else None,\\n 'fb_status_type': post_data['status_type'] if post_data.has_key('status_type') else None,\\n 'fb_message': message\\n }\\n \\n # always insert insights data\\n if is_insights(page_id, config):\\n \\n log.info( \\\"INSIGHTS\\\\tAdding data from %s re: %s\\\" % (page_id, article_slug) )\\n\\n # fetch data\\n insights_value = get_insights_data(api, page_id, post_id)\\n\\n # create datasource name\\n data_source = \\\"facebook_insights_%s\\\" % page_id \\n \\n # upsert url\\n upsert_url(article_url, article_slug, data_source, config)\\n\\n # insert id\\n db.sadd('facebook_post_ids', post_id)\\n\\n # format time bucket\\n current_time_bucket = gen_time_bucket(config)\\n insights_value['time_bucket'] = current_time_bucket\\n post_value.pop('time_bucket', None)\\n \\n value = json.dumps({\\n data_source : dict(post_value.items() + insights_value.items())\\n })\\n\\n # upload data to redis\\n db.zadd(article_slug, current_time_bucket, value) \\n \\n # only insert new posts\\n if not db.sismember('facebook_post_ids', post_id):\\n \\n log.info( \\\"FACEBOOK\\\\tNew post %s\\\\t%s\\\" % (post_id, article_url) )\\n \\n # insert id\\n db.sadd('facebook_post_ids', post_id) \\n \\n # upsert url\\n data_source = \\\"facebook_%s\\\" % page_id\\n upsert_url(article_url, article_slug, data_source, config)\\n\\n value = json.dumps( {data_source : post_value} )\\n\\n\\n # upload data to redis\\n db.zadd(article_slug, time_bucket, value)\",\n \"def proto_post(self, proto_category):\\n return baker.make(\\n Post,\\n content=(\\n \\\"Aliquip excepteur qui mollit labore nulla et culpa \\\"\\n \\\"minim et commodo reprehenderit consequat sint.\\\"\\n ),\\n categories=proto_category,\\n make_m2m=True,\\n _create_files=True,\\n )\",\n \"def post_to_writeas(token, publication, post_title, post_date, post_content):\\n writeas_post_url = 'https://write.as/api/collections/{}/posts'.format(\\n publication)\\n writeas_auth_header = {\\n 'Authorization': token,\\n 'Content-Type': 'application/json'\\n }\\n payload = {\\n 'title': post_title,\\n 'created': post_date,\\n 'body': post_content\\n }\\n r = requests.post(\\n writeas_post_url,\\n headers=writeas_auth_header,\\n json=payload)\\n\\n if r.status_code == 201:\\n result = r.json()\\n post_id = result['data']['id']\\n print('Post {} successfully created!'.format(post_id))\\n else:\\n print(\\\"Publishing post FAILED. Response: {}\\\".format(r.text))\\n writeas_logout(writeas_auth_token)\\n sys.exit(1)\",\n \"async def create_post(self, community: Community, post_id) -> w_Post:\\n post_url = self._api_communities_url + str(community.id) + '/posts/' + str(post_id)\\n async with self.web_session.get(post_url, headers=self._headers) as resp:\\n if self.check_status(resp.status, post_url):\\n data = await resp.json()\\n return (create_post_objects([data], community, new=True))[0]\",\n \"def topic_posting_new(self, request):\\n col = ObjectPostings(self.cdb_object_id, {})\\n return posting_new(col, request)\",\n \"def publish(self):\\n self.published = True\\n self.save()# pylint: disable=no-member\",\n \"def post(self):\\n subject = self.request.get('subject')\\n post_content = self.request.get('post_content')\\n submit = self.request.get('submit')\\n cancel = self.request.get('cancel')\\n user = self.get_active_user()\\n created_by = int(user.key().id())\\n post_id = self.request.get('post_id')\\n\\n if not user:\\n self.redirect('/login')\\n if post_id:\\n post = Posts.get_by_id(int(post_id))\\n else:\\n post = None\\n\\n if cancel == \\\"cancel\\\":\\n self.redirect('/%s' % str(post.key().id()))\\n return\\n if (post and post.submitter_id == user.key().id()) or not post:\\n if submit == \\\"submit\\\" and subject and post_content:\\n if post:\\n post.subject = subject\\n post.content = post_content\\n post.put()\\n else:\\n post = Posts(subject=subject,\\n content=post_content,\\n submitter_id=created_by)\\n post.put()\\n self.redirect('/%s' % str(post.key().id()))\\n else:\\n self.render_newpage(user=user,\\n subject=subject,\\n post_content=post_content,\\n error=\\\"\\\"\\\"Please provide both a subject and a\\n post!\\\"\\\"\\\")\\n else:\\n self.redirect('/login')\",\n \"def create_post_with_comment(pauthor, cauthor, visibility, ptext, ctext):\\n\\n post = Post.objects.create(content = ptext, author = pauthor, visibility=visibility)\\n comment = Comment.objects.create(comment = ctext, post = post, author = cauthor)\\n return (post, comment)\",\n \"def test_auto_set_published_on_create(self):\\n title = ('Transportation Challenges Limit Education Choices for '\\n 'Denver Parents')\\n summary = \\\"\\\"\\\"\\n Many families in the Denver metro area use public\\n transportation instead of a school bus because for them, a\\n quality education is worth hours of daily commuting. Colorado's\\n school choice program is meant to foster educational equity,\\n but the families who benefit most are those who have time and\\n money to travel. Low-income families are often left in a lurch.\\n \\\"\\\"\\\"\\n byline = \\\"Mile High Connects\\\"\\n story = create_story(title=title, summary=summary, byline=byline,\\n status='published')\\n self.assertNowish(story.published)\",\n \"def test_auto_set_published_on_create(self):\\n title = ('Transportation Challenges Limit Education Choices for '\\n 'Denver Parents')\\n summary = \\\"\\\"\\\"\\n Many families in the Denver metro area use public\\n transportation instead of a school bus because for them, a\\n quality education is worth hours of daily commuting. Colorado's\\n school choice program is meant to foster educational equity,\\n but the families who benefit most are those who have time and\\n money to travel. Low-income families are often left in a lurch.\\n \\\"\\\"\\\"\\n byline = \\\"Mile High Connects\\\"\\n story = create_story(title=title, summary=summary, byline=byline,\\n status='published')\\n self.assertNowish(story.published)\",\n \"def proto_post(self, proto_category):\\n return baker.make(\\n Post,\\n content=\\\"Aute non ex nostrud amet ipsum.\\\",\\n categories=proto_category,\\n make_m2m=True,\\n _create_files=True,\\n )\",\n \"def proto_post(self, proto_category):\\n return baker.make(\\n Post,\\n content=\\\"Aute non ex nostrud amet ipsum.\\\",\\n categories=proto_category,\\n make_m2m=True,\\n _create_files=True,\\n )\",\n \"def test_post_creation_success(self):\\n url = reverse('post-list', args=[self.topic1.url_name])\\n payload = {\\n 'author': self.user1.id,\\n 'title': 'Creating a post',\\n 'content': 'Rich content 4',\\n }\\n self.client.credentials(\\n HTTP_AUTHORIZATION = 'Token ' + self.user1.auth_token.key\\n )\\n response = self.client.post(url, payload)\\n\\n self.assertEqual(response.status_code, status.HTTP_201_CREATED)\\n new_post = Post.objects.filter(\\n author=self.user1,\\n title=payload.get('title'),\\n content=payload.get('content'),\\n topic=self.topic1\\n )\\n self.assertTrue(new_post.exists())\",\n \"def write_post(id, args):\\n graph = facebook.GraphAPI(id) \\n graph.put_object(parent_object='me', connection_name = 'feed', **args)\",\n \"def handler(event, _context):\\n model = PostModel()\\n post_id = model.create(**json.loads(event['body']))\\n return dump_result({'post_id': post_id}, status_code=201)\",\n \"def test_show_post_view_with_a_published_post(self):\\n category = create_category('Category 1')\\n author = create_author('Author 1')\\n published_post = create_post(category=category, author=author, name='Published Post',\\n content='Lorem ipsum dolor sit amet, consectetur adipiscing elit. Praesent sollicitudin.',\\n status='Published')\\n url = reverse('blog.post', args=(published_post.id,))\\n create_question(question_text=\\\"Past question.\\\", days=-30)\\n response = self.client.get(url)\\n self.assertContains(response, published_post.content)\",\n \"def post(self):\\n\\n subject = self.request.get('subject')\\n content = self.request.get('content')\\n\\n have_errors = False\\n\\n if not subject:\\n error_subject = \\\"Please write down the subject\\\"\\n have_errors = True\\n if not content:\\n error_content = \\\"Content is required\\\"\\n have_errors = True\\n\\n if have_errors:\\n self.render(\\\"newpost.html\\\",\\n subject=subject,\\n content=content,\\n error_subject=error_subject,\\n error_content=error_content,\\n user=self.user)\\n else:\\n post = Post(parent=blog_key(),\\n subject=subject,\\n content=content,\\n user=self.user)\\n post.put()\\n self.redirect('/blog/%s' % str(post.key().id()))\",\n \"def create(self, request, *args, **kwargs):\\n serializer = self.get_serializer(data=request.data)\\n\\n if serializer.is_valid():\\n data = serializer.data\\n Article.objects.create_article(url=data['url'],\\n authors=data['authors'],\\n publish_time=data['publish_time'],\\n title_image=data['images'])\\n return Response({'message': 'News story added.'})\\n else:\\n return Response({'error': serializer.errors})\",\n \"def new_post(mkp_form, request):\\n newpost = Posts()\\n newpost.init()\\n newpost.authorid = int(request.user.id)\\n newpost.title = mkp_form.cleaned_data['title']\\n newpost.name = mkp_form.cleaned_data['short_title'] # 缩略名\\n newpost.cover = mkp_form.cleaned_data['cover_url']\\n newpost.introduction = mkp_form.cleaned_data['introduction']\\n newpost.content = js_resize_img(mkp_form.cleaned_data['content'])\\n newpost.status = Status.objects.get(id=2) # id为2是已发布的文章，默认为已发布，后面再改\\n tagids = mkp_form.cleaned_data['tags']\\n if len(tagids) != 0:\\n for tagid in tagids:\\n tagid = int(tagid)\\n tag = Tags.objects.get(id=tagid)\\n newpost.tags.add(tag)\\n threadtypeid = mkp_form.cleaned_data['threadtypeid']\\n newpost.threadtypeid = ThreadTypes.objects.get(id=threadtypeid)\\n if mkp_form.cleaned_data['commentnotshow'] != '':\\n newpost.comment_status = False\\n else:\\n newpost.comment_status = True\\n return newpost\",\n \"def post(self):\\n title = self.request.get(\\\"title\\\")\\n body = self.request.get(\\\"body\\\")\\n\\n if title and body:\\n\\n # create a new Post object and store it in the database\\n post = Post(\\n title=title,\\n body=body\\n )\\n post.put()\\n\\n # get the id of the new post, so we can render the post's page (via the permalink)\\n id = post.key().id()\\n self.redirect(\\\"/blog/%s\\\" % id)\\n else:\\n error = \\\"we need both a title and a body!\\\"\\n #self.render_form(title, body, error)\\n self.render(\\\"newpost.html\\\", title, body, error)\",\n \"def newPost(self, useRawHTML):\\n print\\n content, publish = self._fillPost(useRawHTML)\\n\\n # Upload to server\\n try :\\n postid = self.server.metaWeblog.newPost(\\n self.blogid, self.username, self.password,\\n content, publish\\n )\\n except xmlrpclib.Fault as fault:\\n display_XMLRPC_errors(\\\"post the new entry\\\", fault)\\n import pdb\\n pdb.set_trace()\\n else :\\n self._setCategorie(postid)\\n print \\\"New post created with ID =\\\", postid\",\n \"def test_create_post(self):\\n with self.client:\\n result = self.client.post('/users/spike-test/posts', data={\\n \\\"post-title\\\": \\\"autotest title\\\",\\n \\\"post-content\\\": \\\"autotest content for blog post testing\\\"\\n }, follow_redirects=True)\\n self.assertEqual(result.status_code, 200)\\n self.assertIn(b'autotest title', result.data)\",\n \"def test_home_view_with_a_published_post(self):\\n category = create_category('Category 1')\\n author = create_author('Author 1')\\n create_post(category=category, author=author, name='Published Post', content='Lorem ipsum dolor sit amet, consectetur adipiscing elit. Praesent sollicitudin.', status='Published')\\n create_question(question_text=\\\"Past question.\\\", days=-30)\\n response = self.client.get(reverse('blog.home'))\\n self.assertQuerysetEqual(\\n response.context['posts'],\\n ['<Post: Published Post>']\\n )\",\n \"def post(self):\\n subject = self.request.get('subject')\\n content = self.request.get('content')\\n\\n # if user enter good subject and content, redirect them to new post page\\n if subject and content:\\n p = Post(parent = blog_key(), subject = subject, content = content)\\n p.put() # store the post element into database\\n self.redirect('/blog/%s' % str(p.key().id()))\\n # otherwise, render an error page \\n else:\\n error = \\\"subject and content, please!\\\"\\n self.render(\\\"newpost.html\\\", subject=subject, content=content, error=error)\",\n \"def create(cls, headline, text, blog):\\n post = cls()\\n post.headline = headline\\n post.text = text\\n post.blog = blog\\n post.posted_date = timezone.now()\\n try:\\n post.save()\\n return post\\n except(ValueError, IntegrityError, OperationalError):\\n return None\",\n \"def new(request):\\n\\n form = PostForm(request.POST or None)\\n\\n if request.method == \\\"POST\\\":\\n if form.is_valid():\\n instance = form.save(commit=False)\\n instance.user = request.user\\n instance.save()\\n return HttpResponseRedirect(reverse('posts:toggle_publish', args=(instance.id,)))\\n\\n context = {'form':form}\\n\\n return render(request, 'posts/new.html', context)\",\n \"def make_post(request):\\n if request.user.is_authenticated() and request.POST:\\n member = Member.objects.get(user=request.user)\\n thread_id = request.POST.get('thread_id', -1)\\n content = request.POST.get('content', -1)\\n if thread_id != -1 and content != -1 and member:\\n post = Post()\\n post.author = member\\n post.thread = Thread.objects.get(pk=thread_id)\\n post.content = content\\n post.save()\\n return HttpResponse(200)\\n else:\\n return server_error(request)\\n else:\\n return server_error(request)\",\n \"def post(self):\\n\\n title = self.request.get(\\\"title\\\")\\n blogPost = self.request.get(\\\"blogPost\\\")\\n author = self.request.cookies.get('name')\\n\\n if title and blogPost:\\n\\n bp = Blogposts(parent=blog_key(), title=title,\\n blogPost=blogPost, author=check_secure_val(author))\\n\\n bp.put()\\n\\n self.redirect('/%s' % str(bp.key.integer_id()))\\n else:\\n error = \\\"Please submit both a title and a blogpost!\\\"\\n self.render(\\\"newpost.html\\\", title=title,\\n blogPost=blogPost, error=error)\",\n \"def post(self):\\n data = request.json\\n return create_new_blog(data=data)\",\n \"def create_activity(self, created_user, source, action,\\n privacy=Privacy.PRIVATE, **kwargs):\\n Activity = get_activity_model()\\n return Activity.objects.create(\\n about=self,\\n action=action,\\n created_user=created_user,\\n source=source,\\n privacy=privacy,\\n **kwargs\\n )\",\n \"def _create(self, obj, preview=None, include_link=False):\\n # TODO: validation, error handling\\n assert preview in (False, True)\\n type = obj.get('objectType')\\n verb = obj.get('verb')\\n base_id, base_url = self.base_object(obj, verb=verb)\\n if base_id and not base_url:\\n base_url = self.object_url(base_id)\\n\\n content = self._content_for_create(obj)\\n if not content:\\n if type == 'activity':\\n content = verb\\n else:\\n return source.creation_result(\\n abort=False, # keep looking for things to post\\n error_plain='No content text found.',\\n error_html='No content text found.')\\n\\n url = obj.get('url')\\n if include_link and url:\\n content += '\\\\n\\\\n(%s)' % url\\n preview_content = util.linkify(content)\\n msg_data = {'message': content.encode('utf-8')}\\n if appengine_config.DEBUG:\\n msg_data['privacy'] = json.dumps({'value': 'SELF'})\\n msg_data = urllib.urlencode(msg_data)\\n\\n if type == 'comment':\\n if not base_url:\\n return source.creation_result(\\n abort=True,\\n error_plain='Could not find a Facebook status to reply to.',\\n error_html='Could not find a Facebook status to <a href=\\\"http://indiewebcamp.com/comment\\\">reply to</a>. '\\n 'Check that your post has an <a href=\\\"http://indiewebcamp.com/comment\\\">in-reply-to</a> '\\n 'link a Facebook URL or to an original post that publishes a '\\n '<a href=\\\"http://indiewebcamp.com/rel-syndication\\\">rel-syndication</a> link to Facebook.')\\n\\n if preview:\\n return source.creation_result(\\n 'will <span class=\\\"verb\\\">comment</span> <em>%s</em> on '\\n '<a href=\\\"%s\\\">this post</a>:\\\\n%s' %\\n (preview_content, base_url, EMBED_POST % base_url))\\n else:\\n resp = json.loads(self.urlopen(API_COMMENTS_URL % base_id,\\n data=msg_data).read())\\n resp.update({'url': self.comment_url(base_id, resp['id']),\\n 'type': 'comment'})\\n\\n elif type == 'activity' and verb == 'like':\\n if not base_url:\\n return source.creation_result(\\n abort=True,\\n error_plain='Could not find a Facebook status to like.',\\n error_html='Could not find a Facebook status to <a href=\\\"http://indiewebcamp.com/favorite\\\">like</a>. '\\n 'Check that your post has an <a href=\\\"http://indiewebcamp.com/favorite\\\">like-of</a> '\\n 'link a Facebook URL or to an original post that publishes a '\\n '<a href=\\\"http://indiewebcamp.com/rel-syndication\\\">rel-syndication</a> link to Facebook.')\\n\\n if preview:\\n return source.creation_result(\\n 'will <span class=\\\"verb\\\">like</span> <a href=\\\"%s\\\">this post</a>:\\\\n%s' %\\n (base_url, EMBED_POST % base_url))\\n else:\\n resp = json.loads(self.urlopen(API_LIKES_URL % base_id, data='').read())\\n assert resp == True, resp\\n resp = {'type': 'like'}\\n\\n elif type == 'activity' and verb in RSVP_ENDPOINTS:\\n if not base_url:\\n return source.creation_result(\\n abort=True,\\n error_plain=\\\"This looks like an RSVP, but it's missing an \\\"\\n \\\"in-reply-to link to the Facebook event.\\\",\\n error_html=\\\"This looks like an <a href='http://indiewebcamp.com/rsvp'>RSVP</a>, \\\"\\n \\\"but it's missing an <a href='http://indiewebcamp.com/comment'>in-reply-to</a> \\\"\\n \\\"link to the Facebook event.\\\")\\n\\n # TODO: event invites\\n if preview:\\n assert verb.startswith('rsvp-')\\n return source.creation_result(\\n 'will <span class=\\\"verb\\\">RSVP %s</span> to '\\n '<a href=\\\"%s\\\">this event</a>.<br />' % (verb[5:], base_url))\\n else:\\n resp = json.loads(self.urlopen(RSVP_ENDPOINTS[verb] % base_id, data='').read())\\n assert resp == True, resp\\n resp = {'type': 'rsvp'}\\n\\n elif type in ('note', 'article'):\\n if preview:\\n return source.creation_result(\\n 'will <span class=\\\"verb\\\">post</span>:<br /><br />'\\n '<em>%s</em><br />' % preview_content)\\n else:\\n resp = json.loads(self.urlopen(API_FEED_URL, data=msg_data).read())\\n resp.update({'url': self.post_url(resp), 'type': 'post'})\\n\\n elif type == 'activity' and verb == 'share':\\n return source.creation_result(\\n abort=True,\\n error_plain='Cannot publish shares on Facebook.',\\n error_html='Cannot publish <a href=\\\"https://www.facebook.com/help/163779957017799\\\">shares</a> '\\n 'on Facebook. This limitation is imposed by the '\\n '<a href=\\\"https://developers.facebook.com/docs/graph-api/reference/v2.0/object/sharedposts/#publish\\\">Facebook Graph API</a>.')\\n\\n else:\\n return source.creation_result(\\n abort=False,\\n error_plain='Cannot publish type=%s, verb=%s to Facebook' % (type, verb),\\n error_html='Cannot publish type=%s, verb=%s to Facebook' % (type, verb))\\n\\n if 'url' not in resp:\\n resp['url'] = base_url\\n return source.creation_result(resp)\",\n \"def post():\\n\\n form = forms.PostForm()\\n if form.validate_on_submit():\\n models.Post.create(title=form.title.data,\\n date=form.date.data,\\n time_spent=form.time_spent.data,\\n details=form.details.data,\\n remember=form.remember.data)\\n return redirect(url_for('index'))\\n return render_template('new.html', form=form)\",\n \"def post(self, post_id=None):\\n\\n if post_id:\\n abort(400)\\n else:\\n args = parsers.post_post_parser.parse_args(strict=True)\\n\\n new_post = Post(args['title'])\\n new_post.text = args['text']\\n # new_post.user = user\\n\\n if args['tags']:\\n for item in args['tags']:\\n tag = Tag.query.filter_by(name=item).first()\\n # If the tag already exist, append.\\n if tag:\\n new_post.tags.append(tag)\\n # If the tag not exist, create the new one.\\n # Will be write into DB with session do.\\n else:\\n new_tag = Tag(item)\\n new_post.tags.append(new_tag)\\n db.session.add(new_post)\\n db.session.commit()\\n return (new_post.id, 201)\",\n \"def post(self):\\n current_user = self.authenticate_user()\\n\\n if not current_user:\\n self.redirect(\\\"/login\\\")\\n else:\\n content = self.request.get(\\\"content\\\")\\n title = self.request.get(\\\"subject\\\")\\n\\n if not content or not title:\\n self.render_front(title, content, \\\"We need both a title and content\\\")\\n else:\\n post = Post(title=title, content=content, user=current_user.key)\\n post.put()\\n\\n current_user.posts.append(post.key)\\n current_user.put()\\n\\n self.redirect(\\\"/post/\\\" + str(post.key.id()))\",\n \"def create_action(instance, verb, user):\\n return instance.activities.create(action=verb, owner=user)\",\n \"def create_activity(request: Request, activity_type: str, msg_context: dict, object_id: UUID, user: User):\\n dbsession = Session.object_session(user)\\n\\n stream = Stream.get_or_create_user_stream(user)\\n\\n a = Activity()\\n a.object_id = object_id\\n a.activity_type = activity_type\\n a.msg_context = msg_context\\n\\n stream.activities.append(a)\\n dbsession.flush()\\n\\n return a\",\n \"def post(activity, create, extracted, **kwargs):\\n if activity.verb in [\\n models.ReferralActivityVerb.ASSIGNED,\\n models.ReferralActivityVerb.UNASSIGNED,\\n ]:\\n # pylint: disable=attribute-defined-outside-init\\n activity.item_content_object = UserFactory()\\n elif activity.verb in [\\n models.ReferralActivityVerb.DRAFT_ANSWERED,\\n models.ReferralActivityVerb.ANSWERED,\\n ]:\\n # pylint: disable=attribute-defined-outside-init\\n activity.item_content_object = ReferralAnswerFactory(\\n referral=activity.referral\\n )\\n elif activity.verb in [\\n models.ReferralActivityVerb.VALIDATED,\\n models.ReferralActivityVerb.VALIDATION_DENIED,\\n ]:\\n # pylint: disable=attribute-defined-outside-init\\n activity.item_content_object = ReferralAnswerValidationResponseFactory()\\n elif activity.verb == models.ReferralActivityVerb.VALIDATION_REQUESTED:\\n # pylint: disable=attribute-defined-outside-init\\n activity.item_content_object = ReferralAnswerValidationRequestFactory()\\n elif activity.verb == models.ReferralActivityVerb.CREATED:\\n pass\\n else:\\n raise Exception(f\\\"Incorrect activity verb {activity.verb}\\\")\",\n \"def create_multiple_posts(author, num, ptext = TEXT, visibility = ACL_DEFAULT):\\n posts = []\\n\\n for i in range(num):\\n posts.append(Post.objects.create(content = ptext, author = author, visibility=visibility))\\n\\n return posts\",\n \"def get_post(self):\\n\\t\\tself.post = graph.get_object(POST_ID)\",\n \"def addPost(postCreatedEvent):\\n query = generatePostSQLQuery(postCreatedEvent)\\n with conn.cursor() as cur:\\n cur.execute(query)\\n conn.commit()\",\n \"def perform_create(self, serializer):\\n lookup_url_kwarg = self.lookup_url_kwarg or self.lookup_field\\n lookup = self.kwargs.get(lookup_url_kwarg, None)\\n \\n if lookup is not None:\\n filter_kwargs = {self.lookup_field: lookup}\\n video = get_object_or_404(Video, **filter_kwargs)\\n clip = serializer.save(video=video, owner=self.request.user) \\n \\n # register this new activity\\n activity_send(self.request.user, verb='add', object=clip, target=video)\\n else:\\n raise Http404\",\n \"def insert_post(shard, **kwargs):\\n # Create the posting and insert it.\\n post_id = kwargs.pop('post_id', None)\\n if not post_id:\\n post_id = models.human_uuid()\\n\\n new_topic = kwargs.get('new_topic', None)\\n\\n kwargs['post_time'] = datetime.datetime.now()\\n\\n post_key = ndb.Key(models.Post._get_kind(), post_id)\\n post = models.Post(\\n key=post_key,\\n **kwargs)\\n\\n @ndb.tasklet\\n def txn():\\n if (yield post_key.get_async(use_memcache=False, use_cache=False)):\\n logging.warning('Post already exists for shard=%r, post_id=%r',\\n shard, post_id)\\n raise ndb.Rollback()\\n\\n yield post.put_async(use_memcache=False, use_cache=False)\\n\\n # Pull task that indicates the post to apply. This must encode the\\n # new_topic data for this post so the apply_posts() function doesn't\\n # need the models.Post entity in order to make progress.\\n enqueue_post_task(shard, [post_id], new_topic=new_topic)\\n\\n # Notify all users of the post.\\n futures = []\\n futures.append(ndb.transaction_async(txn))\\n futures.append(notify_posts(shard, [post]))\\n\\n # Set the dirty bit for this shard. This causes apply_posts to run a\\n # second time if the Post transaction above completed while apply_posts\\n # was already in flight.\\n dirty_bit(shard, set=True)\\n\\n # Enqueue an apply task to sequence and notify the new post.\\n futures.append(enqueue_apply_task(shard, post_id=post_id))\\n\\n # Wait on futures in case they raise errors.\\n ndb.Future.wait_all(futures)\\n\\n return post_key\",\n \"def post(title, drafts=False):\\n # Parsing options\\n if drafts:\\n dest = _drafts_dest\\n else:\\n dest = _posts_dest\\n\\n # File name\\n date = get_date()\\n name = sanitize(title)\\n fname = \\\"{}-{}{}\\\".format(date, name, _post_ext)\\n\\n # Front Matter\\n front_matter = []\\n front_matter.append('---')\\n front_matter.append('layout: post')\\n front_matter.append('title: {}'.format(title))\\n front_matter.append('description: \\\"\\\"')\\n front_matter.append('tags: \\\"\\\"')\\n front_matter.append('comments: true')\\n front_matter.append('permalink: \\\"\\\"')\\n front_matter.append('sitemap:\\\\n lastmod: {}'.format(date))\\n front_matter.append('---')\\n\\n # Create post file and write Front Matter\\n print(\\\"\\\\nCreating new post '{}' in {}\\\\n\\\".format(fname, dest))\\n try:\\n f = open(dest + fname, 'w')\\n except Exception as e:\\n print(\\\"* [Error] occured: {}\\\\n\\\".format(e))\\n else:\\n f.write('\\\\n'.join(front_matter))\\n f.close()\\n print(\\\"* Done.\\\\n\\\")\",\n \"def post(self, post):\\n\\n self._post = post\",\n \"def post(self, post):\\n\\n self._post = post\",\n \"def post(self):\\n data = request.json\\n create_entry(data)\\n return None, 201\",\n \"def create(self, request, *args, **kwargs):\\n response = super(ReleaseViewSet, self).create(request, *args, **kwargs)\\n if response.status_code == status.HTTP_201_CREATED:\\n signals.release_post_update.send(sender=self.object.__class__,\\n release=self.object,\\n request=request)\\n return response\",\n \"def create_post(request):\\n if request.method == 'POST':\\n title = request.POST['title']\\n content = request.POST['content']\\n user_id = request.POST['author_id']\\n category = request.POST['category']\\n\\n slug = \\\"-\\\".join(list(map(lambda word: word.lower(), title.split())))\\n author = User.objects.get(id=int(user_id))\\n\\n # save info in models\\n post = Post()\\n post.author = author\\n post.category = category\\n post.title = title\\n post.content = content\\n post.slug = slug\\n post.save()\\n return redirect('post')\\n\\n return render(request, 'posts/create_post.html')\",\n \"def post_create(request):\\n\\tform = PostForm(request.POST or None, request.FILES or None)\\n\\tif request.POST:\\n\\t\\tif form.is_valid():\\n\\t\\t\\tinstance = form.save(commit=False)\\n\\t\\t\\tinstance.user = request.user\\n\\t\\t\\tinstance.save()\\n\\t\\t\\tmessages.success(request, \\\"Post created!\\\")\\n\\t\\t\\treturn HttpResponseRedirect(instance.get_absolute_url())\\n\\t\\telse:\\n\\t\\t\\tmessages.error(request, \\\"Sorry! Something went wrong.\\\", extra_tags=\\\"\\\")\\n\\tcontext = {\\n\\t\\t'title': \\\"Create Post\\\",\\n\\t\\t'form' : form,\\n\\t}\\n\\treturn render(request, 'post/create.html', context)\",\n \"def PublishIt(name, path, comments, task=os.getenv('TASK'), status=\\\"WORK IN PROGRESS\\\"):\\n\\n db = get_connection()\\n\\n PubCollections = db['submissions']\\n\\n # creation of the dailies submission entry\\n publishDict = dict()\\n publishDict['date'] = now\\n publishDict['type'] = \\\"publish\\\"\\n publishDict['user_name'] = main_user\\n publishDict['task'] = task\\n publishDict['status'] = status\\n publishDict['asset'] = name\\n publishDict['path'] = path\\n publishDict['comment'] = comments\\n PubCollections.save(publishDict)\\n notifications.push_notifications({\\\"name\\\": main_user, \\\"email\\\": os.getenv('USER_EMAIL')}, users_list, \\\"publish\\\", shot, now)\",\n \"def save(self, *args, **kwargs):\\n if not self.id:\\n self.created = timezone.now()\\n return super(UserActivity, self).save(*args, **kwargs)\",\n \"def generate_presigned_post(Bucket=None, Key=None, Fields=None, Conditions=None, ExpiresIn=None):\\n pass\",\n \"def create_post(bid):\\n form = PostForm(request.form)\\n if request.method == 'POST':\\n if form.validate():\\n DB.session.add(\\n Post(\\n bid,\\n current_user.uid,\\n form.name.data,\\n form.desc.data))\\n DB.session.commit()\\n flash('Post ({}) successfully created!'.format(form.name.data))\\n else:\\n flash(constants.DEFAULT_SUBMISSION_ERR)\\n return redirect(request.referrer)\",\n \"def test_create_post():\\n post = Post(\\\"Test\\\", \\\"Test Content\\\")\\n\\n assert post.title == \\\"Test\\\"\\n assert post.content == \\\"Test Content\\\"\",\n \"def publish():\\n pass\",\n \"def save_post(self, post):\\n return self.collection.insert_one(post.serialize())\",\n \"def should_publish(self, create, extracted, **kwargs):\\n if create and extracted:\\n return True\\n return False\",\n \"def community_post_create_view(request):\\n task = \\\"Create New\\\"\\n form = AddEditPostForm() # An unbound form\\n\\n if request.method == 'POST': # If the form has been submitted...\\n form = AddEditPostForm(request.POST, request.FILES) # A form bound to the POST data\\n if form.is_valid(): # All validation rules pass\\n post = form.save(commit=False) # Create a new object from the form, but don't save it to the database\\n post.author = request.user # Set the author to the current user\\n post.save() # Save the object to the database\\n slug_str = \\\"%s %s\\\" % (post.title, post.date_posted) # Create a slug from the title and date\\n post.slug = slugify(slug_str) # Create the slug\\n post.save() # Save the object to the database\\n return redirect('community-home') # Redirect to the home page\\n\\n context = { # Pass the variables to the template\\n 'task': task,\\n 'form': form,\\n }\\n return render(request,\\n 'pages/patient-community/community-create-update-post.html',\\n context) # render the patient community create post page\",\n \"def announce_board_posting_created(event: BoardPostingCreated) -> None:\\n posting_creator_screen_name = get_screen_name_or_fallback(\\n event.posting_creator_screen_name\\n )\\n board_label = _get_board_label(event.topic_id)\\n\\n if event.topic_muted:\\n return\\n\\n text = (\\n f'{posting_creator_screen_name} hat im {board_label} '\\n f'auf das Thema \\\"{event.topic_title}\\\" geantwortet: {event.url}'\\n )\\n\\n send_message(CHANNEL_ORGA_LOG, text)\\n send_message(CHANNEL_PUBLIC, text)\",\n \"def publish_release(ctx):\\n rel = _get_release()\\n rel.update_release(rel.title, rel.raw_data[\\\"body\\\"], draft=False)\",\n \"def create_post(request):\\n\\n # modified from: http://django-angular.readthedocs.org/en/latest/angular-model-form.html\\n\\n # get data\\n in_data = getRequestData(request)\\n\\n try:\\n # save in database\\n # note that in_data.mytitle throws an error while in_data.get('mytitle') works smoothly\\n post = Thread(pub_date = datetime.datetime.now(pytz.timezone('US/Eastern')), username = in_data.get('myusername'), title = in_data.get('mytitle'), description = in_data.get('mydescription'))\\n post.save()\\n except:\\n return HttpResponseBadRequest('Error saving to database!')\\n\\n return JsonResponse(in_data)\",\n \"def toggle_publish(request,id):\\n \\n instance = get_object_or_404(Post, id=id)\\n\\n if request.method==\\\"POST\\\":\\n instance.published = not instance.published\\n instance.save()\\n\\n t, created = LastUpdate.objects.get_or_create(id=1)\\n\\n t.updated = instance.updated\\n t.save()\\n\\n cache.clear()\\n return redirect('posts:home')\\n\\n context = {'post':instance}\\n\\n return render(request, 'posts/publish.html', context)\",\n \"def create(self, *args, **kwargs):\\n pass\",\n \"def post(self, user):\\n subject = self.request.get(\\\"subject\\\")\\n content = self.request.get(\\\"content\\\")\\n\\n if subject and content:\\n post = Post(subject=subject,\\n content=content,\\n author=self.user)\\n post.put()\\n return self.redirect(\\\"/%s\\\" % post.key().id())\\n else:\\n\\n error = \\\"subject and content, please!\\\"\\n return self.render(\\\"newpost.html\\\",\\n subject=subject,\\n content=content,\\n error=error)\",\n \"def create_url_post(post_id, config):\\n text = \\\"https://graph.facebook.com/\\\"\\n text += config[\\\"graph_api_version\\\"]\\n text += \\\"/\\\" + post_id + \\\"?fields=\\\"\\n text += \\\"id, message, admin_creator, backdated_time,\\\"\\n text += \\\"caption,coordinates,created_time,description,\\\"\\n text += \\\"feed_targeting,from,event,icon,is_popular,link,\\\"\\n text += \\\"message_tags,name,object_id,type,shares,\\\"\\n text += \\\"story,source,properties,place,target,targeting,\\\"\\n text += \\\"status_type,comments.fields(comment_count,from,id,like_count,\\\"\\n text += \\\"permalink_url,created_time,message,comments.fields(comment_count,\\\"\\n text += \\\"from,id,like_count,created_time,message,\\\"\\n text += \\\"reactions.fields(id,name,type,username,profile_type)),\\\"\\n text += \\\"reactions.fields(id,name,type,username,profile_type)),\\\"\\n text += \\\"reactions.fields(id,name,type,username,profile_type), sharedposts\\\"\\n text += \\\"&access_token=\\\" + config[\\\"access_token\\\"]\\n\\n return text\",\n \"def create_post(self: User, content: str, is_public: bool, circles: List[Circle], reshareable: bool,\\n reshared_from: Optional[Post], media_list: List[Media], mentioned_users: List[User],\\n is_update_avatar: bool) \\\\\\n -> Union[Post, bool]:\\n if not content and not media_list:\\n # a post has to have either content or media\\n return False\\n\\n new_post = Post()\\n new_post.eid = make_uuid()\\n new_post.author = self.id\\n if content:\\n new_post.content = bleach.clean(content)\\n new_post.is_public = is_public\\n new_post.circles = circles\\n new_post.media_list = media_list\\n new_post.is_update_avatar = is_update_avatar\\n\\n if reshared_from and not reshareable:\\n # if resharing from a post, this post must also be reshareable, otherwise it's logically wrong\\n return False\\n\\n if reshared_from:\\n if media_list:\\n # when resharing, only allow content (text), e.g. no media\\n return False\\n\\n if reshared_from.reshared_from:\\n # if reshared_from itself is a reshared post, reshare reshared_from's original post\\n # reshared_from.reshared_from is LazyReference so need to retrieve the full post\\n reshared_from = get_in_post_cache(reshared_from.reshared_from.id)\\n\\n # same explanation for context_home_or_profile=False\\n if not sees_post(self, reshared_from, context_home_or_profile=False):\\n return False\\n\\n if not reshared_from.reshareable:\\n return False\\n\\n new_post.reshared_from = reshared_from\\n\\n new_post.reshareable = reshareable\\n new_post.save()\\n\\n if reshared_from:\\n create_notification(\\n self,\\n notifying_href=new_post.make_href(),\\n notifying_summary=new_post.content,\\n notifying_action=NotifyingAction.Reshare,\\n notified_href=reshared_from.make_href(),\\n notified_summary=reshared_from.content,\\n owner=reshared_from.author\\n )\\n # only cache reshared post\\n set_in_post_cache(reshared_from)\\n\\n mention(\\n self,\\n notified_href=new_post.make_href(),\\n notified_summary=new_post.content,\\n mentioned_users=mentioned_users\\n )\\n\\n return new_post\",\n \"def post(self):\\n post_id = self.request.get('post_id')\\n post = Post.get_by_id(int(post_id), parent=blog_key())\\n content = self.request.get('comment')\\n\\n if content:\\n comment = Comment(parent=comment_key(),\\n content=content,\\n user=self.user,\\n post=post)\\n comment.put()\\n\\n time.sleep(0.1)\\n self.redirect('/blog/%s' % str(post.key().id()))\",\n \"def post(self, req):\\n error_messages = []\\n success_message = ''\\n\\n # Creamos owner y se lo pasamos al form con un objeto pre-establecido\\n post_with_owner = Post()\\n post_with_owner.owner = req.user\\n post_with_owner.blog = Blog.objects.filter(owner=req.user)[0]\\n\\n form = PostCreateForm(req.POST, instance=post_with_owner)\\n if form.is_valid():\\n\\n new_post = form.save()\\n form = PostCreateForm()\\n success_message = u'Post guardado con éxito! '\\n success_message += u'<a href=\\\"{0}\\\">'.format(reverse('post_detail', args=[req.user.username, new_post.pk]))\\n success_message += u'(ver post)</a>'\\n else:\\n error_messages.append(u'Formulario incompleto.')\\n\\n context = {\\n 'form': form,\\n 'success_message': success_message\\n }\\n return render(req, 'posts/new_post.html', context)\",\n \"def for_publishing_only(self):\\n self.token['type'] = 'publish'\\n\\n return self\",\n \"def write_page():\\n #1480631468\\n page_id = request.args['page_id']\\n post_time = request.args['time']\\n #bday=2016-11-10\\n args={}\\n if post_time is not None:\\n pattern = '%Y-%m-%d'\\n epoch_time=int(time.mktime(time.strptime(post_time, pattern)))\\n print_me(epoch_time)\\n \\n args['scheduled_publish_time']=epoch_time\\n\\n args['message'] = request.args['text']\\n args['published']= request.args['p_type'] == 'Published'\\n\\n \\\"\\\"\\\"\\n demo purposes\\\"\\n \\\"\\\"\\\"\\n #args['scheduled_publish_time']=\\\"1480631468\\\"\\n #args['published']=\\\"0\\\"\\n write_post(page_id, args)\\n return render_template('write_done.html')\",\n \"def posts_create(request):\\n if request.method == 'POST':\\n form = PostForm(request.POST, request.FILES)\\n if form.is_valid():\\n form.save()\\n return redirect('feed')\\n\\n else:\\n form = PostForm()\\n\\n return render(\\n request=request,\\n template_name='posts/new.html',\\n context={\\n 'form': form,\\n 'user': request.user,\\n 'perfil': request.user.perfil\\n }\\n )\",\n \"def post(self):\\n user_id = get_jwt_identity()\\n user = find_user(user_id)\\n args = post_parser.parse_args()\\n\\n # check circles\\n circles = []\\n for circle_id in args['circle_ids']:\\n found_circle = find_circle(user, circle_id)\\n if not found_circle:\\n return {'msg': f'Circle {circle_id} is not found'}, 404\\n circles.append(found_circle)\\n\\n # check reshare\\n reshared_from = args['reshared_from']\\n reshared_from_post = None\\n if reshared_from:\\n reshared_from_post = dangerously_get_post(reshared_from)\\n if not reshared_from_post:\\n return {\\\"msg\\\": f\\\"Post {reshared_from} is not found\\\"}, 404\\n\\n # check media\\n media_object_names = args['media_object_names']\\n if reshared_from and media_object_names:\\n return {'msg': \\\"Reshared post is not allowed to have media\\\"}, 400\\n\\n post = create_post(\\n user,\\n content=args['content'],\\n is_public=args['is_public'],\\n circles=circles,\\n reshareable=args['reshareable'],\\n reshared_from=reshared_from_post,\\n media_list=check_media_object_names(media_object_names, MaxPostMediaCount),\\n mentioned_users=check_mentioned_user_ids(args['mentioned_user_ids']),\\n is_update_avatar=False\\n )\\n if not post:\\n return {\\\"msg\\\": f\\\"Not allowed to reshare post {reshared_from}\\\"}, 403\\n return post, 201\",\n \"def create_link_page_post(self, page_id, link=None, message=None, picture=None,\\n thumbnail=None, name=None, caption=None,\\n description=None, published=None, call_to_action=None, batch=False):\\n page_access_token = self.get_page_access_token(page_id)\\n if 'error' in page_access_token:\\n return page_access_token\\n if 'access_token' not in page_access_token:\\n raise AdsAPIError('Could not get page access token. (Do you have manage pages permission?)')\\n path = '%s/feed' % page_id\\n args = {\\n 'access_token': page_access_token['access_token'],\\n }\\n files = {}\\n if link is not None:\\n args['link'] = link\\n if message is not None:\\n args['message'] = message\\n if picture is not None:\\n args['picture'] = picture\\n if thumbnail is not None:\\n files['thumbnail'] = thumbnail\\n if published is not None:\\n args['published'] = published\\n if name is not None:\\n args['name'] = name\\n if caption is not None:\\n args['caption'] = caption\\n if description is not None:\\n args['description'] = description\\n if call_to_action is not None:\\n args['call_to_action'] = json.dumps(call_to_action)\\n return self.make_request(path, 'POST', args, files, batch=batch)\",\n \"def post(request):\\n if request.method == \\\"POST\\\":\\n post = Post()\\n post.content = request.POST['content']\\n post.author = request.user\\n post.save()\\n return HttpResponseRedirect(reverse(\\\"index\\\"))\",\n \"def create_post():\\r\\n\\r\\n # Check for and reject empty username or whinge\\r\\n if not request.values.get(\\\"username\\\") or not request.values.get(\\\"whinge\\\"):\\r\\n print(\\\"Ignoring request to with empty username or whinge\\\")\\r\\n else:\\r\\n # Form data ok; add to DB\\r\\n con = get_db()\\r\\n con.execute(\\\"INSERT INTO posts (submitter,content,ts) VALUES (?,?,?);\\\",\\r\\n (\\r\\n request.values.get(\\\"username\\\"), # form field username -> DB column submitter\\r\\n request.values.get(\\\"whinge\\\"), # form field whinge -> DB column content\\r\\n time.time()\\r\\n )\\r\\n )\\r\\n con.commit()\\r\\n con.close()\\r\\n \\r\\n # TODO: Handle possibility of failed INSERT\\r\\n\\r\\n # Send them back to the main page\\r\\n return redirect(url_for(\\\"display_top\\\"))\",\n \"def post(self):\\n post_title = self.request.get(\\\"post_title\\\")\\n post_content = self.request.get(\\\"post_content\\\")\\n param_list = dict(post_title=post_title, post_content=post_content)\\n any_error = False\\n\\n if not post_title:\\n param_list['title_error'] = \\\"Title is missing!\\\"\\n any_error = True\\n if not post_content:\\n param_list['content_error'] = \\\"Content is missing!\\\"\\n any_error = True\\n\\n if any_error:\\n self.render(\\\"blog/addpost.html\\\", **param_list)\\n else:\\n p = Post.add_post(post_title, post_content, self.user)\\n self.redirect('/blog/%s' % str(p.key().id()))\",\n \"def test_publish(self):\\n spiderman = User.objects.create(username=\\\"spiderman\\\", password=\\\"mary_jane\\\")\\n BlogEntry.objects.create(title=\\\"Spam Blog\\\", text=\\\"This is BlogSpam\\\", created_by=spiderman)\",\n \"def add_post(self, post: Post) -> None:\\n self.post_process.append(post)\",\n \"def create_task(author, title, text, **kwargs):\\n mc = MathContent(text=text)\\n mc.save()\\n task = Task(author=author, name=title, content=mc, **kwargs)\\n task.save()\\n return task\",\n \"def post(self):\\n post_id = int(self.request.get('post_id'))\\n post = Posts.get_by_id(post_id)\\n comment = self.request.get('comment')\\n submitter_id = self.get_active_user().key().id()\\n\\n if submitter_id:\\n comment = Comments(post_id=post_id, content=comment,\\n submitter_id=submitter_id)\\n comment.put()\\n self.redirect('/%s' % str(post.key().id()))\\n else:\\n self.error(403)\",\n \"def addPost(self,text,id,url,date):\\n self.topComments.append(Post(text,id,url,date))\\n return None\",\n \"def create_post(session: Session) -> Post:\\n try:\\n admin_user = session.query(User).filter(User.username == \\\"toddthebod\\\").first()\\n post = Post(\\n author_id=admin_user.id,\\n slug=\\\"fake-post-slug\\\",\\n title=\\\"Fake Post Title\\\",\\n summary=\\\"A fake post to have some fake comments.\\\",\\n feature_image=\\\"https://hackersandslackers-cdn.storage.googleapis.com/2021/01/logo-smaller@2x.png\\\",\\n body=\\\"Cheese slices monterey jack cauliflower cheese dolcelatte cheese and wine fromage frais rubber cheese gouda. Rubber cheese cheese and wine cheeseburger cheesy grin paneer paneer taleggio caerphilly. Edam mozzarella.\\\",\\n )\\n session.add(admin_user) # Add the user\\n session.commit() # Commit the change\\n LOGGER.success(f\\\"Created post {post} published by user {admin_user}\\\")\\n return post\\n except IntegrityError as e:\\n LOGGER.error(e.orig)\",\n \"def post(self, request, *args, **kwargs):\\n return self.create(request, *args, **kwargs)\",\n \"def post(self, request, *args, **kwargs):\\n return self.create(request, *args, **kwargs)\",\n \"def post(self, request, *args, **kwargs):\\n return self.create(request, *args, **kwargs)\"\n]","isConversational":false},"negative_scores":{"kind":"list like","value":["0.74787325","0.6533647","0.64678","0.6233577","0.62174255","0.61825395","0.6088378","0.60738593","0.60658085","0.6014791","0.6014791","0.6007368","0.5983524","0.5974287","0.5948988","0.5934463","0.58406353","0.58193","0.58059466","0.5795128","0.5768188","0.57461405","0.57359123","0.5726732","0.5726732","0.56976545","0.56976545","0.5682729","0.56766444","0.56527305","0.5648229","0.56427467","0.56363827","0.5630887","0.56005716","0.55996585","0.55952066","0.5591484","0.5573728","0.5563241","0.55614364","0.5558687","0.55541","0.5544374","0.5509563","0.5501533","0.5476885","0.54750174","0.5473601","0.54705703","0.5466731","0.5429865","0.5419775","0.5418963","0.54171854","0.536608","0.53570604","0.5349897","0.5349437","0.5349437","0.5345016","0.5341026","0.53318655","0.5330204","0.53271306","0.5319822","0.53113914","0.5309986","0.53073776","0.52960557","0.52804416","0.5275282","0.5274261","0.52580476","0.5245762","0.52437705","0.52408564","0.52384305","0.5235155","0.52267885","0.5223141","0.5219743","0.5216944","0.52111655","0.52086025","0.5206281","0.5201329","0.51986057","0.51910335","0.51891345","0.51866406","0.51802355","0.5150144","0.5149991","0.5144655","0.5138089","0.5134593","0.5126126","0.5126126","0.5126126"],"string":"[\n \"0.74787325\",\n \"0.6533647\",\n \"0.64678\",\n \"0.6233577\",\n \"0.62174255\",\n \"0.61825395\",\n \"0.6088378\",\n \"0.60738593\",\n \"0.60658085\",\n \"0.6014791\",\n \"0.6014791\",\n \"0.6007368\",\n \"0.5983524\",\n \"0.5974287\",\n \"0.5948988\",\n \"0.5934463\",\n \"0.58406353\",\n \"0.58193\",\n \"0.58059466\",\n \"0.5795128\",\n \"0.5768188\",\n \"0.57461405\",\n \"0.57359123\",\n \"0.5726732\",\n \"0.5726732\",\n \"0.56976545\",\n \"0.56976545\",\n \"0.5682729\",\n \"0.56766444\",\n \"0.56527305\",\n \"0.5648229\",\n \"0.56427467\",\n \"0.56363827\",\n \"0.5630887\",\n \"0.56005716\",\n \"0.55996585\",\n \"0.55952066\",\n \"0.5591484\",\n \"0.5573728\",\n \"0.5563241\",\n \"0.55614364\",\n \"0.5558687\",\n \"0.55541\",\n \"0.5544374\",\n \"0.5509563\",\n \"0.5501533\",\n \"0.5476885\",\n \"0.54750174\",\n \"0.5473601\",\n \"0.54705703\",\n \"0.5466731\",\n \"0.5429865\",\n \"0.5419775\",\n \"0.5418963\",\n \"0.54171854\",\n \"0.536608\",\n \"0.53570604\",\n \"0.5349897\",\n \"0.5349437\",\n \"0.5349437\",\n \"0.5345016\",\n \"0.5341026\",\n \"0.53318655\",\n \"0.5330204\",\n \"0.53271306\",\n \"0.5319822\",\n \"0.53113914\",\n \"0.5309986\",\n \"0.53073776\",\n \"0.52960557\",\n \"0.52804416\",\n \"0.5275282\",\n \"0.5274261\",\n \"0.52580476\",\n \"0.5245762\",\n \"0.52437705\",\n \"0.52408564\",\n \"0.52384305\",\n \"0.5235155\",\n \"0.52267885\",\n \"0.5223141\",\n \"0.5219743\",\n \"0.5216944\",\n \"0.52111655\",\n \"0.52086025\",\n \"0.5206281\",\n \"0.5201329\",\n \"0.51986057\",\n \"0.51910335\",\n \"0.51891345\",\n \"0.51866406\",\n \"0.51802355\",\n \"0.5150144\",\n \"0.5149991\",\n \"0.5144655\",\n \"0.5138089\",\n \"0.5134593\",\n \"0.5126126\",\n \"0.5126126\",\n \"0.5126126\"\n]","isConversational":false},"document_score":{"kind":"string","value":"0.6908637"},"document_rank":{"kind":"string","value":"1"}}},{"rowIdx":93282,"cells":{"query":{"kind":"string","value":"Assign tags to a comment by parsing the content."},"document":{"kind":"string","value":"def on_created_comment(sender, instance: dillo.models.comments.Comment, created, **kwargs):\n if not created:\n return\n # Extract tags and mentions from text and assign them to the Comment\n tags, mentions = extract_tags_and_mentions(instance.content)\n instance.tags.set(*tags)\n for mentioned_user in mentions:\n log.debug('Mentioning user %s in Comment %i' % (mentioned_user, instance.id))\n dillo.models.mixins.Mentions.objects.create(\n user=mentioned_user,\n content_object=instance,\n )\n # TODO(fsiddi) Generate activity about mention\n\n # Generate activity about comment creation\n log.debug('Generating activity about comment creation')\n verb = 'commented'\n if instance.parent_comment:\n verb = 'replied'\n action.send(instance.user, verb=verb, action_object=instance, target=instance.entity)\n log.debug('Set user %s as follower of own comment %i' % (instance.user, instance.id))\n follow(instance.user, instance, actor_only=False)"},"metadata":{"kind":"string","value":"{\n \"objective\": {\n \"self\": [],\n \"paired\": [],\n \"triplet\": [\n [\n \"query\",\n \"document\",\n \"negatives\"\n ]\n ]\n }\n}"},"negatives":{"kind":"list like","value":["def comment(self, content):\n pass","def __init__(self, node, declare):\n self.comment = getOptionalTag(node, \"comment\", [])\n if self.comment != []:\n self.comment = self.comment.replace(\"\\n\",\" \")\n while self.comment.find(\" \") != -1:\n self.comment = self.comment.replace(\" \", \" \")\n self.comment = self.comment.split(\" \")","def parse(self, content):\n pass","def parse_comment(self, node):\n\n data = []\n\n if node is not None:\n comment_id_pattern = re.compile('comment-(\\d+)')\n for comment_node in node.find_all('div', class_='comment'):\n item = {}\n item['is_deletable'] = False\n item['is_editable'] = False\n \n comment_id_result = comment_id_pattern.search(comment_node.get('id'))\n if comment_id_result:\n item['id'] = int(comment_id_result.group(1))\n \n comment_body_node = comment_node.find('div', class_='comment-body')\n if comment_body_node is not None:\n item['content'] = ''\n for p in comment_body_node.find_all(recursive=False):\n if 'class' in p.attrs and 'author' in p['class']:\n item['author'] = p.get_text()\n item['profile_url'] = self.get_link(p.get('href'))\n author_id = self._parse_user_id_from_url(item['profile_url'])\n if self.userId == author_id:\n item['is_deletable'] = True\n item['is_editable'] = True\n elif 'class' in p.attrs and 'age' in p['class']:\n item['date'] = p.abbr['title']\n item['date_ago'] = timeago.format(self._parse_datetime(item['date']), datetime.now(TIMEZONE))\n elif 'class' in p.attrs and 'edit' in p['class']:\n continue\n elif p.name == 'form':\n continue\n else:\n item['content'] += str(p)\n\n data.append(item)\n\n return data","def comments(self, comments):\n\n self.container['comments'] = comments","def fParseHTMLComments(self, match):\n before, commenttext, after = match.groups()\n commenttext = self.shelve(commenttext)\n return '<!--%s-->' % commenttext","def parseComments(data):\n global comments\n reviewBegins = '<div style=\"margin-left:0.5em;\">'\n reviewEnds = '<div style=\"padding-top: 10px; clear: both; width: 100%;\">'\n stars_line = 'margin-right:5px;'\n stars = re.compile('\\d+.\\d+ out of 5 stars')\n header_line = '<span style=\"vertical-align:middle;\"'\n helpful_line ='people found the following review helpful'\n helpful = re.compile('\\d+ of \\d+ people found the following review helpful')\n reviewText = '<span class=\"h3color tiny\">' # Actual review\n\n boundaries = commentsStartStopLineNmbr(data)\n for i in range(boundaries[0], boundaries[1] + 1):\n if reviewBegins in data[i]:\n curcomment = Comment()\n while reviewEnds not in data[i]:\n # Parse stars\n if stars_line in data[i]:\n stars_found = re.search(stars, data[i])\n if stars_found != None:\n curcomment.stars = stars_found.group()\n # Parse header\n elif header_line in data[i]:\n line = data[i]\n begin = line.find('<b>') + 3\n end = line.find('</b>')\n curcomment.header = line[begin : end]\n # Parse helpfulness\n elif helpful_line in data[i]:\n helpful_found = data[i].replace(\",\", \"\")\n helpful_found = re.search(helpful, helpful_found)\n if helpful_found != None:\n curcomment.helpful = helpful_found.group()\n # Parse body text\n elif reviewText in data[i]:\n i += 3\n if '<span class=\"small\"' in data[i]: # Yep, dirty trick :(\n i += 3\n data[i] = stripHtmlTags(data[i])\n curcomment.comment = re.sub(\"\\s+\", \" \", data[i])\n i += 1\n comments.append(curcomment.getonelinecomment())\n #comments.append(curcomment.__repr__())","def tag(comment):\n words = comment.split()\n doc = spacy.tokens.Doc(nlp.vocab, words)\n doc = nlp.tagger(doc)\n comment=\"\"\n for i in doc:\n comment = comment +\" \"+ i.text+\"/\"+i.tag_\n comment=comment[1:]\n return comment","def gettags(comment):\r\n\r\n tags = []\r\n\r\n tag = None\r\n datatype = None\r\n name = None\r\n tag_lineno = lineno = 0\r\n tag_text = []\r\n\r\n for line in comment.split('\\n'):\r\n line = line.strip()\r\n if line.startswith(\"@\"):\r\n tags.append((tag_lineno, tag, datatype, name, '\\n'.join(tag_text)))\r\n parts = line.split(None, 3)\r\n if len(parts) == 1:\r\n datatype = ''\r\n name = ''\r\n tag_text = []\r\n elif len(parts) == 2:\r\n datatype = parts[1]\r\n name = ''\r\n tag_text = []\r\n elif len(parts) == 3:\r\n datatype = parts[1]\r\n name = parts[2]\r\n tag_text = []\r\n elif len(parts) == 4:\r\n datatype = parts[1]\r\n name = parts[2]\r\n tag_text = [parts[3].lstrip()]\r\n tag = parts[0][1:]\r\n tag_lineno = lineno\r\n else:\r\n if line:\r\n tag_text.append(line)\r\n lineno += 1\r\n\r\n tags.append((tag_lineno, tag, datatype, name, '\\n'.join(tag_text)))\r\n\r\n return tags","def parseComment(self):\n libxml2mod.xmlParseComment(self._o)","def parse_comment(comment, postid):\n urls = get_links_from_body(comment.body)\n if urls:\n # Only insert comment into DB if it contains a link\n comid_db = db.insert('Comments',\n (None,\n postid,\n comment.id,\n comment.author,\n comment.body,\n comment.upvotes,\n comment.downvotes,\n comment.created_utc))\n for url in urls:\n parse_url(url, postid=postid, commentid=comid_db)\n # Recurse over child comments\n for child in comment.children:\n parse_comment(child, postid)","def parse_comment(value):\n try:\n return mark_safe(commentmarkup.parse(value))\n except ValueError:\n return value","def parse_code_comment(self, filepath):\n raise NotImplementedError('Not Implemented')","def visit_comment(self, node):\n self.printer.comment(node.xml_value)\n return","def _parse_comments(self, tokens: TokenIterator):\n metadata = {}\n while tokens.peek().type == 'COMMENT':\n comment = tokens.next().text\n while comment:\n comment, found, meta = comment.rpartition('::')\n if found:\n key, _, value = meta.partition(' ')\n metadata[key] = value.rstrip()\n return metadata","def test_comments(self):\n\n comment_str = \"# This is a comment\\n# This is another comment\"\n doc = parser.parse(comment_str)\n\n self.assertEqual(len(doc.children()), 2)","def parse_contents(self):\n self.parsed_contents = tokenize(self.contents)[0]","def Comment(self, comment):\n self.script.append(\"\")\n for i in comment.split(\"\\n\"):\n self.script.append(\"# \" + i)\n self.script.append(\"\")","def _parse_tags(self):\n tokens = self.tags_str[1:].split(\";\")\n self._tags = {\n k.strip(): v\n for token in tokens\n for k, v in [token.split(\"=\")]\n }","def parse_story_file(content):\n content_raw = content.split(\"@highlight\")[0]\n content = \" \".join(filter(None, [x.strip() for x in content_raw.split(\"\\n\")]))\n return content","def each_comment_from_post(post):\n # first yield the post text body, if any\n if post['text']:\n yield post['text']\n # then yield each comment\n for comment in post['comments']:\n yield comment['text']","def newComment(content):\n ret = libxml2mod.xmlNewComment(content)\n if ret is None:raise treeError('xmlNewComment() failed')\n return xmlNode(_obj=ret)","def make_comment(self, input, start, end, elements):\n return elements[1].text.strip('{}')","def set_comment(self, comment):\n self.comment_text = str(comment)","def unMarkupCommentsAndStrings(self, content):\n\n def replaceMarkups(match):\n groupdict = match.groupdict()\n if groupdict[\"str\"] is not None:\n return self.strings[int(match.group(\"str\"))]\n elif groupdict[\"comment\"] is not None:\n return self.comments[int(match.group(\"comment\"))]\n else:\n assert False\n\n unMarkedup = markups.sub(replaceMarkups, content)\n\n return unMarkedup","def set_comment(self, comment):\n\t\tself.comment_ = comment","def comments(self, comments):\n\n self._comments = comments","def comments(self, comments):\n\n self._comments = comments","def comments(self, comments):\n\n self._comments = comments","def comments(self, comments):\n\n self._comments = comments","def __init__(self, input_str):\n super(CommentParser, self).__init__()\n self.input = input_str\n self._i = iter(input_str.split(','))\n self._result = []","def proceed(self, source):\n\n for line in source:\n line = line.rstrip()\n if istag(line) and isinstance(maketag(line), Time): # here, handle Time tags\n self.time_annotate_date(maketag(line)) # annotate Date tag and queue Time tag\n if not default_options.mdjpp_time_propagate:\n continue\n # not a tag!\n if not istag(line) or isinstance(maketag(line), Time):\n if self.metastable_tag: # and len(line):\n # open metastable tag\n self.open_tag(self.metastable_tag)\n self.metastable_tag = None\n # print line\n if self.printable():\n self.printme_text(line)\n # tag!\n else:\n # make regular tag\n this_tag = maketag(line)\n # no metastable tag\n if not self.metastable_tag:\n # make new metastable tag\n self.metastable_tag = this_tag\n # does new tag close last tag?\n self.close_tag(this_tag)\n # metastable tag present\n else:\n # is metastable tag or this tag of Date type?\n if isinstance(self.metastable_tag, Date) or isinstance(this_tag, Date):\n # open metastable tag\n self.open_tag(self.metastable_tag)\n # recreate metastable tag with new tag\n self.metastable_tag = this_tag\n # does new tag close last tag?\n self.close_tag(this_tag)\n else:\n # add new tag to metastable tag\n self.metastable_tag.append(this_tag)","def _convert_comment(self, data):\n\n output = {}\n output['id'] = int(data['id'])\n output['author'] = data['user_display_name']\n output['profile_url'] = data['user_url']\n output['date'] = data['comment_added_at']\n output['date_ago'] = timeago.format(self._parse_datetime(data['comment_added_at']), datetime.now(TIMEZONE))\n output['content'] = self.convert_content(data['html'].replace('\\n', ''))\n output['is_deletable'] = data['is_deletable']\n output['is_editable'] = data['is_editable']\n\n return output","def comment(self, comment): # type: (str) -> None\n self._tmp_comment = comment","def assign_comments(self, tree, comments):\n comments = list(comments)\n comments.sort(key=lambda c: c.line)\n\n idx_by_line = {0: 0} # {line_no: comment_idx}\n\n for i, c in enumerate(comments):\n if c.line not in idx_by_line:\n idx_by_line[c.line] = i\n\n idx = []\n\n # convert comment tokens to strings, and remove any line breaks\n self.comments = [c.value.strip() for c in comments]\n last_comment_line = max(idx_by_line.keys())\n\n # make a list with an entry for each line\n # number associated with a comment list index\n\n for i in range(last_comment_line, 0, -1):\n if i in idx_by_line:\n # associate line with new comment\n idx.append(idx_by_line[i])\n else:\n # associate line with following comment\n idx.append(idx[-1])\n\n idx.append(0) # line numbers start from 1\n idx.reverse()\n self.idx = idx\n self._assign_comments(tree, 0)","def _readComments(self): \n self.NSCOML = nappy.utils.text_parser.readItemFromLine(self.file.readline(), int)\n self._readSpecialComments()\n self.NNCOML = nappy.utils.text_parser.readItemFromLine(self.file.readline(), int)\n self._readNormalComments()","def _handle_import(contents, use_tags, owner):\n \n lines = contents.decode(\"utf-8\").split(\"\\n\")\n \n title = re.compile(r\"<a.*?>(.+?)</a>\", re.I)\n url = re.compile(r\"\"\"<a.*href=['\"](.+?)['\"]\"\"\", re.I)\n tags = re.compile(r\"\"\"<a.*?tags=[\"'](.+?)[\"']\"\"\", re.I)\n addTime = re.compile(r\"\"\"<a.*?add_date=[\"'](\\d+?)[\"']\"\"\", re.I)\n \n for l in lines:\n if \"<a\" in l.lower() and \"</a>\" in l.lower():\n bookmark = {}\n \n bookmark[\"title\"] = title.search(l)\n if not bookmark[\"title\"]:\n continue\n bookmark[\"title\"] = _unescape(bookmark[\"title\"].group(1))\n \n bookmark[\"url\"] = url.search(l)\n if not bookmark[\"url\"]:\n continue\n bookmark[\"url\"] = _unescape(bookmark[\"url\"].group(1))\n \n bookmark[\"tags\"] = [];\n if use_tags:\n result = tags.search(l)\n if result:\n bookmark[\"tags\"] = map(_unescape, result.group(1).split(\",\"))\n \n bookmark[\"added\"] = addTime.search(l)\n if bookmark[\"added\"]:\n bookmark[\"added\"] = bookmark[\"added\"].group(1)\n \n if not Bookmark.objects.filter(owner=owner, url=bookmark[\"url\"]).exists():\n bm = Bookmark(owner=owner, url=bookmark[\"url\"], title=bookmark[\"title\"])\n \n bm.save()\n if bookmark[\"added\"]:\n bm.added = datetime.datetime.fromtimestamp(int(bookmark[\"added\"]))\n \n for t in bookmark[\"tags\"]:\n bm.tag(t)\n \n bm.save()\n bm.autotag_rules()","def __init__(self, tag_pairs, gcomment, movetext, score):\n self.tag_pairs = tag_pairs\n if type(gcomment) == str:\n self.comment = gcomment\n else:\n self.comment = ''\n self.movetext = movetext\n self.score = score","def comment_tokens(self, value):\n if not isinstance(value, str):\n raise TypeError('comment_tokens attribute must be a string.')\n self._comment_tokens = value","def _parse_comment(i, doc):\n\n if doc[i].strip() != \"/**\":\n raise ParseFailure(i, \"Expected beginning of block comment\")\n\n e = i + 1\n while e < len(doc) and doc[e].strip() != \"*/\":\n e += 1\n\n return e + 1, [x.rstrip() for x in doc[i + 1: e]]","def comment(self, comment) :\n\t\ttry :\n\t\t\tself._comment = comment\n\t\texcept Exception as e:\n\t\t\traise e","def make_parsed_comments(self):\n if not hasattr(self, 'separated_comments'):\n self.separated_comments = self.separate_comments()\n \n # build comments list of dictionaries, one dictionary for each article\n self.comments = []\n for self.separated_comment in self.separated_comments:\n try:\n comment_data = self.get_comment_data(self.separated_comment)\n self.comments.append(comment_data)\n except Exception as e:\n pass\n return self.comments","async def on_comment(self, comment: \"steam.Comment\") -> None:","def read_comment(comment):\n comment_dict = {}\n\n debug(\"parse tab in comment.\")\n comment_dict_from_tab, comment = parse_tab_in_comment(comment)\n debug(\"parsed dict: %s.\" % comment_dict_from_tab)\n comment_dict.update(comment_dict_from_tab)\n\n debug(\"parse space in comment.\")\n comment_dict_from_space, comment = parse_space_in_comment(comment)\n debug(\"parsed dict: %s.\" % comment_dict_from_space)\n comment_dict.update(comment_dict_from_space)\n\n debug(\"parse keyword in comment.\")\n comment_dict_from_keyword, comment = parse_keyword_in_comment(comment)\n debug(\"parsed dict: %s.\" % comment_dict_from_keyword)\n comment_dict.update(comment_dict_from_keyword)\n # keyword based separation.\n return comment_dict","def comment():","def start_comment(self, a, token):\n self.produce(\"comment\", token)\n logging.debug(\"in comment\")\n self.begin('comment')","def _parse(self, content):\n os.environ['ASTER_VERSION_DIR'] = self.dirn\n cfg = {}\n self._content = content\n for l in split_endlines(self._content):\n if not re.search('^[ ]*#', l):\n try:\n typ, nam, ver, val = l.split('|')\n #print '========>', typ, '//', nam, '//', ver, '//', val\n typ = re.sub('^[ ]*', '', re.sub('[ ]*$', '', typ)).strip()\n val = re.sub('^[ ]*', '', re.sub('[ ]*$', '', val)).strip()\n if val != '':\n val = osp.expandvars(val)\n if cfg.has_key(typ):\n cfg[typ].append(val)\n else:\n cfg[typ] = [val]\n except ValueError:\n pass\n return cfg","def __init__(self, *args, **kwargs):\n\n # Construct the base instance.\n super(FilterComment, self).__init__(*args, **kwargs)\n\n # Construct a regular expression tag evaluator.\n regextag = self.thistag.find('CommentRegex')\n if regextag == None:\n raise ValueError('Required tag missing: CommentRegex')\n self.regex = RegexTag(regextag)\n\n # Get the lock comment.\n self.comment = self.context.tokens['Comment']\n logger.debug('comment = \"{0}\"'.format(self.comment))","def comment(self, comment):\n\n self._comment = comment","def comment(self, comment):\n\n self._comment = comment","def comment(self, comment):\n\n self._comment = comment","def comment(self, comment):\n\n self._comment = comment","def comment(self, comment):\n\n self._comment = comment","def comment(self, comment):\n\n self._comment = comment","def comment(self, comment):\n\n self._comment = comment","def comment(self, comment):\n\n self._comment = comment","def comment(self, comment):\n\n self._comment = comment","def comment(self, comment):\n\n self._comment = comment","def taglinks(parser, token):\n bits = token.split_contents()\n\n if len(bits) != 3:\n raise template.TemplateSyntaxError(\n '%r tag requires 1 argument.' % bits[0])\n cmname = bits[1]\n\n try:\n srcm = ContentModule.objects.get(name=cmname)\n\n inputstrl = srcm.comment.strip().split('\\n')\n tuplelist = []\n for tagset in inputstrl:\n tuplelist.append((tagset.split('=')[0].strip(),\n tagset.split('=')[1].strip()))\n\n except:\n tuplelist = []\n urlbase = bits[2]\n\n return TaglinksNode(tuplelist, urlbase)","def set_comment(node_handle, comment):\n content_type = ContentType.objects.get_for_model(NodeHandle)\n object_pk = node_handle.pk\n user = get_user()\n site_id = django_settings.SITE_ID\n c = Comment(content_type=content_type, object_pk=object_pk, user=user, site_id=site_id, comment=comment)\n c.save()","def tag(self, text):\n\t\tpass","def parse_comment(self, comment):\n created_utc = datetime.utcfromtimestamp(comment.created_utc).isoformat()\n\n if comment.author is not None:\n author = comment.author.name\n else:\n author = \"\"\n\n com_obj = CommentMessage(\n comment.id,\n comment.link_id,\n self.subreddit,\n author,\n created_utc,\n comment.body,\n comment.score,\n )\n\n return com_obj","def visit_comment(self, node):\n self.body.append('\\n%% %s \\n' % COMMENT_RE_SUB('\\n% ', node.astext()))\n raise nodes.SkipNode","def comment(self, value: str):\n self._comment = value","def assign_tags_to_content(doc, style_tag):\n header_para = [] # list with headers and paragraphs\n header_dict = {}\n first = True # boolean operator for first header\n previous_s = {} # previous span\n\n for page in doc:\n blocks = page.getText(\"dict\")[\"blocks\"]\n i=0\n for b in blocks: # iterate through the text blocks\n if b['type'] == 0: # this block contains text\n\n # REMEMBER: multiple fonts and sizes are possible IN one block\n\n block_string = \"\" # text found in block\n for l in b[\"lines\"]: # iterate through the text lines\n for s in l[\"spans\"]: # iterate through the text spans\n if s['text'].strip(): # removing whitespaces:\n if first:\n previous_s = s\n first = False\n s_key = (int(s['text'].isupper()), int('bold' in s['font'].lower()), float(s['size']))\n block_string = style_tag[s_key] + s['text']\n else:\n s_key = (int(s['text'].isupper()), int('bold' in s['font'].lower()), float(s['size']))\n previous_key = (int(previous_s['text'].isupper()), int('bold' in previous_s['font'].lower()), float(previous_s['size']))\n if s_key == previous_key:\n\n if block_string and all((c == \"|\") for c in block_string):\n # block_string only contains pipes\n block_string = style_tag[s_key] + s['text']\n if block_string == \"\":\n # new block has started, so append size tag\n block_string = style_tag[s_key] + s['text']\n else: # in the same block, so concatenate strings\n block_string += \" \" + s['text']\n\n else:\n header_para.append(block_string)\n if block_string.startswith(\"<h\"):\n if style_tag[previous_key] in header_dict:\n header_dict[style_tag[previous_key]].append(block_string[block_string.index(\">\")+1:])\n else:\n header_dict[style_tag[previous_key]] = [block_string[block_string.index(\">\")+1:]]\n block_string = style_tag[s_key] + s['text']\n\n previous_s = s\n\n # new block started, indicating with a pipe\n block_string += \"|\"\n\n header_para.append(block_string)\n if block_string.startswith(\"<h\"):\n if style_tag[s_key] in header_dict:\n header_dict[style_tag[s_key]].append(block_string[block_string.index(\">\")+1:])\n else:\n header_dict[style_tag[s_key]] = [block_string[block_string.index(\">\")+1:]]\n return header_para, header_dict","def parse_comment(dom: str) -> Dict[str, str]:\n try:\n pstr = minidom.parseString(dom)\n # If it is not complete dom, mark it as \"partial\"\n # to concatinate to the next data.\n except xml.parsers.expat.ExpatError:\n resp = {\n \"tag\": \"partial\",\n \"data\": dom\n }\n return resp\n\n try:\n chat = pstr.getElementsByTagName(\"chat\")[0]\n except IndexError:\n try:\n _ = pstr.getElementsByTagName(\"thread\")[0]\n except IndexError:\n pass\n # Initial recieved data.\n else:\n tag = \"thread\"\n resp = {\n \"tag\": tag\n }\n else:\n tag = \"chat\"\n # KeyError occurs on official programs.\n try:\n commentno = chat.attributes[\"no\"].value\n except KeyError:\n commentno = \"-\"\n\n time = str(chat.attributes[\"date\"].value)\n userid = chat.attributes[\"user_id\"].value\n\n # Free members don't have premium key.\n try:\n premium = chat.attributes[\"premium\"].value\n except KeyError:\n premium = \"0\"\n\n # ID users don't have anonymity key.\n try:\n anonymity = chat.attributes[\"anonymity\"].value\n except KeyError:\n anonymity = \"0\"\n\n # Owner don't has locale key.\n try:\n locale = chat.attributes[\"locale\"].value\n except KeyError:\n locale = \"ja-jp\"\n\n # If score is 0, dont have score key.\n try:\n score = chat.attributes[\"score\"].value\n except KeyError:\n score = \"0\"\n\n # Comment content.\n content = chat.firstChild.data\n\n resp = {\n \"tag\": tag,\n \"no\": commentno,\n \"time\": time,\n \"id\": userid,\n \"premium\": premium,\n \"anonymity\": anonymity,\n \"locale\": locale,\n \"score\": score,\n \"content\": content\n }\n\n return resp","def tokenize_raw(text):\n while text:\n curr_string = ''\n if text[0] == '[':\n if ']' not in text:\n raise Exception('Found non-terminated tag: '+text[0:100])\n curr_string = text[:text.find(']')+1]\n node_type = 'Tag'\n if text[0] == '!':\n match = re.match('!\\\\w+!', text)\n if match:\n curr_string = match.group()\n node_type = 'Tag'\n if not curr_string:\n if '[' in text:\n curr_string = text[:text.find('[')]\n else:\n curr_string = text\n #check for\n match = re.search('!\\\\w+!', text)\n if match:\n curr_string = curr_string[:match.start()]\n node_type = 'Comment'\n text = text[len(curr_string):]\n yield node_type, curr_string","def start_comments(self, a, text):\n logging.debug(\"in comments\" + text)\n logging.debug(self.bracket_nesting_level)\n\n if self.bracket_nesting_level == 0:\n self.bracket_nesting_level += 1\n logging.debug(\"start state comments\")\n if Str(text) == languages_comment_start1:\n self.begin(\"comments\")\n else:\n self.begin(\"comments2\")\n elif self.bracket_nesting_level == 1:\n self.bracket_nesting_level -= 1\n self.begin('')","def zap_tag_contents(self, tag):\n if hasattr(tag, 'contents'):\n content_length = len(tag.contents)\n\n for x in range(0, content_length):\n if isinstance(tag.contents[x], bs4.element.Comment):\n self.zap_tag_contents(tag.contents[x])\n elif isinstance(tag.contents[x], bs4.element.NavigableString):\n unicode_entry = self.zapper.kill_gremlins(tag.contents[x])\n unicode_entry = unidecode(unicode_entry)\n tag.contents[x].replace_with(unicode_entry)\n elif isinstance(tag.contents[x], bs4.element.Tag):\n self.zap_tag_contents(tag.contents[x])","def __init__(self, _id, _name, _line):\n self.type = 'comment'\n self.id = _id\n self.name = _name\n self.value = _line","def visit(self, node):\n self.body.append('<div id =\"discourse-comments\"></div>')","def get_comments_content_object(parser, token):\n keywords = token.contents.split()\n if len(keywords) != 5:\n raise template.TemplateSyntaxError(\n \"'%s' tag takes exactly 2 arguments\" % (keywords[0],))\n if keywords[1] != 'for':\n raise template.TemplateSyntaxError(\n \"first argument to '%s' tag must be 'for'\" % (keywords[0],))\n if keywords[3] != 'as':\n raise template.TemplateSyntaxError(\n \"first argument to '%s' tag must be 'as'\" % (keywords[0],))\n return GetCommentsContentObject(keywords[2], keywords[4])","def _parse(self, line):\n\n while True:\n sign, before, after = _splitbycharset(line, self._checkstr)\n\n # End of line\n if not sign:\n if self._inside_quote:\n self._buffer.append(before)\n elif self._after_equal_sign:\n self._text(\"\".join(self._buffer) + before.strip())\n self._closetag()\n self._after_equal_sign = False\n elif not self._inside_attrib:\n self._buffer.append(before)\n elif before.strip():\n self._error(SYNTAX_ERROR, (self._currline, self._currline))\n break\n\n # Special character is escaped\n elif before.endswith(\"\\\\\") and not before.endswith(\"\\\\\\\\\"):\n self._buffer.append(before + sign)\n\n # Equal sign\n elif sign == \"=\":\n # Ignore if followed by \"{\" (DFTB+ compatibility)\n if after.lstrip().startswith(\"{\"): # _oldbefore may already contain the tagname, if the # tagname was followed by an attribute -> append\n self._oldbefore += before\n else:\n self._hsdattrib[common.HSD_ATTRIB_EQUAL] = True\n self._starttag(before, False)\n self._after_equal_sign = True\n\n # Opening tag by curly brace\n elif sign == \"{\":\n #self._has_child = True\n self._hsdattrib[common.HSD_ATTRIB_EQUAL] = False\n self._starttag(before, self._after_equal_sign)\n self._buffer = []\n self._after_equal_sign = False\n\n # Closing tag by curly brace\n elif sign == \"}\":\n self._text(\"\".join(self._buffer) + before)\n self._buffer = []\n # If 'test { a = 12 }' occurs, curly brace closes two tags\n if self._after_equal_sign:\n self._after_equal_sign = False\n self._closetag()\n self._closetag()\n\n # Closing tag by semicolon\n elif sign == \";\" and self._after_equal_sign:\n self._after_equal_sign = False\n self._text(before)\n self._closetag()\n\n # Comment line\n elif sign == \"#\":\n self._buffer.append(before)\n after = \"\"\n\n # Opening attribute specification\n elif sign == \"[\":\n if \"\".join(self._buffer).strip():\n self._error(SYNTAX_ERROR, (self._currline, self._currline))\n self._oldbefore = before\n self._buffer = []\n self._inside_attrib = True\n self._opened_tags.append((\"[\", self._currline, None, None, None))\n self._checkstr = _ATTRIB_SPECIALS\n\n # Closing attribute specification\n elif sign == \"]\":\n value = \"\".join(self._buffer) + before\n self._attrib = value.strip()\n self._inside_attrib = False\n self._buffer = []\n self._opened_tags.pop()\n self._checkstr = _GENERAL_SPECIALS\n\n # Quoting strings\n elif sign in (\"'\", '\"'):\n if self._inside_quote:\n self._checkstr = self._oldcheckstr\n self._inside_quote = False\n self._buffer.append(before + sign)\n self._opened_tags.pop()\n else:\n self._oldcheckstr = self._checkstr\n self._checkstr = sign\n self._inside_quote = True\n self._buffer.append(before + sign)\n self._opened_tags.append(('\"', self._currline, None, None, None))\n\n # Interrupt\n elif sign == \"<\" and not self._after_equal_sign:\n txtinc = after.startswith(\"<<\")\n hsdinc = after.startswith(\"<+\")\n if txtinc:\n self._text(\"\".join(self._buffer) + before)\n self._buffer = []\n self._eventhandler.add_text(self._include_txt(after[2:]))\n break\n if hsdinc:\n self._include_hsd(after[2:])\n break\n self._buffer.append(before + sign)\n\n else:\n self._error(SYNTAX_ERROR, (self._currline, self._currline))\n\n line = after","def tag_comment(request, tag_id, comment_id, error='', message=''):\n try:\n tag = Tag.objects.get(id=tag_id)\n except:\n error += 'Couldn\\'t retrieve tag ' + tag_id + '.'\n try:\n comment = Comment.objects.get(id=comment_id)\n except:\n error += 'Couldn\\'t retrieve comment ' + comment_id + '.'\n\n if tag in comment.tags.all():\n error += 'This comment has already been tagged.'\n\n if not error:\n try:\n comment.tags.add(tag)\n message += 'Tagged comment ' + str(comment.id) + ' with ' + tag.name + '.'\n except:\n error += 'Couldn\\'t tag comment.'\n return render_to_response('feedback/taglist.html', {\n 'object': comment, 'type': 'comment', 'error': error,\n }, context_instance=RequestContext(request))","def parse_keyword_in_comment(comment):\n debug(\"into keyword.\")\n debug(comment)\n comment_str = ('\\n'.join(comment)).lower()\n comment_dict = {}\n\n more_keywords = vim.eval(\"g:BHKeywords\")\n keywords = copy(KEYWORDS)\n keywords.update(more_keywords)\n\n keywords = sorted(\n [keyword for keyword in keywords if comment_str.find(keyword.lower()) != -1],\n lambda x, y: cmp(comment_str.index(x.lower()), comment_str.index(y.lower()))\n )\n debug(\"Found keywords in the original header: %s\" % keywords)\n if not keywords:\n return {}, comment\n\n def get_content(comment, keyword, next_keyword=None):\n debug(\"into get content\")\n for index, line in enumerate(comment):\n if line.lower().find(keyword.lower()) != -1:\n break\n\n if next_keyword:\n for jndex, line in enumerate(comment):\n if line.lower().find(next_keyword.lower()) != -1:\n break\n else:\n jndex = len(comment)\n # content in comment[index:jndex-1], further purify it.\n re.I = True\n line_re = re.compile(r\"(?P<key>%s)?:?\\s*(?P<value>.*)\" % keyword)\n re.I = False\n\n first_line = line_re.match(comment[index]).group('value')\n return '\\n'.join([first_line] + comment[index+1:jndex-1])\n\n for i, keyword in enumerate(keywords):\n if i == len(keywords) - 1:\n comment_dict[keyword] = get_content(comment, keyword)\n else:\n comment_dict[keyword] = get_content(comment, keyword, keywords[i+1])\n return comment_dict, comment","def comment_content(c):\n content = str(c)[4:-3]\n return content.strip()","def comment(self, comment):\n\n self.logger.debug(\"In 'comment' setter.\")\n\n self._comment = comment","def addContent(text):","def __init__(self, position, line, original_indentation, type = None):\n super(Comment, self).__init__(position)\n self.text_lines = [line]\n self.type = type\n self.original_indentation = original_indentation\n self.priority = 0","def on_issue_comment(self, payload):\n pass","def visit_BlockComment(self, node):\n\n self.statement(node, '# ', node.text)","def comment(self, comment: str):\n\n self._comment = comment","def comment(self, comment: str):\n\n self._comment = comment","def comment_to_object(self, comment, post_author_id=None):\n # the message_tags field is different in comment vs post. in post, it's a\n # dict of lists, in comment it's just a list. so, convert it to post style\n # here before running post_to_object().\n comment = dict(comment)\n comment['message_tags'] = {'1': comment.get('message_tags', [])}\n\n obj = self.post_to_object(comment)\n if not obj:\n return obj\n\n obj['objectType'] = 'comment'\n\n match = self.COMMENT_ID_RE.match(comment.get('id', ''))\n if match:\n post_author, post_id, comment_id = match.groups()\n obj['url'] = self.comment_url(post_id, comment_id,\n post_author_id=post_author_id)\n obj['inReplyTo'] = [{'id': self.tag_uri(post_id)}]\n\n return self.postprocess_object(obj)","def parse_and_handle_description_tags(rr, text):\n return parse_description_tags(text).handle_tags(\n TagHandler(rr).tag_handlers).replace('\\n', '<br>').replace('\\r', '')","def comment(self, comment_id):\r\n return GistComment(self, comment_id)","def extract_from_comment(self, comment):\n comment = re.sub(r'\\s+', ' ', comment.strip().strip(\"/*\").strip())\n if len(comment) == 0:\n return set(), set()\n domain_terms, code_elements = self.extract_from_sentence(comment)\n return domain_terms, code_elements","def set_comment_editor_value(self, comment_id, new_body):\r\n self._find_within(\"#comment_{} .wmd-input\".format(comment_id)).fill(new_body)","def test_comments(self):\n fp = FilePath(self.mktemp())\n fp.setContent('something\\n#commented\\ncool')\n self.assertEqual(list(inventoryReader(fp.path)), ['something', 'cool'])","def tokenization(self, content: str, doc_id: str):\n self.metadata[doc_id] = dict()\n tokens = list()\n lines = content.splitlines()\n for i in range(200 if self.name == 'Novels' else 0, len(lines)):\n if self.name == 'HillaryEmails' or (lines[i] == '' and lines[i-1] != ''):\n words = 0\n self.metadata[doc_id]['Content'] = str()\n for j in range(i, len(lines)):\n line = lines[j]\n if line:\n words += len(line.split())\n self.metadata[doc_id]['Content'] += line + ''\n if words >= 75:\n self.metadata[doc_id]['Content'] += '...'\n break\n break\n keys = ['Title', 'Author', 'Release Date', 'Language', 'Character set encoding']\n for i, line in enumerate(lines):\n if self.name == 'Novels':\n if i < 30:\n for j in range(len(keys)):\n if keys[j] in line:\n self.metadata[doc_id][keys[j]] = line.strip().replace(keys[j]+': ', '')\n token = line.split() # default split by whitespace\n tokens.extend(zip(token, len(token) * [doc_id]))\n return tokens","def get_caption_comments(content):\n\n if not content.startswith('## fig:'):\n return None, None\n\n content = content.splitlines()\n\n id = content[0].strip('## ')\n\n caption = []\n for line in content[1:]:\n if not line.startswith('# ') or line.startswith('##'):\n break\n else:\n caption.append(line.lstrip('# ').rstrip())\n\n # add \" around the caption. TODO: consider doing this upstream\n # in pandoc-attributes\n caption = '\"' + ' '.join(caption) + '\"'\n return id, caption","def commentinator(coms, ctype, btag, tag):\n # Construct the message\n tag = btag + tag\n \n if ctype.lower() == 'notes':\n coms.notes.append(tag)\n elif ctype.lower() == 'warning':\n coms.warnings.append(tag)\n elif ctype.lower() == 'error':\n coms.errors.append(tag)\n elif ctype.lower() == 'tip':\n coms.tips.append(tag)\n \n return coms","def __init__(self, text, tag, start ,end):\n\n self.text = six.text_type(text)\n self.tag = copy.copy(tag)\n self.end = end\n self.start = start","def addComment(self, page, content):\n return self.pm_getSpaceManager().addComment(self._unbox(page), content)","def do_get_user_comments(parser, token):\r\n error_message = \"%r tag must be of format {%% %r for OBJECT as CONTEXT_VARIABLE %%}\" % (token.contents.split()[0], token.contents.split()[0])\r\n try:\r\n split = token.split_contents()\r\n except ValueError:\r\n raise template.TemplateSyntaxError, error_message\r\n if len(split) != 5:\r\n raise template.TemplateSyntaxError, error_message\r\n return UserCommentsNode(split[2], split[4])","def auto_transform_markup(comment):\r\n try:\r\n from django.utils.html import escape\r\n from threadedcomments.models import MARKDOWN, TEXTILE, REST, PLAINTEXT\r\n if comment.markup == MARKDOWN:\r\n from django.contrib.markup.templatetags.markup import markdown\r\n return markdown(comment.comment)\r\n elif comment.markup == TEXTILE:\r\n from django.contrib.markup.templatetags.markup import textile\r\n return textile(comment.comment)\r\n elif comment.markup == REST:\r\n from django.contrib.markup.templatetags.markup import restructuredtext\r\n return restructuredtext(comment.comment)\r\n# elif comment.markup == HTML:\r\n# return mark_safe(force_unicode(comment.comment))\r\n elif comment.markup == PLAINTEXT:\r\n return escape(comment.comment)\r\n except ImportError:\r\n # Not marking safe, in case tag fails and users input malicious code.\r\n return force_unicode(comment.comment)","def get_comment_data(self, comment):\n # remove double spaces but not triple ones; we use triple spaces to split commenter and parent_commenter\n pattern = '(?<! ) {2}(?! )'\n comment = re.sub(pattern, ' ', comment).strip() # also strip leading and trailing spaces\n\n # get names\n ix = re.search('•', comment).span()[-1]\n names = [x.strip() for x in (comment[:ix]).strip().strip('•').split(' ')]\n try:\n commenter, parent_commenter = names\n except:\n commenter, parent_commenter = names[0], ''\n\n # handle deleted comments\n pattern = 'This comment was deleted.−+−+'\n commenter = commenter.replace(pattern, '').strip()\n \n # get post and upvotes\n comment_upvotes = comment[ix:].split('ago')[-1].strip(' ')\n ix = re.search('(see more)\\w+', comment_upvotes) # redefine ix as index that separates post message from post upvotes\n clean_comment, upvotes = comment_upvotes[:ix.span()[0]], comment_upvotes[ix.span()[0]:].replace('see more', '')\n\n # build dictionary\n d = dict(zip( ['commenter', 'parent_commenter', 'comment', 'upvotes']\n , [commenter, parent_commenter.strip(), clean_comment.strip(), upvotes.strip()]))\n\n return d","def _replace_comment(element):\n replacement = \"\"\n matched = element.group()\n\n # Ensure that only the comment part is being replaced\n if not matched.startswith(\"/\"):\n replacement += matched[0]\n matched = matched[1:]\n\n count = matched.count(\"\\n\")\n\n # Add empty spaces with the size of the content if the size\n # must be kept.\n if keep_content_size:\n _buffer = len(matched) - count\n replacement += \" \" * _buffer + \"\\n\" * count\n\n # Otherwise simply keep the number of lines\n else:\n replacement += \"\\n\" * count\n\n return replacement","def _parse(self):\n soup = BS(self._current_html, 'lxml')\n for item in soup.select('div.c'):\n temp = {}\n # main content\n ctt = item.select('span.ctt')\n if not ctt:\n continue\n weibo_body = item.select('div')\n if len(weibo_body) > 1:\n temp['content'] = weibo_body[0].text\n btn_group = weibo_body[1].text\n else:\n temp['content'] = weibo_body[0].select('span.ctt')[0].text\n btn_group = weibo_body[0].text\n temp['is_repost'] = True if REPO_TEST_PATTERN.match(\n temp['content']) else False\n try:\n temp['like_num'] = LIKE_NUM_PATTERN.findall(btn_group)[0]\n temp['cmt_num'] = COMMENT_NUM_PATTERN.findall(btn_group)[0]\n temp['repo_num'] = REPO_NUM_PATTERN.findall(btn_group)[0]\n except Exception:\n pass\n cmt = item.select('.cmt')\n # visibility\n if cmt:\n try:\n temp['visibility'] = VISIBILITY_PATTERN.findall(\n cmt[0].text)[0]\n except Exception:\n pass\n\n # img in main content\n img = item.select('div a img')\n img_src = img[0].attrs['src'] if img else None\n temp['img_src'] = img_src\n LOGGER.debug('img_src: {}'.format(img_src))\n # time & source device\n ct = item.select('span.ct')\n if ct:\n ct = ct[0]\n text = ct.text\n reg_result = TIME_PATTERN.findall(text)[0]\n\n temp['time'] = ar(\n '{}年{}'.format(self._current_year, reg_result[0]),\n DATE_FMTS[0]\n ).naive if reg_result[0] else ar(\n reg_result[1], DATE_FMTS[1]\n ).naive\n temp['source'] = SOURCE_DEVICE_PATTERN.findall(text)[0]\n self._post_item = Post(**temp)\n self._attachment_item = Attachment(\n uri=img_src, post=self._post_item)\n self._store()","def comment(self, comment):\n self.logger.debug(\"In 'comment' setter.\")\n\n if len(comment) > 512:\n raise Exception(\"Comment is too long, must be less than 512 characters.\")\n\n self._comment = comment"],"string":"[\n \"def comment(self, content):\\n pass\",\n \"def __init__(self, node, declare):\\n self.comment = getOptionalTag(node, \\\"comment\\\", [])\\n if self.comment != []:\\n self.comment = self.comment.replace(\\\"\\\\n\\\",\\\" \\\")\\n while self.comment.find(\\\" \\\") != -1:\\n self.comment = self.comment.replace(\\\" \\\", \\\" \\\")\\n self.comment = self.comment.split(\\\" \\\")\",\n \"def parse(self, content):\\n pass\",\n \"def parse_comment(self, node):\\n\\n data = []\\n\\n if node is not None:\\n comment_id_pattern = re.compile('comment-(\\\\d+)')\\n for comment_node in node.find_all('div', class_='comment'):\\n item = {}\\n item['is_deletable'] = False\\n item['is_editable'] = False\\n \\n comment_id_result = comment_id_pattern.search(comment_node.get('id'))\\n if comment_id_result:\\n item['id'] = int(comment_id_result.group(1))\\n \\n comment_body_node = comment_node.find('div', class_='comment-body')\\n if comment_body_node is not None:\\n item['content'] = ''\\n for p in comment_body_node.find_all(recursive=False):\\n if 'class' in p.attrs and 'author' in p['class']:\\n item['author'] = p.get_text()\\n item['profile_url'] = self.get_link(p.get('href'))\\n author_id = self._parse_user_id_from_url(item['profile_url'])\\n if self.userId == author_id:\\n item['is_deletable'] = True\\n item['is_editable'] = True\\n elif 'class' in p.attrs and 'age' in p['class']:\\n item['date'] = p.abbr['title']\\n item['date_ago'] = timeago.format(self._parse_datetime(item['date']), datetime.now(TIMEZONE))\\n elif 'class' in p.attrs and 'edit' in p['class']:\\n continue\\n elif p.name == 'form':\\n continue\\n else:\\n item['content'] += str(p)\\n\\n data.append(item)\\n\\n return data\",\n \"def comments(self, comments):\\n\\n self.container['comments'] = comments\",\n \"def fParseHTMLComments(self, match):\\n before, commenttext, after = match.groups()\\n commenttext = self.shelve(commenttext)\\n return '<!--%s-->' % commenttext\",\n \"def parseComments(data):\\n global comments\\n reviewBegins = '<div style=\\\"margin-left:0.5em;\\\">'\\n reviewEnds = '<div style=\\\"padding-top: 10px; clear: both; width: 100%;\\\">'\\n stars_line = 'margin-right:5px;'\\n stars = re.compile('\\\\d+.\\\\d+ out of 5 stars')\\n header_line = '<span style=\\\"vertical-align:middle;\\\"'\\n helpful_line ='people found the following review helpful'\\n helpful = re.compile('\\\\d+ of \\\\d+ people found the following review helpful')\\n reviewText = '<span class=\\\"h3color tiny\\\">' # Actual review\\n\\n boundaries = commentsStartStopLineNmbr(data)\\n for i in range(boundaries[0], boundaries[1] + 1):\\n if reviewBegins in data[i]:\\n curcomment = Comment()\\n while reviewEnds not in data[i]:\\n # Parse stars\\n if stars_line in data[i]:\\n stars_found = re.search(stars, data[i])\\n if stars_found != None:\\n curcomment.stars = stars_found.group()\\n # Parse header\\n elif header_line in data[i]:\\n line = data[i]\\n begin = line.find('<b>') + 3\\n end = line.find('</b>')\\n curcomment.header = line[begin : end]\\n # Parse helpfulness\\n elif helpful_line in data[i]:\\n helpful_found = data[i].replace(\\\",\\\", \\\"\\\")\\n helpful_found = re.search(helpful, helpful_found)\\n if helpful_found != None:\\n curcomment.helpful = helpful_found.group()\\n # Parse body text\\n elif reviewText in data[i]:\\n i += 3\\n if '<span class=\\\"small\\\"' in data[i]: # Yep, dirty trick :(\\n i += 3\\n data[i] = stripHtmlTags(data[i])\\n curcomment.comment = re.sub(\\\"\\\\s+\\\", \\\" \\\", data[i])\\n i += 1\\n comments.append(curcomment.getonelinecomment())\\n #comments.append(curcomment.__repr__())\",\n \"def tag(comment):\\n words = comment.split()\\n doc = spacy.tokens.Doc(nlp.vocab, words)\\n doc = nlp.tagger(doc)\\n comment=\\\"\\\"\\n for i in doc:\\n comment = comment +\\\" \\\"+ i.text+\\\"/\\\"+i.tag_\\n comment=comment[1:]\\n return comment\",\n \"def gettags(comment):\\r\\n\\r\\n tags = []\\r\\n\\r\\n tag = None\\r\\n datatype = None\\r\\n name = None\\r\\n tag_lineno = lineno = 0\\r\\n tag_text = []\\r\\n\\r\\n for line in comment.split('\\\\n'):\\r\\n line = line.strip()\\r\\n if line.startswith(\\\"@\\\"):\\r\\n tags.append((tag_lineno, tag, datatype, name, '\\\\n'.join(tag_text)))\\r\\n parts = line.split(None, 3)\\r\\n if len(parts) == 1:\\r\\n datatype = ''\\r\\n name = ''\\r\\n tag_text = []\\r\\n elif len(parts) == 2:\\r\\n datatype = parts[1]\\r\\n name = ''\\r\\n tag_text = []\\r\\n elif len(parts) == 3:\\r\\n datatype = parts[1]\\r\\n name = parts[2]\\r\\n tag_text = []\\r\\n elif len(parts) == 4:\\r\\n datatype = parts[1]\\r\\n name = parts[2]\\r\\n tag_text = [parts[3].lstrip()]\\r\\n tag = parts[0][1:]\\r\\n tag_lineno = lineno\\r\\n else:\\r\\n if line:\\r\\n tag_text.append(line)\\r\\n lineno += 1\\r\\n\\r\\n tags.append((tag_lineno, tag, datatype, name, '\\\\n'.join(tag_text)))\\r\\n\\r\\n return tags\",\n \"def parseComment(self):\\n libxml2mod.xmlParseComment(self._o)\",\n \"def parse_comment(comment, postid):\\n urls = get_links_from_body(comment.body)\\n if urls:\\n # Only insert comment into DB if it contains a link\\n comid_db = db.insert('Comments',\\n (None,\\n postid,\\n comment.id,\\n comment.author,\\n comment.body,\\n comment.upvotes,\\n comment.downvotes,\\n comment.created_utc))\\n for url in urls:\\n parse_url(url, postid=postid, commentid=comid_db)\\n # Recurse over child comments\\n for child in comment.children:\\n parse_comment(child, postid)\",\n \"def parse_comment(value):\\n try:\\n return mark_safe(commentmarkup.parse(value))\\n except ValueError:\\n return value\",\n \"def parse_code_comment(self, filepath):\\n raise NotImplementedError('Not Implemented')\",\n \"def visit_comment(self, node):\\n self.printer.comment(node.xml_value)\\n return\",\n \"def _parse_comments(self, tokens: TokenIterator):\\n metadata = {}\\n while tokens.peek().type == 'COMMENT':\\n comment = tokens.next().text\\n while comment:\\n comment, found, meta = comment.rpartition('::')\\n if found:\\n key, _, value = meta.partition(' ')\\n metadata[key] = value.rstrip()\\n return metadata\",\n \"def test_comments(self):\\n\\n comment_str = \\\"# This is a comment\\\\n# This is another comment\\\"\\n doc = parser.parse(comment_str)\\n\\n self.assertEqual(len(doc.children()), 2)\",\n \"def parse_contents(self):\\n self.parsed_contents = tokenize(self.contents)[0]\",\n \"def Comment(self, comment):\\n self.script.append(\\\"\\\")\\n for i in comment.split(\\\"\\\\n\\\"):\\n self.script.append(\\\"# \\\" + i)\\n self.script.append(\\\"\\\")\",\n \"def _parse_tags(self):\\n tokens = self.tags_str[1:].split(\\\";\\\")\\n self._tags = {\\n k.strip(): v\\n for token in tokens\\n for k, v in [token.split(\\\"=\\\")]\\n }\",\n \"def parse_story_file(content):\\n content_raw = content.split(\\\"@highlight\\\")[0]\\n content = \\\" \\\".join(filter(None, [x.strip() for x in content_raw.split(\\\"\\\\n\\\")]))\\n return content\",\n \"def each_comment_from_post(post):\\n # first yield the post text body, if any\\n if post['text']:\\n yield post['text']\\n # then yield each comment\\n for comment in post['comments']:\\n yield comment['text']\",\n \"def newComment(content):\\n ret = libxml2mod.xmlNewComment(content)\\n if ret is None:raise treeError('xmlNewComment() failed')\\n return xmlNode(_obj=ret)\",\n \"def make_comment(self, input, start, end, elements):\\n return elements[1].text.strip('{}')\",\n \"def set_comment(self, comment):\\n self.comment_text = str(comment)\",\n \"def unMarkupCommentsAndStrings(self, content):\\n\\n def replaceMarkups(match):\\n groupdict = match.groupdict()\\n if groupdict[\\\"str\\\"] is not None:\\n return self.strings[int(match.group(\\\"str\\\"))]\\n elif groupdict[\\\"comment\\\"] is not None:\\n return self.comments[int(match.group(\\\"comment\\\"))]\\n else:\\n assert False\\n\\n unMarkedup = markups.sub(replaceMarkups, content)\\n\\n return unMarkedup\",\n \"def set_comment(self, comment):\\n\\t\\tself.comment_ = comment\",\n \"def comments(self, comments):\\n\\n self._comments = comments\",\n \"def comments(self, comments):\\n\\n self._comments = comments\",\n \"def comments(self, comments):\\n\\n self._comments = comments\",\n \"def comments(self, comments):\\n\\n self._comments = comments\",\n \"def __init__(self, input_str):\\n super(CommentParser, self).__init__()\\n self.input = input_str\\n self._i = iter(input_str.split(','))\\n self._result = []\",\n \"def proceed(self, source):\\n\\n for line in source:\\n line = line.rstrip()\\n if istag(line) and isinstance(maketag(line), Time): # here, handle Time tags\\n self.time_annotate_date(maketag(line)) # annotate Date tag and queue Time tag\\n if not default_options.mdjpp_time_propagate:\\n continue\\n # not a tag!\\n if not istag(line) or isinstance(maketag(line), Time):\\n if self.metastable_tag: # and len(line):\\n # open metastable tag\\n self.open_tag(self.metastable_tag)\\n self.metastable_tag = None\\n # print line\\n if self.printable():\\n self.printme_text(line)\\n # tag!\\n else:\\n # make regular tag\\n this_tag = maketag(line)\\n # no metastable tag\\n if not self.metastable_tag:\\n # make new metastable tag\\n self.metastable_tag = this_tag\\n # does new tag close last tag?\\n self.close_tag(this_tag)\\n # metastable tag present\\n else:\\n # is metastable tag or this tag of Date type?\\n if isinstance(self.metastable_tag, Date) or isinstance(this_tag, Date):\\n # open metastable tag\\n self.open_tag(self.metastable_tag)\\n # recreate metastable tag with new tag\\n self.metastable_tag = this_tag\\n # does new tag close last tag?\\n self.close_tag(this_tag)\\n else:\\n # add new tag to metastable tag\\n self.metastable_tag.append(this_tag)\",\n \"def _convert_comment(self, data):\\n\\n output = {}\\n output['id'] = int(data['id'])\\n output['author'] = data['user_display_name']\\n output['profile_url'] = data['user_url']\\n output['date'] = data['comment_added_at']\\n output['date_ago'] = timeago.format(self._parse_datetime(data['comment_added_at']), datetime.now(TIMEZONE))\\n output['content'] = self.convert_content(data['html'].replace('\\\\n', ''))\\n output['is_deletable'] = data['is_deletable']\\n output['is_editable'] = data['is_editable']\\n\\n return output\",\n \"def comment(self, comment): # type: (str) -> None\\n self._tmp_comment = comment\",\n \"def assign_comments(self, tree, comments):\\n comments = list(comments)\\n comments.sort(key=lambda c: c.line)\\n\\n idx_by_line = {0: 0} # {line_no: comment_idx}\\n\\n for i, c in enumerate(comments):\\n if c.line not in idx_by_line:\\n idx_by_line[c.line] = i\\n\\n idx = []\\n\\n # convert comment tokens to strings, and remove any line breaks\\n self.comments = [c.value.strip() for c in comments]\\n last_comment_line = max(idx_by_line.keys())\\n\\n # make a list with an entry for each line\\n # number associated with a comment list index\\n\\n for i in range(last_comment_line, 0, -1):\\n if i in idx_by_line:\\n # associate line with new comment\\n idx.append(idx_by_line[i])\\n else:\\n # associate line with following comment\\n idx.append(idx[-1])\\n\\n idx.append(0) # line numbers start from 1\\n idx.reverse()\\n self.idx = idx\\n self._assign_comments(tree, 0)\",\n \"def _readComments(self): \\n self.NSCOML = nappy.utils.text_parser.readItemFromLine(self.file.readline(), int)\\n self._readSpecialComments()\\n self.NNCOML = nappy.utils.text_parser.readItemFromLine(self.file.readline(), int)\\n self._readNormalComments()\",\n \"def _handle_import(contents, use_tags, owner):\\n \\n lines = contents.decode(\\\"utf-8\\\").split(\\\"\\\\n\\\")\\n \\n title = re.compile(r\\\"<a.*?>(.+?)</a>\\\", re.I)\\n url = re.compile(r\\\"\\\"\\\"<a.*href=['\\\"](.+?)['\\\"]\\\"\\\"\\\", re.I)\\n tags = re.compile(r\\\"\\\"\\\"<a.*?tags=[\\\"'](.+?)[\\\"']\\\"\\\"\\\", re.I)\\n addTime = re.compile(r\\\"\\\"\\\"<a.*?add_date=[\\\"'](\\\\d+?)[\\\"']\\\"\\\"\\\", re.I)\\n \\n for l in lines:\\n if \\\"<a\\\" in l.lower() and \\\"</a>\\\" in l.lower():\\n bookmark = {}\\n \\n bookmark[\\\"title\\\"] = title.search(l)\\n if not bookmark[\\\"title\\\"]:\\n continue\\n bookmark[\\\"title\\\"] = _unescape(bookmark[\\\"title\\\"].group(1))\\n \\n bookmark[\\\"url\\\"] = url.search(l)\\n if not bookmark[\\\"url\\\"]:\\n continue\\n bookmark[\\\"url\\\"] = _unescape(bookmark[\\\"url\\\"].group(1))\\n \\n bookmark[\\\"tags\\\"] = [];\\n if use_tags:\\n result = tags.search(l)\\n if result:\\n bookmark[\\\"tags\\\"] = map(_unescape, result.group(1).split(\\\",\\\"))\\n \\n bookmark[\\\"added\\\"] = addTime.search(l)\\n if bookmark[\\\"added\\\"]:\\n bookmark[\\\"added\\\"] = bookmark[\\\"added\\\"].group(1)\\n \\n if not Bookmark.objects.filter(owner=owner, url=bookmark[\\\"url\\\"]).exists():\\n bm = Bookmark(owner=owner, url=bookmark[\\\"url\\\"], title=bookmark[\\\"title\\\"])\\n \\n bm.save()\\n if bookmark[\\\"added\\\"]:\\n bm.added = datetime.datetime.fromtimestamp(int(bookmark[\\\"added\\\"]))\\n \\n for t in bookmark[\\\"tags\\\"]:\\n bm.tag(t)\\n \\n bm.save()\\n bm.autotag_rules()\",\n \"def __init__(self, tag_pairs, gcomment, movetext, score):\\n self.tag_pairs = tag_pairs\\n if type(gcomment) == str:\\n self.comment = gcomment\\n else:\\n self.comment = ''\\n self.movetext = movetext\\n self.score = score\",\n \"def comment_tokens(self, value):\\n if not isinstance(value, str):\\n raise TypeError('comment_tokens attribute must be a string.')\\n self._comment_tokens = value\",\n \"def _parse_comment(i, doc):\\n\\n if doc[i].strip() != \\\"/**\\\":\\n raise ParseFailure(i, \\\"Expected beginning of block comment\\\")\\n\\n e = i + 1\\n while e < len(doc) and doc[e].strip() != \\\"*/\\\":\\n e += 1\\n\\n return e + 1, [x.rstrip() for x in doc[i + 1: e]]\",\n \"def comment(self, comment) :\\n\\t\\ttry :\\n\\t\\t\\tself._comment = comment\\n\\t\\texcept Exception as e:\\n\\t\\t\\traise e\",\n \"def make_parsed_comments(self):\\n if not hasattr(self, 'separated_comments'):\\n self.separated_comments = self.separate_comments()\\n \\n # build comments list of dictionaries, one dictionary for each article\\n self.comments = []\\n for self.separated_comment in self.separated_comments:\\n try:\\n comment_data = self.get_comment_data(self.separated_comment)\\n self.comments.append(comment_data)\\n except Exception as e:\\n pass\\n return self.comments\",\n \"async def on_comment(self, comment: \\\"steam.Comment\\\") -> None:\",\n \"def read_comment(comment):\\n comment_dict = {}\\n\\n debug(\\\"parse tab in comment.\\\")\\n comment_dict_from_tab, comment = parse_tab_in_comment(comment)\\n debug(\\\"parsed dict: %s.\\\" % comment_dict_from_tab)\\n comment_dict.update(comment_dict_from_tab)\\n\\n debug(\\\"parse space in comment.\\\")\\n comment_dict_from_space, comment = parse_space_in_comment(comment)\\n debug(\\\"parsed dict: %s.\\\" % comment_dict_from_space)\\n comment_dict.update(comment_dict_from_space)\\n\\n debug(\\\"parse keyword in comment.\\\")\\n comment_dict_from_keyword, comment = parse_keyword_in_comment(comment)\\n debug(\\\"parsed dict: %s.\\\" % comment_dict_from_keyword)\\n comment_dict.update(comment_dict_from_keyword)\\n # keyword based separation.\\n return comment_dict\",\n \"def comment():\",\n \"def start_comment(self, a, token):\\n self.produce(\\\"comment\\\", token)\\n logging.debug(\\\"in comment\\\")\\n self.begin('comment')\",\n \"def _parse(self, content):\\n os.environ['ASTER_VERSION_DIR'] = self.dirn\\n cfg = {}\\n self._content = content\\n for l in split_endlines(self._content):\\n if not re.search('^[ ]*#', l):\\n try:\\n typ, nam, ver, val = l.split('|')\\n #print '========>', typ, '//', nam, '//', ver, '//', val\\n typ = re.sub('^[ ]*', '', re.sub('[ ]*$', '', typ)).strip()\\n val = re.sub('^[ ]*', '', re.sub('[ ]*$', '', val)).strip()\\n if val != '':\\n val = osp.expandvars(val)\\n if cfg.has_key(typ):\\n cfg[typ].append(val)\\n else:\\n cfg[typ] = [val]\\n except ValueError:\\n pass\\n return cfg\",\n \"def __init__(self, *args, **kwargs):\\n\\n # Construct the base instance.\\n super(FilterComment, self).__init__(*args, **kwargs)\\n\\n # Construct a regular expression tag evaluator.\\n regextag = self.thistag.find('CommentRegex')\\n if regextag == None:\\n raise ValueError('Required tag missing: CommentRegex')\\n self.regex = RegexTag(regextag)\\n\\n # Get the lock comment.\\n self.comment = self.context.tokens['Comment']\\n logger.debug('comment = \\\"{0}\\\"'.format(self.comment))\",\n \"def comment(self, comment):\\n\\n self._comment = comment\",\n \"def comment(self, comment):\\n\\n self._comment = comment\",\n \"def comment(self, comment):\\n\\n self._comment = comment\",\n \"def comment(self, comment):\\n\\n self._comment = comment\",\n \"def comment(self, comment):\\n\\n self._comment = comment\",\n \"def comment(self, comment):\\n\\n self._comment = comment\",\n \"def comment(self, comment):\\n\\n self._comment = comment\",\n \"def comment(self, comment):\\n\\n self._comment = comment\",\n \"def comment(self, comment):\\n\\n self._comment = comment\",\n \"def comment(self, comment):\\n\\n self._comment = comment\",\n \"def taglinks(parser, token):\\n bits = token.split_contents()\\n\\n if len(bits) != 3:\\n raise template.TemplateSyntaxError(\\n '%r tag requires 1 argument.' % bits[0])\\n cmname = bits[1]\\n\\n try:\\n srcm = ContentModule.objects.get(name=cmname)\\n\\n inputstrl = srcm.comment.strip().split('\\\\n')\\n tuplelist = []\\n for tagset in inputstrl:\\n tuplelist.append((tagset.split('=')[0].strip(),\\n tagset.split('=')[1].strip()))\\n\\n except:\\n tuplelist = []\\n urlbase = bits[2]\\n\\n return TaglinksNode(tuplelist, urlbase)\",\n \"def set_comment(node_handle, comment):\\n content_type = ContentType.objects.get_for_model(NodeHandle)\\n object_pk = node_handle.pk\\n user = get_user()\\n site_id = django_settings.SITE_ID\\n c = Comment(content_type=content_type, object_pk=object_pk, user=user, site_id=site_id, comment=comment)\\n c.save()\",\n \"def tag(self, text):\\n\\t\\tpass\",\n \"def parse_comment(self, comment):\\n created_utc = datetime.utcfromtimestamp(comment.created_utc).isoformat()\\n\\n if comment.author is not None:\\n author = comment.author.name\\n else:\\n author = \\\"\\\"\\n\\n com_obj = CommentMessage(\\n comment.id,\\n comment.link_id,\\n self.subreddit,\\n author,\\n created_utc,\\n comment.body,\\n comment.score,\\n )\\n\\n return com_obj\",\n \"def visit_comment(self, node):\\n self.body.append('\\\\n%% %s \\\\n' % COMMENT_RE_SUB('\\\\n% ', node.astext()))\\n raise nodes.SkipNode\",\n \"def comment(self, value: str):\\n self._comment = value\",\n \"def assign_tags_to_content(doc, style_tag):\\n header_para = [] # list with headers and paragraphs\\n header_dict = {}\\n first = True # boolean operator for first header\\n previous_s = {} # previous span\\n\\n for page in doc:\\n blocks = page.getText(\\\"dict\\\")[\\\"blocks\\\"]\\n i=0\\n for b in blocks: # iterate through the text blocks\\n if b['type'] == 0: # this block contains text\\n\\n # REMEMBER: multiple fonts and sizes are possible IN one block\\n\\n block_string = \\\"\\\" # text found in block\\n for l in b[\\\"lines\\\"]: # iterate through the text lines\\n for s in l[\\\"spans\\\"]: # iterate through the text spans\\n if s['text'].strip(): # removing whitespaces:\\n if first:\\n previous_s = s\\n first = False\\n s_key = (int(s['text'].isupper()), int('bold' in s['font'].lower()), float(s['size']))\\n block_string = style_tag[s_key] + s['text']\\n else:\\n s_key = (int(s['text'].isupper()), int('bold' in s['font'].lower()), float(s['size']))\\n previous_key = (int(previous_s['text'].isupper()), int('bold' in previous_s['font'].lower()), float(previous_s['size']))\\n if s_key == previous_key:\\n\\n if block_string and all((c == \\\"|\\\") for c in block_string):\\n # block_string only contains pipes\\n block_string = style_tag[s_key] + s['text']\\n if block_string == \\\"\\\":\\n # new block has started, so append size tag\\n block_string = style_tag[s_key] + s['text']\\n else: # in the same block, so concatenate strings\\n block_string += \\\" \\\" + s['text']\\n\\n else:\\n header_para.append(block_string)\\n if block_string.startswith(\\\"<h\\\"):\\n if style_tag[previous_key] in header_dict:\\n header_dict[style_tag[previous_key]].append(block_string[block_string.index(\\\">\\\")+1:])\\n else:\\n header_dict[style_tag[previous_key]] = [block_string[block_string.index(\\\">\\\")+1:]]\\n block_string = style_tag[s_key] + s['text']\\n\\n previous_s = s\\n\\n # new block started, indicating with a pipe\\n block_string += \\\"|\\\"\\n\\n header_para.append(block_string)\\n if block_string.startswith(\\\"<h\\\"):\\n if style_tag[s_key] in header_dict:\\n header_dict[style_tag[s_key]].append(block_string[block_string.index(\\\">\\\")+1:])\\n else:\\n header_dict[style_tag[s_key]] = [block_string[block_string.index(\\\">\\\")+1:]]\\n return header_para, header_dict\",\n \"def parse_comment(dom: str) -> Dict[str, str]:\\n try:\\n pstr = minidom.parseString(dom)\\n # If it is not complete dom, mark it as \\\"partial\\\"\\n # to concatinate to the next data.\\n except xml.parsers.expat.ExpatError:\\n resp = {\\n \\\"tag\\\": \\\"partial\\\",\\n \\\"data\\\": dom\\n }\\n return resp\\n\\n try:\\n chat = pstr.getElementsByTagName(\\\"chat\\\")[0]\\n except IndexError:\\n try:\\n _ = pstr.getElementsByTagName(\\\"thread\\\")[0]\\n except IndexError:\\n pass\\n # Initial recieved data.\\n else:\\n tag = \\\"thread\\\"\\n resp = {\\n \\\"tag\\\": tag\\n }\\n else:\\n tag = \\\"chat\\\"\\n # KeyError occurs on official programs.\\n try:\\n commentno = chat.attributes[\\\"no\\\"].value\\n except KeyError:\\n commentno = \\\"-\\\"\\n\\n time = str(chat.attributes[\\\"date\\\"].value)\\n userid = chat.attributes[\\\"user_id\\\"].value\\n\\n # Free members don't have premium key.\\n try:\\n premium = chat.attributes[\\\"premium\\\"].value\\n except KeyError:\\n premium = \\\"0\\\"\\n\\n # ID users don't have anonymity key.\\n try:\\n anonymity = chat.attributes[\\\"anonymity\\\"].value\\n except KeyError:\\n anonymity = \\\"0\\\"\\n\\n # Owner don't has locale key.\\n try:\\n locale = chat.attributes[\\\"locale\\\"].value\\n except KeyError:\\n locale = \\\"ja-jp\\\"\\n\\n # If score is 0, dont have score key.\\n try:\\n score = chat.attributes[\\\"score\\\"].value\\n except KeyError:\\n score = \\\"0\\\"\\n\\n # Comment content.\\n content = chat.firstChild.data\\n\\n resp = {\\n \\\"tag\\\": tag,\\n \\\"no\\\": commentno,\\n \\\"time\\\": time,\\n \\\"id\\\": userid,\\n \\\"premium\\\": premium,\\n \\\"anonymity\\\": anonymity,\\n \\\"locale\\\": locale,\\n \\\"score\\\": score,\\n \\\"content\\\": content\\n }\\n\\n return resp\",\n \"def tokenize_raw(text):\\n while text:\\n curr_string = ''\\n if text[0] == '[':\\n if ']' not in text:\\n raise Exception('Found non-terminated tag: '+text[0:100])\\n curr_string = text[:text.find(']')+1]\\n node_type = 'Tag'\\n if text[0] == '!':\\n match = re.match('!\\\\\\\\w+!', text)\\n if match:\\n curr_string = match.group()\\n node_type = 'Tag'\\n if not curr_string:\\n if '[' in text:\\n curr_string = text[:text.find('[')]\\n else:\\n curr_string = text\\n #check for\\n match = re.search('!\\\\\\\\w+!', text)\\n if match:\\n curr_string = curr_string[:match.start()]\\n node_type = 'Comment'\\n text = text[len(curr_string):]\\n yield node_type, curr_string\",\n \"def start_comments(self, a, text):\\n logging.debug(\\\"in comments\\\" + text)\\n logging.debug(self.bracket_nesting_level)\\n\\n if self.bracket_nesting_level == 0:\\n self.bracket_nesting_level += 1\\n logging.debug(\\\"start state comments\\\")\\n if Str(text) == languages_comment_start1:\\n self.begin(\\\"comments\\\")\\n else:\\n self.begin(\\\"comments2\\\")\\n elif self.bracket_nesting_level == 1:\\n self.bracket_nesting_level -= 1\\n self.begin('')\",\n \"def zap_tag_contents(self, tag):\\n if hasattr(tag, 'contents'):\\n content_length = len(tag.contents)\\n\\n for x in range(0, content_length):\\n if isinstance(tag.contents[x], bs4.element.Comment):\\n self.zap_tag_contents(tag.contents[x])\\n elif isinstance(tag.contents[x], bs4.element.NavigableString):\\n unicode_entry = self.zapper.kill_gremlins(tag.contents[x])\\n unicode_entry = unidecode(unicode_entry)\\n tag.contents[x].replace_with(unicode_entry)\\n elif isinstance(tag.contents[x], bs4.element.Tag):\\n self.zap_tag_contents(tag.contents[x])\",\n \"def __init__(self, _id, _name, _line):\\n self.type = 'comment'\\n self.id = _id\\n self.name = _name\\n self.value = _line\",\n \"def visit(self, node):\\n self.body.append('<div id =\\\"discourse-comments\\\"></div>')\",\n \"def get_comments_content_object(parser, token):\\n keywords = token.contents.split()\\n if len(keywords) != 5:\\n raise template.TemplateSyntaxError(\\n \\\"'%s' tag takes exactly 2 arguments\\\" % (keywords[0],))\\n if keywords[1] != 'for':\\n raise template.TemplateSyntaxError(\\n \\\"first argument to '%s' tag must be 'for'\\\" % (keywords[0],))\\n if keywords[3] != 'as':\\n raise template.TemplateSyntaxError(\\n \\\"first argument to '%s' tag must be 'as'\\\" % (keywords[0],))\\n return GetCommentsContentObject(keywords[2], keywords[4])\",\n \"def _parse(self, line):\\n\\n while True:\\n sign, before, after = _splitbycharset(line, self._checkstr)\\n\\n # End of line\\n if not sign:\\n if self._inside_quote:\\n self._buffer.append(before)\\n elif self._after_equal_sign:\\n self._text(\\\"\\\".join(self._buffer) + before.strip())\\n self._closetag()\\n self._after_equal_sign = False\\n elif not self._inside_attrib:\\n self._buffer.append(before)\\n elif before.strip():\\n self._error(SYNTAX_ERROR, (self._currline, self._currline))\\n break\\n\\n # Special character is escaped\\n elif before.endswith(\\\"\\\\\\\\\\\") and not before.endswith(\\\"\\\\\\\\\\\\\\\\\\\"):\\n self._buffer.append(before + sign)\\n\\n # Equal sign\\n elif sign == \\\"=\\\":\\n # Ignore if followed by \\\"{\\\" (DFTB+ compatibility)\\n if after.lstrip().startswith(\\\"{\\\"): # _oldbefore may already contain the tagname, if the # tagname was followed by an attribute -> append\\n self._oldbefore += before\\n else:\\n self._hsdattrib[common.HSD_ATTRIB_EQUAL] = True\\n self._starttag(before, False)\\n self._after_equal_sign = True\\n\\n # Opening tag by curly brace\\n elif sign == \\\"{\\\":\\n #self._has_child = True\\n self._hsdattrib[common.HSD_ATTRIB_EQUAL] = False\\n self._starttag(before, self._after_equal_sign)\\n self._buffer = []\\n self._after_equal_sign = False\\n\\n # Closing tag by curly brace\\n elif sign == \\\"}\\\":\\n self._text(\\\"\\\".join(self._buffer) + before)\\n self._buffer = []\\n # If 'test { a = 12 }' occurs, curly brace closes two tags\\n if self._after_equal_sign:\\n self._after_equal_sign = False\\n self._closetag()\\n self._closetag()\\n\\n # Closing tag by semicolon\\n elif sign == \\\";\\\" and self._after_equal_sign:\\n self._after_equal_sign = False\\n self._text(before)\\n self._closetag()\\n\\n # Comment line\\n elif sign == \\\"#\\\":\\n self._buffer.append(before)\\n after = \\\"\\\"\\n\\n # Opening attribute specification\\n elif sign == \\\"[\\\":\\n if \\\"\\\".join(self._buffer).strip():\\n self._error(SYNTAX_ERROR, (self._currline, self._currline))\\n self._oldbefore = before\\n self._buffer = []\\n self._inside_attrib = True\\n self._opened_tags.append((\\\"[\\\", self._currline, None, None, None))\\n self._checkstr = _ATTRIB_SPECIALS\\n\\n # Closing attribute specification\\n elif sign == \\\"]\\\":\\n value = \\\"\\\".join(self._buffer) + before\\n self._attrib = value.strip()\\n self._inside_attrib = False\\n self._buffer = []\\n self._opened_tags.pop()\\n self._checkstr = _GENERAL_SPECIALS\\n\\n # Quoting strings\\n elif sign in (\\\"'\\\", '\\\"'):\\n if self._inside_quote:\\n self._checkstr = self._oldcheckstr\\n self._inside_quote = False\\n self._buffer.append(before + sign)\\n self._opened_tags.pop()\\n else:\\n self._oldcheckstr = self._checkstr\\n self._checkstr = sign\\n self._inside_quote = True\\n self._buffer.append(before + sign)\\n self._opened_tags.append(('\\\"', self._currline, None, None, None))\\n\\n # Interrupt\\n elif sign == \\\"<\\\" and not self._after_equal_sign:\\n txtinc = after.startswith(\\\"<<\\\")\\n hsdinc = after.startswith(\\\"<+\\\")\\n if txtinc:\\n self._text(\\\"\\\".join(self._buffer) + before)\\n self._buffer = []\\n self._eventhandler.add_text(self._include_txt(after[2:]))\\n break\\n if hsdinc:\\n self._include_hsd(after[2:])\\n break\\n self._buffer.append(before + sign)\\n\\n else:\\n self._error(SYNTAX_ERROR, (self._currline, self._currline))\\n\\n line = after\",\n \"def tag_comment(request, tag_id, comment_id, error='', message=''):\\n try:\\n tag = Tag.objects.get(id=tag_id)\\n except:\\n error += 'Couldn\\\\'t retrieve tag ' + tag_id + '.'\\n try:\\n comment = Comment.objects.get(id=comment_id)\\n except:\\n error += 'Couldn\\\\'t retrieve comment ' + comment_id + '.'\\n\\n if tag in comment.tags.all():\\n error += 'This comment has already been tagged.'\\n\\n if not error:\\n try:\\n comment.tags.add(tag)\\n message += 'Tagged comment ' + str(comment.id) + ' with ' + tag.name + '.'\\n except:\\n error += 'Couldn\\\\'t tag comment.'\\n return render_to_response('feedback/taglist.html', {\\n 'object': comment, 'type': 'comment', 'error': error,\\n }, context_instance=RequestContext(request))\",\n \"def parse_keyword_in_comment(comment):\\n debug(\\\"into keyword.\\\")\\n debug(comment)\\n comment_str = ('\\\\n'.join(comment)).lower()\\n comment_dict = {}\\n\\n more_keywords = vim.eval(\\\"g:BHKeywords\\\")\\n keywords = copy(KEYWORDS)\\n keywords.update(more_keywords)\\n\\n keywords = sorted(\\n [keyword for keyword in keywords if comment_str.find(keyword.lower()) != -1],\\n lambda x, y: cmp(comment_str.index(x.lower()), comment_str.index(y.lower()))\\n )\\n debug(\\\"Found keywords in the original header: %s\\\" % keywords)\\n if not keywords:\\n return {}, comment\\n\\n def get_content(comment, keyword, next_keyword=None):\\n debug(\\\"into get content\\\")\\n for index, line in enumerate(comment):\\n if line.lower().find(keyword.lower()) != -1:\\n break\\n\\n if next_keyword:\\n for jndex, line in enumerate(comment):\\n if line.lower().find(next_keyword.lower()) != -1:\\n break\\n else:\\n jndex = len(comment)\\n # content in comment[index:jndex-1], further purify it.\\n re.I = True\\n line_re = re.compile(r\\\"(?P<key>%s)?:?\\\\s*(?P<value>.*)\\\" % keyword)\\n re.I = False\\n\\n first_line = line_re.match(comment[index]).group('value')\\n return '\\\\n'.join([first_line] + comment[index+1:jndex-1])\\n\\n for i, keyword in enumerate(keywords):\\n if i == len(keywords) - 1:\\n comment_dict[keyword] = get_content(comment, keyword)\\n else:\\n comment_dict[keyword] = get_content(comment, keyword, keywords[i+1])\\n return comment_dict, comment\",\n \"def comment_content(c):\\n content = str(c)[4:-3]\\n return content.strip()\",\n \"def comment(self, comment):\\n\\n self.logger.debug(\\\"In 'comment' setter.\\\")\\n\\n self._comment = comment\",\n \"def addContent(text):\",\n \"def __init__(self, position, line, original_indentation, type = None):\\n super(Comment, self).__init__(position)\\n self.text_lines = [line]\\n self.type = type\\n self.original_indentation = original_indentation\\n self.priority = 0\",\n \"def on_issue_comment(self, payload):\\n pass\",\n \"def visit_BlockComment(self, node):\\n\\n self.statement(node, '# ', node.text)\",\n \"def comment(self, comment: str):\\n\\n self._comment = comment\",\n \"def comment(self, comment: str):\\n\\n self._comment = comment\",\n \"def comment_to_object(self, comment, post_author_id=None):\\n # the message_tags field is different in comment vs post. in post, it's a\\n # dict of lists, in comment it's just a list. so, convert it to post style\\n # here before running post_to_object().\\n comment = dict(comment)\\n comment['message_tags'] = {'1': comment.get('message_tags', [])}\\n\\n obj = self.post_to_object(comment)\\n if not obj:\\n return obj\\n\\n obj['objectType'] = 'comment'\\n\\n match = self.COMMENT_ID_RE.match(comment.get('id', ''))\\n if match:\\n post_author, post_id, comment_id = match.groups()\\n obj['url'] = self.comment_url(post_id, comment_id,\\n post_author_id=post_author_id)\\n obj['inReplyTo'] = [{'id': self.tag_uri(post_id)}]\\n\\n return self.postprocess_object(obj)\",\n \"def parse_and_handle_description_tags(rr, text):\\n return parse_description_tags(text).handle_tags(\\n TagHandler(rr).tag_handlers).replace('\\\\n', '<br>').replace('\\\\r', '')\",\n \"def comment(self, comment_id):\\r\\n return GistComment(self, comment_id)\",\n \"def extract_from_comment(self, comment):\\n comment = re.sub(r'\\\\s+', ' ', comment.strip().strip(\\\"/*\\\").strip())\\n if len(comment) == 0:\\n return set(), set()\\n domain_terms, code_elements = self.extract_from_sentence(comment)\\n return domain_terms, code_elements\",\n \"def set_comment_editor_value(self, comment_id, new_body):\\r\\n self._find_within(\\\"#comment_{} .wmd-input\\\".format(comment_id)).fill(new_body)\",\n \"def test_comments(self):\\n fp = FilePath(self.mktemp())\\n fp.setContent('something\\\\n#commented\\\\ncool')\\n self.assertEqual(list(inventoryReader(fp.path)), ['something', 'cool'])\",\n \"def tokenization(self, content: str, doc_id: str):\\n self.metadata[doc_id] = dict()\\n tokens = list()\\n lines = content.splitlines()\\n for i in range(200 if self.name == 'Novels' else 0, len(lines)):\\n if self.name == 'HillaryEmails' or (lines[i] == '' and lines[i-1] != ''):\\n words = 0\\n self.metadata[doc_id]['Content'] = str()\\n for j in range(i, len(lines)):\\n line = lines[j]\\n if line:\\n words += len(line.split())\\n self.metadata[doc_id]['Content'] += line + ''\\n if words >= 75:\\n self.metadata[doc_id]['Content'] += '...'\\n break\\n break\\n keys = ['Title', 'Author', 'Release Date', 'Language', 'Character set encoding']\\n for i, line in enumerate(lines):\\n if self.name == 'Novels':\\n if i < 30:\\n for j in range(len(keys)):\\n if keys[j] in line:\\n self.metadata[doc_id][keys[j]] = line.strip().replace(keys[j]+': ', '')\\n token = line.split() # default split by whitespace\\n tokens.extend(zip(token, len(token) * [doc_id]))\\n return tokens\",\n \"def get_caption_comments(content):\\n\\n if not content.startswith('## fig:'):\\n return None, None\\n\\n content = content.splitlines()\\n\\n id = content[0].strip('## ')\\n\\n caption = []\\n for line in content[1:]:\\n if not line.startswith('# ') or line.startswith('##'):\\n break\\n else:\\n caption.append(line.lstrip('# ').rstrip())\\n\\n # add \\\" around the caption. TODO: consider doing this upstream\\n # in pandoc-attributes\\n caption = '\\\"' + ' '.join(caption) + '\\\"'\\n return id, caption\",\n \"def commentinator(coms, ctype, btag, tag):\\n # Construct the message\\n tag = btag + tag\\n \\n if ctype.lower() == 'notes':\\n coms.notes.append(tag)\\n elif ctype.lower() == 'warning':\\n coms.warnings.append(tag)\\n elif ctype.lower() == 'error':\\n coms.errors.append(tag)\\n elif ctype.lower() == 'tip':\\n coms.tips.append(tag)\\n \\n return coms\",\n \"def __init__(self, text, tag, start ,end):\\n\\n self.text = six.text_type(text)\\n self.tag = copy.copy(tag)\\n self.end = end\\n self.start = start\",\n \"def addComment(self, page, content):\\n return self.pm_getSpaceManager().addComment(self._unbox(page), content)\",\n \"def do_get_user_comments(parser, token):\\r\\n error_message = \\\"%r tag must be of format {%% %r for OBJECT as CONTEXT_VARIABLE %%}\\\" % (token.contents.split()[0], token.contents.split()[0])\\r\\n try:\\r\\n split = token.split_contents()\\r\\n except ValueError:\\r\\n raise template.TemplateSyntaxError, error_message\\r\\n if len(split) != 5:\\r\\n raise template.TemplateSyntaxError, error_message\\r\\n return UserCommentsNode(split[2], split[4])\",\n \"def auto_transform_markup(comment):\\r\\n try:\\r\\n from django.utils.html import escape\\r\\n from threadedcomments.models import MARKDOWN, TEXTILE, REST, PLAINTEXT\\r\\n if comment.markup == MARKDOWN:\\r\\n from django.contrib.markup.templatetags.markup import markdown\\r\\n return markdown(comment.comment)\\r\\n elif comment.markup == TEXTILE:\\r\\n from django.contrib.markup.templatetags.markup import textile\\r\\n return textile(comment.comment)\\r\\n elif comment.markup == REST:\\r\\n from django.contrib.markup.templatetags.markup import restructuredtext\\r\\n return restructuredtext(comment.comment)\\r\\n# elif comment.markup == HTML:\\r\\n# return mark_safe(force_unicode(comment.comment))\\r\\n elif comment.markup == PLAINTEXT:\\r\\n return escape(comment.comment)\\r\\n except ImportError:\\r\\n # Not marking safe, in case tag fails and users input malicious code.\\r\\n return force_unicode(comment.comment)\",\n \"def get_comment_data(self, comment):\\n # remove double spaces but not triple ones; we use triple spaces to split commenter and parent_commenter\\n pattern = '(?<! ) {2}(?! )'\\n comment = re.sub(pattern, ' ', comment).strip() # also strip leading and trailing spaces\\n\\n # get names\\n ix = re.search('•', comment).span()[-1]\\n names = [x.strip() for x in (comment[:ix]).strip().strip('•').split(' ')]\\n try:\\n commenter, parent_commenter = names\\n except:\\n commenter, parent_commenter = names[0], ''\\n\\n # handle deleted comments\\n pattern = 'This comment was deleted.−+−+'\\n commenter = commenter.replace(pattern, '').strip()\\n \\n # get post and upvotes\\n comment_upvotes = comment[ix:].split('ago')[-1].strip(' ')\\n ix = re.search('(see more)\\\\w+', comment_upvotes) # redefine ix as index that separates post message from post upvotes\\n clean_comment, upvotes = comment_upvotes[:ix.span()[0]], comment_upvotes[ix.span()[0]:].replace('see more', '')\\n\\n # build dictionary\\n d = dict(zip( ['commenter', 'parent_commenter', 'comment', 'upvotes']\\n , [commenter, parent_commenter.strip(), clean_comment.strip(), upvotes.strip()]))\\n\\n return d\",\n \"def _replace_comment(element):\\n replacement = \\\"\\\"\\n matched = element.group()\\n\\n # Ensure that only the comment part is being replaced\\n if not matched.startswith(\\\"/\\\"):\\n replacement += matched[0]\\n matched = matched[1:]\\n\\n count = matched.count(\\\"\\\\n\\\")\\n\\n # Add empty spaces with the size of the content if the size\\n # must be kept.\\n if keep_content_size:\\n _buffer = len(matched) - count\\n replacement += \\\" \\\" * _buffer + \\\"\\\\n\\\" * count\\n\\n # Otherwise simply keep the number of lines\\n else:\\n replacement += \\\"\\\\n\\\" * count\\n\\n return replacement\",\n \"def _parse(self):\\n soup = BS(self._current_html, 'lxml')\\n for item in soup.select('div.c'):\\n temp = {}\\n # main content\\n ctt = item.select('span.ctt')\\n if not ctt:\\n continue\\n weibo_body = item.select('div')\\n if len(weibo_body) > 1:\\n temp['content'] = weibo_body[0].text\\n btn_group = weibo_body[1].text\\n else:\\n temp['content'] = weibo_body[0].select('span.ctt')[0].text\\n btn_group = weibo_body[0].text\\n temp['is_repost'] = True if REPO_TEST_PATTERN.match(\\n temp['content']) else False\\n try:\\n temp['like_num'] = LIKE_NUM_PATTERN.findall(btn_group)[0]\\n temp['cmt_num'] = COMMENT_NUM_PATTERN.findall(btn_group)[0]\\n temp['repo_num'] = REPO_NUM_PATTERN.findall(btn_group)[0]\\n except Exception:\\n pass\\n cmt = item.select('.cmt')\\n # visibility\\n if cmt:\\n try:\\n temp['visibility'] = VISIBILITY_PATTERN.findall(\\n cmt[0].text)[0]\\n except Exception:\\n pass\\n\\n # img in main content\\n img = item.select('div a img')\\n img_src = img[0].attrs['src'] if img else None\\n temp['img_src'] = img_src\\n LOGGER.debug('img_src: {}'.format(img_src))\\n # time & source device\\n ct = item.select('span.ct')\\n if ct:\\n ct = ct[0]\\n text = ct.text\\n reg_result = TIME_PATTERN.findall(text)[0]\\n\\n temp['time'] = ar(\\n '{}年{}'.format(self._current_year, reg_result[0]),\\n DATE_FMTS[0]\\n ).naive if reg_result[0] else ar(\\n reg_result[1], DATE_FMTS[1]\\n ).naive\\n temp['source'] = SOURCE_DEVICE_PATTERN.findall(text)[0]\\n self._post_item = Post(**temp)\\n self._attachment_item = Attachment(\\n uri=img_src, post=self._post_item)\\n self._store()\",\n \"def comment(self, comment):\\n self.logger.debug(\\\"In 'comment' setter.\\\")\\n\\n if len(comment) > 512:\\n raise Exception(\\\"Comment is too long, must be less than 512 characters.\\\")\\n\\n self._comment = comment\"\n]","isConversational":false},"negative_scores":{"kind":"list like","value":["0.6160088","0.6029694","0.58862054","0.5713247","0.5622352","0.561385","0.5606108","0.557518","0.5571938","0.5537934","0.54423904","0.54145694","0.5373428","0.5349663","0.53335774","0.5330848","0.5328987","0.5298146","0.52952015","0.52707106","0.5269429","0.5266945","0.52583724","0.52583593","0.52177554","0.52029836","0.5161327","0.5161327","0.5161327","0.5161327","0.51565164","0.5136509","0.5134119","0.51317185","0.5123303","0.5119867","0.5106746","0.5104185","0.5103166","0.510109","0.50991064","0.50805205","0.5078447","0.5077799","0.50615114","0.50483155","0.50374866","0.5010854","0.5000955","0.5000955","0.5000955","0.5000955","0.5000955","0.5000955","0.5000955","0.5000955","0.5000955","0.5000955","0.49983528","0.49862084","0.49733317","0.49658537","0.4965225","0.49567938","0.49275476","0.4913391","0.49099016","0.49098837","0.49014252","0.48931277","0.48911944","0.48806304","0.48781183","0.4872657","0.48720428","0.48709697","0.48609668","0.48471248","0.48360848","0.48262262","0.48260903","0.4824025","0.4824025","0.48226005","0.47903877","0.47823218","0.47763982","0.47678822","0.47588745","0.47585565","0.47448865","0.47386247","0.47310662","0.47245488","0.47242302","0.4718786","0.4717267","0.4710313","0.47028318","0.46993613"],"string":"[\n \"0.6160088\",\n \"0.6029694\",\n \"0.58862054\",\n \"0.5713247\",\n \"0.5622352\",\n \"0.561385\",\n \"0.5606108\",\n \"0.557518\",\n \"0.5571938\",\n \"0.5537934\",\n \"0.54423904\",\n \"0.54145694\",\n \"0.5373428\",\n \"0.5349663\",\n \"0.53335774\",\n \"0.5330848\",\n \"0.5328987\",\n \"0.5298146\",\n \"0.52952015\",\n \"0.52707106\",\n \"0.5269429\",\n \"0.5266945\",\n \"0.52583724\",\n \"0.52583593\",\n \"0.52177554\",\n \"0.52029836\",\n \"0.5161327\",\n \"0.5161327\",\n \"0.5161327\",\n \"0.5161327\",\n \"0.51565164\",\n \"0.5136509\",\n \"0.5134119\",\n \"0.51317185\",\n \"0.5123303\",\n \"0.5119867\",\n \"0.5106746\",\n \"0.5104185\",\n \"0.5103166\",\n \"0.510109\",\n \"0.50991064\",\n \"0.50805205\",\n \"0.5078447\",\n \"0.5077799\",\n \"0.50615114\",\n \"0.50483155\",\n \"0.50374866\",\n \"0.5010854\",\n \"0.5000955\",\n \"0.5000955\",\n \"0.5000955\",\n \"0.5000955\",\n \"0.5000955\",\n \"0.5000955\",\n \"0.5000955\",\n \"0.5000955\",\n \"0.5000955\",\n \"0.5000955\",\n \"0.49983528\",\n \"0.49862084\",\n \"0.49733317\",\n \"0.49658537\",\n \"0.4965225\",\n \"0.49567938\",\n \"0.49275476\",\n \"0.4913391\",\n \"0.49099016\",\n \"0.49098837\",\n \"0.49014252\",\n \"0.48931277\",\n \"0.48911944\",\n \"0.48806304\",\n \"0.48781183\",\n \"0.4872657\",\n \"0.48720428\",\n \"0.48709697\",\n \"0.48609668\",\n \"0.48471248\",\n \"0.48360848\",\n \"0.48262262\",\n \"0.48260903\",\n \"0.4824025\",\n \"0.4824025\",\n \"0.48226005\",\n \"0.47903877\",\n \"0.47823218\",\n \"0.47763982\",\n \"0.47678822\",\n \"0.47588745\",\n \"0.47585565\",\n \"0.47448865\",\n \"0.47386247\",\n \"0.47310662\",\n \"0.47245488\",\n \"0.47242302\",\n \"0.4718786\",\n \"0.4717267\",\n \"0.4710313\",\n \"0.47028318\",\n \"0.46993613\"\n]","isConversational":false},"document_score":{"kind":"string","value":"0.5155063"},"document_rank":{"kind":"string","value":"31"}}},{"rowIdx":93283,"cells":{"query":{"kind":"string","value":"Actions to perform once a Like is created."},"document":{"kind":"string","value":"def on_created_like(sender, instance: dillo.models.mixins.Likes, created, **kwargs):\n\n if (\n isinstance(instance.content_object, dillo.models.comments.Comment)\n and instance.content_object.parent_comment\n ):\n target = instance.content_object.parent_comment\n else:\n target = None\n\n log.debug('Generating like activity')\n # TODO(fsiddi) Prevent duplicate activity.\n # If the user likes a post or comment after having unliked it, do not generate activity.\n action.send(instance.user, verb='liked', action_object=instance.content_object, target=target)\n\n # Increase likes_count for profile of content owner their content is liked.\n if not created:\n return\n target_user = instance.content_object.user\n dillo.models.profiles.Profile.objects.filter(user=target_user).update(\n likes_count=F('likes_count') + 1\n )\n log.debug('Increased like count for user %s' % target_user)"},"metadata":{"kind":"string","value":"{\n \"objective\": {\n \"self\": [],\n \"paired\": [],\n \"triplet\": [\n [\n \"query\",\n \"document\",\n \"negatives\"\n ]\n ]\n }\n}"},"negatives":{"kind":"list like","value":["def like(self):\n self.like_count = self.like_count + 1 if self.like_count else 1","def like():\n if CURR_USER_KEY in session:\n cafe_id = int(request.json[\"cafe_id\"])\n print(type(cafe_id))\n user_id = g.user.id\n print(type(user_id))\n like = Like(\n cafe_id=cafe_id,\n user_id=user_id\n )\n print(like)\n db.session.add(like)\n print(\"====================================\")\n print(like)\n db.session.commit()\n return jsonify({\"liked\": cafe_id})\n return jsonify({\"error\": \"Not logged in\"})","def test_like_model(self):\n u = User(\n username = \"like_test_user\",\n email = \"liketest@test.com\",\n password = \"HASHED_POTATOES\" \n )\n\n db.session.add(u)\n db.session.commit()\n\n m = Message(text=\"warble warble\", user_id=u.id)\n\n u.messages.append(m)\n db.session.commit()\n\n l = Like(user_id=u.id, message_id=m.id)\n\n db.session.add(l)\n db.session.commit()\n\n self.assertEqual(l.user_id, u.id)\n self.assertEqual(l.message_id, m.id)","def like(self, n: int) -> None:\n\n # YOUR CODE HERE\n self.likes += 1","def like(self, request, pk=None):\n\n user_wall_post = self.get_object()\n user_wall_post.likes.add(self.request.user)\n to_user = user_wall_post.owner\n from_user = request.user\n\n UserNotification.create_post_friend_liked_notification(from_user, to_user, 'Right', id=pk)\n return Response(status=201)","def like(self, data_base, user):\n cursor = data_base.cursor()\n cursor.execute(f\"UPDATE post SET likes = likes + 1 WHERE id = '{self.id}'\") # Increments the likes\n cursor.execute(f\"INSERT INTO user_like (user_id, post_id) VALUES ({user.id}, {self.id})\")\n if self.commit_to_db:\n data_base.commit()\n cursor.close()","def test_like_message(self):\n\n self.u.likes.append(self.m_u2)\n db.session.commit()\n\n self.assertEqual(len(self.u.likes), 1)\n self.assertEqual(self.u.likes[0], self.m_u2)","def save_model(self, request, obj, form, change):\n obj.post.likeNumIncrease()\n obj.save()","def like(request, pk):\n try:\n like = Likes()\n like.user = request.user\n like.post = Post.objects.filter(pk=pk).first()\n like.save()\n except Exception:\n like = Likes.objects.filter(user=request.user.pk, post=pk).first()\n like.delete()\n return HttpResponseRedirect(reverse(\"mainapp:post\", kwargs={\"pk\":pk}))","def put_like(self, object_id):\n return self.put_object(object_id, \"likes\")","def put_like(self, object_id):\n return self.put_object(object_id, \"likes\")","def like_tweet(tweet_id):\n twitter.create_favorite(id=tweet_id)","def __init__(self, who: str, when: date, what: str) -> None:\n\n # YOUR CODE HERE\n self.content = what\n self.userid = who\n self.created_at = when\n self.likes = 0","def post(self):\n liked = self.request.get('like')\n unliked = self.request.get('unlike')\n post_id = self.request.get('post_id')\n post = Posts.get_by_id(int(post_id))\n user = self.get_active_user()\n user_id = int(user.key().id())\n\n if liked:\n if user_id in post.liked_by:\n self.render_improper_endpoint_access(\"like\")\n else:\n if post.submitter_id != user_id:\n post.liked_by.append(user.key().id())\n post.put()\n self.redirect('/%s' % str(post.key().id()))\n else:\n self.error(403)\n elif unliked:\n if user_id in post.liked_by:\n index = post.liked_by.index(user_id)\n del post.liked_by[index]\n post.put()\n self.redirect('/%s' % str(post.key().id()))\n else:\n self.error(500)","def post_like(self, entry, **args):\n args.update(entry=entry)\n return self.fetch(\"/like\", post_args=args)","def test_create_like(self):\n pokemon = Pokemon.objects.all().first()\n r = self.client.post(\n reverse('create_like'),\n {'pokemon-like-id': pokemon.id},\n HTTP_X_REQUESTED_WITH='XMLHttpRequest'\n )\n\n self.assertEqual(Like.objects.all().count(), 1)","def test_new_likes(self):\n\n new_user = User.signup(\"newuser\", \"newuser@gmail.com\", \"password123\", None)\n db.session.add(new_user)\n db.session.commit()\n\n new_msg = Message(text=\"A new message from a new user\", user_id=new_user.id)\n db.session.add(new_msg)\n db.session.commit()\n\n new_user.likes.append(new_msg)\n db.session.commit()\n\n likes = Likes.query.all()\n self.assertEqual(len(likes), 1)\n self.assertEqual(likes[0].user_id, new_user.id)\n self.assertEqual(likes[0].message_id, new_msg.id)","def add_like(id):\r\n username = request.json['username']\r\n duplicate_likes_query = models.Like.query.filter(models.Like.username == username, models.Like.post_id == id)\r\n # if like from this user to this tweet already exist\r\n if duplicate_likes_query.count() > 0:\r\n return {'like_id': duplicate_likes_query.first().like_id}\r\n\r\n # if original tweet does not exist -> 404\r\n models.Tweet.query.get_or_404(id)\r\n\r\n like = models.Like(post_id=id, username=username, timestamp=datetime.datetime.now())\r\n db.session.add(like)\r\n db.session.commit()\r\n\r\n return {'like_id': like.like_id}","def test_like(self):\n\n message = Message(text=\"hello world\", user_id=self.testuser2.id)\n \n db.session.add(message)\n db.session.commit()\n \n message_id = message.id\n\n with self.client as c:\n with c.session_transaction() as session:\n session[CURR_USER_KEY] = self.testuser.id\n \n response = c.post(f\"/messages/{message_id}/like\", follow_redirects=True)\n \n self.assertEqual(response.status_code, 200)\n\n likes = Likes.query.filter(Likes.message_id==message_id).all()\n self.assertEqual(len(likes), 1)\n self.assertEqual(likes[0].user_id, self.testuser.id)","def like_event(request, pk):\n template_var = base_template_vals(request)\n previousLike = Like.objects.filter(event=Event.objects.get(id=pk), \n user=UserProfile.objects.filter(\n django_user=request.user)[0])\n if(len(previousLike) == 0):\n like = Like(event=Event.objects.get(id=pk))\n like.user = UserProfile.objects.filter(django_user=request.user)[0]\n like.save()\n #sys notification\n from_user = UserProfile.objects.get(django_user = User.objects.get(\n username__exact='admin')) \n to_user = UserProfile.objects.get(django_user=request.user)\n event_id = pk\n sys_notification(to_user, \"save_event\", from_user, event_id)\n return redirect('index')","def users_like_changed(sender, instance, **kwargs):\n instance.total_likes = instance.users_like.count()\n instance.save()","def likes(self):\r\n return Likes(self)","def like(request, pk):\n update_posts_expiration()\n post = 0;\n post = Post.objects.get(id=pk)\n #if the post is expired you user can't like it\n if post.is_expired == True:\n return Response(\"You can't interact with a expired post\")\n else:\n #if the post is not expired then increment the likes count by 1 and save a serializer of like to the database with user and post information\n if request.method == \"POST\":\n if request.user != post.user:\n post.likes_count += 1\n post.save()\n serializer = LikeSerializer(data=request.data)\n if serializer.is_valid():\n serializer.save(user=request.user, liked_post=post)\n return Response(\"you liked the post with title: \" + post.title)\n else:\n return Response(serializer.errors)\n else:\n return Response(\"you can't like your own post\")\n \n\n return Response(\"you didn't like the post yet\")","def artwork_like(request):\n artwork_number = request.GET.get('id')\n following_page = 'artwork_detail?id='+str(artwork_number)\n if 'following' in request.GET:\n following_page = request.GET.get('following')\n artwork = Artwork.objects.get(id=artwork_number)\n user = request.user\n like = Like(user=user, artwork=artwork)\n \"\"\" Save the like in the database, attached to the correct artwork piece using its unique ID \"\"\"\n like.save()\n return HttpResponseRedirect(following_page)","def send_like(request):\n if request.method == \"POST\":\n if \"token\" in request.data and request.data[\"token\"] != \"\" and request.data[\"token\"] is not None:\n if Token.objects.filter(key=request.data[\"token\"]).exists():\n token = get_object_or_404(Token, key=request.data[\"token\"])\n if Post.objects.filter(pk=request.data[\"post_id\"]).exists():\n post = Post.objects.get(pk=request.data[\"post_id\"])\n if Like.objects.filter(post=post, user_id=token.user_id).exists():\n return Response({\"error\": 31})\n else:\n post.count_likes += 1\n post.save()\n like = Like.objects.create(post=post, user_id=token.user_id)\n serializer = PostSerializer(post, context={'user_id': token.user_id})\n UserFeed.objects.create(user=post.author,\n action_user=token.user,\n like=like,\n action=\"Like\")\n if post.author != token.user:\n message = \"{} likes your post\".format(token.user.username)\n custom = {\n \"post_id\": post.id,\n \"avatar\": UserProfile.objects.get(user=token.user).avatar.url\n }\n\n user_notification = UserNotification.objects.get(user=post.author)\n send_notification(custom, message, user_notification)\n\n return Response({\"success\": 30,\n \"post\": serializer.data})\n else:\n return Response({\"error\": 32})\n else:\n return Response({\"error\": 17})","def test_associate_user_likes(self):\n\n testlike1 = self.user1.likes.append(self.message2)\n testlike2 = self.user2.likes.append(self.message1)\n db.session.commit()\n\n likes = Likes.query.all()\n\n self.assertEqual(len(likes), 2)\n self.assertEqual(likes[0].user_id, self.user1.id)\n self.assertEqual(likes[1].user_id, self.user2.id)","def test_like_activity(self):\n from .mockers import user_status_context\n from .mockers import subscribe_context, create_context\n username = 'messi'\n username_not_me = 'xavi'\n self.create_user(username)\n self.create_user(username_not_me)\n self.create_context(create_context)\n self.admin_subscribe_user_to_context(username, subscribe_context)\n self.admin_subscribe_user_to_context(username_not_me, subscribe_context)\n res = self.create_activity(username, user_status_context)\n activity_id = res.json['id']\n res = self.testapp.post('/activities/%s/likes' % activity_id, '', oauth2Header(username_not_me), status=201)\n activity = self.testapp.get('/activities/%s' % activity_id, '', oauth2Header(username), status=200)\n\n self.assertEqual(res.json['object']['likes'][0]['username'], username_not_me)\n self.assertEqual(res.json['object']['liked'], True)\n self.assertEqual(res.json['object']['likesCount'], 1)\n\n self.assertEqual(activity.json['likes'][0]['username'], username_not_me)\n self.assertEqual(activity.json['liked'], False)\n self.assertEqual(activity.json['likesCount'], 1)","def d_like(request):\n token = request.GET.get('token','')\n\n if not token:\n return HttpResponse('error in user validation')\n \n try:\n api = ApiKey.objects.get(key=token)\n user = api.user\n except ApiKey.DoesNotExist:\n return HttpResponse('error in user validation')\n if request.method == 'POST' and user.is_active:\n try:\n post_id = int(request.POST['post_id'])\n except ValueError:\n return HttpResponse('erro in post id')\n try:\n post = Post.objects.get(pk=post_id)\n except Post.DoesNotExist:\n return HttpResponse('post not found')\n\n like, created = Likes.objects.get_or_create(user=user, post=post)\n if created:\n post.like=post.like+1\n post.save()\n return HttpResponse('+1')\n elif like:\n like.delete()\n post.like=post.like-1\n post.save()\n return HttpResponse('-1')\n\n return HttpResponse('error in parameters')","def test_associate_message_likes(self):\n\n testlike1 = self.user2.likes.append(self.message2)\n testlike2 = self.user2.likes.append(self.message1)\n db.session.commit()\n\n likes = Likes.query.all()\n self.assertEqual(len(likes), 2)\n self.assertEqual(likes[0].message_id, self.message2.id)\n self.assertEqual(likes[1].message_id, self.message1.id)","def update_likes(self):\n self.nb_likes = self.likes.count()\n self.save()","def post_likes(request):\n if request.method == 'POST':\n if not request.user.is_authenticated:\n raise PermissionDenied\n pk = request.POST.get('post_id')\n try:\n post = PostsModel.objects.get(id=pk)\n except PostsModel.DoesNotExist:\n post = None\n try:\n likes = LikesModel.objects.get(user=request.user, post=post)\n except LikesModel.DoesNotExist:\n likes = None\n\n\n if likes == None:\n likes = LikesModel.objects.create(user=request.user, post=post)\n post.likes_count = F('likes_count') + 1\n post.save(update_fields=[\"likes_count\"])\n class_likes = 'fa fa-heart mx-2'\n else:\n if likes.is_liked == True:\n likes.is_liked = False\n likes.save()\n post.likes_count = F('likes_count') - 1\n post.save(update_fields=[\"likes_count\"])\n class_likes = 'fa fa-heart-o mx-2'\n elif likes.is_liked == False:\n likes.is_liked = True\n likes.save()\n post.likes_count = F('likes_count') + 1\n post.save(update_fields=[\"likes_count\"])\n class_likes = 'fa fa-heart mx-2'\n post.refresh_from_db()\n response_data = {'_code' : 0, '_status' : 'ok', '_likes': post.likes_count, '_class_likes': class_likes}\n else:\n response_data = {'_code' : 1, '_status' : 'no' }\n\n return JsonResponse(response_data)","def test_not_create_like_for_current_anon_already_liked(self):\n pokemon = Pokemon.objects.all().first()\n\n r = self.client.post(\n reverse('create_like'),\n {'pokemon-like-id': pokemon.id},\n HTTP_X_REQUESTED_WITH='XMLHttpRequest'\n )\n\n r = self.client.post(\n reverse('create_like'),\n {'pokemon-like-id': pokemon.id},\n HTTP_X_REQUESTED_WITH='XMLHttpRequest'\n )\n\n self.assertEqual(Like.objects.all().count(), 1)","def on_created_comment(sender, instance: dillo.models.comments.Comment, created, **kwargs):\n if not created:\n return\n # Extract tags and mentions from text and assign them to the Comment\n tags, mentions = extract_tags_and_mentions(instance.content)\n instance.tags.set(*tags)\n for mentioned_user in mentions:\n log.debug('Mentioning user %s in Comment %i' % (mentioned_user, instance.id))\n dillo.models.mixins.Mentions.objects.create(\n user=mentioned_user,\n content_object=instance,\n )\n # TODO(fsiddi) Generate activity about mention\n\n # Generate activity about comment creation\n log.debug('Generating activity about comment creation')\n verb = 'commented'\n if instance.parent_comment:\n verb = 'replied'\n action.send(instance.user, verb=verb, action_object=instance, target=instance.entity)\n log.debug('Set user %s as follower of own comment %i' % (instance.user, instance.id))\n follow(instance.user, instance, actor_only=False)","def add_liked_track():\n # Inserts the track information.\n new_liked_by = request.vars.liked_by\n new_liked_title = request.vars.liked_title\n t_id = db.liked_track.insert(\n liked_by = new_liked_by,\n liked_title = new_liked_title\n )\n return response.json(dict(\n id = t_id,\n liked_by = new_liked_by,\n liked_title = new_liked_title\n ))","def test_like_activity_by_various(self):\n from .mockers import user_status_context\n from .mockers import subscribe_context, create_context\n username = 'messi'\n username_not_me = 'xavi'\n self.create_user(username)\n self.create_user(username_not_me)\n self.create_context(create_context)\n self.admin_subscribe_user_to_context(username, subscribe_context)\n self.admin_subscribe_user_to_context(username_not_me, subscribe_context)\n res = self.create_activity(username, user_status_context)\n activity_id = res.json['id']\n res = self.testapp.post('/activities/%s/likes' % activity_id, '', oauth2Header(username_not_me), status=201)\n res = self.testapp.post('/activities/%s/likes' % activity_id, '', oauth2Header(username), status=201)\n\n self.assertEqual(res.json['object']['likes'][0]['username'], username_not_me)\n self.assertEqual(res.json['object']['likes'][1]['username'], username)\n self.assertEqual(res.json['object']['liked'], True)\n self.assertEqual(res.json['object']['likesCount'], 2)","def on_created_post(sender, instance, created, **kwargs):\n if not created:\n return\n if not issubclass(sender, dillo.models.posts.Post):\n return\n instance.hash_id = instance.id\n instance.save()\n log.debug('Set user %s as follower of own post %i' % (instance.user, instance.id))\n follow(instance.user, instance, actor_only=False)","def post(self, request, *args, **kwargs):\n\n user_wall_post_comment = self.get_object()\n user_wall_post_comment.likes.add(self.request.user)\n return Response(status=201)","def test_like_already_liked_activity(self):\n from .mockers import user_status_context\n from .mockers import subscribe_context, create_context\n username = 'messi'\n username_not_me = 'xavi'\n self.create_user(username)\n self.create_user(username_not_me)\n self.create_context(create_context)\n self.admin_subscribe_user_to_context(username, subscribe_context)\n self.admin_subscribe_user_to_context(username_not_me, subscribe_context)\n res = self.create_activity(username, user_status_context)\n activity_id = res.json['id']\n res = self.testapp.post('/activities/%s/likes' % activity_id, '', oauth2Header(username_not_me), status=201)\n res = self.testapp.post('/activities/%s/likes' % activity_id, '', oauth2Header(username_not_me), status=200)\n\n self.assertEqual(res.json['object']['likes'][0]['username'], username_not_me)\n self.assertEqual(res.json['object']['liked'], True)\n self.assertEqual(res.json['object']['likesCount'], 1)","def test_update_like_can_message(self):\n\n self.assertEqual(first=0, second=self.like.likes.all().count())\n url = reverse('like-detail', args=(self.like.id,))\n data = {\n 'likes': [2]\n }\n json_data = json.dumps(data)\n self.client.credentials(HTTP_AUTHORIZATION=self.token)\n self.like_1.likes.add(self.user)\n response = self.client.patch(path=url, content_type='application/json', data=json_data)\n self.assertTrue(response.data.get('can_message', False))\n self.assertEqual(first=200, second=response.status_code)\n self.assertEqual(first=1, second=self.like.likes.all().count())","def test_like_topic_viewset(self):\n\n topic = TopicFactory(author=self.user)\n response = self.client.post(reverse('api:topics-like', kwargs={'topic_id': topic.id}))\n self.assertEqual(response.status_code, status.HTTP_201_CREATED)\n obj_type = ContentType.objects.get_for_model(topic)\n likes = Like.objects.filter(\n content_type=obj_type, object_id=topic.id, user=self.user)\n self.assertTrue(likes.exists())","def test_a_user_can_like_article(self):\n article = self.create_article()\n\n slug = article.data['data']['slug']\n like = self.client.post('/api/articles/{}/like/'.format(slug),\n HTTP_AUTHORIZATION='Bearer ' +\n self.token,\n format='json')\n\n self.assertEqual(like.status_code, 200)","def test_like_mission(self):\n\n with self.client as c:\n with c.session_transaction() as sess:\n sess['user_id'] = self.user2.id\n\n db.session.add(self.mission)\n resp = c.post(f'/v1/mission/like{self.mission.id}')\n\n self.assertEqual(resp.json['success'], 'added')\n mission = Mission.query.get(self.mission.id)\n self.assertEqual(mission.likes, {0, self.user2.id})","def like_photo(self):\n like_button = self.driver.find_elements_by_class_name('wpO6b')[0]\n like_button.click()","def addLikeCount(self,count):\n self.interactionCount += count\n return None","def test_authorized_user_like_field(self):\n lesson_id = 1\n self.create_and_auth(email='user@gmail.com',\n password='qwerty_1234')\n\n response = self.client.get(reverse('lessons-detail',\n args=(lesson_id,)))\n self.assertEqual(response.status_code, status.HTTP_200_OK)\n self.assertIn('like', response.data)\n self.assertFalse(response.data['like'])\n\n self.client.post(reverse('lesson_likes', kwargs={\n 'lesson_id': lesson_id\n }))\n\n response = self.client.get(reverse('lessons-detail',\n args=(lesson_id,)))\n self.assertEqual(response.status_code, status.HTTP_200_OK)\n self.assertIn('like', response.data)\n self.assertTrue(response.data['like'])","def like_tweet(self, tag):\n self.bot.get('https://twitter.com/search?q=' + tag + '&src=typed')\n self.__wait(3, 3)\n for i in range(1, 3):\n self.bot.execute_script('window.scrollTo(0,document.body.scrollHeight)')\n self.__wait(2, 3)\n tweets = self.bot.find_elements_by_tag_name('article')\n\n links = []\n for tweet in tweets:\n sub_links = tweet.find_elements_by_tag_name('a')\n links += [sub_link.get_attribute('href')\n for sub_link in sub_links if 'status' in sub_link.get_attribute('href')]\n\n print('Started to like {} tweets'.format(len(links)))\n\n for link in links:\n self.bot.get(link)\n self.__wait(3, 5)\n likes = self.bot.find_elements_by_css_selector('div[data-testid=\"like\"')\n for like in likes:\n like.click()\n self.__wait(3, 5)","def process_IN_CREATE(self, event):","def create(self):\n ...","def add_to_like(request):\n if request.is_ajax():\n model_id = request.POST.get('model_id')\n model_type = request.POST.get('model_type')\n content_type = ContentType.objects.get(model=model_type)\n ModelType = content_type.model_class()\n model_type_qs = get_object_or_404(ModelType, id=model_id)\n obj = LikeDislike.objects.create_for_instance_model(\n instance=model_type_qs, request=request, likedislike='like')\n logger.info(f\"{model_type_qs} {obj.likedislike}\")\n return JsonResponse({\"response\": f\"شما {model_type_qs.title} را پسندیدید\"}, status=200)\n return HttpResponseRedirect(request.META.get(\"HTTP_REFERER\"))","def create(self, *args, **kwargs):\n pass","def following_changed(sender, action, instance, *args, **kwargs):\n\n # m2mchanged.connect specified in apps.py\n\n following = instance.following.all()\n creator = instance.user\n\n if creator in following:\n raise ValidationError (\"can't like own post\")","def like_view(request, pk):\n cocktail = get_object_or_404(Cocktail, id=pk)\n liked = False\n if cocktail.likes.filter(id=request.user.id).exists():\n cocktail.likes.remove(request.user)\n liked = False\n else:\n cocktail.likes.add(request.user)\n liked = True\n\n return HttpResponseRedirect(reverse('cocktails:cocktail-details', args=[str(pk)]))","def like_comment(username):\n\n user_ID = before_request()\n user_ID = None\n if user_ID != None:\n user_ID = str(g.user['_id'])\n\n message_id = request.args.get('message_id')\n profile_usr = request.args.get('profile_usr')\n count = float(1)\n\n redis_obj.zincrby('add_like', message_id, count)\n\n print \"message ID\", message_id\n\n if redis_obj.get(user_ID):\n return redirect(url_for('timeline', username=profile_usr, userId=pickle.loads(redis_obj.get(user_ID))))\n else:\n \"\"\" Invalidating cache \"\"\"\n redis_obj.delete('leaderboard-key')\n redis_obj.delete(session['user_id'])\n print \"Invalidating cache after Like\"\n return redirect(url_for('timeline', username=profile_usr))","def __init__(self, author, title, text=\"\", haiku=\"False\"):\n self.author = author\n self.title = title\n self.text = text\n self.haiku = haiku\n self.name = self\n Poem.obj_created += 1","def like_tweets(pos_tweets):\n\n for tweet in pos_tweets:\n twitter.CreateFavorite(status_id=tweet.id)\n\n return","def create(self):\n\n pass","def test_is_liked_by(self):\n\n u = self.user\n u2 = User(\n email=\"test2@test.com\" ,\n username=\"testuser2\" ,\n password=\"HASHED_PASSWORD\"\n )\n\n m = Message(\n user_id=u.id ,\n text='test content'\n )\n\n db.session.add_all([u2 , m])\n db.session.commit()\n\n #not like \n self.assertEqual(len(u2.likes) , 0)\n self.assertEqual(len(m.likes_users) , 0)\n self.assertEqual(u2.is_like(m) , False)\n self.assertEqual(m.is_liked_by(u2) , False)\n\n like = Likes(user_id=u2.id , message_id=m.id)\n db.session.add(like)\n db.session.commit()\n\n self.assertEqual(len(u2.likes) , 1)\n self.assertEqual(len(m.likes_users) , 1)\n self.assertEqual(u2.is_like(m) , True)\n self.assertEqual(m.is_liked_by(u2), True)","def order_posts_to_likes(sender, instance, created=False, **kwargs):\n user = instance.user\n post = instance.post\n tags = post.tags.all()\n\n likes = Like.objects.filter(user=user, post=post)\n liked_posts = [ i.post for i in likes if i.liked_status==True]\n\n for post in liked_posts:\n liked_tags = post.tags.all()\n for tag in liked_tags:\n if not tag in tags:\n tags.append(tag)\n\n similar_posts = Post.objects.filter(tags__in=tags).distinct()\n try:\n UserPostWeight.objects.filter(user=user).delete()\n except:\n pass\n weight = 2\n for post in similar_posts:\n UserPostWeight.objects.create(user=user, post=post, weight=weight)\n weight+=2\n \n\n\n pass","def like(id):\n spinner = Halo(text=\"Liking photo...\", spinner=\"dots\").start()\n try:\n api.like(id)\n spinner.succeed(\"Liked\")\n except Exception:\n spinner.fail(\"Failed to like photo.\")","def get(self, post_id):\n Post.add_like(int(post_id), self.user.get_id())\n self.redirect('/blog')","def create(self):\n pass","def create(self):\n pass","def create(self):\n pass","def process_like(self):\n medias = self.prepare_process_like()\n wait_time = 3600//(self.limits_per_hour.get('like') + 1)\n for m in medias:\n time.sleep(wait_time + trunc_gauss(0, 10, -40, 40))\n self.liking(m)","def on_created_action(sender, instance: models_actstream.Action, created, **kwargs):\n if not created:\n return\n\n # If content_type is Post or Profile\n if (\n instance.action_object_content_type\n == ContentType.objects.get_for_model(dillo.models.posts.Post)\n and instance.verb == 'posted'\n ) or (\n instance.action_object_content_type\n == ContentType.objects.get_for_model(dillo.models.profiles.Profile)\n and instance.verb == 'updated their reel'\n ):\n dillo.tasks.feeds.establish_time_proximity(instance)\n\n log.debug('Creating background fanout operation for action %i' % instance.id)\n dillo.tasks.feeds.activity_fanout_to_feeds(instance.id)","def process_like_and_follow(self):\n follow, media, unfollow = self.prepare_process_like_and_follow()\n follow_acts, media_acts, unfollow_acts = len(follow), len(media), len(unfollow)\n all_acts = round(self.limits_per_hour.get('follow') + self.limits_per_hour.get('like') +\n self.limits_per_hour.get('unfollow'))\n wait_time = 3600 // all_acts + 1\n while follow_acts or media_acts or unfollow_acts:\n time.sleep(wait_time + trunc_gauss(0, 10, -40, 40))\n rc = random.choices(['f', 'l', 'u'], [follow_acts, media_acts, unfollow_acts])[0]\n\n if rc == 'f':\n fo = follow.pop(0)\n self.following_and_storing(fo)\n follow_acts -= 1\n elif rc == 'l':\n mo = media.pop(0)\n self.liking(mo)\n media_acts -= 1\n elif rc == 'u':\n uo = unfollow.pop(0)\n self.unfollowing_and_removing(uo)\n unfollow_acts -= 1","def like(request, post_id):\n if request.method == \"PUT\":\n liked = None\n user = request.user\n post = Post.objects.get(id=post_id)\n # If user already liked, decrement the like count and remove as 'liker'\n if user in post.liked_by.all():\n post.liked_by.remove(user)\n post.likes -= 1\n post.save()\n liked = False\n # Else increase like count and add user\n else:\n post.liked_by.add(user)\n post.likes += 1\n post.save()\n liked = True\n # Return data for updating dynamically with javascript\n return JsonResponse({\"total_likes\": post.likes, \"liked\": liked})","def test_is_liked_by_method(self):\n\n self.u2.likes.append(self.m_u1)\n db.session.commit()\n \n self.assertTrue(self.m_u1.is_liked_by(self.u2))\n self.assertFalse(self.m_u1.is_liked_by(self.u))","def post_liker(a, args):\n if args.likes.filter(user=a).exists():\n exists = True\n else:\n exists = False\n return exists","def sample(self, like_params):\n\t\traise NotImplementedError","def test_a_user_can_delete_like(self):\n article = self.create_article()\n\n slug = article.data['data']['slug']\n self.client.post('/api/articles/{}/like/'.format(slug),\n HTTP_AUTHORIZATION='Bearer ' +\n self.token,\n format='json')\n like = self.client.post('/api/articles/{}/like/'.format(slug),\n HTTP_AUTHORIZATION='Bearer ' +\n self.token,\n format='json')\n\n self.assertEqual(like.status_code, 200)","def LikeArticle(request):\n try:\n if not request.user.is_authenticated():\n return HttpResponse(json.dumps({\n \"result\": \"1\"\n }))\n\n like = request.GET.get('like', '')\n article_id = request.GET.get('article_id', '')\n\n if str(like).lower() != 'true':\n like = 'false'\n else:\n like = 'true'\n \n record = UserLikedArticles.objects.filter(user=request.user.id, article=article_id)\n\n if like == 'false':\n if record != None and len(record) > 0:\n record[0].delete()\n else:\n if record == None or len(record) == 0:\n like = UserLikedArticles()\n like.user = request.user.id\n like.article=Article.objects.filter(id=article_id)[0]\n like.save()\n \n article = Article.objects.filter(id=article_id)[0]\n article.likes = len(UserLikedArticles.objects.filter(article=article_id))\n article.save()\n \n return HttpResponse(json.dumps({\n \"result\": 0\n }))\n except Exception, e:\n return HttpResponse(json.dumps({\n \"result\": e\n }))","def post_tweet(instance, created, raw, **kwargs):\n if created and not raw:\n _twitter.statuses.update(status='%s %s' % (\n instance.title,\n instance.url,\n ))","def test_unlike(self):\n\n message = Message(text=\"hello world\", user_id=self.testuser2.id)\n \n db.session.add(message)\n db.session.commit()\n\n like = Likes(message_id=message.id, user_id=self.testuser.id)\n \n db.session.add(like)\n db.session.commit()\n \n message_id = message.id\n\n with self.client as c:\n with c.session_transaction() as session:\n session[CURR_USER_KEY] = self.testuser.id\n \n response = c.post(f\"/messages/{message_id}/like\", follow_redirects=True)\n \n self.assertEqual(response.status_code, 200)\n\n likes = Likes.query.filter(Likes.message_id==message_id).all()\n self.assertEqual(len(likes), 0)","def created(self, created):\n\n self._created = created","def created(self, created):\n\n self._created = created","def created(self, created):\n\n self._created = created","def created(self, created):\n\n self._created = created","def created(self, created):\n\n self._created = created","def created(self, created):\n\n self._created = created","def created(self, created):\n\n self._created = created","def created(self, created):\n\n self._created = created","def created(self, created):\n\n self._created = created","def created(self, created):\n\n self._created = created","def created(self, created):\n\n self._created = created","def created(self, created):\n\n self._created = created","def created(self, created):\n\n self._created = created","def like(self, event, number):\n\n if not hasattr(self, 'user_id') or not hasattr(self, 'top_id'):\n vk_common.dprint(2, vk_common.func_name(), f\"Не заданы атрибуты user_id = или top_id.\")\n return\n\n item_id = self.item_ids[self.top_id-1]\n if item_id is None:\n vk_common.dprint(2, vk_common.func_name(), f\"Атрибут item_id = None.\")\n return\n else:\n vk_common.dprint(2, vk_common.func_name(), f\"number = {number}; item_id = {item_id}\")\n\n vk_common.dprint(1, f\"имя файла с фото = \\\"{self.user_id}_{self.top_id}\\\"; \"\n f\"ссылка = https://vk.com/id{self.user_id}\")\n\n err, msg, likes = vk_common.like_photo(self.user_id, item_id)\n if err != 0:\n vk_common.dprint(2, \"Не смогли поставить лайк.\")\n mb.showinfo(\"Информация\", f\"Не смогли поставить лайк фото пользователя c id = {self.user_id}.\")\n return\n vk_common.dprint(2, f\"Поставили лайк: owner_id = {self.user_id}; item_id = {item_id}\")\n mb.showinfo(\"Информация\", f\"Поставили лайк фото пользователя c id = {self.user_id}.\")\n # err, msg, likes = user.like_photo(234068204, 345376029)","def log_create(sender, instance, created, **kwargs):\n if created:\n stracks.user(instance).log(\"? has been created\")","def create(self):","def on_create(self):","def like_post(self,id):\n try:\n data = self.api.create_favorite(id)\n if isinstance(data,tweepy.models.Status):\n if hasattr(data,'_json'): \n return json.dumps(data._json)\n except tweepy.error.TweepError as e:\n print(e.__dict__)\n if e.__dict__['api_code']==139:\n data = unlike_post(id)\n return data","def add_like(db, filename, usernick=None):\n cur = db.cursor()\n\n sql = \"\"\"\n select exists(\n select 1 from users where nick=?\n );\n \"\"\"\n cur.execute(sql, (usernick,))\n user_exists = cur.fetchone()[0]\n\n if user_exists==1: # or usernick is None\n sql = \"\"\"\n select exists(\n select 1 from images where filename=?\n );\n \"\"\"\n cur.execute(sql, (filename,))\n img_exists = cur.fetchone()[0]\n if img_exists==1:\n\n sql = \"\"\"\n insert into likes values (?, ?);\n \"\"\"\n cur.execute(sql, (filename, usernick))\n db.commit()","def test_like(self):\n client = Client()\n slug = ['bryan-fox-snowboard-2017', 'some_slug'] # Correct and incorrect slug\n response = client.get('/products/{0}/'.format(slug[0]))\n self.assertEqual(response.status_code, 200)\n Like.objects.filter(product__slug=slug).count() # numbers like of post\n product = Product.objects.get(slug=slug[0])\n if response.context[\"user\"].is_authenticated():\n product.like_set.filter(user=response.context[\"user\"])","def help_create(self):\n print(\"create instances\")","def cursor_created(self, cursor, action, *args, **kw):\n pass","def like_following(self):\n self.logger.log(\"starting like_following...\")\n count_following = self.account.follows_count\n follows_accounts = self.following\n random.shuffle(follows_accounts)\n for acc in follows_accounts:\n acc = perform_with_ran_delay(self.instagram.get_account_by_id, acc)\n self.logger.log(\" {} > {} posts\".format(acc.username, acc.media_count))\n if acc.media_count > 0:\n\n posts = perform_with_ran_delay(self.instagram.get_medias, acc.username, 50)\n if posts:\n for m in posts:\n try:\n perform_with_ran_delay(self.instagram.like, m.identifier)\n self.logger.log(\"liking 1 post from \"+acc.username)\n random_delay()\n except Exception as e:\n self.logger.log(\"skipping 1 post from \"+acc.username)\n self.logger.log(e)\n random_delay()\n continue","def prepare_process_like_and_follow(self):\n follow = []\n media = []\n unfollow = []\n\n coef = self.users_to_follow / self.limits_per_hour.get('follow', 1)\n media_to_like = round(coef*self.limits_per_hour.get('like'))\n num_to_unfollow = round(coef*self.limits_per_hour.get('unfollow'))\n\n feed_likes = media_to_like // 2\n feed_likes_list = []\n following_likes = round((media_to_like // 2) * 3 / 4)\n following_likes_list = []\n followers_likes = media_to_like - feed_likes - following_likes\n\n monitored_ids = [i[\"user\"] for i in self.monitored_users]\n\n for posts in self.hashtag_feed_list(self.search_hashtags):\n if len(follow) < self.users_to_follow:\n for m in posts:\n if self.check_if_suit(m):\n user_id, username = self.get_user_from_post(m)\n if user_id and user_id not in [i[\"user\"] for i in follow] \\\n and user_id not in monitored_ids:\n follow.append({'user': user_id, 'username': username})\n following_likes_list.append(m)\n\n if len(follow) >= self.users_to_follow:\n break\n\n for p in following_likes_list:\n if p in posts:\n posts.remove(p)\n\n if feed_likes > 0:\n if len(posts) > feed_likes:\n feed_likes_list.extend([i['id'] for i in (random.choice(posts) for _ in range(feed_likes))\n if i['id'] not in feed_likes_list])\n else:\n feed_likes_list.extend([i['id'] for i in posts[:feed_likes] if i['id'] not in feed_likes_list])\n feed_likes -= len(feed_likes_list)\n if feed_likes <= 0:\n if len(follow) >= self.users_to_follow:\n break\n if len(follow) >= self.users_to_follow and feed_likes <= 0:\n break\n\n media.extend(feed_likes_list)\n\n if len(following_likes_list) < following_likes:\n followings = []\n get_n_followings = following_likes - len(following_likes_list)\n if following_likes_list:\n following_likes_list = [self.get_media_id_from_post(i) for i in following_likes_list]\n following_likes_list.extend([i for i in self.get_following_likes(followings, get_n_followings)\n if i and i not in media])\n media.extend(following_likes_list)\n else:\n media.extend([self.get_media_id_from_post(i) for i in following_likes_list[:following_likes]])\n\n media.extend([i for i in self.get_followers_likes(followers_likes) if i and i not in media])\n\n unfollow = self.get_to_unfollow(unfollow, num_to_unfollow)\n\n return follow, media, unfollow","def on_saved_post(sender, instance, created, **kwargs):\n # If the post was just created, stop here since we have no tags yet\n if created:\n return\n if not issubclass(sender, dillo.models.posts.Post):\n return\n # Extract tags and mentions from text and assign them to the Post\n tags, mentions = extract_tags_and_mentions(instance.title)\n instance.tags.set(*tags)\n # Delete all existing mentions\n dillo.models.mixins.Mentions.objects.filter(\n content_type_id=instance.content_type_id, object_id=instance.id\n ).delete()\n for mentioned_user in mentions:\n log.debug('Mentioning user %s in Post %i' % (mentioned_user, instance.id))\n dillo.models.mixins.Mentions.objects.create(\n user=mentioned_user,\n content_object=instance,\n )","def like_movie(request):\n\n movie = get_object_or_404(Movie, id=request.POST.get('movie_id'))\n is_liked = False\n if Like.objects.filter(user=request.user, movie=movie).exists():\n Like.objects.get(user=request.user, movie=movie).delete()\n is_liked = False\n else:\n new = Like(user=request.user, movie=movie)\n new.save()\n if Dislike.objects.filter(user=request.user, movie=movie).count():\n Dislike.objects.get(user=request.user, movie=movie).delete()\n is_liked = True\n return HttpResponseRedirect(movie.get_absolute_url())"],"string":"[\n \"def like(self):\\n self.like_count = self.like_count + 1 if self.like_count else 1\",\n \"def like():\\n if CURR_USER_KEY in session:\\n cafe_id = int(request.json[\\\"cafe_id\\\"])\\n print(type(cafe_id))\\n user_id = g.user.id\\n print(type(user_id))\\n like = Like(\\n cafe_id=cafe_id,\\n user_id=user_id\\n )\\n print(like)\\n db.session.add(like)\\n print(\\\"====================================\\\")\\n print(like)\\n db.session.commit()\\n return jsonify({\\\"liked\\\": cafe_id})\\n return jsonify({\\\"error\\\": \\\"Not logged in\\\"})\",\n \"def test_like_model(self):\\n u = User(\\n username = \\\"like_test_user\\\",\\n email = \\\"liketest@test.com\\\",\\n password = \\\"HASHED_POTATOES\\\" \\n )\\n\\n db.session.add(u)\\n db.session.commit()\\n\\n m = Message(text=\\\"warble warble\\\", user_id=u.id)\\n\\n u.messages.append(m)\\n db.session.commit()\\n\\n l = Like(user_id=u.id, message_id=m.id)\\n\\n db.session.add(l)\\n db.session.commit()\\n\\n self.assertEqual(l.user_id, u.id)\\n self.assertEqual(l.message_id, m.id)\",\n \"def like(self, n: int) -> None:\\n\\n # YOUR CODE HERE\\n self.likes += 1\",\n \"def like(self, request, pk=None):\\n\\n user_wall_post = self.get_object()\\n user_wall_post.likes.add(self.request.user)\\n to_user = user_wall_post.owner\\n from_user = request.user\\n\\n UserNotification.create_post_friend_liked_notification(from_user, to_user, 'Right', id=pk)\\n return Response(status=201)\",\n \"def like(self, data_base, user):\\n cursor = data_base.cursor()\\n cursor.execute(f\\\"UPDATE post SET likes = likes + 1 WHERE id = '{self.id}'\\\") # Increments the likes\\n cursor.execute(f\\\"INSERT INTO user_like (user_id, post_id) VALUES ({user.id}, {self.id})\\\")\\n if self.commit_to_db:\\n data_base.commit()\\n cursor.close()\",\n \"def test_like_message(self):\\n\\n self.u.likes.append(self.m_u2)\\n db.session.commit()\\n\\n self.assertEqual(len(self.u.likes), 1)\\n self.assertEqual(self.u.likes[0], self.m_u2)\",\n \"def save_model(self, request, obj, form, change):\\n obj.post.likeNumIncrease()\\n obj.save()\",\n \"def like(request, pk):\\n try:\\n like = Likes()\\n like.user = request.user\\n like.post = Post.objects.filter(pk=pk).first()\\n like.save()\\n except Exception:\\n like = Likes.objects.filter(user=request.user.pk, post=pk).first()\\n like.delete()\\n return HttpResponseRedirect(reverse(\\\"mainapp:post\\\", kwargs={\\\"pk\\\":pk}))\",\n \"def put_like(self, object_id):\\n return self.put_object(object_id, \\\"likes\\\")\",\n \"def put_like(self, object_id):\\n return self.put_object(object_id, \\\"likes\\\")\",\n \"def like_tweet(tweet_id):\\n twitter.create_favorite(id=tweet_id)\",\n \"def __init__(self, who: str, when: date, what: str) -> None:\\n\\n # YOUR CODE HERE\\n self.content = what\\n self.userid = who\\n self.created_at = when\\n self.likes = 0\",\n \"def post(self):\\n liked = self.request.get('like')\\n unliked = self.request.get('unlike')\\n post_id = self.request.get('post_id')\\n post = Posts.get_by_id(int(post_id))\\n user = self.get_active_user()\\n user_id = int(user.key().id())\\n\\n if liked:\\n if user_id in post.liked_by:\\n self.render_improper_endpoint_access(\\\"like\\\")\\n else:\\n if post.submitter_id != user_id:\\n post.liked_by.append(user.key().id())\\n post.put()\\n self.redirect('/%s' % str(post.key().id()))\\n else:\\n self.error(403)\\n elif unliked:\\n if user_id in post.liked_by:\\n index = post.liked_by.index(user_id)\\n del post.liked_by[index]\\n post.put()\\n self.redirect('/%s' % str(post.key().id()))\\n else:\\n self.error(500)\",\n \"def post_like(self, entry, **args):\\n args.update(entry=entry)\\n return self.fetch(\\\"/like\\\", post_args=args)\",\n \"def test_create_like(self):\\n pokemon = Pokemon.objects.all().first()\\n r = self.client.post(\\n reverse('create_like'),\\n {'pokemon-like-id': pokemon.id},\\n HTTP_X_REQUESTED_WITH='XMLHttpRequest'\\n )\\n\\n self.assertEqual(Like.objects.all().count(), 1)\",\n \"def test_new_likes(self):\\n\\n new_user = User.signup(\\\"newuser\\\", \\\"newuser@gmail.com\\\", \\\"password123\\\", None)\\n db.session.add(new_user)\\n db.session.commit()\\n\\n new_msg = Message(text=\\\"A new message from a new user\\\", user_id=new_user.id)\\n db.session.add(new_msg)\\n db.session.commit()\\n\\n new_user.likes.append(new_msg)\\n db.session.commit()\\n\\n likes = Likes.query.all()\\n self.assertEqual(len(likes), 1)\\n self.assertEqual(likes[0].user_id, new_user.id)\\n self.assertEqual(likes[0].message_id, new_msg.id)\",\n \"def add_like(id):\\r\\n username = request.json['username']\\r\\n duplicate_likes_query = models.Like.query.filter(models.Like.username == username, models.Like.post_id == id)\\r\\n # if like from this user to this tweet already exist\\r\\n if duplicate_likes_query.count() > 0:\\r\\n return {'like_id': duplicate_likes_query.first().like_id}\\r\\n\\r\\n # if original tweet does not exist -> 404\\r\\n models.Tweet.query.get_or_404(id)\\r\\n\\r\\n like = models.Like(post_id=id, username=username, timestamp=datetime.datetime.now())\\r\\n db.session.add(like)\\r\\n db.session.commit()\\r\\n\\r\\n return {'like_id': like.like_id}\",\n \"def test_like(self):\\n\\n message = Message(text=\\\"hello world\\\", user_id=self.testuser2.id)\\n \\n db.session.add(message)\\n db.session.commit()\\n \\n message_id = message.id\\n\\n with self.client as c:\\n with c.session_transaction() as session:\\n session[CURR_USER_KEY] = self.testuser.id\\n \\n response = c.post(f\\\"/messages/{message_id}/like\\\", follow_redirects=True)\\n \\n self.assertEqual(response.status_code, 200)\\n\\n likes = Likes.query.filter(Likes.message_id==message_id).all()\\n self.assertEqual(len(likes), 1)\\n self.assertEqual(likes[0].user_id, self.testuser.id)\",\n \"def like_event(request, pk):\\n template_var = base_template_vals(request)\\n previousLike = Like.objects.filter(event=Event.objects.get(id=pk), \\n user=UserProfile.objects.filter(\\n django_user=request.user)[0])\\n if(len(previousLike) == 0):\\n like = Like(event=Event.objects.get(id=pk))\\n like.user = UserProfile.objects.filter(django_user=request.user)[0]\\n like.save()\\n #sys notification\\n from_user = UserProfile.objects.get(django_user = User.objects.get(\\n username__exact='admin')) \\n to_user = UserProfile.objects.get(django_user=request.user)\\n event_id = pk\\n sys_notification(to_user, \\\"save_event\\\", from_user, event_id)\\n return redirect('index')\",\n \"def users_like_changed(sender, instance, **kwargs):\\n instance.total_likes = instance.users_like.count()\\n instance.save()\",\n \"def likes(self):\\r\\n return Likes(self)\",\n \"def like(request, pk):\\n update_posts_expiration()\\n post = 0;\\n post = Post.objects.get(id=pk)\\n #if the post is expired you user can't like it\\n if post.is_expired == True:\\n return Response(\\\"You can't interact with a expired post\\\")\\n else:\\n #if the post is not expired then increment the likes count by 1 and save a serializer of like to the database with user and post information\\n if request.method == \\\"POST\\\":\\n if request.user != post.user:\\n post.likes_count += 1\\n post.save()\\n serializer = LikeSerializer(data=request.data)\\n if serializer.is_valid():\\n serializer.save(user=request.user, liked_post=post)\\n return Response(\\\"you liked the post with title: \\\" + post.title)\\n else:\\n return Response(serializer.errors)\\n else:\\n return Response(\\\"you can't like your own post\\\")\\n \\n\\n return Response(\\\"you didn't like the post yet\\\")\",\n \"def artwork_like(request):\\n artwork_number = request.GET.get('id')\\n following_page = 'artwork_detail?id='+str(artwork_number)\\n if 'following' in request.GET:\\n following_page = request.GET.get('following')\\n artwork = Artwork.objects.get(id=artwork_number)\\n user = request.user\\n like = Like(user=user, artwork=artwork)\\n \\\"\\\"\\\" Save the like in the database, attached to the correct artwork piece using its unique ID \\\"\\\"\\\"\\n like.save()\\n return HttpResponseRedirect(following_page)\",\n \"def send_like(request):\\n if request.method == \\\"POST\\\":\\n if \\\"token\\\" in request.data and request.data[\\\"token\\\"] != \\\"\\\" and request.data[\\\"token\\\"] is not None:\\n if Token.objects.filter(key=request.data[\\\"token\\\"]).exists():\\n token = get_object_or_404(Token, key=request.data[\\\"token\\\"])\\n if Post.objects.filter(pk=request.data[\\\"post_id\\\"]).exists():\\n post = Post.objects.get(pk=request.data[\\\"post_id\\\"])\\n if Like.objects.filter(post=post, user_id=token.user_id).exists():\\n return Response({\\\"error\\\": 31})\\n else:\\n post.count_likes += 1\\n post.save()\\n like = Like.objects.create(post=post, user_id=token.user_id)\\n serializer = PostSerializer(post, context={'user_id': token.user_id})\\n UserFeed.objects.create(user=post.author,\\n action_user=token.user,\\n like=like,\\n action=\\\"Like\\\")\\n if post.author != token.user:\\n message = \\\"{} likes your post\\\".format(token.user.username)\\n custom = {\\n \\\"post_id\\\": post.id,\\n \\\"avatar\\\": UserProfile.objects.get(user=token.user).avatar.url\\n }\\n\\n user_notification = UserNotification.objects.get(user=post.author)\\n send_notification(custom, message, user_notification)\\n\\n return Response({\\\"success\\\": 30,\\n \\\"post\\\": serializer.data})\\n else:\\n return Response({\\\"error\\\": 32})\\n else:\\n return Response({\\\"error\\\": 17})\",\n \"def test_associate_user_likes(self):\\n\\n testlike1 = self.user1.likes.append(self.message2)\\n testlike2 = self.user2.likes.append(self.message1)\\n db.session.commit()\\n\\n likes = Likes.query.all()\\n\\n self.assertEqual(len(likes), 2)\\n self.assertEqual(likes[0].user_id, self.user1.id)\\n self.assertEqual(likes[1].user_id, self.user2.id)\",\n \"def test_like_activity(self):\\n from .mockers import user_status_context\\n from .mockers import subscribe_context, create_context\\n username = 'messi'\\n username_not_me = 'xavi'\\n self.create_user(username)\\n self.create_user(username_not_me)\\n self.create_context(create_context)\\n self.admin_subscribe_user_to_context(username, subscribe_context)\\n self.admin_subscribe_user_to_context(username_not_me, subscribe_context)\\n res = self.create_activity(username, user_status_context)\\n activity_id = res.json['id']\\n res = self.testapp.post('/activities/%s/likes' % activity_id, '', oauth2Header(username_not_me), status=201)\\n activity = self.testapp.get('/activities/%s' % activity_id, '', oauth2Header(username), status=200)\\n\\n self.assertEqual(res.json['object']['likes'][0]['username'], username_not_me)\\n self.assertEqual(res.json['object']['liked'], True)\\n self.assertEqual(res.json['object']['likesCount'], 1)\\n\\n self.assertEqual(activity.json['likes'][0]['username'], username_not_me)\\n self.assertEqual(activity.json['liked'], False)\\n self.assertEqual(activity.json['likesCount'], 1)\",\n \"def d_like(request):\\n token = request.GET.get('token','')\\n\\n if not token:\\n return HttpResponse('error in user validation')\\n \\n try:\\n api = ApiKey.objects.get(key=token)\\n user = api.user\\n except ApiKey.DoesNotExist:\\n return HttpResponse('error in user validation')\\n if request.method == 'POST' and user.is_active:\\n try:\\n post_id = int(request.POST['post_id'])\\n except ValueError:\\n return HttpResponse('erro in post id')\\n try:\\n post = Post.objects.get(pk=post_id)\\n except Post.DoesNotExist:\\n return HttpResponse('post not found')\\n\\n like, created = Likes.objects.get_or_create(user=user, post=post)\\n if created:\\n post.like=post.like+1\\n post.save()\\n return HttpResponse('+1')\\n elif like:\\n like.delete()\\n post.like=post.like-1\\n post.save()\\n return HttpResponse('-1')\\n\\n return HttpResponse('error in parameters')\",\n \"def test_associate_message_likes(self):\\n\\n testlike1 = self.user2.likes.append(self.message2)\\n testlike2 = self.user2.likes.append(self.message1)\\n db.session.commit()\\n\\n likes = Likes.query.all()\\n self.assertEqual(len(likes), 2)\\n self.assertEqual(likes[0].message_id, self.message2.id)\\n self.assertEqual(likes[1].message_id, self.message1.id)\",\n \"def update_likes(self):\\n self.nb_likes = self.likes.count()\\n self.save()\",\n \"def post_likes(request):\\n if request.method == 'POST':\\n if not request.user.is_authenticated:\\n raise PermissionDenied\\n pk = request.POST.get('post_id')\\n try:\\n post = PostsModel.objects.get(id=pk)\\n except PostsModel.DoesNotExist:\\n post = None\\n try:\\n likes = LikesModel.objects.get(user=request.user, post=post)\\n except LikesModel.DoesNotExist:\\n likes = None\\n\\n\\n if likes == None:\\n likes = LikesModel.objects.create(user=request.user, post=post)\\n post.likes_count = F('likes_count') + 1\\n post.save(update_fields=[\\\"likes_count\\\"])\\n class_likes = 'fa fa-heart mx-2'\\n else:\\n if likes.is_liked == True:\\n likes.is_liked = False\\n likes.save()\\n post.likes_count = F('likes_count') - 1\\n post.save(update_fields=[\\\"likes_count\\\"])\\n class_likes = 'fa fa-heart-o mx-2'\\n elif likes.is_liked == False:\\n likes.is_liked = True\\n likes.save()\\n post.likes_count = F('likes_count') + 1\\n post.save(update_fields=[\\\"likes_count\\\"])\\n class_likes = 'fa fa-heart mx-2'\\n post.refresh_from_db()\\n response_data = {'_code' : 0, '_status' : 'ok', '_likes': post.likes_count, '_class_likes': class_likes}\\n else:\\n response_data = {'_code' : 1, '_status' : 'no' }\\n\\n return JsonResponse(response_data)\",\n \"def test_not_create_like_for_current_anon_already_liked(self):\\n pokemon = Pokemon.objects.all().first()\\n\\n r = self.client.post(\\n reverse('create_like'),\\n {'pokemon-like-id': pokemon.id},\\n HTTP_X_REQUESTED_WITH='XMLHttpRequest'\\n )\\n\\n r = self.client.post(\\n reverse('create_like'),\\n {'pokemon-like-id': pokemon.id},\\n HTTP_X_REQUESTED_WITH='XMLHttpRequest'\\n )\\n\\n self.assertEqual(Like.objects.all().count(), 1)\",\n \"def on_created_comment(sender, instance: dillo.models.comments.Comment, created, **kwargs):\\n if not created:\\n return\\n # Extract tags and mentions from text and assign them to the Comment\\n tags, mentions = extract_tags_and_mentions(instance.content)\\n instance.tags.set(*tags)\\n for mentioned_user in mentions:\\n log.debug('Mentioning user %s in Comment %i' % (mentioned_user, instance.id))\\n dillo.models.mixins.Mentions.objects.create(\\n user=mentioned_user,\\n content_object=instance,\\n )\\n # TODO(fsiddi) Generate activity about mention\\n\\n # Generate activity about comment creation\\n log.debug('Generating activity about comment creation')\\n verb = 'commented'\\n if instance.parent_comment:\\n verb = 'replied'\\n action.send(instance.user, verb=verb, action_object=instance, target=instance.entity)\\n log.debug('Set user %s as follower of own comment %i' % (instance.user, instance.id))\\n follow(instance.user, instance, actor_only=False)\",\n \"def add_liked_track():\\n # Inserts the track information.\\n new_liked_by = request.vars.liked_by\\n new_liked_title = request.vars.liked_title\\n t_id = db.liked_track.insert(\\n liked_by = new_liked_by,\\n liked_title = new_liked_title\\n )\\n return response.json(dict(\\n id = t_id,\\n liked_by = new_liked_by,\\n liked_title = new_liked_title\\n ))\",\n \"def test_like_activity_by_various(self):\\n from .mockers import user_status_context\\n from .mockers import subscribe_context, create_context\\n username = 'messi'\\n username_not_me = 'xavi'\\n self.create_user(username)\\n self.create_user(username_not_me)\\n self.create_context(create_context)\\n self.admin_subscribe_user_to_context(username, subscribe_context)\\n self.admin_subscribe_user_to_context(username_not_me, subscribe_context)\\n res = self.create_activity(username, user_status_context)\\n activity_id = res.json['id']\\n res = self.testapp.post('/activities/%s/likes' % activity_id, '', oauth2Header(username_not_me), status=201)\\n res = self.testapp.post('/activities/%s/likes' % activity_id, '', oauth2Header(username), status=201)\\n\\n self.assertEqual(res.json['object']['likes'][0]['username'], username_not_me)\\n self.assertEqual(res.json['object']['likes'][1]['username'], username)\\n self.assertEqual(res.json['object']['liked'], True)\\n self.assertEqual(res.json['object']['likesCount'], 2)\",\n \"def on_created_post(sender, instance, created, **kwargs):\\n if not created:\\n return\\n if not issubclass(sender, dillo.models.posts.Post):\\n return\\n instance.hash_id = instance.id\\n instance.save()\\n log.debug('Set user %s as follower of own post %i' % (instance.user, instance.id))\\n follow(instance.user, instance, actor_only=False)\",\n \"def post(self, request, *args, **kwargs):\\n\\n user_wall_post_comment = self.get_object()\\n user_wall_post_comment.likes.add(self.request.user)\\n return Response(status=201)\",\n \"def test_like_already_liked_activity(self):\\n from .mockers import user_status_context\\n from .mockers import subscribe_context, create_context\\n username = 'messi'\\n username_not_me = 'xavi'\\n self.create_user(username)\\n self.create_user(username_not_me)\\n self.create_context(create_context)\\n self.admin_subscribe_user_to_context(username, subscribe_context)\\n self.admin_subscribe_user_to_context(username_not_me, subscribe_context)\\n res = self.create_activity(username, user_status_context)\\n activity_id = res.json['id']\\n res = self.testapp.post('/activities/%s/likes' % activity_id, '', oauth2Header(username_not_me), status=201)\\n res = self.testapp.post('/activities/%s/likes' % activity_id, '', oauth2Header(username_not_me), status=200)\\n\\n self.assertEqual(res.json['object']['likes'][0]['username'], username_not_me)\\n self.assertEqual(res.json['object']['liked'], True)\\n self.assertEqual(res.json['object']['likesCount'], 1)\",\n \"def test_update_like_can_message(self):\\n\\n self.assertEqual(first=0, second=self.like.likes.all().count())\\n url = reverse('like-detail', args=(self.like.id,))\\n data = {\\n 'likes': [2]\\n }\\n json_data = json.dumps(data)\\n self.client.credentials(HTTP_AUTHORIZATION=self.token)\\n self.like_1.likes.add(self.user)\\n response = self.client.patch(path=url, content_type='application/json', data=json_data)\\n self.assertTrue(response.data.get('can_message', False))\\n self.assertEqual(first=200, second=response.status_code)\\n self.assertEqual(first=1, second=self.like.likes.all().count())\",\n \"def test_like_topic_viewset(self):\\n\\n topic = TopicFactory(author=self.user)\\n response = self.client.post(reverse('api:topics-like', kwargs={'topic_id': topic.id}))\\n self.assertEqual(response.status_code, status.HTTP_201_CREATED)\\n obj_type = ContentType.objects.get_for_model(topic)\\n likes = Like.objects.filter(\\n content_type=obj_type, object_id=topic.id, user=self.user)\\n self.assertTrue(likes.exists())\",\n \"def test_a_user_can_like_article(self):\\n article = self.create_article()\\n\\n slug = article.data['data']['slug']\\n like = self.client.post('/api/articles/{}/like/'.format(slug),\\n HTTP_AUTHORIZATION='Bearer ' +\\n self.token,\\n format='json')\\n\\n self.assertEqual(like.status_code, 200)\",\n \"def test_like_mission(self):\\n\\n with self.client as c:\\n with c.session_transaction() as sess:\\n sess['user_id'] = self.user2.id\\n\\n db.session.add(self.mission)\\n resp = c.post(f'/v1/mission/like{self.mission.id}')\\n\\n self.assertEqual(resp.json['success'], 'added')\\n mission = Mission.query.get(self.mission.id)\\n self.assertEqual(mission.likes, {0, self.user2.id})\",\n \"def like_photo(self):\\n like_button = self.driver.find_elements_by_class_name('wpO6b')[0]\\n like_button.click()\",\n \"def addLikeCount(self,count):\\n self.interactionCount += count\\n return None\",\n \"def test_authorized_user_like_field(self):\\n lesson_id = 1\\n self.create_and_auth(email='user@gmail.com',\\n password='qwerty_1234')\\n\\n response = self.client.get(reverse('lessons-detail',\\n args=(lesson_id,)))\\n self.assertEqual(response.status_code, status.HTTP_200_OK)\\n self.assertIn('like', response.data)\\n self.assertFalse(response.data['like'])\\n\\n self.client.post(reverse('lesson_likes', kwargs={\\n 'lesson_id': lesson_id\\n }))\\n\\n response = self.client.get(reverse('lessons-detail',\\n args=(lesson_id,)))\\n self.assertEqual(response.status_code, status.HTTP_200_OK)\\n self.assertIn('like', response.data)\\n self.assertTrue(response.data['like'])\",\n \"def like_tweet(self, tag):\\n self.bot.get('https://twitter.com/search?q=' + tag + '&src=typed')\\n self.__wait(3, 3)\\n for i in range(1, 3):\\n self.bot.execute_script('window.scrollTo(0,document.body.scrollHeight)')\\n self.__wait(2, 3)\\n tweets = self.bot.find_elements_by_tag_name('article')\\n\\n links = []\\n for tweet in tweets:\\n sub_links = tweet.find_elements_by_tag_name('a')\\n links += [sub_link.get_attribute('href')\\n for sub_link in sub_links if 'status' in sub_link.get_attribute('href')]\\n\\n print('Started to like {} tweets'.format(len(links)))\\n\\n for link in links:\\n self.bot.get(link)\\n self.__wait(3, 5)\\n likes = self.bot.find_elements_by_css_selector('div[data-testid=\\\"like\\\"')\\n for like in likes:\\n like.click()\\n self.__wait(3, 5)\",\n \"def process_IN_CREATE(self, event):\",\n \"def create(self):\\n ...\",\n \"def add_to_like(request):\\n if request.is_ajax():\\n model_id = request.POST.get('model_id')\\n model_type = request.POST.get('model_type')\\n content_type = ContentType.objects.get(model=model_type)\\n ModelType = content_type.model_class()\\n model_type_qs = get_object_or_404(ModelType, id=model_id)\\n obj = LikeDislike.objects.create_for_instance_model(\\n instance=model_type_qs, request=request, likedislike='like')\\n logger.info(f\\\"{model_type_qs} {obj.likedislike}\\\")\\n return JsonResponse({\\\"response\\\": f\\\"شما {model_type_qs.title} را پسندیدید\\\"}, status=200)\\n return HttpResponseRedirect(request.META.get(\\\"HTTP_REFERER\\\"))\",\n \"def create(self, *args, **kwargs):\\n pass\",\n \"def following_changed(sender, action, instance, *args, **kwargs):\\n\\n # m2mchanged.connect specified in apps.py\\n\\n following = instance.following.all()\\n creator = instance.user\\n\\n if creator in following:\\n raise ValidationError (\\\"can't like own post\\\")\",\n \"def like_view(request, pk):\\n cocktail = get_object_or_404(Cocktail, id=pk)\\n liked = False\\n if cocktail.likes.filter(id=request.user.id).exists():\\n cocktail.likes.remove(request.user)\\n liked = False\\n else:\\n cocktail.likes.add(request.user)\\n liked = True\\n\\n return HttpResponseRedirect(reverse('cocktails:cocktail-details', args=[str(pk)]))\",\n \"def like_comment(username):\\n\\n user_ID = before_request()\\n user_ID = None\\n if user_ID != None:\\n user_ID = str(g.user['_id'])\\n\\n message_id = request.args.get('message_id')\\n profile_usr = request.args.get('profile_usr')\\n count = float(1)\\n\\n redis_obj.zincrby('add_like', message_id, count)\\n\\n print \\\"message ID\\\", message_id\\n\\n if redis_obj.get(user_ID):\\n return redirect(url_for('timeline', username=profile_usr, userId=pickle.loads(redis_obj.get(user_ID))))\\n else:\\n \\\"\\\"\\\" Invalidating cache \\\"\\\"\\\"\\n redis_obj.delete('leaderboard-key')\\n redis_obj.delete(session['user_id'])\\n print \\\"Invalidating cache after Like\\\"\\n return redirect(url_for('timeline', username=profile_usr))\",\n \"def __init__(self, author, title, text=\\\"\\\", haiku=\\\"False\\\"):\\n self.author = author\\n self.title = title\\n self.text = text\\n self.haiku = haiku\\n self.name = self\\n Poem.obj_created += 1\",\n \"def like_tweets(pos_tweets):\\n\\n for tweet in pos_tweets:\\n twitter.CreateFavorite(status_id=tweet.id)\\n\\n return\",\n \"def create(self):\\n\\n pass\",\n \"def test_is_liked_by(self):\\n\\n u = self.user\\n u2 = User(\\n email=\\\"test2@test.com\\\" ,\\n username=\\\"testuser2\\\" ,\\n password=\\\"HASHED_PASSWORD\\\"\\n )\\n\\n m = Message(\\n user_id=u.id ,\\n text='test content'\\n )\\n\\n db.session.add_all([u2 , m])\\n db.session.commit()\\n\\n #not like \\n self.assertEqual(len(u2.likes) , 0)\\n self.assertEqual(len(m.likes_users) , 0)\\n self.assertEqual(u2.is_like(m) , False)\\n self.assertEqual(m.is_liked_by(u2) , False)\\n\\n like = Likes(user_id=u2.id , message_id=m.id)\\n db.session.add(like)\\n db.session.commit()\\n\\n self.assertEqual(len(u2.likes) , 1)\\n self.assertEqual(len(m.likes_users) , 1)\\n self.assertEqual(u2.is_like(m) , True)\\n self.assertEqual(m.is_liked_by(u2), True)\",\n \"def order_posts_to_likes(sender, instance, created=False, **kwargs):\\n user = instance.user\\n post = instance.post\\n tags = post.tags.all()\\n\\n likes = Like.objects.filter(user=user, post=post)\\n liked_posts = [ i.post for i in likes if i.liked_status==True]\\n\\n for post in liked_posts:\\n liked_tags = post.tags.all()\\n for tag in liked_tags:\\n if not tag in tags:\\n tags.append(tag)\\n\\n similar_posts = Post.objects.filter(tags__in=tags).distinct()\\n try:\\n UserPostWeight.objects.filter(user=user).delete()\\n except:\\n pass\\n weight = 2\\n for post in similar_posts:\\n UserPostWeight.objects.create(user=user, post=post, weight=weight)\\n weight+=2\\n \\n\\n\\n pass\",\n \"def like(id):\\n spinner = Halo(text=\\\"Liking photo...\\\", spinner=\\\"dots\\\").start()\\n try:\\n api.like(id)\\n spinner.succeed(\\\"Liked\\\")\\n except Exception:\\n spinner.fail(\\\"Failed to like photo.\\\")\",\n \"def get(self, post_id):\\n Post.add_like(int(post_id), self.user.get_id())\\n self.redirect('/blog')\",\n \"def create(self):\\n pass\",\n \"def create(self):\\n pass\",\n \"def create(self):\\n pass\",\n \"def process_like(self):\\n medias = self.prepare_process_like()\\n wait_time = 3600//(self.limits_per_hour.get('like') + 1)\\n for m in medias:\\n time.sleep(wait_time + trunc_gauss(0, 10, -40, 40))\\n self.liking(m)\",\n \"def on_created_action(sender, instance: models_actstream.Action, created, **kwargs):\\n if not created:\\n return\\n\\n # If content_type is Post or Profile\\n if (\\n instance.action_object_content_type\\n == ContentType.objects.get_for_model(dillo.models.posts.Post)\\n and instance.verb == 'posted'\\n ) or (\\n instance.action_object_content_type\\n == ContentType.objects.get_for_model(dillo.models.profiles.Profile)\\n and instance.verb == 'updated their reel'\\n ):\\n dillo.tasks.feeds.establish_time_proximity(instance)\\n\\n log.debug('Creating background fanout operation for action %i' % instance.id)\\n dillo.tasks.feeds.activity_fanout_to_feeds(instance.id)\",\n \"def process_like_and_follow(self):\\n follow, media, unfollow = self.prepare_process_like_and_follow()\\n follow_acts, media_acts, unfollow_acts = len(follow), len(media), len(unfollow)\\n all_acts = round(self.limits_per_hour.get('follow') + self.limits_per_hour.get('like') +\\n self.limits_per_hour.get('unfollow'))\\n wait_time = 3600 // all_acts + 1\\n while follow_acts or media_acts or unfollow_acts:\\n time.sleep(wait_time + trunc_gauss(0, 10, -40, 40))\\n rc = random.choices(['f', 'l', 'u'], [follow_acts, media_acts, unfollow_acts])[0]\\n\\n if rc == 'f':\\n fo = follow.pop(0)\\n self.following_and_storing(fo)\\n follow_acts -= 1\\n elif rc == 'l':\\n mo = media.pop(0)\\n self.liking(mo)\\n media_acts -= 1\\n elif rc == 'u':\\n uo = unfollow.pop(0)\\n self.unfollowing_and_removing(uo)\\n unfollow_acts -= 1\",\n \"def like(request, post_id):\\n if request.method == \\\"PUT\\\":\\n liked = None\\n user = request.user\\n post = Post.objects.get(id=post_id)\\n # If user already liked, decrement the like count and remove as 'liker'\\n if user in post.liked_by.all():\\n post.liked_by.remove(user)\\n post.likes -= 1\\n post.save()\\n liked = False\\n # Else increase like count and add user\\n else:\\n post.liked_by.add(user)\\n post.likes += 1\\n post.save()\\n liked = True\\n # Return data for updating dynamically with javascript\\n return JsonResponse({\\\"total_likes\\\": post.likes, \\\"liked\\\": liked})\",\n \"def test_is_liked_by_method(self):\\n\\n self.u2.likes.append(self.m_u1)\\n db.session.commit()\\n \\n self.assertTrue(self.m_u1.is_liked_by(self.u2))\\n self.assertFalse(self.m_u1.is_liked_by(self.u))\",\n \"def post_liker(a, args):\\n if args.likes.filter(user=a).exists():\\n exists = True\\n else:\\n exists = False\\n return exists\",\n \"def sample(self, like_params):\\n\\t\\traise NotImplementedError\",\n \"def test_a_user_can_delete_like(self):\\n article = self.create_article()\\n\\n slug = article.data['data']['slug']\\n self.client.post('/api/articles/{}/like/'.format(slug),\\n HTTP_AUTHORIZATION='Bearer ' +\\n self.token,\\n format='json')\\n like = self.client.post('/api/articles/{}/like/'.format(slug),\\n HTTP_AUTHORIZATION='Bearer ' +\\n self.token,\\n format='json')\\n\\n self.assertEqual(like.status_code, 200)\",\n \"def LikeArticle(request):\\n try:\\n if not request.user.is_authenticated():\\n return HttpResponse(json.dumps({\\n \\\"result\\\": \\\"1\\\"\\n }))\\n\\n like = request.GET.get('like', '')\\n article_id = request.GET.get('article_id', '')\\n\\n if str(like).lower() != 'true':\\n like = 'false'\\n else:\\n like = 'true'\\n \\n record = UserLikedArticles.objects.filter(user=request.user.id, article=article_id)\\n\\n if like == 'false':\\n if record != None and len(record) > 0:\\n record[0].delete()\\n else:\\n if record == None or len(record) == 0:\\n like = UserLikedArticles()\\n like.user = request.user.id\\n like.article=Article.objects.filter(id=article_id)[0]\\n like.save()\\n \\n article = Article.objects.filter(id=article_id)[0]\\n article.likes = len(UserLikedArticles.objects.filter(article=article_id))\\n article.save()\\n \\n return HttpResponse(json.dumps({\\n \\\"result\\\": 0\\n }))\\n except Exception, e:\\n return HttpResponse(json.dumps({\\n \\\"result\\\": e\\n }))\",\n \"def post_tweet(instance, created, raw, **kwargs):\\n if created and not raw:\\n _twitter.statuses.update(status='%s %s' % (\\n instance.title,\\n instance.url,\\n ))\",\n \"def test_unlike(self):\\n\\n message = Message(text=\\\"hello world\\\", user_id=self.testuser2.id)\\n \\n db.session.add(message)\\n db.session.commit()\\n\\n like = Likes(message_id=message.id, user_id=self.testuser.id)\\n \\n db.session.add(like)\\n db.session.commit()\\n \\n message_id = message.id\\n\\n with self.client as c:\\n with c.session_transaction() as session:\\n session[CURR_USER_KEY] = self.testuser.id\\n \\n response = c.post(f\\\"/messages/{message_id}/like\\\", follow_redirects=True)\\n \\n self.assertEqual(response.status_code, 200)\\n\\n likes = Likes.query.filter(Likes.message_id==message_id).all()\\n self.assertEqual(len(likes), 0)\",\n \"def created(self, created):\\n\\n self._created = created\",\n \"def created(self, created):\\n\\n self._created = created\",\n \"def created(self, created):\\n\\n self._created = created\",\n \"def created(self, created):\\n\\n self._created = created\",\n \"def created(self, created):\\n\\n self._created = created\",\n \"def created(self, created):\\n\\n self._created = created\",\n \"def created(self, created):\\n\\n self._created = created\",\n \"def created(self, created):\\n\\n self._created = created\",\n \"def created(self, created):\\n\\n self._created = created\",\n \"def created(self, created):\\n\\n self._created = created\",\n \"def created(self, created):\\n\\n self._created = created\",\n \"def created(self, created):\\n\\n self._created = created\",\n \"def created(self, created):\\n\\n self._created = created\",\n \"def like(self, event, number):\\n\\n if not hasattr(self, 'user_id') or not hasattr(self, 'top_id'):\\n vk_common.dprint(2, vk_common.func_name(), f\\\"Не заданы атрибуты user_id = или top_id.\\\")\\n return\\n\\n item_id = self.item_ids[self.top_id-1]\\n if item_id is None:\\n vk_common.dprint(2, vk_common.func_name(), f\\\"Атрибут item_id = None.\\\")\\n return\\n else:\\n vk_common.dprint(2, vk_common.func_name(), f\\\"number = {number}; item_id = {item_id}\\\")\\n\\n vk_common.dprint(1, f\\\"имя файла с фото = \\\\\\\"{self.user_id}_{self.top_id}\\\\\\\"; \\\"\\n f\\\"ссылка = https://vk.com/id{self.user_id}\\\")\\n\\n err, msg, likes = vk_common.like_photo(self.user_id, item_id)\\n if err != 0:\\n vk_common.dprint(2, \\\"Не смогли поставить лайк.\\\")\\n mb.showinfo(\\\"Информация\\\", f\\\"Не смогли поставить лайк фото пользователя c id = {self.user_id}.\\\")\\n return\\n vk_common.dprint(2, f\\\"Поставили лайк: owner_id = {self.user_id}; item_id = {item_id}\\\")\\n mb.showinfo(\\\"Информация\\\", f\\\"Поставили лайк фото пользователя c id = {self.user_id}.\\\")\\n # err, msg, likes = user.like_photo(234068204, 345376029)\",\n \"def log_create(sender, instance, created, **kwargs):\\n if created:\\n stracks.user(instance).log(\\\"? has been created\\\")\",\n \"def create(self):\",\n \"def on_create(self):\",\n \"def like_post(self,id):\\n try:\\n data = self.api.create_favorite(id)\\n if isinstance(data,tweepy.models.Status):\\n if hasattr(data,'_json'): \\n return json.dumps(data._json)\\n except tweepy.error.TweepError as e:\\n print(e.__dict__)\\n if e.__dict__['api_code']==139:\\n data = unlike_post(id)\\n return data\",\n \"def add_like(db, filename, usernick=None):\\n cur = db.cursor()\\n\\n sql = \\\"\\\"\\\"\\n select exists(\\n select 1 from users where nick=?\\n );\\n \\\"\\\"\\\"\\n cur.execute(sql, (usernick,))\\n user_exists = cur.fetchone()[0]\\n\\n if user_exists==1: # or usernick is None\\n sql = \\\"\\\"\\\"\\n select exists(\\n select 1 from images where filename=?\\n );\\n \\\"\\\"\\\"\\n cur.execute(sql, (filename,))\\n img_exists = cur.fetchone()[0]\\n if img_exists==1:\\n\\n sql = \\\"\\\"\\\"\\n insert into likes values (?, ?);\\n \\\"\\\"\\\"\\n cur.execute(sql, (filename, usernick))\\n db.commit()\",\n \"def test_like(self):\\n client = Client()\\n slug = ['bryan-fox-snowboard-2017', 'some_slug'] # Correct and incorrect slug\\n response = client.get('/products/{0}/'.format(slug[0]))\\n self.assertEqual(response.status_code, 200)\\n Like.objects.filter(product__slug=slug).count() # numbers like of post\\n product = Product.objects.get(slug=slug[0])\\n if response.context[\\\"user\\\"].is_authenticated():\\n product.like_set.filter(user=response.context[\\\"user\\\"])\",\n \"def help_create(self):\\n print(\\\"create instances\\\")\",\n \"def cursor_created(self, cursor, action, *args, **kw):\\n pass\",\n \"def like_following(self):\\n self.logger.log(\\\"starting like_following...\\\")\\n count_following = self.account.follows_count\\n follows_accounts = self.following\\n random.shuffle(follows_accounts)\\n for acc in follows_accounts:\\n acc = perform_with_ran_delay(self.instagram.get_account_by_id, acc)\\n self.logger.log(\\\" {} > {} posts\\\".format(acc.username, acc.media_count))\\n if acc.media_count > 0:\\n\\n posts = perform_with_ran_delay(self.instagram.get_medias, acc.username, 50)\\n if posts:\\n for m in posts:\\n try:\\n perform_with_ran_delay(self.instagram.like, m.identifier)\\n self.logger.log(\\\"liking 1 post from \\\"+acc.username)\\n random_delay()\\n except Exception as e:\\n self.logger.log(\\\"skipping 1 post from \\\"+acc.username)\\n self.logger.log(e)\\n random_delay()\\n continue\",\n \"def prepare_process_like_and_follow(self):\\n follow = []\\n media = []\\n unfollow = []\\n\\n coef = self.users_to_follow / self.limits_per_hour.get('follow', 1)\\n media_to_like = round(coef*self.limits_per_hour.get('like'))\\n num_to_unfollow = round(coef*self.limits_per_hour.get('unfollow'))\\n\\n feed_likes = media_to_like // 2\\n feed_likes_list = []\\n following_likes = round((media_to_like // 2) * 3 / 4)\\n following_likes_list = []\\n followers_likes = media_to_like - feed_likes - following_likes\\n\\n monitored_ids = [i[\\\"user\\\"] for i in self.monitored_users]\\n\\n for posts in self.hashtag_feed_list(self.search_hashtags):\\n if len(follow) < self.users_to_follow:\\n for m in posts:\\n if self.check_if_suit(m):\\n user_id, username = self.get_user_from_post(m)\\n if user_id and user_id not in [i[\\\"user\\\"] for i in follow] \\\\\\n and user_id not in monitored_ids:\\n follow.append({'user': user_id, 'username': username})\\n following_likes_list.append(m)\\n\\n if len(follow) >= self.users_to_follow:\\n break\\n\\n for p in following_likes_list:\\n if p in posts:\\n posts.remove(p)\\n\\n if feed_likes > 0:\\n if len(posts) > feed_likes:\\n feed_likes_list.extend([i['id'] for i in (random.choice(posts) for _ in range(feed_likes))\\n if i['id'] not in feed_likes_list])\\n else:\\n feed_likes_list.extend([i['id'] for i in posts[:feed_likes] if i['id'] not in feed_likes_list])\\n feed_likes -= len(feed_likes_list)\\n if feed_likes <= 0:\\n if len(follow) >= self.users_to_follow:\\n break\\n if len(follow) >= self.users_to_follow and feed_likes <= 0:\\n break\\n\\n media.extend(feed_likes_list)\\n\\n if len(following_likes_list) < following_likes:\\n followings = []\\n get_n_followings = following_likes - len(following_likes_list)\\n if following_likes_list:\\n following_likes_list = [self.get_media_id_from_post(i) for i in following_likes_list]\\n following_likes_list.extend([i for i in self.get_following_likes(followings, get_n_followings)\\n if i and i not in media])\\n media.extend(following_likes_list)\\n else:\\n media.extend([self.get_media_id_from_post(i) for i in following_likes_list[:following_likes]])\\n\\n media.extend([i for i in self.get_followers_likes(followers_likes) if i and i not in media])\\n\\n unfollow = self.get_to_unfollow(unfollow, num_to_unfollow)\\n\\n return follow, media, unfollow\",\n \"def on_saved_post(sender, instance, created, **kwargs):\\n # If the post was just created, stop here since we have no tags yet\\n if created:\\n return\\n if not issubclass(sender, dillo.models.posts.Post):\\n return\\n # Extract tags and mentions from text and assign them to the Post\\n tags, mentions = extract_tags_and_mentions(instance.title)\\n instance.tags.set(*tags)\\n # Delete all existing mentions\\n dillo.models.mixins.Mentions.objects.filter(\\n content_type_id=instance.content_type_id, object_id=instance.id\\n ).delete()\\n for mentioned_user in mentions:\\n log.debug('Mentioning user %s in Post %i' % (mentioned_user, instance.id))\\n dillo.models.mixins.Mentions.objects.create(\\n user=mentioned_user,\\n content_object=instance,\\n )\",\n \"def like_movie(request):\\n\\n movie = get_object_or_404(Movie, id=request.POST.get('movie_id'))\\n is_liked = False\\n if Like.objects.filter(user=request.user, movie=movie).exists():\\n Like.objects.get(user=request.user, movie=movie).delete()\\n is_liked = False\\n else:\\n new = Like(user=request.user, movie=movie)\\n new.save()\\n if Dislike.objects.filter(user=request.user, movie=movie).count():\\n Dislike.objects.get(user=request.user, movie=movie).delete()\\n is_liked = True\\n return HttpResponseRedirect(movie.get_absolute_url())\"\n]","isConversational":false},"negative_scores":{"kind":"list like","value":["0.6593319","0.6169452","0.6148844","0.6112476","0.6108443","0.6105551","0.60661376","0.60419214","0.6040893","0.59646094","0.59646094","0.59571815","0.5945115","0.5939536","0.5915928","0.5865272","0.58126223","0.5807308","0.5788529","0.57406265","0.57226074","0.57064927","0.5669241","0.5642736","0.56357634","0.5594599","0.5582276","0.55618525","0.5548304","0.55464184","0.55239004","0.55127364","0.5492005","0.5487842","0.54525405","0.5445634","0.5437827","0.5427737","0.54225945","0.54147553","0.5411805","0.5385287","0.5362955","0.5345345","0.5341431","0.5339212","0.5323452","0.5319627","0.53021246","0.5270436","0.52651584","0.52562314","0.5251975","0.52496314","0.52386326","0.5215764","0.5209985","0.5200202","0.519358","0.51725554","0.5167937","0.5167937","0.5167937","0.51657695","0.5165095","0.51465756","0.5146465","0.51252115","0.5118905","0.5114063","0.5111687","0.5099054","0.5087212","0.5085984","0.50724626","0.50724626","0.50724626","0.50724626","0.50724626","0.50724626","0.50724626","0.50724626","0.50724626","0.50724626","0.50724626","0.50724626","0.50724626","0.50564903","0.5046971","0.50377893","0.50322646","0.5031709","0.50284326","0.5021658","0.5012041","0.50110614","0.5004727","0.5004375","0.50023675","0.49979663"],"string":"[\n \"0.6593319\",\n \"0.6169452\",\n \"0.6148844\",\n \"0.6112476\",\n \"0.6108443\",\n \"0.6105551\",\n \"0.60661376\",\n \"0.60419214\",\n \"0.6040893\",\n \"0.59646094\",\n \"0.59646094\",\n \"0.59571815\",\n \"0.5945115\",\n \"0.5939536\",\n \"0.5915928\",\n \"0.5865272\",\n \"0.58126223\",\n \"0.5807308\",\n \"0.5788529\",\n \"0.57406265\",\n \"0.57226074\",\n \"0.57064927\",\n \"0.5669241\",\n \"0.5642736\",\n \"0.56357634\",\n \"0.5594599\",\n \"0.5582276\",\n \"0.55618525\",\n \"0.5548304\",\n \"0.55464184\",\n \"0.55239004\",\n \"0.55127364\",\n \"0.5492005\",\n \"0.5487842\",\n \"0.54525405\",\n \"0.5445634\",\n \"0.5437827\",\n \"0.5427737\",\n \"0.54225945\",\n \"0.54147553\",\n \"0.5411805\",\n \"0.5385287\",\n \"0.5362955\",\n \"0.5345345\",\n \"0.5341431\",\n \"0.5339212\",\n \"0.5323452\",\n \"0.5319627\",\n \"0.53021246\",\n \"0.5270436\",\n \"0.52651584\",\n \"0.52562314\",\n \"0.5251975\",\n \"0.52496314\",\n \"0.52386326\",\n \"0.5215764\",\n \"0.5209985\",\n \"0.5200202\",\n \"0.519358\",\n \"0.51725554\",\n \"0.5167937\",\n \"0.5167937\",\n \"0.5167937\",\n \"0.51657695\",\n \"0.5165095\",\n \"0.51465756\",\n \"0.5146465\",\n \"0.51252115\",\n \"0.5118905\",\n \"0.5114063\",\n \"0.5111687\",\n \"0.5099054\",\n \"0.5087212\",\n \"0.5085984\",\n \"0.50724626\",\n \"0.50724626\",\n \"0.50724626\",\n \"0.50724626\",\n \"0.50724626\",\n \"0.50724626\",\n \"0.50724626\",\n \"0.50724626\",\n \"0.50724626\",\n \"0.50724626\",\n \"0.50724626\",\n \"0.50724626\",\n \"0.50724626\",\n \"0.50564903\",\n \"0.5046971\",\n \"0.50377893\",\n \"0.50322646\",\n \"0.5031709\",\n \"0.50284326\",\n \"0.5021658\",\n \"0.5012041\",\n \"0.50110614\",\n \"0.5004727\",\n \"0.5004375\",\n \"0.50023675\",\n \"0.49979663\"\n]","isConversational":false},"document_score":{"kind":"string","value":"0.74319637"},"document_rank":{"kind":"string","value":"0"}}},{"rowIdx":93284,"cells":{"query":{"kind":"string","value":"Decrease user profile likes of 1. The function is available as standalone for easier unit testing."},"document":{"kind":"string","value":"def profile_likes_count_decrease(target_user: User):\n try:\n with transaction.atomic():\n dillo.models.profiles.Profile.objects.filter(user=target_user).update(\n likes_count=F('likes_count') - 1\n )\n except IntegrityError:\n log.warning('Integrity error when incrementing likes count for user %i' % target_user.id)\n target_user.profile.recalculate_likes()"},"metadata":{"kind":"string","value":"{\n \"objective\": {\n \"self\": [],\n \"paired\": [],\n \"triplet\": [\n [\n \"query\",\n \"document\",\n \"negatives\"\n ]\n ]\n }\n}"},"negatives":{"kind":"list like","value":["def on_deleted_like(sender, instance: dillo.models.mixins.Likes, **kwargs):\n if not instance.content_object:\n return\n target_user = instance.content_object.user\n profile_likes_count_decrease(target_user)\n log.debug('Decreased like count for user %s' % target_user)","def unlike(self, data_base, user):\n cursor = data_base.cursor()\n cursor.execute(f\"UPDATE post SET likes = likes - 1 WHERE id = '{self.id}'\") # Decrements the likes\n cursor.execute(f\"DELETE FROM user_like WHERE user_id = {user.id} AND post_id = {self.id}\")\n if self.commit_to_db:\n data_base.commit()\n cursor.close()","def get_user_likes(self) -> int:\n return -1","def _decrement_num_user_stars(user_id, match_id, now):\n\tnum_user_stars = session.query(CalendarEntry.num_user_stars)\\\n\t\t\t.filter(\n\t\t\t\tCalendarEntry.user_id == user_id,\n\t\t\t\tCalendarEntry.match_id == match_id)\\\n\t\t\t.one()\\\n\t\t\t.num_user_stars\n\tif num_user_stars > 1:\n\t\t# Decrement the count of stars for the CalendarEntry to a positive integer.\n\t\tsession.execute(CalendarEntries.update()\n\t\t\t\t.where(sa.and_(\n\t\t\t\t\tCalendarEntry.user_id == user_id,\n\t\t\t\t\tCalendarEntry.match_id == match_id))\n\t\t\t\t.values({CalendarEntry.num_user_stars: num_user_stars - 1}))\n\telse:\n\t\t# Delete the CalendarEntry becuase the count of stars is now zero.\n\t\tsession.execute(CalendarEntries.delete().where(sa.and_(\n\t\t\t\tCalendarEntry.user_id == user_id,\n\t\t\t\tCalendarEntry.match_id == match_id)))","def users_like_changed(sender, instance, **kwargs):\n instance.total_likes = instance.users_like.count()\n instance.save()","def update_likes(self):\n self.nb_likes = self.likes.count()\n self.save()","def like(self, n: int) -> None:\n\n # YOUR CODE HERE\n self.likes += 1","def test_unauthenticated_user_disliking(self):\n self.like_dislike(self.like_url(6))","def unlike():\n if CURR_USER_KEY in session:\n cafe_id = int(request.json[\"cafe_id\"])\n user_id = g.user.id\n Like.query.filter_by(cafe_id=cafe_id, user_id=user_id).delete()\n db.session.commit()\n\n return jsonify({\"unliked\": cafe_id})\n return jsonify({\"error\": \"Not logged in\"})","def test_unauthenticated_user_liking(self):\n self.like_dislike(self.dislike_url(5))","def unfollow_profile(self):\n self.find_clickable_element(self.ISFOLLOWED_BTN).click()","def unlike(self, request, pk=None):\n\n user_wall_post = self.get_object()\n user_wall_post.likes.remove(self.request.user)\n return Response(status=201)","def unlike_comment(username):\n user_ID = before_request()\n user_ID = None\n if user_ID != None:\n user_ID = str(g.user['_id'])\n\n message_id = request.args.get('message_id')\n profile_usr = request.args.get('profile_usr')\n count=float(-1)\n\n print \"username who liked message\", username\n print \"Id of the message\", message_id\n\n redis_obj.zincrby('add_like', message_id, count)\n\n print redis_obj.scard(message_id)\n\n if redis_obj.get(user_ID):\n return redirect(url_for('timeline', username=profile_usr, userId=pickle.loads(redis_obj.get(user_ID))))\n else:\n \"\"\" Invalidating cache \"\"\"\n redis_obj.delete('leaderboard-key')\n redis_obj.delete(session['user_id'])\n print \"Invalidating cache after unlike\"\n return redirect(url_for('timeline', username=profile_usr))","def like(self):\n self.like_count = self.like_count + 1 if self.like_count else 1","def score_up(self, increment_by):\n self.user_score += increment_by","def toggle_vote(self):\n\n self.vote = 1 - self.vote","def toggle_like(post_id):\n if CURRENT_USER_KEY not in session:\n raise Unauthorized()\n\n liked_post = Post.query.get_or_404(post_id)\n user = User.query.get(session[CURRENT_USER_KEY])\n # toggle the like by removing from user likes\n if liked_post in user.likes:\n user.likes.remove(liked_post)\n else:\n user.likes.append(liked_post)\n\n db.session.commit()\n\n return redirect(url_for('show_likes', user_id=user.id))","def remove_like(obj, user):\n\tobj_type = ContentType.objects.get_for_model(obj)\n\tLike.objects.filter(\n\t\tcontent_type = obj_type,\n\t\tobject_id = obj.id,\n\t\tuser = user).delete()","def unlike(self, request, pk=None):\n\n obj = self.get_object()\n like_func.remove_like(obj, request.user)\n return Response(status=status.HTTP_202_ACCEPTED)","def remove_vote_weight(var, target : users.User, amount : int = 1) -> None:\n add_vote_weight(var, target, -amount)","def up_vote(cls, user, message):\r\n pass","def add_unsaved_likes_to_user(sender, user, request, **kwargs):\n session_likes = request.session.get('likes')\n if session_likes:\n user.userprofile.liked_products.add(*session_likes)","def unfollow(self, user_index, following_index):\n if user_index >= self.num_users or following_index >= self.num_users:\n raise ValueError(\n f\"Number of users is {self.num_users}, but indices \"\n f\"{user_index} and {following_index} were requested.\"\n )\n if self.users_hat[following_index, user_index] == 1:\n self.users_hat[following_index, user_index] = 0\n elif self.is_verbose():\n self.log(f\"User {following_index} was not following user {user_index}\")","def updateUserRating(definition, increase):\n user = mongo.db.users.find_one({\"_id\": definition[\"submitted_by\"]})\n mongo.db.users.update_one(\n {\"_id\": user[\"_id\"]},\n {\"$inc\": {\"total_rating\": increase}})","def get_number_of_likes_for_user(self, user):\n\t\tfrom pins.models import Pin\n\t\tpin_ctype = ContentType.objects.get_for_model(Pin)\n\t\tpin_list = Pin.objects.active_pins().filter(board__user=user).values_list('pk', flat=True)\n\t\treturn self.filter(content_type=pin_ctype, object_id__in=pin_list).count()","def unlike(db, filename, user):\n cur = db.cursor()\n if like_exists(db, filename, user):\n sql = \"\"\"\n delete from likes where filename=? and usernick=?;\n \"\"\"\n cur.execute(sql, (filename, user))\n db.commit()","def up_vote(cls, user, message):\n pass","def unlike_comment_API(id):\n\n count = float(-1)\n redis_obj.zincrby('add_like', id, count)\n\n \"\"\" Invalidating cache \"\"\"\n redis_obj.delete('leaderboard-key')\n return jsonify(username=g.user[1], Message=\"Success! Message unliked successfully. \", Status_code=status.HTTP_200_OK)","def dislike(self, event, number):\n\n if not hasattr(self, 'user_id') or not hasattr(self, 'top_id'):\n vk_common.dprint(2, vk_common.func_name(), f\"Не заданы атрибуты user_id = или top_id.\")\n return\n\n item_id = self.item_ids[self.top_id - 1]\n if item_id is None:\n vk_common.dprint(2, vk_common.func_name(), f\"Атрибут item_id = None.\")\n return\n else:\n vk_common.dprint(2, vk_common.func_name(), f\"number = {number}; item_id = {item_id}\")\n\n vk_common.dprint(2, vk_common.func_name(), f\"number = {number}; item_id = {item_id}\")\n vk_common.dprint(1, f\"имя файла с фото = \\\"{self.user_id}_{self.top_id}\\\"; \"\n f\"ссылка = https://vk.com/id{self.user_id}\")\n\n err, msg, likes = vk_common.dislike_photo(self.user_id, item_id)\n if err != 0:\n mb.showinfo(\"Информация\", f\"Не смогли убрать лайк у фото пользователя с id = {self.user_id}.\")\n vk_common.dprint(2, \"Не смогли убрать лайк.\")\n return\n mb.showinfo(\"Информация\", f\"Убрали лайк у фото пользователя с id = {self.user_id}.\")\n vk_common.dprint(2, f\"Убрали лайк: owner_id = {self.user_id}; item_id = {item_id}\")","def meal_disliked(request, meal_pk):\n meal_pk = int(meal_pk)\n meal = Meal.objects.get(pk=meal_pk)\n like = False\n member = request.user\n\n try:\n rating = Rating.objects.get(member=member, meal=meal)\n r_percent = meal.percent()\n except ObjectDoesNotExist:\n Rating.objects.create_rating(member, meal, like)\n r_percent = meal.percent()\n\n rating.like = like\n rating.save()\n r_percent = meal.percent()\n return HttpResponse(r_percent)","def unfollow_user(username):\n user_ID = before_request()\n user_ID = None\n if user_ID != None:\n user_ID = str(g.user['_id'])\n if not g.user:\n abort(401)\n whom_id = get_user_id(username)\n if whom_id is None:\n abort(404)\n mongo.db.users.update({'_id': g.user['_id']}, {\n '$pull': {'follows': whom_id}})\n flash('You are no longer following \"%s\"' % username)\n if redis_obj.get(user_ID):\n return redirect(url_for('user_timeline', username=username, userId=pickle.loads(redis_obj.get(user_ID))))\n else:\n redis_obj.delete(session['user_id'])\n print \"Invalidating cache after Unfollow\"\n return redirect(url_for('user_timeline', username=username))","def latest_likes(self, user, number_posts, likes):\n WAIT = 1\n if likes:\n action = 'Like'\n else:\n action = 'Unlike'\n self.nav_user(user)\n image_container = []\n image_container.extend(self.driver.find_elements_by_class_name('_9AhH0'))\n for image in image_container[:number_posts]:\n image.click()\n time.sleep(WAIT)\n try:\n self.driver.find_element_by_xpath(\"//*[@aria-label='{}']\".format(action).click())\n except Exception as e:\n print(e)\n self.driver.find_elements_by_xpath('/html/body/div[4]/div[2]/div/article/div[3]/section[1]/span[1]/button')[0].click() # clicks the heart symbol\n time.sleep(WAIT)\n self.driver.find_elements_by_xpath('/html/body/div[4]/div[3]/button')[0].click() #Makes sure to close out of current picture\n time.sleep(WAIT)\n \n # Tested\n users_list = []\n def get_likes_list(self, username):\n \"\"\"\n Method gets a list of users who like a post\n\n \"\"\"\n api = self.api\n api.searchUsername(username) \n result = api.LastJson\n username_id = result['user']['pk'] #Gets the user ID\n user_posts = api.getUserFeed(username_id) # gets the user feed\n result = api.LastJson\n media_id = result['items'][0]['id'] #gets the most recent post\n api.getMediaLikers(media_id) #gets users who liked\n users = api.LastJson('users')\n for user in users: #appends the users to the list\n users_list.append({'pk':user['pk'], 'username':user['username']})","def test_unlike_activity_get_other_likes(self):\n from .mockers import user_status_context\n from .mockers import subscribe_context, create_context\n username = 'messi'\n username_not_me = 'xavi'\n self.create_user(username)\n self.create_user(username_not_me)\n self.create_context(create_context)\n self.admin_subscribe_user_to_context(username, subscribe_context)\n self.admin_subscribe_user_to_context(username_not_me, subscribe_context)\n res = self.create_activity(username, user_status_context)\n activity_id = res.json['id']\n res = self.testapp.post('/activities/%s/likes' % activity_id, '', oauth2Header(username_not_me), status=201)\n res = self.testapp.post('/activities/%s/likes' % activity_id, '', oauth2Header(username), status=201)\n res = self.testapp.delete('/activities/%s/likes/%s' % (activity_id, username_not_me), '', oauth2Header(username_not_me), status=204)\n\n activity = self.testapp.get('/activities/%s' % activity_id, '', oauth2Header(username), status=200)\n\n self.assertEqual(activity.json['likes'][0]['username'], username)\n self.assertEqual(activity.json['liked'], True)\n self.assertEqual(activity.json['likesCount'], 1)","def unlike_event(request, pk):\n template_var = base_template_vals(request)\n previousLike = Like.objects.filter(event=Event.objects.get(id=pk), \n user=UserProfile.objects.filter(\n django_user=request.user)[0])\n if(len(previousLike) != 0):\n like = previousLike[0]\n like.delete()\n return redirect('index')","def user_unfollow():\n data = request.get_json(force=True)\n follower = User.query.get(data['follower'])\n following = User.query.get(data['following'])\n follower.followcheck.remove(following)\n db.session.commit()\n return {'unfollowed': True}","def remove_like(request):\n if request.method == \"POST\":\n if \"token\" in request.data and request.data[\"token\"] != \"\" and request.data[\"token\"] is not None:\n if Token.objects.filter(key=request.data[\"token\"]).exists():\n token = get_object_or_404(Token, key=request.data[\"token\"])\n if Post.objects.filter(pk=request.data[\"post_id\"]).exists():\n post = Post.objects.get(pk=request.data[\"post_id\"])\n if Like.objects.filter(post=post, user_id=token.user_id).exists():\n post.count_likes -= 1\n post.save()\n Like.objects.filter(post=post, user_id=token.user_id).delete()\n serializer = PostSerializer(post, context={'user_id': token.user_id})\n return Response({\"success\": 33,\n \"post\": serializer.data})\n else:\n return Response({\"error\": 34})\n\n else:\n return Response({\"error\": 32})\n else:\n return Response({\"error\": 17})","def like(self, data_base, user):\n cursor = data_base.cursor()\n cursor.execute(f\"UPDATE post SET likes = likes + 1 WHERE id = '{self.id}'\") # Increments the likes\n cursor.execute(f\"INSERT INTO user_like (user_id, post_id) VALUES ({user.id}, {self.id})\")\n if self.commit_to_db:\n data_base.commit()\n cursor.close()","def _force_vote(self, user, value):\n previous = 0\n if value == 0:\n # Delete any previous vote object\n for v in Vote.objects.filter(user=user, content=self):\n previous = v.value\n v.delete()\n else:\n # Create or change vote object\n v, created = Vote.objects.get_or_create(user=user, content=self)\n previous = v.value\n v.value = value\n v.save(update_fields=['value'])\n return (previous-value)*(-1)","def test_unlike_report(self):\n\n self.report.likes = {0, self.user2.id}\n db.session.commit()\n report = Report.query.get(self.report.id)\n # make sure like is there in report.likes.\n self.assertEqual(report.likes, {0, self.user2.id})\n\n with self.client as c:\n with c.session_transaction() as sess:\n sess['user_id'] = self.user2.id\n\n resp = c.post(f'/v1/report/like{self.report.id}')\n\n self.assertEqual(resp.json['success'], 'removed')\n report = Report.query.get(self.report.id)\n self.assertEqual(report.likes, {0})","def unfollow_friend(username):\n\n if not g.user:\n print \"401\"\n abort(401)\n whom_id = get_user_id(username)\n print whom_id\n if whom_id is None:\n abort(404)\n unfollow_query(whom_id)\n flash('You are no longer following \"%s\"' % username)\n name = {'name of unfollowing user': username}\n ############### REDIS cache invalidate #####################\n R_SERVER.delete(user_timeline_key)\n return jsonify(Username=name, Status_code=status.HTTP_200_OK)","def post(self, request, *args, **kwargs):\n\n user_wall_post_comment = self.get_object()\n user_wall_post_comment.likes.remove(self.request.user)\n return Response(status=201)","def like(request, post_id):\n if request.method == \"PUT\":\n liked = None\n user = request.user\n post = Post.objects.get(id=post_id)\n # If user already liked, decrement the like count and remove as 'liker'\n if user in post.liked_by.all():\n post.liked_by.remove(user)\n post.likes -= 1\n post.save()\n liked = False\n # Else increase like count and add user\n else:\n post.liked_by.add(user)\n post.likes += 1\n post.save()\n liked = True\n # Return data for updating dynamically with javascript\n return JsonResponse({\"total_likes\": post.likes, \"liked\": liked})","def unlike(id):\n spinner = Halo(text=\"Unliking photo...\", spinner=\"dots\").start()\n try:\n api.unlike(id)\n spinner.succeed(\"Unliked\")\n except Exception:\n spinner.fail(\"Failed to unlike photo.\")","def remove_user(self, u: \"Node\") -> None:\n\n if u in self.users_:\n self.users_[u] -= 1\n if self.users_[u] == 0:\n del self.users_[u]","def unfollow(request, usertostopfollow):\n stop_follow = Member.objects.get(user__username=usertostopfollow)\n user = Member.objects.get(user=request.user)\n user.following.remove(stop_follow)\n user.save()\n return redirect(request.META['HTTP_REFERER'])","async def handle_rank_ups(self, user: discord.User, brawler: str):\n\n brawler_data = await self.get_player_stat(\n user, 'brawlers', is_iter=True, substat=brawler)\n\n pb = brawler_data['pb']\n rank = brawler_data['rank']\n\n rank_as_per_pb = self.get_rank(pb)\n\n if rank_as_per_pb <= rank:\n return False\n\n await self.update_player_stat(\n user, 'brawlers', rank_as_per_pb, brawler, 'rank')\n\n rank_up_tokens = self.RANKS[str(rank)][\"PrimaryLvlUpRewardCount\"]\n\n token_doubler = await self.get_player_stat(user, 'token_doubler')\n\n upd_td = token_doubler - rank_up_tokens\n if upd_td < 0:\n upd_td = 0\n\n if token_doubler > rank_up_tokens:\n rank_up_tokens *= 2\n else:\n rank_up_tokens += token_doubler\n\n rank_up_starpoints = self.RANKS[str(rank)][\"SecondaryLvlUpRewardCount\"]\n\n await self.update_player_stat(\n user, 'tokens', rank_up_tokens, add_self=True)\n await self.update_player_stat(\n user, 'starpoints', rank_up_starpoints, add_self=True)\n await self.update_player_stat(\n user, 'token_doubler', upd_td)\n\n embed = discord.Embed(\n color=EMBED_COLOR,\n title=f\"Brawler Rank Up! {rank} → {rank_as_per_pb}\"\n )\n embed.set_author(name=user.name, icon_url=user.avatar_url)\n embed.add_field(\n name=\"Brawler\", value=f\"{brawler_emojis[brawler]} {brawler}\")\n embed.add_field(\n name=\"Tokens\", value=f\"{emojis['token']} {rank_up_tokens}\")\n if rank_up_starpoints:\n embed.add_field(\n name=\"Star Points\",\n value=f\"{emojis['starpoints']} {rank_up_starpoints}\"\n )\n if token_doubler > 0:\n embed.add_field(\n name=\"Token Doubler\",\n value=f\"{emojis['tokendoubler']} x{upd_td} remaining!\",\n inline=False\n )\n return embed","def clear_user_module_score(self, user):\r\n self.set_user_module_score(user, None, None)","def dislike(request, pk):\n update_posts_expiration()\n post = 0;\n post = Post.objects.get(id=pk)\n if post.is_expired == True:\n return Response(\"You can't interact with a expired post\")\n else:\n if request.method == \"POST\":\n if request.user != post.user:\n post.dislikes_count += 1\n post.save()\n serializer = DislikeSerializer(data=request.data)\n if serializer.is_valid():\n serializer.save(user=request.user, disliked_post=post)\n return Response(\"you disliked the post with title: \" + post.title)\n else:\n return Response(serializer.errors)\n else:\n return Response(\"you can't dislike your own post\")\n \n\n return Response(\"you didn't dislike the post yet\")","def get_dislikes(self, obj):\n return QuestionPersonLike.objects.filter(question=obj,\n like=False).count()","def reset_new_bookmarks_count(user_id):\n cacheAPI._reset_new_bookmarks_count(user_id)","async def changescore(self, ctx, num: int, *, user: discord.Member):\n self.data[ctx.guild.id]['score'][user.id] += num\n return await ctx.send(f\"{user}'s score has been changed to {self.data[ctx.guild.id]['score'][user.id]}.\")","def level_up(self):\n self.user_access_level += 1","def count_karma(user):\r\n\r\n karma = 0\r\n posts = Post.objects.filter(author=user).all()\r\n for post in posts:\r\n karma += (int(post.likes.count()) - int(post.dislikes.count()))\r\n \r\n return karma","def unfollow_user(username):\n if not g.user:\n abort(401)\n whom_id = get_user_id(username)\n if whom_id is None:\n abort(404)\n db = get_db()\n db.execute('delete from follower where who_id=? and whom_id=?',\n [session['user_id'], whom_id])\n db.commit()\n flash('You are no longer following \"%s\"' % username)\n return redirect(url_for('user_timeline', username=username))","def save_model(self, request, obj, form, change):\n obj.post.likeNumIncrease()\n obj.save()","def delete_user_profile(IamUserArn=None):\n pass","def new_unfollow(self, user_id, user_name):\n url_unfollow = self.url_unfollow % (user_id)\n try:\n unfollow = self._send_post_request(url_unfollow)\n if unfollow.status_code == 200:\n self.unfollow_counter += 1\n log_string = \"Unfollow: %s #%i.\" % (user_name,\n self.unfollow_counter)\n self.log.debug(log_string)\n return unfollow\n except:\n self.log.debug(\"Exept on unfollow!\")\n return False","def delete(self, request, username):\n\n # Retrieve the user from the user table if the user exists\n try:\n user_to_unfollow = User.objects.get(username=username)\n current_user = request.user\n # If a user is trying to unfollow themselves then stop the request\n if user_to_unfollow.profile.id == current_user.profile.id:\n return Response(\n {\"errors\": FOLLOW_USER_MSGS['CANNOT_UNFOLLOW_SELF']},\n status=status.HTTP_400_BAD_REQUEST\n )\n # Check if the user to be unfollowed\n # is in the current users following list\n try:\n profile_id = user_to_unfollow.profile.id\n user_being_followed = CustomFollows.objects.get(\n to_profile_id=profile_id,\n from_profile_id=current_user.profile.id\n )\n # If not tell the user the request can't happen\n # Because they don't follow the user\n except Exception as e:\n return Response(\n {\n \"errors\": FOLLOW_USER_MSGS['USER_UNFOLLOWED_ALREADY']\n },\n status=status.HTTP_400_BAD_REQUEST\n )\n # Otherwise unfollow the user as requested\n current_user.profile.follows.remove(user_to_unfollow.profile)\n # Get the following & followers username list\n # And the following & followers count for the current user\n user_following_data = get_user_following_data(current_user)\n return Response(\n {\n \"message\": FOLLOW_USER_MSGS['USER_UNFOLLOW_SUCCESSFUL'],\n \"following\": user_following_data[\"following\"],\n \"followers\": user_following_data[\"followers\"],\n \"followingCount\": user_following_data[\"followingCount\"],\n \"followersCount\": user_following_data[\"followersCount\"]\n },\n status=status.HTTP_200_OK\n )\n # End request if we cannot find the user we want to unfollow.\n except User.DoesNotExist:\n return Response(\n {\"errors\": FOLLOW_USER_MSGS['USER_NOT_FOUND']},\n status=status.HTTP_404_NOT_FOUND\n )","def unfollow_user(request, course_id, followed_user_id):\r\n user = cc.User.from_django_user(request.user)\r\n followed_user = cc.User.find(followed_user_id)\r\n user.unfollow(followed_user)\r\n return JsonResponse({})","def user_response_to_post(self, request, pk):\n post_objects_count = Post.objects.filter(id=pk, liked_users__id=request.user.id).count()\n post_objects = Post.objects.get(id=pk)\n if post_objects_count !=0:\n post_objects.liked_users.remove(request.user)\n response_msg = \"You disliked the post\"\n else:\n post_objects.liked_users.add(request.user)\n response_msg = \"You have liked the post\"\n return Response({'data': response_msg}, status=status.HTTP_200_OK)","def toggle_vote(self, user, value):\n try:\n v = Vote.objects.get(user=user, content=self)\n except Vote.DoesNotExist:\n Vote.objects.create(user=user, content=self, value=value)\n else:\n if v.value == value:\n v.delete()\n else:\n v.value = value\n v.save(update_fields=['value'])\n\n self.up = self.votes.count_upvotes()\n self.down = self.votes.count_downvotes()\n self.set_points()\n self.set_timepoints()\n self.save(update_fields=['up', 'down', 'points', 'timepoints'])","def test_unlike_mission(self):\n\n self.mission.likes = {0, self.user2.id}\n db.session.commit()\n mission = Mission.query.get(self.mission.id)\n # make sure like is there in mission.likes.\n self.assertEqual(mission.likes, {0, self.user2.id})\n\n with self.client as c:\n with c.session_transaction() as sess:\n sess['user_id'] = self.user2.id\n\n resp = c.post(f'/v1/mission/like{self.mission.id}')\n\n self.assertEqual(resp.json['success'], 'removed')\n mission = Mission.query.get(self.mission.id)\n self.assertEqual(mission.likes, {0})","def downvote(self):\n url = \"https://api.imgur.com/3/gallery/{0}/vote/down\".format(self.id)\n return self._imgur._send_request(url, needs_auth=True, method='POST')","def decrease(self):\n self.score -= self.score","def rate_down(self,req):\n try:\n self.rating=self.downratings[self.rating+4]\n self.flush()\n self.log_track(self.uid,'rated',self.rating)\n except:\n pass\n return self.rating_return(req)","def dislikes(self):\n return self.get_queryset().filter(vote__lt=0)","async def love(ctx, user: discord.Member):\r\n author = ctx.message.author\r\n if user.id == ctx.bot.user.id:\r\n await ctx.send(\"I am not capable of loving like you can. I'm sorry.\" )\r\n else:\r\n await ctx.send(author.mention + \" is capable of loving \" + user.mention + \" a whopping \" +\r\n str(randint(0, 100)) + \"%!\")\r\n ctx.counter(n)","def _decrease_money(self, amount):\n if 1 in self.money:\n self.money[1] -= amount","def users_likes(user_id):\n\n if not g.user:\n flash(\"Access unauthorized.\", \"danger\")\n return redirect(\"/\")\n\n users_blocking = [block.user_blocking_id for block in Blocks.query.all() if block.user_being_blocked_id == g.user.id]\n # likes = Message.query.filter(Message.user_id.notin_(users_blocking)).all()\n user = User.query.get_or_404(user_id)\n likes = [message for message in user.likes if message.user_id not in users_blocking]\n return render_template('users/likes.html', user=user, likes=likes)","def show_likes(user_id):\n\n\n user = User.query.get_or_404(user_id)\n\n return render_template('users/likes.html', user=user)","def test_removal_from_public_index(self):\n before = (S(UserProfile)\n .indexes(UserProfile.get_index(public_index=True))\n .count())\n profile = self.mozillian2.userprofile\n profile.privacy_full_name = MOZILLIANS\n profile.save()\n sleep(1)\n after = (S(UserProfile)\n .indexes(UserProfile.get_index(public_index=True))\n .count())\n eq_(after+1, before)","def test_unlike(self):\n\n message = Message(text=\"hello world\", user_id=self.testuser2.id)\n \n db.session.add(message)\n db.session.commit()\n\n like = Likes(message_id=message.id, user_id=self.testuser.id)\n \n db.session.add(like)\n db.session.commit()\n \n message_id = message.id\n\n with self.client as c:\n with c.session_transaction() as session:\n session[CURR_USER_KEY] = self.testuser.id\n \n response = c.post(f\"/messages/{message_id}/like\", follow_redirects=True)\n \n self.assertEqual(response.status_code, 200)\n\n likes = Likes.query.filter(Likes.message_id==message_id).all()\n self.assertEqual(len(likes), 0)","def update_collection_num(user_id, another_user_id, is_add):\n\n user = db_session.query(User).filter_by(user_id=user_id).scalar()\n another_user = db_session.query(User).filter_by(\n user_id=another_user_id).scalar()\n if is_add:\n user.follow_num += 1\n another_user.be_followed_num += 1\n else:\n user.follow_num -= 1\n another_user.be_followed_num -= 1\n db_session.commit()","def toggle_favorite(self, user, article, is_favoriting):\n if user not in article.favorited_by.all() and is_favoriting:\n article.favorited_by.add(user)\n if user in article.favorited_by.all() and not is_favoriting:\n article.favorited_by.remove(user)\n article.favoritesCount = article.favorited_by.all().count()\n article.save()","def score_minus_one():\r\n # TODO: Avoid duplicated code with score_plus_one; have\r\n # both call a single add_to_score function.\r\n postid = request.values.get(\"postid\")\r\n con = get_db()\r\n con.execute(\"\"\"\r\n UPDATE posts SET score=score-1 WHERE postid=?;\r\n \"\"\",\r\n (postid,))\r\n con.commit()\r\n con.close()\r\n return redirect(url_for(\"display_top\"))","def get_likes_count(self, instance):\n return instance.likes.count()","def get_likes_count(self, instance):\n return instance.likes.count()","def get_likes_count(self, instance):\n return instance.likes.count()","def incr_downvote(self, delta, kind):\r\n try:\r\n g.cache.incr(self.downvote_cache_key(kind), delta)\r\n except ValueError, e:\r\n print 'Account.incr_downvote failed with: %s' % e","def total_likes(self):\n\n return (self.likes.count()-self.dislikes.count())","def user_disappears(self, user):\n pass","def relation_downvote(request, pk):\n try:\n relation = Relation.objects.get(pk=pk)\n except Relation.DoesNotExist:\n return Response(status=status.HTTP_404_NOT_FOUND)\n\n if request.method == 'PUT':\n relation.negative_reaction_count += 1\n relation.save()\n serializer = RelationSerializer(relation)\n return Response(serializer.data)","async def _user_update_threshold(self, user_config: dict):\n return 30.0 * user_config['backoff_factor']","def decrement(self, stats, sample_rate=1):\n self.update_stats(stats, -1, sample_rate=sample_rate)","def rate_up(self,req):\n try:\n self.rating=self.upratings[self.rating+4]\n self.flush()\n self.log_track(self.uid,'rated',self.rating)\n except:\n pass\n return self.rating_return(req)","def like_comment(username):\n\n user_ID = before_request()\n user_ID = None\n if user_ID != None:\n user_ID = str(g.user['_id'])\n\n message_id = request.args.get('message_id')\n profile_usr = request.args.get('profile_usr')\n count = float(1)\n\n redis_obj.zincrby('add_like', message_id, count)\n\n print \"message ID\", message_id\n\n if redis_obj.get(user_ID):\n return redirect(url_for('timeline', username=profile_usr, userId=pickle.loads(redis_obj.get(user_ID))))\n else:\n \"\"\" Invalidating cache \"\"\"\n redis_obj.delete('leaderboard-key')\n redis_obj.delete(session['user_id'])\n print \"Invalidating cache after Like\"\n return redirect(url_for('timeline', username=profile_usr))","def user_reset(user_id):\n\n con = connect()\n cursor = con.cursor()\n cursor.execute(f\"\"\"\n START TRANSACTION;\n \n DELETE FROM ratings\n WHERE user_id = {user_id};\n \n DELETE FROM counts_by_rating_type\n WHERE user_id = {user_id};\n \n {_count_reset_query(user_id, 1)}\n {_count_reset_query(user_id, 2)}\n {_count_reset_query(user_id, 3)}\n \n COMMIT;\n \"\"\")\n cursor.close()\n return jsonify({'message': 'ok'}), 200","def test_downvote_then_upvote_same_user_leaves_comment_score_one_greater(self):\n comment = Comment.objects.get(body=\"987XYZ\")\n # self.assertEqual(len(post_qs), 1)\n self.assertEqual(comment.score, DEFAULT_SCORE)\n comment = Comment.objects.get(body=\"987XYZ\")\n\n vote1 = Vote.create(comment=comment, value=-1, voter=self.user)\n comment = Comment.objects.get(body=\"987XYZ\")\n self.assertEqual(comment.score, DEFAULT_SCORE - 1)\n\n vote2 = Vote.create(comment=comment, value=1, voter=self.user)\n comment = Comment.objects.get(body=\"987XYZ\")\n self.assertEqual(comment.score, DEFAULT_SCORE + 1)","def deduct_skill_points(self, amount):\r\n self.__skill_points -= amount\r\n self.__skill_points_indicator.configure(\r\n text=\"Available skill points: \" + str(self.__skill_points))","def _decrease_counter(self, response):\n response_id = response.meta['__id']\n spot = self._request_registry[response_id]\n spot['counter'] = spot.get('counter', 0) - 1","def unfollow(self, followerId: int, followeeId: int) -> None:\n if followerId not in self.users.keys() or followeeId not in self.users.keys():\n return\n if followeeId not in self.users[followerId].followees.keys():\n return\n self.users[followerId].followees.pop(followeeId)\n\n\n\n # Your Twitter object will be instantiated and called as such:\n # obj = Twitter()\n # obj.postTweet(userId,tweetId)\n # param_2 = obj.getNewsFeed(userId)\n # obj.follow(followerId,followeeId)\n # obj.unfollow(followerId,followeeId)","def test_unlike_activity(self):\n from .mockers import user_status_context\n from .mockers import subscribe_context, create_context\n username = 'messi'\n username_not_me = 'xavi'\n self.create_user(username)\n self.create_user(username_not_me)\n self.create_context(create_context)\n self.admin_subscribe_user_to_context(username, subscribe_context)\n self.admin_subscribe_user_to_context(username_not_me, subscribe_context)\n res = self.create_activity(username, user_status_context)\n activity_id = res.json['id']\n self.testapp.post('/activities/%s/likes' % activity_id, '', oauth2Header(username_not_me), status=201)\n self.testapp.delete('/activities/%s/likes/%s' % (activity_id, username_not_me), '', oauth2Header(username_not_me), status=204)\n activity = self.testapp.get('/activities/%s' % activity_id, '', oauth2Header(username), status=200)\n\n self.assertEqual(activity.json['likes'], [])\n self.assertEqual(activity.json['liked'], False)\n self.assertEqual(activity.json['likesCount'], 0)","def test_unpopular(self):\n self.assertFalse(self.user3.is_popular())\n self.user3.receive_upvotes(randint(101, 10000))\n self.assertTrue(self.user3.is_popular())","def increase_score(self):\n\n old_score = self.get_score()\n new_score = old_score + 1\n sql = \"UPDATE Users SET score = ? WHERE username = ?\"\n self.conn.execute(sql, (new_score, self.username))\n self.conn.commit()","def backwards(self, orm):\n \n user_type = orm['auth.User']\n user = user_type.objects.get(username=self.suUsername)\n \n profile_type = orm['mooi.Profile']\n userProfile = profile_type.objects.get(user=user)\n \n userProfile.delete()\n user.delete()","def is_liked(value, user: User):\n return value.is_liked(user)","def bass_decrease():\n request_command(tv_command=TVCommand.bassDecrease)","def decrement_comment_count():\n\n with _conn.cursor() as cur:\n cur.execute(\"UPDATE comment_count SET count = count - 1;\")\n _conn.commit()","def upgrad_one_contribute_score(user_total_click_num):\n return 1/math.log(1 + user_total_click_num)","def unfollowUser(following):\n \n cur, user_id, con = initialise(3, True)\n cur.execute(\"DELETE FROM followers WHERE user = (SELECT username FROM users WHERE id = ?) AND following = ?\", (user_id, following))\n finish(con)"],"string":"[\n \"def on_deleted_like(sender, instance: dillo.models.mixins.Likes, **kwargs):\\n if not instance.content_object:\\n return\\n target_user = instance.content_object.user\\n profile_likes_count_decrease(target_user)\\n log.debug('Decreased like count for user %s' % target_user)\",\n \"def unlike(self, data_base, user):\\n cursor = data_base.cursor()\\n cursor.execute(f\\\"UPDATE post SET likes = likes - 1 WHERE id = '{self.id}'\\\") # Decrements the likes\\n cursor.execute(f\\\"DELETE FROM user_like WHERE user_id = {user.id} AND post_id = {self.id}\\\")\\n if self.commit_to_db:\\n data_base.commit()\\n cursor.close()\",\n \"def get_user_likes(self) -> int:\\n return -1\",\n \"def _decrement_num_user_stars(user_id, match_id, now):\\n\\tnum_user_stars = session.query(CalendarEntry.num_user_stars)\\\\\\n\\t\\t\\t.filter(\\n\\t\\t\\t\\tCalendarEntry.user_id == user_id,\\n\\t\\t\\t\\tCalendarEntry.match_id == match_id)\\\\\\n\\t\\t\\t.one()\\\\\\n\\t\\t\\t.num_user_stars\\n\\tif num_user_stars > 1:\\n\\t\\t# Decrement the count of stars for the CalendarEntry to a positive integer.\\n\\t\\tsession.execute(CalendarEntries.update()\\n\\t\\t\\t\\t.where(sa.and_(\\n\\t\\t\\t\\t\\tCalendarEntry.user_id == user_id,\\n\\t\\t\\t\\t\\tCalendarEntry.match_id == match_id))\\n\\t\\t\\t\\t.values({CalendarEntry.num_user_stars: num_user_stars - 1}))\\n\\telse:\\n\\t\\t# Delete the CalendarEntry becuase the count of stars is now zero.\\n\\t\\tsession.execute(CalendarEntries.delete().where(sa.and_(\\n\\t\\t\\t\\tCalendarEntry.user_id == user_id,\\n\\t\\t\\t\\tCalendarEntry.match_id == match_id)))\",\n \"def users_like_changed(sender, instance, **kwargs):\\n instance.total_likes = instance.users_like.count()\\n instance.save()\",\n \"def update_likes(self):\\n self.nb_likes = self.likes.count()\\n self.save()\",\n \"def like(self, n: int) -> None:\\n\\n # YOUR CODE HERE\\n self.likes += 1\",\n \"def test_unauthenticated_user_disliking(self):\\n self.like_dislike(self.like_url(6))\",\n \"def unlike():\\n if CURR_USER_KEY in session:\\n cafe_id = int(request.json[\\\"cafe_id\\\"])\\n user_id = g.user.id\\n Like.query.filter_by(cafe_id=cafe_id, user_id=user_id).delete()\\n db.session.commit()\\n\\n return jsonify({\\\"unliked\\\": cafe_id})\\n return jsonify({\\\"error\\\": \\\"Not logged in\\\"})\",\n \"def test_unauthenticated_user_liking(self):\\n self.like_dislike(self.dislike_url(5))\",\n \"def unfollow_profile(self):\\n self.find_clickable_element(self.ISFOLLOWED_BTN).click()\",\n \"def unlike(self, request, pk=None):\\n\\n user_wall_post = self.get_object()\\n user_wall_post.likes.remove(self.request.user)\\n return Response(status=201)\",\n \"def unlike_comment(username):\\n user_ID = before_request()\\n user_ID = None\\n if user_ID != None:\\n user_ID = str(g.user['_id'])\\n\\n message_id = request.args.get('message_id')\\n profile_usr = request.args.get('profile_usr')\\n count=float(-1)\\n\\n print \\\"username who liked message\\\", username\\n print \\\"Id of the message\\\", message_id\\n\\n redis_obj.zincrby('add_like', message_id, count)\\n\\n print redis_obj.scard(message_id)\\n\\n if redis_obj.get(user_ID):\\n return redirect(url_for('timeline', username=profile_usr, userId=pickle.loads(redis_obj.get(user_ID))))\\n else:\\n \\\"\\\"\\\" Invalidating cache \\\"\\\"\\\"\\n redis_obj.delete('leaderboard-key')\\n redis_obj.delete(session['user_id'])\\n print \\\"Invalidating cache after unlike\\\"\\n return redirect(url_for('timeline', username=profile_usr))\",\n \"def like(self):\\n self.like_count = self.like_count + 1 if self.like_count else 1\",\n \"def score_up(self, increment_by):\\n self.user_score += increment_by\",\n \"def toggle_vote(self):\\n\\n self.vote = 1 - self.vote\",\n \"def toggle_like(post_id):\\n if CURRENT_USER_KEY not in session:\\n raise Unauthorized()\\n\\n liked_post = Post.query.get_or_404(post_id)\\n user = User.query.get(session[CURRENT_USER_KEY])\\n # toggle the like by removing from user likes\\n if liked_post in user.likes:\\n user.likes.remove(liked_post)\\n else:\\n user.likes.append(liked_post)\\n\\n db.session.commit()\\n\\n return redirect(url_for('show_likes', user_id=user.id))\",\n \"def remove_like(obj, user):\\n\\tobj_type = ContentType.objects.get_for_model(obj)\\n\\tLike.objects.filter(\\n\\t\\tcontent_type = obj_type,\\n\\t\\tobject_id = obj.id,\\n\\t\\tuser = user).delete()\",\n \"def unlike(self, request, pk=None):\\n\\n obj = self.get_object()\\n like_func.remove_like(obj, request.user)\\n return Response(status=status.HTTP_202_ACCEPTED)\",\n \"def remove_vote_weight(var, target : users.User, amount : int = 1) -> None:\\n add_vote_weight(var, target, -amount)\",\n \"def up_vote(cls, user, message):\\r\\n pass\",\n \"def add_unsaved_likes_to_user(sender, user, request, **kwargs):\\n session_likes = request.session.get('likes')\\n if session_likes:\\n user.userprofile.liked_products.add(*session_likes)\",\n \"def unfollow(self, user_index, following_index):\\n if user_index >= self.num_users or following_index >= self.num_users:\\n raise ValueError(\\n f\\\"Number of users is {self.num_users}, but indices \\\"\\n f\\\"{user_index} and {following_index} were requested.\\\"\\n )\\n if self.users_hat[following_index, user_index] == 1:\\n self.users_hat[following_index, user_index] = 0\\n elif self.is_verbose():\\n self.log(f\\\"User {following_index} was not following user {user_index}\\\")\",\n \"def updateUserRating(definition, increase):\\n user = mongo.db.users.find_one({\\\"_id\\\": definition[\\\"submitted_by\\\"]})\\n mongo.db.users.update_one(\\n {\\\"_id\\\": user[\\\"_id\\\"]},\\n {\\\"$inc\\\": {\\\"total_rating\\\": increase}})\",\n \"def get_number_of_likes_for_user(self, user):\\n\\t\\tfrom pins.models import Pin\\n\\t\\tpin_ctype = ContentType.objects.get_for_model(Pin)\\n\\t\\tpin_list = Pin.objects.active_pins().filter(board__user=user).values_list('pk', flat=True)\\n\\t\\treturn self.filter(content_type=pin_ctype, object_id__in=pin_list).count()\",\n \"def unlike(db, filename, user):\\n cur = db.cursor()\\n if like_exists(db, filename, user):\\n sql = \\\"\\\"\\\"\\n delete from likes where filename=? and usernick=?;\\n \\\"\\\"\\\"\\n cur.execute(sql, (filename, user))\\n db.commit()\",\n \"def up_vote(cls, user, message):\\n pass\",\n \"def unlike_comment_API(id):\\n\\n count = float(-1)\\n redis_obj.zincrby('add_like', id, count)\\n\\n \\\"\\\"\\\" Invalidating cache \\\"\\\"\\\"\\n redis_obj.delete('leaderboard-key')\\n return jsonify(username=g.user[1], Message=\\\"Success! Message unliked successfully. \\\", Status_code=status.HTTP_200_OK)\",\n \"def dislike(self, event, number):\\n\\n if not hasattr(self, 'user_id') or not hasattr(self, 'top_id'):\\n vk_common.dprint(2, vk_common.func_name(), f\\\"Не заданы атрибуты user_id = или top_id.\\\")\\n return\\n\\n item_id = self.item_ids[self.top_id - 1]\\n if item_id is None:\\n vk_common.dprint(2, vk_common.func_name(), f\\\"Атрибут item_id = None.\\\")\\n return\\n else:\\n vk_common.dprint(2, vk_common.func_name(), f\\\"number = {number}; item_id = {item_id}\\\")\\n\\n vk_common.dprint(2, vk_common.func_name(), f\\\"number = {number}; item_id = {item_id}\\\")\\n vk_common.dprint(1, f\\\"имя файла с фото = \\\\\\\"{self.user_id}_{self.top_id}\\\\\\\"; \\\"\\n f\\\"ссылка = https://vk.com/id{self.user_id}\\\")\\n\\n err, msg, likes = vk_common.dislike_photo(self.user_id, item_id)\\n if err != 0:\\n mb.showinfo(\\\"Информация\\\", f\\\"Не смогли убрать лайк у фото пользователя с id = {self.user_id}.\\\")\\n vk_common.dprint(2, \\\"Не смогли убрать лайк.\\\")\\n return\\n mb.showinfo(\\\"Информация\\\", f\\\"Убрали лайк у фото пользователя с id = {self.user_id}.\\\")\\n vk_common.dprint(2, f\\\"Убрали лайк: owner_id = {self.user_id}; item_id = {item_id}\\\")\",\n \"def meal_disliked(request, meal_pk):\\n meal_pk = int(meal_pk)\\n meal = Meal.objects.get(pk=meal_pk)\\n like = False\\n member = request.user\\n\\n try:\\n rating = Rating.objects.get(member=member, meal=meal)\\n r_percent = meal.percent()\\n except ObjectDoesNotExist:\\n Rating.objects.create_rating(member, meal, like)\\n r_percent = meal.percent()\\n\\n rating.like = like\\n rating.save()\\n r_percent = meal.percent()\\n return HttpResponse(r_percent)\",\n \"def unfollow_user(username):\\n user_ID = before_request()\\n user_ID = None\\n if user_ID != None:\\n user_ID = str(g.user['_id'])\\n if not g.user:\\n abort(401)\\n whom_id = get_user_id(username)\\n if whom_id is None:\\n abort(404)\\n mongo.db.users.update({'_id': g.user['_id']}, {\\n '$pull': {'follows': whom_id}})\\n flash('You are no longer following \\\"%s\\\"' % username)\\n if redis_obj.get(user_ID):\\n return redirect(url_for('user_timeline', username=username, userId=pickle.loads(redis_obj.get(user_ID))))\\n else:\\n redis_obj.delete(session['user_id'])\\n print \\\"Invalidating cache after Unfollow\\\"\\n return redirect(url_for('user_timeline', username=username))\",\n \"def latest_likes(self, user, number_posts, likes):\\n WAIT = 1\\n if likes:\\n action = 'Like'\\n else:\\n action = 'Unlike'\\n self.nav_user(user)\\n image_container = []\\n image_container.extend(self.driver.find_elements_by_class_name('_9AhH0'))\\n for image in image_container[:number_posts]:\\n image.click()\\n time.sleep(WAIT)\\n try:\\n self.driver.find_element_by_xpath(\\\"//*[@aria-label='{}']\\\".format(action).click())\\n except Exception as e:\\n print(e)\\n self.driver.find_elements_by_xpath('/html/body/div[4]/div[2]/div/article/div[3]/section[1]/span[1]/button')[0].click() # clicks the heart symbol\\n time.sleep(WAIT)\\n self.driver.find_elements_by_xpath('/html/body/div[4]/div[3]/button')[0].click() #Makes sure to close out of current picture\\n time.sleep(WAIT)\\n \\n # Tested\\n users_list = []\\n def get_likes_list(self, username):\\n \\\"\\\"\\\"\\n Method gets a list of users who like a post\\n\\n \\\"\\\"\\\"\\n api = self.api\\n api.searchUsername(username) \\n result = api.LastJson\\n username_id = result['user']['pk'] #Gets the user ID\\n user_posts = api.getUserFeed(username_id) # gets the user feed\\n result = api.LastJson\\n media_id = result['items'][0]['id'] #gets the most recent post\\n api.getMediaLikers(media_id) #gets users who liked\\n users = api.LastJson('users')\\n for user in users: #appends the users to the list\\n users_list.append({'pk':user['pk'], 'username':user['username']})\",\n \"def test_unlike_activity_get_other_likes(self):\\n from .mockers import user_status_context\\n from .mockers import subscribe_context, create_context\\n username = 'messi'\\n username_not_me = 'xavi'\\n self.create_user(username)\\n self.create_user(username_not_me)\\n self.create_context(create_context)\\n self.admin_subscribe_user_to_context(username, subscribe_context)\\n self.admin_subscribe_user_to_context(username_not_me, subscribe_context)\\n res = self.create_activity(username, user_status_context)\\n activity_id = res.json['id']\\n res = self.testapp.post('/activities/%s/likes' % activity_id, '', oauth2Header(username_not_me), status=201)\\n res = self.testapp.post('/activities/%s/likes' % activity_id, '', oauth2Header(username), status=201)\\n res = self.testapp.delete('/activities/%s/likes/%s' % (activity_id, username_not_me), '', oauth2Header(username_not_me), status=204)\\n\\n activity = self.testapp.get('/activities/%s' % activity_id, '', oauth2Header(username), status=200)\\n\\n self.assertEqual(activity.json['likes'][0]['username'], username)\\n self.assertEqual(activity.json['liked'], True)\\n self.assertEqual(activity.json['likesCount'], 1)\",\n \"def unlike_event(request, pk):\\n template_var = base_template_vals(request)\\n previousLike = Like.objects.filter(event=Event.objects.get(id=pk), \\n user=UserProfile.objects.filter(\\n django_user=request.user)[0])\\n if(len(previousLike) != 0):\\n like = previousLike[0]\\n like.delete()\\n return redirect('index')\",\n \"def user_unfollow():\\n data = request.get_json(force=True)\\n follower = User.query.get(data['follower'])\\n following = User.query.get(data['following'])\\n follower.followcheck.remove(following)\\n db.session.commit()\\n return {'unfollowed': True}\",\n \"def remove_like(request):\\n if request.method == \\\"POST\\\":\\n if \\\"token\\\" in request.data and request.data[\\\"token\\\"] != \\\"\\\" and request.data[\\\"token\\\"] is not None:\\n if Token.objects.filter(key=request.data[\\\"token\\\"]).exists():\\n token = get_object_or_404(Token, key=request.data[\\\"token\\\"])\\n if Post.objects.filter(pk=request.data[\\\"post_id\\\"]).exists():\\n post = Post.objects.get(pk=request.data[\\\"post_id\\\"])\\n if Like.objects.filter(post=post, user_id=token.user_id).exists():\\n post.count_likes -= 1\\n post.save()\\n Like.objects.filter(post=post, user_id=token.user_id).delete()\\n serializer = PostSerializer(post, context={'user_id': token.user_id})\\n return Response({\\\"success\\\": 33,\\n \\\"post\\\": serializer.data})\\n else:\\n return Response({\\\"error\\\": 34})\\n\\n else:\\n return Response({\\\"error\\\": 32})\\n else:\\n return Response({\\\"error\\\": 17})\",\n \"def like(self, data_base, user):\\n cursor = data_base.cursor()\\n cursor.execute(f\\\"UPDATE post SET likes = likes + 1 WHERE id = '{self.id}'\\\") # Increments the likes\\n cursor.execute(f\\\"INSERT INTO user_like (user_id, post_id) VALUES ({user.id}, {self.id})\\\")\\n if self.commit_to_db:\\n data_base.commit()\\n cursor.close()\",\n \"def _force_vote(self, user, value):\\n previous = 0\\n if value == 0:\\n # Delete any previous vote object\\n for v in Vote.objects.filter(user=user, content=self):\\n previous = v.value\\n v.delete()\\n else:\\n # Create or change vote object\\n v, created = Vote.objects.get_or_create(user=user, content=self)\\n previous = v.value\\n v.value = value\\n v.save(update_fields=['value'])\\n return (previous-value)*(-1)\",\n \"def test_unlike_report(self):\\n\\n self.report.likes = {0, self.user2.id}\\n db.session.commit()\\n report = Report.query.get(self.report.id)\\n # make sure like is there in report.likes.\\n self.assertEqual(report.likes, {0, self.user2.id})\\n\\n with self.client as c:\\n with c.session_transaction() as sess:\\n sess['user_id'] = self.user2.id\\n\\n resp = c.post(f'/v1/report/like{self.report.id}')\\n\\n self.assertEqual(resp.json['success'], 'removed')\\n report = Report.query.get(self.report.id)\\n self.assertEqual(report.likes, {0})\",\n \"def unfollow_friend(username):\\n\\n if not g.user:\\n print \\\"401\\\"\\n abort(401)\\n whom_id = get_user_id(username)\\n print whom_id\\n if whom_id is None:\\n abort(404)\\n unfollow_query(whom_id)\\n flash('You are no longer following \\\"%s\\\"' % username)\\n name = {'name of unfollowing user': username}\\n ############### REDIS cache invalidate #####################\\n R_SERVER.delete(user_timeline_key)\\n return jsonify(Username=name, Status_code=status.HTTP_200_OK)\",\n \"def post(self, request, *args, **kwargs):\\n\\n user_wall_post_comment = self.get_object()\\n user_wall_post_comment.likes.remove(self.request.user)\\n return Response(status=201)\",\n \"def like(request, post_id):\\n if request.method == \\\"PUT\\\":\\n liked = None\\n user = request.user\\n post = Post.objects.get(id=post_id)\\n # If user already liked, decrement the like count and remove as 'liker'\\n if user in post.liked_by.all():\\n post.liked_by.remove(user)\\n post.likes -= 1\\n post.save()\\n liked = False\\n # Else increase like count and add user\\n else:\\n post.liked_by.add(user)\\n post.likes += 1\\n post.save()\\n liked = True\\n # Return data for updating dynamically with javascript\\n return JsonResponse({\\\"total_likes\\\": post.likes, \\\"liked\\\": liked})\",\n \"def unlike(id):\\n spinner = Halo(text=\\\"Unliking photo...\\\", spinner=\\\"dots\\\").start()\\n try:\\n api.unlike(id)\\n spinner.succeed(\\\"Unliked\\\")\\n except Exception:\\n spinner.fail(\\\"Failed to unlike photo.\\\")\",\n \"def remove_user(self, u: \\\"Node\\\") -> None:\\n\\n if u in self.users_:\\n self.users_[u] -= 1\\n if self.users_[u] == 0:\\n del self.users_[u]\",\n \"def unfollow(request, usertostopfollow):\\n stop_follow = Member.objects.get(user__username=usertostopfollow)\\n user = Member.objects.get(user=request.user)\\n user.following.remove(stop_follow)\\n user.save()\\n return redirect(request.META['HTTP_REFERER'])\",\n \"async def handle_rank_ups(self, user: discord.User, brawler: str):\\n\\n brawler_data = await self.get_player_stat(\\n user, 'brawlers', is_iter=True, substat=brawler)\\n\\n pb = brawler_data['pb']\\n rank = brawler_data['rank']\\n\\n rank_as_per_pb = self.get_rank(pb)\\n\\n if rank_as_per_pb <= rank:\\n return False\\n\\n await self.update_player_stat(\\n user, 'brawlers', rank_as_per_pb, brawler, 'rank')\\n\\n rank_up_tokens = self.RANKS[str(rank)][\\\"PrimaryLvlUpRewardCount\\\"]\\n\\n token_doubler = await self.get_player_stat(user, 'token_doubler')\\n\\n upd_td = token_doubler - rank_up_tokens\\n if upd_td < 0:\\n upd_td = 0\\n\\n if token_doubler > rank_up_tokens:\\n rank_up_tokens *= 2\\n else:\\n rank_up_tokens += token_doubler\\n\\n rank_up_starpoints = self.RANKS[str(rank)][\\\"SecondaryLvlUpRewardCount\\\"]\\n\\n await self.update_player_stat(\\n user, 'tokens', rank_up_tokens, add_self=True)\\n await self.update_player_stat(\\n user, 'starpoints', rank_up_starpoints, add_self=True)\\n await self.update_player_stat(\\n user, 'token_doubler', upd_td)\\n\\n embed = discord.Embed(\\n color=EMBED_COLOR,\\n title=f\\\"Brawler Rank Up! {rank} → {rank_as_per_pb}\\\"\\n )\\n embed.set_author(name=user.name, icon_url=user.avatar_url)\\n embed.add_field(\\n name=\\\"Brawler\\\", value=f\\\"{brawler_emojis[brawler]} {brawler}\\\")\\n embed.add_field(\\n name=\\\"Tokens\\\", value=f\\\"{emojis['token']} {rank_up_tokens}\\\")\\n if rank_up_starpoints:\\n embed.add_field(\\n name=\\\"Star Points\\\",\\n value=f\\\"{emojis['starpoints']} {rank_up_starpoints}\\\"\\n )\\n if token_doubler > 0:\\n embed.add_field(\\n name=\\\"Token Doubler\\\",\\n value=f\\\"{emojis['tokendoubler']} x{upd_td} remaining!\\\",\\n inline=False\\n )\\n return embed\",\n \"def clear_user_module_score(self, user):\\r\\n self.set_user_module_score(user, None, None)\",\n \"def dislike(request, pk):\\n update_posts_expiration()\\n post = 0;\\n post = Post.objects.get(id=pk)\\n if post.is_expired == True:\\n return Response(\\\"You can't interact with a expired post\\\")\\n else:\\n if request.method == \\\"POST\\\":\\n if request.user != post.user:\\n post.dislikes_count += 1\\n post.save()\\n serializer = DislikeSerializer(data=request.data)\\n if serializer.is_valid():\\n serializer.save(user=request.user, disliked_post=post)\\n return Response(\\\"you disliked the post with title: \\\" + post.title)\\n else:\\n return Response(serializer.errors)\\n else:\\n return Response(\\\"you can't dislike your own post\\\")\\n \\n\\n return Response(\\\"you didn't dislike the post yet\\\")\",\n \"def get_dislikes(self, obj):\\n return QuestionPersonLike.objects.filter(question=obj,\\n like=False).count()\",\n \"def reset_new_bookmarks_count(user_id):\\n cacheAPI._reset_new_bookmarks_count(user_id)\",\n \"async def changescore(self, ctx, num: int, *, user: discord.Member):\\n self.data[ctx.guild.id]['score'][user.id] += num\\n return await ctx.send(f\\\"{user}'s score has been changed to {self.data[ctx.guild.id]['score'][user.id]}.\\\")\",\n \"def level_up(self):\\n self.user_access_level += 1\",\n \"def count_karma(user):\\r\\n\\r\\n karma = 0\\r\\n posts = Post.objects.filter(author=user).all()\\r\\n for post in posts:\\r\\n karma += (int(post.likes.count()) - int(post.dislikes.count()))\\r\\n \\r\\n return karma\",\n \"def unfollow_user(username):\\n if not g.user:\\n abort(401)\\n whom_id = get_user_id(username)\\n if whom_id is None:\\n abort(404)\\n db = get_db()\\n db.execute('delete from follower where who_id=? and whom_id=?',\\n [session['user_id'], whom_id])\\n db.commit()\\n flash('You are no longer following \\\"%s\\\"' % username)\\n return redirect(url_for('user_timeline', username=username))\",\n \"def save_model(self, request, obj, form, change):\\n obj.post.likeNumIncrease()\\n obj.save()\",\n \"def delete_user_profile(IamUserArn=None):\\n pass\",\n \"def new_unfollow(self, user_id, user_name):\\n url_unfollow = self.url_unfollow % (user_id)\\n try:\\n unfollow = self._send_post_request(url_unfollow)\\n if unfollow.status_code == 200:\\n self.unfollow_counter += 1\\n log_string = \\\"Unfollow: %s #%i.\\\" % (user_name,\\n self.unfollow_counter)\\n self.log.debug(log_string)\\n return unfollow\\n except:\\n self.log.debug(\\\"Exept on unfollow!\\\")\\n return False\",\n \"def delete(self, request, username):\\n\\n # Retrieve the user from the user table if the user exists\\n try:\\n user_to_unfollow = User.objects.get(username=username)\\n current_user = request.user\\n # If a user is trying to unfollow themselves then stop the request\\n if user_to_unfollow.profile.id == current_user.profile.id:\\n return Response(\\n {\\\"errors\\\": FOLLOW_USER_MSGS['CANNOT_UNFOLLOW_SELF']},\\n status=status.HTTP_400_BAD_REQUEST\\n )\\n # Check if the user to be unfollowed\\n # is in the current users following list\\n try:\\n profile_id = user_to_unfollow.profile.id\\n user_being_followed = CustomFollows.objects.get(\\n to_profile_id=profile_id,\\n from_profile_id=current_user.profile.id\\n )\\n # If not tell the user the request can't happen\\n # Because they don't follow the user\\n except Exception as e:\\n return Response(\\n {\\n \\\"errors\\\": FOLLOW_USER_MSGS['USER_UNFOLLOWED_ALREADY']\\n },\\n status=status.HTTP_400_BAD_REQUEST\\n )\\n # Otherwise unfollow the user as requested\\n current_user.profile.follows.remove(user_to_unfollow.profile)\\n # Get the following & followers username list\\n # And the following & followers count for the current user\\n user_following_data = get_user_following_data(current_user)\\n return Response(\\n {\\n \\\"message\\\": FOLLOW_USER_MSGS['USER_UNFOLLOW_SUCCESSFUL'],\\n \\\"following\\\": user_following_data[\\\"following\\\"],\\n \\\"followers\\\": user_following_data[\\\"followers\\\"],\\n \\\"followingCount\\\": user_following_data[\\\"followingCount\\\"],\\n \\\"followersCount\\\": user_following_data[\\\"followersCount\\\"]\\n },\\n status=status.HTTP_200_OK\\n )\\n # End request if we cannot find the user we want to unfollow.\\n except User.DoesNotExist:\\n return Response(\\n {\\\"errors\\\": FOLLOW_USER_MSGS['USER_NOT_FOUND']},\\n status=status.HTTP_404_NOT_FOUND\\n )\",\n \"def unfollow_user(request, course_id, followed_user_id):\\r\\n user = cc.User.from_django_user(request.user)\\r\\n followed_user = cc.User.find(followed_user_id)\\r\\n user.unfollow(followed_user)\\r\\n return JsonResponse({})\",\n \"def user_response_to_post(self, request, pk):\\n post_objects_count = Post.objects.filter(id=pk, liked_users__id=request.user.id).count()\\n post_objects = Post.objects.get(id=pk)\\n if post_objects_count !=0:\\n post_objects.liked_users.remove(request.user)\\n response_msg = \\\"You disliked the post\\\"\\n else:\\n post_objects.liked_users.add(request.user)\\n response_msg = \\\"You have liked the post\\\"\\n return Response({'data': response_msg}, status=status.HTTP_200_OK)\",\n \"def toggle_vote(self, user, value):\\n try:\\n v = Vote.objects.get(user=user, content=self)\\n except Vote.DoesNotExist:\\n Vote.objects.create(user=user, content=self, value=value)\\n else:\\n if v.value == value:\\n v.delete()\\n else:\\n v.value = value\\n v.save(update_fields=['value'])\\n\\n self.up = self.votes.count_upvotes()\\n self.down = self.votes.count_downvotes()\\n self.set_points()\\n self.set_timepoints()\\n self.save(update_fields=['up', 'down', 'points', 'timepoints'])\",\n \"def test_unlike_mission(self):\\n\\n self.mission.likes = {0, self.user2.id}\\n db.session.commit()\\n mission = Mission.query.get(self.mission.id)\\n # make sure like is there in mission.likes.\\n self.assertEqual(mission.likes, {0, self.user2.id})\\n\\n with self.client as c:\\n with c.session_transaction() as sess:\\n sess['user_id'] = self.user2.id\\n\\n resp = c.post(f'/v1/mission/like{self.mission.id}')\\n\\n self.assertEqual(resp.json['success'], 'removed')\\n mission = Mission.query.get(self.mission.id)\\n self.assertEqual(mission.likes, {0})\",\n \"def downvote(self):\\n url = \\\"https://api.imgur.com/3/gallery/{0}/vote/down\\\".format(self.id)\\n return self._imgur._send_request(url, needs_auth=True, method='POST')\",\n \"def decrease(self):\\n self.score -= self.score\",\n \"def rate_down(self,req):\\n try:\\n self.rating=self.downratings[self.rating+4]\\n self.flush()\\n self.log_track(self.uid,'rated',self.rating)\\n except:\\n pass\\n return self.rating_return(req)\",\n \"def dislikes(self):\\n return self.get_queryset().filter(vote__lt=0)\",\n \"async def love(ctx, user: discord.Member):\\r\\n author = ctx.message.author\\r\\n if user.id == ctx.bot.user.id:\\r\\n await ctx.send(\\\"I am not capable of loving like you can. I'm sorry.\\\" )\\r\\n else:\\r\\n await ctx.send(author.mention + \\\" is capable of loving \\\" + user.mention + \\\" a whopping \\\" +\\r\\n str(randint(0, 100)) + \\\"%!\\\")\\r\\n ctx.counter(n)\",\n \"def _decrease_money(self, amount):\\n if 1 in self.money:\\n self.money[1] -= amount\",\n \"def users_likes(user_id):\\n\\n if not g.user:\\n flash(\\\"Access unauthorized.\\\", \\\"danger\\\")\\n return redirect(\\\"/\\\")\\n\\n users_blocking = [block.user_blocking_id for block in Blocks.query.all() if block.user_being_blocked_id == g.user.id]\\n # likes = Message.query.filter(Message.user_id.notin_(users_blocking)).all()\\n user = User.query.get_or_404(user_id)\\n likes = [message for message in user.likes if message.user_id not in users_blocking]\\n return render_template('users/likes.html', user=user, likes=likes)\",\n \"def show_likes(user_id):\\n\\n\\n user = User.query.get_or_404(user_id)\\n\\n return render_template('users/likes.html', user=user)\",\n \"def test_removal_from_public_index(self):\\n before = (S(UserProfile)\\n .indexes(UserProfile.get_index(public_index=True))\\n .count())\\n profile = self.mozillian2.userprofile\\n profile.privacy_full_name = MOZILLIANS\\n profile.save()\\n sleep(1)\\n after = (S(UserProfile)\\n .indexes(UserProfile.get_index(public_index=True))\\n .count())\\n eq_(after+1, before)\",\n \"def test_unlike(self):\\n\\n message = Message(text=\\\"hello world\\\", user_id=self.testuser2.id)\\n \\n db.session.add(message)\\n db.session.commit()\\n\\n like = Likes(message_id=message.id, user_id=self.testuser.id)\\n \\n db.session.add(like)\\n db.session.commit()\\n \\n message_id = message.id\\n\\n with self.client as c:\\n with c.session_transaction() as session:\\n session[CURR_USER_KEY] = self.testuser.id\\n \\n response = c.post(f\\\"/messages/{message_id}/like\\\", follow_redirects=True)\\n \\n self.assertEqual(response.status_code, 200)\\n\\n likes = Likes.query.filter(Likes.message_id==message_id).all()\\n self.assertEqual(len(likes), 0)\",\n \"def update_collection_num(user_id, another_user_id, is_add):\\n\\n user = db_session.query(User).filter_by(user_id=user_id).scalar()\\n another_user = db_session.query(User).filter_by(\\n user_id=another_user_id).scalar()\\n if is_add:\\n user.follow_num += 1\\n another_user.be_followed_num += 1\\n else:\\n user.follow_num -= 1\\n another_user.be_followed_num -= 1\\n db_session.commit()\",\n \"def toggle_favorite(self, user, article, is_favoriting):\\n if user not in article.favorited_by.all() and is_favoriting:\\n article.favorited_by.add(user)\\n if user in article.favorited_by.all() and not is_favoriting:\\n article.favorited_by.remove(user)\\n article.favoritesCount = article.favorited_by.all().count()\\n article.save()\",\n \"def score_minus_one():\\r\\n # TODO: Avoid duplicated code with score_plus_one; have\\r\\n # both call a single add_to_score function.\\r\\n postid = request.values.get(\\\"postid\\\")\\r\\n con = get_db()\\r\\n con.execute(\\\"\\\"\\\"\\r\\n UPDATE posts SET score=score-1 WHERE postid=?;\\r\\n \\\"\\\"\\\",\\r\\n (postid,))\\r\\n con.commit()\\r\\n con.close()\\r\\n return redirect(url_for(\\\"display_top\\\"))\",\n \"def get_likes_count(self, instance):\\n return instance.likes.count()\",\n \"def get_likes_count(self, instance):\\n return instance.likes.count()\",\n \"def get_likes_count(self, instance):\\n return instance.likes.count()\",\n \"def incr_downvote(self, delta, kind):\\r\\n try:\\r\\n g.cache.incr(self.downvote_cache_key(kind), delta)\\r\\n except ValueError, e:\\r\\n print 'Account.incr_downvote failed with: %s' % e\",\n \"def total_likes(self):\\n\\n return (self.likes.count()-self.dislikes.count())\",\n \"def user_disappears(self, user):\\n pass\",\n \"def relation_downvote(request, pk):\\n try:\\n relation = Relation.objects.get(pk=pk)\\n except Relation.DoesNotExist:\\n return Response(status=status.HTTP_404_NOT_FOUND)\\n\\n if request.method == 'PUT':\\n relation.negative_reaction_count += 1\\n relation.save()\\n serializer = RelationSerializer(relation)\\n return Response(serializer.data)\",\n \"async def _user_update_threshold(self, user_config: dict):\\n return 30.0 * user_config['backoff_factor']\",\n \"def decrement(self, stats, sample_rate=1):\\n self.update_stats(stats, -1, sample_rate=sample_rate)\",\n \"def rate_up(self,req):\\n try:\\n self.rating=self.upratings[self.rating+4]\\n self.flush()\\n self.log_track(self.uid,'rated',self.rating)\\n except:\\n pass\\n return self.rating_return(req)\",\n \"def like_comment(username):\\n\\n user_ID = before_request()\\n user_ID = None\\n if user_ID != None:\\n user_ID = str(g.user['_id'])\\n\\n message_id = request.args.get('message_id')\\n profile_usr = request.args.get('profile_usr')\\n count = float(1)\\n\\n redis_obj.zincrby('add_like', message_id, count)\\n\\n print \\\"message ID\\\", message_id\\n\\n if redis_obj.get(user_ID):\\n return redirect(url_for('timeline', username=profile_usr, userId=pickle.loads(redis_obj.get(user_ID))))\\n else:\\n \\\"\\\"\\\" Invalidating cache \\\"\\\"\\\"\\n redis_obj.delete('leaderboard-key')\\n redis_obj.delete(session['user_id'])\\n print \\\"Invalidating cache after Like\\\"\\n return redirect(url_for('timeline', username=profile_usr))\",\n \"def user_reset(user_id):\\n\\n con = connect()\\n cursor = con.cursor()\\n cursor.execute(f\\\"\\\"\\\"\\n START TRANSACTION;\\n \\n DELETE FROM ratings\\n WHERE user_id = {user_id};\\n \\n DELETE FROM counts_by_rating_type\\n WHERE user_id = {user_id};\\n \\n {_count_reset_query(user_id, 1)}\\n {_count_reset_query(user_id, 2)}\\n {_count_reset_query(user_id, 3)}\\n \\n COMMIT;\\n \\\"\\\"\\\")\\n cursor.close()\\n return jsonify({'message': 'ok'}), 200\",\n \"def test_downvote_then_upvote_same_user_leaves_comment_score_one_greater(self):\\n comment = Comment.objects.get(body=\\\"987XYZ\\\")\\n # self.assertEqual(len(post_qs), 1)\\n self.assertEqual(comment.score, DEFAULT_SCORE)\\n comment = Comment.objects.get(body=\\\"987XYZ\\\")\\n\\n vote1 = Vote.create(comment=comment, value=-1, voter=self.user)\\n comment = Comment.objects.get(body=\\\"987XYZ\\\")\\n self.assertEqual(comment.score, DEFAULT_SCORE - 1)\\n\\n vote2 = Vote.create(comment=comment, value=1, voter=self.user)\\n comment = Comment.objects.get(body=\\\"987XYZ\\\")\\n self.assertEqual(comment.score, DEFAULT_SCORE + 1)\",\n \"def deduct_skill_points(self, amount):\\r\\n self.__skill_points -= amount\\r\\n self.__skill_points_indicator.configure(\\r\\n text=\\\"Available skill points: \\\" + str(self.__skill_points))\",\n \"def _decrease_counter(self, response):\\n response_id = response.meta['__id']\\n spot = self._request_registry[response_id]\\n spot['counter'] = spot.get('counter', 0) - 1\",\n \"def unfollow(self, followerId: int, followeeId: int) -> None:\\n if followerId not in self.users.keys() or followeeId not in self.users.keys():\\n return\\n if followeeId not in self.users[followerId].followees.keys():\\n return\\n self.users[followerId].followees.pop(followeeId)\\n\\n\\n\\n # Your Twitter object will be instantiated and called as such:\\n # obj = Twitter()\\n # obj.postTweet(userId,tweetId)\\n # param_2 = obj.getNewsFeed(userId)\\n # obj.follow(followerId,followeeId)\\n # obj.unfollow(followerId,followeeId)\",\n \"def test_unlike_activity(self):\\n from .mockers import user_status_context\\n from .mockers import subscribe_context, create_context\\n username = 'messi'\\n username_not_me = 'xavi'\\n self.create_user(username)\\n self.create_user(username_not_me)\\n self.create_context(create_context)\\n self.admin_subscribe_user_to_context(username, subscribe_context)\\n self.admin_subscribe_user_to_context(username_not_me, subscribe_context)\\n res = self.create_activity(username, user_status_context)\\n activity_id = res.json['id']\\n self.testapp.post('/activities/%s/likes' % activity_id, '', oauth2Header(username_not_me), status=201)\\n self.testapp.delete('/activities/%s/likes/%s' % (activity_id, username_not_me), '', oauth2Header(username_not_me), status=204)\\n activity = self.testapp.get('/activities/%s' % activity_id, '', oauth2Header(username), status=200)\\n\\n self.assertEqual(activity.json['likes'], [])\\n self.assertEqual(activity.json['liked'], False)\\n self.assertEqual(activity.json['likesCount'], 0)\",\n \"def test_unpopular(self):\\n self.assertFalse(self.user3.is_popular())\\n self.user3.receive_upvotes(randint(101, 10000))\\n self.assertTrue(self.user3.is_popular())\",\n \"def increase_score(self):\\n\\n old_score = self.get_score()\\n new_score = old_score + 1\\n sql = \\\"UPDATE Users SET score = ? WHERE username = ?\\\"\\n self.conn.execute(sql, (new_score, self.username))\\n self.conn.commit()\",\n \"def backwards(self, orm):\\n \\n user_type = orm['auth.User']\\n user = user_type.objects.get(username=self.suUsername)\\n \\n profile_type = orm['mooi.Profile']\\n userProfile = profile_type.objects.get(user=user)\\n \\n userProfile.delete()\\n user.delete()\",\n \"def is_liked(value, user: User):\\n return value.is_liked(user)\",\n \"def bass_decrease():\\n request_command(tv_command=TVCommand.bassDecrease)\",\n \"def decrement_comment_count():\\n\\n with _conn.cursor() as cur:\\n cur.execute(\\\"UPDATE comment_count SET count = count - 1;\\\")\\n _conn.commit()\",\n \"def upgrad_one_contribute_score(user_total_click_num):\\n return 1/math.log(1 + user_total_click_num)\",\n \"def unfollowUser(following):\\n \\n cur, user_id, con = initialise(3, True)\\n cur.execute(\\\"DELETE FROM followers WHERE user = (SELECT username FROM users WHERE id = ?) AND following = ?\\\", (user_id, following))\\n finish(con)\"\n]","isConversational":false},"negative_scores":{"kind":"list like","value":["0.7092867","0.69812095","0.6517912","0.61881864","0.61731464","0.6097247","0.6096408","0.6036876","0.5903274","0.58781475","0.58778524","0.5823332","0.5778853","0.5737949","0.5679264","0.563741","0.55705523","0.5514937","0.54584485","0.5454046","0.5450492","0.54445386","0.54407024","0.5418183","0.5418168","0.5417053","0.54000807","0.53698355","0.5369245","0.5329137","0.53256106","0.5323234","0.5299713","0.5294256","0.52934587","0.52898693","0.5286788","0.5279086","0.5268312","0.5268302","0.52497464","0.5248356","0.52102375","0.5202129","0.5198457","0.5189717","0.51837957","0.51785535","0.51781523","0.51711315","0.51637644","0.5156253","0.51352984","0.5130796","0.5122848","0.51200557","0.50891304","0.5086492","0.50818616","0.50803304","0.5070948","0.50699675","0.50542384","0.5048177","0.5043408","0.5039869","0.50293344","0.50247985","0.50213426","0.5010534","0.50077003","0.50020516","0.50006795","0.49949592","0.49766383","0.4963524","0.4963524","0.4963524","0.49626923","0.49572334","0.49548087","0.49470112","0.49246806","0.49240986","0.4922845","0.49226567","0.4917498","0.48935694","0.48924816","0.48895052","0.4885018","0.48766908","0.48706737","0.48693994","0.48664388","0.4865741","0.48634017","0.48612535","0.48537165","0.48515657"],"string":"[\n \"0.7092867\",\n \"0.69812095\",\n \"0.6517912\",\n \"0.61881864\",\n \"0.61731464\",\n \"0.6097247\",\n \"0.6096408\",\n \"0.6036876\",\n \"0.5903274\",\n \"0.58781475\",\n \"0.58778524\",\n \"0.5823332\",\n \"0.5778853\",\n \"0.5737949\",\n \"0.5679264\",\n \"0.563741\",\n \"0.55705523\",\n \"0.5514937\",\n \"0.54584485\",\n \"0.5454046\",\n \"0.5450492\",\n \"0.54445386\",\n \"0.54407024\",\n \"0.5418183\",\n \"0.5418168\",\n \"0.5417053\",\n \"0.54000807\",\n \"0.53698355\",\n \"0.5369245\",\n \"0.5329137\",\n \"0.53256106\",\n \"0.5323234\",\n \"0.5299713\",\n \"0.5294256\",\n \"0.52934587\",\n \"0.52898693\",\n \"0.5286788\",\n \"0.5279086\",\n \"0.5268312\",\n \"0.5268302\",\n \"0.52497464\",\n \"0.5248356\",\n \"0.52102375\",\n \"0.5202129\",\n \"0.5198457\",\n \"0.5189717\",\n \"0.51837957\",\n \"0.51785535\",\n \"0.51781523\",\n \"0.51711315\",\n \"0.51637644\",\n \"0.5156253\",\n \"0.51352984\",\n \"0.5130796\",\n \"0.5122848\",\n \"0.51200557\",\n \"0.50891304\",\n \"0.5086492\",\n \"0.50818616\",\n \"0.50803304\",\n \"0.5070948\",\n \"0.50699675\",\n \"0.50542384\",\n \"0.5048177\",\n \"0.5043408\",\n \"0.5039869\",\n \"0.50293344\",\n \"0.50247985\",\n \"0.50213426\",\n \"0.5010534\",\n \"0.50077003\",\n \"0.50020516\",\n \"0.50006795\",\n \"0.49949592\",\n \"0.49766383\",\n \"0.4963524\",\n \"0.4963524\",\n \"0.4963524\",\n \"0.49626923\",\n \"0.49572334\",\n \"0.49548087\",\n \"0.49470112\",\n \"0.49246806\",\n \"0.49240986\",\n \"0.4922845\",\n \"0.49226567\",\n \"0.4917498\",\n \"0.48935694\",\n \"0.48924816\",\n \"0.48895052\",\n \"0.4885018\",\n \"0.48766908\",\n \"0.48706737\",\n \"0.48693994\",\n \"0.48664388\",\n \"0.4865741\",\n \"0.48634017\",\n \"0.48612535\",\n \"0.48537165\",\n \"0.48515657\"\n]","isConversational":false},"document_score":{"kind":"string","value":"0.8061975"},"document_rank":{"kind":"string","value":"0"}}},{"rowIdx":93285,"cells":{"query":{"kind":"string","value":"Decrease likes_count for profile when Entity or Comment is unliked."},"document":{"kind":"string","value":"def on_deleted_like(sender, instance: dillo.models.mixins.Likes, **kwargs):\n if not instance.content_object:\n return\n target_user = instance.content_object.user\n profile_likes_count_decrease(target_user)\n log.debug('Decreased like count for user %s' % target_user)"},"metadata":{"kind":"string","value":"{\n \"objective\": {\n \"self\": [],\n \"paired\": [],\n \"triplet\": [\n [\n \"query\",\n \"document\",\n \"negatives\"\n ]\n ]\n }\n}"},"negatives":{"kind":"list like","value":["def profile_likes_count_decrease(target_user: User):\n try:\n with transaction.atomic():\n dillo.models.profiles.Profile.objects.filter(user=target_user).update(\n likes_count=F('likes_count') - 1\n )\n except IntegrityError:\n log.warning('Integrity error when incrementing likes count for user %i' % target_user.id)\n target_user.profile.recalculate_likes()","def unlike(self, data_base, user):\n cursor = data_base.cursor()\n cursor.execute(f\"UPDATE post SET likes = likes - 1 WHERE id = '{self.id}'\") # Decrements the likes\n cursor.execute(f\"DELETE FROM user_like WHERE user_id = {user.id} AND post_id = {self.id}\")\n if self.commit_to_db:\n data_base.commit()\n cursor.close()","def unlike(self, request, pk=None):\n\n user_wall_post = self.get_object()\n user_wall_post.likes.remove(self.request.user)\n return Response(status=201)","def update_likes(self):\n self.nb_likes = self.likes.count()\n self.save()","def get_dislikes(self, obj):\n return QuestionPersonLike.objects.filter(question=obj,\n like=False).count()","def test_unauthenticated_user_disliking(self):\n self.like_dislike(self.like_url(6))","def unfollow_profile(self):\n self.find_clickable_element(self.ISFOLLOWED_BTN).click()","def users_like_changed(sender, instance, **kwargs):\n instance.total_likes = instance.users_like.count()\n instance.save()","def unlike():\n if CURR_USER_KEY in session:\n cafe_id = int(request.json[\"cafe_id\"])\n user_id = g.user.id\n Like.query.filter_by(cafe_id=cafe_id, user_id=user_id).delete()\n db.session.commit()\n\n return jsonify({\"unliked\": cafe_id})\n return jsonify({\"error\": \"Not logged in\"})","def unlike(self, request, pk=None):\n\n obj = self.get_object()\n like_func.remove_like(obj, request.user)\n return Response(status=status.HTTP_202_ACCEPTED)","def test_unauthenticated_user_liking(self):\n self.like_dislike(self.dislike_url(5))","def decrement_comment_count():\n\n with _conn.cursor() as cur:\n cur.execute(\"UPDATE comment_count SET count = count - 1;\")\n _conn.commit()","def post(self, request, *args, **kwargs):\n\n user_wall_post_comment = self.get_object()\n user_wall_post_comment.likes.remove(self.request.user)\n return Response(status=201)","def remove_like(request):\n if request.method == \"POST\":\n if \"token\" in request.data and request.data[\"token\"] != \"\" and request.data[\"token\"] is not None:\n if Token.objects.filter(key=request.data[\"token\"]).exists():\n token = get_object_or_404(Token, key=request.data[\"token\"])\n if Post.objects.filter(pk=request.data[\"post_id\"]).exists():\n post = Post.objects.get(pk=request.data[\"post_id\"])\n if Like.objects.filter(post=post, user_id=token.user_id).exists():\n post.count_likes -= 1\n post.save()\n Like.objects.filter(post=post, user_id=token.user_id).delete()\n serializer = PostSerializer(post, context={'user_id': token.user_id})\n return Response({\"success\": 33,\n \"post\": serializer.data})\n else:\n return Response({\"error\": 34})\n\n else:\n return Response({\"error\": 32})\n else:\n return Response({\"error\": 17})","def get_user_likes(self) -> int:\n return -1","def get_likes_count(self, instance):\n return instance.likes.count()","def get_likes_count(self, instance):\n return instance.likes.count()","def get_likes_count(self, instance):\n return instance.likes.count()","def test_unlike_report(self):\n\n self.report.likes = {0, self.user2.id}\n db.session.commit()\n report = Report.query.get(self.report.id)\n # make sure like is there in report.likes.\n self.assertEqual(report.likes, {0, self.user2.id})\n\n with self.client as c:\n with c.session_transaction() as sess:\n sess['user_id'] = self.user2.id\n\n resp = c.post(f'/v1/report/like{self.report.id}')\n\n self.assertEqual(resp.json['success'], 'removed')\n report = Report.query.get(self.report.id)\n self.assertEqual(report.likes, {0})","def like(self):\n self.like_count = self.like_count + 1 if self.like_count else 1","def unlike_comment(username):\n user_ID = before_request()\n user_ID = None\n if user_ID != None:\n user_ID = str(g.user['_id'])\n\n message_id = request.args.get('message_id')\n profile_usr = request.args.get('profile_usr')\n count=float(-1)\n\n print \"username who liked message\", username\n print \"Id of the message\", message_id\n\n redis_obj.zincrby('add_like', message_id, count)\n\n print redis_obj.scard(message_id)\n\n if redis_obj.get(user_ID):\n return redirect(url_for('timeline', username=profile_usr, userId=pickle.loads(redis_obj.get(user_ID))))\n else:\n \"\"\" Invalidating cache \"\"\"\n redis_obj.delete('leaderboard-key')\n redis_obj.delete(session['user_id'])\n print \"Invalidating cache after unlike\"\n return redirect(url_for('timeline', username=profile_usr))","def dislikes(self):\n return self.get_queryset().filter(vote__lt=0)","def unlike_event(request, pk):\n template_var = base_template_vals(request)\n previousLike = Like.objects.filter(event=Event.objects.get(id=pk), \n user=UserProfile.objects.filter(\n django_user=request.user)[0])\n if(len(previousLike) != 0):\n like = previousLike[0]\n like.delete()\n return redirect('index')","def like(self, n: int) -> None:\n\n # YOUR CODE HERE\n self.likes += 1","def test_unlike_activity_get_other_likes(self):\n from .mockers import user_status_context\n from .mockers import subscribe_context, create_context\n username = 'messi'\n username_not_me = 'xavi'\n self.create_user(username)\n self.create_user(username_not_me)\n self.create_context(create_context)\n self.admin_subscribe_user_to_context(username, subscribe_context)\n self.admin_subscribe_user_to_context(username_not_me, subscribe_context)\n res = self.create_activity(username, user_status_context)\n activity_id = res.json['id']\n res = self.testapp.post('/activities/%s/likes' % activity_id, '', oauth2Header(username_not_me), status=201)\n res = self.testapp.post('/activities/%s/likes' % activity_id, '', oauth2Header(username), status=201)\n res = self.testapp.delete('/activities/%s/likes/%s' % (activity_id, username_not_me), '', oauth2Header(username_not_me), status=204)\n\n activity = self.testapp.get('/activities/%s' % activity_id, '', oauth2Header(username), status=200)\n\n self.assertEqual(activity.json['likes'][0]['username'], username)\n self.assertEqual(activity.json['liked'], True)\n self.assertEqual(activity.json['likesCount'], 1)","def test_like_unlike_works_without_error(self):\n create_response = self.client.post(reverse('posts:comment_create'),\n data={\n 'post': self.post.id,\n 'user': self.user.id,\n 'comment_description':\n 'This is a test_comment'\n }, format='json')\n self.assertEqual(create_response.status_code, status.HTTP_201_CREATED)\n # initial stars_count is zero, so when liked must increase by 1\n like_response = self.like_unlike_comment()\n self.assertEqual(like_response.status_code, status.HTTP_200_OK)\n self.assertEqual(like_response.data['stars_count'], 1)\n like_again = self.like_unlike_comment()\n self.assertEqual(like_again.status_code, status.HTTP_200_OK)\n self.assertEqual(like_again.data['stars_count'], 2)\n # current likes/stars count is 2 so if unliked should become 1\n unlike_comment_response = self.like_unlike_comment(pk='1',\n action='unlike')\n self.assertEqual(unlike_comment_response.status_code,\n status.HTTP_200_OK)\n self.assertEqual(unlike_comment_response.data['stars_count'], 1)\n # current likes/stars count is 1 so if unliked should become 0\n unlike_comment_response = self.like_unlike_comment(pk='1',\n action='unlike')\n self.assertEqual(unlike_comment_response.status_code,\n status.HTTP_200_OK)\n self.assertEqual(unlike_comment_response.data['stars_count'], 0)\n # current likes/stars count is 0 so if unliked should remain 0\n unlike_comment_response = self.like_unlike_comment(pk='1',\n action='unlike')\n self.assertEqual(unlike_comment_response.status_code,\n status.HTTP_200_OK)\n self.assertEqual(unlike_comment_response.data['stars_count'], 0)","def user_response_to_post(self, request, pk):\n post_objects_count = Post.objects.filter(id=pk, liked_users__id=request.user.id).count()\n post_objects = Post.objects.get(id=pk)\n if post_objects_count !=0:\n post_objects.liked_users.remove(request.user)\n response_msg = \"You disliked the post\"\n else:\n post_objects.liked_users.add(request.user)\n response_msg = \"You have liked the post\"\n return Response({'data': response_msg}, status=status.HTTP_200_OK)","def total_likes(self):\n\n return (self.likes.count()-self.dislikes.count())","def test_unlike(self):\n\n message = Message(text=\"hello world\", user_id=self.testuser2.id)\n \n db.session.add(message)\n db.session.commit()\n\n like = Likes(message_id=message.id, user_id=self.testuser.id)\n \n db.session.add(like)\n db.session.commit()\n \n message_id = message.id\n\n with self.client as c:\n with c.session_transaction() as session:\n session[CURR_USER_KEY] = self.testuser.id\n \n response = c.post(f\"/messages/{message_id}/like\", follow_redirects=True)\n \n self.assertEqual(response.status_code, 200)\n\n likes = Likes.query.filter(Likes.message_id==message_id).all()\n self.assertEqual(len(likes), 0)","def dislike(self, event, number):\n\n if not hasattr(self, 'user_id') or not hasattr(self, 'top_id'):\n vk_common.dprint(2, vk_common.func_name(), f\"Не заданы атрибуты user_id = или top_id.\")\n return\n\n item_id = self.item_ids[self.top_id - 1]\n if item_id is None:\n vk_common.dprint(2, vk_common.func_name(), f\"Атрибут item_id = None.\")\n return\n else:\n vk_common.dprint(2, vk_common.func_name(), f\"number = {number}; item_id = {item_id}\")\n\n vk_common.dprint(2, vk_common.func_name(), f\"number = {number}; item_id = {item_id}\")\n vk_common.dprint(1, f\"имя файла с фото = \\\"{self.user_id}_{self.top_id}\\\"; \"\n f\"ссылка = https://vk.com/id{self.user_id}\")\n\n err, msg, likes = vk_common.dislike_photo(self.user_id, item_id)\n if err != 0:\n mb.showinfo(\"Информация\", f\"Не смогли убрать лайк у фото пользователя с id = {self.user_id}.\")\n vk_common.dprint(2, \"Не смогли убрать лайк.\")\n return\n mb.showinfo(\"Информация\", f\"Убрали лайк у фото пользователя с id = {self.user_id}.\")\n vk_common.dprint(2, f\"Убрали лайк: owner_id = {self.user_id}; item_id = {item_id}\")","def add_unsaved_likes_to_user(sender, user, request, **kwargs):\n session_likes = request.session.get('likes')\n if session_likes:\n user.userprofile.liked_products.add(*session_likes)","def like_unlike_comment(self, pk='1', action='like'):\n return self.client.post(reverse('posts:like_unlike_comment',\n kwargs={\n 'pk': pk,\n 'action': action\n }))","def remove_like(obj, user):\n\tobj_type = ContentType.objects.get_for_model(obj)\n\tLike.objects.filter(\n\t\tcontent_type = obj_type,\n\t\tobject_id = obj.id,\n\t\tuser = user).delete()","def unlike_comment_API(id):\n\n count = float(-1)\n redis_obj.zincrby('add_like', id, count)\n\n \"\"\" Invalidating cache \"\"\"\n redis_obj.delete('leaderboard-key')\n return jsonify(username=g.user[1], Message=\"Success! Message unliked successfully. \", Status_code=status.HTTP_200_OK)","def count_likes(self):\n likes = self.event_likes\n num_likes = len(likes)\n return num_likes","def unlike(db, filename, user):\n cur = db.cursor()\n if like_exists(db, filename, user):\n sql = \"\"\"\n delete from likes where filename=? and usernick=?;\n \"\"\"\n cur.execute(sql, (filename, user))\n db.commit()","def test_unlike_mission(self):\n\n self.mission.likes = {0, self.user2.id}\n db.session.commit()\n mission = Mission.query.get(self.mission.id)\n # make sure like is there in mission.likes.\n self.assertEqual(mission.likes, {0, self.user2.id})\n\n with self.client as c:\n with c.session_transaction() as sess:\n sess['user_id'] = self.user2.id\n\n resp = c.post(f'/v1/mission/like{self.mission.id}')\n\n self.assertEqual(resp.json['success'], 'removed')\n mission = Mission.query.get(self.mission.id)\n self.assertEqual(mission.likes, {0})","def unblock_profile(self, request, *args, **kwargs):\n context = {\n 'conversation': self.get_object(),\n 'request': request\n }\n serializer = UnblockProfileSerializer(data=request.data, context=context)\n serializer.is_valid(raise_exception=True)\n serializer.save()\n return Response(serializer.data, status=status.HTTP_202_ACCEPTED)","def unseen_count_for(self, user):\r\n return self.filter(user=user, unseen=True).count()","def test_unlike_activity(self):\n from .mockers import user_status_context\n from .mockers import subscribe_context, create_context\n username = 'messi'\n username_not_me = 'xavi'\n self.create_user(username)\n self.create_user(username_not_me)\n self.create_context(create_context)\n self.admin_subscribe_user_to_context(username, subscribe_context)\n self.admin_subscribe_user_to_context(username_not_me, subscribe_context)\n res = self.create_activity(username, user_status_context)\n activity_id = res.json['id']\n self.testapp.post('/activities/%s/likes' % activity_id, '', oauth2Header(username_not_me), status=201)\n self.testapp.delete('/activities/%s/likes/%s' % (activity_id, username_not_me), '', oauth2Header(username_not_me), status=204)\n activity = self.testapp.get('/activities/%s' % activity_id, '', oauth2Header(username), status=200)\n\n self.assertEqual(activity.json['likes'], [])\n self.assertEqual(activity.json['liked'], False)\n self.assertEqual(activity.json['likesCount'], 0)","def dislike(request, pk):\n update_posts_expiration()\n post = 0;\n post = Post.objects.get(id=pk)\n if post.is_expired == True:\n return Response(\"You can't interact with a expired post\")\n else:\n if request.method == \"POST\":\n if request.user != post.user:\n post.dislikes_count += 1\n post.save()\n serializer = DislikeSerializer(data=request.data)\n if serializer.is_valid():\n serializer.save(user=request.user, disliked_post=post)\n return Response(\"you disliked the post with title: \" + post.title)\n else:\n return Response(serializer.errors)\n else:\n return Response(\"you can't dislike your own post\")\n \n\n return Response(\"you didn't dislike the post yet\")","def backwards(self, orm):\n \n user_type = orm['auth.User']\n user = user_type.objects.get(username=self.suUsername)\n \n profile_type = orm['mooi.Profile']\n userProfile = profile_type.objects.get(user=user)\n \n userProfile.delete()\n user.delete()","def delete(self, request, username):\n\n # Retrieve the user from the user table if the user exists\n try:\n user_to_unfollow = User.objects.get(username=username)\n current_user = request.user\n # If a user is trying to unfollow themselves then stop the request\n if user_to_unfollow.profile.id == current_user.profile.id:\n return Response(\n {\"errors\": FOLLOW_USER_MSGS['CANNOT_UNFOLLOW_SELF']},\n status=status.HTTP_400_BAD_REQUEST\n )\n # Check if the user to be unfollowed\n # is in the current users following list\n try:\n profile_id = user_to_unfollow.profile.id\n user_being_followed = CustomFollows.objects.get(\n to_profile_id=profile_id,\n from_profile_id=current_user.profile.id\n )\n # If not tell the user the request can't happen\n # Because they don't follow the user\n except Exception as e:\n return Response(\n {\n \"errors\": FOLLOW_USER_MSGS['USER_UNFOLLOWED_ALREADY']\n },\n status=status.HTTP_400_BAD_REQUEST\n )\n # Otherwise unfollow the user as requested\n current_user.profile.follows.remove(user_to_unfollow.profile)\n # Get the following & followers username list\n # And the following & followers count for the current user\n user_following_data = get_user_following_data(current_user)\n return Response(\n {\n \"message\": FOLLOW_USER_MSGS['USER_UNFOLLOW_SUCCESSFUL'],\n \"following\": user_following_data[\"following\"],\n \"followers\": user_following_data[\"followers\"],\n \"followingCount\": user_following_data[\"followingCount\"],\n \"followersCount\": user_following_data[\"followersCount\"]\n },\n status=status.HTTP_200_OK\n )\n # End request if we cannot find the user we want to unfollow.\n except User.DoesNotExist:\n return Response(\n {\"errors\": FOLLOW_USER_MSGS['USER_NOT_FOUND']},\n status=status.HTTP_404_NOT_FOUND\n )","def test_downvote_then_upvote_same_user_leaves_comment_score_one_greater(self):\n comment = Comment.objects.get(body=\"987XYZ\")\n # self.assertEqual(len(post_qs), 1)\n self.assertEqual(comment.score, DEFAULT_SCORE)\n comment = Comment.objects.get(body=\"987XYZ\")\n\n vote1 = Vote.create(comment=comment, value=-1, voter=self.user)\n comment = Comment.objects.get(body=\"987XYZ\")\n self.assertEqual(comment.score, DEFAULT_SCORE - 1)\n\n vote2 = Vote.create(comment=comment, value=1, voter=self.user)\n comment = Comment.objects.get(body=\"987XYZ\")\n self.assertEqual(comment.score, DEFAULT_SCORE + 1)","def user_unfollow():\n data = request.get_json(force=True)\n follower = User.query.get(data['follower'])\n following = User.query.get(data['following'])\n follower.followcheck.remove(following)\n db.session.commit()\n return {'unfollowed': True}","def count_downvotes(self):\n return self.filter(value=-1).count()","def addLikeCount(self,count):\n self.interactionCount += count\n return None","def test_removal_from_public_index(self):\n before = (S(UserProfile)\n .indexes(UserProfile.get_index(public_index=True))\n .count())\n profile = self.mozillian2.userprofile\n profile.privacy_full_name = MOZILLIANS\n profile.save()\n sleep(1)\n after = (S(UserProfile)\n .indexes(UserProfile.get_index(public_index=True))\n .count())\n eq_(after+1, before)","def decrement_count(self, state):\n if state == domain.REVIEW_STATE_ASSIGNED:\n self.assigned_count -= 1\n elif state == domain.REVIEW_STATE_COMPLETED:\n self.completed_count -= 1\n elif state == domain.REVIEW_STATE_EXPIRED:\n self.expired_count -= 1\n else:\n raise ValueError('%s not in %s' % (state, domain.REVIEW_STATES))","def relation_downvote(request, pk):\n try:\n relation = Relation.objects.get(pk=pk)\n except Relation.DoesNotExist:\n return Response(status=status.HTTP_404_NOT_FOUND)\n\n if request.method == 'PUT':\n relation.negative_reaction_count += 1\n relation.save()\n serializer = RelationSerializer(relation)\n return Response(serializer.data)","def remove_count(self, denom: CashDenomination, count: int) -> None:\n if not self.__open:\n raise RuntimeError(\"Cash drawer must be open to modify.\")\n if count < 0:\n raise ValueError(\"Count must not be negative.\")\n if count > self.__contents[denom]:\n raise ValueError(\"Cannot remove more than are present.\")\n self.__contents[denom] -= count","def latest_likes(self, user, number_posts, likes):\n WAIT = 1\n if likes:\n action = 'Like'\n else:\n action = 'Unlike'\n self.nav_user(user)\n image_container = []\n image_container.extend(self.driver.find_elements_by_class_name('_9AhH0'))\n for image in image_container[:number_posts]:\n image.click()\n time.sleep(WAIT)\n try:\n self.driver.find_element_by_xpath(\"//*[@aria-label='{}']\".format(action).click())\n except Exception as e:\n print(e)\n self.driver.find_elements_by_xpath('/html/body/div[4]/div[2]/div/article/div[3]/section[1]/span[1]/button')[0].click() # clicks the heart symbol\n time.sleep(WAIT)\n self.driver.find_elements_by_xpath('/html/body/div[4]/div[3]/button')[0].click() #Makes sure to close out of current picture\n time.sleep(WAIT)\n \n # Tested\n users_list = []\n def get_likes_list(self, username):\n \"\"\"\n Method gets a list of users who like a post\n\n \"\"\"\n api = self.api\n api.searchUsername(username) \n result = api.LastJson\n username_id = result['user']['pk'] #Gets the user ID\n user_posts = api.getUserFeed(username_id) # gets the user feed\n result = api.LastJson\n media_id = result['items'][0]['id'] #gets the most recent post\n api.getMediaLikers(media_id) #gets users who liked\n users = api.LastJson('users')\n for user in users: #appends the users to the list\n users_list.append({'pk':user['pk'], 'username':user['username']})","def unfollow_me(self):\n return self.follow_me('unfollow_topic')","def dislike_post(request, pk):\r\n post = get_object_or_404(Post, id=request.POST.get('id'))\r\n disliked = False\r\n if post.dislikes.filter(id=request.user.id).exists():\r\n post.dislikes.remove(request.user)\r\n disliked = False\r\n else:\r\n if post.likes.filter(id=request.user.id).exists():\r\n post.likes.remove(request.user)\r\n post.dislikes.add(request.user)\r\n disliked = True\r\n\r\n if request.is_ajax():\r\n return JsonResponse({'total_dislikes': post.dislikes.count(),\r\n 'total_likes': post.likes.count(),\r\n 'disliked': disliked})","def test_upvote_then_downvote_same_user_leaves_comment_score_one_less(self):\n comment = Comment.objects.get(body=\"987XYZ\")\n # self.assertEqual(len(post_qs), 1)\n self.assertEqual(comment.score, DEFAULT_SCORE)\n comment = Comment.objects.get(body=\"987XYZ\")\n\n vote1 = Vote.create(comment=comment, value=1, voter=self.user)\n comment = Comment.objects.get(body=\"987XYZ\")\n self.assertEqual(comment.score, DEFAULT_SCORE + 1)\n\n vote2 = Vote.create(comment=comment, value=-1, voter=self.user)\n comment = Comment.objects.get(body=\"987XYZ\")\n self.assertEqual(comment.score, DEFAULT_SCORE - 1)","def unlike(id):\n spinner = Halo(text=\"Unliking photo...\", spinner=\"dots\").start()\n try:\n api.unlike(id)\n spinner.succeed(\"Unliked\")\n except Exception:\n spinner.fail(\"Failed to unlike photo.\")","def _decrement_num_user_stars(user_id, match_id, now):\n\tnum_user_stars = session.query(CalendarEntry.num_user_stars)\\\n\t\t\t.filter(\n\t\t\t\tCalendarEntry.user_id == user_id,\n\t\t\t\tCalendarEntry.match_id == match_id)\\\n\t\t\t.one()\\\n\t\t\t.num_user_stars\n\tif num_user_stars > 1:\n\t\t# Decrement the count of stars for the CalendarEntry to a positive integer.\n\t\tsession.execute(CalendarEntries.update()\n\t\t\t\t.where(sa.and_(\n\t\t\t\t\tCalendarEntry.user_id == user_id,\n\t\t\t\t\tCalendarEntry.match_id == match_id))\n\t\t\t\t.values({CalendarEntry.num_user_stars: num_user_stars - 1}))\n\telse:\n\t\t# Delete the CalendarEntry becuase the count of stars is now zero.\n\t\tsession.execute(CalendarEntries.delete().where(sa.and_(\n\t\t\t\tCalendarEntry.user_id == user_id,\n\t\t\t\tCalendarEntry.match_id == match_id)))","def unsubscribed_member_count(self, unsubscribed_member_count):\n\n self._unsubscribed_member_count = unsubscribed_member_count","def unfollow_friend(username):\n\n if not g.user:\n print \"401\"\n abort(401)\n whom_id = get_user_id(username)\n print whom_id\n if whom_id is None:\n abort(404)\n unfollow_query(whom_id)\n flash('You are no longer following \"%s\"' % username)\n name = {'name of unfollowing user': username}\n ############### REDIS cache invalidate #####################\n R_SERVER.delete(user_timeline_key)\n return jsonify(Username=name, Status_code=status.HTTP_200_OK)","def _decrease_counter(self, response):\n response_id = response.meta['__id']\n spot = self._request_registry[response_id]\n spot['counter'] = spot.get('counter', 0) - 1","def toggle_like(post_id):\n if CURRENT_USER_KEY not in session:\n raise Unauthorized()\n\n liked_post = Post.query.get_or_404(post_id)\n user = User.query.get(session[CURRENT_USER_KEY])\n # toggle the like by removing from user likes\n if liked_post in user.likes:\n user.likes.remove(liked_post)\n else:\n user.likes.append(liked_post)\n\n db.session.commit()\n\n return redirect(url_for('show_likes', user_id=user.id))","def like(request, post_id):\n if request.method == \"PUT\":\n liked = None\n user = request.user\n post = Post.objects.get(id=post_id)\n # If user already liked, decrement the like count and remove as 'liker'\n if user in post.liked_by.all():\n post.liked_by.remove(user)\n post.likes -= 1\n post.save()\n liked = False\n # Else increase like count and add user\n else:\n post.liked_by.add(user)\n post.likes += 1\n post.save()\n liked = True\n # Return data for updating dynamically with javascript\n return JsonResponse({\"total_likes\": post.likes, \"liked\": liked})","def decrement(self, stats, sample_rate=1):\n self.update_stats(stats, -1, sample_rate=sample_rate)","def delete(self, *args, **kwargs):\n self.item.comments_total -= 1\n self.item.save()\n super(Comment, self).delete(*args, **kwargs)","def decrease(self):\n self.score -= self.score","def meal_disliked(request, meal_pk):\n meal_pk = int(meal_pk)\n meal = Meal.objects.get(pk=meal_pk)\n like = False\n member = request.user\n\n try:\n rating = Rating.objects.get(member=member, meal=meal)\n r_percent = meal.percent()\n except ObjectDoesNotExist:\n Rating.objects.create_rating(member, meal, like)\n r_percent = meal.percent()\n\n rating.like = like\n rating.save()\n r_percent = meal.percent()\n return HttpResponse(r_percent)","def dislike_comment(request, pk, pk1):\r\n comment = get_object_or_404(Comment, id=request.POST.get('id'))\r\n disliked = False\r\n if comment.dislikes.filter(id=request.user.id).exists():\r\n comment.dislikes.remove(request.user)\r\n disliked = False\r\n else:\r\n comment.dislikes.add(request.user)\r\n disliked = True\r\n\r\n context = {\r\n 'is_liked': disliked,\r\n 'total_dislikes': comment.dislikes.count()\r\n }\r\n\r\n if request.method == 'POST':\r\n return HttpResponse(\r\n json.dumps(context),\r\n content_type=\"application/json\")","def delete_like(self, entry, **args): \n args.update(entry=entry)\n return self.fetch(\"/like/delete\", post_args=args)","def update_count(self):\n pass # Do nothing","def unfollow(self, other):\n\t\tif self.follows(other):\n\t\t\tself.followed.remove(other)","def get_likes_count():\n return Flag.objects.filter(flag=Flag.LIKE_FLAG).count()","def toggle_vote(self):\n\n self.vote = 1 - self.vote","def test_unlike_comment(comment, mocker):\n unlike_mock = mocker.patch(\"instapi.client.client.comment_unlike\")\n\n comment.unlike()\n\n unlike_mock.assert_called_once_with(comment.pk)","def reset_count(self):\n self.count = 0","def update_count(self):\n pass","def decrement_misses_remaining(self):\n self.misses_remaining -=1","def order_posts_to_likes(sender, instance, created=False, **kwargs):\n user = instance.user\n post = instance.post\n tags = post.tags.all()\n\n likes = Like.objects.filter(user=user, post=post)\n liked_posts = [ i.post for i in likes if i.liked_status==True]\n\n for post in liked_posts:\n liked_tags = post.tags.all()\n for tag in liked_tags:\n if not tag in tags:\n tags.append(tag)\n\n similar_posts = Post.objects.filter(tags__in=tags).distinct()\n try:\n UserPostWeight.objects.filter(user=user).delete()\n except:\n pass\n weight = 2\n for post in similar_posts:\n UserPostWeight.objects.create(user=user, post=post, weight=weight)\n weight+=2\n \n\n\n pass","def is_disliked_by(self, user):\n return user.disliked_articles.filter(pk=self.pk).exists()","def downvote(self):\n url = \"https://api.imgur.com/3/gallery/{0}/vote/down\".format(self.id)\n return self._imgur._send_request(url, needs_auth=True, method='POST')","def unfollow(self, user_index, following_index):\n if user_index >= self.num_users or following_index >= self.num_users:\n raise ValueError(\n f\"Number of users is {self.num_users}, but indices \"\n f\"{user_index} and {following_index} were requested.\"\n )\n if self.users_hat[following_index, user_index] == 1:\n self.users_hat[following_index, user_index] = 0\n elif self.is_verbose():\n self.log(f\"User {following_index} was not following user {user_index}\")","def delete_model(self, request, obj):\n obj.post.likeNumDreacase()\n obj.delete()","async def twitter_unfollow(self, ctx, handle):\n sane_handle = handle.lower().lstrip('@')\n conf = dutils.get(self.conf.follows, screen_name=sane_handle)\n chan_conf = dutils.get(conf.discord_channels, id=ctx.message.channel.id) if conf is not None else None\n\n if chan_conf is None:\n raise TwitterError('Not following {} on this channel.'.format(handle))\n\n # Remove the Discord channel from the Twitter channel conf\n conf.discord_channels.remove(chan_conf)\n if not conf.discord_channels:\n # If there are no more Discord channel to feed, unfollow the Twitter channel\n self.conf.follows.remove(conf)\n del conf\n\n # Update the tweepy stream\n if len(self.conf.follows) > 0:\n await self.stream.start()\n else:\n self.stream.stop()\n\n self.conf.save()\n\n await self.bot.say('\\N{OK HAND SIGN}')","def unfollow_user(username):\n user_ID = before_request()\n user_ID = None\n if user_ID != None:\n user_ID = str(g.user['_id'])\n if not g.user:\n abort(401)\n whom_id = get_user_id(username)\n if whom_id is None:\n abort(404)\n mongo.db.users.update({'_id': g.user['_id']}, {\n '$pull': {'follows': whom_id}})\n flash('You are no longer following \"%s\"' % username)\n if redis_obj.get(user_ID):\n return redirect(url_for('user_timeline', username=username, userId=pickle.loads(redis_obj.get(user_ID))))\n else:\n redis_obj.delete(session['user_id'])\n print \"Invalidating cache after Unfollow\"\n return redirect(url_for('user_timeline', username=username))","def unfollow_user(username):\n if not g.user:\n abort(401)\n whom_id = get_user_id(username)\n if whom_id is None:\n abort(404)\n db = get_db()\n db.execute('delete from follower where who_id=? and whom_id=?',\n [session['user_id'], whom_id])\n db.commit()\n flash('You are no longer following \"%s\"' % username)\n return redirect(url_for('user_timeline', username=username))","def delete(self, request, *args, **kwargs):\n\n following_username = token_payload(request)\n\n followed_username = self.kwargs.get(self.look_url_kwarg)\n try:\n followed_user = User.objects.get(username=followed_username)\n followed_user_id = followed_user.id\n\n is_following = FollowUser.objects.filter(\n following_username=following_username)\n for user in is_following:\n if user.following_username == following_username and user.followed_username == followed_username:\n returned_user = FollowUser.objects.get(pk=user.id)\n returned_user.delete()\n\n followed_user_profile = Profile.objects.get(\n profile_user=followed_user_id)\n serializer_profile = self.serializer_classprofile(\n followed_user_profile)\n return Response({\n \"user\": following_username,\n \"unfollowing\": serializer_profile.data\n })\n return Response({\n \"message\": 'User has been unfollowed or You are unfollowing a user you were not orignally following'\n }, status=status.HTTP_400_BAD_REQUEST)\n except:\n return Response({\n \"message\": 'User does not exist'\n }, status=status.HTTP_400_BAD_REQUEST)","def get_number_of_likes_for_user(self, user):\n\t\tfrom pins.models import Pin\n\t\tpin_ctype = ContentType.objects.get_for_model(Pin)\n\t\tpin_list = Pin.objects.active_pins().filter(board__user=user).values_list('pk', flat=True)\n\t\treturn self.filter(content_type=pin_ctype, object_id__in=pin_list).count()","def like(self, data_base, user):\n cursor = data_base.cursor()\n cursor.execute(f\"UPDATE post SET likes = likes + 1 WHERE id = '{self.id}'\") # Increments the likes\n cursor.execute(f\"INSERT INTO user_like (user_id, post_id) VALUES ({user.id}, {self.id})\")\n if self.commit_to_db:\n data_base.commit()\n cursor.close()","def get_likes(self, obj):\n return QuestionPersonLike.objects.filter(question=obj,\n like=True).count()","def test_unpopular(self):\n self.assertFalse(self.user3.is_popular())\n self.user3.receive_upvotes(randint(101, 10000))\n self.assertTrue(self.user3.is_popular())","def delete_profile_pic(sender, instance, **kwargs):\n if instance.profile_picture:\n if instance.profile_picture.name != \"default.png\":\n path = instance.profile_picture.path\n os.remove(path)","def test_downvote_then_upvote_same_user_leaves_post_score_one_greater(self):\n post = Post.objects.get(body=\"123ABC Body\")\n # self.assertEqual(len(post_qs), 1)\n self.assertEqual(post.score, DEFAULT_SCORE)\n post = Post.objects.get(body=\"123ABC Body\")\n\n vote1 = Vote.create(post=post, value=-1, voter=self.user)\n post = Post.objects.get(body=\"123ABC Body\")\n self.assertEqual(post.score, DEFAULT_SCORE - 1)\n\n vote2 = Vote.create(post=post, value=1, voter=self.user)\n post = Post.objects.get(body=\"123ABC Body\")\n self.assertEqual(post.score, DEFAULT_SCORE + 1)","def lose_life(self):\n self.__num_lives -= 1","def test_update_like_cant_message(self):\n\n self.assertEqual(first=0, second=self.like.likes.all().count())\n url = reverse('like-detail', args=(self.like.id,))\n data = {\n 'likes': [2]\n }\n json_data = json.dumps(data)\n self.client.credentials(HTTP_AUTHORIZATION=self.token)\n response = self.client.patch(path=url, content_type='application/json', data=json_data)\n self.assertFalse(response.data.get('can_message', False))\n self.assertEqual(first=200, second=response.status_code)\n self.assertEqual(first=1, second=self.like.likes.all().count())","def post_likes(request):\n if request.method == 'POST':\n if not request.user.is_authenticated:\n raise PermissionDenied\n pk = request.POST.get('post_id')\n try:\n post = PostsModel.objects.get(id=pk)\n except PostsModel.DoesNotExist:\n post = None\n try:\n likes = LikesModel.objects.get(user=request.user, post=post)\n except LikesModel.DoesNotExist:\n likes = None\n\n\n if likes == None:\n likes = LikesModel.objects.create(user=request.user, post=post)\n post.likes_count = F('likes_count') + 1\n post.save(update_fields=[\"likes_count\"])\n class_likes = 'fa fa-heart mx-2'\n else:\n if likes.is_liked == True:\n likes.is_liked = False\n likes.save()\n post.likes_count = F('likes_count') - 1\n post.save(update_fields=[\"likes_count\"])\n class_likes = 'fa fa-heart-o mx-2'\n elif likes.is_liked == False:\n likes.is_liked = True\n likes.save()\n post.likes_count = F('likes_count') + 1\n post.save(update_fields=[\"likes_count\"])\n class_likes = 'fa fa-heart mx-2'\n post.refresh_from_db()\n response_data = {'_code' : 0, '_status' : 'ok', '_likes': post.likes_count, '_class_likes': class_likes}\n else:\n response_data = {'_code' : 1, '_status' : 'no' }\n\n return JsonResponse(response_data)","def remove_vote_weight(var, target : users.User, amount : int = 1) -> None:\n add_vote_weight(var, target, -amount)","def test_downvote_modifies_comment_score(self):\n comment = Comment.objects.get(body=\"987XYZ\")\n self.assertEqual(comment.score, DEFAULT_SCORE)\n vote = Vote.create(comment=comment, value=-1, voter=self.user)\n comment = Comment.objects.get(body=\"987XYZ\")\n self.assertEqual(comment.score, DEFAULT_SCORE - 1)","def perform_destroy(self, instance):\n self.object.comments -= 1\n self.object.save()\n instance.delete()","def unfollow(self, followerId, followeeId):\r\n self.follows[followerId].discard(followeeId)","def get_user_has_liked(self, instance):\n request = self.context.get(\"request\")\n return instance.likes.filter(pk=request.user.pk).exists()","def get_user_has_liked(self, instance):\n request = self.context.get(\"request\")\n return instance.likes.filter(pk=request.user.pk).exists()"],"string":"[\n \"def profile_likes_count_decrease(target_user: User):\\n try:\\n with transaction.atomic():\\n dillo.models.profiles.Profile.objects.filter(user=target_user).update(\\n likes_count=F('likes_count') - 1\\n )\\n except IntegrityError:\\n log.warning('Integrity error when incrementing likes count for user %i' % target_user.id)\\n target_user.profile.recalculate_likes()\",\n \"def unlike(self, data_base, user):\\n cursor = data_base.cursor()\\n cursor.execute(f\\\"UPDATE post SET likes = likes - 1 WHERE id = '{self.id}'\\\") # Decrements the likes\\n cursor.execute(f\\\"DELETE FROM user_like WHERE user_id = {user.id} AND post_id = {self.id}\\\")\\n if self.commit_to_db:\\n data_base.commit()\\n cursor.close()\",\n \"def unlike(self, request, pk=None):\\n\\n user_wall_post = self.get_object()\\n user_wall_post.likes.remove(self.request.user)\\n return Response(status=201)\",\n \"def update_likes(self):\\n self.nb_likes = self.likes.count()\\n self.save()\",\n \"def get_dislikes(self, obj):\\n return QuestionPersonLike.objects.filter(question=obj,\\n like=False).count()\",\n \"def test_unauthenticated_user_disliking(self):\\n self.like_dislike(self.like_url(6))\",\n \"def unfollow_profile(self):\\n self.find_clickable_element(self.ISFOLLOWED_BTN).click()\",\n \"def users_like_changed(sender, instance, **kwargs):\\n instance.total_likes = instance.users_like.count()\\n instance.save()\",\n \"def unlike():\\n if CURR_USER_KEY in session:\\n cafe_id = int(request.json[\\\"cafe_id\\\"])\\n user_id = g.user.id\\n Like.query.filter_by(cafe_id=cafe_id, user_id=user_id).delete()\\n db.session.commit()\\n\\n return jsonify({\\\"unliked\\\": cafe_id})\\n return jsonify({\\\"error\\\": \\\"Not logged in\\\"})\",\n \"def unlike(self, request, pk=None):\\n\\n obj = self.get_object()\\n like_func.remove_like(obj, request.user)\\n return Response(status=status.HTTP_202_ACCEPTED)\",\n \"def test_unauthenticated_user_liking(self):\\n self.like_dislike(self.dislike_url(5))\",\n \"def decrement_comment_count():\\n\\n with _conn.cursor() as cur:\\n cur.execute(\\\"UPDATE comment_count SET count = count - 1;\\\")\\n _conn.commit()\",\n \"def post(self, request, *args, **kwargs):\\n\\n user_wall_post_comment = self.get_object()\\n user_wall_post_comment.likes.remove(self.request.user)\\n return Response(status=201)\",\n \"def remove_like(request):\\n if request.method == \\\"POST\\\":\\n if \\\"token\\\" in request.data and request.data[\\\"token\\\"] != \\\"\\\" and request.data[\\\"token\\\"] is not None:\\n if Token.objects.filter(key=request.data[\\\"token\\\"]).exists():\\n token = get_object_or_404(Token, key=request.data[\\\"token\\\"])\\n if Post.objects.filter(pk=request.data[\\\"post_id\\\"]).exists():\\n post = Post.objects.get(pk=request.data[\\\"post_id\\\"])\\n if Like.objects.filter(post=post, user_id=token.user_id).exists():\\n post.count_likes -= 1\\n post.save()\\n Like.objects.filter(post=post, user_id=token.user_id).delete()\\n serializer = PostSerializer(post, context={'user_id': token.user_id})\\n return Response({\\\"success\\\": 33,\\n \\\"post\\\": serializer.data})\\n else:\\n return Response({\\\"error\\\": 34})\\n\\n else:\\n return Response({\\\"error\\\": 32})\\n else:\\n return Response({\\\"error\\\": 17})\",\n \"def get_user_likes(self) -> int:\\n return -1\",\n \"def get_likes_count(self, instance):\\n return instance.likes.count()\",\n \"def get_likes_count(self, instance):\\n return instance.likes.count()\",\n \"def get_likes_count(self, instance):\\n return instance.likes.count()\",\n \"def test_unlike_report(self):\\n\\n self.report.likes = {0, self.user2.id}\\n db.session.commit()\\n report = Report.query.get(self.report.id)\\n # make sure like is there in report.likes.\\n self.assertEqual(report.likes, {0, self.user2.id})\\n\\n with self.client as c:\\n with c.session_transaction() as sess:\\n sess['user_id'] = self.user2.id\\n\\n resp = c.post(f'/v1/report/like{self.report.id}')\\n\\n self.assertEqual(resp.json['success'], 'removed')\\n report = Report.query.get(self.report.id)\\n self.assertEqual(report.likes, {0})\",\n \"def like(self):\\n self.like_count = self.like_count + 1 if self.like_count else 1\",\n \"def unlike_comment(username):\\n user_ID = before_request()\\n user_ID = None\\n if user_ID != None:\\n user_ID = str(g.user['_id'])\\n\\n message_id = request.args.get('message_id')\\n profile_usr = request.args.get('profile_usr')\\n count=float(-1)\\n\\n print \\\"username who liked message\\\", username\\n print \\\"Id of the message\\\", message_id\\n\\n redis_obj.zincrby('add_like', message_id, count)\\n\\n print redis_obj.scard(message_id)\\n\\n if redis_obj.get(user_ID):\\n return redirect(url_for('timeline', username=profile_usr, userId=pickle.loads(redis_obj.get(user_ID))))\\n else:\\n \\\"\\\"\\\" Invalidating cache \\\"\\\"\\\"\\n redis_obj.delete('leaderboard-key')\\n redis_obj.delete(session['user_id'])\\n print \\\"Invalidating cache after unlike\\\"\\n return redirect(url_for('timeline', username=profile_usr))\",\n \"def dislikes(self):\\n return self.get_queryset().filter(vote__lt=0)\",\n \"def unlike_event(request, pk):\\n template_var = base_template_vals(request)\\n previousLike = Like.objects.filter(event=Event.objects.get(id=pk), \\n user=UserProfile.objects.filter(\\n django_user=request.user)[0])\\n if(len(previousLike) != 0):\\n like = previousLike[0]\\n like.delete()\\n return redirect('index')\",\n \"def like(self, n: int) -> None:\\n\\n # YOUR CODE HERE\\n self.likes += 1\",\n \"def test_unlike_activity_get_other_likes(self):\\n from .mockers import user_status_context\\n from .mockers import subscribe_context, create_context\\n username = 'messi'\\n username_not_me = 'xavi'\\n self.create_user(username)\\n self.create_user(username_not_me)\\n self.create_context(create_context)\\n self.admin_subscribe_user_to_context(username, subscribe_context)\\n self.admin_subscribe_user_to_context(username_not_me, subscribe_context)\\n res = self.create_activity(username, user_status_context)\\n activity_id = res.json['id']\\n res = self.testapp.post('/activities/%s/likes' % activity_id, '', oauth2Header(username_not_me), status=201)\\n res = self.testapp.post('/activities/%s/likes' % activity_id, '', oauth2Header(username), status=201)\\n res = self.testapp.delete('/activities/%s/likes/%s' % (activity_id, username_not_me), '', oauth2Header(username_not_me), status=204)\\n\\n activity = self.testapp.get('/activities/%s' % activity_id, '', oauth2Header(username), status=200)\\n\\n self.assertEqual(activity.json['likes'][0]['username'], username)\\n self.assertEqual(activity.json['liked'], True)\\n self.assertEqual(activity.json['likesCount'], 1)\",\n \"def test_like_unlike_works_without_error(self):\\n create_response = self.client.post(reverse('posts:comment_create'),\\n data={\\n 'post': self.post.id,\\n 'user': self.user.id,\\n 'comment_description':\\n 'This is a test_comment'\\n }, format='json')\\n self.assertEqual(create_response.status_code, status.HTTP_201_CREATED)\\n # initial stars_count is zero, so when liked must increase by 1\\n like_response = self.like_unlike_comment()\\n self.assertEqual(like_response.status_code, status.HTTP_200_OK)\\n self.assertEqual(like_response.data['stars_count'], 1)\\n like_again = self.like_unlike_comment()\\n self.assertEqual(like_again.status_code, status.HTTP_200_OK)\\n self.assertEqual(like_again.data['stars_count'], 2)\\n # current likes/stars count is 2 so if unliked should become 1\\n unlike_comment_response = self.like_unlike_comment(pk='1',\\n action='unlike')\\n self.assertEqual(unlike_comment_response.status_code,\\n status.HTTP_200_OK)\\n self.assertEqual(unlike_comment_response.data['stars_count'], 1)\\n # current likes/stars count is 1 so if unliked should become 0\\n unlike_comment_response = self.like_unlike_comment(pk='1',\\n action='unlike')\\n self.assertEqual(unlike_comment_response.status_code,\\n status.HTTP_200_OK)\\n self.assertEqual(unlike_comment_response.data['stars_count'], 0)\\n # current likes/stars count is 0 so if unliked should remain 0\\n unlike_comment_response = self.like_unlike_comment(pk='1',\\n action='unlike')\\n self.assertEqual(unlike_comment_response.status_code,\\n status.HTTP_200_OK)\\n self.assertEqual(unlike_comment_response.data['stars_count'], 0)\",\n \"def user_response_to_post(self, request, pk):\\n post_objects_count = Post.objects.filter(id=pk, liked_users__id=request.user.id).count()\\n post_objects = Post.objects.get(id=pk)\\n if post_objects_count !=0:\\n post_objects.liked_users.remove(request.user)\\n response_msg = \\\"You disliked the post\\\"\\n else:\\n post_objects.liked_users.add(request.user)\\n response_msg = \\\"You have liked the post\\\"\\n return Response({'data': response_msg}, status=status.HTTP_200_OK)\",\n \"def total_likes(self):\\n\\n return (self.likes.count()-self.dislikes.count())\",\n \"def test_unlike(self):\\n\\n message = Message(text=\\\"hello world\\\", user_id=self.testuser2.id)\\n \\n db.session.add(message)\\n db.session.commit()\\n\\n like = Likes(message_id=message.id, user_id=self.testuser.id)\\n \\n db.session.add(like)\\n db.session.commit()\\n \\n message_id = message.id\\n\\n with self.client as c:\\n with c.session_transaction() as session:\\n session[CURR_USER_KEY] = self.testuser.id\\n \\n response = c.post(f\\\"/messages/{message_id}/like\\\", follow_redirects=True)\\n \\n self.assertEqual(response.status_code, 200)\\n\\n likes = Likes.query.filter(Likes.message_id==message_id).all()\\n self.assertEqual(len(likes), 0)\",\n \"def dislike(self, event, number):\\n\\n if not hasattr(self, 'user_id') or not hasattr(self, 'top_id'):\\n vk_common.dprint(2, vk_common.func_name(), f\\\"Не заданы атрибуты user_id = или top_id.\\\")\\n return\\n\\n item_id = self.item_ids[self.top_id - 1]\\n if item_id is None:\\n vk_common.dprint(2, vk_common.func_name(), f\\\"Атрибут item_id = None.\\\")\\n return\\n else:\\n vk_common.dprint(2, vk_common.func_name(), f\\\"number = {number}; item_id = {item_id}\\\")\\n\\n vk_common.dprint(2, vk_common.func_name(), f\\\"number = {number}; item_id = {item_id}\\\")\\n vk_common.dprint(1, f\\\"имя файла с фото = \\\\\\\"{self.user_id}_{self.top_id}\\\\\\\"; \\\"\\n f\\\"ссылка = https://vk.com/id{self.user_id}\\\")\\n\\n err, msg, likes = vk_common.dislike_photo(self.user_id, item_id)\\n if err != 0:\\n mb.showinfo(\\\"Информация\\\", f\\\"Не смогли убрать лайк у фото пользователя с id = {self.user_id}.\\\")\\n vk_common.dprint(2, \\\"Не смогли убрать лайк.\\\")\\n return\\n mb.showinfo(\\\"Информация\\\", f\\\"Убрали лайк у фото пользователя с id = {self.user_id}.\\\")\\n vk_common.dprint(2, f\\\"Убрали лайк: owner_id = {self.user_id}; item_id = {item_id}\\\")\",\n \"def add_unsaved_likes_to_user(sender, user, request, **kwargs):\\n session_likes = request.session.get('likes')\\n if session_likes:\\n user.userprofile.liked_products.add(*session_likes)\",\n \"def like_unlike_comment(self, pk='1', action='like'):\\n return self.client.post(reverse('posts:like_unlike_comment',\\n kwargs={\\n 'pk': pk,\\n 'action': action\\n }))\",\n \"def remove_like(obj, user):\\n\\tobj_type = ContentType.objects.get_for_model(obj)\\n\\tLike.objects.filter(\\n\\t\\tcontent_type = obj_type,\\n\\t\\tobject_id = obj.id,\\n\\t\\tuser = user).delete()\",\n \"def unlike_comment_API(id):\\n\\n count = float(-1)\\n redis_obj.zincrby('add_like', id, count)\\n\\n \\\"\\\"\\\" Invalidating cache \\\"\\\"\\\"\\n redis_obj.delete('leaderboard-key')\\n return jsonify(username=g.user[1], Message=\\\"Success! Message unliked successfully. \\\", Status_code=status.HTTP_200_OK)\",\n \"def count_likes(self):\\n likes = self.event_likes\\n num_likes = len(likes)\\n return num_likes\",\n \"def unlike(db, filename, user):\\n cur = db.cursor()\\n if like_exists(db, filename, user):\\n sql = \\\"\\\"\\\"\\n delete from likes where filename=? and usernick=?;\\n \\\"\\\"\\\"\\n cur.execute(sql, (filename, user))\\n db.commit()\",\n \"def test_unlike_mission(self):\\n\\n self.mission.likes = {0, self.user2.id}\\n db.session.commit()\\n mission = Mission.query.get(self.mission.id)\\n # make sure like is there in mission.likes.\\n self.assertEqual(mission.likes, {0, self.user2.id})\\n\\n with self.client as c:\\n with c.session_transaction() as sess:\\n sess['user_id'] = self.user2.id\\n\\n resp = c.post(f'/v1/mission/like{self.mission.id}')\\n\\n self.assertEqual(resp.json['success'], 'removed')\\n mission = Mission.query.get(self.mission.id)\\n self.assertEqual(mission.likes, {0})\",\n \"def unblock_profile(self, request, *args, **kwargs):\\n context = {\\n 'conversation': self.get_object(),\\n 'request': request\\n }\\n serializer = UnblockProfileSerializer(data=request.data, context=context)\\n serializer.is_valid(raise_exception=True)\\n serializer.save()\\n return Response(serializer.data, status=status.HTTP_202_ACCEPTED)\",\n \"def unseen_count_for(self, user):\\r\\n return self.filter(user=user, unseen=True).count()\",\n \"def test_unlike_activity(self):\\n from .mockers import user_status_context\\n from .mockers import subscribe_context, create_context\\n username = 'messi'\\n username_not_me = 'xavi'\\n self.create_user(username)\\n self.create_user(username_not_me)\\n self.create_context(create_context)\\n self.admin_subscribe_user_to_context(username, subscribe_context)\\n self.admin_subscribe_user_to_context(username_not_me, subscribe_context)\\n res = self.create_activity(username, user_status_context)\\n activity_id = res.json['id']\\n self.testapp.post('/activities/%s/likes' % activity_id, '', oauth2Header(username_not_me), status=201)\\n self.testapp.delete('/activities/%s/likes/%s' % (activity_id, username_not_me), '', oauth2Header(username_not_me), status=204)\\n activity = self.testapp.get('/activities/%s' % activity_id, '', oauth2Header(username), status=200)\\n\\n self.assertEqual(activity.json['likes'], [])\\n self.assertEqual(activity.json['liked'], False)\\n self.assertEqual(activity.json['likesCount'], 0)\",\n \"def dislike(request, pk):\\n update_posts_expiration()\\n post = 0;\\n post = Post.objects.get(id=pk)\\n if post.is_expired == True:\\n return Response(\\\"You can't interact with a expired post\\\")\\n else:\\n if request.method == \\\"POST\\\":\\n if request.user != post.user:\\n post.dislikes_count += 1\\n post.save()\\n serializer = DislikeSerializer(data=request.data)\\n if serializer.is_valid():\\n serializer.save(user=request.user, disliked_post=post)\\n return Response(\\\"you disliked the post with title: \\\" + post.title)\\n else:\\n return Response(serializer.errors)\\n else:\\n return Response(\\\"you can't dislike your own post\\\")\\n \\n\\n return Response(\\\"you didn't dislike the post yet\\\")\",\n \"def backwards(self, orm):\\n \\n user_type = orm['auth.User']\\n user = user_type.objects.get(username=self.suUsername)\\n \\n profile_type = orm['mooi.Profile']\\n userProfile = profile_type.objects.get(user=user)\\n \\n userProfile.delete()\\n user.delete()\",\n \"def delete(self, request, username):\\n\\n # Retrieve the user from the user table if the user exists\\n try:\\n user_to_unfollow = User.objects.get(username=username)\\n current_user = request.user\\n # If a user is trying to unfollow themselves then stop the request\\n if user_to_unfollow.profile.id == current_user.profile.id:\\n return Response(\\n {\\\"errors\\\": FOLLOW_USER_MSGS['CANNOT_UNFOLLOW_SELF']},\\n status=status.HTTP_400_BAD_REQUEST\\n )\\n # Check if the user to be unfollowed\\n # is in the current users following list\\n try:\\n profile_id = user_to_unfollow.profile.id\\n user_being_followed = CustomFollows.objects.get(\\n to_profile_id=profile_id,\\n from_profile_id=current_user.profile.id\\n )\\n # If not tell the user the request can't happen\\n # Because they don't follow the user\\n except Exception as e:\\n return Response(\\n {\\n \\\"errors\\\": FOLLOW_USER_MSGS['USER_UNFOLLOWED_ALREADY']\\n },\\n status=status.HTTP_400_BAD_REQUEST\\n )\\n # Otherwise unfollow the user as requested\\n current_user.profile.follows.remove(user_to_unfollow.profile)\\n # Get the following & followers username list\\n # And the following & followers count for the current user\\n user_following_data = get_user_following_data(current_user)\\n return Response(\\n {\\n \\\"message\\\": FOLLOW_USER_MSGS['USER_UNFOLLOW_SUCCESSFUL'],\\n \\\"following\\\": user_following_data[\\\"following\\\"],\\n \\\"followers\\\": user_following_data[\\\"followers\\\"],\\n \\\"followingCount\\\": user_following_data[\\\"followingCount\\\"],\\n \\\"followersCount\\\": user_following_data[\\\"followersCount\\\"]\\n },\\n status=status.HTTP_200_OK\\n )\\n # End request if we cannot find the user we want to unfollow.\\n except User.DoesNotExist:\\n return Response(\\n {\\\"errors\\\": FOLLOW_USER_MSGS['USER_NOT_FOUND']},\\n status=status.HTTP_404_NOT_FOUND\\n )\",\n \"def test_downvote_then_upvote_same_user_leaves_comment_score_one_greater(self):\\n comment = Comment.objects.get(body=\\\"987XYZ\\\")\\n # self.assertEqual(len(post_qs), 1)\\n self.assertEqual(comment.score, DEFAULT_SCORE)\\n comment = Comment.objects.get(body=\\\"987XYZ\\\")\\n\\n vote1 = Vote.create(comment=comment, value=-1, voter=self.user)\\n comment = Comment.objects.get(body=\\\"987XYZ\\\")\\n self.assertEqual(comment.score, DEFAULT_SCORE - 1)\\n\\n vote2 = Vote.create(comment=comment, value=1, voter=self.user)\\n comment = Comment.objects.get(body=\\\"987XYZ\\\")\\n self.assertEqual(comment.score, DEFAULT_SCORE + 1)\",\n \"def user_unfollow():\\n data = request.get_json(force=True)\\n follower = User.query.get(data['follower'])\\n following = User.query.get(data['following'])\\n follower.followcheck.remove(following)\\n db.session.commit()\\n return {'unfollowed': True}\",\n \"def count_downvotes(self):\\n return self.filter(value=-1).count()\",\n \"def addLikeCount(self,count):\\n self.interactionCount += count\\n return None\",\n \"def test_removal_from_public_index(self):\\n before = (S(UserProfile)\\n .indexes(UserProfile.get_index(public_index=True))\\n .count())\\n profile = self.mozillian2.userprofile\\n profile.privacy_full_name = MOZILLIANS\\n profile.save()\\n sleep(1)\\n after = (S(UserProfile)\\n .indexes(UserProfile.get_index(public_index=True))\\n .count())\\n eq_(after+1, before)\",\n \"def decrement_count(self, state):\\n if state == domain.REVIEW_STATE_ASSIGNED:\\n self.assigned_count -= 1\\n elif state == domain.REVIEW_STATE_COMPLETED:\\n self.completed_count -= 1\\n elif state == domain.REVIEW_STATE_EXPIRED:\\n self.expired_count -= 1\\n else:\\n raise ValueError('%s not in %s' % (state, domain.REVIEW_STATES))\",\n \"def relation_downvote(request, pk):\\n try:\\n relation = Relation.objects.get(pk=pk)\\n except Relation.DoesNotExist:\\n return Response(status=status.HTTP_404_NOT_FOUND)\\n\\n if request.method == 'PUT':\\n relation.negative_reaction_count += 1\\n relation.save()\\n serializer = RelationSerializer(relation)\\n return Response(serializer.data)\",\n \"def remove_count(self, denom: CashDenomination, count: int) -> None:\\n if not self.__open:\\n raise RuntimeError(\\\"Cash drawer must be open to modify.\\\")\\n if count < 0:\\n raise ValueError(\\\"Count must not be negative.\\\")\\n if count > self.__contents[denom]:\\n raise ValueError(\\\"Cannot remove more than are present.\\\")\\n self.__contents[denom] -= count\",\n \"def latest_likes(self, user, number_posts, likes):\\n WAIT = 1\\n if likes:\\n action = 'Like'\\n else:\\n action = 'Unlike'\\n self.nav_user(user)\\n image_container = []\\n image_container.extend(self.driver.find_elements_by_class_name('_9AhH0'))\\n for image in image_container[:number_posts]:\\n image.click()\\n time.sleep(WAIT)\\n try:\\n self.driver.find_element_by_xpath(\\\"//*[@aria-label='{}']\\\".format(action).click())\\n except Exception as e:\\n print(e)\\n self.driver.find_elements_by_xpath('/html/body/div[4]/div[2]/div/article/div[3]/section[1]/span[1]/button')[0].click() # clicks the heart symbol\\n time.sleep(WAIT)\\n self.driver.find_elements_by_xpath('/html/body/div[4]/div[3]/button')[0].click() #Makes sure to close out of current picture\\n time.sleep(WAIT)\\n \\n # Tested\\n users_list = []\\n def get_likes_list(self, username):\\n \\\"\\\"\\\"\\n Method gets a list of users who like a post\\n\\n \\\"\\\"\\\"\\n api = self.api\\n api.searchUsername(username) \\n result = api.LastJson\\n username_id = result['user']['pk'] #Gets the user ID\\n user_posts = api.getUserFeed(username_id) # gets the user feed\\n result = api.LastJson\\n media_id = result['items'][0]['id'] #gets the most recent post\\n api.getMediaLikers(media_id) #gets users who liked\\n users = api.LastJson('users')\\n for user in users: #appends the users to the list\\n users_list.append({'pk':user['pk'], 'username':user['username']})\",\n \"def unfollow_me(self):\\n return self.follow_me('unfollow_topic')\",\n \"def dislike_post(request, pk):\\r\\n post = get_object_or_404(Post, id=request.POST.get('id'))\\r\\n disliked = False\\r\\n if post.dislikes.filter(id=request.user.id).exists():\\r\\n post.dislikes.remove(request.user)\\r\\n disliked = False\\r\\n else:\\r\\n if post.likes.filter(id=request.user.id).exists():\\r\\n post.likes.remove(request.user)\\r\\n post.dislikes.add(request.user)\\r\\n disliked = True\\r\\n\\r\\n if request.is_ajax():\\r\\n return JsonResponse({'total_dislikes': post.dislikes.count(),\\r\\n 'total_likes': post.likes.count(),\\r\\n 'disliked': disliked})\",\n \"def test_upvote_then_downvote_same_user_leaves_comment_score_one_less(self):\\n comment = Comment.objects.get(body=\\\"987XYZ\\\")\\n # self.assertEqual(len(post_qs), 1)\\n self.assertEqual(comment.score, DEFAULT_SCORE)\\n comment = Comment.objects.get(body=\\\"987XYZ\\\")\\n\\n vote1 = Vote.create(comment=comment, value=1, voter=self.user)\\n comment = Comment.objects.get(body=\\\"987XYZ\\\")\\n self.assertEqual(comment.score, DEFAULT_SCORE + 1)\\n\\n vote2 = Vote.create(comment=comment, value=-1, voter=self.user)\\n comment = Comment.objects.get(body=\\\"987XYZ\\\")\\n self.assertEqual(comment.score, DEFAULT_SCORE - 1)\",\n \"def unlike(id):\\n spinner = Halo(text=\\\"Unliking photo...\\\", spinner=\\\"dots\\\").start()\\n try:\\n api.unlike(id)\\n spinner.succeed(\\\"Unliked\\\")\\n except Exception:\\n spinner.fail(\\\"Failed to unlike photo.\\\")\",\n \"def _decrement_num_user_stars(user_id, match_id, now):\\n\\tnum_user_stars = session.query(CalendarEntry.num_user_stars)\\\\\\n\\t\\t\\t.filter(\\n\\t\\t\\t\\tCalendarEntry.user_id == user_id,\\n\\t\\t\\t\\tCalendarEntry.match_id == match_id)\\\\\\n\\t\\t\\t.one()\\\\\\n\\t\\t\\t.num_user_stars\\n\\tif num_user_stars > 1:\\n\\t\\t# Decrement the count of stars for the CalendarEntry to a positive integer.\\n\\t\\tsession.execute(CalendarEntries.update()\\n\\t\\t\\t\\t.where(sa.and_(\\n\\t\\t\\t\\t\\tCalendarEntry.user_id == user_id,\\n\\t\\t\\t\\t\\tCalendarEntry.match_id == match_id))\\n\\t\\t\\t\\t.values({CalendarEntry.num_user_stars: num_user_stars - 1}))\\n\\telse:\\n\\t\\t# Delete the CalendarEntry becuase the count of stars is now zero.\\n\\t\\tsession.execute(CalendarEntries.delete().where(sa.and_(\\n\\t\\t\\t\\tCalendarEntry.user_id == user_id,\\n\\t\\t\\t\\tCalendarEntry.match_id == match_id)))\",\n \"def unsubscribed_member_count(self, unsubscribed_member_count):\\n\\n self._unsubscribed_member_count = unsubscribed_member_count\",\n \"def unfollow_friend(username):\\n\\n if not g.user:\\n print \\\"401\\\"\\n abort(401)\\n whom_id = get_user_id(username)\\n print whom_id\\n if whom_id is None:\\n abort(404)\\n unfollow_query(whom_id)\\n flash('You are no longer following \\\"%s\\\"' % username)\\n name = {'name of unfollowing user': username}\\n ############### REDIS cache invalidate #####################\\n R_SERVER.delete(user_timeline_key)\\n return jsonify(Username=name, Status_code=status.HTTP_200_OK)\",\n \"def _decrease_counter(self, response):\\n response_id = response.meta['__id']\\n spot = self._request_registry[response_id]\\n spot['counter'] = spot.get('counter', 0) - 1\",\n \"def toggle_like(post_id):\\n if CURRENT_USER_KEY not in session:\\n raise Unauthorized()\\n\\n liked_post = Post.query.get_or_404(post_id)\\n user = User.query.get(session[CURRENT_USER_KEY])\\n # toggle the like by removing from user likes\\n if liked_post in user.likes:\\n user.likes.remove(liked_post)\\n else:\\n user.likes.append(liked_post)\\n\\n db.session.commit()\\n\\n return redirect(url_for('show_likes', user_id=user.id))\",\n \"def like(request, post_id):\\n if request.method == \\\"PUT\\\":\\n liked = None\\n user = request.user\\n post = Post.objects.get(id=post_id)\\n # If user already liked, decrement the like count and remove as 'liker'\\n if user in post.liked_by.all():\\n post.liked_by.remove(user)\\n post.likes -= 1\\n post.save()\\n liked = False\\n # Else increase like count and add user\\n else:\\n post.liked_by.add(user)\\n post.likes += 1\\n post.save()\\n liked = True\\n # Return data for updating dynamically with javascript\\n return JsonResponse({\\\"total_likes\\\": post.likes, \\\"liked\\\": liked})\",\n \"def decrement(self, stats, sample_rate=1):\\n self.update_stats(stats, -1, sample_rate=sample_rate)\",\n \"def delete(self, *args, **kwargs):\\n self.item.comments_total -= 1\\n self.item.save()\\n super(Comment, self).delete(*args, **kwargs)\",\n \"def decrease(self):\\n self.score -= self.score\",\n \"def meal_disliked(request, meal_pk):\\n meal_pk = int(meal_pk)\\n meal = Meal.objects.get(pk=meal_pk)\\n like = False\\n member = request.user\\n\\n try:\\n rating = Rating.objects.get(member=member, meal=meal)\\n r_percent = meal.percent()\\n except ObjectDoesNotExist:\\n Rating.objects.create_rating(member, meal, like)\\n r_percent = meal.percent()\\n\\n rating.like = like\\n rating.save()\\n r_percent = meal.percent()\\n return HttpResponse(r_percent)\",\n \"def dislike_comment(request, pk, pk1):\\r\\n comment = get_object_or_404(Comment, id=request.POST.get('id'))\\r\\n disliked = False\\r\\n if comment.dislikes.filter(id=request.user.id).exists():\\r\\n comment.dislikes.remove(request.user)\\r\\n disliked = False\\r\\n else:\\r\\n comment.dislikes.add(request.user)\\r\\n disliked = True\\r\\n\\r\\n context = {\\r\\n 'is_liked': disliked,\\r\\n 'total_dislikes': comment.dislikes.count()\\r\\n }\\r\\n\\r\\n if request.method == 'POST':\\r\\n return HttpResponse(\\r\\n json.dumps(context),\\r\\n content_type=\\\"application/json\\\")\",\n \"def delete_like(self, entry, **args): \\n args.update(entry=entry)\\n return self.fetch(\\\"/like/delete\\\", post_args=args)\",\n \"def update_count(self):\\n pass # Do nothing\",\n \"def unfollow(self, other):\\n\\t\\tif self.follows(other):\\n\\t\\t\\tself.followed.remove(other)\",\n \"def get_likes_count():\\n return Flag.objects.filter(flag=Flag.LIKE_FLAG).count()\",\n \"def toggle_vote(self):\\n\\n self.vote = 1 - self.vote\",\n \"def test_unlike_comment(comment, mocker):\\n unlike_mock = mocker.patch(\\\"instapi.client.client.comment_unlike\\\")\\n\\n comment.unlike()\\n\\n unlike_mock.assert_called_once_with(comment.pk)\",\n \"def reset_count(self):\\n self.count = 0\",\n \"def update_count(self):\\n pass\",\n \"def decrement_misses_remaining(self):\\n self.misses_remaining -=1\",\n \"def order_posts_to_likes(sender, instance, created=False, **kwargs):\\n user = instance.user\\n post = instance.post\\n tags = post.tags.all()\\n\\n likes = Like.objects.filter(user=user, post=post)\\n liked_posts = [ i.post for i in likes if i.liked_status==True]\\n\\n for post in liked_posts:\\n liked_tags = post.tags.all()\\n for tag in liked_tags:\\n if not tag in tags:\\n tags.append(tag)\\n\\n similar_posts = Post.objects.filter(tags__in=tags).distinct()\\n try:\\n UserPostWeight.objects.filter(user=user).delete()\\n except:\\n pass\\n weight = 2\\n for post in similar_posts:\\n UserPostWeight.objects.create(user=user, post=post, weight=weight)\\n weight+=2\\n \\n\\n\\n pass\",\n \"def is_disliked_by(self, user):\\n return user.disliked_articles.filter(pk=self.pk).exists()\",\n \"def downvote(self):\\n url = \\\"https://api.imgur.com/3/gallery/{0}/vote/down\\\".format(self.id)\\n return self._imgur._send_request(url, needs_auth=True, method='POST')\",\n \"def unfollow(self, user_index, following_index):\\n if user_index >= self.num_users or following_index >= self.num_users:\\n raise ValueError(\\n f\\\"Number of users is {self.num_users}, but indices \\\"\\n f\\\"{user_index} and {following_index} were requested.\\\"\\n )\\n if self.users_hat[following_index, user_index] == 1:\\n self.users_hat[following_index, user_index] = 0\\n elif self.is_verbose():\\n self.log(f\\\"User {following_index} was not following user {user_index}\\\")\",\n \"def delete_model(self, request, obj):\\n obj.post.likeNumDreacase()\\n obj.delete()\",\n \"async def twitter_unfollow(self, ctx, handle):\\n sane_handle = handle.lower().lstrip('@')\\n conf = dutils.get(self.conf.follows, screen_name=sane_handle)\\n chan_conf = dutils.get(conf.discord_channels, id=ctx.message.channel.id) if conf is not None else None\\n\\n if chan_conf is None:\\n raise TwitterError('Not following {} on this channel.'.format(handle))\\n\\n # Remove the Discord channel from the Twitter channel conf\\n conf.discord_channels.remove(chan_conf)\\n if not conf.discord_channels:\\n # If there are no more Discord channel to feed, unfollow the Twitter channel\\n self.conf.follows.remove(conf)\\n del conf\\n\\n # Update the tweepy stream\\n if len(self.conf.follows) > 0:\\n await self.stream.start()\\n else:\\n self.stream.stop()\\n\\n self.conf.save()\\n\\n await self.bot.say('\\\\N{OK HAND SIGN}')\",\n \"def unfollow_user(username):\\n user_ID = before_request()\\n user_ID = None\\n if user_ID != None:\\n user_ID = str(g.user['_id'])\\n if not g.user:\\n abort(401)\\n whom_id = get_user_id(username)\\n if whom_id is None:\\n abort(404)\\n mongo.db.users.update({'_id': g.user['_id']}, {\\n '$pull': {'follows': whom_id}})\\n flash('You are no longer following \\\"%s\\\"' % username)\\n if redis_obj.get(user_ID):\\n return redirect(url_for('user_timeline', username=username, userId=pickle.loads(redis_obj.get(user_ID))))\\n else:\\n redis_obj.delete(session['user_id'])\\n print \\\"Invalidating cache after Unfollow\\\"\\n return redirect(url_for('user_timeline', username=username))\",\n \"def unfollow_user(username):\\n if not g.user:\\n abort(401)\\n whom_id = get_user_id(username)\\n if whom_id is None:\\n abort(404)\\n db = get_db()\\n db.execute('delete from follower where who_id=? and whom_id=?',\\n [session['user_id'], whom_id])\\n db.commit()\\n flash('You are no longer following \\\"%s\\\"' % username)\\n return redirect(url_for('user_timeline', username=username))\",\n \"def delete(self, request, *args, **kwargs):\\n\\n following_username = token_payload(request)\\n\\n followed_username = self.kwargs.get(self.look_url_kwarg)\\n try:\\n followed_user = User.objects.get(username=followed_username)\\n followed_user_id = followed_user.id\\n\\n is_following = FollowUser.objects.filter(\\n following_username=following_username)\\n for user in is_following:\\n if user.following_username == following_username and user.followed_username == followed_username:\\n returned_user = FollowUser.objects.get(pk=user.id)\\n returned_user.delete()\\n\\n followed_user_profile = Profile.objects.get(\\n profile_user=followed_user_id)\\n serializer_profile = self.serializer_classprofile(\\n followed_user_profile)\\n return Response({\\n \\\"user\\\": following_username,\\n \\\"unfollowing\\\": serializer_profile.data\\n })\\n return Response({\\n \\\"message\\\": 'User has been unfollowed or You are unfollowing a user you were not orignally following'\\n }, status=status.HTTP_400_BAD_REQUEST)\\n except:\\n return Response({\\n \\\"message\\\": 'User does not exist'\\n }, status=status.HTTP_400_BAD_REQUEST)\",\n \"def get_number_of_likes_for_user(self, user):\\n\\t\\tfrom pins.models import Pin\\n\\t\\tpin_ctype = ContentType.objects.get_for_model(Pin)\\n\\t\\tpin_list = Pin.objects.active_pins().filter(board__user=user).values_list('pk', flat=True)\\n\\t\\treturn self.filter(content_type=pin_ctype, object_id__in=pin_list).count()\",\n \"def like(self, data_base, user):\\n cursor = data_base.cursor()\\n cursor.execute(f\\\"UPDATE post SET likes = likes + 1 WHERE id = '{self.id}'\\\") # Increments the likes\\n cursor.execute(f\\\"INSERT INTO user_like (user_id, post_id) VALUES ({user.id}, {self.id})\\\")\\n if self.commit_to_db:\\n data_base.commit()\\n cursor.close()\",\n \"def get_likes(self, obj):\\n return QuestionPersonLike.objects.filter(question=obj,\\n like=True).count()\",\n \"def test_unpopular(self):\\n self.assertFalse(self.user3.is_popular())\\n self.user3.receive_upvotes(randint(101, 10000))\\n self.assertTrue(self.user3.is_popular())\",\n \"def delete_profile_pic(sender, instance, **kwargs):\\n if instance.profile_picture:\\n if instance.profile_picture.name != \\\"default.png\\\":\\n path = instance.profile_picture.path\\n os.remove(path)\",\n \"def test_downvote_then_upvote_same_user_leaves_post_score_one_greater(self):\\n post = Post.objects.get(body=\\\"123ABC Body\\\")\\n # self.assertEqual(len(post_qs), 1)\\n self.assertEqual(post.score, DEFAULT_SCORE)\\n post = Post.objects.get(body=\\\"123ABC Body\\\")\\n\\n vote1 = Vote.create(post=post, value=-1, voter=self.user)\\n post = Post.objects.get(body=\\\"123ABC Body\\\")\\n self.assertEqual(post.score, DEFAULT_SCORE - 1)\\n\\n vote2 = Vote.create(post=post, value=1, voter=self.user)\\n post = Post.objects.get(body=\\\"123ABC Body\\\")\\n self.assertEqual(post.score, DEFAULT_SCORE + 1)\",\n \"def lose_life(self):\\n self.__num_lives -= 1\",\n \"def test_update_like_cant_message(self):\\n\\n self.assertEqual(first=0, second=self.like.likes.all().count())\\n url = reverse('like-detail', args=(self.like.id,))\\n data = {\\n 'likes': [2]\\n }\\n json_data = json.dumps(data)\\n self.client.credentials(HTTP_AUTHORIZATION=self.token)\\n response = self.client.patch(path=url, content_type='application/json', data=json_data)\\n self.assertFalse(response.data.get('can_message', False))\\n self.assertEqual(first=200, second=response.status_code)\\n self.assertEqual(first=1, second=self.like.likes.all().count())\",\n \"def post_likes(request):\\n if request.method == 'POST':\\n if not request.user.is_authenticated:\\n raise PermissionDenied\\n pk = request.POST.get('post_id')\\n try:\\n post = PostsModel.objects.get(id=pk)\\n except PostsModel.DoesNotExist:\\n post = None\\n try:\\n likes = LikesModel.objects.get(user=request.user, post=post)\\n except LikesModel.DoesNotExist:\\n likes = None\\n\\n\\n if likes == None:\\n likes = LikesModel.objects.create(user=request.user, post=post)\\n post.likes_count = F('likes_count') + 1\\n post.save(update_fields=[\\\"likes_count\\\"])\\n class_likes = 'fa fa-heart mx-2'\\n else:\\n if likes.is_liked == True:\\n likes.is_liked = False\\n likes.save()\\n post.likes_count = F('likes_count') - 1\\n post.save(update_fields=[\\\"likes_count\\\"])\\n class_likes = 'fa fa-heart-o mx-2'\\n elif likes.is_liked == False:\\n likes.is_liked = True\\n likes.save()\\n post.likes_count = F('likes_count') + 1\\n post.save(update_fields=[\\\"likes_count\\\"])\\n class_likes = 'fa fa-heart mx-2'\\n post.refresh_from_db()\\n response_data = {'_code' : 0, '_status' : 'ok', '_likes': post.likes_count, '_class_likes': class_likes}\\n else:\\n response_data = {'_code' : 1, '_status' : 'no' }\\n\\n return JsonResponse(response_data)\",\n \"def remove_vote_weight(var, target : users.User, amount : int = 1) -> None:\\n add_vote_weight(var, target, -amount)\",\n \"def test_downvote_modifies_comment_score(self):\\n comment = Comment.objects.get(body=\\\"987XYZ\\\")\\n self.assertEqual(comment.score, DEFAULT_SCORE)\\n vote = Vote.create(comment=comment, value=-1, voter=self.user)\\n comment = Comment.objects.get(body=\\\"987XYZ\\\")\\n self.assertEqual(comment.score, DEFAULT_SCORE - 1)\",\n \"def perform_destroy(self, instance):\\n self.object.comments -= 1\\n self.object.save()\\n instance.delete()\",\n \"def unfollow(self, followerId, followeeId):\\r\\n self.follows[followerId].discard(followeeId)\",\n \"def get_user_has_liked(self, instance):\\n request = self.context.get(\\\"request\\\")\\n return instance.likes.filter(pk=request.user.pk).exists()\",\n \"def get_user_has_liked(self, instance):\\n request = self.context.get(\\\"request\\\")\\n return instance.likes.filter(pk=request.user.pk).exists()\"\n]","isConversational":false},"negative_scores":{"kind":"list like","value":["0.74382937","0.68135536","0.62009215","0.61690074","0.6078528","0.5979719","0.5800343","0.5772288","0.5705149","0.56960106","0.5675965","0.56723356","0.5662375","0.56615674","0.5642525","0.56078047","0.56078047","0.56078047","0.5589605","0.5575719","0.5571602","0.5571017","0.55612946","0.54670876","0.5451424","0.54350483","0.5424731","0.5379507","0.5375072","0.5276599","0.52553254","0.52311957","0.52061963","0.5196879","0.5183039","0.51808625","0.5132592","0.5109533","0.5108023","0.51018405","0.5080324","0.50657994","0.5018532","0.4992783","0.49864173","0.49556902","0.49372804","0.49359408","0.493583","0.49335515","0.4912882","0.49108195","0.49061006","0.49044812","0.48986524","0.4896988","0.48859552","0.48790115","0.48638326","0.48529992","0.48511755","0.4839292","0.48349914","0.48311698","0.4820532","0.48131287","0.48115182","0.48082355","0.47863573","0.4782388","0.47637552","0.4757786","0.47543883","0.4749852","0.47441941","0.47339696","0.47332487","0.47253078","0.47243497","0.4719879","0.47176155","0.46982047","0.4684297","0.46802357","0.46756968","0.46754527","0.4664367","0.46623802","0.46530935","0.46449846","0.46339247","0.46249196","0.46235156","0.46227404","0.46211964","0.46166578","0.4612197","0.46099323","0.46074864","0.46074864"],"string":"[\n \"0.74382937\",\n \"0.68135536\",\n \"0.62009215\",\n \"0.61690074\",\n \"0.6078528\",\n \"0.5979719\",\n \"0.5800343\",\n \"0.5772288\",\n \"0.5705149\",\n \"0.56960106\",\n \"0.5675965\",\n \"0.56723356\",\n \"0.5662375\",\n \"0.56615674\",\n \"0.5642525\",\n \"0.56078047\",\n \"0.56078047\",\n \"0.56078047\",\n \"0.5589605\",\n \"0.5575719\",\n \"0.5571602\",\n \"0.5571017\",\n \"0.55612946\",\n \"0.54670876\",\n \"0.5451424\",\n \"0.54350483\",\n \"0.5424731\",\n \"0.5379507\",\n \"0.5375072\",\n \"0.5276599\",\n \"0.52553254\",\n \"0.52311957\",\n \"0.52061963\",\n \"0.5196879\",\n \"0.5183039\",\n \"0.51808625\",\n \"0.5132592\",\n \"0.5109533\",\n \"0.5108023\",\n \"0.51018405\",\n \"0.5080324\",\n \"0.50657994\",\n \"0.5018532\",\n \"0.4992783\",\n \"0.49864173\",\n \"0.49556902\",\n \"0.49372804\",\n \"0.49359408\",\n \"0.493583\",\n \"0.49335515\",\n \"0.4912882\",\n \"0.49108195\",\n \"0.49061006\",\n \"0.49044812\",\n \"0.48986524\",\n \"0.4896988\",\n \"0.48859552\",\n \"0.48790115\",\n \"0.48638326\",\n \"0.48529992\",\n \"0.48511755\",\n \"0.4839292\",\n \"0.48349914\",\n \"0.48311698\",\n \"0.4820532\",\n \"0.48131287\",\n \"0.48115182\",\n \"0.48082355\",\n \"0.47863573\",\n \"0.4782388\",\n \"0.47637552\",\n \"0.4757786\",\n \"0.47543883\",\n \"0.4749852\",\n \"0.47441941\",\n \"0.47339696\",\n \"0.47332487\",\n \"0.47253078\",\n \"0.47243497\",\n \"0.4719879\",\n \"0.47176155\",\n \"0.46982047\",\n \"0.4684297\",\n \"0.46802357\",\n \"0.46756968\",\n \"0.46754527\",\n \"0.4664367\",\n \"0.46623802\",\n \"0.46530935\",\n \"0.46449846\",\n \"0.46339247\",\n \"0.46249196\",\n \"0.46235156\",\n \"0.46227404\",\n \"0.46211964\",\n \"0.46166578\",\n \"0.4612197\",\n \"0.46099323\",\n \"0.46074864\",\n \"0.46074864\"\n]","isConversational":false},"document_score":{"kind":"string","value":"0.7310249"},"document_rank":{"kind":"string","value":"1"}}},{"rowIdx":93286,"cells":{"query":{"kind":"string","value":"Fetch social account name and avatar, and add them to the Profile."},"document":{"kind":"string","value":"def on_social_account_added(sender, instance: SocialAccount, created, **kwargs):\n if not created:\n return\n name = instance.extra_data.get('name')\n if name:\n log.debug('Updating name via socialaccount for user %i' % instance.user.id)\n instance.user.profile.name = name\n instance.user.profile.save()\n\n # Look for Social Account avatar and update the avatar\n url_avatar = instance.get_avatar_url()\n if url_avatar:\n log.debug('Updating avatar via socialaccount for user %i' % instance.user.id)\n image_content = ContentFile(requests.get(url_avatar).content)\n instance.user.profile.avatar.save(\"profile.jpg\", image_content)"},"metadata":{"kind":"string","value":"{\n \"objective\": {\n \"self\": [],\n \"paired\": [],\n \"triplet\": [\n [\n \"query\",\n \"document\",\n \"negatives\"\n ]\n ]\n }\n}"},"negatives":{"kind":"list like","value":["def get_profile(self,soup,Name,Profile_url):\n h3=soup.find(\"h3\")\n \n if h3 !=None:\n if h3.find(\"a\")!=None: \n name=h3.a.get_text()\n if h3.a.has_attr('href') :\n h3_a_tag=h3.a['href']\n h3_a_tag=h3_a_tag.replace(\"&__tn__=C-R\",'')\n \n profile_url='https://www.facebook.com'+h3_a_tag\n else :\n profile_url=\"None\"\n \n else :\n h31=soup.find('a',class_=\"actor-link\")\n if h31!=None:\n name=h31.get_text()\n if h31.has_attr('href'):\n profile_url='https://www.facebook.com'+h31['href']\n else: \n profile_url=\"None\"\n else:\n \n name=\"None\"\n profile_url=\"None\"\n \n \n Name.append(name)\n Profile_url.append(profile_url)\n \n return Name,Profile_url","def get_profile(self,fields=('id','first-name','last-name','headline','summary')):\n\n if not self._access_token:\n raise FBError(\"Authentication needed!\")\n \n token = oauth.Token(self._access_token['oauth_token'], self._access_token['oauth_token_secret'])\n client = oauth.Client(self.consumer, token)\n profile_url = self.profile_url % \",\".join(fields)\n resp, content = client.request(profile_url,headers={\"x-li-format\":'json'})\n \n if resp['status'] != '200':\n print resp\n raise FBError(\"Invalid response %s.\" % resp['status'])\n \n try:\n return json.loads(content)\n except Exception, e:\n raise FBError(\"Invalid json %s.\" % unicode(e))","def add_profile_photo():\n pass","def get_profile_image(strategy, details, response, uid, user, social, is_new=False, *args, **kwargs):\n\n # If we don't have a user then just return\n if user is None:\n return\n\n # Save photo from FB\n if kwargs['backend'].name == \"facebook\":\n try:\n image_url = \"https://graph.facebook.com/%s/picture?type=large\" % uid\n result = urllib.urlretrieve(image_url)\n\n def save_image(user, uid, result):\n user.original_photo.save(\"%s.jpg\" % uid, File(open(result[0])))\n user.save(update_fields=['original_photo'])\n\n retry_cloudfiles(save_image, user, uid, result)\n except URLError:\n pass\n elif kwargs['backend'].name == \"twitter\" and social:\n try:\n # Get profile image to save\n if response['profile_image_url'] != '':\n image_result = urllib.urlretrieve(response['profile_image_url'])\n\n def save_image(user, uid, image_result):\n user.original_photo.save(\"%s.jpg\" % uid, File(open(image_result[0])))\n user.save(update_fields=['original_photo'])\n\n retry_cloudfiles(save_image, user, uid, image_result)\n except URLError:\n pass","def profile(self) -> dict:\n endpoint = \"/api/users/profile/\"\n ret = self._request(endpoint=endpoint)\n return ret","def fetch_profile(access_token):\n\n headers = {\"Authorization\": \"Bearer \" + access_token}\n\n # Fetch athlete profile\n r = requests.get(API_URL + \"/athlete\", headers=headers)\n profile = r.json()\n if \"errors\" in profile:\n raise AuthError(profile[\"message\"])\n\n return {\n \"firstName\": profile[\"firstname\"],\n \"lastName\": profile[\"lastname\"],\n \"imgUrl\": profile[\"profile\"],\n \"profileUrl\": \"https://www.strava.com/athletes/{}\".format(profile[\"id\"]),\n }","def profile(self, name=\"johndoe\"):\r\n url = \"/account/%s\" % name\r\n return self.app.get(url, follow_redirects=True)","def get_my_profile(self):\n\n url = self.api_base_url + \"user/profile\"\n\n try:\n raw_response = self.request_handler.make_request(ApiRequestHandler.GET, url)\n except RequestFailed:\n raise\n\n jsonified_response = json.loads(raw_response.text)\n user_profile = jsonified_response\n\n return user_profile","def create(self, request, *args, **kwargs):\n response = super().create(request, *args, **kwargs)\n profile = response.data\n user_name = profile.get(\"username\")\n cache.set(f\"{USER_PROFILE_PREFIX}{user_name}\", profile)\n return response","def get_user_profile_info(token):\n user_profile_endpoint = 'https://api.spotify.com/v1/me'\n headers = {'Authorization': 'Bearer %s' % token}\n\n r = requests.get(user_profile_endpoint, headers=headers)\n if r.status_code != 200:\n return None\n profile = r.json()\n\n display_name = profile['display_name']\n profile_id = profile['id']\n email = profile['email']\n\n return User(display_name=display_name,\n profile_id=profile_id,\n email=email)","def extra_data(self, user, uid, response, details):\n try:\n return self.get_steam_profile(response)\n except:\n return \"\"","def social_associate_and_load_data(backend, details, response, uid, user,\n social_user=None, *args, **kwargs):\n extra_data = backend.extra_data(user, uid, response, details)\n created = False\n if not social_user and user:\n social_user, created = UserSocialAuth.objects.get_or_create(\n user_id=user.id,\n provider=backend.name,\n uid=uid,\n defaults={'extra_data': extra_data})\n\n if not created and extra_data and social_user.extra_data != extra_data:\n social_user.extra_data.update(extra_data)\n social_user.save()\n return {'social_user': social_user}","def build_avatar(login, request):\n log.debug(\"# Building avatar\")\n from {{package}}.models import DBSESSION\n from {{package}}.models.model import User\n avatar = DBSESSION.query(Account).filter_by(account_lid=login).first()\n request.session['user'] = avatar\n if avatar:\n return []","def user_profile(first, last, **add_info):\n profile = {}\n profile['firstname'] = first\n profile['lastname'] = last\n\n for key, value in add_info.items():\n profile[key] = value\n \n return profile","def profile_details(self, profile_name):\n url = get_url('profile details', profile=profile_name)\n response = self._get(url)\n if response.status_code == 404:\n return None\n raise_on_error(response)\n return Profile(response.json())","def post(self, request):\n profile = Profile.get_by_id(request.user.id)\n if not profile:\n return HttpResponse(status=403)\n if not imageValidator(request.FILES.get('name')):\n return HttpResponse(status=400)\n imageToUpload = request.FILES.get('name')\n key = 'avatar=' + imageToUpload.name\n url = upload(key, imageToUpload)\n profile.update(avatar=url)\n return JsonResponse(profile.to_dict(), status=200)","async def profile(self, ctx, member: discord.Member = None):\n if member is None:\n member = ctx.author\n name = member.display_name\n id = member.id\n avatar = member.avatar_url\n created = member.created_at\n embed = discord.Embed(\n title=name,\n description=name + \"'s profile!\",\n color=discord.Colour.purple()\n )\n embed.set_thumbnail(url=avatar)\n embed.add_field(name=\"ID:\", value=str(id))\n embed.add_field(name=\"Created At:\", value=str(created))\n if ctx.guild is not None:\n joined_at = member.joined_at\n embed.add_field(name=\"Joined Server At:\", value=str(joined_at))\n if member.bot:\n embed.add_field(name=\"Bot?\", value=\"Yeah\")\n if member.premium_since is not None:\n premium = str(member.premium_since)\n embed.add_field(name=\"Premium Member Since: \", value=premium)\n else:\n embed.add_field(name=\"Premium?\", value=\"Nope\")\n top_role = str(member.top_role)\n embed.add_field(name=\"Top Role:\", value=top_role)\n await ctx.send(\"\", embed=embed)","def set_image_from_socialaccount(self, socialaccount: Union[str, SocialAccount]):\n if not self.image_is_identicon():\n return\n\n if not isinstance(socialaccount, SocialAccount):\n try:\n socialaccount = SocialAccount.objects.get(\n user=self.user, provider=socialaccount\n )\n except SocialAccount.DoesNotExist:\n return\n\n url = None\n provider = socialaccount.provider\n data = socialaccount.extra_data\n if provider == \"google\":\n url = data.get(\"picture\")\n elif provider == \"github\":\n url = data.get(\"avatar_url\")\n elif provider == \"twitter\":\n url = data.get(\"profile_image_url\")\n\n if url:\n self.set_image_from_url(url)","def fetch_my_profile(self, api_token: str) -> dict:\n query = \"\"\"\n query myProfile {\n myProfile {\n id\n firstName\n lastName\n }\n }\n \"\"\"\n path = jmespath.compile(\n \"\"\"\n data.myProfile.{\n id: id\n first_name: firstName\n last_name: lastName\n }\n \"\"\"\n )\n\n data = self.do_query(query, api_token=api_token)\n\n parsed_data = path.search(data)\n self.contains_keys(parsed_data, [\"id\", \"first_name\", \"last_name\"])\n return parsed_data","def get_facebook_user_info(access_token):\n required_data_list = []\n for per in settings.FACEBOOK_EXTENDED_PERMISSIONS:\n required_data_list.append(per.replace(\"user_\",\"\"))\n \n required_data_list.append(\"picture.type(large)\")\n required_data = (\", \").join([data for data in required_data_list])\n \n graph_url = \"https://graph.facebook.com/me?access_token=%s&fields=%s\" % (access_token,required_data)\n public_info_url = \"https://graph.facebook.com/me?access_token=%s\" % access_token\n \n profile = json.load(urllib.urlopen(graph_url))\n profile_info = json.load(urllib.urlopen(public_info_url))\n \n profile_response_dict = {}\n profile_response_dict.update(profile)\n profile_response_dict.update(profile_info)\n profile_response_json = json.dumps(profile_response_dict)\n\n return (profile_response_json, profile_response_dict)","def user_profileImg(id):\n data = request.get_json(force=True)\n\n user = User.query.get(id)\n user.profileImg = data['profileImg']\n db.session.commit()\n return {'user': user.to_dict()}","def retrieve_profile(self, name):\n\n url = get_url('profile details', profile=name)\n response = self._get(url)\n raise_on_error(response)\n if response.status_code == 404:\n raise QarnotGenericException(response.json()['message'])\n return Profile(response.json())","def profile_pic(request):\n if request.user.is_authenticated:\n profile_obj = CustomUser.objects.get(id__exact=request.user.id)\n pic = profile_obj.avatar\n return {'picture': pic}\n return {}","async def github_user_info(self, ctx: commands.Context, username: str) -> None:\n async with ctx.typing():\n user_data = await self.fetch_data(f\"{GITHUB_API_URL}/users/{quote_plus(username)}\")\n\n # User_data will not have a message key if the user exists\n if \"message\" in user_data:\n embed = discord.Embed(\n title=random.choice(NEGATIVE_REPLIES),\n description=f\"The profile for `{username}` was not found.\",\n colour=Colours.soft_red\n )\n\n await ctx.send(embed=embed)\n return\n\n org_data = await self.fetch_data(user_data[\"organizations_url\"])\n orgs = [f\"[{org['login']}](https://github.com/{org['login']})\" for org in org_data]\n orgs_to_add = \" | \".join(orgs)\n\n gists = user_data[\"public_gists\"]\n\n # Forming blog link\n if user_data[\"blog\"].startswith(\"http\"): # Blog link is complete\n blog = user_data[\"blog\"]\n elif user_data[\"blog\"]: # Blog exists but the link is not complete\n blog = f\"https://{user_data['blog']}\"\n else:\n blog = \"No website link available\"\n\n embed = discord.Embed(\n title=f\"`{user_data['login']}`'s GitHub profile info\",\n description=f\"```{user_data['bio']}```\\n\" if user_data[\"bio\"] else \"\",\n colour=discord.Colour.blurple(),\n url=user_data[\"html_url\"],\n timestamp=datetime.strptime(user_data[\"created_at\"], \"%Y-%m-%dT%H:%M:%SZ\")\n )\n embed.set_thumbnail(url=user_data[\"avatar_url\"])\n embed.set_footer(text=\"Account created at\")\n\n if user_data[\"type\"] == \"User\":\n\n embed.add_field(\n name=\"Followers\",\n value=f\"[{user_data['followers']}]({user_data['html_url']}?tab=followers)\"\n )\n embed.add_field(\n name=\"Following\",\n value=f\"[{user_data['following']}]({user_data['html_url']}?tab=following)\"\n )\n\n embed.add_field(\n name=\"Public repos\",\n value=f\"[{user_data['public_repos']}]({user_data['html_url']}?tab=repositories)\"\n )\n\n if user_data[\"type\"] == \"User\":\n embed.add_field(\n name=\"Gists\",\n value=f\"[{gists}](https://gist.github.com/{quote_plus(username, safe='')})\"\n )\n\n embed.add_field(\n name=f\"Organization{'s' if len(orgs)!=1 else ''}\",\n value=orgs_to_add if orgs else \"No organizations.\"\n )\n embed.add_field(name=\"Website\", value=blog)\n\n await ctx.send(embed=embed)","def loadInfo(self):\n \n response, = self.client.query( 'get_user_account_by_login', (self.name, ), version = 14 )\n result = utils.responseToDict(self.info_query_description, response)\n self.__dict__.update(result)\n\n if self.sponsor_type != 'NONE':\n self.sponsor = ListMember.create(self.client, self.sponsor_type, self.sponsor_name)\n else:\n self.sponsor = None","def get_profile_data(auth, db):\n\n id_team, user, team, money, color_prim, color_sec = analyze_init(auth, db)\n id_user, seats, fans, ranking, streak = analyze_team_page(auth, db, id_team)\n\n \n v_profile = profile.Profile(\n id_user, user, id_team, team, money, color_prim, \n color_sec, seats, fans, ranking, streak\n )\n\n if (db.profile.find_one({\"id\": int(id_team)}) is not None):\n db.profile.replace_one(\n {\"id\": int(id_team)}, v_profile.to_db_collection())\n else:\n db.profile.insert_one(v_profile.to_db_collection())\n\n print(show(\"profile\") + \" > Perfil actualizado\")\n\n return id_team","def save_profile_picture(current_user, url):\n # save profile_pic\n current_user.profile_pic = url\n current_user.save(current_user)","def synchronize(self):\n if self.oauth_token.expired:\n raise ValueError('Signed request expired.')\n\n profile = get_facebook_profile(self.oauth_token.token)\n\n self.facebook_id = profile.get('id')\n self.facebook_username = profile.get('username')\n self.first_name = profile.get('first_name')\n self.middle_name = profile.get('middle_name', None)\n self.last_name = profile.get('last_name')\n self.profile_url = profile.get('link')\n self.gender = profile.get('gender')\n self.hometown = profile['hometown']['name'] if profile.has_key('hometown') else None\n self.location = profile['location']['name'] if profile.has_key('location') else None\n self.bio = profile.get('bio')\n self.relationship_status = profile.get('relationship_status')\n self.political_views = profile.get('political')\n self.email = profile.get('email')\n self.website = profile.get('website')\n self.locale = profile.get('locale')\n self.verified = profile.get('verified')\n self.birthday = datetime.strptime(profile['birthday'], '%m/%d/%Y') if profile.has_key('birthday') else None\n self.timezone = profile.get('timezone', None)\n self.quotes = profile.get('quotes', None)\n\n self.save()","def get_profiles(count, credentials_type=constants.API_KEY_CREDENTIALS_TYPE, verbose=False):\n print('getting profiles through {}...'.format(credentials_type))\n\n # getting profiles information\n profile_ids = plus_api.get_unique_profile_ids(count, credentials_type=credentials_type, verbose=verbose)\n profile_urls = list(map(plus_api.profile_id_to_google_plus_url, profile_ids))\n # display_names, image_urls = get_profile_display_names_and_image_urls(profile_ids, credentials_type=credentials_type,\n # verbose=verbose)\n display_names, image_urls = get_profile_display_names_and_image_urls_with_getBatchGet(profile_ids,\n credentials_type=credentials_type,\n verbose=verbose)\n\n # saving as plain text\n txt_filename = 'results/people-api-profiles.txt'\n if verbose:\n print('saving profiles as plain text to {}...'.format(txt_filename))\n result_str = ''\n for profile in zip(profile_ids, profile_urls, display_names, image_urls):\n result_str += '\\n'.join(profile) + '\\n\\n'\n profiles_file = open(txt_filename, 'w', encoding='utf-8')\n profiles_file.write(result_str)\n profiles_file.close()\n\n # saving as json\n json_filename = 'results/people-api-profiles.json'\n if verbose:\n print('saving profiles as json to {}...'.format(json_filename))\n result_list = []\n json_fields = ('profileId', 'profileUrl', 'displayName', 'imageUrl')\n for profile in zip(profile_ids, profile_urls, display_names, image_urls):\n result_list.append(dict(zip(json_fields, profile)))\n json_str = json.dumps(result_list)\n profiles_file = open(json_filename, 'w', encoding='utf-8')\n profiles_file.write(json_str)\n profiles_file.close()","def user_profile():\n access_token = _request_ctx_stack.top.current_user_token\n message_log(\"Got access token for user profile\")\n user_profile = get_user_profile(access_token)\n user_profile['access_token'] = access_token\n return json.dumps(user_profile)","def getProfileInfo(self):\n doc = minidom.parse(urllib.urlopen(serverString + \"/rest/user/\" + self.name))\n for element in doc.getElementsByTagName(\"user\")[0].childNodes:\n if element.nodeType != minidom.Node.ELEMENT_NODE:\n continue\n elif element.tagName == \"status\" and int(element.firstChild.data) != 1:\n raise ServerError(element.firstChild.data)\n elif element.tagName == \"input\":\n self.name = element.firstChild.data\n elif element.tagName == \"id\":\n self.id = element.firstChild.data\n elif element.tagName == \"image\":\n self.image = element.firstChild.data\n elif element.tagName == \"tagline\":\n if element.firstChild == None:\n self.tagline = None\n else:\n self.tagline = element.firstChild.data\n elif element.tagName == \"creation\":\n self.created = datetime.datetime.strptime(element.firstChild.data[:element.firstChild.data.rfind(\".\")]+\".GMT\", \"%Y-%m-%d %H:%M:%S.%Z\")","def registerAvatar(self, avatar):\r\n assert self._avatar is None\r\n self._avatar = avatar","def _account_info(remote, resp):\n g.oauth_logged_in_with_remote = remote\n resource = get_resource(remote, resp)\n\n valid_roles = current_app.config.get(\n \"OAUTHCLIENT_CERN_OPENID_ALLOWED_ROLES\",\n OAUTHCLIENT_CERN_OPENID_ALLOWED_ROLES,\n )\n cern_roles = resource.get(\"cern_roles\")\n if cern_roles is None or not set(cern_roles).issubset(valid_roles):\n raise OAuthCERNRejectedAccountError(\n \"User roles {0} are not one of {1}\".format(cern_roles, valid_roles),\n remote,\n resp,\n )\n\n email = resource[\"email\"]\n external_id = resource[\"cern_upn\"]\n nice = resource[\"preferred_username\"]\n name = resource[\"name\"]\n\n return dict(\n user=dict(email=email.lower(), profile=dict(username=nice, full_name=name)),\n external_id=external_id,\n external_method=\"cern_openid\",\n active=True,\n )","async def profilepicture(self, ctx):\n if not ctx.message.mentions:\n await ctx.send(embed=discord.Embed().set_image(\n url=ctx.author.avatar_url))\n else:\n for member in ctx.message.mentions:\n await ctx.send(embed=discord.Embed().set_image(\n url=member.avatar_url))","def get_profile(self):\n endpoint = '/profile'\n return self.get_request(endpoint)","def build_profile(first, last, **user_info):\n user_info['first_name'] = first\n user_info['last_name'] = last\n return user_info","def build_profile(first, last, **user_info):\n user_info['first_name'] = first\n user_info['last_name'] = last\n return user_info","def update_avatar(self, url):\n if self.avatar:\n logging.info(f'{self} already has an avatar: {self.avatar}')\n # TODO: check if image has been updated\n else:\n logging.info(f'{self} has no profile image.')\n img_temp = NamedTemporaryFile(delete=True)\n # TODO: Use requests instead of urllib?\n img_temp.write(urlopen(url).read()) # noqa: S310\n img_temp.flush()\n self.avatar.save(f'{self.pk}', File(img_temp))","def create(self, validated_data):\n user_data = validated_data.pop('user')\n user = UserSerializer.create(UserSerializer(), validated_data=user_data)\n profile, created = Profile.objects.update_or_create(\n user=user,\n avatar=validated_data.pop('avatar'),\n biography=validated_data.pop('biography'),\n link=validated_data.pop('link') \n )\n return profile","def GetAvatar(self):\n\n return self.__GetJsonOrNone(\"/users/\"+self.userName+\"/avatar\", False)","def get_profile():\n logger.debug(\"entering function get_profile\")\n response = read_user_profile()\n logger.debug(\"exiting function get_profile\")\n return jsonify(response)","def get(self, request):\n current_user = request.user\n user_profile = self.serializer_class(current_user.profile)\n return Response({\n \"message\": PROFILE_MSGS['MY_PROFILE'],\n \"profile\": user_profile.data\n }, status=status.HTTP_200_OK)","def display_profile(self):\n print(f\"Id: {self._id}\")\n print(f\"username: {self.username}\")\n print(f\"name: {self.name}\")\n print(f\"contact: {self.contact}\")\n print(f\"address: {self.address}\")","def account(self):\n provider_name = self.provider_entry.get_text()\n provider = Provider.get_by_name(provider_name)\n\n # Create a new provider if we don't find one\n if not provider:\n provider_image = self.provider_image.image\n provider_website = self.provider_website_entry.get_text()\n provider = Provider.create(provider_name, provider_website, None,\n provider_image)\n # Update the provider image if it changed\n elif provider and self.provider_image.image != provider.image:\n provider.update(image=self.provider_image.image)\n\n account = {\n \"username\": self.account_name_entry.get_text(),\n \"provider\": provider\n }\n if not self.props.is_edit:\n # remove spaces\n token = self.token_entry.get_text()\n account[\"token\"] = \"\".join(token.split())\n return account","def makeProfile(request):\n upr = UserProfile()\n upr.user = request.user\n upr.image = \"images/no-pic.png\"\n upr.save()","def handle_callback(self, response):\n\n client = BitbucketClient(access_token=response['access_token'])\n user_info = client.user()\n\n return {\n 'provider_id': user_info['uuid'],\n 'profile_url': user_info['links']['html']['href'],\n 'display_name': user_info['username']\n }","def profile():\n from flickrAPI import FlickrAPI\n #flickr = FlickrAPI(key=session['resource_owner_key'], secret=session['resource_owner_secret'])\n flickr = FlickrAPI(key=request.cookies.get('oauth_token'), secret=request.cookies.get('oauth_token_secret'))\n faves = flickr.favorites_getList(user_id=\"44124394781@N01\", page=1, per_page=5, extras='owner_name')\n return str(faves)","def avatar(self) -> str:\n\t\tif self.hasAvatar:\n\t\t\treturn f'https://avatars.atelier801.com/{self.id % 10000}/{self.id}.jpg'\n\n\t\t# default avatar\n\t\treturn 'https://avatars.atelier801.com/0/0.jpg'","def load_extra_data(backend, details, response, uid, user, social_user=None,\n *args, **kwargs):\n social_user = social_user or \\\n UserSocialAuth.get_social_auth(backend.name, uid)\n if social_user:\n extra_data = backend.extra_data(user, uid, response, details)\n extra_data_field_name = \"{0}_extra_data\".format(backend.name)\n social_user_extra_data = getattr(social_user, extra_data_field_name)\n if extra_data and social_user_extra_data != extra_data:\n if social_user_extra_data:\n social_user_extra_data.update(extra_data)\n else:\n social_user_extra_data = extra_data\n\n #Getting the access token\n access_token_field_name = \"{0}_access_token\".format(backend.name)\n setattr(social_user, access_token_field_name, extra_data['access_token'])\n\n #Storing extra data\n social_user_extra_data.pop('access_token', None)\n social_user_extra_data.pop('id', None)\n setattr(social_user, extra_data_field_name, social_user_extra_data)\n\n social_user.save()\n return {'social_user': social_user}","def build_profile(self, update, context, phone=None, raw_text=None):\n user = update.effective_user\n chat_id = update.effective_chat.id\n log.info(\"PROFILE from %s `%s`\", chat_id, raw_text)\n # import pdb; pdb.set_trace()\n\n # If necessary, create the part of the state that holds data about registration procedures\n if \"registrations\" not in context.bot_data:\n context.bot_data[\"registrations\"] = {}\n\n if chat_id not in context.bot_data[\"registrations\"]:\n # create a new user profile and add it to the bot's state, so we can populate it\n # as we ask the user to provide info about themselves; keep in mind that it is an ORDERED dict, we'll\n # rely on this later!\n profile = OrderedDict(\n {\n c.PROFILE_FIRST_NAME: user.first_name, # may be empty at first\n c.PROFILE_LAST_NAME: user.last_name, # may be empty at first\n c.PROFILE_AVAILABILITY: None,\n c.PROFILE_ACTIVITIES: [],\n c.PROFILE_PHONE: phone,\n c.PROFILE_EMAIL: None,\n }\n )\n\n if not phone.startswith(c.LOCAL_PREFIX):\n # If the Telegram phone number is not a local number (i.e. it was registered abroad), we're moving it\n # to a different attribute, and clearing the original one, such that later in this function we shall\n # ask for a local phone number\n log.debug(\"Phone number is foreign, will ask for a local one\")\n profile[c.PROFILE_PHONE_FOREIGN] = phone\n profile[c.PROFILE_PHONE] = None\n\n context.bot_data[\"registrations\"][chat_id] = profile\n else:\n profile = context.bot_data[\"registrations\"][chat_id]\n\n for key, value in profile.items():\n if not value:\n # a part of the profile is empty, maybe we should ask about it?\n if raw_text:\n # This seems to be yet another call of this function, so raw_text contains the answer to the\n # question asked earlier - let's populate it.\n # NOTE that we use an OrderedDict when building the profile, so we know for sure this answer\n # goes to that particular question (i.e. key in the dict)\n profile[key] = raw_text\n raw_text = None\n continue\n\n # if we got this far, we stumbled upon the next missing part of the profile\n context.user_data[\"state\"] = c.State.EXPECTING_PROFILE_DETAILS\n\n self.updater.bot.send_message(\n chat_id=chat_id,\n text=c.PROFILE_QUESTIONS[key],\n parse_mode=ParseMode.MARKDOWN_V2,\n )\n\n if key == c.PROFILE_ACTIVITIES:\n # this is a special case, because we'll send them an interactive keyboard with options to chose from\n self.confirm_activities(update, context)\n return\n\n return\n\n # if we got this far, it means the profile is complete, inform the user about it\n self.updater.bot.send_message(\n chat_id=chat_id, text=c.MSG_ONBOARD_NEXT_STEPS, parse_mode=ParseMode.MARKDOWN,\n )\n\n # and the backend, but first let's augment the profile with more data\n profile[c.PROFILE_CHAT_ID] = chat_id\n self.backend.register_pending_volunteer(profile)\n context.user_data[\"state\"] = c.State.AVAILABLE\n\n # remove if from the state, because we don't need it anymore\n del context.bot_data[\"registrations\"][chat_id]\n\n # Also get rid of this user's individual keyboard for assitance activities\n context.user_data.pop(\"assist_keyboard\", None)","def getprofile(self, *args, **kwargs):\n return _image.image_getprofile(self, *args, **kwargs)","def head(self, entity):\n link = '<{url}>; rel=\"https://tent.io/rels/profile\"'.format(\n url=url_for('entity.profile', entity=entity.name, _external=True))\n resp = jsonify(entity.to_json())\n resp.headers['Link'] = link\n \n return resp","def build_profile(first, last, **user_info):\n\n user_info['first_name'] = first\n user_info['last_name'] = last\n return user_info","def _getProfileFromUser(self):\n # Make sure user is authenticated\n user = endpoints.get_current_user()\n if not user:\n raise endpoints.UnauthorizedException('Authorization required')\n # Get Profile from datastore\n user_id = user.email()\n p_key = ndb.Key(Profile, user_id)\n profile = p_key.get()\n # Create new Profile if not there\n if not profile:\n profile = Profile(\n key = p_key,\n displayName = user.nickname(),\n mainEmail= user.email(),\n teeShirtSize = str(TeeShirtSize.NOT_SPECIFIED),\n )\n profile.put()\n return profile","def get_user_details(self, response):\n # Build the username with the team $username@$team_url\n # Necessary to get unique names for all of slack\n username = response.get('user')\n if self.setting('USERNAME_WITH_TEAM', True):\n match = re.search(r'//([^.]+)\\.slack\\.com', response['url'])\n username = '{0}@{1}'.format(username, match.group(1))\n\n out = {'username': username}\n if 'profile' in response:\n out.update({\n 'email': response['profile'].get('email'),\n 'fullname': response['profile'].get('real_name'),\n 'first_name': response['profile'].get('first_name'),\n 'last_name': response['profile'].get('last_name'),\n 'team_name': response.get('team_name')\n })\n return out","def get_profile_picture_url(cls, filename):\n if filename is None:\n return None\n profile_picture = bucket.blob('images/users/'+filename)\n if profile_picture.exists():\n profile_picture.make_public()\n return profile_picture.public_url\n return None","def profile(request):\n profile = request.user.profile\n return render(request, 'accounts/profile.html', {'profile': profile})","def save_profile(self, request):\n return self.profile_service.do_profile(request)","def add_gamer_profile():\n user_id = current_user.get().id\n user = User.objects.get_or_404(id=user_id)\n\n # should we validate object_id here?\n\n # auto populate with g.formdata\n game_form = GamerForm(g.formdata)\n game_form.populate_obj(user.gamer)\n\n if request.files and request.files['avatar']:\n filename = Avatars.save(request.files['avatar'])\n user.gamer.avatar = Avatars.url(filename)\n\n # manually handle games\n user.gamer.games = [Game(id=_id)\n for _id in request.form.getlist('games')]\n\n user.save()\n return user","def resolve_avatar(self, info):\n if self.avatar:\n self.avatar = info.context.build_absolute_uri(self.avatar.url)\n return self.avatar","def get_gallery_profile(self):\n url = (\"https://api.imgur.com/3/account/{0}/\"\n \"gallery_profile\".format(self.name))\n return self._imgur._send_request(url)","def getAvatarInfo(self):\n return \", \".join(self._get_avatar_info())","def _profile_to_InstagramUser(profile: Dict[str, Any]) -> _InstagramUser:\n # Navigate to the user JSON that is coincidentally used by the provided API methods\n user = profile['users'][0]['user']\n\n # Simply build our InstagramUser, as the user JSON is the same\n return _InstagramUser(user)","def update_info(self):\n\n r = requests.get(self.url + 'getMe')\n if r.status_code == 200:\n response = json.loads(r.text)\n if response['ok']:\n bot_info = response['result']\n self.user_id = bot_info['id']\n self.first_name = bot_info['first_name']\n if 'last_name' in bot_info:\n self.last_name = bot_info['last_name']\n if 'username' in bot_info:\n self.username = bot_info['username']\n else:\n raise TelegramError('The result was not \"ok\"')\n else:\n raise TelegramError('Did not get a 200 response', r.status_code)","def create_or_update_profile(sender, instance, created, **kwargs):\n if created:\n Profile.objects.get_or_create(user=instance)\n instance.profile.save()","def google_extra_values(sender, user, response, details, **kwargs):\n LOGGER.debug('profile.models.google_extra_values')\n user_info_url = \"https://www.googleapis.com/oauth2/v1/userinfo\"\n\n data = {'access_token': response.get('access_token', ''), 'alt': 'json'}\n params = urlencode(data)\n try:\n request = Request(user_info_url + '?' + params, headers={'Authorization': params})\n result = simplejson.loads(urlopen(request).read())\n\n user.last_name = result.get('family_name', '')\n user.first_name = result.get('given_name', '')\n profile = user.profile\n profile.gender = result.get('gender', '')\n profile.image_url = result.get('picture', '')\n profile.url = result.get('link', '')\n\n profile.save()\n except:\n pass\n\n return True","def player_avatar(player_obj):\n avatar = \"http://www.gravatar.com/avatar/%s.jpg?d=monsterid\" % md5(player_obj.user.email).hexdigest()\n\n return avatar","def _username_to_profile(self, username: str) -> Dict[str, Any]:\n\n base_url = self.keys.pre_profile + username + self.keys.rank_token + self.keys.post_profile\n\n # Build the page source url for the given user's account\n con = urllib.request.urlopen(base_url)\n user_profile = con.read().decode('utf-8')\n\n # Convert the webpage to a profile JSON\n profile: dict = json.loads(str(user_profile))\n return profile","def get_user_details(self, response):\n\n kaccount_email = \"\"\n kakao_account = response.get(\"kakao_account\", \"\")\n if kakao_account:\n kaccount_email = kakao_account.get(\"email\", \"\")\n properties = response.get(\"properties\", \"\")\n nickname = properties.get(\"nickname\") if properties else \"\"\n return {\n \"username\": nickname,\n \"email\": kaccount_email,\n \"fullname\": nickname,\n \"first_name\": nickname[1:] if nickname else \"\",\n \"last_name\": nickname[0] if nickname else \"\",\n }","def save_profile_picture(backend, user, response, *args, **kwargs):\n profile = Profile.objects.filter(user=user).first()\n if profile is None:\n profile = Profile(user=user)\n\n if \"locale\" in response:\n profile.locale = response[\"locale\"]\n\n if \"picture\" in response:\n with TempDir() as d:\n _name, ext = splitext(response[\"picture\"])\n r = requests.get(response[\"picture\"], stream=True)\n\n if r.ok:\n temp_file = d / f\"{user.pk}{ext or '.jpg'}\"\n\n with open(temp_file, \"wb\") as fd:\n for chunk in r.iter_content(chunk_size=128):\n fd.write(chunk)\n\n with temp_file.open(\"rb\") as fd:\n profile.picture.save(\n f\"{user.pk}{ext or '.jpg'}\", File(fd), save=False\n )\n\n profile.save()","def set_user_profile_picture(user_id, file_name):\n\n user = User.query.get(user_id)\n \n user.profile_picture = file_name\n db.session.commit()","def _parse_profile_data (self, netflix_page_data):\n profiles = {};\n important_fields = [\n 'profileName',\n 'isActive',\n 'isAccountOwner',\n 'isKids'\n ]\n # values are accessible via dict (sloppy parsing successfull)\n if type(netflix_page_data) == dict:\n for profile_id in netflix_page_data.get('profiles'):\n if self._is_size_key(key=profile_id) == False and type(netflix_page_data['profiles'][profile_id]) == dict and netflix_page_data['profiles'][profile_id].get('avatar', False) != False:\n profile = {'id': profile_id}\n for important_field in important_fields:\n if important_field == 'profileName':\n profile.update({important_field: HTMLParser.HTMLParser().unescape(netflix_page_data['profiles'][profile_id]['summary'][important_field]).encode('utf8')})\n else:\n profile.update({important_field: netflix_page_data['profiles'][profile_id]['summary'][important_field]})\n avatar_base = netflix_page_data['nf'].get(netflix_page_data['profiles'][profile_id]['summary']['avatarName'], False);\n avatar = 'https://secure.netflix.com/ffe/profiles/avatars_v2/320x320/PICON_029.png' if avatar_base == False else avatar_base['images']['byWidth']['320']['value']\n profile.update({'avatar': avatar, 'isFirstUse': False})\n profiles.update({profile_id: profile})\n return profiles\n\n # values are stored in lists (returned from JS parser)\n # TODO: get rid of this christmas tree of doom\n for item in netflix_page_data:\n if 'hasViewedRatingWelcomeModal' in dict(item).keys():\n for profile_id in item:\n if self._is_size_key(key=profile_id) == False and type(item[profile_id]) == dict and item[profile_id].get('avatar', False) != False:\n profile = {'id': profile_id}\n for important_field in important_fields:\n profile.update({important_field: item[profile_id]['summary'][important_field]})\n avatar_base = item['nf'].get(item[profile_id]['summary']['avatarName'], False);\n avatar = 'https://secure.netflix.com/ffe/profiles/avatars_v2/320x320/PICON_029.png' if avatar_base == False else avatar_base['images']['byWidth']['320']['value']\n profile.update({'avatar': avatar})\n profiles.update({profile_id: profile})\n return profiles","def save_user_profile(instance, **_):\n instance.profile.save()","def build_profile(first, last, **user_info):\n profile = {}\n profile['first_name'] = first\n profile['last_name'] = last\n for key, value in user_info.items():\n profile[key] = value\n return profile","def build_profile(first, last, **user_info):\n profile = {}\n profile['first_name'] = first\n profile['last_name'] = last\n for key, value in user_info.items():\n profile[key] = value\n return profile","def update_user_profile(id):\n token = request.json['token']\n u = user.User.query.filter(user.User.token == token).first()\n if u is None:\n abort(404)\n if u.id != id:\n print \"user id is wrong.\" #TODO: Support log system\n abort(500)\n u.name = request.json['name']\n u.nickname = request.json['nickname']\n u.company = request.json['nickname']\n with store_context(fs_store):\n with open(files.path(request.json['header'])) as f:\n u.header_icon.from_file(f)\n db.session.merge(u)\n db.session.commit()\n db.session.merge(u)\n db.session.commit()\n return jsonify(u.to_dict())","def _request_user_info(credentials):\n http = httplib2.Http()\n\n # User information stored here\n credentials.authorize(http)\n resp, content = http.request('https://www.googleapis.com/plus/v1/people/me')\n\n if resp.status != 200:\n current_app.logger.error(\"Error while obtaining user profile: %s\" % resp)\n return None\n\n # Check whether user is authenticating with the allowed domain.\n if (current_app.config['RESTRICT_DOMAIN'] is True and \n credentials.id_token.get('hd') != current_app.config['REQUIRED_DOMAIN']):\n\n # Replace with logging for a real app\n print(\"\\n------------------------------------------------------\")\n print(\"User attempted to authenticate with disallowed domain.\")\n print(\"------------------------------------------------------\\n\")\n\n # User information deleted here\n oauth2.storage.delete()\n return None\n\n # Stores the users information in the session profile dictionary\n session['profile'] = json.loads(content.decode('utf-8'))\n\n # Remove this in production. It's here so you can see what information is stored.\n print(\"\\n------------------------------------------------------\")\n print(\"SESSION PROFILE INFORMATION\")\n print(\"------------------------------------------------------\")\n for k,v in session['profile'].items():\n print(k,\"--->\",v)\n print(\"------------------------------------------------------\\n\")","def prefetch_login(self):\n self.profiles = []\n email = self.credentials.get('email', '')\n password = self.credentials.get('password', '')\n if email != '' and password != '':\n if self.netflix_session.is_logged_in(account=self.credentials):\n refresh_session = self.netflix_session.refresh_session_data(\n account=self.credentials)\n if refresh_session:\n self.profiles = self.netflix_session.profiles\n else:\n if self.netflix_session.login(account=self.credentials):\n self.profiles = self.netflix_session.profiles\n self.kodi_helper.set_esn(self.netflix_session.esn)","def get_user_profile(self, username: str = None) -> Profile:\n if username:\n res: dict[str, Any] = self.api.users.get(user_ids=username).pop()\n else:\n res = self.api.users.get().pop()\n return VkontakteConnector.build_profile(res)","async def steam_avatar(message):\n s = message.content.strip()\n id = await parse_steam_id(s)\n profile = await steam_community.steam_profile(id.to_64(), id64=True)\n if profile.avatar:\n return profile.avatar\n else:\n raise CommandError(\"The user has no avatar.\")","def __str__(self):\n profile = {\n 'address': self.address,\n 'bio': self.bio,\n 'website': self.website,\n 'hireable': self.hireable,\n 'travel_radius': self.travel_radius,\n 'phone number': self.phone_number,\n 'camera type': self.camera_type,\n 'type of photography': self.photography_type\n }\n return str(self.user.username)","def build_profile(first, last, **user_info):\n profile = {}\n profile['first_name'] = first\n profile['last_name'] = last\n for key, value in user_info.items():\n profile[key] = value\n return(profile)","def build_profile(first, last, **user_info):\n profile = {}\n profile['first_name'] = first\n profile['last_name'] = last\n for key, value in user_info.items():\n profile[key] = value\n return(profile)","def get_author_data(uid, gravatar_on_right_side=True, linked_with_users_page=True, profile_picture_size=20):\n db_user = DBDiscussionSession.query(User).get(int(uid))\n if not db_user:\n return None, 'Missing author with uid ' + str(uid), False\n\n nick = db_user.global_nickname\n img_src = get_profile_picture(db_user, profile_picture_size)\n link_begin = ''\n link_end = ''\n if linked_with_users_page:\n link_begin = '<a href=\"/user/{}\" title=\"{}\">'.format(db_user.uid, nick)\n link_end = '</a>'\n\n side = 'left' if gravatar_on_right_side else 'right'\n img = '<img class=\"img-circle\" src=\"{}\" style=\"padding-{}: 0.3em\">'.format(img_src, side)\n\n if gravatar_on_right_side:\n return db_user, '{}{}{}{}'.format(link_begin, nick, img, link_end), True\n else:\n return db_user, '{}{}{}{}'.format(link_begin, img, nick, link_end), True","def get_user_details(self, response):\n name = response.get(\"name\")\n return {\n \"username\": str(response.get(\"account_id\")),\n \"email\": response.get(\"email\"),\n \"fullname\": name.get(\"display_name\"),\n \"first_name\": name.get(\"given_name\"),\n \"last_name\": name.get(\"surname\"),\n }","def build_profile(first, last, **user_info):\n profile = {}\n profile['first_name'] = first\n profile['last_name'] = last\n for key, value in user_info.items():\n profile[key] = value\n return profile","def build_profile(first, last, **user_info):\n profile = {}\n profile['first_name'] = first\n profile['last_name'] = last\n for key, value in user_info.items():\n profile[key] = value\n return profile","def build_profile(first, last, **user_info):\n profile = {}\n profile['first_name'] = first\n profile['last_name'] = last\n for key, value in user_info.items():\n profile[key] = value\n return profile","async def get_my_profile_info_async(\n namespace: Optional[str] = None,\n x_additional_headers: Optional[Dict[str, str]] = None,\n **kwargs\n):\n if namespace is None:\n namespace, error = get_services_namespace()\n if error:\n return None, error\n request = GetMyProfileInfo.create(\n namespace=namespace,\n )\n return await run_request_async(\n request, additional_headers=x_additional_headers, **kwargs\n )","def google_profile_extra_details(backend, details, response, user=None, is_new=False,\n *args, **kwargs):\n profile = {}\n\n logger.info('Pipeline logging in with google-oauth2')\n\n if user is not None:\n client_profile = user.profile\n # save all updated fields\n if isinstance(response, dict):\n logger.info('Pipeline.google updating client.profile_data from google %s - %s' % (user, response,) )\n\n client_profile.profile_data.update(response) \n\n client_profile.save(update_fields=['profile_data'])","def profile_pic(self, client_file_storage):\n\n # If we already have a profile picture, remove it\n if self.profile_pic_filename:\n filepath = os.path.join(\n current_app.config['UPLOADED_IMAGES_DEST'],\n self.profile_pic_filename)\n os.remove(filepath)\n self.profile_pic_filename = None\n self.profile_pic_url = None\n\n # This uploads & saves the file on the server\n # NOTE: It uses the secure_filename function...\n server_filename = images.save(client_file_storage)\n\n # Generate the URL to this file\n url = images.url(server_filename)\n\n # Store information with the user\n self.profile_pic_filename = server_filename\n self.profile_pic_url = url","def __str__(self):\n return self.user.username + \"'s Profile\"","def _get_linkedin_accounts(self, linkedin_access_token):\n response = requests.get(\n 'https://api.linkedin.com/v2/me?projection='\n + '(id,localizedLastName,localizedFirstName,'\n + 'profilePicture(displayImage~:playableStreams))',\n headers={\n 'Authorization': 'Bearer ' + linkedin_access_token,\n 'cache-control': 'no-cache',\n 'X-Restli-Protocol-Version': '2.0.0'\n }\n ).json()\n\n if ('id' in response and 'localizedLastName' in response\n and 'localizedFirstName' in response):\n linkedin_account_id = 'urn:li:person:' + response['id']\n\n try:\n image_url = response['profilePicture']['displayImage~']['elements'][0]['identifiers'][0]['identifier']\n linkedin_profile_image = base64.b64encode(requests.get(image_url).content)\n except Exception:\n linkedin_profile_image = ''\n\n # TODO - STD: add each companies page\n return [{\n 'name': response['localizedLastName'] + ' ' + response['localizedFirstName'],\n 'linkedin_account_id': linkedin_account_id,\n 'linkedin_access_token': linkedin_access_token,\n 'image': linkedin_profile_image\n }]\n\n return []","def get_profile(tag, platform=\"pc\", region=\"eu\"):\n #\n try:\n context = ssl._create_unverified_context()\n profile = json.load(\n const.codec(urlopen(const.URL + platform + \"/\" + region + \"/\" + tag + \"/profile\", context=context)))\n #\n if \"error\" in profile:\n raise BattleTagNotFound(profile['error'])\n exit(1)\n #\n result = pr.Profile(profile['data']['username'],\n profile['data']['level'],\n profile['data']['games']['quick']['wins'],\n profile['data']['games']['competitive']['wins'],\n profile['data']['games']['competitive']['lost'],\n profile['data']['playtime']['quick'],\n profile['data']['playtime']['competitive'],\n profile['data']['avatar'],\n profile['data']['competitive']['rank'])\n return result\n except urllib.error.URLError as e:\n print(\"An error occurred when fetching stats\\n\" + e)\n exit(1)\n except Exception as e:\n print(\"An error occurred:\\n \" + str(e))\n exit(1)","def save(self, profile_callback=None):\n\n # First, save the parent form\n new_user = super(BodbRegistrationForm, self).save(profile_callback=profile_callback)\n\n # Update user with first, last names\n new_user.first_name = self.cleaned_data['first_name']\n new_user.last_name = self.cleaned_data['last_name']\n new_user.save()\n\n # Update profile with affiliation\n profile = new_user.get_profile()\n profile.affiliation = self.cleaned_data['affiliation']\n profile.save()\n\n cache.set('%d.profile' % new_user.id, profile)\n\n return new_user","def profiles():\n images = get_uploaded_images()\n records = db.session.query(UserProfile).all()\n return render_template('profiles.html', images=images, records =records)","def getProfile(self):\n # GET /profile\n debugMain('getProfile')\n return self._genericGet('/profile')","def get_user_profile(self):\n return self.user.profile","def get(self, username):\n\t\tdb = getattr(g, 'db', None)\n\n\t\tqry = \"SELECT username,email,active,steamid FROM\\\n\t\t\tprofiles WHERE username = %s;\"\n\t\twith db as cursor:\n\t\t\tcursor.execute(qry, (username,))\n\n\t\treturn {'profile':cursor.fetchone()}","def profile():\n token = request.json['token']\n u = user.User.query.filter(user.User.token == token).first()\n if u is None:\n abort(404)\n return jsonify(u.to_dict())"],"string":"[\n \"def get_profile(self,soup,Name,Profile_url):\\n h3=soup.find(\\\"h3\\\")\\n \\n if h3 !=None:\\n if h3.find(\\\"a\\\")!=None: \\n name=h3.a.get_text()\\n if h3.a.has_attr('href') :\\n h3_a_tag=h3.a['href']\\n h3_a_tag=h3_a_tag.replace(\\\"&__tn__=C-R\\\",'')\\n \\n profile_url='https://www.facebook.com'+h3_a_tag\\n else :\\n profile_url=\\\"None\\\"\\n \\n else :\\n h31=soup.find('a',class_=\\\"actor-link\\\")\\n if h31!=None:\\n name=h31.get_text()\\n if h31.has_attr('href'):\\n profile_url='https://www.facebook.com'+h31['href']\\n else: \\n profile_url=\\\"None\\\"\\n else:\\n \\n name=\\\"None\\\"\\n profile_url=\\\"None\\\"\\n \\n \\n Name.append(name)\\n Profile_url.append(profile_url)\\n \\n return Name,Profile_url\",\n \"def get_profile(self,fields=('id','first-name','last-name','headline','summary')):\\n\\n if not self._access_token:\\n raise FBError(\\\"Authentication needed!\\\")\\n \\n token = oauth.Token(self._access_token['oauth_token'], self._access_token['oauth_token_secret'])\\n client = oauth.Client(self.consumer, token)\\n profile_url = self.profile_url % \\\",\\\".join(fields)\\n resp, content = client.request(profile_url,headers={\\\"x-li-format\\\":'json'})\\n \\n if resp['status'] != '200':\\n print resp\\n raise FBError(\\\"Invalid response %s.\\\" % resp['status'])\\n \\n try:\\n return json.loads(content)\\n except Exception, e:\\n raise FBError(\\\"Invalid json %s.\\\" % unicode(e))\",\n \"def add_profile_photo():\\n pass\",\n \"def get_profile_image(strategy, details, response, uid, user, social, is_new=False, *args, **kwargs):\\n\\n # If we don't have a user then just return\\n if user is None:\\n return\\n\\n # Save photo from FB\\n if kwargs['backend'].name == \\\"facebook\\\":\\n try:\\n image_url = \\\"https://graph.facebook.com/%s/picture?type=large\\\" % uid\\n result = urllib.urlretrieve(image_url)\\n\\n def save_image(user, uid, result):\\n user.original_photo.save(\\\"%s.jpg\\\" % uid, File(open(result[0])))\\n user.save(update_fields=['original_photo'])\\n\\n retry_cloudfiles(save_image, user, uid, result)\\n except URLError:\\n pass\\n elif kwargs['backend'].name == \\\"twitter\\\" and social:\\n try:\\n # Get profile image to save\\n if response['profile_image_url'] != '':\\n image_result = urllib.urlretrieve(response['profile_image_url'])\\n\\n def save_image(user, uid, image_result):\\n user.original_photo.save(\\\"%s.jpg\\\" % uid, File(open(image_result[0])))\\n user.save(update_fields=['original_photo'])\\n\\n retry_cloudfiles(save_image, user, uid, image_result)\\n except URLError:\\n pass\",\n \"def profile(self) -> dict:\\n endpoint = \\\"/api/users/profile/\\\"\\n ret = self._request(endpoint=endpoint)\\n return ret\",\n \"def fetch_profile(access_token):\\n\\n headers = {\\\"Authorization\\\": \\\"Bearer \\\" + access_token}\\n\\n # Fetch athlete profile\\n r = requests.get(API_URL + \\\"/athlete\\\", headers=headers)\\n profile = r.json()\\n if \\\"errors\\\" in profile:\\n raise AuthError(profile[\\\"message\\\"])\\n\\n return {\\n \\\"firstName\\\": profile[\\\"firstname\\\"],\\n \\\"lastName\\\": profile[\\\"lastname\\\"],\\n \\\"imgUrl\\\": profile[\\\"profile\\\"],\\n \\\"profileUrl\\\": \\\"https://www.strava.com/athletes/{}\\\".format(profile[\\\"id\\\"]),\\n }\",\n \"def profile(self, name=\\\"johndoe\\\"):\\r\\n url = \\\"/account/%s\\\" % name\\r\\n return self.app.get(url, follow_redirects=True)\",\n \"def get_my_profile(self):\\n\\n url = self.api_base_url + \\\"user/profile\\\"\\n\\n try:\\n raw_response = self.request_handler.make_request(ApiRequestHandler.GET, url)\\n except RequestFailed:\\n raise\\n\\n jsonified_response = json.loads(raw_response.text)\\n user_profile = jsonified_response\\n\\n return user_profile\",\n \"def create(self, request, *args, **kwargs):\\n response = super().create(request, *args, **kwargs)\\n profile = response.data\\n user_name = profile.get(\\\"username\\\")\\n cache.set(f\\\"{USER_PROFILE_PREFIX}{user_name}\\\", profile)\\n return response\",\n \"def get_user_profile_info(token):\\n user_profile_endpoint = 'https://api.spotify.com/v1/me'\\n headers = {'Authorization': 'Bearer %s' % token}\\n\\n r = requests.get(user_profile_endpoint, headers=headers)\\n if r.status_code != 200:\\n return None\\n profile = r.json()\\n\\n display_name = profile['display_name']\\n profile_id = profile['id']\\n email = profile['email']\\n\\n return User(display_name=display_name,\\n profile_id=profile_id,\\n email=email)\",\n \"def extra_data(self, user, uid, response, details):\\n try:\\n return self.get_steam_profile(response)\\n except:\\n return \\\"\\\"\",\n \"def social_associate_and_load_data(backend, details, response, uid, user,\\n social_user=None, *args, **kwargs):\\n extra_data = backend.extra_data(user, uid, response, details)\\n created = False\\n if not social_user and user:\\n social_user, created = UserSocialAuth.objects.get_or_create(\\n user_id=user.id,\\n provider=backend.name,\\n uid=uid,\\n defaults={'extra_data': extra_data})\\n\\n if not created and extra_data and social_user.extra_data != extra_data:\\n social_user.extra_data.update(extra_data)\\n social_user.save()\\n return {'social_user': social_user}\",\n \"def build_avatar(login, request):\\n log.debug(\\\"# Building avatar\\\")\\n from {{package}}.models import DBSESSION\\n from {{package}}.models.model import User\\n avatar = DBSESSION.query(Account).filter_by(account_lid=login).first()\\n request.session['user'] = avatar\\n if avatar:\\n return []\",\n \"def user_profile(first, last, **add_info):\\n profile = {}\\n profile['firstname'] = first\\n profile['lastname'] = last\\n\\n for key, value in add_info.items():\\n profile[key] = value\\n \\n return profile\",\n \"def profile_details(self, profile_name):\\n url = get_url('profile details', profile=profile_name)\\n response = self._get(url)\\n if response.status_code == 404:\\n return None\\n raise_on_error(response)\\n return Profile(response.json())\",\n \"def post(self, request):\\n profile = Profile.get_by_id(request.user.id)\\n if not profile:\\n return HttpResponse(status=403)\\n if not imageValidator(request.FILES.get('name')):\\n return HttpResponse(status=400)\\n imageToUpload = request.FILES.get('name')\\n key = 'avatar=' + imageToUpload.name\\n url = upload(key, imageToUpload)\\n profile.update(avatar=url)\\n return JsonResponse(profile.to_dict(), status=200)\",\n \"async def profile(self, ctx, member: discord.Member = None):\\n if member is None:\\n member = ctx.author\\n name = member.display_name\\n id = member.id\\n avatar = member.avatar_url\\n created = member.created_at\\n embed = discord.Embed(\\n title=name,\\n description=name + \\\"'s profile!\\\",\\n color=discord.Colour.purple()\\n )\\n embed.set_thumbnail(url=avatar)\\n embed.add_field(name=\\\"ID:\\\", value=str(id))\\n embed.add_field(name=\\\"Created At:\\\", value=str(created))\\n if ctx.guild is not None:\\n joined_at = member.joined_at\\n embed.add_field(name=\\\"Joined Server At:\\\", value=str(joined_at))\\n if member.bot:\\n embed.add_field(name=\\\"Bot?\\\", value=\\\"Yeah\\\")\\n if member.premium_since is not None:\\n premium = str(member.premium_since)\\n embed.add_field(name=\\\"Premium Member Since: \\\", value=premium)\\n else:\\n embed.add_field(name=\\\"Premium?\\\", value=\\\"Nope\\\")\\n top_role = str(member.top_role)\\n embed.add_field(name=\\\"Top Role:\\\", value=top_role)\\n await ctx.send(\\\"\\\", embed=embed)\",\n \"def set_image_from_socialaccount(self, socialaccount: Union[str, SocialAccount]):\\n if not self.image_is_identicon():\\n return\\n\\n if not isinstance(socialaccount, SocialAccount):\\n try:\\n socialaccount = SocialAccount.objects.get(\\n user=self.user, provider=socialaccount\\n )\\n except SocialAccount.DoesNotExist:\\n return\\n\\n url = None\\n provider = socialaccount.provider\\n data = socialaccount.extra_data\\n if provider == \\\"google\\\":\\n url = data.get(\\\"picture\\\")\\n elif provider == \\\"github\\\":\\n url = data.get(\\\"avatar_url\\\")\\n elif provider == \\\"twitter\\\":\\n url = data.get(\\\"profile_image_url\\\")\\n\\n if url:\\n self.set_image_from_url(url)\",\n \"def fetch_my_profile(self, api_token: str) -> dict:\\n query = \\\"\\\"\\\"\\n query myProfile {\\n myProfile {\\n id\\n firstName\\n lastName\\n }\\n }\\n \\\"\\\"\\\"\\n path = jmespath.compile(\\n \\\"\\\"\\\"\\n data.myProfile.{\\n id: id\\n first_name: firstName\\n last_name: lastName\\n }\\n \\\"\\\"\\\"\\n )\\n\\n data = self.do_query(query, api_token=api_token)\\n\\n parsed_data = path.search(data)\\n self.contains_keys(parsed_data, [\\\"id\\\", \\\"first_name\\\", \\\"last_name\\\"])\\n return parsed_data\",\n \"def get_facebook_user_info(access_token):\\n required_data_list = []\\n for per in settings.FACEBOOK_EXTENDED_PERMISSIONS:\\n required_data_list.append(per.replace(\\\"user_\\\",\\\"\\\"))\\n \\n required_data_list.append(\\\"picture.type(large)\\\")\\n required_data = (\\\", \\\").join([data for data in required_data_list])\\n \\n graph_url = \\\"https://graph.facebook.com/me?access_token=%s&fields=%s\\\" % (access_token,required_data)\\n public_info_url = \\\"https://graph.facebook.com/me?access_token=%s\\\" % access_token\\n \\n profile = json.load(urllib.urlopen(graph_url))\\n profile_info = json.load(urllib.urlopen(public_info_url))\\n \\n profile_response_dict = {}\\n profile_response_dict.update(profile)\\n profile_response_dict.update(profile_info)\\n profile_response_json = json.dumps(profile_response_dict)\\n\\n return (profile_response_json, profile_response_dict)\",\n \"def user_profileImg(id):\\n data = request.get_json(force=True)\\n\\n user = User.query.get(id)\\n user.profileImg = data['profileImg']\\n db.session.commit()\\n return {'user': user.to_dict()}\",\n \"def retrieve_profile(self, name):\\n\\n url = get_url('profile details', profile=name)\\n response = self._get(url)\\n raise_on_error(response)\\n if response.status_code == 404:\\n raise QarnotGenericException(response.json()['message'])\\n return Profile(response.json())\",\n \"def profile_pic(request):\\n if request.user.is_authenticated:\\n profile_obj = CustomUser.objects.get(id__exact=request.user.id)\\n pic = profile_obj.avatar\\n return {'picture': pic}\\n return {}\",\n \"async def github_user_info(self, ctx: commands.Context, username: str) -> None:\\n async with ctx.typing():\\n user_data = await self.fetch_data(f\\\"{GITHUB_API_URL}/users/{quote_plus(username)}\\\")\\n\\n # User_data will not have a message key if the user exists\\n if \\\"message\\\" in user_data:\\n embed = discord.Embed(\\n title=random.choice(NEGATIVE_REPLIES),\\n description=f\\\"The profile for `{username}` was not found.\\\",\\n colour=Colours.soft_red\\n )\\n\\n await ctx.send(embed=embed)\\n return\\n\\n org_data = await self.fetch_data(user_data[\\\"organizations_url\\\"])\\n orgs = [f\\\"[{org['login']}](https://github.com/{org['login']})\\\" for org in org_data]\\n orgs_to_add = \\\" | \\\".join(orgs)\\n\\n gists = user_data[\\\"public_gists\\\"]\\n\\n # Forming blog link\\n if user_data[\\\"blog\\\"].startswith(\\\"http\\\"): # Blog link is complete\\n blog = user_data[\\\"blog\\\"]\\n elif user_data[\\\"blog\\\"]: # Blog exists but the link is not complete\\n blog = f\\\"https://{user_data['blog']}\\\"\\n else:\\n blog = \\\"No website link available\\\"\\n\\n embed = discord.Embed(\\n title=f\\\"`{user_data['login']}`'s GitHub profile info\\\",\\n description=f\\\"```{user_data['bio']}```\\\\n\\\" if user_data[\\\"bio\\\"] else \\\"\\\",\\n colour=discord.Colour.blurple(),\\n url=user_data[\\\"html_url\\\"],\\n timestamp=datetime.strptime(user_data[\\\"created_at\\\"], \\\"%Y-%m-%dT%H:%M:%SZ\\\")\\n )\\n embed.set_thumbnail(url=user_data[\\\"avatar_url\\\"])\\n embed.set_footer(text=\\\"Account created at\\\")\\n\\n if user_data[\\\"type\\\"] == \\\"User\\\":\\n\\n embed.add_field(\\n name=\\\"Followers\\\",\\n value=f\\\"[{user_data['followers']}]({user_data['html_url']}?tab=followers)\\\"\\n )\\n embed.add_field(\\n name=\\\"Following\\\",\\n value=f\\\"[{user_data['following']}]({user_data['html_url']}?tab=following)\\\"\\n )\\n\\n embed.add_field(\\n name=\\\"Public repos\\\",\\n value=f\\\"[{user_data['public_repos']}]({user_data['html_url']}?tab=repositories)\\\"\\n )\\n\\n if user_data[\\\"type\\\"] == \\\"User\\\":\\n embed.add_field(\\n name=\\\"Gists\\\",\\n value=f\\\"[{gists}](https://gist.github.com/{quote_plus(username, safe='')})\\\"\\n )\\n\\n embed.add_field(\\n name=f\\\"Organization{'s' if len(orgs)!=1 else ''}\\\",\\n value=orgs_to_add if orgs else \\\"No organizations.\\\"\\n )\\n embed.add_field(name=\\\"Website\\\", value=blog)\\n\\n await ctx.send(embed=embed)\",\n \"def loadInfo(self):\\n \\n response, = self.client.query( 'get_user_account_by_login', (self.name, ), version = 14 )\\n result = utils.responseToDict(self.info_query_description, response)\\n self.__dict__.update(result)\\n\\n if self.sponsor_type != 'NONE':\\n self.sponsor = ListMember.create(self.client, self.sponsor_type, self.sponsor_name)\\n else:\\n self.sponsor = None\",\n \"def get_profile_data(auth, db):\\n\\n id_team, user, team, money, color_prim, color_sec = analyze_init(auth, db)\\n id_user, seats, fans, ranking, streak = analyze_team_page(auth, db, id_team)\\n\\n \\n v_profile = profile.Profile(\\n id_user, user, id_team, team, money, color_prim, \\n color_sec, seats, fans, ranking, streak\\n )\\n\\n if (db.profile.find_one({\\\"id\\\": int(id_team)}) is not None):\\n db.profile.replace_one(\\n {\\\"id\\\": int(id_team)}, v_profile.to_db_collection())\\n else:\\n db.profile.insert_one(v_profile.to_db_collection())\\n\\n print(show(\\\"profile\\\") + \\\" > Perfil actualizado\\\")\\n\\n return id_team\",\n \"def save_profile_picture(current_user, url):\\n # save profile_pic\\n current_user.profile_pic = url\\n current_user.save(current_user)\",\n \"def synchronize(self):\\n if self.oauth_token.expired:\\n raise ValueError('Signed request expired.')\\n\\n profile = get_facebook_profile(self.oauth_token.token)\\n\\n self.facebook_id = profile.get('id')\\n self.facebook_username = profile.get('username')\\n self.first_name = profile.get('first_name')\\n self.middle_name = profile.get('middle_name', None)\\n self.last_name = profile.get('last_name')\\n self.profile_url = profile.get('link')\\n self.gender = profile.get('gender')\\n self.hometown = profile['hometown']['name'] if profile.has_key('hometown') else None\\n self.location = profile['location']['name'] if profile.has_key('location') else None\\n self.bio = profile.get('bio')\\n self.relationship_status = profile.get('relationship_status')\\n self.political_views = profile.get('political')\\n self.email = profile.get('email')\\n self.website = profile.get('website')\\n self.locale = profile.get('locale')\\n self.verified = profile.get('verified')\\n self.birthday = datetime.strptime(profile['birthday'], '%m/%d/%Y') if profile.has_key('birthday') else None\\n self.timezone = profile.get('timezone', None)\\n self.quotes = profile.get('quotes', None)\\n\\n self.save()\",\n \"def get_profiles(count, credentials_type=constants.API_KEY_CREDENTIALS_TYPE, verbose=False):\\n print('getting profiles through {}...'.format(credentials_type))\\n\\n # getting profiles information\\n profile_ids = plus_api.get_unique_profile_ids(count, credentials_type=credentials_type, verbose=verbose)\\n profile_urls = list(map(plus_api.profile_id_to_google_plus_url, profile_ids))\\n # display_names, image_urls = get_profile_display_names_and_image_urls(profile_ids, credentials_type=credentials_type,\\n # verbose=verbose)\\n display_names, image_urls = get_profile_display_names_and_image_urls_with_getBatchGet(profile_ids,\\n credentials_type=credentials_type,\\n verbose=verbose)\\n\\n # saving as plain text\\n txt_filename = 'results/people-api-profiles.txt'\\n if verbose:\\n print('saving profiles as plain text to {}...'.format(txt_filename))\\n result_str = ''\\n for profile in zip(profile_ids, profile_urls, display_names, image_urls):\\n result_str += '\\\\n'.join(profile) + '\\\\n\\\\n'\\n profiles_file = open(txt_filename, 'w', encoding='utf-8')\\n profiles_file.write(result_str)\\n profiles_file.close()\\n\\n # saving as json\\n json_filename = 'results/people-api-profiles.json'\\n if verbose:\\n print('saving profiles as json to {}...'.format(json_filename))\\n result_list = []\\n json_fields = ('profileId', 'profileUrl', 'displayName', 'imageUrl')\\n for profile in zip(profile_ids, profile_urls, display_names, image_urls):\\n result_list.append(dict(zip(json_fields, profile)))\\n json_str = json.dumps(result_list)\\n profiles_file = open(json_filename, 'w', encoding='utf-8')\\n profiles_file.write(json_str)\\n profiles_file.close()\",\n \"def user_profile():\\n access_token = _request_ctx_stack.top.current_user_token\\n message_log(\\\"Got access token for user profile\\\")\\n user_profile = get_user_profile(access_token)\\n user_profile['access_token'] = access_token\\n return json.dumps(user_profile)\",\n \"def getProfileInfo(self):\\n doc = minidom.parse(urllib.urlopen(serverString + \\\"/rest/user/\\\" + self.name))\\n for element in doc.getElementsByTagName(\\\"user\\\")[0].childNodes:\\n if element.nodeType != minidom.Node.ELEMENT_NODE:\\n continue\\n elif element.tagName == \\\"status\\\" and int(element.firstChild.data) != 1:\\n raise ServerError(element.firstChild.data)\\n elif element.tagName == \\\"input\\\":\\n self.name = element.firstChild.data\\n elif element.tagName == \\\"id\\\":\\n self.id = element.firstChild.data\\n elif element.tagName == \\\"image\\\":\\n self.image = element.firstChild.data\\n elif element.tagName == \\\"tagline\\\":\\n if element.firstChild == None:\\n self.tagline = None\\n else:\\n self.tagline = element.firstChild.data\\n elif element.tagName == \\\"creation\\\":\\n self.created = datetime.datetime.strptime(element.firstChild.data[:element.firstChild.data.rfind(\\\".\\\")]+\\\".GMT\\\", \\\"%Y-%m-%d %H:%M:%S.%Z\\\")\",\n \"def registerAvatar(self, avatar):\\r\\n assert self._avatar is None\\r\\n self._avatar = avatar\",\n \"def _account_info(remote, resp):\\n g.oauth_logged_in_with_remote = remote\\n resource = get_resource(remote, resp)\\n\\n valid_roles = current_app.config.get(\\n \\\"OAUTHCLIENT_CERN_OPENID_ALLOWED_ROLES\\\",\\n OAUTHCLIENT_CERN_OPENID_ALLOWED_ROLES,\\n )\\n cern_roles = resource.get(\\\"cern_roles\\\")\\n if cern_roles is None or not set(cern_roles).issubset(valid_roles):\\n raise OAuthCERNRejectedAccountError(\\n \\\"User roles {0} are not one of {1}\\\".format(cern_roles, valid_roles),\\n remote,\\n resp,\\n )\\n\\n email = resource[\\\"email\\\"]\\n external_id = resource[\\\"cern_upn\\\"]\\n nice = resource[\\\"preferred_username\\\"]\\n name = resource[\\\"name\\\"]\\n\\n return dict(\\n user=dict(email=email.lower(), profile=dict(username=nice, full_name=name)),\\n external_id=external_id,\\n external_method=\\\"cern_openid\\\",\\n active=True,\\n )\",\n \"async def profilepicture(self, ctx):\\n if not ctx.message.mentions:\\n await ctx.send(embed=discord.Embed().set_image(\\n url=ctx.author.avatar_url))\\n else:\\n for member in ctx.message.mentions:\\n await ctx.send(embed=discord.Embed().set_image(\\n url=member.avatar_url))\",\n \"def get_profile(self):\\n endpoint = '/profile'\\n return self.get_request(endpoint)\",\n \"def build_profile(first, last, **user_info):\\n user_info['first_name'] = first\\n user_info['last_name'] = last\\n return user_info\",\n \"def build_profile(first, last, **user_info):\\n user_info['first_name'] = first\\n user_info['last_name'] = last\\n return user_info\",\n \"def update_avatar(self, url):\\n if self.avatar:\\n logging.info(f'{self} already has an avatar: {self.avatar}')\\n # TODO: check if image has been updated\\n else:\\n logging.info(f'{self} has no profile image.')\\n img_temp = NamedTemporaryFile(delete=True)\\n # TODO: Use requests instead of urllib?\\n img_temp.write(urlopen(url).read()) # noqa: S310\\n img_temp.flush()\\n self.avatar.save(f'{self.pk}', File(img_temp))\",\n \"def create(self, validated_data):\\n user_data = validated_data.pop('user')\\n user = UserSerializer.create(UserSerializer(), validated_data=user_data)\\n profile, created = Profile.objects.update_or_create(\\n user=user,\\n avatar=validated_data.pop('avatar'),\\n biography=validated_data.pop('biography'),\\n link=validated_data.pop('link') \\n )\\n return profile\",\n \"def GetAvatar(self):\\n\\n return self.__GetJsonOrNone(\\\"/users/\\\"+self.userName+\\\"/avatar\\\", False)\",\n \"def get_profile():\\n logger.debug(\\\"entering function get_profile\\\")\\n response = read_user_profile()\\n logger.debug(\\\"exiting function get_profile\\\")\\n return jsonify(response)\",\n \"def get(self, request):\\n current_user = request.user\\n user_profile = self.serializer_class(current_user.profile)\\n return Response({\\n \\\"message\\\": PROFILE_MSGS['MY_PROFILE'],\\n \\\"profile\\\": user_profile.data\\n }, status=status.HTTP_200_OK)\",\n \"def display_profile(self):\\n print(f\\\"Id: {self._id}\\\")\\n print(f\\\"username: {self.username}\\\")\\n print(f\\\"name: {self.name}\\\")\\n print(f\\\"contact: {self.contact}\\\")\\n print(f\\\"address: {self.address}\\\")\",\n \"def account(self):\\n provider_name = self.provider_entry.get_text()\\n provider = Provider.get_by_name(provider_name)\\n\\n # Create a new provider if we don't find one\\n if not provider:\\n provider_image = self.provider_image.image\\n provider_website = self.provider_website_entry.get_text()\\n provider = Provider.create(provider_name, provider_website, None,\\n provider_image)\\n # Update the provider image if it changed\\n elif provider and self.provider_image.image != provider.image:\\n provider.update(image=self.provider_image.image)\\n\\n account = {\\n \\\"username\\\": self.account_name_entry.get_text(),\\n \\\"provider\\\": provider\\n }\\n if not self.props.is_edit:\\n # remove spaces\\n token = self.token_entry.get_text()\\n account[\\\"token\\\"] = \\\"\\\".join(token.split())\\n return account\",\n \"def makeProfile(request):\\n upr = UserProfile()\\n upr.user = request.user\\n upr.image = \\\"images/no-pic.png\\\"\\n upr.save()\",\n \"def handle_callback(self, response):\\n\\n client = BitbucketClient(access_token=response['access_token'])\\n user_info = client.user()\\n\\n return {\\n 'provider_id': user_info['uuid'],\\n 'profile_url': user_info['links']['html']['href'],\\n 'display_name': user_info['username']\\n }\",\n \"def profile():\\n from flickrAPI import FlickrAPI\\n #flickr = FlickrAPI(key=session['resource_owner_key'], secret=session['resource_owner_secret'])\\n flickr = FlickrAPI(key=request.cookies.get('oauth_token'), secret=request.cookies.get('oauth_token_secret'))\\n faves = flickr.favorites_getList(user_id=\\\"44124394781@N01\\\", page=1, per_page=5, extras='owner_name')\\n return str(faves)\",\n \"def avatar(self) -> str:\\n\\t\\tif self.hasAvatar:\\n\\t\\t\\treturn f'https://avatars.atelier801.com/{self.id % 10000}/{self.id}.jpg'\\n\\n\\t\\t# default avatar\\n\\t\\treturn 'https://avatars.atelier801.com/0/0.jpg'\",\n \"def load_extra_data(backend, details, response, uid, user, social_user=None,\\n *args, **kwargs):\\n social_user = social_user or \\\\\\n UserSocialAuth.get_social_auth(backend.name, uid)\\n if social_user:\\n extra_data = backend.extra_data(user, uid, response, details)\\n extra_data_field_name = \\\"{0}_extra_data\\\".format(backend.name)\\n social_user_extra_data = getattr(social_user, extra_data_field_name)\\n if extra_data and social_user_extra_data != extra_data:\\n if social_user_extra_data:\\n social_user_extra_data.update(extra_data)\\n else:\\n social_user_extra_data = extra_data\\n\\n #Getting the access token\\n access_token_field_name = \\\"{0}_access_token\\\".format(backend.name)\\n setattr(social_user, access_token_field_name, extra_data['access_token'])\\n\\n #Storing extra data\\n social_user_extra_data.pop('access_token', None)\\n social_user_extra_data.pop('id', None)\\n setattr(social_user, extra_data_field_name, social_user_extra_data)\\n\\n social_user.save()\\n return {'social_user': social_user}\",\n \"def build_profile(self, update, context, phone=None, raw_text=None):\\n user = update.effective_user\\n chat_id = update.effective_chat.id\\n log.info(\\\"PROFILE from %s `%s`\\\", chat_id, raw_text)\\n # import pdb; pdb.set_trace()\\n\\n # If necessary, create the part of the state that holds data about registration procedures\\n if \\\"registrations\\\" not in context.bot_data:\\n context.bot_data[\\\"registrations\\\"] = {}\\n\\n if chat_id not in context.bot_data[\\\"registrations\\\"]:\\n # create a new user profile and add it to the bot's state, so we can populate it\\n # as we ask the user to provide info about themselves; keep in mind that it is an ORDERED dict, we'll\\n # rely on this later!\\n profile = OrderedDict(\\n {\\n c.PROFILE_FIRST_NAME: user.first_name, # may be empty at first\\n c.PROFILE_LAST_NAME: user.last_name, # may be empty at first\\n c.PROFILE_AVAILABILITY: None,\\n c.PROFILE_ACTIVITIES: [],\\n c.PROFILE_PHONE: phone,\\n c.PROFILE_EMAIL: None,\\n }\\n )\\n\\n if not phone.startswith(c.LOCAL_PREFIX):\\n # If the Telegram phone number is not a local number (i.e. it was registered abroad), we're moving it\\n # to a different attribute, and clearing the original one, such that later in this function we shall\\n # ask for a local phone number\\n log.debug(\\\"Phone number is foreign, will ask for a local one\\\")\\n profile[c.PROFILE_PHONE_FOREIGN] = phone\\n profile[c.PROFILE_PHONE] = None\\n\\n context.bot_data[\\\"registrations\\\"][chat_id] = profile\\n else:\\n profile = context.bot_data[\\\"registrations\\\"][chat_id]\\n\\n for key, value in profile.items():\\n if not value:\\n # a part of the profile is empty, maybe we should ask about it?\\n if raw_text:\\n # This seems to be yet another call of this function, so raw_text contains the answer to the\\n # question asked earlier - let's populate it.\\n # NOTE that we use an OrderedDict when building the profile, so we know for sure this answer\\n # goes to that particular question (i.e. key in the dict)\\n profile[key] = raw_text\\n raw_text = None\\n continue\\n\\n # if we got this far, we stumbled upon the next missing part of the profile\\n context.user_data[\\\"state\\\"] = c.State.EXPECTING_PROFILE_DETAILS\\n\\n self.updater.bot.send_message(\\n chat_id=chat_id,\\n text=c.PROFILE_QUESTIONS[key],\\n parse_mode=ParseMode.MARKDOWN_V2,\\n )\\n\\n if key == c.PROFILE_ACTIVITIES:\\n # this is a special case, because we'll send them an interactive keyboard with options to chose from\\n self.confirm_activities(update, context)\\n return\\n\\n return\\n\\n # if we got this far, it means the profile is complete, inform the user about it\\n self.updater.bot.send_message(\\n chat_id=chat_id, text=c.MSG_ONBOARD_NEXT_STEPS, parse_mode=ParseMode.MARKDOWN,\\n )\\n\\n # and the backend, but first let's augment the profile with more data\\n profile[c.PROFILE_CHAT_ID] = chat_id\\n self.backend.register_pending_volunteer(profile)\\n context.user_data[\\\"state\\\"] = c.State.AVAILABLE\\n\\n # remove if from the state, because we don't need it anymore\\n del context.bot_data[\\\"registrations\\\"][chat_id]\\n\\n # Also get rid of this user's individual keyboard for assitance activities\\n context.user_data.pop(\\\"assist_keyboard\\\", None)\",\n \"def getprofile(self, *args, **kwargs):\\n return _image.image_getprofile(self, *args, **kwargs)\",\n \"def head(self, entity):\\n link = '<{url}>; rel=\\\"https://tent.io/rels/profile\\\"'.format(\\n url=url_for('entity.profile', entity=entity.name, _external=True))\\n resp = jsonify(entity.to_json())\\n resp.headers['Link'] = link\\n \\n return resp\",\n \"def build_profile(first, last, **user_info):\\n\\n user_info['first_name'] = first\\n user_info['last_name'] = last\\n return user_info\",\n \"def _getProfileFromUser(self):\\n # Make sure user is authenticated\\n user = endpoints.get_current_user()\\n if not user:\\n raise endpoints.UnauthorizedException('Authorization required')\\n # Get Profile from datastore\\n user_id = user.email()\\n p_key = ndb.Key(Profile, user_id)\\n profile = p_key.get()\\n # Create new Profile if not there\\n if not profile:\\n profile = Profile(\\n key = p_key,\\n displayName = user.nickname(),\\n mainEmail= user.email(),\\n teeShirtSize = str(TeeShirtSize.NOT_SPECIFIED),\\n )\\n profile.put()\\n return profile\",\n \"def get_user_details(self, response):\\n # Build the username with the team $username@$team_url\\n # Necessary to get unique names for all of slack\\n username = response.get('user')\\n if self.setting('USERNAME_WITH_TEAM', True):\\n match = re.search(r'//([^.]+)\\\\.slack\\\\.com', response['url'])\\n username = '{0}@{1}'.format(username, match.group(1))\\n\\n out = {'username': username}\\n if 'profile' in response:\\n out.update({\\n 'email': response['profile'].get('email'),\\n 'fullname': response['profile'].get('real_name'),\\n 'first_name': response['profile'].get('first_name'),\\n 'last_name': response['profile'].get('last_name'),\\n 'team_name': response.get('team_name')\\n })\\n return out\",\n \"def get_profile_picture_url(cls, filename):\\n if filename is None:\\n return None\\n profile_picture = bucket.blob('images/users/'+filename)\\n if profile_picture.exists():\\n profile_picture.make_public()\\n return profile_picture.public_url\\n return None\",\n \"def profile(request):\\n profile = request.user.profile\\n return render(request, 'accounts/profile.html', {'profile': profile})\",\n \"def save_profile(self, request):\\n return self.profile_service.do_profile(request)\",\n \"def add_gamer_profile():\\n user_id = current_user.get().id\\n user = User.objects.get_or_404(id=user_id)\\n\\n # should we validate object_id here?\\n\\n # auto populate with g.formdata\\n game_form = GamerForm(g.formdata)\\n game_form.populate_obj(user.gamer)\\n\\n if request.files and request.files['avatar']:\\n filename = Avatars.save(request.files['avatar'])\\n user.gamer.avatar = Avatars.url(filename)\\n\\n # manually handle games\\n user.gamer.games = [Game(id=_id)\\n for _id in request.form.getlist('games')]\\n\\n user.save()\\n return user\",\n \"def resolve_avatar(self, info):\\n if self.avatar:\\n self.avatar = info.context.build_absolute_uri(self.avatar.url)\\n return self.avatar\",\n \"def get_gallery_profile(self):\\n url = (\\\"https://api.imgur.com/3/account/{0}/\\\"\\n \\\"gallery_profile\\\".format(self.name))\\n return self._imgur._send_request(url)\",\n \"def getAvatarInfo(self):\\n return \\\", \\\".join(self._get_avatar_info())\",\n \"def _profile_to_InstagramUser(profile: Dict[str, Any]) -> _InstagramUser:\\n # Navigate to the user JSON that is coincidentally used by the provided API methods\\n user = profile['users'][0]['user']\\n\\n # Simply build our InstagramUser, as the user JSON is the same\\n return _InstagramUser(user)\",\n \"def update_info(self):\\n\\n r = requests.get(self.url + 'getMe')\\n if r.status_code == 200:\\n response = json.loads(r.text)\\n if response['ok']:\\n bot_info = response['result']\\n self.user_id = bot_info['id']\\n self.first_name = bot_info['first_name']\\n if 'last_name' in bot_info:\\n self.last_name = bot_info['last_name']\\n if 'username' in bot_info:\\n self.username = bot_info['username']\\n else:\\n raise TelegramError('The result was not \\\"ok\\\"')\\n else:\\n raise TelegramError('Did not get a 200 response', r.status_code)\",\n \"def create_or_update_profile(sender, instance, created, **kwargs):\\n if created:\\n Profile.objects.get_or_create(user=instance)\\n instance.profile.save()\",\n \"def google_extra_values(sender, user, response, details, **kwargs):\\n LOGGER.debug('profile.models.google_extra_values')\\n user_info_url = \\\"https://www.googleapis.com/oauth2/v1/userinfo\\\"\\n\\n data = {'access_token': response.get('access_token', ''), 'alt': 'json'}\\n params = urlencode(data)\\n try:\\n request = Request(user_info_url + '?' + params, headers={'Authorization': params})\\n result = simplejson.loads(urlopen(request).read())\\n\\n user.last_name = result.get('family_name', '')\\n user.first_name = result.get('given_name', '')\\n profile = user.profile\\n profile.gender = result.get('gender', '')\\n profile.image_url = result.get('picture', '')\\n profile.url = result.get('link', '')\\n\\n profile.save()\\n except:\\n pass\\n\\n return True\",\n \"def player_avatar(player_obj):\\n avatar = \\\"http://www.gravatar.com/avatar/%s.jpg?d=monsterid\\\" % md5(player_obj.user.email).hexdigest()\\n\\n return avatar\",\n \"def _username_to_profile(self, username: str) -> Dict[str, Any]:\\n\\n base_url = self.keys.pre_profile + username + self.keys.rank_token + self.keys.post_profile\\n\\n # Build the page source url for the given user's account\\n con = urllib.request.urlopen(base_url)\\n user_profile = con.read().decode('utf-8')\\n\\n # Convert the webpage to a profile JSON\\n profile: dict = json.loads(str(user_profile))\\n return profile\",\n \"def get_user_details(self, response):\\n\\n kaccount_email = \\\"\\\"\\n kakao_account = response.get(\\\"kakao_account\\\", \\\"\\\")\\n if kakao_account:\\n kaccount_email = kakao_account.get(\\\"email\\\", \\\"\\\")\\n properties = response.get(\\\"properties\\\", \\\"\\\")\\n nickname = properties.get(\\\"nickname\\\") if properties else \\\"\\\"\\n return {\\n \\\"username\\\": nickname,\\n \\\"email\\\": kaccount_email,\\n \\\"fullname\\\": nickname,\\n \\\"first_name\\\": nickname[1:] if nickname else \\\"\\\",\\n \\\"last_name\\\": nickname[0] if nickname else \\\"\\\",\\n }\",\n \"def save_profile_picture(backend, user, response, *args, **kwargs):\\n profile = Profile.objects.filter(user=user).first()\\n if profile is None:\\n profile = Profile(user=user)\\n\\n if \\\"locale\\\" in response:\\n profile.locale = response[\\\"locale\\\"]\\n\\n if \\\"picture\\\" in response:\\n with TempDir() as d:\\n _name, ext = splitext(response[\\\"picture\\\"])\\n r = requests.get(response[\\\"picture\\\"], stream=True)\\n\\n if r.ok:\\n temp_file = d / f\\\"{user.pk}{ext or '.jpg'}\\\"\\n\\n with open(temp_file, \\\"wb\\\") as fd:\\n for chunk in r.iter_content(chunk_size=128):\\n fd.write(chunk)\\n\\n with temp_file.open(\\\"rb\\\") as fd:\\n profile.picture.save(\\n f\\\"{user.pk}{ext or '.jpg'}\\\", File(fd), save=False\\n )\\n\\n profile.save()\",\n \"def set_user_profile_picture(user_id, file_name):\\n\\n user = User.query.get(user_id)\\n \\n user.profile_picture = file_name\\n db.session.commit()\",\n \"def _parse_profile_data (self, netflix_page_data):\\n profiles = {};\\n important_fields = [\\n 'profileName',\\n 'isActive',\\n 'isAccountOwner',\\n 'isKids'\\n ]\\n # values are accessible via dict (sloppy parsing successfull)\\n if type(netflix_page_data) == dict:\\n for profile_id in netflix_page_data.get('profiles'):\\n if self._is_size_key(key=profile_id) == False and type(netflix_page_data['profiles'][profile_id]) == dict and netflix_page_data['profiles'][profile_id].get('avatar', False) != False:\\n profile = {'id': profile_id}\\n for important_field in important_fields:\\n if important_field == 'profileName':\\n profile.update({important_field: HTMLParser.HTMLParser().unescape(netflix_page_data['profiles'][profile_id]['summary'][important_field]).encode('utf8')})\\n else:\\n profile.update({important_field: netflix_page_data['profiles'][profile_id]['summary'][important_field]})\\n avatar_base = netflix_page_data['nf'].get(netflix_page_data['profiles'][profile_id]['summary']['avatarName'], False);\\n avatar = 'https://secure.netflix.com/ffe/profiles/avatars_v2/320x320/PICON_029.png' if avatar_base == False else avatar_base['images']['byWidth']['320']['value']\\n profile.update({'avatar': avatar, 'isFirstUse': False})\\n profiles.update({profile_id: profile})\\n return profiles\\n\\n # values are stored in lists (returned from JS parser)\\n # TODO: get rid of this christmas tree of doom\\n for item in netflix_page_data:\\n if 'hasViewedRatingWelcomeModal' in dict(item).keys():\\n for profile_id in item:\\n if self._is_size_key(key=profile_id) == False and type(item[profile_id]) == dict and item[profile_id].get('avatar', False) != False:\\n profile = {'id': profile_id}\\n for important_field in important_fields:\\n profile.update({important_field: item[profile_id]['summary'][important_field]})\\n avatar_base = item['nf'].get(item[profile_id]['summary']['avatarName'], False);\\n avatar = 'https://secure.netflix.com/ffe/profiles/avatars_v2/320x320/PICON_029.png' if avatar_base == False else avatar_base['images']['byWidth']['320']['value']\\n profile.update({'avatar': avatar})\\n profiles.update({profile_id: profile})\\n return profiles\",\n \"def save_user_profile(instance, **_):\\n instance.profile.save()\",\n \"def build_profile(first, last, **user_info):\\n profile = {}\\n profile['first_name'] = first\\n profile['last_name'] = last\\n for key, value in user_info.items():\\n profile[key] = value\\n return profile\",\n \"def build_profile(first, last, **user_info):\\n profile = {}\\n profile['first_name'] = first\\n profile['last_name'] = last\\n for key, value in user_info.items():\\n profile[key] = value\\n return profile\",\n \"def update_user_profile(id):\\n token = request.json['token']\\n u = user.User.query.filter(user.User.token == token).first()\\n if u is None:\\n abort(404)\\n if u.id != id:\\n print \\\"user id is wrong.\\\" #TODO: Support log system\\n abort(500)\\n u.name = request.json['name']\\n u.nickname = request.json['nickname']\\n u.company = request.json['nickname']\\n with store_context(fs_store):\\n with open(files.path(request.json['header'])) as f:\\n u.header_icon.from_file(f)\\n db.session.merge(u)\\n db.session.commit()\\n db.session.merge(u)\\n db.session.commit()\\n return jsonify(u.to_dict())\",\n \"def _request_user_info(credentials):\\n http = httplib2.Http()\\n\\n # User information stored here\\n credentials.authorize(http)\\n resp, content = http.request('https://www.googleapis.com/plus/v1/people/me')\\n\\n if resp.status != 200:\\n current_app.logger.error(\\\"Error while obtaining user profile: %s\\\" % resp)\\n return None\\n\\n # Check whether user is authenticating with the allowed domain.\\n if (current_app.config['RESTRICT_DOMAIN'] is True and \\n credentials.id_token.get('hd') != current_app.config['REQUIRED_DOMAIN']):\\n\\n # Replace with logging for a real app\\n print(\\\"\\\\n------------------------------------------------------\\\")\\n print(\\\"User attempted to authenticate with disallowed domain.\\\")\\n print(\\\"------------------------------------------------------\\\\n\\\")\\n\\n # User information deleted here\\n oauth2.storage.delete()\\n return None\\n\\n # Stores the users information in the session profile dictionary\\n session['profile'] = json.loads(content.decode('utf-8'))\\n\\n # Remove this in production. It's here so you can see what information is stored.\\n print(\\\"\\\\n------------------------------------------------------\\\")\\n print(\\\"SESSION PROFILE INFORMATION\\\")\\n print(\\\"------------------------------------------------------\\\")\\n for k,v in session['profile'].items():\\n print(k,\\\"--->\\\",v)\\n print(\\\"------------------------------------------------------\\\\n\\\")\",\n \"def prefetch_login(self):\\n self.profiles = []\\n email = self.credentials.get('email', '')\\n password = self.credentials.get('password', '')\\n if email != '' and password != '':\\n if self.netflix_session.is_logged_in(account=self.credentials):\\n refresh_session = self.netflix_session.refresh_session_data(\\n account=self.credentials)\\n if refresh_session:\\n self.profiles = self.netflix_session.profiles\\n else:\\n if self.netflix_session.login(account=self.credentials):\\n self.profiles = self.netflix_session.profiles\\n self.kodi_helper.set_esn(self.netflix_session.esn)\",\n \"def get_user_profile(self, username: str = None) -> Profile:\\n if username:\\n res: dict[str, Any] = self.api.users.get(user_ids=username).pop()\\n else:\\n res = self.api.users.get().pop()\\n return VkontakteConnector.build_profile(res)\",\n \"async def steam_avatar(message):\\n s = message.content.strip()\\n id = await parse_steam_id(s)\\n profile = await steam_community.steam_profile(id.to_64(), id64=True)\\n if profile.avatar:\\n return profile.avatar\\n else:\\n raise CommandError(\\\"The user has no avatar.\\\")\",\n \"def __str__(self):\\n profile = {\\n 'address': self.address,\\n 'bio': self.bio,\\n 'website': self.website,\\n 'hireable': self.hireable,\\n 'travel_radius': self.travel_radius,\\n 'phone number': self.phone_number,\\n 'camera type': self.camera_type,\\n 'type of photography': self.photography_type\\n }\\n return str(self.user.username)\",\n \"def build_profile(first, last, **user_info):\\n profile = {}\\n profile['first_name'] = first\\n profile['last_name'] = last\\n for key, value in user_info.items():\\n profile[key] = value\\n return(profile)\",\n \"def build_profile(first, last, **user_info):\\n profile = {}\\n profile['first_name'] = first\\n profile['last_name'] = last\\n for key, value in user_info.items():\\n profile[key] = value\\n return(profile)\",\n \"def get_author_data(uid, gravatar_on_right_side=True, linked_with_users_page=True, profile_picture_size=20):\\n db_user = DBDiscussionSession.query(User).get(int(uid))\\n if not db_user:\\n return None, 'Missing author with uid ' + str(uid), False\\n\\n nick = db_user.global_nickname\\n img_src = get_profile_picture(db_user, profile_picture_size)\\n link_begin = ''\\n link_end = ''\\n if linked_with_users_page:\\n link_begin = '<a href=\\\"/user/{}\\\" title=\\\"{}\\\">'.format(db_user.uid, nick)\\n link_end = '</a>'\\n\\n side = 'left' if gravatar_on_right_side else 'right'\\n img = '<img class=\\\"img-circle\\\" src=\\\"{}\\\" style=\\\"padding-{}: 0.3em\\\">'.format(img_src, side)\\n\\n if gravatar_on_right_side:\\n return db_user, '{}{}{}{}'.format(link_begin, nick, img, link_end), True\\n else:\\n return db_user, '{}{}{}{}'.format(link_begin, img, nick, link_end), True\",\n \"def get_user_details(self, response):\\n name = response.get(\\\"name\\\")\\n return {\\n \\\"username\\\": str(response.get(\\\"account_id\\\")),\\n \\\"email\\\": response.get(\\\"email\\\"),\\n \\\"fullname\\\": name.get(\\\"display_name\\\"),\\n \\\"first_name\\\": name.get(\\\"given_name\\\"),\\n \\\"last_name\\\": name.get(\\\"surname\\\"),\\n }\",\n \"def build_profile(first, last, **user_info):\\n profile = {}\\n profile['first_name'] = first\\n profile['last_name'] = last\\n for key, value in user_info.items():\\n profile[key] = value\\n return profile\",\n \"def build_profile(first, last, **user_info):\\n profile = {}\\n profile['first_name'] = first\\n profile['last_name'] = last\\n for key, value in user_info.items():\\n profile[key] = value\\n return profile\",\n \"def build_profile(first, last, **user_info):\\n profile = {}\\n profile['first_name'] = first\\n profile['last_name'] = last\\n for key, value in user_info.items():\\n profile[key] = value\\n return profile\",\n \"async def get_my_profile_info_async(\\n namespace: Optional[str] = None,\\n x_additional_headers: Optional[Dict[str, str]] = None,\\n **kwargs\\n):\\n if namespace is None:\\n namespace, error = get_services_namespace()\\n if error:\\n return None, error\\n request = GetMyProfileInfo.create(\\n namespace=namespace,\\n )\\n return await run_request_async(\\n request, additional_headers=x_additional_headers, **kwargs\\n )\",\n \"def google_profile_extra_details(backend, details, response, user=None, is_new=False,\\n *args, **kwargs):\\n profile = {}\\n\\n logger.info('Pipeline logging in with google-oauth2')\\n\\n if user is not None:\\n client_profile = user.profile\\n # save all updated fields\\n if isinstance(response, dict):\\n logger.info('Pipeline.google updating client.profile_data from google %s - %s' % (user, response,) )\\n\\n client_profile.profile_data.update(response) \\n\\n client_profile.save(update_fields=['profile_data'])\",\n \"def profile_pic(self, client_file_storage):\\n\\n # If we already have a profile picture, remove it\\n if self.profile_pic_filename:\\n filepath = os.path.join(\\n current_app.config['UPLOADED_IMAGES_DEST'],\\n self.profile_pic_filename)\\n os.remove(filepath)\\n self.profile_pic_filename = None\\n self.profile_pic_url = None\\n\\n # This uploads & saves the file on the server\\n # NOTE: It uses the secure_filename function...\\n server_filename = images.save(client_file_storage)\\n\\n # Generate the URL to this file\\n url = images.url(server_filename)\\n\\n # Store information with the user\\n self.profile_pic_filename = server_filename\\n self.profile_pic_url = url\",\n \"def __str__(self):\\n return self.user.username + \\\"'s Profile\\\"\",\n \"def _get_linkedin_accounts(self, linkedin_access_token):\\n response = requests.get(\\n 'https://api.linkedin.com/v2/me?projection='\\n + '(id,localizedLastName,localizedFirstName,'\\n + 'profilePicture(displayImage~:playableStreams))',\\n headers={\\n 'Authorization': 'Bearer ' + linkedin_access_token,\\n 'cache-control': 'no-cache',\\n 'X-Restli-Protocol-Version': '2.0.0'\\n }\\n ).json()\\n\\n if ('id' in response and 'localizedLastName' in response\\n and 'localizedFirstName' in response):\\n linkedin_account_id = 'urn:li:person:' + response['id']\\n\\n try:\\n image_url = response['profilePicture']['displayImage~']['elements'][0]['identifiers'][0]['identifier']\\n linkedin_profile_image = base64.b64encode(requests.get(image_url).content)\\n except Exception:\\n linkedin_profile_image = ''\\n\\n # TODO - STD: add each companies page\\n return [{\\n 'name': response['localizedLastName'] + ' ' + response['localizedFirstName'],\\n 'linkedin_account_id': linkedin_account_id,\\n 'linkedin_access_token': linkedin_access_token,\\n 'image': linkedin_profile_image\\n }]\\n\\n return []\",\n \"def get_profile(tag, platform=\\\"pc\\\", region=\\\"eu\\\"):\\n #\\n try:\\n context = ssl._create_unverified_context()\\n profile = json.load(\\n const.codec(urlopen(const.URL + platform + \\\"/\\\" + region + \\\"/\\\" + tag + \\\"/profile\\\", context=context)))\\n #\\n if \\\"error\\\" in profile:\\n raise BattleTagNotFound(profile['error'])\\n exit(1)\\n #\\n result = pr.Profile(profile['data']['username'],\\n profile['data']['level'],\\n profile['data']['games']['quick']['wins'],\\n profile['data']['games']['competitive']['wins'],\\n profile['data']['games']['competitive']['lost'],\\n profile['data']['playtime']['quick'],\\n profile['data']['playtime']['competitive'],\\n profile['data']['avatar'],\\n profile['data']['competitive']['rank'])\\n return result\\n except urllib.error.URLError as e:\\n print(\\\"An error occurred when fetching stats\\\\n\\\" + e)\\n exit(1)\\n except Exception as e:\\n print(\\\"An error occurred:\\\\n \\\" + str(e))\\n exit(1)\",\n \"def save(self, profile_callback=None):\\n\\n # First, save the parent form\\n new_user = super(BodbRegistrationForm, self).save(profile_callback=profile_callback)\\n\\n # Update user with first, last names\\n new_user.first_name = self.cleaned_data['first_name']\\n new_user.last_name = self.cleaned_data['last_name']\\n new_user.save()\\n\\n # Update profile with affiliation\\n profile = new_user.get_profile()\\n profile.affiliation = self.cleaned_data['affiliation']\\n profile.save()\\n\\n cache.set('%d.profile' % new_user.id, profile)\\n\\n return new_user\",\n \"def profiles():\\n images = get_uploaded_images()\\n records = db.session.query(UserProfile).all()\\n return render_template('profiles.html', images=images, records =records)\",\n \"def getProfile(self):\\n # GET /profile\\n debugMain('getProfile')\\n return self._genericGet('/profile')\",\n \"def get_user_profile(self):\\n return self.user.profile\",\n \"def get(self, username):\\n\\t\\tdb = getattr(g, 'db', None)\\n\\n\\t\\tqry = \\\"SELECT username,email,active,steamid FROM\\\\\\n\\t\\t\\tprofiles WHERE username = %s;\\\"\\n\\t\\twith db as cursor:\\n\\t\\t\\tcursor.execute(qry, (username,))\\n\\n\\t\\treturn {'profile':cursor.fetchone()}\",\n \"def profile():\\n token = request.json['token']\\n u = user.User.query.filter(user.User.token == token).first()\\n if u is None:\\n abort(404)\\n return jsonify(u.to_dict())\"\n]","isConversational":false},"negative_scores":{"kind":"list like","value":["0.62179977","0.61199266","0.61171144","0.6097713","0.60976326","0.60648817","0.605172","0.60389084","0.6031145","0.60300606","0.60167927","0.6015737","0.6003023","0.5930042","0.5929104","0.5922604","0.59004813","0.59002423","0.58768433","0.58611965","0.58558357","0.58234036","0.58162713","0.5813563","0.5743932","0.5711935","0.5702799","0.5681541","0.5680092","0.56783867","0.56743634","0.56511","0.5649045","0.5640262","0.56206745","0.56180173","0.56180173","0.5617542","0.5605671","0.55969334","0.55914146","0.55909824","0.5586745","0.55848545","0.557691","0.55758595","0.55747646","0.556392","0.55616677","0.55609655","0.5560606","0.5558067","0.5555299","0.554781","0.5545324","0.5543319","0.5533971","0.55332166","0.55324984","0.5527128","0.5520784","0.55195165","0.55182767","0.5516057","0.5501652","0.5500739","0.5496724","0.54960644","0.5490521","0.5477958","0.54558736","0.5444795","0.54423726","0.54397976","0.54397976","0.5438897","0.54382765","0.54318035","0.54311377","0.5424502","0.5424179","0.5423448","0.5423448","0.541164","0.5409118","0.5407208","0.5407208","0.5407208","0.54056674","0.5390446","0.5387945","0.53864557","0.53779626","0.5373978","0.53732765","0.5370817","0.53683716","0.536679","0.5366611","0.5351645"],"string":"[\n \"0.62179977\",\n \"0.61199266\",\n \"0.61171144\",\n \"0.6097713\",\n \"0.60976326\",\n \"0.60648817\",\n \"0.605172\",\n \"0.60389084\",\n \"0.6031145\",\n \"0.60300606\",\n \"0.60167927\",\n \"0.6015737\",\n \"0.6003023\",\n \"0.5930042\",\n \"0.5929104\",\n \"0.5922604\",\n \"0.59004813\",\n \"0.59002423\",\n \"0.58768433\",\n \"0.58611965\",\n \"0.58558357\",\n \"0.58234036\",\n \"0.58162713\",\n \"0.5813563\",\n \"0.5743932\",\n \"0.5711935\",\n \"0.5702799\",\n \"0.5681541\",\n \"0.5680092\",\n \"0.56783867\",\n \"0.56743634\",\n \"0.56511\",\n \"0.5649045\",\n \"0.5640262\",\n \"0.56206745\",\n \"0.56180173\",\n \"0.56180173\",\n \"0.5617542\",\n \"0.5605671\",\n \"0.55969334\",\n \"0.55914146\",\n \"0.55909824\",\n \"0.5586745\",\n \"0.55848545\",\n \"0.557691\",\n \"0.55758595\",\n \"0.55747646\",\n \"0.556392\",\n \"0.55616677\",\n \"0.55609655\",\n \"0.5560606\",\n \"0.5558067\",\n \"0.5555299\",\n \"0.554781\",\n \"0.5545324\",\n \"0.5543319\",\n \"0.5533971\",\n \"0.55332166\",\n \"0.55324984\",\n \"0.5527128\",\n \"0.5520784\",\n \"0.55195165\",\n \"0.55182767\",\n \"0.5516057\",\n \"0.5501652\",\n \"0.5500739\",\n \"0.5496724\",\n \"0.54960644\",\n \"0.5490521\",\n \"0.5477958\",\n \"0.54558736\",\n \"0.5444795\",\n \"0.54423726\",\n \"0.54397976\",\n \"0.54397976\",\n \"0.5438897\",\n \"0.54382765\",\n \"0.54318035\",\n \"0.54311377\",\n \"0.5424502\",\n \"0.5424179\",\n \"0.5423448\",\n \"0.5423448\",\n \"0.541164\",\n \"0.5409118\",\n \"0.5407208\",\n \"0.5407208\",\n \"0.5407208\",\n \"0.54056674\",\n \"0.5390446\",\n \"0.5387945\",\n \"0.53864557\",\n \"0.53779626\",\n \"0.5373978\",\n \"0.53732765\",\n \"0.5370817\",\n \"0.53683716\",\n \"0.536679\",\n \"0.5366611\",\n \"0.5351645\"\n]","isConversational":false},"document_score":{"kind":"string","value":"0.76166373"},"document_rank":{"kind":"string","value":"0"}}},{"rowIdx":93287,"cells":{"query":{"kind":"string","value":"On action created, fan out notifications"},"document":{"kind":"string","value":"def on_created_action(sender, instance: models_actstream.Action, created, **kwargs):\n if not created:\n return\n\n # If content_type is Post or Profile\n if (\n instance.action_object_content_type\n == ContentType.objects.get_for_model(dillo.models.posts.Post)\n and instance.verb == 'posted'\n ) or (\n instance.action_object_content_type\n == ContentType.objects.get_for_model(dillo.models.profiles.Profile)\n and instance.verb == 'updated their reel'\n ):\n dillo.tasks.feeds.establish_time_proximity(instance)\n\n log.debug('Creating background fanout operation for action %i' % instance.id)\n dillo.tasks.feeds.activity_fanout_to_feeds(instance.id)"},"metadata":{"kind":"string","value":"{\n \"objective\": {\n \"self\": [],\n \"paired\": [],\n \"triplet\": [\n [\n \"query\",\n \"document\",\n \"negatives\"\n ]\n ]\n }\n}"},"negatives":{"kind":"list like","value":["def action_defined(sender, instance, created, raw, using, **kwargs):\n if created:\n raw_hook_event.send(\n sender=None,\n event_name=\"action_defined\",\n instance=instance,\n payload=ActionSerializer(instance).data,\n user=instance.team,\n )","def onAction(*args):","def onAction(*args):","def onAction(*args):","def onAction(*args):","def publish_action(self, action):\n raise NotImplementedError","def on_action(self, connection, event):\r\n print('[{}] OnAction from {}' .format(event.type.upper(), event.source))","def send_actions(self, actions):\n pass","def actions():\n pass","def on_publish(client, userdata, mid):\n print(\"Message Published.\")","def onActionTaken(self, agent):\n\n pass","def post_activities():\n pass","def action(self):\n pass","def action(self):\n pass","def on_publish( client, userdata, mid ):\n logging.info( \"Data published successfully.\" )","async def send_event_created(self, action_id: int):\n async with self.pg.acquire() as conn:\n data = await conn.fetchrow(\n \"\"\"\n SELECT a.company AS company_id, u.role AS host_role, u.id AS host_user_id,\n full_name(u.first_name, u.last_name, u.email) AS host_name,\n e.id AS event_id, e.name AS event_name,\n (e.start_ts AT TIME ZONE e.timezone)::date AS event_date,\n cat.name AS cat_name, cat.slug AS cat_slug,\n event_link(cat.slug, e.slug, e.public, $2) AS event_link\n FROM actions AS a\n JOIN users AS u ON a.user_id = u.id\n JOIN events AS e ON a.event = e.id\n JOIN categories AS cat ON e.category = cat.id\n WHERE a.id=$1\n \"\"\",\n action_id,\n self.settings.auth_key,\n )\n\n link = f'/dashboard/events/{data[\"event_id\"]}/'\n ctx = dict(\n summary='{host_name} created an event \"{event_name}\"'.format(**data),\n details=(\n 'Event \"{event_name}\" ({cat_name}) created by \"{host_name}\" ({host_role}), '\n 'click the link below to view the event.'\n ).format(**data),\n action_label='View Event',\n action_link=link,\n )\n users = [\n UserEmail(id=r['id'], ctx=ctx)\n for r in await conn.fetch(\"SELECT id FROM users WHERE role='admin' AND company=$1\", data['company_id'])\n ]\n await self.send_emails.direct(data['company_id'], Triggers.admin_notification, users)\n if data['host_role'] != 'admin':\n ctx = {\n 'event_link': data['event_link'],\n 'event_dashboard_link': link,\n 'event_name': data['event_name'],\n 'event_date': format_dt(data['event_date']),\n 'category_name': data['cat_name'],\n is_cat(data['cat_slug']): True,\n }\n await self.send_emails.direct(\n data['company_id'], Triggers.event_host_created, [UserEmail(data['host_user_id'], ctx)]\n )","def take_action(self, *args, **kwargs):\r\n pass","def on_publish(unused_client, unused_userdata, unused_mid):\n print('on_publish')\n status_light.on()","def write(self, notification):","def on_publish(unused_client, unused_userdata, unused_mid):\n\tprint('on_publish')","def onShowed(self):\n self.parent.actionTagTwo=\"\"\n pass","def broadcast_publisher_info(sender, instance, created, **kwargs):\n if created:\n # notify creation to generic channel\n broadcast('pulse', {\n 'status': instance.status,\n 'pk': instance.pk,\n 'slug': instance.slug,\n 'date_last_crawled': '{0}'.format(instance.date_last_crawled)\n }, event_type=settings.OBER_EVENTS_CREATE_PUBLISHER)\n else:\n # notify updates, with basic serialized instance\n broadcast('pulse', {\n 'status': instance.status,\n 'status_changed': instance.status != instance._original.get('status'),\n 'pk': instance.pk,\n 'slug': instance.slug,\n 'date_last_crawled': '{0}'.format(instance.date_last_crawled)\n }, event_type=settings.OBER_EVENTS_UPDATE_PUBLISHER)","def receiveAction(self, action):\n self.action = action","def _action(self):\n pass","def mqtt_publish_fan_state():\n mqtt_publish_fan_status()\n mqtt_publish_fan_percon()\n mqtt_publish_fan_percoff()\n mqtt_publish_fan_tempon()\n mqtt_publish_fan_tempoff()","def upgrade_notifications_action(context, logger=None):\n #Run the workflow setup\n setup = getToolByName(context, 'portal_setup')\n setup.runImportStepFromProfile(PROFILE_ID, 'actions')\n return","def receive_notification(self, *args, **kwargs):\n\t\tprint(f\"{self.__location} is now hearing \\\"{args[0]}\\\" on {args[1]}\")","def log_create(sender, instance, created, **kwargs):\n if created:\n stracks.user(instance).log(\"? has been created\")","def on_notify(self, name):\r\n pass","def _after_serve_actions(self):\n pass","def on_new_notification(data, index, *args):\n print(data)","def on_publish(client, userdata, mid):\n print('on_publish')\n print(\" userdata:\" + str(userdata))\n print(\" mid:\" + str(mid))\n print()","def process_IN_CREATE(self, event):","def triggered(self, *args, **kwargs): # real signature unknown\n pass","def log(self, obj, action):\n action_dict = {'time': time.time(),\n 'action': action}\n self.log_data[obj.get_obj_id()]['actions'].append(action_dict)","def _generate_actions(self) -> list:\n pass","def send_back(self, action, payload):\n self.outbox.put({\n 'action': action,\n 'payload': payload\n })","def send_notification (event):\n Publisher.sendMessage (event)","def execute(self):\n return LOGGER.info(f\"{datetime.datetime.now()} - Sending notification in Slack\")","def device_action(self, client, action):\r\n client.deviceAction(action)","def actions() -> None:\n pass","def send_fanout_message(context, region, sns_topic, cli=False):\n\n message = json.dumps({'region': region})\n LOG.debug('send_fanout_message: %s', message)\n\n if cli:\n perform_snapshot(context, region)\n else:\n utils.sns_publish(TopicArn=sns_topic, Message=message)","def record_user_action(self, user, action):\n if user in self.users:\n if ((action == MessageAction.unread and not self.user_has_read(user))\n or (action == MessageAction.respond and (not self.requires_response or self.is_responded()[0]))):\n return\n elif action == MessageAction.delete:\n self.users.remove(user)\n self.history.append(MessageHistory(user, action))","def onActionChosen(self, agent, action):\n\n pass","def act(self):\n pass","def call_action(self, action):\n pass","def task_created(user, friends, task_description):\n\n title = f\"{user.username}'s task\"\n body = f\"{task_description}\"\n\n for user in friends:\n if user.fir_push_notif_token is not None and user.task_notifs:\n notify_user(user, title, body)","def on_post_published(sender, instance: dillo.models.posts.Post, **kwargs):\n # If activity was already created (the post was already published),\n # skip this signal to prevent the activity from appearing twice in the feeds.\n if models_actstream.Action.objects.filter(\n actor_object_id=instance.user.id, action_object_object_id=instance.id, verb='posted'\n ).exists():\n log.info('Content already published, skipping notification generation')\n return\n action.send(instance.user, verb='posted', action_object=instance)","def action_done(self):","def publish(self):\n return","def action(self, action):\n self._action = action","def notify_on_action(self):\n desired_action = self.pr.config.actionToBeNotifiedFor\n if self.pr.action == desired_action:\n if self.pr.base_branch == self.pr.config.mainBranch:\n msg = MSG_OPENED_TO_MAIN_BRANCH.format(repo=self.pr.repo, pr_by=self.created_by,\n main_branch=self.pr.config.mainBranch, title=self.pr.title,\n pr=self.pr.link_pretty, action=self.pr.action)\n for person in self.pr.config.techLeadsToBeNotified:\n self.slack.postToSlack(person, msg + MSG_RELEASE_PREPARATION)\n LOG.info(\"Notified to %s on action %s\" % (self.pr.config.techLeadsToBeNotified, self.pr.action))\n return {\"msg\": \"Notified to %s on action %s\" % (self.pr.config.techLeadsToBeNotified, self.pr.action)}\n else:\n msg = MSG_OPENED_TO_PREVENTED_BRANCH.format(repo=self.pr.repo, pr_by=self.created_by,\n base_branch=self.pr.base_branch, title=self.pr.title,\n pr=self.pr.link_pretty, action=self.pr.action)\n self.slack.postToSlack(self.pr.config.personToBeNotified, msg)\n LOG.info(\"Notified to %s on action %s\" % (self.pr.config.personToBeNotified, self.pr.action))\n return {\"msg\": \"Notified to %s on action %s\" % (self.pr.config.personToBeNotified, self.pr.action)}\n return {\"msg\": \"Skipped notify because its (%s) not desired event '%s'\" %(self.pr.action, desired_action)}","def action_run(self):\n pass","def acknowledged(self):\n ...","def events(self):","def take_action(self, obj):\n self.logger.info(\n \"Rule \\\"%s\\\" triggered and now I'm supposed to run the action \\\"%s\\\" with the log's object...\" %\n (self.name, self.action))","def on(self) -> None:","def on(self):","def on_create(self, payload):\n pass","def action(self, action):\n\n self._action = action","def notifications_n_email_after_event_creation(sender, instance, **kwargs):\n alarm = instance.alarm # Alarm which generated the event\n\n subscriptions = Subscription.objects.filter(alarm=alarm) # Getting the subscriptions associated with alarms\n sub_serializer = SubscriptionSerializer(subscriptions, many=True)\n send = [] # list of emails which the mail was send\n notificated = [] # list with users notificated\n\n # If no device, no variable and no content_type, there is nothing to send yet. Cancel notification and email\n if instance.device is None and instance.variables is None and len(instance.content_type.all()) == 0 :\n return\n for sub in sub_serializer.data: # Itering for subscription\n if sub['user'] is not None: # if user field isn't NULL AND not Group\n user = User.objects.get(id=sub['user'])\n if sub['active'] and user not in notificated: # if subscription is active\n Notification.objects.create(user=user, event=instance) # creating notification\n notificated.append(user) # adding user to the notified list\n if sub['email']: # if email option is checked\n email = user.email\n if email not in send: # for dont repeat email\n # Get a dict with relevant information about the event\n context = {'event': instance,\n 'alarm': instance.alarm,\n 'user': user,\n 'device': instance.device,\n 'var': instance.variables,\n 'content_type': instance.content_type.all()}\n plain_text = get_template('mail.txt') # Plain text template\n text_content = plain_text.render(context)\n subject = 'Event Alert: ' + instance.__str__()\n from_email = 'noreply@localhost.com'\n to = email\n msg = EmailMultiAlternatives(subject, text_content, from_email, [to])\n try:\n if sub['staff_template'] is not None:\n htmly = get_template(sub['staff_template']) # Define the HTML template\n html_content = htmly.render(context) # Rendering the templates with context information\n elif sub['staff_template_text'] != \"\":\n htmly = Template(sub['staff_template_text'])\n html_content = htmly.render(Context(context))\n elif sub['user_template'] is not None:\n htmly = get_template(sub['user_template']) # Define the HTML template\n html_content = htmly.render(context) # Rendering the templates with context information\n elif sub['user_template_text'] != \"\":\n htmly = Template(sub['user_template_text'])\n html_content = htmly.render(Context(context))\n msg.attach_alternative(html_content, 'text/html')\n msg.send()\n except:\n msg.send()\n print('Mail send to %s' % email)\n\n if sub['group'] is not None: # if is group and not user\n users_mail_list = [] # list with staff users instances\n if sub['active']: # if subscription is active\n group = Group.objects.get(pk=sub['group']) # Getting the group by id\n users = User.objects.filter(groups__name=group) # getting the users for group\n context = {'event': instance,\n 'alarm': instance.alarm,\n 'user': group,\n 'device': instance.device,\n 'var': instance.variables}\n for user in users: # Iterating users\n if user not in notificated:\n Notification.objects.create(user=user, event=instance) # creating notification\n notificated.append(user) # adding user to notificated list\n if sub['email']:\n mail = user.email # Adding the email for users in the user list\n if mail not in send: # for don't repeat email\n users_mail_list.append(mail)\n send.append(mail)\n # After getting all the emails and classifying it for staff and not staff members\n plain_text = get_template('mail.txt') # Plain text template\n text_content = plain_text.render(context)\n subject = 'Event Alert: ' + instance.__str__()\n from_email = 'noreply@localhost.com'\n msg = EmailMultiAlternatives(subject, text_content, from_email, users_mail_list)\n try:\n if sub['staff_template'] is not None:\n htmly = get_template(sub['staff_template']) # Define the HTML template\n html_content = htmly.render(context) # Rendering the templates with context information\n elif sub['staff_template_text'] != \"\":\n htmly = Template(sub['staff_template_text'])\n html_content = htmly.render(Context(context))\n elif sub['user_template'] is not None:\n htmly = get_template(sub['user_template']) # Define the HTML template\n html_content = htmly.render(context) # Rendering the templates with context information\n elif sub['user_template_text'] != \"\":\n htmly = Template(sub['user_template_text'])\n html_content = htmly.render(Context(context))\n msg.attach_alternative(html_content, 'text/html')\n msg.send()\n except:\n msg.send()\n print('Mail send to %s' % str(users_mail_list))","def on_action(atype, action):\n if atype == Actions.CREATE_OFFER:\n assert action.structure[\"which\"] in (\"local\", \"remote\")\n c = self.client if action.structure[\"which\"] == \"local\" else self.remote_client\n c.create_offer()\n return GstValidate.ActionReturn.OK\n elif atype == Actions.CREATE_ANSWER:\n assert action.structure[\"which\"] in (\"local\", \"remote\")\n c = self.client if action.structure[\"which\"] == \"local\" else self.remote_client\n c.create_answer()\n return GstValidate.ActionReturn.OK\n elif atype == Actions.WAIT_FOR_NEGOTIATION_STATE:\n states = [NegotiationState(action.structure[\"state\"]), NegotiationState.ERROR]\n assert action.structure[\"which\"] in (\"local\", \"remote\")\n c = self.client if action.structure[\"which\"] == \"local\" else self.remote_client\n state = c.wait_for_negotiation_states(states)\n return GstValidate.ActionReturn.OK if state != NegotiationState.ERROR else GstValidate.ActionReturn.ERROR\n elif atype == Actions.ADD_STREAM:\n self.client.add_stream(action.structure[\"pipeline\"])\n return GstValidate.ActionReturn.OK\n elif atype == Actions.ADD_DATA_CHANNEL:\n assert action.structure[\"which\"] in (\"local\", \"remote\")\n c = self.client if action.structure[\"which\"] == \"local\" else self.remote_client\n c.add_data_channel(action.structure[\"id\"])\n return GstValidate.ActionReturn.OK\n elif atype == Actions.SEND_DATA_CHANNEL_STRING:\n assert action.structure[\"which\"] in (\"local\", \"remote\")\n c = self.client if action.structure[\"which\"] == \"local\" else self.remote_client\n channel = c.find_channel (action.structure[\"id\"])\n channel.send_string (action.structure[\"msg\"])\n return GstValidate.ActionReturn.OK\n elif atype == Actions.WAIT_FOR_DATA_CHANNEL_STATE:\n assert action.structure[\"which\"] in (\"local\", \"remote\")\n c = self.client if action.structure[\"which\"] == \"local\" else self.remote_client\n states = [DataChannelState(action.structure[\"state\"]), DataChannelState.ERROR]\n channel = c.find_channel (action.structure[\"id\"])\n state = channel.wait_for_states(states)\n return GstValidate.ActionReturn.OK if state != DataChannelState.ERROR else GstValidate.ActionReturn.ERROR\n elif atype == Actions.CLOSE_DATA_CHANNEL:\n assert action.structure[\"which\"] in (\"local\", \"remote\")\n c = self.client if action.structure[\"which\"] == \"local\" else self.remote_client\n channel = c.find_channel (action.structure[\"id\"])\n channel.close()\n return GstValidate.ActionReturn.OK\n elif atype == Actions.WAIT_FOR_DATA_CHANNEL:\n assert action.structure[\"which\"] in (\"local\", \"remote\")\n c = self.client if action.structure[\"which\"] == \"local\" else self.remote_client\n state = c.wait_for_data_channel(action.structure[\"id\"])\n return GstValidate.ActionReturn.OK\n elif atype == Actions.WAIT_FOR_DATA_CHANNEL_STRING:\n assert action.structure[\"which\"] in (\"local\", \"remote\")\n c = self.client if action.structure[\"which\"] == \"local\" else self.remote_client\n channel = c.find_channel (action.structure[\"id\"])\n channel.wait_for_message(action.structure[\"msg\"])\n return GstValidate.ActionReturn.OK\n elif atype == Actions.WAIT_FOR_NEGOTIATION_NEEDED:\n self.client.wait_for_negotiation_needed(action.structure[\"generation\"])\n return GstValidate.ActionReturn.OK\n elif atype == Actions.SET_WEBRTC_OPTIONS:\n self.client.set_options (action.structure)\n self.remote_client.set_options (action.structure)\n return GstValidate.ActionReturn.OK\n else:\n assert \"Not reached\" == \"\"","def _do_action(self):\n pass","def _do_action(self):\n pass","def post_delete(cls, sender, instance, **kwargs):\n super(AbstractActivityModelMixin, cls).post_delete(sender, instance,\n **kwargs)\n # need to delete any notifications about this object\n get_activity_model().objects.delete_all_about_object(about=instance)","def notify(self) -> None:\n pass","def notify(self) -> None:\n pass","def on(self) -> None:\n ...","def actions(self):\n raise NotImplementedError","def publish_status(client):\n client.publish(config.topic_get, payload=getlight())","def on_add(self):\n self.notify(on_add())","def post(self, event, *args, **kwargs):\n self.inq.Signal((event, args, kwargs))","def step(self) -> None:\n info = self._check_status()\n for key in info:\n self.client.publish(\n topic=f\"home/watering/{key}\", payload=info[key], retain=True\n )","def addPostStartAction ( action ) :\n global __Bender_PostStart_Actions\n if action : __Bender_PostStart_Actions.append ( action ) \n return tuple(__Bender_PostStart_Actions)","def trigger(self, type, event):","def handle_sent(self, instance):\n if not instance.created_by:\n return\n\n activity = Activity(\n actor=instance.created_by,\n verb=RestrictedMailSent,\n object=instance,\n time=instance.used,\n extra_context={},\n )\n self.manager.add_activity(\n activity, [instance.created_by.pk], [NotificationFeed]\n )\n\n # Send notification\n notification = RestrictedMailSentNotification(instance.created_by)\n notification.notify()","def add_action(self, action):\n self.action = action","def add_action(self, action):\n self.action = action","def actions(self, actions):\n\n self._actions = actions","def actions(self, actions):\n\n self._actions = actions","def get_user_notifications(self, login):","def get_action(self, context):\n pass","def send_notifications_to_all_users(sender,\n instance,\n created, *args, **kwargs):\n\n if instance and created:\n users_followers = instance.author.followers.all()\n\n link = f\"\"\"{os.getenv(\"HEROKU_BACKEND_URL\")}/articles/\\n\"\"\"\n f\"\"\"{instance.slug}\"\"\"\n for user in users_followers:\n if user.get_notifications:\n uuid = urlsafe_base64_encode(force_bytes(user)\n ).decode(\"utf-8\")\n subscription = f'{os.getenv(\"HEROKU_BACKEND_URL\")}/api/' +\\\n 'v1/users/' +\\\n f'unsubscribe/{uuid}/'\n sender = os.getenv('EMAIL_HOST_USER')\n email = user.email\n email_subject = \"Author's Haven Email Notification\"\n message = render_to_string('create_article.html', {\n 'title': email_subject,\n 'username': user.username,\n 'link': link,\n 'subscription': subscription\n })\n\n send_mail(email_subject, '', sender, [\n email, ], html_message=message)\n notify.send(instance.author, recipient=user,\n verb='A user you follow has a new post',\n action_object=instance)","def actions(self, state):\n raise NotImplementedError # Override this!","def notification_handler(sender, data):\n print(\"{0}: {1}\".format(sender, data))","def create_action(instance, verb, user):\n return instance.activities.create(action=verb, owner=user)","def __init__(self):\n self.actions = []","def handler(event, context):\n alert_message = json.loads(event['Records'][0]['Sns']['Message'])\n alarm_name = alert_message['AlarmName']\n reason = alert_message['NewStateReason']\n new_state = alert_message['NewStateValue']\n color = \"good\" if new_state == 'OK' else \"danger\"\n\n region = os.getenv('AWS_DEFAULT_REGION')\n alert_url = f'https://console.aws.amazon.com/cloudwatch/home?region={region}#alarm:alarmFilter=ANY;name={alarm_name}'\n link = f\"<{alert_url}|{alarm_name}>\"\n\n secrets = json.loads(get_secret()['SecretString'])\n default_slack_channel = secrets['slack_alert_channel']\n alarm_description = json.loads(alert_message.get('AlarmDescription', '{}'))\n slack_channel = alarm_description.get(\"slack_channel\", default_slack_channel)\n description = alarm_description.get(\"description\")\n slack_message = '\\n'.join(\n [f\"New state: {new_state}\", f\"Description: {description}\", reason]\n )\n\n attachments = [{\n \"fallback\": f\"{link} {slack_message}\",\n \"title\": alarm_name,\n \"title_link\": alert_url,\n \"text\": slack_message,\n \"color\": color\n }]\n\n slack_url = secrets['slack_webhooks'][slack_channel]\n\n post_message_to_url(slack_url, {\"attachments\": attachments})","def send_action(self, *args, **kwargs):\n return self.bot.send_chat_action(self.id, *args, **kwargs)","def write(cls, activity_type, actor, target, data):\n\n activity_log = ActivityLog.objects.create(\n activity_type=activity_type, actor=actor, target=target, data=data\n )\n cls.notify(activity_log)","def action_done(self):\n pass","def perform_action(self, action_data):\n pass","def on_watch(self, payload):\n pass","def obtain_action(self):\r\n\t\treturn","def get_actions(self, request):\n actions = super().get_actions(request)\n if not settings.PUBLISHER_CODE:\n del actions['create_cwr']\n if 'delete_selected' in actions:\n del actions['delete_selected']\n return actions","def handle_action_received(msg: ReceiveMessage) -> None:\n payload = self.render_template(msg, CONF_ACTION_TEMPLATE)\n if not payload or payload == PAYLOAD_NONE:\n _LOGGER.debug(\n \"Invalid %s action: %s, ignoring\",\n [e.value for e in HVACAction],\n payload,\n )\n return\n try:\n self._attr_hvac_action = HVACAction(str(payload))\n except ValueError:\n _LOGGER.warning(\n \"Invalid %s action: %s\",\n [e.value for e in HVACAction],\n payload,\n )\n return\n get_mqtt_data(self.hass).state_write_requests.write_state_request(self)","def act(self) -> None:\n pass","def on_run(self):\n pass","def on_publish(client: mqtt.Client, userdata: Any, mid: int) -> None:\n logging.info(f\"Successfully published a message: mid={mid}\")","def subscribeViewerClosed(self, action):\n self.sub_id += 1\n self._viewer_closed_subscribers[self.sub_id] = action\n return self.sub_id"],"string":"[\n \"def action_defined(sender, instance, created, raw, using, **kwargs):\\n if created:\\n raw_hook_event.send(\\n sender=None,\\n event_name=\\\"action_defined\\\",\\n instance=instance,\\n payload=ActionSerializer(instance).data,\\n user=instance.team,\\n )\",\n \"def onAction(*args):\",\n \"def onAction(*args):\",\n \"def onAction(*args):\",\n \"def onAction(*args):\",\n \"def publish_action(self, action):\\n raise NotImplementedError\",\n \"def on_action(self, connection, event):\\r\\n print('[{}] OnAction from {}' .format(event.type.upper(), event.source))\",\n \"def send_actions(self, actions):\\n pass\",\n \"def actions():\\n pass\",\n \"def on_publish(client, userdata, mid):\\n print(\\\"Message Published.\\\")\",\n \"def onActionTaken(self, agent):\\n\\n pass\",\n \"def post_activities():\\n pass\",\n \"def action(self):\\n pass\",\n \"def action(self):\\n pass\",\n \"def on_publish( client, userdata, mid ):\\n logging.info( \\\"Data published successfully.\\\" )\",\n \"async def send_event_created(self, action_id: int):\\n async with self.pg.acquire() as conn:\\n data = await conn.fetchrow(\\n \\\"\\\"\\\"\\n SELECT a.company AS company_id, u.role AS host_role, u.id AS host_user_id,\\n full_name(u.first_name, u.last_name, u.email) AS host_name,\\n e.id AS event_id, e.name AS event_name,\\n (e.start_ts AT TIME ZONE e.timezone)::date AS event_date,\\n cat.name AS cat_name, cat.slug AS cat_slug,\\n event_link(cat.slug, e.slug, e.public, $2) AS event_link\\n FROM actions AS a\\n JOIN users AS u ON a.user_id = u.id\\n JOIN events AS e ON a.event = e.id\\n JOIN categories AS cat ON e.category = cat.id\\n WHERE a.id=$1\\n \\\"\\\"\\\",\\n action_id,\\n self.settings.auth_key,\\n )\\n\\n link = f'/dashboard/events/{data[\\\"event_id\\\"]}/'\\n ctx = dict(\\n summary='{host_name} created an event \\\"{event_name}\\\"'.format(**data),\\n details=(\\n 'Event \\\"{event_name}\\\" ({cat_name}) created by \\\"{host_name}\\\" ({host_role}), '\\n 'click the link below to view the event.'\\n ).format(**data),\\n action_label='View Event',\\n action_link=link,\\n )\\n users = [\\n UserEmail(id=r['id'], ctx=ctx)\\n for r in await conn.fetch(\\\"SELECT id FROM users WHERE role='admin' AND company=$1\\\", data['company_id'])\\n ]\\n await self.send_emails.direct(data['company_id'], Triggers.admin_notification, users)\\n if data['host_role'] != 'admin':\\n ctx = {\\n 'event_link': data['event_link'],\\n 'event_dashboard_link': link,\\n 'event_name': data['event_name'],\\n 'event_date': format_dt(data['event_date']),\\n 'category_name': data['cat_name'],\\n is_cat(data['cat_slug']): True,\\n }\\n await self.send_emails.direct(\\n data['company_id'], Triggers.event_host_created, [UserEmail(data['host_user_id'], ctx)]\\n )\",\n \"def take_action(self, *args, **kwargs):\\r\\n pass\",\n \"def on_publish(unused_client, unused_userdata, unused_mid):\\n print('on_publish')\\n status_light.on()\",\n \"def write(self, notification):\",\n \"def on_publish(unused_client, unused_userdata, unused_mid):\\n\\tprint('on_publish')\",\n \"def onShowed(self):\\n self.parent.actionTagTwo=\\\"\\\"\\n pass\",\n \"def broadcast_publisher_info(sender, instance, created, **kwargs):\\n if created:\\n # notify creation to generic channel\\n broadcast('pulse', {\\n 'status': instance.status,\\n 'pk': instance.pk,\\n 'slug': instance.slug,\\n 'date_last_crawled': '{0}'.format(instance.date_last_crawled)\\n }, event_type=settings.OBER_EVENTS_CREATE_PUBLISHER)\\n else:\\n # notify updates, with basic serialized instance\\n broadcast('pulse', {\\n 'status': instance.status,\\n 'status_changed': instance.status != instance._original.get('status'),\\n 'pk': instance.pk,\\n 'slug': instance.slug,\\n 'date_last_crawled': '{0}'.format(instance.date_last_crawled)\\n }, event_type=settings.OBER_EVENTS_UPDATE_PUBLISHER)\",\n \"def receiveAction(self, action):\\n self.action = action\",\n \"def _action(self):\\n pass\",\n \"def mqtt_publish_fan_state():\\n mqtt_publish_fan_status()\\n mqtt_publish_fan_percon()\\n mqtt_publish_fan_percoff()\\n mqtt_publish_fan_tempon()\\n mqtt_publish_fan_tempoff()\",\n \"def upgrade_notifications_action(context, logger=None):\\n #Run the workflow setup\\n setup = getToolByName(context, 'portal_setup')\\n setup.runImportStepFromProfile(PROFILE_ID, 'actions')\\n return\",\n \"def receive_notification(self, *args, **kwargs):\\n\\t\\tprint(f\\\"{self.__location} is now hearing \\\\\\\"{args[0]}\\\\\\\" on {args[1]}\\\")\",\n \"def log_create(sender, instance, created, **kwargs):\\n if created:\\n stracks.user(instance).log(\\\"? has been created\\\")\",\n \"def on_notify(self, name):\\r\\n pass\",\n \"def _after_serve_actions(self):\\n pass\",\n \"def on_new_notification(data, index, *args):\\n print(data)\",\n \"def on_publish(client, userdata, mid):\\n print('on_publish')\\n print(\\\" userdata:\\\" + str(userdata))\\n print(\\\" mid:\\\" + str(mid))\\n print()\",\n \"def process_IN_CREATE(self, event):\",\n \"def triggered(self, *args, **kwargs): # real signature unknown\\n pass\",\n \"def log(self, obj, action):\\n action_dict = {'time': time.time(),\\n 'action': action}\\n self.log_data[obj.get_obj_id()]['actions'].append(action_dict)\",\n \"def _generate_actions(self) -> list:\\n pass\",\n \"def send_back(self, action, payload):\\n self.outbox.put({\\n 'action': action,\\n 'payload': payload\\n })\",\n \"def send_notification (event):\\n Publisher.sendMessage (event)\",\n \"def execute(self):\\n return LOGGER.info(f\\\"{datetime.datetime.now()} - Sending notification in Slack\\\")\",\n \"def device_action(self, client, action):\\r\\n client.deviceAction(action)\",\n \"def actions() -> None:\\n pass\",\n \"def send_fanout_message(context, region, sns_topic, cli=False):\\n\\n message = json.dumps({'region': region})\\n LOG.debug('send_fanout_message: %s', message)\\n\\n if cli:\\n perform_snapshot(context, region)\\n else:\\n utils.sns_publish(TopicArn=sns_topic, Message=message)\",\n \"def record_user_action(self, user, action):\\n if user in self.users:\\n if ((action == MessageAction.unread and not self.user_has_read(user))\\n or (action == MessageAction.respond and (not self.requires_response or self.is_responded()[0]))):\\n return\\n elif action == MessageAction.delete:\\n self.users.remove(user)\\n self.history.append(MessageHistory(user, action))\",\n \"def onActionChosen(self, agent, action):\\n\\n pass\",\n \"def act(self):\\n pass\",\n \"def call_action(self, action):\\n pass\",\n \"def task_created(user, friends, task_description):\\n\\n title = f\\\"{user.username}'s task\\\"\\n body = f\\\"{task_description}\\\"\\n\\n for user in friends:\\n if user.fir_push_notif_token is not None and user.task_notifs:\\n notify_user(user, title, body)\",\n \"def on_post_published(sender, instance: dillo.models.posts.Post, **kwargs):\\n # If activity was already created (the post was already published),\\n # skip this signal to prevent the activity from appearing twice in the feeds.\\n if models_actstream.Action.objects.filter(\\n actor_object_id=instance.user.id, action_object_object_id=instance.id, verb='posted'\\n ).exists():\\n log.info('Content already published, skipping notification generation')\\n return\\n action.send(instance.user, verb='posted', action_object=instance)\",\n \"def action_done(self):\",\n \"def publish(self):\\n return\",\n \"def action(self, action):\\n self._action = action\",\n \"def notify_on_action(self):\\n desired_action = self.pr.config.actionToBeNotifiedFor\\n if self.pr.action == desired_action:\\n if self.pr.base_branch == self.pr.config.mainBranch:\\n msg = MSG_OPENED_TO_MAIN_BRANCH.format(repo=self.pr.repo, pr_by=self.created_by,\\n main_branch=self.pr.config.mainBranch, title=self.pr.title,\\n pr=self.pr.link_pretty, action=self.pr.action)\\n for person in self.pr.config.techLeadsToBeNotified:\\n self.slack.postToSlack(person, msg + MSG_RELEASE_PREPARATION)\\n LOG.info(\\\"Notified to %s on action %s\\\" % (self.pr.config.techLeadsToBeNotified, self.pr.action))\\n return {\\\"msg\\\": \\\"Notified to %s on action %s\\\" % (self.pr.config.techLeadsToBeNotified, self.pr.action)}\\n else:\\n msg = MSG_OPENED_TO_PREVENTED_BRANCH.format(repo=self.pr.repo, pr_by=self.created_by,\\n base_branch=self.pr.base_branch, title=self.pr.title,\\n pr=self.pr.link_pretty, action=self.pr.action)\\n self.slack.postToSlack(self.pr.config.personToBeNotified, msg)\\n LOG.info(\\\"Notified to %s on action %s\\\" % (self.pr.config.personToBeNotified, self.pr.action))\\n return {\\\"msg\\\": \\\"Notified to %s on action %s\\\" % (self.pr.config.personToBeNotified, self.pr.action)}\\n return {\\\"msg\\\": \\\"Skipped notify because its (%s) not desired event '%s'\\\" %(self.pr.action, desired_action)}\",\n \"def action_run(self):\\n pass\",\n \"def acknowledged(self):\\n ...\",\n \"def events(self):\",\n \"def take_action(self, obj):\\n self.logger.info(\\n \\\"Rule \\\\\\\"%s\\\\\\\" triggered and now I'm supposed to run the action \\\\\\\"%s\\\\\\\" with the log's object...\\\" %\\n (self.name, self.action))\",\n \"def on(self) -> None:\",\n \"def on(self):\",\n \"def on_create(self, payload):\\n pass\",\n \"def action(self, action):\\n\\n self._action = action\",\n \"def notifications_n_email_after_event_creation(sender, instance, **kwargs):\\n alarm = instance.alarm # Alarm which generated the event\\n\\n subscriptions = Subscription.objects.filter(alarm=alarm) # Getting the subscriptions associated with alarms\\n sub_serializer = SubscriptionSerializer(subscriptions, many=True)\\n send = [] # list of emails which the mail was send\\n notificated = [] # list with users notificated\\n\\n # If no device, no variable and no content_type, there is nothing to send yet. Cancel notification and email\\n if instance.device is None and instance.variables is None and len(instance.content_type.all()) == 0 :\\n return\\n for sub in sub_serializer.data: # Itering for subscription\\n if sub['user'] is not None: # if user field isn't NULL AND not Group\\n user = User.objects.get(id=sub['user'])\\n if sub['active'] and user not in notificated: # if subscription is active\\n Notification.objects.create(user=user, event=instance) # creating notification\\n notificated.append(user) # adding user to the notified list\\n if sub['email']: # if email option is checked\\n email = user.email\\n if email not in send: # for dont repeat email\\n # Get a dict with relevant information about the event\\n context = {'event': instance,\\n 'alarm': instance.alarm,\\n 'user': user,\\n 'device': instance.device,\\n 'var': instance.variables,\\n 'content_type': instance.content_type.all()}\\n plain_text = get_template('mail.txt') # Plain text template\\n text_content = plain_text.render(context)\\n subject = 'Event Alert: ' + instance.__str__()\\n from_email = 'noreply@localhost.com'\\n to = email\\n msg = EmailMultiAlternatives(subject, text_content, from_email, [to])\\n try:\\n if sub['staff_template'] is not None:\\n htmly = get_template(sub['staff_template']) # Define the HTML template\\n html_content = htmly.render(context) # Rendering the templates with context information\\n elif sub['staff_template_text'] != \\\"\\\":\\n htmly = Template(sub['staff_template_text'])\\n html_content = htmly.render(Context(context))\\n elif sub['user_template'] is not None:\\n htmly = get_template(sub['user_template']) # Define the HTML template\\n html_content = htmly.render(context) # Rendering the templates with context information\\n elif sub['user_template_text'] != \\\"\\\":\\n htmly = Template(sub['user_template_text'])\\n html_content = htmly.render(Context(context))\\n msg.attach_alternative(html_content, 'text/html')\\n msg.send()\\n except:\\n msg.send()\\n print('Mail send to %s' % email)\\n\\n if sub['group'] is not None: # if is group and not user\\n users_mail_list = [] # list with staff users instances\\n if sub['active']: # if subscription is active\\n group = Group.objects.get(pk=sub['group']) # Getting the group by id\\n users = User.objects.filter(groups__name=group) # getting the users for group\\n context = {'event': instance,\\n 'alarm': instance.alarm,\\n 'user': group,\\n 'device': instance.device,\\n 'var': instance.variables}\\n for user in users: # Iterating users\\n if user not in notificated:\\n Notification.objects.create(user=user, event=instance) # creating notification\\n notificated.append(user) # adding user to notificated list\\n if sub['email']:\\n mail = user.email # Adding the email for users in the user list\\n if mail not in send: # for don't repeat email\\n users_mail_list.append(mail)\\n send.append(mail)\\n # After getting all the emails and classifying it for staff and not staff members\\n plain_text = get_template('mail.txt') # Plain text template\\n text_content = plain_text.render(context)\\n subject = 'Event Alert: ' + instance.__str__()\\n from_email = 'noreply@localhost.com'\\n msg = EmailMultiAlternatives(subject, text_content, from_email, users_mail_list)\\n try:\\n if sub['staff_template'] is not None:\\n htmly = get_template(sub['staff_template']) # Define the HTML template\\n html_content = htmly.render(context) # Rendering the templates with context information\\n elif sub['staff_template_text'] != \\\"\\\":\\n htmly = Template(sub['staff_template_text'])\\n html_content = htmly.render(Context(context))\\n elif sub['user_template'] is not None:\\n htmly = get_template(sub['user_template']) # Define the HTML template\\n html_content = htmly.render(context) # Rendering the templates with context information\\n elif sub['user_template_text'] != \\\"\\\":\\n htmly = Template(sub['user_template_text'])\\n html_content = htmly.render(Context(context))\\n msg.attach_alternative(html_content, 'text/html')\\n msg.send()\\n except:\\n msg.send()\\n print('Mail send to %s' % str(users_mail_list))\",\n \"def on_action(atype, action):\\n if atype == Actions.CREATE_OFFER:\\n assert action.structure[\\\"which\\\"] in (\\\"local\\\", \\\"remote\\\")\\n c = self.client if action.structure[\\\"which\\\"] == \\\"local\\\" else self.remote_client\\n c.create_offer()\\n return GstValidate.ActionReturn.OK\\n elif atype == Actions.CREATE_ANSWER:\\n assert action.structure[\\\"which\\\"] in (\\\"local\\\", \\\"remote\\\")\\n c = self.client if action.structure[\\\"which\\\"] == \\\"local\\\" else self.remote_client\\n c.create_answer()\\n return GstValidate.ActionReturn.OK\\n elif atype == Actions.WAIT_FOR_NEGOTIATION_STATE:\\n states = [NegotiationState(action.structure[\\\"state\\\"]), NegotiationState.ERROR]\\n assert action.structure[\\\"which\\\"] in (\\\"local\\\", \\\"remote\\\")\\n c = self.client if action.structure[\\\"which\\\"] == \\\"local\\\" else self.remote_client\\n state = c.wait_for_negotiation_states(states)\\n return GstValidate.ActionReturn.OK if state != NegotiationState.ERROR else GstValidate.ActionReturn.ERROR\\n elif atype == Actions.ADD_STREAM:\\n self.client.add_stream(action.structure[\\\"pipeline\\\"])\\n return GstValidate.ActionReturn.OK\\n elif atype == Actions.ADD_DATA_CHANNEL:\\n assert action.structure[\\\"which\\\"] in (\\\"local\\\", \\\"remote\\\")\\n c = self.client if action.structure[\\\"which\\\"] == \\\"local\\\" else self.remote_client\\n c.add_data_channel(action.structure[\\\"id\\\"])\\n return GstValidate.ActionReturn.OK\\n elif atype == Actions.SEND_DATA_CHANNEL_STRING:\\n assert action.structure[\\\"which\\\"] in (\\\"local\\\", \\\"remote\\\")\\n c = self.client if action.structure[\\\"which\\\"] == \\\"local\\\" else self.remote_client\\n channel = c.find_channel (action.structure[\\\"id\\\"])\\n channel.send_string (action.structure[\\\"msg\\\"])\\n return GstValidate.ActionReturn.OK\\n elif atype == Actions.WAIT_FOR_DATA_CHANNEL_STATE:\\n assert action.structure[\\\"which\\\"] in (\\\"local\\\", \\\"remote\\\")\\n c = self.client if action.structure[\\\"which\\\"] == \\\"local\\\" else self.remote_client\\n states = [DataChannelState(action.structure[\\\"state\\\"]), DataChannelState.ERROR]\\n channel = c.find_channel (action.structure[\\\"id\\\"])\\n state = channel.wait_for_states(states)\\n return GstValidate.ActionReturn.OK if state != DataChannelState.ERROR else GstValidate.ActionReturn.ERROR\\n elif atype == Actions.CLOSE_DATA_CHANNEL:\\n assert action.structure[\\\"which\\\"] in (\\\"local\\\", \\\"remote\\\")\\n c = self.client if action.structure[\\\"which\\\"] == \\\"local\\\" else self.remote_client\\n channel = c.find_channel (action.structure[\\\"id\\\"])\\n channel.close()\\n return GstValidate.ActionReturn.OK\\n elif atype == Actions.WAIT_FOR_DATA_CHANNEL:\\n assert action.structure[\\\"which\\\"] in (\\\"local\\\", \\\"remote\\\")\\n c = self.client if action.structure[\\\"which\\\"] == \\\"local\\\" else self.remote_client\\n state = c.wait_for_data_channel(action.structure[\\\"id\\\"])\\n return GstValidate.ActionReturn.OK\\n elif atype == Actions.WAIT_FOR_DATA_CHANNEL_STRING:\\n assert action.structure[\\\"which\\\"] in (\\\"local\\\", \\\"remote\\\")\\n c = self.client if action.structure[\\\"which\\\"] == \\\"local\\\" else self.remote_client\\n channel = c.find_channel (action.structure[\\\"id\\\"])\\n channel.wait_for_message(action.structure[\\\"msg\\\"])\\n return GstValidate.ActionReturn.OK\\n elif atype == Actions.WAIT_FOR_NEGOTIATION_NEEDED:\\n self.client.wait_for_negotiation_needed(action.structure[\\\"generation\\\"])\\n return GstValidate.ActionReturn.OK\\n elif atype == Actions.SET_WEBRTC_OPTIONS:\\n self.client.set_options (action.structure)\\n self.remote_client.set_options (action.structure)\\n return GstValidate.ActionReturn.OK\\n else:\\n assert \\\"Not reached\\\" == \\\"\\\"\",\n \"def _do_action(self):\\n pass\",\n \"def _do_action(self):\\n pass\",\n \"def post_delete(cls, sender, instance, **kwargs):\\n super(AbstractActivityModelMixin, cls).post_delete(sender, instance,\\n **kwargs)\\n # need to delete any notifications about this object\\n get_activity_model().objects.delete_all_about_object(about=instance)\",\n \"def notify(self) -> None:\\n pass\",\n \"def notify(self) -> None:\\n pass\",\n \"def on(self) -> None:\\n ...\",\n \"def actions(self):\\n raise NotImplementedError\",\n \"def publish_status(client):\\n client.publish(config.topic_get, payload=getlight())\",\n \"def on_add(self):\\n self.notify(on_add())\",\n \"def post(self, event, *args, **kwargs):\\n self.inq.Signal((event, args, kwargs))\",\n \"def step(self) -> None:\\n info = self._check_status()\\n for key in info:\\n self.client.publish(\\n topic=f\\\"home/watering/{key}\\\", payload=info[key], retain=True\\n )\",\n \"def addPostStartAction ( action ) :\\n global __Bender_PostStart_Actions\\n if action : __Bender_PostStart_Actions.append ( action ) \\n return tuple(__Bender_PostStart_Actions)\",\n \"def trigger(self, type, event):\",\n \"def handle_sent(self, instance):\\n if not instance.created_by:\\n return\\n\\n activity = Activity(\\n actor=instance.created_by,\\n verb=RestrictedMailSent,\\n object=instance,\\n time=instance.used,\\n extra_context={},\\n )\\n self.manager.add_activity(\\n activity, [instance.created_by.pk], [NotificationFeed]\\n )\\n\\n # Send notification\\n notification = RestrictedMailSentNotification(instance.created_by)\\n notification.notify()\",\n \"def add_action(self, action):\\n self.action = action\",\n \"def add_action(self, action):\\n self.action = action\",\n \"def actions(self, actions):\\n\\n self._actions = actions\",\n \"def actions(self, actions):\\n\\n self._actions = actions\",\n \"def get_user_notifications(self, login):\",\n \"def get_action(self, context):\\n pass\",\n \"def send_notifications_to_all_users(sender,\\n instance,\\n created, *args, **kwargs):\\n\\n if instance and created:\\n users_followers = instance.author.followers.all()\\n\\n link = f\\\"\\\"\\\"{os.getenv(\\\"HEROKU_BACKEND_URL\\\")}/articles/\\\\n\\\"\\\"\\\"\\n f\\\"\\\"\\\"{instance.slug}\\\"\\\"\\\"\\n for user in users_followers:\\n if user.get_notifications:\\n uuid = urlsafe_base64_encode(force_bytes(user)\\n ).decode(\\\"utf-8\\\")\\n subscription = f'{os.getenv(\\\"HEROKU_BACKEND_URL\\\")}/api/' +\\\\\\n 'v1/users/' +\\\\\\n f'unsubscribe/{uuid}/'\\n sender = os.getenv('EMAIL_HOST_USER')\\n email = user.email\\n email_subject = \\\"Author's Haven Email Notification\\\"\\n message = render_to_string('create_article.html', {\\n 'title': email_subject,\\n 'username': user.username,\\n 'link': link,\\n 'subscription': subscription\\n })\\n\\n send_mail(email_subject, '', sender, [\\n email, ], html_message=message)\\n notify.send(instance.author, recipient=user,\\n verb='A user you follow has a new post',\\n action_object=instance)\",\n \"def actions(self, state):\\n raise NotImplementedError # Override this!\",\n \"def notification_handler(sender, data):\\n print(\\\"{0}: {1}\\\".format(sender, data))\",\n \"def create_action(instance, verb, user):\\n return instance.activities.create(action=verb, owner=user)\",\n \"def __init__(self):\\n self.actions = []\",\n \"def handler(event, context):\\n alert_message = json.loads(event['Records'][0]['Sns']['Message'])\\n alarm_name = alert_message['AlarmName']\\n reason = alert_message['NewStateReason']\\n new_state = alert_message['NewStateValue']\\n color = \\\"good\\\" if new_state == 'OK' else \\\"danger\\\"\\n\\n region = os.getenv('AWS_DEFAULT_REGION')\\n alert_url = f'https://console.aws.amazon.com/cloudwatch/home?region={region}#alarm:alarmFilter=ANY;name={alarm_name}'\\n link = f\\\"<{alert_url}|{alarm_name}>\\\"\\n\\n secrets = json.loads(get_secret()['SecretString'])\\n default_slack_channel = secrets['slack_alert_channel']\\n alarm_description = json.loads(alert_message.get('AlarmDescription', '{}'))\\n slack_channel = alarm_description.get(\\\"slack_channel\\\", default_slack_channel)\\n description = alarm_description.get(\\\"description\\\")\\n slack_message = '\\\\n'.join(\\n [f\\\"New state: {new_state}\\\", f\\\"Description: {description}\\\", reason]\\n )\\n\\n attachments = [{\\n \\\"fallback\\\": f\\\"{link} {slack_message}\\\",\\n \\\"title\\\": alarm_name,\\n \\\"title_link\\\": alert_url,\\n \\\"text\\\": slack_message,\\n \\\"color\\\": color\\n }]\\n\\n slack_url = secrets['slack_webhooks'][slack_channel]\\n\\n post_message_to_url(slack_url, {\\\"attachments\\\": attachments})\",\n \"def send_action(self, *args, **kwargs):\\n return self.bot.send_chat_action(self.id, *args, **kwargs)\",\n \"def write(cls, activity_type, actor, target, data):\\n\\n activity_log = ActivityLog.objects.create(\\n activity_type=activity_type, actor=actor, target=target, data=data\\n )\\n cls.notify(activity_log)\",\n \"def action_done(self):\\n pass\",\n \"def perform_action(self, action_data):\\n pass\",\n \"def on_watch(self, payload):\\n pass\",\n \"def obtain_action(self):\\r\\n\\t\\treturn\",\n \"def get_actions(self, request):\\n actions = super().get_actions(request)\\n if not settings.PUBLISHER_CODE:\\n del actions['create_cwr']\\n if 'delete_selected' in actions:\\n del actions['delete_selected']\\n return actions\",\n \"def handle_action_received(msg: ReceiveMessage) -> None:\\n payload = self.render_template(msg, CONF_ACTION_TEMPLATE)\\n if not payload or payload == PAYLOAD_NONE:\\n _LOGGER.debug(\\n \\\"Invalid %s action: %s, ignoring\\\",\\n [e.value for e in HVACAction],\\n payload,\\n )\\n return\\n try:\\n self._attr_hvac_action = HVACAction(str(payload))\\n except ValueError:\\n _LOGGER.warning(\\n \\\"Invalid %s action: %s\\\",\\n [e.value for e in HVACAction],\\n payload,\\n )\\n return\\n get_mqtt_data(self.hass).state_write_requests.write_state_request(self)\",\n \"def act(self) -> None:\\n pass\",\n \"def on_run(self):\\n pass\",\n \"def on_publish(client: mqtt.Client, userdata: Any, mid: int) -> None:\\n logging.info(f\\\"Successfully published a message: mid={mid}\\\")\",\n \"def subscribeViewerClosed(self, action):\\n self.sub_id += 1\\n self._viewer_closed_subscribers[self.sub_id] = action\\n return self.sub_id\"\n]","isConversational":false},"negative_scores":{"kind":"list like","value":["0.63869154","0.6225311","0.6225311","0.6225311","0.6225311","0.6144163","0.5890172","0.58513165","0.58451253","0.5807999","0.57855886","0.57809436","0.5763071","0.5763071","0.57493734","0.5724776","0.572035","0.5714243","0.5679732","0.5625895","0.55907756","0.55830616","0.55718523","0.555276","0.553306","0.5528688","0.55265915","0.55073893","0.54892355","0.54816544","0.54693997","0.54672605","0.5447692","0.5399855","0.53814375","0.5361442","0.53557247","0.5355606","0.535154","0.53432065","0.53325236","0.5330406","0.5323166","0.53179884","0.5307609","0.5303047","0.52949667","0.52912444","0.5285322","0.5279463","0.527885","0.5276315","0.52585864","0.5255246","0.5253941","0.52438843","0.523733","0.5231569","0.52264756","0.5224379","0.52192605","0.5215848","0.5213145","0.5213145","0.52023673","0.52003676","0.52003676","0.52003264","0.519784","0.5194341","0.5183786","0.51832813","0.51829386","0.5180746","0.5177552","0.5177111","0.515961","0.515961","0.5153012","0.5153012","0.51518196","0.515104","0.51489824","0.51480114","0.5146603","0.5132463","0.51258785","0.51216644","0.51199245","0.5118482","0.51113147","0.5106885","0.5105249","0.5104812","0.50971913","0.5096562","0.50960606","0.5085358","0.5080008","0.50798965"],"string":"[\n \"0.63869154\",\n \"0.6225311\",\n \"0.6225311\",\n \"0.6225311\",\n \"0.6225311\",\n \"0.6144163\",\n \"0.5890172\",\n \"0.58513165\",\n \"0.58451253\",\n \"0.5807999\",\n \"0.57855886\",\n \"0.57809436\",\n \"0.5763071\",\n \"0.5763071\",\n \"0.57493734\",\n \"0.5724776\",\n \"0.572035\",\n \"0.5714243\",\n \"0.5679732\",\n \"0.5625895\",\n \"0.55907756\",\n \"0.55830616\",\n \"0.55718523\",\n \"0.555276\",\n \"0.553306\",\n \"0.5528688\",\n \"0.55265915\",\n \"0.55073893\",\n \"0.54892355\",\n \"0.54816544\",\n \"0.54693997\",\n \"0.54672605\",\n \"0.5447692\",\n \"0.5399855\",\n \"0.53814375\",\n \"0.5361442\",\n \"0.53557247\",\n \"0.5355606\",\n \"0.535154\",\n \"0.53432065\",\n \"0.53325236\",\n \"0.5330406\",\n \"0.5323166\",\n \"0.53179884\",\n \"0.5307609\",\n \"0.5303047\",\n \"0.52949667\",\n \"0.52912444\",\n \"0.5285322\",\n \"0.5279463\",\n \"0.527885\",\n \"0.5276315\",\n \"0.52585864\",\n \"0.5255246\",\n \"0.5253941\",\n \"0.52438843\",\n \"0.523733\",\n \"0.5231569\",\n \"0.52264756\",\n \"0.5224379\",\n \"0.52192605\",\n \"0.5215848\",\n \"0.5213145\",\n \"0.5213145\",\n \"0.52023673\",\n \"0.52003676\",\n \"0.52003676\",\n \"0.52003264\",\n \"0.519784\",\n \"0.5194341\",\n \"0.5183786\",\n \"0.51832813\",\n \"0.51829386\",\n \"0.5180746\",\n \"0.5177552\",\n \"0.5177111\",\n \"0.515961\",\n \"0.515961\",\n \"0.5153012\",\n \"0.5153012\",\n \"0.51518196\",\n \"0.515104\",\n \"0.51489824\",\n \"0.51480114\",\n \"0.5146603\",\n \"0.5132463\",\n \"0.51258785\",\n \"0.51216644\",\n \"0.51199245\",\n \"0.5118482\",\n \"0.51113147\",\n \"0.5106885\",\n \"0.5105249\",\n \"0.5104812\",\n \"0.50971913\",\n \"0.5096562\",\n \"0.50960606\",\n \"0.5085358\",\n \"0.5080008\",\n \"0.50798965\"\n]","isConversational":false},"document_score":{"kind":"string","value":"0.7528839"},"document_rank":{"kind":"string","value":"0"}}},{"rowIdx":93288,"cells":{"query":{"kind":"string","value":"User stops following something."},"document":{"kind":"string","value":"def on_deleted_follow(sender, instance: models_actstream.Follow, **kwargs):\n content_type = ContentType.objects.get_for_id(instance.content_type_id)\n log.debug(\"Unfollowing %s %s\" % (content_type.name, instance.object_id))\n dillo.tasks.feeds.repopulate_timeline_content(\n instance.content_type_id, instance.object_id, instance.user_id, 'unfollow'\n )"},"metadata":{"kind":"string","value":"{\n \"objective\": {\n \"self\": [],\n \"paired\": [],\n \"triplet\": [\n [\n \"query\",\n \"document\",\n \"negatives\"\n ]\n ]\n }\n}"},"negatives":{"kind":"list like","value":["def stop(self):\r\n self.terminating = True","def stop(self) -> None:","def stop(self) -> None:","def stop(self):\r\n pass","def stop(self):","def stop(self):","def stop(self):\n return","def stop() -> None:","def stop(self):\n pass","def stop(self):\n pass","def stop(self):\n pass","def stop(self):\n pass","def stop(self):\n pass","def stop(self):\n pass","def stop(self):\n pass","def stop(self):\n pass","def stop(self):\n pass","def stop(self):\n pass","def stop(self):\n pass","def stop (self):\n pass","def stop (self):\n pass","def stop(self):\n\t\tpass","def _stop(self):\n self.display_end_message()","def stop():","def stop():","def stop():","def stop():","def stop(self) -> None:\n ...","def stop(self) -> None:\n pass","def stop(self):\n self.idle = True\n # pass","def Stop(self) :\n\t\t...","def kill(self):\n if self.living == True:\n self.living = False\n self.arrow_enter_callback(self)","def _stop(self):","def stop(self):\r\n self.stopped = True","def stop(self):\n self.active = False","def stop(self):\n self.halt = True","def stop(self):\n\n self.active = False","def stop(self):\n raise NotImplementedError","def stop(self):\n raise NotImplementedError","def stop(self):\n self.stopped = True\n # FIXME?: Why is this not doing anything? Shouldn't it be calling into the player API?","def _stop(self):\n return True","def stop(self):\r\n self.running = False","def stop(self):\r\n self.running = False","def stop(self):\n raise NotImplementedError()","def stop(self):\n raise NotImplementedError()","def stop(self):\n self.alive = False","def stop(self):\n self.alive = False","def stop(self):\r\n self.stopped = True\r\n time.sleep(3)","def stop(self):\n self.stopped = True","def stop(self):\n self.stopped = True","def stop(self):\n self.finished = True","def stop(self):\n self.gripper_io.set_signal_value(\"go\", False)","def stop(self):\r\n self.inst.write(':STOP')","def stop(self):\r\n self.stopped = True\r\n time.sleep(1)","def eStop(self):\n Step(speed=0, coils=1, steps=0, dir=Step.BRAKE)\n # #######################################################\n # Need to blink Stop and wait until Stop is pressed again\n # #######################################################","def post_stop(self):","def _stop_bot(_event):\n pass","def stop(self):\n self.unhook()","def stop(self):\n self._stop_flag = True","def on_stop(self):\n pass","def on_stop(self):\n pass","def on_stop(self):\n pass","def on_stop(self):\n pass","def on_stop(self):\n pass","def on_stop(self):\n pass","def on_stop(self):\n pass","def foreceStop(self):\n self.__success = False\n self.stop()","def stop(self) -> None:\n raise NotImplementedError()","def stop(self) -> None:\n raise NotImplementedError()","def bail(self, msg):\n self.logger.error(\"Can't handle follower result: {}\".format(msg))\n self.call('ctrl', 'stop_full')\n sys.exit(1)","def stop(self):\n self.running = False","def stop(self):\n self.running = False","def stop(self):\n self.running = False","def stop(self):\n self.running = False","def stop(self):\n self.running = False","def stop(self):\n self.on_stop()","def stop(self):\n self.ctrl_obj.finish = True","def stop(self):\n self.ctrl_obj.finish = True","def stop(self):\n print(\"Stopping accessory.\")","def __followerExit(self):\r\n self.phone.select('KBD_KEY_KEYLOCK_TOGGLE', doNotReport=True)\r\n self.phone.delay(5000, False)\r\n self.phone.select('KBD_KEY_KEYLOCK_TOGGLE', doNotReport=True)\r\n self.phone.delay(1000, False)\r\n self.__backToIdleWithSwipe() # open screen lock\r\n self.phone.delay(1500, False)\r\n self.__backToIdleWithSwipe() # close application with swipe\r\n self.phone.delay(1000, False)\r\n\r\n self.__backToIdleWithBackPress(True) # if still in some application, this could close it\r\n self.phone.delay(500, False)\r\n self.__backToIdleWithBackPress() # return to idle screen top with 2 back press\r\n self.phone.delay(500, False)\r\n self.__backToIdleWithBackPress()\r\n self.phone.delay(2000, False)","def stop(self):\n self.exit.set()","def stop(self):\n self.exit.set()","def on_stop(self):\n print(\"策略停止\")\n\n self.put_event()","def stop(self):\n \r\n self.__stopevent.set()\r\n\n return True","def on_stop(self):\n self.write_log(\"策略停止\")\n self.put_event()","def stop(self):\n # All done!\n super().stop()","def stopDetection(self):\n self.statusWrite(\"stop\")\n self.p.sleep()\n self.birdHere = 0","def on_stop(self):\n self.write_log(\"策略停止\")\n\n self.put_event()","def stop(self):\n self.stop_event.set()","def stop(self):\n self.stop_event.set()","def stop(self):\n self.__publish_cmd(0.0, 0.0)\n\n return","def stop(self):\n self._stop_signal = True","def stop(self):\n self.move(0, 0)","def stop_check(self):\n pass","def stop(self) -> None:\n raise NotImplementedError(\"Base method not implemented\")","def stop(self):\n self.send_stop()\n self.join()","def stop(self):\n\n self.keep_running = False","def stop(self):\n self.running = False\n self.hop_channel(\"auto\")","def force_stop(self):\n #cancel any current request:\n self._cancel_current_request()","def stop(self):\r\n raise NotImplementedError('method stop() is not implemented')","def stop(self):\n super().stop()"],"string":"[\n \"def stop(self):\\r\\n self.terminating = True\",\n \"def stop(self) -> None:\",\n \"def stop(self) -> None:\",\n \"def stop(self):\\r\\n pass\",\n \"def stop(self):\",\n \"def stop(self):\",\n \"def stop(self):\\n return\",\n \"def stop() -> None:\",\n \"def stop(self):\\n pass\",\n \"def stop(self):\\n pass\",\n \"def stop(self):\\n pass\",\n \"def stop(self):\\n pass\",\n \"def stop(self):\\n pass\",\n \"def stop(self):\\n pass\",\n \"def stop(self):\\n pass\",\n \"def stop(self):\\n pass\",\n \"def stop(self):\\n pass\",\n \"def stop(self):\\n pass\",\n \"def stop(self):\\n pass\",\n \"def stop (self):\\n pass\",\n \"def stop (self):\\n pass\",\n \"def stop(self):\\n\\t\\tpass\",\n \"def _stop(self):\\n self.display_end_message()\",\n \"def stop():\",\n \"def stop():\",\n \"def stop():\",\n \"def stop():\",\n \"def stop(self) -> None:\\n ...\",\n \"def stop(self) -> None:\\n pass\",\n \"def stop(self):\\n self.idle = True\\n # pass\",\n \"def Stop(self) :\\n\\t\\t...\",\n \"def kill(self):\\n if self.living == True:\\n self.living = False\\n self.arrow_enter_callback(self)\",\n \"def _stop(self):\",\n \"def stop(self):\\r\\n self.stopped = True\",\n \"def stop(self):\\n self.active = False\",\n \"def stop(self):\\n self.halt = True\",\n \"def stop(self):\\n\\n self.active = False\",\n \"def stop(self):\\n raise NotImplementedError\",\n \"def stop(self):\\n raise NotImplementedError\",\n \"def stop(self):\\n self.stopped = True\\n # FIXME?: Why is this not doing anything? Shouldn't it be calling into the player API?\",\n \"def _stop(self):\\n return True\",\n \"def stop(self):\\r\\n self.running = False\",\n \"def stop(self):\\r\\n self.running = False\",\n \"def stop(self):\\n raise NotImplementedError()\",\n \"def stop(self):\\n raise NotImplementedError()\",\n \"def stop(self):\\n self.alive = False\",\n \"def stop(self):\\n self.alive = False\",\n \"def stop(self):\\r\\n self.stopped = True\\r\\n time.sleep(3)\",\n \"def stop(self):\\n self.stopped = True\",\n \"def stop(self):\\n self.stopped = True\",\n \"def stop(self):\\n self.finished = True\",\n \"def stop(self):\\n self.gripper_io.set_signal_value(\\\"go\\\", False)\",\n \"def stop(self):\\r\\n self.inst.write(':STOP')\",\n \"def stop(self):\\r\\n self.stopped = True\\r\\n time.sleep(1)\",\n \"def eStop(self):\\n Step(speed=0, coils=1, steps=0, dir=Step.BRAKE)\\n # #######################################################\\n # Need to blink Stop and wait until Stop is pressed again\\n # #######################################################\",\n \"def post_stop(self):\",\n \"def _stop_bot(_event):\\n pass\",\n \"def stop(self):\\n self.unhook()\",\n \"def stop(self):\\n self._stop_flag = True\",\n \"def on_stop(self):\\n pass\",\n \"def on_stop(self):\\n pass\",\n \"def on_stop(self):\\n pass\",\n \"def on_stop(self):\\n pass\",\n \"def on_stop(self):\\n pass\",\n \"def on_stop(self):\\n pass\",\n \"def on_stop(self):\\n pass\",\n \"def foreceStop(self):\\n self.__success = False\\n self.stop()\",\n \"def stop(self) -> None:\\n raise NotImplementedError()\",\n \"def stop(self) -> None:\\n raise NotImplementedError()\",\n \"def bail(self, msg):\\n self.logger.error(\\\"Can't handle follower result: {}\\\".format(msg))\\n self.call('ctrl', 'stop_full')\\n sys.exit(1)\",\n \"def stop(self):\\n self.running = False\",\n \"def stop(self):\\n self.running = False\",\n \"def stop(self):\\n self.running = False\",\n \"def stop(self):\\n self.running = False\",\n \"def stop(self):\\n self.running = False\",\n \"def stop(self):\\n self.on_stop()\",\n \"def stop(self):\\n self.ctrl_obj.finish = True\",\n \"def stop(self):\\n self.ctrl_obj.finish = True\",\n \"def stop(self):\\n print(\\\"Stopping accessory.\\\")\",\n \"def __followerExit(self):\\r\\n self.phone.select('KBD_KEY_KEYLOCK_TOGGLE', doNotReport=True)\\r\\n self.phone.delay(5000, False)\\r\\n self.phone.select('KBD_KEY_KEYLOCK_TOGGLE', doNotReport=True)\\r\\n self.phone.delay(1000, False)\\r\\n self.__backToIdleWithSwipe() # open screen lock\\r\\n self.phone.delay(1500, False)\\r\\n self.__backToIdleWithSwipe() # close application with swipe\\r\\n self.phone.delay(1000, False)\\r\\n\\r\\n self.__backToIdleWithBackPress(True) # if still in some application, this could close it\\r\\n self.phone.delay(500, False)\\r\\n self.__backToIdleWithBackPress() # return to idle screen top with 2 back press\\r\\n self.phone.delay(500, False)\\r\\n self.__backToIdleWithBackPress()\\r\\n self.phone.delay(2000, False)\",\n \"def stop(self):\\n self.exit.set()\",\n \"def stop(self):\\n self.exit.set()\",\n \"def on_stop(self):\\n print(\\\"策略停止\\\")\\n\\n self.put_event()\",\n \"def stop(self):\\n \\r\\n self.__stopevent.set()\\r\\n\\n return True\",\n \"def on_stop(self):\\n self.write_log(\\\"策略停止\\\")\\n self.put_event()\",\n \"def stop(self):\\n # All done!\\n super().stop()\",\n \"def stopDetection(self):\\n self.statusWrite(\\\"stop\\\")\\n self.p.sleep()\\n self.birdHere = 0\",\n \"def on_stop(self):\\n self.write_log(\\\"策略停止\\\")\\n\\n self.put_event()\",\n \"def stop(self):\\n self.stop_event.set()\",\n \"def stop(self):\\n self.stop_event.set()\",\n \"def stop(self):\\n self.__publish_cmd(0.0, 0.0)\\n\\n return\",\n \"def stop(self):\\n self._stop_signal = True\",\n \"def stop(self):\\n self.move(0, 0)\",\n \"def stop_check(self):\\n pass\",\n \"def stop(self) -> None:\\n raise NotImplementedError(\\\"Base method not implemented\\\")\",\n \"def stop(self):\\n self.send_stop()\\n self.join()\",\n \"def stop(self):\\n\\n self.keep_running = False\",\n \"def stop(self):\\n self.running = False\\n self.hop_channel(\\\"auto\\\")\",\n \"def force_stop(self):\\n #cancel any current request:\\n self._cancel_current_request()\",\n \"def stop(self):\\r\\n raise NotImplementedError('method stop() is not implemented')\",\n \"def stop(self):\\n super().stop()\"\n]","isConversational":false},"negative_scores":{"kind":"list like","value":["0.758307","0.7484069","0.7484069","0.7418194","0.73704475","0.73704475","0.7342334","0.733363","0.73308474","0.73308474","0.73308474","0.73308474","0.73308474","0.73308474","0.73308474","0.73308474","0.73308474","0.73308474","0.73308474","0.7324925","0.7324925","0.72930723","0.72908324","0.7244097","0.7244097","0.7244097","0.7244097","0.7204508","0.7120815","0.7097954","0.7086419","0.7018418","0.7017975","0.6991467","0.6976136","0.6967564","0.69059896","0.68981564","0.68981564","0.68929225","0.68862945","0.6883779","0.6883779","0.6867247","0.6867247","0.68545496","0.68545496","0.68450063","0.68343157","0.68343157","0.68312025","0.68290526","0.6826115","0.6824478","0.6808517","0.6806112","0.6802405","0.679905","0.6789301","0.6787819","0.6787819","0.6787819","0.6787819","0.6787819","0.6787819","0.6787819","0.6776511","0.67740303","0.67740303","0.67681134","0.67591","0.67591","0.67591","0.67591","0.67591","0.6756133","0.6754188","0.6754188","0.6751404","0.6748778","0.6747652","0.6747652","0.67437917","0.6738537","0.6731641","0.67087644","0.6693993","0.66835546","0.66691285","0.66691285","0.6667604","0.6665123","0.6656895","0.66551805","0.66530263","0.66529006","0.6652325","0.66487235","0.6646065","0.6628231","0.66262805"],"string":"[\n \"0.758307\",\n \"0.7484069\",\n \"0.7484069\",\n \"0.7418194\",\n \"0.73704475\",\n \"0.73704475\",\n \"0.7342334\",\n \"0.733363\",\n \"0.73308474\",\n \"0.73308474\",\n \"0.73308474\",\n \"0.73308474\",\n \"0.73308474\",\n \"0.73308474\",\n \"0.73308474\",\n \"0.73308474\",\n \"0.73308474\",\n \"0.73308474\",\n \"0.73308474\",\n \"0.7324925\",\n \"0.7324925\",\n \"0.72930723\",\n \"0.72908324\",\n \"0.7244097\",\n \"0.7244097\",\n \"0.7244097\",\n \"0.7244097\",\n \"0.7204508\",\n \"0.7120815\",\n \"0.7097954\",\n \"0.7086419\",\n \"0.7018418\",\n \"0.7017975\",\n \"0.6991467\",\n \"0.6976136\",\n \"0.6967564\",\n \"0.69059896\",\n \"0.68981564\",\n \"0.68981564\",\n \"0.68929225\",\n \"0.68862945\",\n \"0.6883779\",\n \"0.6883779\",\n \"0.6867247\",\n \"0.6867247\",\n \"0.68545496\",\n \"0.68545496\",\n \"0.68450063\",\n \"0.68343157\",\n \"0.68343157\",\n \"0.68312025\",\n \"0.68290526\",\n \"0.6826115\",\n \"0.6824478\",\n \"0.6808517\",\n \"0.6806112\",\n \"0.6802405\",\n \"0.679905\",\n \"0.6789301\",\n \"0.6787819\",\n \"0.6787819\",\n \"0.6787819\",\n \"0.6787819\",\n \"0.6787819\",\n \"0.6787819\",\n \"0.6787819\",\n \"0.6776511\",\n \"0.67740303\",\n \"0.67740303\",\n \"0.67681134\",\n \"0.67591\",\n \"0.67591\",\n \"0.67591\",\n \"0.67591\",\n \"0.67591\",\n \"0.6756133\",\n \"0.6754188\",\n \"0.6754188\",\n \"0.6751404\",\n \"0.6748778\",\n \"0.6747652\",\n \"0.6747652\",\n \"0.67437917\",\n \"0.6738537\",\n \"0.6731641\",\n \"0.67087644\",\n \"0.6693993\",\n \"0.66835546\",\n \"0.66691285\",\n \"0.66691285\",\n \"0.6667604\",\n \"0.6665123\",\n \"0.6656895\",\n \"0.66551805\",\n \"0.66530263\",\n \"0.66529006\",\n \"0.6652325\",\n \"0.66487235\",\n \"0.6646065\",\n \"0.6628231\",\n \"0.66262805\"\n]","isConversational":false},"document_score":{"kind":"string","value":"0.0"},"document_rank":{"kind":"string","value":"-1"}}},{"rowIdx":93289,"cells":{"query":{"kind":"string","value":"If confirmed, subscribe to newsletter."},"document":{"kind":"string","value":"def on_email_confirmed(request, email_address: EmailAddress, **kwargs):\n dillo.tasks.profile.update_mailing_list_subscription(email_address.email, True)"},"metadata":{"kind":"string","value":"{\n \"objective\": {\n \"self\": [],\n \"paired\": [],\n \"triplet\": [\n [\n \"query\",\n \"document\",\n \"negatives\"\n ]\n ]\n }\n}"},"negatives":{"kind":"list like","value":["def subscribe(self, request):\n email = self.cleaned_data.get('email')\n\n email_name, domain_part = email.rsplit('@', 1)\n domain_name = '@' + domain_part\n email_domain, created = Domain.objects.get_or_create(name=domain_name)\n\n subscriber, created = Subscriber.objects.get_or_create(email=email, mailing_list=self.mailing_list, defaults={\n 'domain': email_domain\n })\n subscriber.status = Status.PENDING\n subscriber.optin_ip_address = get_client_ip(request)\n subscriber.optin_date = timezone.now()\n subscriber.save()\n\n if not created:\n subscriber.tokens.filter(description='confirm_subscription').delete()\n\n token = subscriber.tokens.create(description='confirm_subscription')\n current_site = get_current_site(request)\n protocol = 'https' if request.is_secure() else 'http'\n domain = current_site.domain\n path = reverse('subscribers:confirm_double_optin_token', kwargs={\n 'mailing_list_uuid': self.mailing_list.uuid,\n 'token': token.text\n })\n confirm_link = '%s://%s%s' % (protocol, domain, path)\n\n confirm_email = self.mailing_list.get_confirm_email_template()\n confirm_email.send(subscriber.get_email(), {\n 'confirm_link': confirm_link\n })\n\n return subscriber","async def aoc_subscribe(self, ctx: commands.Context) -> None:\n if ctx.channel.id != settings.aoc.channel_id:\n await ctx.send(f\"Please use the <#{settings.aoc.channel_id}> channel\")\n return\n\n role = ctx.guild.get_role(settings.aoc.role_id)\n unsubscribe_command = f\"{ctx.prefix}{ctx.command.root_parent} unsubscribe\"\n\n if role not in ctx.author.roles:\n await ctx.author.add_roles(role)\n await ctx.send(\n \"Okay! You have been __subscribed__ to notifications about new Advent of Code tasks. \"\n f\"You can run `{unsubscribe_command}` to disable them again for you.\"\n )\n else:\n await ctx.send(\n \"Hey, you already are receiving notifications about new Advent of Code tasks. \"\n f\"If you don't want them any more, run `{unsubscribe_command}` instead.\"\n )","async def aoc_subscribe(self, ctx: commands.Context) -> None:\n role = ctx.guild.get_role(AocConfig.role_id)\n unsubscribe_command = f\"{ctx.prefix}{ctx.command.root_parent} unsubscribe\"\n\n if role not in ctx.author.roles:\n await ctx.author.add_roles(role)\n await ctx.send(\"Okay! You have been __subscribed__ to notifications about new Advent of Code tasks. \"\n f\"You can run `{unsubscribe_command}` to disable them again for you.\")\n else:\n await ctx.send(\"Hey, you already are receiving notifications about new Advent of Code tasks. \"\n f\"If you don't want them any more, run `{unsubscribe_command}` instead.\")","def _subscribe(self):\n self.subscribed = True\n self.subscribe_date = now()\n self.unsubscribed = False","def subscribe(self, update, context):\n # add or update to the sqlite table.\n chat = update.message.chat\n user_tuple = self.db_manager.create_user_tuple(chat.id)\n try:\n status = self.db_manager.add_new_user(user_tuple)\n except IntegrityError:\n self.logger.info('Username: %s and chat_id: %s is already subscribed to the list.')\n update.message.reply_text('You are already subscribed to real time notification for a six '\n 'scored in IPL 2019 cricket match.')\n return True\n self.logger.info(\n 'Username: %s and chat_id: %s subscribed to the list.' % (chat.username, chat.id)\n )\n update.message.reply_text('Welcome! You have successfully subscribed to real time notification for a six '\n 'scored in IPL 2019 cricket match.')","def view_mailinglist_subscribe(request, form_class, mailing_list_id=None):\n subscribed = False\n mailing_list = None\n if mailing_list_id:\n mailing_list = get_object_or_404(MailingList, id=mailing_list_id)\n\n if request.POST and not subscribed:\n form = form_class(request.POST)\n if form.is_valid():\n form.save(mailing_list)\n subscribed = True\n else:\n form = form_class()\n\n return render_to_response('newsletter/mailing_list_subscribe.html',\n {'subscribed': subscribed,\n 'mailing_list': mailing_list,\n 'form': form},\n context_instance=RequestContext(request))","def _handle_subscribe(self, presence):\n presence.reply()\n presence['to'] = presence['to'].bare\n\n # We are using trinary logic, so conditions have to be\n # more explicit than usual.\n if self.auto_authorize == True:\n presence['type'] = 'subscribed'\n presence.send()\n if self.auto_subscribe:\n presence['type'] = 'subscribe'\n presence.send()\n elif self.auto_authorize == False:\n presence['type'] = 'unsubscribed'\n presence.send()","def subscribe( self, mess, args):\n user = self.get_sender_username(mess)\n if user in self.users:\n return 'You are already subscribed.'\n else:\n self.users[user] = user\n self.invited.pop(user)\n self.message_queue.append('_%s has joined the channel_' % user)\n self.log.info('%s subscribed to the broadcast.' % user)\n self.save_state()\n return 'You are now subscribed.'","def email_subscribe_pending_confirm(hexdomain):\n domain = tools.parse_domain(hexdomain)\n if domain is None:\n flask.abort(400, 'Malformed domain or domain not represented in hexadecimal format.')\n\n email_address = flask.request.form['email_address']\n\n if email_address.strip() == '':\n return flask.redirect('/email/subscribe/{}/0'.format(hexdomain))\n\n verify_code = tools.random_id()\n verify_url = flask.request.url_root + 'email/verify/{}'.format(verify_code)\n email_body = email_tools.render_email(\n 'confirm.html',\n domain=domain,\n verify_url=verify_url\n )\n\n repository.propose_subscription(verify_code, email_address, domain)\n emailer.send(\n email_address, 'Please verify your subscription', email_body\n )\n\n return flask.render_template('www/email/pending_verify.html', domain=domain)","def test_send_subscribe_email(self):\n #Verifica se foi enviado 1 e-mail, o este não envia e-mail\n self.assertEqual(1, len(mail.outbox))","def on_subscribe(self, request, ssn):\n if 'hub.challenge' not in request.GET:\n logger.error(f'Missing hub.challenge in subscription verification {ssn.pk}!')\n tasks.save.delay(\n pk = ssn.pk,\n subscribe_status = 'verifyerror',\n verifyerror_count = ssn.verifyerror_count + 1\n )\n return Response('Missing hub.challenge', status=HTTP_400_BAD_REQUEST)\n\n if not request.GET.get('hub.lease_seconds', '').isdigit():\n logger.error(f'Missing integer hub.lease_seconds in subscription verification {ssn.pk}!')\n tasks.save.delay(\n pk = ssn.pk,\n subscribe_status = 'verifyerror',\n verifyerror_count = ssn.verifyerror_count + 1\n )\n return Response('hub.lease_seconds required and must be integer', status=HTTP_400_BAD_REQUEST)\n\n if ssn.unsubscribe_status is not None:\n logger.error(f'Subscription {ssn.pk} received subscription verification request,'\n f' but its was explicitly unsubscribed before.')\n return Response('Unsubscribed')\n\n tasks.save.delay(\n pk = ssn.pk,\n subscribe_status = 'verified',\n lease_expiration_time = now() + timedelta(seconds=int(request.GET['hub.lease_seconds'])),\n connerror_count = 0,\n huberror_count = 0,\n verifyerror_count = 0,\n verifytimeout_count = 0\n )\n logger.info(f'Got {ssn.pk} subscribe confirmation from hub.')\n return HttpResponse(request.GET['hub.challenge'])","def _want_subscription() -> bool:\n prompt = (\n 'Would you be willing, once your first certificate is successfully issued, '\n 'to share your email address with the Electronic Frontier Foundation, a '\n \"founding partner of the Let's Encrypt project and the non-profit organization \"\n \"that develops Certbot? We'd like to send you email about our work encrypting \"\n \"the web, EFF news, campaigns, and ways to support digital freedom. \")\n return display_util.yesno(prompt, default=False)","def subscribe():\n form = SubscribeForm()\n if form.validate_on_submit():\n subscription = Subscription(email=form.email.data)\n db.session.add(subscription)\n db.session.commit()\n return redirect(url_for('main.index'))\n\n return render_template('subscribes.html', form=form)","def cbMqtt_on_subscribe(client, userdata, mid, granted_qos):\n # logger.debug('Subscribed to MQTT topic with message id %d', mid)\n pass","def cb_start(self, update, context):\n chat_id = update.message.chat_id\n\n if chat_id in self.clientChatIds:\n update.message.reply_text('You are already subscribed.')\n return\n\n print(f\"Subscribing chat_id '{chat_id}'\")\n self.clientChatIds.add(chat_id)\n update.message.reply_text(\"You have successfully subscribed.\")\n self.saveToFile(self.configFile)","def send_confirmation(self):\r\n c.user.email_validated = False\r\n c.user.confirmation_code = random_key(6)\r\n c.user._commit()\r\n emailer.confirmation_email(c.user)","async def enable_publish_confirms(self) -> None:\n if self.channel:\n if not self.channel.publisher_confirms:\n await self.channel.confirm_select()","def mqtt_on_connect(client, userdata, flags, rc):\n logging.debug('successfully connected to mqtt broker')\n client.subscribe(config['mqtt']['subscribe_topic'])","def form_valid(self, form):\n form.save()\n return super(SubscribeView, self).form_valid(form)","def confirm_email(self):\n self.active = True\n self.save()","def subscription(bot, update):\n chat_id = update.message.chat_id\n bot.sendMessage(chat_id=chat_id, text=SUBSCRIPTION_MSG, parse_mode='markdown', \n disable_web_page_preview=True)\n \n mp.track(get_user_info(chat_id)['PID'], 'Checked Subscription')","def subscribe(self, feed, **args):\n args.update(feed=feed)\n return self.fetch(\"/subscribe\", post_args=args)","def send_confirm_email(request,uid):\n user=models.UserProfile.objects.get(id=uid)\n current_site=get_current_site(request)\n email_subject='Activate Your Account'\n message=render_to_string('activate_account.html',{\n 'user':user,\n 'domain':current_site.domain,\n 'uid':urlsafe_base64_encode(force_bytes(uid)),\n 'token':account_activation_token.make_token(user),\n })\n to_email= user.email\n email= EmailMessage(email_subject,message,to=[to_email])\n email.send()\n return JsonResponse(\n {\n \"status\":\"The confirmation email has been sent.\",\n }\n )","def subscribe( self, topic ):\n logging.info( \"Subscribing to topic %s\" %topic )\n try:\n self.client.subscribe( topic )\n except Exception as error:\n print( error )","def subscribe(self, subject):\n pass","def test_successful_subscriptions_notifies_pm(self) -> None:\n invitee = self.example_user(\"iago\")\n\n current_stream = self.get_streams(invitee)[0]\n invite_streams = self.make_random_stream_names([current_stream])[:1]\n self.common_subscribe_to_streams(\n invitee,\n invite_streams,\n extra_post_data={\n \"announce\": \"true\",\n \"principals\": orjson.dumps([self.user_profile.id]).decode(),\n },\n )","def confirmed_email_notification(sender, **kwargs):\n subject_template_name = 'shop_simplenotifications/confirmed_subject.txt'\n body_text_template_name = 'shop_simplenotifications/confirmed_body.txt'\n body_html_template_name = 'shop_simplenotifications/confirmed_body.html'\n request = kwargs.get('request')\n order = kwargs.get('order')\n subject = loader.render_to_string(\n subject_template_name,\n RequestContext(request, {'order': order})\n )\n subject = subject.join(subject.splitlines())\n\n text_content = loader.render_to_string(\n body_text_template_name,\n RequestContext(request, {'order': order})\n )\n try:\n html_content = loader.render_to_string(\n body_html_template_name,\n RequestContext(request, {'order': order})\n )\n except TemplateDoesNotExist:\n html_content = None\n\n from_email = getattr(settings, 'SN_FROM_EMAIL',\n settings.DEFAULT_FROM_EMAIL)\n owners = getattr(settings, 'SN_OWNERS', settings.ADMINS)\n\n message = EmailMultiAlternatives(subject, text_content, from_email,\n [owner[1] for owner in owners])\n if html_content:\n message.attach_alternative(html_content, \"text/html\")\n message.send()","def on_subscribe(\n client: mqtt.Client,\n userdata: Any,\n mid: int,\n granted_qos: int,\n properties: Properties = None,\n ) -> None:\n logging.info(\n f\"Successfully subscribed to topic: mid={mid}, granted qos={granted_qos}, properties={properties}\"\n )","def activate(self):\n if not self.is_active:\n self.is_active = True\n self.activated_at = datetime.datetime.utcnow()\n import messaging # avoid circular import\n messaging.send_activated_emails(self)\n self.save()","def subscribe(self, feedUrl):\r\n response = self.httpPost(\r\n ReaderUrl.SUBSCRIPTION_EDIT_URL,\r\n {'ac':'subscribe', 's': feedUrl})\r\n # FIXME - need better return API\r\n if response and 'OK' in response:\r\n return True\r\n else:\r\n return False","def subscribe(self, user_token, topic):\n response = _request('POST',\n url=self.url_v1('/user/subscriptions/' + topic),\n user_agent=self.user_agent,\n user_token=user_token,\n )\n _raise_for_status(response)","async def send_confirm_link(email: TextData, background_tasks: BackgroundTasks):\n email = email.data\n mail, subject, body = await AccountProcessor.send_confirmation_link(email)\n background_tasks.add_task(Utility.validate_and_send_mail, email=mail, subject=subject, body=body)\n return {\"message\": \"Success! Confirmation link sent\"}","def notify_thesis_accepted(thesis: 'Thesis'):\n recipients = []\n if thesis.advisor:\n recipients.append(thesis.advisor.user.email)\n if thesis.supporting_advisor:\n recipients.append(thesis.supporting_advisor.user.email)\n if recipients:\n formatted_body = ACCEPTED_BODY.format(thesis.title)\n send_mail(ACCEPTED_SUBJECT, formatted_body, settings.MASS_MAIL_FROM, recipients)","def confirm_subscription(self, sub_code):\n\t\tresult = {}\n\t\tconnection = DbHelper.connect()\n\t\tsub_id = 0\n\t\t# print(\"Subscription Code: \" + sub_code)\n\n\t\ttry:\n\t\t\twith connection.cursor() as cursor:\n\t\t\t\tsql = \"SELECT * FROM mail_list \\\n\t\t\t\t\t WHERE email_hash=%s;\"\n\t\t\t\tcursor.execute(sql, [sub_code])\n\t\t\t\tresult = cursor.fetchone()\n\t\t\t\t\n\t\t\t\tif not result:\n\t\t\t\t\tconnection.close()\n\t\t\t\t\treturn \"CODE_DOES_NOT_EXIST\"\n\t\t\t\t\n\t\t\t\telif result['is_activated']:\n\t\t\t\t\tconnection.close()\n\t\t\t\t\treturn \"CODE_ALREADY_ACTIVATED\"\n\n\t\t\t\tsub_id = result['sub_id']\n\n\t\t\t\tsql = \"UPDATE mail_list \\\n\t\t\t\t\t SET is_activated=is_activated+1 \\\n\t\t\t\t\t WHERE sub_id=%s;\"\n\t\t\t\tcursor.execute(sql, [sub_id])\n\t\t\t\tconnection.commit()\n\t\t\t\tconnection.close()\n\t\t\t\treturn result\n\t\texcept pymysql.MySQLError as e:\n\t\t\tconnection.close()\n\t\t\treturn \"DATABASE_ERROR\"","def test_subscribe(self):\n self.service.clientConnected()\n self.service.subscribe(u'url', None)\n pubsubClient = self.service.pubsubClient\n self.assertIn(u'url', pubsubClient.subscriptions)","def _subscribe(self):\n if self.subscribed:\n return False\n return {}","def on_subscribe( client, userdata, mid, granted_qos ):\n logging.info( \"Topic successfully subcribed with QoS: %s\" %granted_qos )","def handle_email_confirmed(sender, **kwargs):\n email = kwargs['email_address']\n email.user.userprofile.member.cast().confirm_email()","def subscribe(self, topic):\n\t\tself.topic=topic\n\t\tself.client.subscribe(self.topic)","def _subscribe(self, callback_path: str, topic_url: str, mode: str = \"subscribe\"):\n data = {'hub.callback': self.callback_url + callback_path,\n 'hub.mode': mode,\n 'hub.topic': topic_url,\n 'hub.lease_seconds': self.subscribe_least_seconds}\n if self.secret is not None:\n data['hub.secret'] = self.secret\n result = self.__api_post_request(TWITCH_API_BASE_URL + \"webhooks/hub\", data=data)\n if result.status_code != 202:\n logging.error(f'Subscription failed! status code: {result.status_code}, body: {result.text}')\n return result.status_code == 202","def on_subscribe(self, mqtt_client, userdata, mid, granted_qos):\n logging.debug(\"DEBUG - subscribe ack received\")","def send_confirmation_email(user_pk):\n pass","def _on_connect(self, client, userdata, flags, rc):\n self.subscribe(self.topic)","def confirm_subscription(self, topic, token,\r\n authenticate_on_unsubscribe=False):\r\n params = {'ContentType' : 'JSON',\r\n 'TopicArn' : topic,\r\n 'Token' : token}\r\n if authenticate_on_unsubscribe:\r\n params['AuthenticateOnUnsubscribe'] = 'true'\r\n response = self.make_request('ConfirmSubscription', params, '/', 'GET')\r\n body = response.read()\r\n if response.status == 200:\r\n return json.loads(body)\r\n else:\r\n boto.log.error('%s %s' % (response.status, response.reason))\r\n boto.log.error('%s' % body)\r\n raise self.ResponseError(response.status, response.reason, body)","def onSubscribed(self, connection:MQTTConnection, topic:str) -> bool:\n\t\tconnection.subscribedCount += 1\n\t\treturn True","def render_email(self, user):\n if user.get_profile().subscribed_to_news:\n return EmailMessage(self.subject, self.body, from_email='anon@example.com', to=[user.email])","def _subscribed(self, account_id):\n sql = \"\"\"SELECT 1 FROM hive_subscriptions\n WHERE community_id = :community_id\n AND account_id = :account_id\"\"\"\n return bool(DB.query_one(\n sql, community_id=self.community_id, account_id=account_id))","def confirm_email(request, key):\n alt_email = cpm.Email.objects.filter(activation_key=key)\n if alt_email.exists():\n alt_email[0].confirm()\n return redirect('/')\n hero_title = 'We weren\\'t able to complete your request...'\n return render_err_msg(request, hero_title)","def subscribe_email_to_cloudwatch(self, email):\n return self.conn.subscribe(self.cloudwatch_arn, 'email', email)","def speaker_confirmation():\n\n talks = Talk.objects.all()\n\n template = 'notifications/speaker_confirmation_mail.html'\n\n for talk in talks:\n subject = 'SciPy.in 2011: Requesting for confirmation of your talk'\n message = loader.render_to_string(\n template, dictionary={'name': talk.speaker.username,\n 'title': talk.title})\n\n talk.speaker.email_user(subject=subject, message=message,\n from_email='admin@scipy.in')","def subscribe(request):\n address = request.POST.get('address')\n\n new_sub = Subscription(**{\n \"address\": address\n })\n new_sub.save()\n\n return HttpResponse(json.dumps({\n \"status\": \"success\"\n }, default=helpers.json_custom_parser), content_type='application/json')","def send_confirmation_email(self, *args, **kwargs):\n raise NotImplementedError","def subscribe_user(self, user):\n self.ensure_one()\n if self.has_user(user):\n # already subscribed\n return False\n return self.emulate_request(user)","def on_connect(client, userdata, flags, rc):\n if rc == 0:\n client.subscribe(topic_subscribe)\n print(\"connected OK with returned code=\", rc)\n else:\n print(\"Bad connection with returned code=\", rc)","def test_manually_confirm(self):\n data = {}\n response = self.client.post(self.url, data)\n self.assertRedirects(response, reverse('reminders_dashboard'))\n\n reminder = reminders.SentNotification.objects.get(pk=self.unconfirmed.pk)\n self.assertEqual(reminder.status, 'manual')\n self.assertEqual(reminder.date_confirmed.date(), datetime.date.today())","def test_manually_confirm(self):\n data = {}\n response = self.client.post(self.url, data)\n self.assertRedirects(response, reverse('reminders_dashboard'))\n\n reminder = reminders.SentNotification.objects.get(pk=self.unconfirmed.pk)\n self.assertEqual(reminder.status, 'manual')\n self.assertEqual(reminder.date_confirmed.date(), datetime.date.today())","def send_emails_to_subscribers(creator_id, exploration_id, exploration_title):\n\n creator_name = user_services.get_username(creator_id)\n email_subject = ('%s has published a new exploration!' % creator_name)\n email_body_template = (\n 'Hi %s,<br>'\n '<br>'\n '%s has published a new exploration! You can play it here: '\n '<a href=\"https://www.oppia.org/explore/%s\">%s</a><br>'\n '<br>'\n 'Thanks, and happy learning!<br>'\n '<br>'\n 'Best wishes,<br>'\n '- The Oppia Team<br>'\n '<br>%s')\n\n if not feconf.CAN_SEND_EMAILS:\n log_new_error('This app cannot send emails to users.')\n return\n\n if not feconf.CAN_SEND_SUBSCRIPTION_EMAILS:\n log_new_error('This app cannot send subscription emails to users.')\n return\n\n recipient_list = subscription_services.get_all_subscribers_of_creator(\n creator_id)\n recipients_usernames = user_services.get_usernames(recipient_list)\n recipients_preferences = user_services.get_users_email_preferences(\n recipient_list)\n for index, username in enumerate(recipients_usernames):\n if recipients_preferences[index].can_receive_subscription_email:\n email_body = email_body_template % (\n username, creator_name, exploration_id,\n exploration_title, EMAIL_FOOTER.value)\n _send_email(\n recipient_list[index], feconf.SYSTEM_COMMITTER_ID,\n feconf.EMAIL_INTENT_SUBSCRIPTION_NOTIFICATION,\n email_subject, email_body, feconf.NOREPLY_EMAIL_ADDRESS)","def subscribe(self, path):\n box = self._getMailbox(path)\n\n box.metadata['subscribed'] = True\n\n box.saveMetadata()\n\n return True","def post(self):\n data = request.get_json()\n user = actions.get_user_by_email(data['email'])\n html = '<p>Confirming your account will give you </p> <b>full access to Kwikker</b>'\n subject = 'Confirm your Kwikker account, ' + user['username']\n actions.send_email(data['email'], user['username'], user['password'], subject,\n '/confirm/', html, True)\n return \"\", 200\n pass","def _confirm_email(user, email):\n mail_subject = 'Подтверждение почты'\n message = render_to_string('accounts/account_verification_email.html', {\n 'user': user,\n 'uid': urlsafe_base64_encode(force_bytes(user.pk)),\n 'token': default_token_generator.make_token(user),\n 'email': email,\n })\n to_email = email\n send_email = EmailMessage(mail_subject, message, to=[to_email])\n send_email.send()","def subscribe(client, mqtt_topic):\n\n print()\n print(\"Subscribe\")\n print(\"================================================\")\n print()\n\n # Subscribe to the config topic.\n print(\"Subscribing\")\n print(mqtt_topic)\n print()\n client.subscribe(mqtt_topic, qos=1)","def view_subscriber_verification(request, form_class):\n context = {}\n\n if request.POST:\n context['form'] = form_class(request.POST)\n subscription = SubscriberVerification()\n if context['form'].is_valid():\n contact = context['form'].save()\n\n subscription.contact = context['form'].instance\n subscription.save()\n link_id = subscription.link_id\n\n # select smtpserver, maybe, a boolean field for 'standard' will be\n # a nice idea.\n server = SMTPServer.objects.get(id=1)\n\n # set server settings to django\n settings.EMAIL_HOST = smart_str(server.host)\n settings.EMAIL_HOST_PASSWORD = smart_str(server.password)\n settings.EMAIL_HOST_USER = smart_str(server.user)\n settings.EMAIL_PORT = int(server.port)\n settings.EMAIL_USE_TLS = server.tls\n\n # mail settings\n subject = _('Subriber verification')\n message = _('Thanks for subscription, please click the following link to verificate your email address.')\n from_mail = DEFAULT_HEADER_REPLY\n to_mail = context['form'].instance.email\n\n link = 'http://{0}/newsletters/subscribe/{1}'.format(str(request.get_host()), str(link_id))\n text_content = u'{0!s}\\n\\n {1}'.format(message, link)\n html_content = u'<body><p>{0!s}</p><p><a href=\"{1}\">{2}</a></p></body>'.format(message, link, _('verify link'))\n msg = EmailMultiAlternatives(\n subject,\n text_content,\n from_mail,\n [to_mail]\n )\n msg.attach_alternative(html_content, 'text/html')\n msg.send()\n\n context['send'] = True\n\n else:\n context['form'] = form_class()\n\n return render_to_response('newsletter/subscriber_verification.html',\n context,\n context_instance=RequestContext(request))","def on_connect(client, userdata, flags, rc):\n print(f\"Re/Suscribing to TOPIC: {TOPIC}\")\n client.subscribe(TOPIC)\n if rc == 0:\n print(f'Connected OK Returned code={rc}')\n else:\n print('Bad connection Returned code={rc}')","def save(self, *args, **kwargs):\n if not self.require_confirm_email:\n User.objects.filter(is_active=False, deactivation_reason=\"pending\").update(\n is_active=True, deactivation_reason=None\n )\n if not self.invite_question_text:\n self.invite_question_text = \"What is your favourite book?\"\n super().save(*args, **kwargs)","def subscribe(self, broker):\n if self.subscribed == False:\n for attr in self.parm_list:\n if attr.direction == attr.SUB:\n self.logging.debug(\"Subscribing: \"+attr.label)\n self.mqtt_client.subscribe(attr.topic)\n self.subscribed = True\n else:\n self.logging.debug(\"Already subscribed ... ignoring\")","def get_url(cls):\n return reverse('shop_subscribe:confirm')","def send_activation_notifications(modeladmin, request, queryset):\n for rec in queryset:\n if not rec.is_activated:\n send_activation_notification(rec)\n modeladmin.message_user(request, \"Письма с оповещением отправлены\")\n send_activation_notifications.short_description = 'Отправка писем с оповещением об активации'","def test_admin_approval_email(self):\n new_user = UserModel().objects.create_user(**self.user_info)\n profile = self.registration_profile.objects.create_profile(new_user)\n profile.activated = True\n self.registration_profile.objects.send_admin_approve_email(\n new_user, Site.objects.get_current())\n self.assertEqual(len(mail.outbox), 1)\n admins_emails = [value[1] for value in settings.REGISTRATION_ADMINS]\n for email in mail.outbox[0].to:\n self.assertIn(email, admins_emails)","def test_subscribe(mocker, api: API, account: Account, order, sku_id, activation):\n api.regnum.order.return_value = order\n api.activation.activate.return_value = activation\n assert account.subscribe(sku_id) == activation[\"id\"]\n api.regnum.order.assert_called_with(\n \"USERNAME\", sku_id, 1, date.today(), timedelta(days=365)\n )\n api.activation.activate.assert_called_with(\"USERNAME\", 5678, mocker.ANY, date.today())\n assert order in account.orders\n assert activation in account.activations","def process(self):\n received_message = SubscribeMessage(*self.message.value)\n allow, msg = customize.authorize_subscription(received_message.topic, self.connection)\n if allow:\n subscription_id = tornwamp_topic.topics.add_subscriber(\n received_message.topic,\n self.connection,\n )\n answer = SubscribedMessage(\n request_id=received_message.request_id,\n subscription_id=subscription_id\n )\n self.broadcast_messages = customize.get_subscribe_broadcast_messages(received_message, subscription_id, self.connection.id)\n else:\n answer = ErrorMessage(\n request_id=received_message.request_id,\n request_code=received_message.code,\n uri=\"tornwamp.subscribe.unauthorized\"\n )\n answer.error(msg)\n self.answer_message = answer","def web_channels_subscriptions():\n\n try:\n auth_check()\n except Exception as e:\n return flask.redirect(str(e))\n\n update = flask.request.args.get('update', None)\n\n if update is not None:\n update_data = json.loads(urllib.parse.unquote(update))\n\n if update_data['subscribe']:\n yt_create_subscription(update_data['id'])\n time.sleep(10)\n else:\n yt_remove_subscription(update_data['id'])\n time.sleep(10)\n\n return flask.redirect(update_data['redirect'])\n\n return flask.redirect('channels')","def confirm_email(self):\n # The base class' implementation does nothing\n pass","def confirmed(self):","def subscribe_topic(self):\n req = {\n \"op\": \"subscribe\",\n \"args\": [\n \"instrument\",\n \"trade\",\n \"orderBook10\",\n \"execution\",\n \"order\",\n \"position\",\n \"margin\",\n ],\n }\n self.send_packet(req)","def subscribe(self) -> None:\n events = [\n HathorEvents.NETWORK_NEW_TX_ACCEPTED,\n HathorEvents.NETWORK_PEER_CONNECTING,\n HathorEvents.NETWORK_PEER_READY,\n HathorEvents.NETWORK_PEER_CONNECTED,\n HathorEvents.NETWORK_PEER_DISCONNECTED,\n HathorEvents.NETWORK_PEER_CONNECTION_FAILED\n ]\n\n for event in events:\n self.pubsub.subscribe(event, self.handle_publish)","def newsletter():\n mongo.db.newsletter.insert_one(\n {\"email\": request.form.get(\"email\")})\n return jsonify(success=True)","def set_subscription(request, token, subscribe): # pylint: disable=unused-argument\r\n try:\r\n username = UsernameCipher().decrypt(token.encode())\r\n user = User.objects.get(username=username)\r\n except UnicodeDecodeError:\r\n raise Http404(\"base64url\")\r\n except UsernameDecryptionException as exn:\r\n raise Http404(exn.message)\r\n except User.DoesNotExist:\r\n raise Http404(\"username\")\r\n\r\n if subscribe:\r\n UserPreference.objects.get_or_create(user=user,\r\n key=NOTIFICATION_PREF_KEY,\r\n defaults={\r\n \"value\": UsernameCipher.encrypt(user.username)\r\n })\r\n return render_to_response(\"resubscribe.html\", {'token': token})\r\n else:\r\n UserPreference.objects.filter(user=user, key=NOTIFICATION_PREF_KEY).delete()\r\n return render_to_response(\"unsubscribe.html\", {'token': token})","def test_admin_approval_complete_email(self):\n new_user = UserModel().objects.create_user(**self.user_info)\n profile = self.registration_profile.objects.create_profile(new_user)\n profile.send_admin_approve_complete_email(Site.objects.get_current())\n self.assertEqual(len(mail.outbox), 1)\n self.assertEqual(mail.outbox[0].to, [self.user_info['email']])","def subscribe_mqtt(self, topic):\n if topic not in self.subscriptions:\n self.subscriptions.append(topic)\n self.mqtt.subscribe(topic)","def send_confirmation_email(self, registration_profile,\n text_template=None, html_template=None,\n subject=None, email_data=None, **kwargs):\n user_id = registration_profile.user.id\n key = registration_profile.activation_key\n self._send_email(\n confirmation_profile=registration_profile,\n url=reverse('users.activate', args=[user_id, key]),\n subject=subject or _('Please confirm your email address'),\n text_template=text_template or 'users/email/activate.ltxt',\n html_template=html_template or 'users/email/activate.html',\n send_to=registration_profile.user.email,\n expiration_days=settings.ACCOUNT_ACTIVATION_DAYS,\n username=registration_profile.user.username,\n email_data=email_data,\n **kwargs)","def subscribe(self):\n res = self._subscribe()\n if res is not None:\n self._subscribed = True\n return res","def save(self, **kwargs):\n if CustomerModel.objects.filter(user__email=self.cleaned_data['email']).exists():\n logger.info('Subscription from {} dropped, email address exists.'.format(self.cleaned_data['email']))\n return self.instance\n # email is not assigned by the form probably because it is a related user object field\n self.instance.email = self.cleaned_data['email']\n if send_confirmation_email(self.request, self.instance):\n self.instance = super(SubscribeForm, self).save(**kwargs)\n return self.instance","def test_email_signup(self, subscribe):\n params = {'display_name': 'newbie',\n 'email': 'newbie@example.com',\n 'password': 'asdf1234',\n 'agreement': 'on',\n 'email_subscribe': 'on'}\n response = self.client.post(reverse('users.register'), params)\n eq_(200, response.status_code)\n\n source_url = 'http://testserver%s' % response.request['PATH_INFO']\n subscribe.assert_called_with('testcamp', u'newbie@example.com',\n lang='en-us', source_url=source_url)","def _subscribe(topic:str) -> None:\n\t\t\tif topic in self.subscribedTopics:\n\t\t\t\tself.messageHandler and self.messageHandler.logging(self.mqttClient, logging.WARNING, f'MQTT: topic already subscribed: {topic}')\n\t\t\t\treturn\n\t\t\tif (r := self.mqttClient.subscribe(topic))[0] == 0:\n\t\t\t\tt = MQTTTopic(topic = topic, mid=r[1], callback=callback, callbackArgs=kwargs)\n\t\t\t\tself.subscribedTopics[topic] = t\n\t\t\telse:\n\t\t\t\tself.messageHandler and self.messageHandler.logging(self.mqttClient, logging.ERROR, f'MQTT: cannot subscribe: {r[0]}')","def confirm_email(self, request, email_address):\n email_address.verified = True\n email_address.set_as_primary(conditional=True)\n email_address.save()\n\n u = get_user_model().objects.get(pk=email_address.user.id)\n u.is_active = True\n u.save()","def account_confirm_email(request, profile_id, confirmation_code):\n user = request.user\n profile = user\n if profile.valid:\n return render(request, 'agda/account/confirm_email.html')\n\n # Check if the confirmation code works\n code_ok = profile.check_email_confirmation_code(confirmation_code)\n\n # Check that this email_pending is not the confirmed email of another User\n users = User.objects.filter(username=profile.email)\n if users and (len(users) > 1 or users[0] != profile):\n code_ok = False\n\n if code_ok and request.POST.get(\"confirmed\"):\n # Register the email as confirmed, and the account therefore as valid.\n user.username = profile.email\n user.save()\n profile.save()\n profile.log_change(profile, message=\"Agda: Email %s confirmed, user level now %s.\" % (user.username))\n profile.send_email(\"%s %s: Agda account created\" % (user.first_name, user.last_name),\n \"agda/account/validated_email.txt\",\n cc_recipients=[settings.AGDA_MAIL_WATCHERS])\n # Redirect with message\n messages.success(request, \"Your email address has now been confirmed.\")\n return redirect(top)\n\n logger.warn(\"Failed email confirmation for profile %s.\" % profile)\n return render(request, 'agda/account/confirm_email.html', dict(code_ok=code_ok, confirmed=False))","def start(self, update, context):\n # add or update to the sqlite table.\n chat = update.message.chat\n user_tuple = self.db_manager.create_user_tuple(chat.id)\n self.db_manager.add_new_user(user_tuple)\n self.logger.info(\n 'A new user with username: %s and chat_id: %s subscribed to the list.' % (chat.username, chat.id)\n )\n update.message.reply_text('Welcome! You have successfully subscribed to real time notification for a six '\n 'scored in IPL 2019 cricket match.')","def on_connect(client, userdata, flags, rc):\n print('Connected with result code ' + str(rc))\n client.subscribe(MQTT_TOPIC)","def on_connect(client, userdata, flags, rc):\n print('Connected with result code ' + str(rc))\n client.subscribe(MQTT_TOPIC)","def test_customer_notified(self, mocked_notify_client):\n order = OrderPaidFactory()\n\n notify.quote_accepted(order)\n\n assert mocked_notify_client.send_email_notification.called\n call_args = mocked_notify_client.send_email_notification.call_args_list[0][1]\n assert call_args['email_address'] == order.get_current_contact_email()\n assert call_args['template_id'] == Template.quote_accepted_for_customer.value\n assert call_args['personalisation']['recipient name'] == order.contact.name\n assert call_args['personalisation']['embedded link'] == order.get_public_facing_url()","def test_activation_email(self):\n new_user = UserModel().objects.create_user(**self.user_info)\n profile = self.registration_profile.objects.create_profile(new_user)\n profile.send_activation_email(Site.objects.get_current())\n self.assertEqual(len(mail.outbox), 1)\n self.assertEqual(mail.outbox[0].to, [self.user_info['email']])","def index(request):\n if request.method == \"POST\":\n subscribe_form = UserSubscribeForm(request.POST)\n\n if subscribe_form.is_valid():\n messages.success(request, 'Thank you for your email subscription')\n\n else:\n subscribe_form = UserSubscribeForm()\n return render(request, 'index.html', {'index': True, 'subscribe_form': subscribe_form})","def sendSubcribe(self):\n user_subcribe = {\"channel\": \"/meta/subscribe\",\n \"clientId\": self.client_id,\n \"subscription\": \"/user/{}\".format(self.user_id),\n \"id\": str(self.id),\n \"ext\": {\"access_token\": self.token,\n \"timestamp\": time.time()}\n }\n self.sendMessage(json.dumps(user_subcribe).encode('utf-8'))\n # Increment sending id, see GroupMe push api\n self.id += 1","def send_confirmation_email(self, email_change, new_email):\n self._send_email(\n confirmation_profile=email_change,\n url=reverse('users.confirm_email',\n args=[email_change.activation_key]),\n subject=_('Please confirm your email address'),\n text_template='users/email/confirm_email.ltxt',\n html_template='users/email/confirm_email.html',\n send_to=new_email)","def account_activation_sent(request):\n current_user = request.user\n if current_user.is_authenticated():\n return HttpResponseRedirect('/')\n return render(request, 'registration/activation_complete.html')","def is_confirmed(self, is_confirmed: bool):\n\n self._is_confirmed = is_confirmed","def maybe_notify_lessee(request, response):\n if request.get('pubsub_topic'):\n pubsub.publish(\n pubsub.full_topic_name(\n request['pubsub_project'], request['pubsub_topic']),\n json.dumps(response),\n {},\n )\n metrics.pubsub_messages_sent.increment(fields={'target': 'lessee'})","def on_connect(client, userdata, flags, rc):\n print('Connected with result code ' + str(rc))\n client.subscribe(MQTT_TOPIC, qos=1)","def test_resend_activation_email(self):\n\n data = {\n 'email': self.user.email,\n }\n\n response = self.client.post(\n reverse('user-resend-activation-email'),\n data,\n format='json',\n )\n\n self.assertEqual(\n response.status_code,\n status.HTTP_200_OK,\n response.content\n )\n\n self.assertEqual(\n response.content,\n b'',\n )\n\n self.assertEqual(len(mail.outbox), 1)","def subscribe(self):\n if not self._subscribed and self._connected:\n if ATTR_STREAM_ID not in self.data:\n msg = self._create_message(strings.SUB_MSG)\n self.write(msg)\n else:\n msg = self._create_message(strings.RESUB_MSG)\n self.write(msg)\n self._subscribed = True"],"string":"[\n \"def subscribe(self, request):\\n email = self.cleaned_data.get('email')\\n\\n email_name, domain_part = email.rsplit('@', 1)\\n domain_name = '@' + domain_part\\n email_domain, created = Domain.objects.get_or_create(name=domain_name)\\n\\n subscriber, created = Subscriber.objects.get_or_create(email=email, mailing_list=self.mailing_list, defaults={\\n 'domain': email_domain\\n })\\n subscriber.status = Status.PENDING\\n subscriber.optin_ip_address = get_client_ip(request)\\n subscriber.optin_date = timezone.now()\\n subscriber.save()\\n\\n if not created:\\n subscriber.tokens.filter(description='confirm_subscription').delete()\\n\\n token = subscriber.tokens.create(description='confirm_subscription')\\n current_site = get_current_site(request)\\n protocol = 'https' if request.is_secure() else 'http'\\n domain = current_site.domain\\n path = reverse('subscribers:confirm_double_optin_token', kwargs={\\n 'mailing_list_uuid': self.mailing_list.uuid,\\n 'token': token.text\\n })\\n confirm_link = '%s://%s%s' % (protocol, domain, path)\\n\\n confirm_email = self.mailing_list.get_confirm_email_template()\\n confirm_email.send(subscriber.get_email(), {\\n 'confirm_link': confirm_link\\n })\\n\\n return subscriber\",\n \"async def aoc_subscribe(self, ctx: commands.Context) -> None:\\n if ctx.channel.id != settings.aoc.channel_id:\\n await ctx.send(f\\\"Please use the <#{settings.aoc.channel_id}> channel\\\")\\n return\\n\\n role = ctx.guild.get_role(settings.aoc.role_id)\\n unsubscribe_command = f\\\"{ctx.prefix}{ctx.command.root_parent} unsubscribe\\\"\\n\\n if role not in ctx.author.roles:\\n await ctx.author.add_roles(role)\\n await ctx.send(\\n \\\"Okay! You have been __subscribed__ to notifications about new Advent of Code tasks. \\\"\\n f\\\"You can run `{unsubscribe_command}` to disable them again for you.\\\"\\n )\\n else:\\n await ctx.send(\\n \\\"Hey, you already are receiving notifications about new Advent of Code tasks. \\\"\\n f\\\"If you don't want them any more, run `{unsubscribe_command}` instead.\\\"\\n )\",\n \"async def aoc_subscribe(self, ctx: commands.Context) -> None:\\n role = ctx.guild.get_role(AocConfig.role_id)\\n unsubscribe_command = f\\\"{ctx.prefix}{ctx.command.root_parent} unsubscribe\\\"\\n\\n if role not in ctx.author.roles:\\n await ctx.author.add_roles(role)\\n await ctx.send(\\\"Okay! You have been __subscribed__ to notifications about new Advent of Code tasks. \\\"\\n f\\\"You can run `{unsubscribe_command}` to disable them again for you.\\\")\\n else:\\n await ctx.send(\\\"Hey, you already are receiving notifications about new Advent of Code tasks. \\\"\\n f\\\"If you don't want them any more, run `{unsubscribe_command}` instead.\\\")\",\n \"def _subscribe(self):\\n self.subscribed = True\\n self.subscribe_date = now()\\n self.unsubscribed = False\",\n \"def subscribe(self, update, context):\\n # add or update to the sqlite table.\\n chat = update.message.chat\\n user_tuple = self.db_manager.create_user_tuple(chat.id)\\n try:\\n status = self.db_manager.add_new_user(user_tuple)\\n except IntegrityError:\\n self.logger.info('Username: %s and chat_id: %s is already subscribed to the list.')\\n update.message.reply_text('You are already subscribed to real time notification for a six '\\n 'scored in IPL 2019 cricket match.')\\n return True\\n self.logger.info(\\n 'Username: %s and chat_id: %s subscribed to the list.' % (chat.username, chat.id)\\n )\\n update.message.reply_text('Welcome! You have successfully subscribed to real time notification for a six '\\n 'scored in IPL 2019 cricket match.')\",\n \"def view_mailinglist_subscribe(request, form_class, mailing_list_id=None):\\n subscribed = False\\n mailing_list = None\\n if mailing_list_id:\\n mailing_list = get_object_or_404(MailingList, id=mailing_list_id)\\n\\n if request.POST and not subscribed:\\n form = form_class(request.POST)\\n if form.is_valid():\\n form.save(mailing_list)\\n subscribed = True\\n else:\\n form = form_class()\\n\\n return render_to_response('newsletter/mailing_list_subscribe.html',\\n {'subscribed': subscribed,\\n 'mailing_list': mailing_list,\\n 'form': form},\\n context_instance=RequestContext(request))\",\n \"def _handle_subscribe(self, presence):\\n presence.reply()\\n presence['to'] = presence['to'].bare\\n\\n # We are using trinary logic, so conditions have to be\\n # more explicit than usual.\\n if self.auto_authorize == True:\\n presence['type'] = 'subscribed'\\n presence.send()\\n if self.auto_subscribe:\\n presence['type'] = 'subscribe'\\n presence.send()\\n elif self.auto_authorize == False:\\n presence['type'] = 'unsubscribed'\\n presence.send()\",\n \"def subscribe( self, mess, args):\\n user = self.get_sender_username(mess)\\n if user in self.users:\\n return 'You are already subscribed.'\\n else:\\n self.users[user] = user\\n self.invited.pop(user)\\n self.message_queue.append('_%s has joined the channel_' % user)\\n self.log.info('%s subscribed to the broadcast.' % user)\\n self.save_state()\\n return 'You are now subscribed.'\",\n \"def email_subscribe_pending_confirm(hexdomain):\\n domain = tools.parse_domain(hexdomain)\\n if domain is None:\\n flask.abort(400, 'Malformed domain or domain not represented in hexadecimal format.')\\n\\n email_address = flask.request.form['email_address']\\n\\n if email_address.strip() == '':\\n return flask.redirect('/email/subscribe/{}/0'.format(hexdomain))\\n\\n verify_code = tools.random_id()\\n verify_url = flask.request.url_root + 'email/verify/{}'.format(verify_code)\\n email_body = email_tools.render_email(\\n 'confirm.html',\\n domain=domain,\\n verify_url=verify_url\\n )\\n\\n repository.propose_subscription(verify_code, email_address, domain)\\n emailer.send(\\n email_address, 'Please verify your subscription', email_body\\n )\\n\\n return flask.render_template('www/email/pending_verify.html', domain=domain)\",\n \"def test_send_subscribe_email(self):\\n #Verifica se foi enviado 1 e-mail, o este não envia e-mail\\n self.assertEqual(1, len(mail.outbox))\",\n \"def on_subscribe(self, request, ssn):\\n if 'hub.challenge' not in request.GET:\\n logger.error(f'Missing hub.challenge in subscription verification {ssn.pk}!')\\n tasks.save.delay(\\n pk = ssn.pk,\\n subscribe_status = 'verifyerror',\\n verifyerror_count = ssn.verifyerror_count + 1\\n )\\n return Response('Missing hub.challenge', status=HTTP_400_BAD_REQUEST)\\n\\n if not request.GET.get('hub.lease_seconds', '').isdigit():\\n logger.error(f'Missing integer hub.lease_seconds in subscription verification {ssn.pk}!')\\n tasks.save.delay(\\n pk = ssn.pk,\\n subscribe_status = 'verifyerror',\\n verifyerror_count = ssn.verifyerror_count + 1\\n )\\n return Response('hub.lease_seconds required and must be integer', status=HTTP_400_BAD_REQUEST)\\n\\n if ssn.unsubscribe_status is not None:\\n logger.error(f'Subscription {ssn.pk} received subscription verification request,'\\n f' but its was explicitly unsubscribed before.')\\n return Response('Unsubscribed')\\n\\n tasks.save.delay(\\n pk = ssn.pk,\\n subscribe_status = 'verified',\\n lease_expiration_time = now() + timedelta(seconds=int(request.GET['hub.lease_seconds'])),\\n connerror_count = 0,\\n huberror_count = 0,\\n verifyerror_count = 0,\\n verifytimeout_count = 0\\n )\\n logger.info(f'Got {ssn.pk} subscribe confirmation from hub.')\\n return HttpResponse(request.GET['hub.challenge'])\",\n \"def _want_subscription() -> bool:\\n prompt = (\\n 'Would you be willing, once your first certificate is successfully issued, '\\n 'to share your email address with the Electronic Frontier Foundation, a '\\n \\\"founding partner of the Let's Encrypt project and the non-profit organization \\\"\\n \\\"that develops Certbot? We'd like to send you email about our work encrypting \\\"\\n \\\"the web, EFF news, campaigns, and ways to support digital freedom. \\\")\\n return display_util.yesno(prompt, default=False)\",\n \"def subscribe():\\n form = SubscribeForm()\\n if form.validate_on_submit():\\n subscription = Subscription(email=form.email.data)\\n db.session.add(subscription)\\n db.session.commit()\\n return redirect(url_for('main.index'))\\n\\n return render_template('subscribes.html', form=form)\",\n \"def cbMqtt_on_subscribe(client, userdata, mid, granted_qos):\\n # logger.debug('Subscribed to MQTT topic with message id %d', mid)\\n pass\",\n \"def cb_start(self, update, context):\\n chat_id = update.message.chat_id\\n\\n if chat_id in self.clientChatIds:\\n update.message.reply_text('You are already subscribed.')\\n return\\n\\n print(f\\\"Subscribing chat_id '{chat_id}'\\\")\\n self.clientChatIds.add(chat_id)\\n update.message.reply_text(\\\"You have successfully subscribed.\\\")\\n self.saveToFile(self.configFile)\",\n \"def send_confirmation(self):\\r\\n c.user.email_validated = False\\r\\n c.user.confirmation_code = random_key(6)\\r\\n c.user._commit()\\r\\n emailer.confirmation_email(c.user)\",\n \"async def enable_publish_confirms(self) -> None:\\n if self.channel:\\n if not self.channel.publisher_confirms:\\n await self.channel.confirm_select()\",\n \"def mqtt_on_connect(client, userdata, flags, rc):\\n logging.debug('successfully connected to mqtt broker')\\n client.subscribe(config['mqtt']['subscribe_topic'])\",\n \"def form_valid(self, form):\\n form.save()\\n return super(SubscribeView, self).form_valid(form)\",\n \"def confirm_email(self):\\n self.active = True\\n self.save()\",\n \"def subscription(bot, update):\\n chat_id = update.message.chat_id\\n bot.sendMessage(chat_id=chat_id, text=SUBSCRIPTION_MSG, parse_mode='markdown', \\n disable_web_page_preview=True)\\n \\n mp.track(get_user_info(chat_id)['PID'], 'Checked Subscription')\",\n \"def subscribe(self, feed, **args):\\n args.update(feed=feed)\\n return self.fetch(\\\"/subscribe\\\", post_args=args)\",\n \"def send_confirm_email(request,uid):\\n user=models.UserProfile.objects.get(id=uid)\\n current_site=get_current_site(request)\\n email_subject='Activate Your Account'\\n message=render_to_string('activate_account.html',{\\n 'user':user,\\n 'domain':current_site.domain,\\n 'uid':urlsafe_base64_encode(force_bytes(uid)),\\n 'token':account_activation_token.make_token(user),\\n })\\n to_email= user.email\\n email= EmailMessage(email_subject,message,to=[to_email])\\n email.send()\\n return JsonResponse(\\n {\\n \\\"status\\\":\\\"The confirmation email has been sent.\\\",\\n }\\n )\",\n \"def subscribe( self, topic ):\\n logging.info( \\\"Subscribing to topic %s\\\" %topic )\\n try:\\n self.client.subscribe( topic )\\n except Exception as error:\\n print( error )\",\n \"def subscribe(self, subject):\\n pass\",\n \"def test_successful_subscriptions_notifies_pm(self) -> None:\\n invitee = self.example_user(\\\"iago\\\")\\n\\n current_stream = self.get_streams(invitee)[0]\\n invite_streams = self.make_random_stream_names([current_stream])[:1]\\n self.common_subscribe_to_streams(\\n invitee,\\n invite_streams,\\n extra_post_data={\\n \\\"announce\\\": \\\"true\\\",\\n \\\"principals\\\": orjson.dumps([self.user_profile.id]).decode(),\\n },\\n )\",\n \"def confirmed_email_notification(sender, **kwargs):\\n subject_template_name = 'shop_simplenotifications/confirmed_subject.txt'\\n body_text_template_name = 'shop_simplenotifications/confirmed_body.txt'\\n body_html_template_name = 'shop_simplenotifications/confirmed_body.html'\\n request = kwargs.get('request')\\n order = kwargs.get('order')\\n subject = loader.render_to_string(\\n subject_template_name,\\n RequestContext(request, {'order': order})\\n )\\n subject = subject.join(subject.splitlines())\\n\\n text_content = loader.render_to_string(\\n body_text_template_name,\\n RequestContext(request, {'order': order})\\n )\\n try:\\n html_content = loader.render_to_string(\\n body_html_template_name,\\n RequestContext(request, {'order': order})\\n )\\n except TemplateDoesNotExist:\\n html_content = None\\n\\n from_email = getattr(settings, 'SN_FROM_EMAIL',\\n settings.DEFAULT_FROM_EMAIL)\\n owners = getattr(settings, 'SN_OWNERS', settings.ADMINS)\\n\\n message = EmailMultiAlternatives(subject, text_content, from_email,\\n [owner[1] for owner in owners])\\n if html_content:\\n message.attach_alternative(html_content, \\\"text/html\\\")\\n message.send()\",\n \"def on_subscribe(\\n client: mqtt.Client,\\n userdata: Any,\\n mid: int,\\n granted_qos: int,\\n properties: Properties = None,\\n ) -> None:\\n logging.info(\\n f\\\"Successfully subscribed to topic: mid={mid}, granted qos={granted_qos}, properties={properties}\\\"\\n )\",\n \"def activate(self):\\n if not self.is_active:\\n self.is_active = True\\n self.activated_at = datetime.datetime.utcnow()\\n import messaging # avoid circular import\\n messaging.send_activated_emails(self)\\n self.save()\",\n \"def subscribe(self, feedUrl):\\r\\n response = self.httpPost(\\r\\n ReaderUrl.SUBSCRIPTION_EDIT_URL,\\r\\n {'ac':'subscribe', 's': feedUrl})\\r\\n # FIXME - need better return API\\r\\n if response and 'OK' in response:\\r\\n return True\\r\\n else:\\r\\n return False\",\n \"def subscribe(self, user_token, topic):\\n response = _request('POST',\\n url=self.url_v1('/user/subscriptions/' + topic),\\n user_agent=self.user_agent,\\n user_token=user_token,\\n )\\n _raise_for_status(response)\",\n \"async def send_confirm_link(email: TextData, background_tasks: BackgroundTasks):\\n email = email.data\\n mail, subject, body = await AccountProcessor.send_confirmation_link(email)\\n background_tasks.add_task(Utility.validate_and_send_mail, email=mail, subject=subject, body=body)\\n return {\\\"message\\\": \\\"Success! Confirmation link sent\\\"}\",\n \"def notify_thesis_accepted(thesis: 'Thesis'):\\n recipients = []\\n if thesis.advisor:\\n recipients.append(thesis.advisor.user.email)\\n if thesis.supporting_advisor:\\n recipients.append(thesis.supporting_advisor.user.email)\\n if recipients:\\n formatted_body = ACCEPTED_BODY.format(thesis.title)\\n send_mail(ACCEPTED_SUBJECT, formatted_body, settings.MASS_MAIL_FROM, recipients)\",\n \"def confirm_subscription(self, sub_code):\\n\\t\\tresult = {}\\n\\t\\tconnection = DbHelper.connect()\\n\\t\\tsub_id = 0\\n\\t\\t# print(\\\"Subscription Code: \\\" + sub_code)\\n\\n\\t\\ttry:\\n\\t\\t\\twith connection.cursor() as cursor:\\n\\t\\t\\t\\tsql = \\\"SELECT * FROM mail_list \\\\\\n\\t\\t\\t\\t\\t WHERE email_hash=%s;\\\"\\n\\t\\t\\t\\tcursor.execute(sql, [sub_code])\\n\\t\\t\\t\\tresult = cursor.fetchone()\\n\\t\\t\\t\\t\\n\\t\\t\\t\\tif not result:\\n\\t\\t\\t\\t\\tconnection.close()\\n\\t\\t\\t\\t\\treturn \\\"CODE_DOES_NOT_EXIST\\\"\\n\\t\\t\\t\\t\\n\\t\\t\\t\\telif result['is_activated']:\\n\\t\\t\\t\\t\\tconnection.close()\\n\\t\\t\\t\\t\\treturn \\\"CODE_ALREADY_ACTIVATED\\\"\\n\\n\\t\\t\\t\\tsub_id = result['sub_id']\\n\\n\\t\\t\\t\\tsql = \\\"UPDATE mail_list \\\\\\n\\t\\t\\t\\t\\t SET is_activated=is_activated+1 \\\\\\n\\t\\t\\t\\t\\t WHERE sub_id=%s;\\\"\\n\\t\\t\\t\\tcursor.execute(sql, [sub_id])\\n\\t\\t\\t\\tconnection.commit()\\n\\t\\t\\t\\tconnection.close()\\n\\t\\t\\t\\treturn result\\n\\t\\texcept pymysql.MySQLError as e:\\n\\t\\t\\tconnection.close()\\n\\t\\t\\treturn \\\"DATABASE_ERROR\\\"\",\n \"def test_subscribe(self):\\n self.service.clientConnected()\\n self.service.subscribe(u'url', None)\\n pubsubClient = self.service.pubsubClient\\n self.assertIn(u'url', pubsubClient.subscriptions)\",\n \"def _subscribe(self):\\n if self.subscribed:\\n return False\\n return {}\",\n \"def on_subscribe( client, userdata, mid, granted_qos ):\\n logging.info( \\\"Topic successfully subcribed with QoS: %s\\\" %granted_qos )\",\n \"def handle_email_confirmed(sender, **kwargs):\\n email = kwargs['email_address']\\n email.user.userprofile.member.cast().confirm_email()\",\n \"def subscribe(self, topic):\\n\\t\\tself.topic=topic\\n\\t\\tself.client.subscribe(self.topic)\",\n \"def _subscribe(self, callback_path: str, topic_url: str, mode: str = \\\"subscribe\\\"):\\n data = {'hub.callback': self.callback_url + callback_path,\\n 'hub.mode': mode,\\n 'hub.topic': topic_url,\\n 'hub.lease_seconds': self.subscribe_least_seconds}\\n if self.secret is not None:\\n data['hub.secret'] = self.secret\\n result = self.__api_post_request(TWITCH_API_BASE_URL + \\\"webhooks/hub\\\", data=data)\\n if result.status_code != 202:\\n logging.error(f'Subscription failed! status code: {result.status_code}, body: {result.text}')\\n return result.status_code == 202\",\n \"def on_subscribe(self, mqtt_client, userdata, mid, granted_qos):\\n logging.debug(\\\"DEBUG - subscribe ack received\\\")\",\n \"def send_confirmation_email(user_pk):\\n pass\",\n \"def _on_connect(self, client, userdata, flags, rc):\\n self.subscribe(self.topic)\",\n \"def confirm_subscription(self, topic, token,\\r\\n authenticate_on_unsubscribe=False):\\r\\n params = {'ContentType' : 'JSON',\\r\\n 'TopicArn' : topic,\\r\\n 'Token' : token}\\r\\n if authenticate_on_unsubscribe:\\r\\n params['AuthenticateOnUnsubscribe'] = 'true'\\r\\n response = self.make_request('ConfirmSubscription', params, '/', 'GET')\\r\\n body = response.read()\\r\\n if response.status == 200:\\r\\n return json.loads(body)\\r\\n else:\\r\\n boto.log.error('%s %s' % (response.status, response.reason))\\r\\n boto.log.error('%s' % body)\\r\\n raise self.ResponseError(response.status, response.reason, body)\",\n \"def onSubscribed(self, connection:MQTTConnection, topic:str) -> bool:\\n\\t\\tconnection.subscribedCount += 1\\n\\t\\treturn True\",\n \"def render_email(self, user):\\n if user.get_profile().subscribed_to_news:\\n return EmailMessage(self.subject, self.body, from_email='anon@example.com', to=[user.email])\",\n \"def _subscribed(self, account_id):\\n sql = \\\"\\\"\\\"SELECT 1 FROM hive_subscriptions\\n WHERE community_id = :community_id\\n AND account_id = :account_id\\\"\\\"\\\"\\n return bool(DB.query_one(\\n sql, community_id=self.community_id, account_id=account_id))\",\n \"def confirm_email(request, key):\\n alt_email = cpm.Email.objects.filter(activation_key=key)\\n if alt_email.exists():\\n alt_email[0].confirm()\\n return redirect('/')\\n hero_title = 'We weren\\\\'t able to complete your request...'\\n return render_err_msg(request, hero_title)\",\n \"def subscribe_email_to_cloudwatch(self, email):\\n return self.conn.subscribe(self.cloudwatch_arn, 'email', email)\",\n \"def speaker_confirmation():\\n\\n talks = Talk.objects.all()\\n\\n template = 'notifications/speaker_confirmation_mail.html'\\n\\n for talk in talks:\\n subject = 'SciPy.in 2011: Requesting for confirmation of your talk'\\n message = loader.render_to_string(\\n template, dictionary={'name': talk.speaker.username,\\n 'title': talk.title})\\n\\n talk.speaker.email_user(subject=subject, message=message,\\n from_email='admin@scipy.in')\",\n \"def subscribe(request):\\n address = request.POST.get('address')\\n\\n new_sub = Subscription(**{\\n \\\"address\\\": address\\n })\\n new_sub.save()\\n\\n return HttpResponse(json.dumps({\\n \\\"status\\\": \\\"success\\\"\\n }, default=helpers.json_custom_parser), content_type='application/json')\",\n \"def send_confirmation_email(self, *args, **kwargs):\\n raise NotImplementedError\",\n \"def subscribe_user(self, user):\\n self.ensure_one()\\n if self.has_user(user):\\n # already subscribed\\n return False\\n return self.emulate_request(user)\",\n \"def on_connect(client, userdata, flags, rc):\\n if rc == 0:\\n client.subscribe(topic_subscribe)\\n print(\\\"connected OK with returned code=\\\", rc)\\n else:\\n print(\\\"Bad connection with returned code=\\\", rc)\",\n \"def test_manually_confirm(self):\\n data = {}\\n response = self.client.post(self.url, data)\\n self.assertRedirects(response, reverse('reminders_dashboard'))\\n\\n reminder = reminders.SentNotification.objects.get(pk=self.unconfirmed.pk)\\n self.assertEqual(reminder.status, 'manual')\\n self.assertEqual(reminder.date_confirmed.date(), datetime.date.today())\",\n \"def test_manually_confirm(self):\\n data = {}\\n response = self.client.post(self.url, data)\\n self.assertRedirects(response, reverse('reminders_dashboard'))\\n\\n reminder = reminders.SentNotification.objects.get(pk=self.unconfirmed.pk)\\n self.assertEqual(reminder.status, 'manual')\\n self.assertEqual(reminder.date_confirmed.date(), datetime.date.today())\",\n \"def send_emails_to_subscribers(creator_id, exploration_id, exploration_title):\\n\\n creator_name = user_services.get_username(creator_id)\\n email_subject = ('%s has published a new exploration!' % creator_name)\\n email_body_template = (\\n 'Hi %s,<br>'\\n '<br>'\\n '%s has published a new exploration! You can play it here: '\\n '<a href=\\\"https://www.oppia.org/explore/%s\\\">%s</a><br>'\\n '<br>'\\n 'Thanks, and happy learning!<br>'\\n '<br>'\\n 'Best wishes,<br>'\\n '- The Oppia Team<br>'\\n '<br>%s')\\n\\n if not feconf.CAN_SEND_EMAILS:\\n log_new_error('This app cannot send emails to users.')\\n return\\n\\n if not feconf.CAN_SEND_SUBSCRIPTION_EMAILS:\\n log_new_error('This app cannot send subscription emails to users.')\\n return\\n\\n recipient_list = subscription_services.get_all_subscribers_of_creator(\\n creator_id)\\n recipients_usernames = user_services.get_usernames(recipient_list)\\n recipients_preferences = user_services.get_users_email_preferences(\\n recipient_list)\\n for index, username in enumerate(recipients_usernames):\\n if recipients_preferences[index].can_receive_subscription_email:\\n email_body = email_body_template % (\\n username, creator_name, exploration_id,\\n exploration_title, EMAIL_FOOTER.value)\\n _send_email(\\n recipient_list[index], feconf.SYSTEM_COMMITTER_ID,\\n feconf.EMAIL_INTENT_SUBSCRIPTION_NOTIFICATION,\\n email_subject, email_body, feconf.NOREPLY_EMAIL_ADDRESS)\",\n \"def subscribe(self, path):\\n box = self._getMailbox(path)\\n\\n box.metadata['subscribed'] = True\\n\\n box.saveMetadata()\\n\\n return True\",\n \"def post(self):\\n data = request.get_json()\\n user = actions.get_user_by_email(data['email'])\\n html = '<p>Confirming your account will give you </p> <b>full access to Kwikker</b>'\\n subject = 'Confirm your Kwikker account, ' + user['username']\\n actions.send_email(data['email'], user['username'], user['password'], subject,\\n '/confirm/', html, True)\\n return \\\"\\\", 200\\n pass\",\n \"def _confirm_email(user, email):\\n mail_subject = 'Подтверждение почты'\\n message = render_to_string('accounts/account_verification_email.html', {\\n 'user': user,\\n 'uid': urlsafe_base64_encode(force_bytes(user.pk)),\\n 'token': default_token_generator.make_token(user),\\n 'email': email,\\n })\\n to_email = email\\n send_email = EmailMessage(mail_subject, message, to=[to_email])\\n send_email.send()\",\n \"def subscribe(client, mqtt_topic):\\n\\n print()\\n print(\\\"Subscribe\\\")\\n print(\\\"================================================\\\")\\n print()\\n\\n # Subscribe to the config topic.\\n print(\\\"Subscribing\\\")\\n print(mqtt_topic)\\n print()\\n client.subscribe(mqtt_topic, qos=1)\",\n \"def view_subscriber_verification(request, form_class):\\n context = {}\\n\\n if request.POST:\\n context['form'] = form_class(request.POST)\\n subscription = SubscriberVerification()\\n if context['form'].is_valid():\\n contact = context['form'].save()\\n\\n subscription.contact = context['form'].instance\\n subscription.save()\\n link_id = subscription.link_id\\n\\n # select smtpserver, maybe, a boolean field for 'standard' will be\\n # a nice idea.\\n server = SMTPServer.objects.get(id=1)\\n\\n # set server settings to django\\n settings.EMAIL_HOST = smart_str(server.host)\\n settings.EMAIL_HOST_PASSWORD = smart_str(server.password)\\n settings.EMAIL_HOST_USER = smart_str(server.user)\\n settings.EMAIL_PORT = int(server.port)\\n settings.EMAIL_USE_TLS = server.tls\\n\\n # mail settings\\n subject = _('Subriber verification')\\n message = _('Thanks for subscription, please click the following link to verificate your email address.')\\n from_mail = DEFAULT_HEADER_REPLY\\n to_mail = context['form'].instance.email\\n\\n link = 'http://{0}/newsletters/subscribe/{1}'.format(str(request.get_host()), str(link_id))\\n text_content = u'{0!s}\\\\n\\\\n {1}'.format(message, link)\\n html_content = u'<body><p>{0!s}</p><p><a href=\\\"{1}\\\">{2}</a></p></body>'.format(message, link, _('verify link'))\\n msg = EmailMultiAlternatives(\\n subject,\\n text_content,\\n from_mail,\\n [to_mail]\\n )\\n msg.attach_alternative(html_content, 'text/html')\\n msg.send()\\n\\n context['send'] = True\\n\\n else:\\n context['form'] = form_class()\\n\\n return render_to_response('newsletter/subscriber_verification.html',\\n context,\\n context_instance=RequestContext(request))\",\n \"def on_connect(client, userdata, flags, rc):\\n print(f\\\"Re/Suscribing to TOPIC: {TOPIC}\\\")\\n client.subscribe(TOPIC)\\n if rc == 0:\\n print(f'Connected OK Returned code={rc}')\\n else:\\n print('Bad connection Returned code={rc}')\",\n \"def save(self, *args, **kwargs):\\n if not self.require_confirm_email:\\n User.objects.filter(is_active=False, deactivation_reason=\\\"pending\\\").update(\\n is_active=True, deactivation_reason=None\\n )\\n if not self.invite_question_text:\\n self.invite_question_text = \\\"What is your favourite book?\\\"\\n super().save(*args, **kwargs)\",\n \"def subscribe(self, broker):\\n if self.subscribed == False:\\n for attr in self.parm_list:\\n if attr.direction == attr.SUB:\\n self.logging.debug(\\\"Subscribing: \\\"+attr.label)\\n self.mqtt_client.subscribe(attr.topic)\\n self.subscribed = True\\n else:\\n self.logging.debug(\\\"Already subscribed ... ignoring\\\")\",\n \"def get_url(cls):\\n return reverse('shop_subscribe:confirm')\",\n \"def send_activation_notifications(modeladmin, request, queryset):\\n for rec in queryset:\\n if not rec.is_activated:\\n send_activation_notification(rec)\\n modeladmin.message_user(request, \\\"Письма с оповещением отправлены\\\")\\n send_activation_notifications.short_description = 'Отправка писем с оповещением об активации'\",\n \"def test_admin_approval_email(self):\\n new_user = UserModel().objects.create_user(**self.user_info)\\n profile = self.registration_profile.objects.create_profile(new_user)\\n profile.activated = True\\n self.registration_profile.objects.send_admin_approve_email(\\n new_user, Site.objects.get_current())\\n self.assertEqual(len(mail.outbox), 1)\\n admins_emails = [value[1] for value in settings.REGISTRATION_ADMINS]\\n for email in mail.outbox[0].to:\\n self.assertIn(email, admins_emails)\",\n \"def test_subscribe(mocker, api: API, account: Account, order, sku_id, activation):\\n api.regnum.order.return_value = order\\n api.activation.activate.return_value = activation\\n assert account.subscribe(sku_id) == activation[\\\"id\\\"]\\n api.regnum.order.assert_called_with(\\n \\\"USERNAME\\\", sku_id, 1, date.today(), timedelta(days=365)\\n )\\n api.activation.activate.assert_called_with(\\\"USERNAME\\\", 5678, mocker.ANY, date.today())\\n assert order in account.orders\\n assert activation in account.activations\",\n \"def process(self):\\n received_message = SubscribeMessage(*self.message.value)\\n allow, msg = customize.authorize_subscription(received_message.topic, self.connection)\\n if allow:\\n subscription_id = tornwamp_topic.topics.add_subscriber(\\n received_message.topic,\\n self.connection,\\n )\\n answer = SubscribedMessage(\\n request_id=received_message.request_id,\\n subscription_id=subscription_id\\n )\\n self.broadcast_messages = customize.get_subscribe_broadcast_messages(received_message, subscription_id, self.connection.id)\\n else:\\n answer = ErrorMessage(\\n request_id=received_message.request_id,\\n request_code=received_message.code,\\n uri=\\\"tornwamp.subscribe.unauthorized\\\"\\n )\\n answer.error(msg)\\n self.answer_message = answer\",\n \"def web_channels_subscriptions():\\n\\n try:\\n auth_check()\\n except Exception as e:\\n return flask.redirect(str(e))\\n\\n update = flask.request.args.get('update', None)\\n\\n if update is not None:\\n update_data = json.loads(urllib.parse.unquote(update))\\n\\n if update_data['subscribe']:\\n yt_create_subscription(update_data['id'])\\n time.sleep(10)\\n else:\\n yt_remove_subscription(update_data['id'])\\n time.sleep(10)\\n\\n return flask.redirect(update_data['redirect'])\\n\\n return flask.redirect('channels')\",\n \"def confirm_email(self):\\n # The base class' implementation does nothing\\n pass\",\n \"def confirmed(self):\",\n \"def subscribe_topic(self):\\n req = {\\n \\\"op\\\": \\\"subscribe\\\",\\n \\\"args\\\": [\\n \\\"instrument\\\",\\n \\\"trade\\\",\\n \\\"orderBook10\\\",\\n \\\"execution\\\",\\n \\\"order\\\",\\n \\\"position\\\",\\n \\\"margin\\\",\\n ],\\n }\\n self.send_packet(req)\",\n \"def subscribe(self) -> None:\\n events = [\\n HathorEvents.NETWORK_NEW_TX_ACCEPTED,\\n HathorEvents.NETWORK_PEER_CONNECTING,\\n HathorEvents.NETWORK_PEER_READY,\\n HathorEvents.NETWORK_PEER_CONNECTED,\\n HathorEvents.NETWORK_PEER_DISCONNECTED,\\n HathorEvents.NETWORK_PEER_CONNECTION_FAILED\\n ]\\n\\n for event in events:\\n self.pubsub.subscribe(event, self.handle_publish)\",\n \"def newsletter():\\n mongo.db.newsletter.insert_one(\\n {\\\"email\\\": request.form.get(\\\"email\\\")})\\n return jsonify(success=True)\",\n \"def set_subscription(request, token, subscribe): # pylint: disable=unused-argument\\r\\n try:\\r\\n username = UsernameCipher().decrypt(token.encode())\\r\\n user = User.objects.get(username=username)\\r\\n except UnicodeDecodeError:\\r\\n raise Http404(\\\"base64url\\\")\\r\\n except UsernameDecryptionException as exn:\\r\\n raise Http404(exn.message)\\r\\n except User.DoesNotExist:\\r\\n raise Http404(\\\"username\\\")\\r\\n\\r\\n if subscribe:\\r\\n UserPreference.objects.get_or_create(user=user,\\r\\n key=NOTIFICATION_PREF_KEY,\\r\\n defaults={\\r\\n \\\"value\\\": UsernameCipher.encrypt(user.username)\\r\\n })\\r\\n return render_to_response(\\\"resubscribe.html\\\", {'token': token})\\r\\n else:\\r\\n UserPreference.objects.filter(user=user, key=NOTIFICATION_PREF_KEY).delete()\\r\\n return render_to_response(\\\"unsubscribe.html\\\", {'token': token})\",\n \"def test_admin_approval_complete_email(self):\\n new_user = UserModel().objects.create_user(**self.user_info)\\n profile = self.registration_profile.objects.create_profile(new_user)\\n profile.send_admin_approve_complete_email(Site.objects.get_current())\\n self.assertEqual(len(mail.outbox), 1)\\n self.assertEqual(mail.outbox[0].to, [self.user_info['email']])\",\n \"def subscribe_mqtt(self, topic):\\n if topic not in self.subscriptions:\\n self.subscriptions.append(topic)\\n self.mqtt.subscribe(topic)\",\n \"def send_confirmation_email(self, registration_profile,\\n text_template=None, html_template=None,\\n subject=None, email_data=None, **kwargs):\\n user_id = registration_profile.user.id\\n key = registration_profile.activation_key\\n self._send_email(\\n confirmation_profile=registration_profile,\\n url=reverse('users.activate', args=[user_id, key]),\\n subject=subject or _('Please confirm your email address'),\\n text_template=text_template or 'users/email/activate.ltxt',\\n html_template=html_template or 'users/email/activate.html',\\n send_to=registration_profile.user.email,\\n expiration_days=settings.ACCOUNT_ACTIVATION_DAYS,\\n username=registration_profile.user.username,\\n email_data=email_data,\\n **kwargs)\",\n \"def subscribe(self):\\n res = self._subscribe()\\n if res is not None:\\n self._subscribed = True\\n return res\",\n \"def save(self, **kwargs):\\n if CustomerModel.objects.filter(user__email=self.cleaned_data['email']).exists():\\n logger.info('Subscription from {} dropped, email address exists.'.format(self.cleaned_data['email']))\\n return self.instance\\n # email is not assigned by the form probably because it is a related user object field\\n self.instance.email = self.cleaned_data['email']\\n if send_confirmation_email(self.request, self.instance):\\n self.instance = super(SubscribeForm, self).save(**kwargs)\\n return self.instance\",\n \"def test_email_signup(self, subscribe):\\n params = {'display_name': 'newbie',\\n 'email': 'newbie@example.com',\\n 'password': 'asdf1234',\\n 'agreement': 'on',\\n 'email_subscribe': 'on'}\\n response = self.client.post(reverse('users.register'), params)\\n eq_(200, response.status_code)\\n\\n source_url = 'http://testserver%s' % response.request['PATH_INFO']\\n subscribe.assert_called_with('testcamp', u'newbie@example.com',\\n lang='en-us', source_url=source_url)\",\n \"def _subscribe(topic:str) -> None:\\n\\t\\t\\tif topic in self.subscribedTopics:\\n\\t\\t\\t\\tself.messageHandler and self.messageHandler.logging(self.mqttClient, logging.WARNING, f'MQTT: topic already subscribed: {topic}')\\n\\t\\t\\t\\treturn\\n\\t\\t\\tif (r := self.mqttClient.subscribe(topic))[0] == 0:\\n\\t\\t\\t\\tt = MQTTTopic(topic = topic, mid=r[1], callback=callback, callbackArgs=kwargs)\\n\\t\\t\\t\\tself.subscribedTopics[topic] = t\\n\\t\\t\\telse:\\n\\t\\t\\t\\tself.messageHandler and self.messageHandler.logging(self.mqttClient, logging.ERROR, f'MQTT: cannot subscribe: {r[0]}')\",\n \"def confirm_email(self, request, email_address):\\n email_address.verified = True\\n email_address.set_as_primary(conditional=True)\\n email_address.save()\\n\\n u = get_user_model().objects.get(pk=email_address.user.id)\\n u.is_active = True\\n u.save()\",\n \"def account_confirm_email(request, profile_id, confirmation_code):\\n user = request.user\\n profile = user\\n if profile.valid:\\n return render(request, 'agda/account/confirm_email.html')\\n\\n # Check if the confirmation code works\\n code_ok = profile.check_email_confirmation_code(confirmation_code)\\n\\n # Check that this email_pending is not the confirmed email of another User\\n users = User.objects.filter(username=profile.email)\\n if users and (len(users) > 1 or users[0] != profile):\\n code_ok = False\\n\\n if code_ok and request.POST.get(\\\"confirmed\\\"):\\n # Register the email as confirmed, and the account therefore as valid.\\n user.username = profile.email\\n user.save()\\n profile.save()\\n profile.log_change(profile, message=\\\"Agda: Email %s confirmed, user level now %s.\\\" % (user.username))\\n profile.send_email(\\\"%s %s: Agda account created\\\" % (user.first_name, user.last_name),\\n \\\"agda/account/validated_email.txt\\\",\\n cc_recipients=[settings.AGDA_MAIL_WATCHERS])\\n # Redirect with message\\n messages.success(request, \\\"Your email address has now been confirmed.\\\")\\n return redirect(top)\\n\\n logger.warn(\\\"Failed email confirmation for profile %s.\\\" % profile)\\n return render(request, 'agda/account/confirm_email.html', dict(code_ok=code_ok, confirmed=False))\",\n \"def start(self, update, context):\\n # add or update to the sqlite table.\\n chat = update.message.chat\\n user_tuple = self.db_manager.create_user_tuple(chat.id)\\n self.db_manager.add_new_user(user_tuple)\\n self.logger.info(\\n 'A new user with username: %s and chat_id: %s subscribed to the list.' % (chat.username, chat.id)\\n )\\n update.message.reply_text('Welcome! You have successfully subscribed to real time notification for a six '\\n 'scored in IPL 2019 cricket match.')\",\n \"def on_connect(client, userdata, flags, rc):\\n print('Connected with result code ' + str(rc))\\n client.subscribe(MQTT_TOPIC)\",\n \"def on_connect(client, userdata, flags, rc):\\n print('Connected with result code ' + str(rc))\\n client.subscribe(MQTT_TOPIC)\",\n \"def test_customer_notified(self, mocked_notify_client):\\n order = OrderPaidFactory()\\n\\n notify.quote_accepted(order)\\n\\n assert mocked_notify_client.send_email_notification.called\\n call_args = mocked_notify_client.send_email_notification.call_args_list[0][1]\\n assert call_args['email_address'] == order.get_current_contact_email()\\n assert call_args['template_id'] == Template.quote_accepted_for_customer.value\\n assert call_args['personalisation']['recipient name'] == order.contact.name\\n assert call_args['personalisation']['embedded link'] == order.get_public_facing_url()\",\n \"def test_activation_email(self):\\n new_user = UserModel().objects.create_user(**self.user_info)\\n profile = self.registration_profile.objects.create_profile(new_user)\\n profile.send_activation_email(Site.objects.get_current())\\n self.assertEqual(len(mail.outbox), 1)\\n self.assertEqual(mail.outbox[0].to, [self.user_info['email']])\",\n \"def index(request):\\n if request.method == \\\"POST\\\":\\n subscribe_form = UserSubscribeForm(request.POST)\\n\\n if subscribe_form.is_valid():\\n messages.success(request, 'Thank you for your email subscription')\\n\\n else:\\n subscribe_form = UserSubscribeForm()\\n return render(request, 'index.html', {'index': True, 'subscribe_form': subscribe_form})\",\n \"def sendSubcribe(self):\\n user_subcribe = {\\\"channel\\\": \\\"/meta/subscribe\\\",\\n \\\"clientId\\\": self.client_id,\\n \\\"subscription\\\": \\\"/user/{}\\\".format(self.user_id),\\n \\\"id\\\": str(self.id),\\n \\\"ext\\\": {\\\"access_token\\\": self.token,\\n \\\"timestamp\\\": time.time()}\\n }\\n self.sendMessage(json.dumps(user_subcribe).encode('utf-8'))\\n # Increment sending id, see GroupMe push api\\n self.id += 1\",\n \"def send_confirmation_email(self, email_change, new_email):\\n self._send_email(\\n confirmation_profile=email_change,\\n url=reverse('users.confirm_email',\\n args=[email_change.activation_key]),\\n subject=_('Please confirm your email address'),\\n text_template='users/email/confirm_email.ltxt',\\n html_template='users/email/confirm_email.html',\\n send_to=new_email)\",\n \"def account_activation_sent(request):\\n current_user = request.user\\n if current_user.is_authenticated():\\n return HttpResponseRedirect('/')\\n return render(request, 'registration/activation_complete.html')\",\n \"def is_confirmed(self, is_confirmed: bool):\\n\\n self._is_confirmed = is_confirmed\",\n \"def maybe_notify_lessee(request, response):\\n if request.get('pubsub_topic'):\\n pubsub.publish(\\n pubsub.full_topic_name(\\n request['pubsub_project'], request['pubsub_topic']),\\n json.dumps(response),\\n {},\\n )\\n metrics.pubsub_messages_sent.increment(fields={'target': 'lessee'})\",\n \"def on_connect(client, userdata, flags, rc):\\n print('Connected with result code ' + str(rc))\\n client.subscribe(MQTT_TOPIC, qos=1)\",\n \"def test_resend_activation_email(self):\\n\\n data = {\\n 'email': self.user.email,\\n }\\n\\n response = self.client.post(\\n reverse('user-resend-activation-email'),\\n data,\\n format='json',\\n )\\n\\n self.assertEqual(\\n response.status_code,\\n status.HTTP_200_OK,\\n response.content\\n )\\n\\n self.assertEqual(\\n response.content,\\n b'',\\n )\\n\\n self.assertEqual(len(mail.outbox), 1)\",\n \"def subscribe(self):\\n if not self._subscribed and self._connected:\\n if ATTR_STREAM_ID not in self.data:\\n msg = self._create_message(strings.SUB_MSG)\\n self.write(msg)\\n else:\\n msg = self._create_message(strings.RESUB_MSG)\\n self.write(msg)\\n self._subscribed = True\"\n]","isConversational":false},"negative_scores":{"kind":"list like","value":["0.7161424","0.6835099","0.6736429","0.65938777","0.6326111","0.6227474","0.6216683","0.6181522","0.6166237","0.61440307","0.6120497","0.61191845","0.60010856","0.5983212","0.595965","0.5947182","0.5929845","0.5913996","0.5880866","0.5878438","0.58783746","0.5854324","0.5838064","0.58184385","0.5812429","0.5805716","0.5793445","0.57694453","0.576241","0.5754764","0.575086","0.5743451","0.57420033","0.5738133","0.57375056","0.5737393","0.57274103","0.572582","0.5708896","0.569854","0.56944036","0.5655669","0.56538945","0.5645436","0.5643742","0.5633316","0.5620609","0.56121445","0.5607224","0.5604434","0.5583541","0.55741954","0.55724263","0.55682874","0.5562985","0.5562985","0.555806","0.5553148","0.554474","0.5543126","0.5516758","0.5485422","0.5481461","0.5473959","0.54549557","0.5454939","0.54421055","0.5439339","0.54345787","0.5423172","0.5418822","0.54087937","0.54059875","0.54022354","0.5400946","0.53978074","0.53920954","0.5388167","0.53848636","0.536616","0.53486633","0.53439665","0.53402424","0.5340184","0.5339595","0.5338725","0.53336275","0.53220546","0.53220546","0.5314708","0.5314634","0.5313908","0.53132284","0.52949697","0.5292222","0.528708","0.5282741","0.52822906","0.5275533","0.52748466"],"string":"[\n \"0.7161424\",\n \"0.6835099\",\n \"0.6736429\",\n \"0.65938777\",\n \"0.6326111\",\n \"0.6227474\",\n \"0.6216683\",\n \"0.6181522\",\n \"0.6166237\",\n \"0.61440307\",\n \"0.6120497\",\n \"0.61191845\",\n \"0.60010856\",\n \"0.5983212\",\n \"0.595965\",\n \"0.5947182\",\n \"0.5929845\",\n \"0.5913996\",\n \"0.5880866\",\n \"0.5878438\",\n \"0.58783746\",\n \"0.5854324\",\n \"0.5838064\",\n \"0.58184385\",\n \"0.5812429\",\n \"0.5805716\",\n \"0.5793445\",\n \"0.57694453\",\n \"0.576241\",\n \"0.5754764\",\n \"0.575086\",\n \"0.5743451\",\n \"0.57420033\",\n \"0.5738133\",\n \"0.57375056\",\n \"0.5737393\",\n \"0.57274103\",\n \"0.572582\",\n \"0.5708896\",\n \"0.569854\",\n \"0.56944036\",\n \"0.5655669\",\n \"0.56538945\",\n \"0.5645436\",\n \"0.5643742\",\n \"0.5633316\",\n \"0.5620609\",\n \"0.56121445\",\n \"0.5607224\",\n \"0.5604434\",\n \"0.5583541\",\n \"0.55741954\",\n \"0.55724263\",\n \"0.55682874\",\n \"0.5562985\",\n \"0.5562985\",\n \"0.555806\",\n \"0.5553148\",\n \"0.554474\",\n \"0.5543126\",\n \"0.5516758\",\n \"0.5485422\",\n \"0.5481461\",\n \"0.5473959\",\n \"0.54549557\",\n \"0.5454939\",\n \"0.54421055\",\n \"0.5439339\",\n \"0.54345787\",\n \"0.5423172\",\n \"0.5418822\",\n \"0.54087937\",\n \"0.54059875\",\n \"0.54022354\",\n \"0.5400946\",\n \"0.53978074\",\n \"0.53920954\",\n \"0.5388167\",\n \"0.53848636\",\n \"0.536616\",\n \"0.53486633\",\n \"0.53439665\",\n \"0.53402424\",\n \"0.5340184\",\n \"0.5339595\",\n \"0.5338725\",\n \"0.53336275\",\n \"0.53220546\",\n \"0.53220546\",\n \"0.5314708\",\n \"0.5314634\",\n \"0.5313908\",\n \"0.53132284\",\n \"0.52949697\",\n \"0.5292222\",\n \"0.528708\",\n \"0.5282741\",\n \"0.52822906\",\n \"0.5275533\",\n \"0.52748466\"\n]","isConversational":false},"document_score":{"kind":"string","value":"0.66522044"},"document_rank":{"kind":"string","value":"3"}}},{"rowIdx":93290,"cells":{"query":{"kind":"string","value":"Unsubscribe previous email, wait for new one to be confirmed."},"document":{"kind":"string","value":"def on_email_email_changed(\n request, user, from_email_address: EmailAddress, to_email_address, **kwargs\n):\n if from_email_address:\n dillo.tasks.profile.update_mailing_list_subscription(from_email_address.email, False)"},"metadata":{"kind":"string","value":"{\n \"objective\": {\n \"self\": [],\n \"paired\": [],\n \"triplet\": [\n [\n \"query\",\n \"document\",\n \"negatives\"\n ]\n ]\n }\n}"},"negatives":{"kind":"list like","value":["def unsubscribe_all_emails_from_cloudwatch(self):\n for subscription in self._get_cloudwatch_subscriptions():\n subscription_arn = subscription['SubscriptionArn']\n if not subscription_arn == 'PendingConfirmation':\n self.conn.unsubscribe(subscription_arn)","def unsubscribe():\n from alot.helper import mailto_to_envelope\n from alot.buffers import EnvelopeBuffer\n msg = ui.current_buffer.get_selected_message()\n e = msg.get_email()\n uheader = e['List-Unsubscribe']\n dtheader = e.get('Delivered-To', None)\n\n if uheader is not None:\n M = re.search(r'<(mailto:\\S*)>', uheader)\n if M is not None:\n env = mailto_to_envelope(M.group(1))\n if dtheader is not None:\n env['From'] = [dtheader]\n ui.buffer_open(EnvelopeBuffer(ui, env))\n else:\n ui.notify('focussed mail contains no \\'List-Unsubscribe\\' header',\n 'error')","def _unsubscribe(self):\n self.unsubscribe_date = now()\n self.unsubscribed = True\n self.subscribed = False","def unsubscribe(self, request):\n email = self.cleaned_data.get('email')\n subscriber = Subscriber.objects.get(email=email, mailing_list=self.mailing_list)\n subscriber.unsubscribe(request)","async def aoc_unsubscribe(self, ctx: commands.Context) -> None:\n role = ctx.guild.get_role(AocConfig.role_id)\n\n if role in ctx.author.roles:\n await ctx.author.remove_roles(role)\n await ctx.send(\"Okay! You have been __unsubscribed__ from notifications about new Advent of Code tasks.\")\n else:\n await ctx.send(\"Hey, you don't even get any notifications about new Advent of Code tasks currently anyway.\")","def unsubscribe(receiver):","def unsubscribe(receiver):","def unsubscribe(receiver):","def unsubscribe(receiver):","def unsubscribe(receiver):","async def aoc_unsubscribe(self, ctx: commands.Context) -> None:\n if ctx.channel.id != settings.aoc.channel_id:\n await ctx.send(f\"Please use the <#{settings.aoc.channel_id}> channel\")\n return\n\n role = ctx.guild.get_role(settings.aoc.role_id)\n\n if role in ctx.author.roles:\n await ctx.author.remove_roles(role)\n await ctx.send(\"Okay! You have been __unsubscribed__ from notifications about new Advent of Code tasks.\")\n else:\n await ctx.send(\"Hey, you don't even get any notifications about new Advent of Code tasks currently anyway.\")","def unsubscribe(self, subject):\n pass","def unsubscribe(self, email, id):\n self.db.commit([\n 'UPDATE comments SET',\n ' notification=0',\n 'WHERE email=%s AND (id=%s OR parent=%s);'], (email, id, id))","def unsubscribe(self, feed, **args):\n args.update(feed=feed)\n return self.fetch(\"/unsubscribe\", post_args=args)","def Promote_from_waitlist(self, email):\n if email in self.Waitlist:\n self.Waitlist.remove(email)\n self.Add_attendee(email)\n else:\n raise MissingAddressException(email)","def test_reactivation_for_unregistered_user(self, email_user):\r\n response_data = self.reactivation_email(self.unregisteredUser)\r\n\r\n self.assertFalse(response_data['success'])","def unsubscribe(request):\n user = None\n if request.user.is_authenticated:\n user = request.user\n elif \"email\" in request.GET:\n try:\n notification = Notification.objects.get(pk=request.GET[\"email\"])\n user = User.objects.get(email=notification.email)\n except (Notification.DoesNotExist, User.DoesNotExist):\n pass\n\n if not user:\n return redirect(\"/accounts/login/?next=/accounts/profile/unsubscribe/\")\n\n subscriptions = user.subscriptions.filter(active=True).order_by(\"-created_at\")\n if request.method == \"POST\":\n count = subscriptions.update(active=False)\n messages.success(request, f\"Successfully deactivated {count} subscriptions.\")\n return redirect(\"/accounts/profile/\")\n else:\n return render(\n request, \"account/unsubscribe.html\", {\"subscriptions\": subscriptions}\n )","def unsubscribe(self):\n pass # pragma: no cover","def pop():\n task = connection.zrange(QUEUE_KEY, 0, 0)\n if not task:\n return False, 'No emails now!'\n msg_id = task[0]\n timestamp = connection.zscore(QUEUE_KEY, msg_id)\n now = datetime.datetime.now().timestamp()\n if timestamp < now or abs(timestamp - now) <= 1e-6:\n message = connection.get(msg_id)\n pipeline = connection.pipeline()\n pipeline.zrem(QUEUE_KEY, msg_id)\n pipeline.delete(msg_id)\n pipeline.execute()\n return True, message\n return False, \"It's too early now!\"","def unsubscribe( self, mess, args):\n user = self.get_sender_username(mess)\n if not user in self.users:\n return 'You are not subscribed!'\n else:\n user = self.users.pop(user)\n self.message_queue.append('_%s has left the channel_' % user)\n self.log.info( '%s unsubscribed from the broadcast.' % user)\n self.save_state()\n return 'You are now unsubscribed.'","def _unsubscribe(self):\n if not self.subscribed:\n return False\n return {}","def unsubscribe(self, update, context):\n # remove or update to the sqlite table.\n chat = update.message.chat\n self.db_manager.remove_user(chat.id)\n self.logger.info(\n 'Username: %s and chat_id: %s unsubscribed to the list.' % (chat.username, chat.id)\n )\n update.message.reply_text('You have successfully unsubscribed the notifications forever.')","def _async_unsubscribe(self, topic: str) -> None:\n if self._is_active_subscription(topic):\n if self._max_qos[topic] == 0:\n return\n subs = self._matching_subscriptions(topic)\n self._max_qos[topic] = max(sub.qos for sub in subs)\n # Other subscriptions on topic remaining - don't unsubscribe.\n return\n if topic in self._max_qos:\n del self._max_qos[topic]\n if topic in self._pending_subscriptions:\n # Avoid any pending subscription to be executed\n del self._pending_subscriptions[topic]\n\n self._pending_unsubscribes.add(topic)\n self._unsubscribe_debouncer.async_schedule()","def cb_stop(self, update, context):\n\n print(f\"Unsubscribing chat_id '{update.message.chat_id}'\")\n try:\n self.clientChatIds.remove(update.message.chat_id)\n answer = \"You sucessfully unsubscribed.\"\n self.saveToFile(self.configFile)\n except KeyError:\n answer = \"You are not subscribed.\"\n\n update.message.reply_text(answer)","def opt_out_email_updates(request, token, course_id):\n try:\n username = UsernameCipher().decrypt(token).decode(\"utf-8\")\n user = User.objects.get(username=username)\n course_key = CourseKey.from_string(course_id)\n course = get_course_by_id(course_key, depth=0)\n except UnicodeDecodeError:\n raise Http404(\"base64url\") # lint-amnesty, pylint: disable=raise-missing-from\n except UsernameDecryptionException as exn:\n raise Http404(str(exn)) # lint-amnesty, pylint: disable=raise-missing-from\n except User.DoesNotExist:\n raise Http404(\"username\") # lint-amnesty, pylint: disable=raise-missing-from\n except InvalidKeyError:\n raise Http404(\"course\") # lint-amnesty, pylint: disable=raise-missing-from\n\n unsub_check = request.POST.get('unsubscribe', False)\n context = {\n 'course': course,\n 'unsubscribe': unsub_check\n }\n\n if request.method == 'GET':\n return render_to_response('bulk_email/confirm_unsubscribe.html', context)\n\n if request.method == 'POST' and unsub_check:\n Optout.objects.get_or_create(user=user, course_id=course_key)\n log.info(\n \"User %s (%s) opted out of receiving emails from course %s\",\n user.username,\n user.email,\n course_id,\n )\n\n return render_to_response('bulk_email/unsubscribe_success.html', context)","def on_email_confirmed(request, email_address: EmailAddress, **kwargs):\n dillo.tasks.profile.update_mailing_list_subscription(email_address.email, True)","def remove_from_mailing_list(self, email):\n if not self.list_removal_endpoint or not self.list_key:\n # fail silently, log info\n print('REMOVE_ON_DELETE is turned on, but required '\n 'fields are missing.')\n return\n\n try:\n email = email.encode('utf-8').lower()\n email_hash = hashlib.md5(email).hexdigest()\n res = requests.delete(self.list_removal_endpoint.format(email_hash),\n auth=('nop', self.list_key),\n timeout=1.5)\n\n if res.status_code != 204:\n print('Unexpected mailing list API response.. '\n 'status code: {0.status_code}\\n'\n 'content: {0.content}'.format(res))\n\n except Exception as e:\n if e is requests.exceptions.Timeout:\n print('Mailing list API timed out..')\n else:\n print('Removing from mailing list failed:', e)","def test_notification_cancellation_email(self):\n api.content.transition(obj=self.event, transition='cancel')\n mailhost = api.portal.get_tool('MailHost')\n self.assertEqual(len(mailhost.messages), 2)\n msg = message_from_string(mailhost.messages[1])\n\n self.assertEqual(msg['To'], BOARD_LIST_ADDRESS)\n self.assertEqual(\n msg['From'], 'EESTEC International <noreply@eestec.net>')\n self.assertEqual(\n msg['Subject'],\n '=?utf-8?q?=5BEVENTS=5D=5BCancelled=5D_T=C3=A9st_event?=',\n )\n self.assertIn('an Event has been cancelled', msg.get_payload())\n self.assertIn('http://nohost/plone/lc/test-event', msg.get_payload())","def unsubscribe(self, destination, *args, **kwargs):","def unsubscribe(self, item_name):\n self.subscribed = None","def unsubscribe_by_mail(self):\n qry = ServiceOperationQuery(self, \"unsubscribeByMail\")\n self.context.add_query(qry)\n return self","def test_email_after_contest_end(self):\n self.prep_consumer()\n temp_date = settings.CONTEST_END_DATE\n settings.CONTEST_END_DATE = str(\n datetime.today().date() - timedelta(days=1))\n UnqualifiedConsumerEmailTask().run(test_mode=self.consumer)\n log = get_last_db_log(\n 'email_gateway.tasks.send_unqualified_emails', 'EMAIL')\n if log:\n self.fail('Performed task even though contest ended.')\n settings.CONTEST_END_DATE = temp_date","def unsubscribe(self):\n res = self._unsubscribe()\n if res is not None:\n self._subscribed = False\n return res","def delete_email_from_blacklist(self, email):\n logger.info(\"Function call: delete_email_from_blacklist for '{}'\".format(email, ))\n return self.__handle_error('Empty email') if not email else self.__handle_result(self.__send_request('blacklist', 'DELETE', {'emails': base64.b64encode(email)}))","def unsubscribe(self, feedUrl):\r\n response = self.httpPost(\r\n ReaderUrl.SUBSCRIPTION_EDIT_URL,\r\n {'ac':'unsubscribe', 's': feedUrl})\r\n # FIXME - need better return API\r\n if response and 'OK' in response:\r\n return True\r\n else:\r\n return False","def unsubscribe(self, sub_code):\n\t\tresult = {}\n\t\tconnection = DbHelper.connect()\n\t\tsub_id = 0\n\n\t\ttry:\n\t\t\twith connection.cursor() as cursor:\n\t\t\t\tsql = \"SELECT * FROM mail_list \\\n\t\t\t\t\t WHERE email_hash=%s;\"\n\t\t\t\tcursor.execute(sql, [sub_code])\n\t\t\t\tresult = cursor.fetchone()\n\t\t\t\t\n\t\t\t\tif not result:\n\t\t\t\t\tconnection.close()\n\t\t\t\t\treturn \"CODE_DOES_NOT_EXIST\"\n\t\t\t\telif not result['is_activated']:\n\t\t\t\t\tconnection.close()\n\t\t\t\t\treturn \"CODE_NOT_ACTIVATED\"\n\n\t\t\t\tsub_id = result['sub_id']\n\n\t\t\t\tsql = \"DELETE FROM mail_list \\\n\t\t\t\t\t WHERE sub_id=%s;\"\n\t\t\t\tcursor.execute(sql, [sub_id])\n\t\t\t\tconnection.commit()\n\t\t\t\tconnection.close()\n\t\t\t\treturn \"UNSUBSCRIBED\"\n\t\texcept pymysql.MySQLError as e:\n\t\t\tconnection.close()\n\t\t\treturn \"DATABASE_ERROR\"","def notify_email_confirmed(self, user, email):\n \n # make sure user isn't still invited to groups he owns or is a member of\n for g in self.users_groups(user):\n g.remove_invitation(user)","def unsubscribe(endpoint: str, topic: str, timeout: int = 5):\n global logger\n logger.info(f\"Unsubscribing from topic '{topic}' ...\")\n action = {\"action\": \"unsubscribe\", \"topic\": topic}\n reply = send_manage_message(endpoint, action, timeout)\n if not reply_is_success(reply):\n logger.warning(\"Unsubscription failed\")\n return\n logger.info(\"Unsubscription successful\")","def test_unsubscribe(self):\n dest = '/topic/dest'\n\n self.tm.subscribe(self.conn, dest)\n f = Frame(frames.MESSAGE, headers={'destination': dest}, body='Empty')\n self.tm.send(f)\n\n self.assertEqual(len(self.conn.frames), 1)\n subscription = self.conn.frames[0].headers.pop(\"subscription\", None)\n self.assertEqual(subscription, 0)\n self.assertEqual(self.conn.frames[0], f)\n\n self.tm.unsubscribe(self.conn, dest)\n f = Frame(frames.MESSAGE, headers={'destination': dest}, body='Empty')\n self.tm.send(f)\n\n self.assertEqual(len(self.conn.frames), 1)","def on_unsubscribe(self, mqtt_client, userdata, mid ):\n logging.debug(\"DEBUG - unsubscribe ack received\")","def test_unsubscribe(self):\n self.service.clientConnected()\n\n unsubscribers = []\n self.service.subscribe(u'url', 1\n ).addCallback(lambda fn: unsubscribers.append(fn))\n self.service.subscribe(u'url', 2\n ).addCallback(lambda fn: unsubscribers.append(fn))\n\n pubsubClient = self.service.pubsubClient\n self.assertIn(u'url', pubsubClient.subscriptions)\n\n unsubscribers.pop()()\n self.service.unsubscribe(u'url')\n self.assertIn(u'url', pubsubClient.subscriptions)\n\n unsubscribers.pop()()\n self.service.unsubscribe(u'url')\n self.assertNotIn(u'url', pubsubClient.subscriptions)","def _unsubscribe(self, signal):\n while signal in self._downstream:\n self._downstream.remove(signal)\n while signal in self._downstream_reconnect:\n self._downstream_reconnect.remove(signal)","async def unsubscribe(self):\n LOGGER.info('Subscription removed')\n await self._ros.send(self._unsubscribe_msg)","def test_cancelling_more_than_once_only_emails_once(self, mock_tz):\n mock_tz.now.return_value = datetime(2015, 2, 13, 17, 15, tzinfo=dt_timezone.utc)\n\n # make full event (setup has one paid and one unpaid)\n # cancellation period =1, date = 2015, 2, 13, 18, 0\n self.event.max_participants = 3\n self.event.save()\n\n # make another booking that will be cancelled\n baker.make_recipe(\n 'booking.booking', event=self.event, paid=False,\n payment_confirmed=False, status='OPEN',\n user__email=\"unpaid@test.com\",\n date_booked=datetime(2015, 2, 9, 18, 0, tzinfo=dt_timezone.utc),\n warning_sent=True,\n date_warning_sent=datetime(2015, 2, 9, 20, 0, tzinfo=dt_timezone.utc),\n )\n\n # make some waiting list users\n for i in range(3):\n baker.make_recipe(\n 'booking.waiting_list_user', event=self.event,\n user__email='test{}@test.com'.format(i)\n )\n\n management.call_command('cancel_unpaid_bookings')\n # emails are sent to user per cancelled booking (2) and\n # one email with bcc to waiting list (1) and studio (1)\n # waiting list email sent after the two user emails\n self.assertEqual(len(mail.outbox), 4)\n self.assertEqual(\n sorted(mail.outbox[2].bcc),\n ['test0@test.com', 'test1@test.com', 'test2@test.com']\n )\n for email in [mail.outbox[0], mail.outbox[1], mail.outbox[3]]:\n self.assertEqual(email.bcc, [])\n\n self.assertEqual(Booking.objects.filter(status='CANCELLED').count(), 2)","def rpc_campaign_alerts_unsubscribe(self, campaign_id):\n\t\tusername = self.basic_auth_user\n\t\tsession = db_manager.Session()\n\t\tquery = session.query(db_models.AlertSubscription)\n\t\tquery = query.filter_by(campaign_id=campaign_id, user_id=username)\n\t\tsubscription = query.first()\n\t\tif subscription:\n\t\t\tsession.delete(subscription)\n\t\t\tsession.commit()\n\t\tsession.close()\n\t\treturn","async def unsubscribe(self, topic: str, subscription_id: int = None) -> None:\n ...","def unsubscribe(self, path):\n\n box = self._getMailbox(path)\n\n box.metadata['subscribed'] = False\n\n box.saveMetadata()\n\n return True","async def _async_perform_unsubscribes(self) -> None:\n if not self._pending_unsubscribes:\n return\n\n topics = list(self._pending_unsubscribes)\n self._pending_unsubscribes = set()\n\n async with self._paho_lock:\n result, mid = await self.hass.async_add_executor_job(\n self._mqttc.unsubscribe, topics\n )\n _raise_on_error(result)\n for topic in topics:\n _LOGGER.debug(\"Unsubscribing from %s, mid: %s\", topic, mid)\n\n await self._wait_for_mid(mid)","def unsubscribe_user(sub_id):\n repository.unsubscribe(sub_id)\n return flask.render_template('www/email/unsubscribed.html')","def confirm_email(self):\n # The base class' implementation does nothing\n pass","def test_unsubscribe_offer(self):\n pass","def unsubscribe(request, action):\n data = {}\n if action and action.lower() == 'updates':\n user_profile = request.user.revolvuserprofile\n user_profile.subscribed_to_repayment_notifications = False\n user_profile.save()\n data = {'msg': \"You have successfully unsubscribed\"}\n elif action and action.lower() == 'confirmation':\n return render_to_response('base/unsub_conf.html', context_instance=RequestContext(request, data))\n else:\n data = {'msg': 'Please specify which section you want to unsubscribe'}\n\n return render_to_response('base/minimal_message.html',\n context_instance=RequestContext(request, data))","def view_mailinglist_unsubscribe(request, slug, uidb36, token):\n newsletter = get_object_or_404(Newsletter, slug=slug)\n contact = untokenize(uidb36, token)\n\n already_unsubscribed = contact in newsletter.mailing_list.unsubscribers.all()\n\n if request.POST.get('email') and not already_unsubscribed:\n newsletter.mailing_list.unsubscribers.add(contact)\n newsletter.mailing_list.save()\n already_unsubscribed = True\n ContactMailingStatus.objects.create(newsletter=newsletter, contact=contact,\n status=ContactMailingStatus.UNSUBSCRIPTION)\n\n return render_to_response('newsletter/mailing_list_unsubscribe.html',\n {'email': contact.email,\n 'already_unsubscribed': already_unsubscribed},\n context_instance=RequestContext(request))","def _logout(self):\n if self.connected:\n try:\n self._smtp.quit()\n except smtplib.SMTPServerDisconnected:\n pass\n self._connected = False","def unsubscribe(self):\n if self._subscribed and self._connected:\n try:\n msg = self._create_message(strings.UNSUB_MSG)\n self.write(msg)\n except (OSError, KeyError) as ex:\n _LOGGER.error(\n \"PyISY encountered a socket error while writing unsubscribe message to the socket: %s.\",\n ex,\n )\n self._subscribed = False\n self.disconnect()","def handle_email_confirmed(sender, **kwargs):\n email = kwargs['email_address']\n email.user.userprofile.member.cast().confirm_email()","def unsubscribe_all(self):\n ids = []\n repeat = True\n cursor = None\n # get all ids\n while repeat:\n ret = self.__twitch.get_eventsub_subscriptions(after=cursor)\n for d in ret.get('data', []):\n ids.append(d.get('id'))\n cursor = ret.get('pagination', {}).get('cursor')\n repeat = cursor is not None\n for _id in ids:\n succ = self.__twitch.delete_eventsub_subscription(_id)\n if not succ:\n self.__logger.warning(f'failed to unsubscribe from event {_id}')\n self.__callbacks.clear()","def _handle_unsubscribe_request(state, client, request):\n # TODO: create me","def email_subscribe_pending_confirm(hexdomain):\n domain = tools.parse_domain(hexdomain)\n if domain is None:\n flask.abort(400, 'Malformed domain or domain not represented in hexadecimal format.')\n\n email_address = flask.request.form['email_address']\n\n if email_address.strip() == '':\n return flask.redirect('/email/subscribe/{}/0'.format(hexdomain))\n\n verify_code = tools.random_id()\n verify_url = flask.request.url_root + 'email/verify/{}'.format(verify_code)\n email_body = email_tools.render_email(\n 'confirm.html',\n domain=domain,\n verify_url=verify_url\n )\n\n repository.propose_subscription(verify_code, email_address, domain)\n emailer.send(\n email_address, 'Please verify your subscription', email_body\n )\n\n return flask.render_template('www/email/pending_verify.html', domain=domain)","def unenroll_email(course_id, student_email, email_students=False, email_params=None):\r\n previous_state = EmailEnrollmentState(course_id, student_email)\r\n\r\n if previous_state.enrollment:\r\n CourseEnrollment.unenroll_by_email(student_email, course_id)\r\n if email_students:\r\n email_params['message'] = 'enrolled_unenroll'\r\n email_params['email_address'] = student_email\r\n email_params['full_name'] = previous_state.full_name\r\n send_mail_to_student(student_email, email_params)\r\n\r\n if previous_state.allowed:\r\n CourseEnrollmentAllowed.objects.get(course_id=course_id, email=student_email).delete()\r\n if email_students:\r\n email_params['message'] = 'allowed_unenroll'\r\n email_params['email_address'] = student_email\r\n # Since no User object exists for this student there is no \"full_name\" available.\r\n send_mail_to_student(student_email, email_params)\r\n\r\n after_state = EmailEnrollmentState(course_id, student_email)\r\n\r\n return previous_state, after_state","async def async_unsubscribe_services(self) -> None:\n # Delete list of subscriptions and cancel renewal before unsubcribing\n # to avoid unsub-resub race.\n sids = list(self._subscriptions)\n self._subscriptions.clear()\n await self._update_resubscriber_task()\n\n await asyncio.gather(*(self._async_unsubscribe_service(sid) for sid in sids))","def test_unverified_subscriber(self):\n self.prep_consumer()\n subscriber = Subscriber.objects.get(id=7)\n self.consumer.subscriber = subscriber\n self.consumer.save()\n UnqualifiedConsumerEmailTask().run(test_mode=self.consumer)\n self.common_asserts()\n self.assertTrue('71010' in mail.outbox[0].alternatives[0][0])\n self.assertTrue('71010' in mail.outbox[0].body)","async def async_will_remove_from_hass(self) -> None:\n async_unsubscribe_topics(self.hass, self._sub_state)\n self._sub_state = None","def onUnsubscribed(self, connection:MQTTConnection, topic:str) -> bool:\n\t\tconnection.subscribedCount -= 1\n\t\treturn True","def unsubscribe(self, destination, extra_headers=None):\n unsubscribe = frame.UnsubscribeFrame(destination, extra_headers=extra_headers)\n res = self.send_frame(unsubscribe)\n with self.subscription_lock:\n self.subscribed_destinations.pop(destination)\n return res","def unsubscribe_from_guest_statistics(self):\n\t\treturn Job(SDK.PrlVm_UnsubscribeFromGuestStatistics(self.handle)[0])","def test_cancelling_for_full_event_emails_waiting_list(self, mock_tz):\n mock_tz.now.return_value = datetime(\n 2015, 2, 13, 17, 15, tzinfo=dt_timezone.utc\n )\n\n # make full event (setup has one paid and one unpaid)\n # cancellation period =1, date = 2015, 2, 13, 18, 0\n self.event.max_participants = 2\n self.event.save()\n\n # make some waiting list users\n for i in range(3):\n baker.make_recipe(\n 'booking.waiting_list_user', event=self.event,\n user__email='test{}@test.com'.format(i)\n )\n\n management.call_command('cancel_unpaid_bookings')\n # emails are sent to user per cancelled booking (1) and\n # one email with bcc to waiting list (1) and studio (1)\n self.assertEqual(len(mail.outbox), 3)\n self.assertEqual(\n sorted(mail.outbox[1].bcc),\n ['test0@test.com', 'test1@test.com', 'test2@test.com']\n )","def unsubscribe(self):\r\n self._unregister()","def unsubscribe(self):\n\n # Unsubscribe\n self.pyrps.redis.srem(self.pyrps._ns_subscriptions(self.queue), self.consumer_id) \n\n # Remove message queue\n self.pyrps.redis.delete(self.pyrps._ns_queue(self.queue, self.consumer_id))","def email_ml_unsubscribe(self, maillist=None, maillist_uid=None, subscriber=None, subscriber_uid=None):\n if not maillist and not maillist_uid:\n raise ValueError('Maillist or uid required')\n if not subscriber and not subscriber_uid:\n raise ValueError('Subscriber or uid required')\n return self._request('email/ml/unsubscribe', inspect_args_func(inspect.currentframe()))","def test_resend_activation_email_nonunique_email(self):\n user1 = self.registration_profile.objects.create_inactive_user(\n site=Site.objects.get_current(), send_email=False, **self.user_info)\n user2_info = copy(self.user_info)\n user2_info['username'] = 'bob'\n user2 = self.registration_profile.objects.create_inactive_user(\n site=Site.objects.get_current(), send_email=False, **user2_info)\n self.assertEqual(user1.email, user2.email)\n self.assertFalse(self.registration_profile.objects.resend_activation_mail(\n email=self.user_info['email'],\n site=Site.objects.get_current(),\n ))\n self.assertEqual(len(mail.outbox), 0)","def unsubscribe(request):\n address = request.POST.get('address')\n\n Subscription.objects.filter(address=address).delete()\n\n return HttpResponse(json.dumps({\n \"status\": \"success\"\n }, default=helpers.json_custom_parser), content_type='application/json')","async def _send_message_for_unsubscribe(self, data_id):\n body = json.dumps(\n dict(\n action='unsub',\n data_id=data_id\n )\n ).encode('utf-8')\n await self._send_message_in_queue(self.queue_crypto_quotes_service, body)","def terminate(self):\n self.mailQueue.put(None)\n self.mailQueue.join()\n self.join()\n self.logger.info(\"Mailer terminated\")","def unsubscribe_all_known(self):\n for key, value in self.__callbacks.items():\n self.__logger.debug(f'unsubscribe from event {key}')\n succ = self.__twitch.delete_eventsub_subscription(key)\n if not succ:\n self.__logger.warning(f'failed to unsubscribe from event {key}')\n self.__callbacks.clear()","def test_resend_activation_email(self):\n user = self.registration_profile.objects.create_inactive_user(\n site=Site.objects.get_current(), send_email=False, **self.user_info)\n self.assertEqual(len(mail.outbox), 0)\n\n profile = self.registration_profile.objects.get(user=user)\n orig_activation_key = profile.activation_key\n\n self.assertTrue(self.registration_profile.objects.resend_activation_mail(\n email=self.user_info['email'],\n site=Site.objects.get_current(),\n ))\n\n profile = self.registration_profile.objects.get(pk=profile.pk)\n new_activation_key = profile.activation_key\n\n self.assertNotEqual(orig_activation_key, new_activation_key)\n self.assertEqual(len(mail.outbox), 1)","def unsubscribe(self, topic):\n request = protos.RequestUnsubscribe(topic=topic)\n return self.stub.unsubscribe(request)","def stop(self):\r\n self._incompleteMsgs = set()\r\n\r\n if self._cleaner.running:\r\n self._cleaner.stop()","async def unsubscribe(connection, message):\n from high_templar.hub import NotSubscribedException\n\n if 'requestId' not in message:\n return await connection.send({\n 'code': 'error',\n 'message': 'no-request-id'\n })\n\n for subscription in connection.app.hub.subscriptions[connection]:\n if subscription.request_id == message['requestId']:\n try:\n connection.app.hub.unsubscribe(subscription)\n await connection.send({ 'code': 'success' })\n except NotSubscribedException:\n await connection.send({\n 'code': 'error',\n 'message': 'not-subscribed'\n })\n break\n else:\n return await connection.send({\n 'code': 'error',\n 'message': 'not-subscribed'\n })","def remove_user(self, email):\n return run_transaction(\n self.sessionfactory,\n lambda session: remove_user_txn(session, email))","def disconnect(self, extra_headers=None):\n try:\n if self.connection.connected:\n with self.subscription_lock:\n # Need a copy since unsubscribe() removes the destination from the collection.\n subcpy = copy(self.subscribed_destinations)\n for destination in subcpy:\n self.unsubscribe(destination)\n disconnect = frame.DisconnectFrame(extra_headers=extra_headers)\n result = self.send_frame(disconnect)\n try:\n self.connection.disconnect()\n except NotConnectedError:\n pass\n return result\n finally:\n self.shutdown_event.set()","def unsubscribe(self, jid=None, username=None, domain=None):\n if jid is not None:\n self._pres_manager.flag_offline(jid)\n elif username is not None and domain is not None:\n self._pres_manager.flag_offline(xmpp.JID(node=username, domain=domain))\n self._roster.unsubscribe(jid=jid, username=username, domain=domain)","def teardown(self):\n try:\n self.loop.run_until_complete(self.webhook_connection.disconnect())\n except Exception:\n print_exc()\n raise","def _onUnsubscribe(self, client:mqtt.Client, userdata:Any, mid:int) -> None:\n\t\t# TODO doc, error check when not connected, not subscribed\n\t\tfor t in self.subscribedTopics.values():\n\t\t\tif t.mid == mid:\n\t\t\t\tdel self.subscribedTopics[t.topic]\n\t\t\t\tself.messageHandler and self.messageHandler.onUnsubscribed(self, t.topic)\n\t\t\t\tbreak","def delete_email_from_all_addressooks(self, email):\n logger.info(\"Function call: delete_email_from_all_addressooks for '{}'\".format(email, ))\n return self.__handle_error('Empty email') if not email else self.__handle_result(self.__send_request('emails/{}'.format(email, ), 'DELETE'))","def unsubscribe(observer):","def unsubscribe(observer):","def test_generate_confirmational_email_for_student_that_already_received_email(self):\n self.student2.has_received_email = True\n self.student2.save()\n\n template = self.email_confirmation_template.name\n confirm_interview_url = \"confirm-interview-test-url\"\n choose_interview_url = \"choose-interview-test-url\"\n\n confirm_email_generator = GenerateConfirmEmails(\n template, confirm_interview_url, choose_interview_url)\n\n confirm_email_generator.generate_confirmation_emails()\n\n client = Client()\n client.login(\n email=self.teacher_admin.email,\n password='123'\n )\n\n url = reverse('admin:post_office_email_changelist')\n response = client.get(url, follow=True)\n\n result_list = response.context_data['cl'].result_list\n\n # Two students dont have interviews and the third already received email\n # There should be no emails generated in Email change list\n self.assertEqual(len(result_list), 0)","def unsubscribe(self, user_token, topic):\n response = _request('DELETE',\n url=self.url_v1('/user/subscriptions/' + topic),\n user_agent=self.user_agent,\n user_token=user_token,\n )\n _raise_for_status(response)","def unsubscribe_feed():\n\n source_id = request.args.get('source_id', type=int)\n if source_id:\n current_user.unsubscribe_feed(source_id)\n flash('Unsubscribe successful source=%s' % source_id)\n\n return redirect(url_for('mod_feed.manage_subscriptions'))","def unconfirm(self):\n self.automatically_detected=True\n self.save()","def confirm_email_change(request, key):\r\n try:\r\n try:\r\n pec = PendingEmailChange.objects.get(activation_key=key)\r\n except PendingEmailChange.DoesNotExist:\r\n response = render_to_response(\"invalid_email_key.html\", {})\r\n transaction.rollback()\r\n return response\r\n\r\n user = pec.user\r\n address_context = {\r\n 'old_email': user.email,\r\n 'new_email': pec.new_email\r\n }\r\n\r\n if len(User.objects.filter(email=pec.new_email)) != 0:\r\n response = render_to_response(\"email_exists.html\", {})\r\n transaction.rollback()\r\n return response\r\n\r\n subject = render_to_string('emails/email_change_subject.txt', address_context)\r\n subject = ''.join(subject.splitlines())\r\n message = render_to_string('emails/confirm_email_change.txt', address_context)\r\n up = UserProfile.objects.get(user=user)\r\n meta = up.get_meta()\r\n if 'old_emails' not in meta:\r\n meta['old_emails'] = []\r\n meta['old_emails'].append([user.email, datetime.datetime.now(UTC).isoformat()])\r\n up.set_meta(meta)\r\n up.save()\r\n # Send it to the old email...\r\n try:\r\n user.email_user(subject, message, settings.DEFAULT_FROM_EMAIL)\r\n except Exception:\r\n log.warning('Unable to send confirmation email to old address', exc_info=True)\r\n response = render_to_response(\"email_change_failed.html\", {'email': user.email})\r\n transaction.rollback()\r\n return response\r\n\r\n user.email = pec.new_email\r\n user.save()\r\n pec.delete()\r\n # And send it to the new email...\r\n try:\r\n user.email_user(subject, message, settings.DEFAULT_FROM_EMAIL)\r\n except Exception:\r\n log.warning('Unable to send confirmation email to new address', exc_info=True)\r\n response = render_to_response(\"email_change_failed.html\", {'email': pec.new_email})\r\n transaction.rollback()\r\n return response\r\n\r\n response = render_to_response(\"email_change_successful.html\", address_context)\r\n transaction.commit()\r\n return response\r\n except Exception:\r\n # If we get an unexpected exception, be sure to rollback the transaction\r\n transaction.rollback()\r\n raise","def test_resend_activation_email(self):\n\n data = {\n 'email': self.user.email,\n }\n\n response = self.client.post(\n reverse('user-resend-activation-email'),\n data,\n format='json',\n )\n\n self.assertEqual(\n response.status_code,\n status.HTTP_200_OK,\n response.content\n )\n\n self.assertEqual(\n response.content,\n b'',\n )\n\n self.assertEqual(len(mail.outbox), 1)","def unsubscribe(self, namespace, unsub_strings=None):\n req = JSONRPCRequest('unsubscribe', [namespace, unsub_strings])\n result = yield self._send(req)\n self._cache_jsonrpc_request(req)\n raise tornado.gen.Return(result)","def unsend_scheduled_messages_after(self, time_cutoff):\n for user_id in self.user_id_to_scheduled_message_ts:\n for scheduled_ts in list(self.user_id_to_scheduled_message_ts[user_id]):\n if scheduled_ts >= time_cutoff:\n # The below if statement is likley redundant\n if scheduled_ts in self.user_id_to_scheduled_message_ts[user_id]:\n self.user_id_to_scheduled_message_ts[user_id].remove(scheduled_ts)","async def delete_email_address(self, ctx):\n author = ctx.message.author\n\n if not self.email_list:\n with open(\"data/email/emails.json\", \"r\", encoding='utf-8') as file:\n self.email_list = json.load(file)\n\n if str(author.id) in self.email_list:\n del self.email_list[str(author.id)]\n with open(\"data/email/emails.json\", \"w\", encoding='utf-8') as file:\n json.dump(self.email_list, file)\n await ctx.send(\"Email address has been deleted successfully..!\")\n else:\n await ctx.send(\"There is no email address configured..!\")\n return","def unsubscribe(self, instrument_ids, exchange_id=b''):\n pass","def invalidate_email(self, accountid, verificationUrl = '', emailTemplate = ''):\n auth = 'appkey='+ self._lr_object._get_api_key()+ '&appsecret='+ self._lr_object._get_api_secret() + '&accountid=' + accountid + '&emailTemplate='+emailTemplate+'&verificationUrl='+verificationUrl\n payload = {}\n url = SECURE_API_URL + \"raas/v1/account/invalidateemail\" + \"?\" + auth\n return self._lr_object._put_json(url, payload)","async def test_account_admin_removed(app, session, stan_server, event_loop, client_id, events_stan, future):\n # Call back for the subscription\n from account_mailer.worker import cb_subscription_handler\n\n # vars\n user = factory_user_model_with_contact()\n org = factory_org_model()\n factory_membership_model(user.id, org.id)\n id = org.id\n\n events_subject = 'test_subject'\n events_queue = 'test_queue'\n events_durable_name = 'test_durable'\n with patch.object(notification_service, 'send_email', return_value=None) as mock_send:\n # register the handler to test it\n await subscribe_to_queue(events_stan,\n events_subject,\n events_queue,\n events_durable_name,\n cb_subscription_handler)\n\n # add an event to queue\n email = 'foo@testbar.com'\n mail_details = {\n 'accountId': id,\n 'recipientEmail': email\n }\n await helper_add_event_to_queue(events_stan, events_subject, org_id=id,\n msg_type=MessageType.ADMIN_REMOVED.value, mail_details=mail_details)\n\n mock_send.assert_called\n assert mock_send.call_args.args[0].get('recipients') == email\n assert mock_send.call_args.args[0].get('content').get('subject') == SubjectType.ADMIN_REMOVED_SUBJECT.value\n assert mock_send.call_args.args[0].get('attachments') is None\n assert mock_send.call_args.args[0].get('content').get('body') is not None","def test_cancelling_not_full_event_emails_waiting_list(self, mock_tz):\n mock_tz.now.return_value = datetime(\n 2015, 2, 13, 17, 15, tzinfo=dt_timezone.utc\n )\n\n # make full event (setup has one paid and one unpaid)\n # cancellation period =1, date = 2015, 2, 13, 18, 0\n self.event.max_participants = 3\n self.event.save()\n\n # make some waiting list users\n for i in range(3):\n baker.make_recipe(\n 'booking.waiting_list_user', event=self.event,\n user__email='test{}@test.com'.format(i)\n )\n\n management.call_command('cancel_unpaid_bookings')\n # emails are sent to user per cancelled booking (1) studio (1) and waitinglist (1, with bcc)\n self.assertEqual(len(mail.outbox), 3)","def unsubscribe(cls,sender,receiver):\n cls._unsubscribe(id(sender),receiver)"],"string":"[\n \"def unsubscribe_all_emails_from_cloudwatch(self):\\n for subscription in self._get_cloudwatch_subscriptions():\\n subscription_arn = subscription['SubscriptionArn']\\n if not subscription_arn == 'PendingConfirmation':\\n self.conn.unsubscribe(subscription_arn)\",\n \"def unsubscribe():\\n from alot.helper import mailto_to_envelope\\n from alot.buffers import EnvelopeBuffer\\n msg = ui.current_buffer.get_selected_message()\\n e = msg.get_email()\\n uheader = e['List-Unsubscribe']\\n dtheader = e.get('Delivered-To', None)\\n\\n if uheader is not None:\\n M = re.search(r'<(mailto:\\\\S*)>', uheader)\\n if M is not None:\\n env = mailto_to_envelope(M.group(1))\\n if dtheader is not None:\\n env['From'] = [dtheader]\\n ui.buffer_open(EnvelopeBuffer(ui, env))\\n else:\\n ui.notify('focussed mail contains no \\\\'List-Unsubscribe\\\\' header',\\n 'error')\",\n \"def _unsubscribe(self):\\n self.unsubscribe_date = now()\\n self.unsubscribed = True\\n self.subscribed = False\",\n \"def unsubscribe(self, request):\\n email = self.cleaned_data.get('email')\\n subscriber = Subscriber.objects.get(email=email, mailing_list=self.mailing_list)\\n subscriber.unsubscribe(request)\",\n \"async def aoc_unsubscribe(self, ctx: commands.Context) -> None:\\n role = ctx.guild.get_role(AocConfig.role_id)\\n\\n if role in ctx.author.roles:\\n await ctx.author.remove_roles(role)\\n await ctx.send(\\\"Okay! You have been __unsubscribed__ from notifications about new Advent of Code tasks.\\\")\\n else:\\n await ctx.send(\\\"Hey, you don't even get any notifications about new Advent of Code tasks currently anyway.\\\")\",\n \"def unsubscribe(receiver):\",\n \"def unsubscribe(receiver):\",\n \"def unsubscribe(receiver):\",\n \"def unsubscribe(receiver):\",\n \"def unsubscribe(receiver):\",\n \"async def aoc_unsubscribe(self, ctx: commands.Context) -> None:\\n if ctx.channel.id != settings.aoc.channel_id:\\n await ctx.send(f\\\"Please use the <#{settings.aoc.channel_id}> channel\\\")\\n return\\n\\n role = ctx.guild.get_role(settings.aoc.role_id)\\n\\n if role in ctx.author.roles:\\n await ctx.author.remove_roles(role)\\n await ctx.send(\\\"Okay! You have been __unsubscribed__ from notifications about new Advent of Code tasks.\\\")\\n else:\\n await ctx.send(\\\"Hey, you don't even get any notifications about new Advent of Code tasks currently anyway.\\\")\",\n \"def unsubscribe(self, subject):\\n pass\",\n \"def unsubscribe(self, email, id):\\n self.db.commit([\\n 'UPDATE comments SET',\\n ' notification=0',\\n 'WHERE email=%s AND (id=%s OR parent=%s);'], (email, id, id))\",\n \"def unsubscribe(self, feed, **args):\\n args.update(feed=feed)\\n return self.fetch(\\\"/unsubscribe\\\", post_args=args)\",\n \"def Promote_from_waitlist(self, email):\\n if email in self.Waitlist:\\n self.Waitlist.remove(email)\\n self.Add_attendee(email)\\n else:\\n raise MissingAddressException(email)\",\n \"def test_reactivation_for_unregistered_user(self, email_user):\\r\\n response_data = self.reactivation_email(self.unregisteredUser)\\r\\n\\r\\n self.assertFalse(response_data['success'])\",\n \"def unsubscribe(request):\\n user = None\\n if request.user.is_authenticated:\\n user = request.user\\n elif \\\"email\\\" in request.GET:\\n try:\\n notification = Notification.objects.get(pk=request.GET[\\\"email\\\"])\\n user = User.objects.get(email=notification.email)\\n except (Notification.DoesNotExist, User.DoesNotExist):\\n pass\\n\\n if not user:\\n return redirect(\\\"/accounts/login/?next=/accounts/profile/unsubscribe/\\\")\\n\\n subscriptions = user.subscriptions.filter(active=True).order_by(\\\"-created_at\\\")\\n if request.method == \\\"POST\\\":\\n count = subscriptions.update(active=False)\\n messages.success(request, f\\\"Successfully deactivated {count} subscriptions.\\\")\\n return redirect(\\\"/accounts/profile/\\\")\\n else:\\n return render(\\n request, \\\"account/unsubscribe.html\\\", {\\\"subscriptions\\\": subscriptions}\\n )\",\n \"def unsubscribe(self):\\n pass # pragma: no cover\",\n \"def pop():\\n task = connection.zrange(QUEUE_KEY, 0, 0)\\n if not task:\\n return False, 'No emails now!'\\n msg_id = task[0]\\n timestamp = connection.zscore(QUEUE_KEY, msg_id)\\n now = datetime.datetime.now().timestamp()\\n if timestamp < now or abs(timestamp - now) <= 1e-6:\\n message = connection.get(msg_id)\\n pipeline = connection.pipeline()\\n pipeline.zrem(QUEUE_KEY, msg_id)\\n pipeline.delete(msg_id)\\n pipeline.execute()\\n return True, message\\n return False, \\\"It's too early now!\\\"\",\n \"def unsubscribe( self, mess, args):\\n user = self.get_sender_username(mess)\\n if not user in self.users:\\n return 'You are not subscribed!'\\n else:\\n user = self.users.pop(user)\\n self.message_queue.append('_%s has left the channel_' % user)\\n self.log.info( '%s unsubscribed from the broadcast.' % user)\\n self.save_state()\\n return 'You are now unsubscribed.'\",\n \"def _unsubscribe(self):\\n if not self.subscribed:\\n return False\\n return {}\",\n \"def unsubscribe(self, update, context):\\n # remove or update to the sqlite table.\\n chat = update.message.chat\\n self.db_manager.remove_user(chat.id)\\n self.logger.info(\\n 'Username: %s and chat_id: %s unsubscribed to the list.' % (chat.username, chat.id)\\n )\\n update.message.reply_text('You have successfully unsubscribed the notifications forever.')\",\n \"def _async_unsubscribe(self, topic: str) -> None:\\n if self._is_active_subscription(topic):\\n if self._max_qos[topic] == 0:\\n return\\n subs = self._matching_subscriptions(topic)\\n self._max_qos[topic] = max(sub.qos for sub in subs)\\n # Other subscriptions on topic remaining - don't unsubscribe.\\n return\\n if topic in self._max_qos:\\n del self._max_qos[topic]\\n if topic in self._pending_subscriptions:\\n # Avoid any pending subscription to be executed\\n del self._pending_subscriptions[topic]\\n\\n self._pending_unsubscribes.add(topic)\\n self._unsubscribe_debouncer.async_schedule()\",\n \"def cb_stop(self, update, context):\\n\\n print(f\\\"Unsubscribing chat_id '{update.message.chat_id}'\\\")\\n try:\\n self.clientChatIds.remove(update.message.chat_id)\\n answer = \\\"You sucessfully unsubscribed.\\\"\\n self.saveToFile(self.configFile)\\n except KeyError:\\n answer = \\\"You are not subscribed.\\\"\\n\\n update.message.reply_text(answer)\",\n \"def opt_out_email_updates(request, token, course_id):\\n try:\\n username = UsernameCipher().decrypt(token).decode(\\\"utf-8\\\")\\n user = User.objects.get(username=username)\\n course_key = CourseKey.from_string(course_id)\\n course = get_course_by_id(course_key, depth=0)\\n except UnicodeDecodeError:\\n raise Http404(\\\"base64url\\\") # lint-amnesty, pylint: disable=raise-missing-from\\n except UsernameDecryptionException as exn:\\n raise Http404(str(exn)) # lint-amnesty, pylint: disable=raise-missing-from\\n except User.DoesNotExist:\\n raise Http404(\\\"username\\\") # lint-amnesty, pylint: disable=raise-missing-from\\n except InvalidKeyError:\\n raise Http404(\\\"course\\\") # lint-amnesty, pylint: disable=raise-missing-from\\n\\n unsub_check = request.POST.get('unsubscribe', False)\\n context = {\\n 'course': course,\\n 'unsubscribe': unsub_check\\n }\\n\\n if request.method == 'GET':\\n return render_to_response('bulk_email/confirm_unsubscribe.html', context)\\n\\n if request.method == 'POST' and unsub_check:\\n Optout.objects.get_or_create(user=user, course_id=course_key)\\n log.info(\\n \\\"User %s (%s) opted out of receiving emails from course %s\\\",\\n user.username,\\n user.email,\\n course_id,\\n )\\n\\n return render_to_response('bulk_email/unsubscribe_success.html', context)\",\n \"def on_email_confirmed(request, email_address: EmailAddress, **kwargs):\\n dillo.tasks.profile.update_mailing_list_subscription(email_address.email, True)\",\n \"def remove_from_mailing_list(self, email):\\n if not self.list_removal_endpoint or not self.list_key:\\n # fail silently, log info\\n print('REMOVE_ON_DELETE is turned on, but required '\\n 'fields are missing.')\\n return\\n\\n try:\\n email = email.encode('utf-8').lower()\\n email_hash = hashlib.md5(email).hexdigest()\\n res = requests.delete(self.list_removal_endpoint.format(email_hash),\\n auth=('nop', self.list_key),\\n timeout=1.5)\\n\\n if res.status_code != 204:\\n print('Unexpected mailing list API response.. '\\n 'status code: {0.status_code}\\\\n'\\n 'content: {0.content}'.format(res))\\n\\n except Exception as e:\\n if e is requests.exceptions.Timeout:\\n print('Mailing list API timed out..')\\n else:\\n print('Removing from mailing list failed:', e)\",\n \"def test_notification_cancellation_email(self):\\n api.content.transition(obj=self.event, transition='cancel')\\n mailhost = api.portal.get_tool('MailHost')\\n self.assertEqual(len(mailhost.messages), 2)\\n msg = message_from_string(mailhost.messages[1])\\n\\n self.assertEqual(msg['To'], BOARD_LIST_ADDRESS)\\n self.assertEqual(\\n msg['From'], 'EESTEC International <noreply@eestec.net>')\\n self.assertEqual(\\n msg['Subject'],\\n '=?utf-8?q?=5BEVENTS=5D=5BCancelled=5D_T=C3=A9st_event?=',\\n )\\n self.assertIn('an Event has been cancelled', msg.get_payload())\\n self.assertIn('http://nohost/plone/lc/test-event', msg.get_payload())\",\n \"def unsubscribe(self, destination, *args, **kwargs):\",\n \"def unsubscribe(self, item_name):\\n self.subscribed = None\",\n \"def unsubscribe_by_mail(self):\\n qry = ServiceOperationQuery(self, \\\"unsubscribeByMail\\\")\\n self.context.add_query(qry)\\n return self\",\n \"def test_email_after_contest_end(self):\\n self.prep_consumer()\\n temp_date = settings.CONTEST_END_DATE\\n settings.CONTEST_END_DATE = str(\\n datetime.today().date() - timedelta(days=1))\\n UnqualifiedConsumerEmailTask().run(test_mode=self.consumer)\\n log = get_last_db_log(\\n 'email_gateway.tasks.send_unqualified_emails', 'EMAIL')\\n if log:\\n self.fail('Performed task even though contest ended.')\\n settings.CONTEST_END_DATE = temp_date\",\n \"def unsubscribe(self):\\n res = self._unsubscribe()\\n if res is not None:\\n self._subscribed = False\\n return res\",\n \"def delete_email_from_blacklist(self, email):\\n logger.info(\\\"Function call: delete_email_from_blacklist for '{}'\\\".format(email, ))\\n return self.__handle_error('Empty email') if not email else self.__handle_result(self.__send_request('blacklist', 'DELETE', {'emails': base64.b64encode(email)}))\",\n \"def unsubscribe(self, feedUrl):\\r\\n response = self.httpPost(\\r\\n ReaderUrl.SUBSCRIPTION_EDIT_URL,\\r\\n {'ac':'unsubscribe', 's': feedUrl})\\r\\n # FIXME - need better return API\\r\\n if response and 'OK' in response:\\r\\n return True\\r\\n else:\\r\\n return False\",\n \"def unsubscribe(self, sub_code):\\n\\t\\tresult = {}\\n\\t\\tconnection = DbHelper.connect()\\n\\t\\tsub_id = 0\\n\\n\\t\\ttry:\\n\\t\\t\\twith connection.cursor() as cursor:\\n\\t\\t\\t\\tsql = \\\"SELECT * FROM mail_list \\\\\\n\\t\\t\\t\\t\\t WHERE email_hash=%s;\\\"\\n\\t\\t\\t\\tcursor.execute(sql, [sub_code])\\n\\t\\t\\t\\tresult = cursor.fetchone()\\n\\t\\t\\t\\t\\n\\t\\t\\t\\tif not result:\\n\\t\\t\\t\\t\\tconnection.close()\\n\\t\\t\\t\\t\\treturn \\\"CODE_DOES_NOT_EXIST\\\"\\n\\t\\t\\t\\telif not result['is_activated']:\\n\\t\\t\\t\\t\\tconnection.close()\\n\\t\\t\\t\\t\\treturn \\\"CODE_NOT_ACTIVATED\\\"\\n\\n\\t\\t\\t\\tsub_id = result['sub_id']\\n\\n\\t\\t\\t\\tsql = \\\"DELETE FROM mail_list \\\\\\n\\t\\t\\t\\t\\t WHERE sub_id=%s;\\\"\\n\\t\\t\\t\\tcursor.execute(sql, [sub_id])\\n\\t\\t\\t\\tconnection.commit()\\n\\t\\t\\t\\tconnection.close()\\n\\t\\t\\t\\treturn \\\"UNSUBSCRIBED\\\"\\n\\t\\texcept pymysql.MySQLError as e:\\n\\t\\t\\tconnection.close()\\n\\t\\t\\treturn \\\"DATABASE_ERROR\\\"\",\n \"def notify_email_confirmed(self, user, email):\\n \\n # make sure user isn't still invited to groups he owns or is a member of\\n for g in self.users_groups(user):\\n g.remove_invitation(user)\",\n \"def unsubscribe(endpoint: str, topic: str, timeout: int = 5):\\n global logger\\n logger.info(f\\\"Unsubscribing from topic '{topic}' ...\\\")\\n action = {\\\"action\\\": \\\"unsubscribe\\\", \\\"topic\\\": topic}\\n reply = send_manage_message(endpoint, action, timeout)\\n if not reply_is_success(reply):\\n logger.warning(\\\"Unsubscription failed\\\")\\n return\\n logger.info(\\\"Unsubscription successful\\\")\",\n \"def test_unsubscribe(self):\\n dest = '/topic/dest'\\n\\n self.tm.subscribe(self.conn, dest)\\n f = Frame(frames.MESSAGE, headers={'destination': dest}, body='Empty')\\n self.tm.send(f)\\n\\n self.assertEqual(len(self.conn.frames), 1)\\n subscription = self.conn.frames[0].headers.pop(\\\"subscription\\\", None)\\n self.assertEqual(subscription, 0)\\n self.assertEqual(self.conn.frames[0], f)\\n\\n self.tm.unsubscribe(self.conn, dest)\\n f = Frame(frames.MESSAGE, headers={'destination': dest}, body='Empty')\\n self.tm.send(f)\\n\\n self.assertEqual(len(self.conn.frames), 1)\",\n \"def on_unsubscribe(self, mqtt_client, userdata, mid ):\\n logging.debug(\\\"DEBUG - unsubscribe ack received\\\")\",\n \"def test_unsubscribe(self):\\n self.service.clientConnected()\\n\\n unsubscribers = []\\n self.service.subscribe(u'url', 1\\n ).addCallback(lambda fn: unsubscribers.append(fn))\\n self.service.subscribe(u'url', 2\\n ).addCallback(lambda fn: unsubscribers.append(fn))\\n\\n pubsubClient = self.service.pubsubClient\\n self.assertIn(u'url', pubsubClient.subscriptions)\\n\\n unsubscribers.pop()()\\n self.service.unsubscribe(u'url')\\n self.assertIn(u'url', pubsubClient.subscriptions)\\n\\n unsubscribers.pop()()\\n self.service.unsubscribe(u'url')\\n self.assertNotIn(u'url', pubsubClient.subscriptions)\",\n \"def _unsubscribe(self, signal):\\n while signal in self._downstream:\\n self._downstream.remove(signal)\\n while signal in self._downstream_reconnect:\\n self._downstream_reconnect.remove(signal)\",\n \"async def unsubscribe(self):\\n LOGGER.info('Subscription removed')\\n await self._ros.send(self._unsubscribe_msg)\",\n \"def test_cancelling_more_than_once_only_emails_once(self, mock_tz):\\n mock_tz.now.return_value = datetime(2015, 2, 13, 17, 15, tzinfo=dt_timezone.utc)\\n\\n # make full event (setup has one paid and one unpaid)\\n # cancellation period =1, date = 2015, 2, 13, 18, 0\\n self.event.max_participants = 3\\n self.event.save()\\n\\n # make another booking that will be cancelled\\n baker.make_recipe(\\n 'booking.booking', event=self.event, paid=False,\\n payment_confirmed=False, status='OPEN',\\n user__email=\\\"unpaid@test.com\\\",\\n date_booked=datetime(2015, 2, 9, 18, 0, tzinfo=dt_timezone.utc),\\n warning_sent=True,\\n date_warning_sent=datetime(2015, 2, 9, 20, 0, tzinfo=dt_timezone.utc),\\n )\\n\\n # make some waiting list users\\n for i in range(3):\\n baker.make_recipe(\\n 'booking.waiting_list_user', event=self.event,\\n user__email='test{}@test.com'.format(i)\\n )\\n\\n management.call_command('cancel_unpaid_bookings')\\n # emails are sent to user per cancelled booking (2) and\\n # one email with bcc to waiting list (1) and studio (1)\\n # waiting list email sent after the two user emails\\n self.assertEqual(len(mail.outbox), 4)\\n self.assertEqual(\\n sorted(mail.outbox[2].bcc),\\n ['test0@test.com', 'test1@test.com', 'test2@test.com']\\n )\\n for email in [mail.outbox[0], mail.outbox[1], mail.outbox[3]]:\\n self.assertEqual(email.bcc, [])\\n\\n self.assertEqual(Booking.objects.filter(status='CANCELLED').count(), 2)\",\n \"def rpc_campaign_alerts_unsubscribe(self, campaign_id):\\n\\t\\tusername = self.basic_auth_user\\n\\t\\tsession = db_manager.Session()\\n\\t\\tquery = session.query(db_models.AlertSubscription)\\n\\t\\tquery = query.filter_by(campaign_id=campaign_id, user_id=username)\\n\\t\\tsubscription = query.first()\\n\\t\\tif subscription:\\n\\t\\t\\tsession.delete(subscription)\\n\\t\\t\\tsession.commit()\\n\\t\\tsession.close()\\n\\t\\treturn\",\n \"async def unsubscribe(self, topic: str, subscription_id: int = None) -> None:\\n ...\",\n \"def unsubscribe(self, path):\\n\\n box = self._getMailbox(path)\\n\\n box.metadata['subscribed'] = False\\n\\n box.saveMetadata()\\n\\n return True\",\n \"async def _async_perform_unsubscribes(self) -> None:\\n if not self._pending_unsubscribes:\\n return\\n\\n topics = list(self._pending_unsubscribes)\\n self._pending_unsubscribes = set()\\n\\n async with self._paho_lock:\\n result, mid = await self.hass.async_add_executor_job(\\n self._mqttc.unsubscribe, topics\\n )\\n _raise_on_error(result)\\n for topic in topics:\\n _LOGGER.debug(\\\"Unsubscribing from %s, mid: %s\\\", topic, mid)\\n\\n await self._wait_for_mid(mid)\",\n \"def unsubscribe_user(sub_id):\\n repository.unsubscribe(sub_id)\\n return flask.render_template('www/email/unsubscribed.html')\",\n \"def confirm_email(self):\\n # The base class' implementation does nothing\\n pass\",\n \"def test_unsubscribe_offer(self):\\n pass\",\n \"def unsubscribe(request, action):\\n data = {}\\n if action and action.lower() == 'updates':\\n user_profile = request.user.revolvuserprofile\\n user_profile.subscribed_to_repayment_notifications = False\\n user_profile.save()\\n data = {'msg': \\\"You have successfully unsubscribed\\\"}\\n elif action and action.lower() == 'confirmation':\\n return render_to_response('base/unsub_conf.html', context_instance=RequestContext(request, data))\\n else:\\n data = {'msg': 'Please specify which section you want to unsubscribe'}\\n\\n return render_to_response('base/minimal_message.html',\\n context_instance=RequestContext(request, data))\",\n \"def view_mailinglist_unsubscribe(request, slug, uidb36, token):\\n newsletter = get_object_or_404(Newsletter, slug=slug)\\n contact = untokenize(uidb36, token)\\n\\n already_unsubscribed = contact in newsletter.mailing_list.unsubscribers.all()\\n\\n if request.POST.get('email') and not already_unsubscribed:\\n newsletter.mailing_list.unsubscribers.add(contact)\\n newsletter.mailing_list.save()\\n already_unsubscribed = True\\n ContactMailingStatus.objects.create(newsletter=newsletter, contact=contact,\\n status=ContactMailingStatus.UNSUBSCRIPTION)\\n\\n return render_to_response('newsletter/mailing_list_unsubscribe.html',\\n {'email': contact.email,\\n 'already_unsubscribed': already_unsubscribed},\\n context_instance=RequestContext(request))\",\n \"def _logout(self):\\n if self.connected:\\n try:\\n self._smtp.quit()\\n except smtplib.SMTPServerDisconnected:\\n pass\\n self._connected = False\",\n \"def unsubscribe(self):\\n if self._subscribed and self._connected:\\n try:\\n msg = self._create_message(strings.UNSUB_MSG)\\n self.write(msg)\\n except (OSError, KeyError) as ex:\\n _LOGGER.error(\\n \\\"PyISY encountered a socket error while writing unsubscribe message to the socket: %s.\\\",\\n ex,\\n )\\n self._subscribed = False\\n self.disconnect()\",\n \"def handle_email_confirmed(sender, **kwargs):\\n email = kwargs['email_address']\\n email.user.userprofile.member.cast().confirm_email()\",\n \"def unsubscribe_all(self):\\n ids = []\\n repeat = True\\n cursor = None\\n # get all ids\\n while repeat:\\n ret = self.__twitch.get_eventsub_subscriptions(after=cursor)\\n for d in ret.get('data', []):\\n ids.append(d.get('id'))\\n cursor = ret.get('pagination', {}).get('cursor')\\n repeat = cursor is not None\\n for _id in ids:\\n succ = self.__twitch.delete_eventsub_subscription(_id)\\n if not succ:\\n self.__logger.warning(f'failed to unsubscribe from event {_id}')\\n self.__callbacks.clear()\",\n \"def _handle_unsubscribe_request(state, client, request):\\n # TODO: create me\",\n \"def email_subscribe_pending_confirm(hexdomain):\\n domain = tools.parse_domain(hexdomain)\\n if domain is None:\\n flask.abort(400, 'Malformed domain or domain not represented in hexadecimal format.')\\n\\n email_address = flask.request.form['email_address']\\n\\n if email_address.strip() == '':\\n return flask.redirect('/email/subscribe/{}/0'.format(hexdomain))\\n\\n verify_code = tools.random_id()\\n verify_url = flask.request.url_root + 'email/verify/{}'.format(verify_code)\\n email_body = email_tools.render_email(\\n 'confirm.html',\\n domain=domain,\\n verify_url=verify_url\\n )\\n\\n repository.propose_subscription(verify_code, email_address, domain)\\n emailer.send(\\n email_address, 'Please verify your subscription', email_body\\n )\\n\\n return flask.render_template('www/email/pending_verify.html', domain=domain)\",\n \"def unenroll_email(course_id, student_email, email_students=False, email_params=None):\\r\\n previous_state = EmailEnrollmentState(course_id, student_email)\\r\\n\\r\\n if previous_state.enrollment:\\r\\n CourseEnrollment.unenroll_by_email(student_email, course_id)\\r\\n if email_students:\\r\\n email_params['message'] = 'enrolled_unenroll'\\r\\n email_params['email_address'] = student_email\\r\\n email_params['full_name'] = previous_state.full_name\\r\\n send_mail_to_student(student_email, email_params)\\r\\n\\r\\n if previous_state.allowed:\\r\\n CourseEnrollmentAllowed.objects.get(course_id=course_id, email=student_email).delete()\\r\\n if email_students:\\r\\n email_params['message'] = 'allowed_unenroll'\\r\\n email_params['email_address'] = student_email\\r\\n # Since no User object exists for this student there is no \\\"full_name\\\" available.\\r\\n send_mail_to_student(student_email, email_params)\\r\\n\\r\\n after_state = EmailEnrollmentState(course_id, student_email)\\r\\n\\r\\n return previous_state, after_state\",\n \"async def async_unsubscribe_services(self) -> None:\\n # Delete list of subscriptions and cancel renewal before unsubcribing\\n # to avoid unsub-resub race.\\n sids = list(self._subscriptions)\\n self._subscriptions.clear()\\n await self._update_resubscriber_task()\\n\\n await asyncio.gather(*(self._async_unsubscribe_service(sid) for sid in sids))\",\n \"def test_unverified_subscriber(self):\\n self.prep_consumer()\\n subscriber = Subscriber.objects.get(id=7)\\n self.consumer.subscriber = subscriber\\n self.consumer.save()\\n UnqualifiedConsumerEmailTask().run(test_mode=self.consumer)\\n self.common_asserts()\\n self.assertTrue('71010' in mail.outbox[0].alternatives[0][0])\\n self.assertTrue('71010' in mail.outbox[0].body)\",\n \"async def async_will_remove_from_hass(self) -> None:\\n async_unsubscribe_topics(self.hass, self._sub_state)\\n self._sub_state = None\",\n \"def onUnsubscribed(self, connection:MQTTConnection, topic:str) -> bool:\\n\\t\\tconnection.subscribedCount -= 1\\n\\t\\treturn True\",\n \"def unsubscribe(self, destination, extra_headers=None):\\n unsubscribe = frame.UnsubscribeFrame(destination, extra_headers=extra_headers)\\n res = self.send_frame(unsubscribe)\\n with self.subscription_lock:\\n self.subscribed_destinations.pop(destination)\\n return res\",\n \"def unsubscribe_from_guest_statistics(self):\\n\\t\\treturn Job(SDK.PrlVm_UnsubscribeFromGuestStatistics(self.handle)[0])\",\n \"def test_cancelling_for_full_event_emails_waiting_list(self, mock_tz):\\n mock_tz.now.return_value = datetime(\\n 2015, 2, 13, 17, 15, tzinfo=dt_timezone.utc\\n )\\n\\n # make full event (setup has one paid and one unpaid)\\n # cancellation period =1, date = 2015, 2, 13, 18, 0\\n self.event.max_participants = 2\\n self.event.save()\\n\\n # make some waiting list users\\n for i in range(3):\\n baker.make_recipe(\\n 'booking.waiting_list_user', event=self.event,\\n user__email='test{}@test.com'.format(i)\\n )\\n\\n management.call_command('cancel_unpaid_bookings')\\n # emails are sent to user per cancelled booking (1) and\\n # one email with bcc to waiting list (1) and studio (1)\\n self.assertEqual(len(mail.outbox), 3)\\n self.assertEqual(\\n sorted(mail.outbox[1].bcc),\\n ['test0@test.com', 'test1@test.com', 'test2@test.com']\\n )\",\n \"def unsubscribe(self):\\r\\n self._unregister()\",\n \"def unsubscribe(self):\\n\\n # Unsubscribe\\n self.pyrps.redis.srem(self.pyrps._ns_subscriptions(self.queue), self.consumer_id) \\n\\n # Remove message queue\\n self.pyrps.redis.delete(self.pyrps._ns_queue(self.queue, self.consumer_id))\",\n \"def email_ml_unsubscribe(self, maillist=None, maillist_uid=None, subscriber=None, subscriber_uid=None):\\n if not maillist and not maillist_uid:\\n raise ValueError('Maillist or uid required')\\n if not subscriber and not subscriber_uid:\\n raise ValueError('Subscriber or uid required')\\n return self._request('email/ml/unsubscribe', inspect_args_func(inspect.currentframe()))\",\n \"def test_resend_activation_email_nonunique_email(self):\\n user1 = self.registration_profile.objects.create_inactive_user(\\n site=Site.objects.get_current(), send_email=False, **self.user_info)\\n user2_info = copy(self.user_info)\\n user2_info['username'] = 'bob'\\n user2 = self.registration_profile.objects.create_inactive_user(\\n site=Site.objects.get_current(), send_email=False, **user2_info)\\n self.assertEqual(user1.email, user2.email)\\n self.assertFalse(self.registration_profile.objects.resend_activation_mail(\\n email=self.user_info['email'],\\n site=Site.objects.get_current(),\\n ))\\n self.assertEqual(len(mail.outbox), 0)\",\n \"def unsubscribe(request):\\n address = request.POST.get('address')\\n\\n Subscription.objects.filter(address=address).delete()\\n\\n return HttpResponse(json.dumps({\\n \\\"status\\\": \\\"success\\\"\\n }, default=helpers.json_custom_parser), content_type='application/json')\",\n \"async def _send_message_for_unsubscribe(self, data_id):\\n body = json.dumps(\\n dict(\\n action='unsub',\\n data_id=data_id\\n )\\n ).encode('utf-8')\\n await self._send_message_in_queue(self.queue_crypto_quotes_service, body)\",\n \"def terminate(self):\\n self.mailQueue.put(None)\\n self.mailQueue.join()\\n self.join()\\n self.logger.info(\\\"Mailer terminated\\\")\",\n \"def unsubscribe_all_known(self):\\n for key, value in self.__callbacks.items():\\n self.__logger.debug(f'unsubscribe from event {key}')\\n succ = self.__twitch.delete_eventsub_subscription(key)\\n if not succ:\\n self.__logger.warning(f'failed to unsubscribe from event {key}')\\n self.__callbacks.clear()\",\n \"def test_resend_activation_email(self):\\n user = self.registration_profile.objects.create_inactive_user(\\n site=Site.objects.get_current(), send_email=False, **self.user_info)\\n self.assertEqual(len(mail.outbox), 0)\\n\\n profile = self.registration_profile.objects.get(user=user)\\n orig_activation_key = profile.activation_key\\n\\n self.assertTrue(self.registration_profile.objects.resend_activation_mail(\\n email=self.user_info['email'],\\n site=Site.objects.get_current(),\\n ))\\n\\n profile = self.registration_profile.objects.get(pk=profile.pk)\\n new_activation_key = profile.activation_key\\n\\n self.assertNotEqual(orig_activation_key, new_activation_key)\\n self.assertEqual(len(mail.outbox), 1)\",\n \"def unsubscribe(self, topic):\\n request = protos.RequestUnsubscribe(topic=topic)\\n return self.stub.unsubscribe(request)\",\n \"def stop(self):\\r\\n self._incompleteMsgs = set()\\r\\n\\r\\n if self._cleaner.running:\\r\\n self._cleaner.stop()\",\n \"async def unsubscribe(connection, message):\\n from high_templar.hub import NotSubscribedException\\n\\n if 'requestId' not in message:\\n return await connection.send({\\n 'code': 'error',\\n 'message': 'no-request-id'\\n })\\n\\n for subscription in connection.app.hub.subscriptions[connection]:\\n if subscription.request_id == message['requestId']:\\n try:\\n connection.app.hub.unsubscribe(subscription)\\n await connection.send({ 'code': 'success' })\\n except NotSubscribedException:\\n await connection.send({\\n 'code': 'error',\\n 'message': 'not-subscribed'\\n })\\n break\\n else:\\n return await connection.send({\\n 'code': 'error',\\n 'message': 'not-subscribed'\\n })\",\n \"def remove_user(self, email):\\n return run_transaction(\\n self.sessionfactory,\\n lambda session: remove_user_txn(session, email))\",\n \"def disconnect(self, extra_headers=None):\\n try:\\n if self.connection.connected:\\n with self.subscription_lock:\\n # Need a copy since unsubscribe() removes the destination from the collection.\\n subcpy = copy(self.subscribed_destinations)\\n for destination in subcpy:\\n self.unsubscribe(destination)\\n disconnect = frame.DisconnectFrame(extra_headers=extra_headers)\\n result = self.send_frame(disconnect)\\n try:\\n self.connection.disconnect()\\n except NotConnectedError:\\n pass\\n return result\\n finally:\\n self.shutdown_event.set()\",\n \"def unsubscribe(self, jid=None, username=None, domain=None):\\n if jid is not None:\\n self._pres_manager.flag_offline(jid)\\n elif username is not None and domain is not None:\\n self._pres_manager.flag_offline(xmpp.JID(node=username, domain=domain))\\n self._roster.unsubscribe(jid=jid, username=username, domain=domain)\",\n \"def teardown(self):\\n try:\\n self.loop.run_until_complete(self.webhook_connection.disconnect())\\n except Exception:\\n print_exc()\\n raise\",\n \"def _onUnsubscribe(self, client:mqtt.Client, userdata:Any, mid:int) -> None:\\n\\t\\t# TODO doc, error check when not connected, not subscribed\\n\\t\\tfor t in self.subscribedTopics.values():\\n\\t\\t\\tif t.mid == mid:\\n\\t\\t\\t\\tdel self.subscribedTopics[t.topic]\\n\\t\\t\\t\\tself.messageHandler and self.messageHandler.onUnsubscribed(self, t.topic)\\n\\t\\t\\t\\tbreak\",\n \"def delete_email_from_all_addressooks(self, email):\\n logger.info(\\\"Function call: delete_email_from_all_addressooks for '{}'\\\".format(email, ))\\n return self.__handle_error('Empty email') if not email else self.__handle_result(self.__send_request('emails/{}'.format(email, ), 'DELETE'))\",\n \"def unsubscribe(observer):\",\n \"def unsubscribe(observer):\",\n \"def test_generate_confirmational_email_for_student_that_already_received_email(self):\\n self.student2.has_received_email = True\\n self.student2.save()\\n\\n template = self.email_confirmation_template.name\\n confirm_interview_url = \\\"confirm-interview-test-url\\\"\\n choose_interview_url = \\\"choose-interview-test-url\\\"\\n\\n confirm_email_generator = GenerateConfirmEmails(\\n template, confirm_interview_url, choose_interview_url)\\n\\n confirm_email_generator.generate_confirmation_emails()\\n\\n client = Client()\\n client.login(\\n email=self.teacher_admin.email,\\n password='123'\\n )\\n\\n url = reverse('admin:post_office_email_changelist')\\n response = client.get(url, follow=True)\\n\\n result_list = response.context_data['cl'].result_list\\n\\n # Two students dont have interviews and the third already received email\\n # There should be no emails generated in Email change list\\n self.assertEqual(len(result_list), 0)\",\n \"def unsubscribe(self, user_token, topic):\\n response = _request('DELETE',\\n url=self.url_v1('/user/subscriptions/' + topic),\\n user_agent=self.user_agent,\\n user_token=user_token,\\n )\\n _raise_for_status(response)\",\n \"def unsubscribe_feed():\\n\\n source_id = request.args.get('source_id', type=int)\\n if source_id:\\n current_user.unsubscribe_feed(source_id)\\n flash('Unsubscribe successful source=%s' % source_id)\\n\\n return redirect(url_for('mod_feed.manage_subscriptions'))\",\n \"def unconfirm(self):\\n self.automatically_detected=True\\n self.save()\",\n \"def confirm_email_change(request, key):\\r\\n try:\\r\\n try:\\r\\n pec = PendingEmailChange.objects.get(activation_key=key)\\r\\n except PendingEmailChange.DoesNotExist:\\r\\n response = render_to_response(\\\"invalid_email_key.html\\\", {})\\r\\n transaction.rollback()\\r\\n return response\\r\\n\\r\\n user = pec.user\\r\\n address_context = {\\r\\n 'old_email': user.email,\\r\\n 'new_email': pec.new_email\\r\\n }\\r\\n\\r\\n if len(User.objects.filter(email=pec.new_email)) != 0:\\r\\n response = render_to_response(\\\"email_exists.html\\\", {})\\r\\n transaction.rollback()\\r\\n return response\\r\\n\\r\\n subject = render_to_string('emails/email_change_subject.txt', address_context)\\r\\n subject = ''.join(subject.splitlines())\\r\\n message = render_to_string('emails/confirm_email_change.txt', address_context)\\r\\n up = UserProfile.objects.get(user=user)\\r\\n meta = up.get_meta()\\r\\n if 'old_emails' not in meta:\\r\\n meta['old_emails'] = []\\r\\n meta['old_emails'].append([user.email, datetime.datetime.now(UTC).isoformat()])\\r\\n up.set_meta(meta)\\r\\n up.save()\\r\\n # Send it to the old email...\\r\\n try:\\r\\n user.email_user(subject, message, settings.DEFAULT_FROM_EMAIL)\\r\\n except Exception:\\r\\n log.warning('Unable to send confirmation email to old address', exc_info=True)\\r\\n response = render_to_response(\\\"email_change_failed.html\\\", {'email': user.email})\\r\\n transaction.rollback()\\r\\n return response\\r\\n\\r\\n user.email = pec.new_email\\r\\n user.save()\\r\\n pec.delete()\\r\\n # And send it to the new email...\\r\\n try:\\r\\n user.email_user(subject, message, settings.DEFAULT_FROM_EMAIL)\\r\\n except Exception:\\r\\n log.warning('Unable to send confirmation email to new address', exc_info=True)\\r\\n response = render_to_response(\\\"email_change_failed.html\\\", {'email': pec.new_email})\\r\\n transaction.rollback()\\r\\n return response\\r\\n\\r\\n response = render_to_response(\\\"email_change_successful.html\\\", address_context)\\r\\n transaction.commit()\\r\\n return response\\r\\n except Exception:\\r\\n # If we get an unexpected exception, be sure to rollback the transaction\\r\\n transaction.rollback()\\r\\n raise\",\n \"def test_resend_activation_email(self):\\n\\n data = {\\n 'email': self.user.email,\\n }\\n\\n response = self.client.post(\\n reverse('user-resend-activation-email'),\\n data,\\n format='json',\\n )\\n\\n self.assertEqual(\\n response.status_code,\\n status.HTTP_200_OK,\\n response.content\\n )\\n\\n self.assertEqual(\\n response.content,\\n b'',\\n )\\n\\n self.assertEqual(len(mail.outbox), 1)\",\n \"def unsubscribe(self, namespace, unsub_strings=None):\\n req = JSONRPCRequest('unsubscribe', [namespace, unsub_strings])\\n result = yield self._send(req)\\n self._cache_jsonrpc_request(req)\\n raise tornado.gen.Return(result)\",\n \"def unsend_scheduled_messages_after(self, time_cutoff):\\n for user_id in self.user_id_to_scheduled_message_ts:\\n for scheduled_ts in list(self.user_id_to_scheduled_message_ts[user_id]):\\n if scheduled_ts >= time_cutoff:\\n # The below if statement is likley redundant\\n if scheduled_ts in self.user_id_to_scheduled_message_ts[user_id]:\\n self.user_id_to_scheduled_message_ts[user_id].remove(scheduled_ts)\",\n \"async def delete_email_address(self, ctx):\\n author = ctx.message.author\\n\\n if not self.email_list:\\n with open(\\\"data/email/emails.json\\\", \\\"r\\\", encoding='utf-8') as file:\\n self.email_list = json.load(file)\\n\\n if str(author.id) in self.email_list:\\n del self.email_list[str(author.id)]\\n with open(\\\"data/email/emails.json\\\", \\\"w\\\", encoding='utf-8') as file:\\n json.dump(self.email_list, file)\\n await ctx.send(\\\"Email address has been deleted successfully..!\\\")\\n else:\\n await ctx.send(\\\"There is no email address configured..!\\\")\\n return\",\n \"def unsubscribe(self, instrument_ids, exchange_id=b''):\\n pass\",\n \"def invalidate_email(self, accountid, verificationUrl = '', emailTemplate = ''):\\n auth = 'appkey='+ self._lr_object._get_api_key()+ '&appsecret='+ self._lr_object._get_api_secret() + '&accountid=' + accountid + '&emailTemplate='+emailTemplate+'&verificationUrl='+verificationUrl\\n payload = {}\\n url = SECURE_API_URL + \\\"raas/v1/account/invalidateemail\\\" + \\\"?\\\" + auth\\n return self._lr_object._put_json(url, payload)\",\n \"async def test_account_admin_removed(app, session, stan_server, event_loop, client_id, events_stan, future):\\n # Call back for the subscription\\n from account_mailer.worker import cb_subscription_handler\\n\\n # vars\\n user = factory_user_model_with_contact()\\n org = factory_org_model()\\n factory_membership_model(user.id, org.id)\\n id = org.id\\n\\n events_subject = 'test_subject'\\n events_queue = 'test_queue'\\n events_durable_name = 'test_durable'\\n with patch.object(notification_service, 'send_email', return_value=None) as mock_send:\\n # register the handler to test it\\n await subscribe_to_queue(events_stan,\\n events_subject,\\n events_queue,\\n events_durable_name,\\n cb_subscription_handler)\\n\\n # add an event to queue\\n email = 'foo@testbar.com'\\n mail_details = {\\n 'accountId': id,\\n 'recipientEmail': email\\n }\\n await helper_add_event_to_queue(events_stan, events_subject, org_id=id,\\n msg_type=MessageType.ADMIN_REMOVED.value, mail_details=mail_details)\\n\\n mock_send.assert_called\\n assert mock_send.call_args.args[0].get('recipients') == email\\n assert mock_send.call_args.args[0].get('content').get('subject') == SubjectType.ADMIN_REMOVED_SUBJECT.value\\n assert mock_send.call_args.args[0].get('attachments') is None\\n assert mock_send.call_args.args[0].get('content').get('body') is not None\",\n \"def test_cancelling_not_full_event_emails_waiting_list(self, mock_tz):\\n mock_tz.now.return_value = datetime(\\n 2015, 2, 13, 17, 15, tzinfo=dt_timezone.utc\\n )\\n\\n # make full event (setup has one paid and one unpaid)\\n # cancellation period =1, date = 2015, 2, 13, 18, 0\\n self.event.max_participants = 3\\n self.event.save()\\n\\n # make some waiting list users\\n for i in range(3):\\n baker.make_recipe(\\n 'booking.waiting_list_user', event=self.event,\\n user__email='test{}@test.com'.format(i)\\n )\\n\\n management.call_command('cancel_unpaid_bookings')\\n # emails are sent to user per cancelled booking (1) studio (1) and waitinglist (1, with bcc)\\n self.assertEqual(len(mail.outbox), 3)\",\n \"def unsubscribe(cls,sender,receiver):\\n cls._unsubscribe(id(sender),receiver)\"\n]","isConversational":false},"negative_scores":{"kind":"list like","value":["0.6626395","0.64564025","0.6453663","0.6379453","0.6170811","0.61062586","0.61062586","0.61062586","0.61062586","0.61062586","0.60732317","0.6026578","0.58910805","0.5749465","0.57386756","0.5738046","0.5716363","0.5714279","0.57047355","0.5697949","0.56593245","0.56568694","0.56494164","0.56274873","0.56220424","0.5573087","0.55684745","0.5562139","0.55148065","0.5511321","0.549591","0.54330575","0.5427303","0.53831947","0.5377717","0.53704673","0.53653884","0.53619933","0.53498286","0.5346127","0.5336579","0.53358597","0.53229135","0.5314983","0.53144604","0.5309337","0.53061193","0.5276512","0.5251661","0.52168566","0.5215222","0.5215091","0.519419","0.5177541","0.51646894","0.51382333","0.51326233","0.51307577","0.5121972","0.51205784","0.50991315","0.5092092","0.5089654","0.50879407","0.5083061","0.5081658","0.50802225","0.5063257","0.50510687","0.50437534","0.5041081","0.5040269","0.50372285","0.5033088","0.5026242","0.5018926","0.50151","0.50132805","0.50117785","0.5011202","0.5008567","0.50082326","0.50037014","0.4992751","0.4989968","0.49868098","0.49868098","0.4984998","0.4979844","0.49750414","0.4971317","0.4970092","0.4969535","0.49662217","0.4961967","0.49591482","0.4946072","0.49437115","0.49434003","0.49398217","0.4915655"],"string":"[\n \"0.6626395\",\n \"0.64564025\",\n \"0.6453663\",\n \"0.6379453\",\n \"0.6170811\",\n \"0.61062586\",\n \"0.61062586\",\n \"0.61062586\",\n \"0.61062586\",\n \"0.61062586\",\n \"0.60732317\",\n \"0.6026578\",\n \"0.58910805\",\n \"0.5749465\",\n \"0.57386756\",\n \"0.5738046\",\n \"0.5716363\",\n \"0.5714279\",\n \"0.57047355\",\n \"0.5697949\",\n \"0.56593245\",\n \"0.56568694\",\n \"0.56494164\",\n \"0.56274873\",\n \"0.56220424\",\n \"0.5573087\",\n \"0.55684745\",\n \"0.5562139\",\n \"0.55148065\",\n \"0.5511321\",\n \"0.549591\",\n \"0.54330575\",\n \"0.5427303\",\n \"0.53831947\",\n \"0.5377717\",\n \"0.53704673\",\n \"0.53653884\",\n \"0.53619933\",\n \"0.53498286\",\n \"0.5346127\",\n \"0.5336579\",\n \"0.53358597\",\n \"0.53229135\",\n \"0.5314983\",\n \"0.53144604\",\n \"0.5309337\",\n \"0.53061193\",\n \"0.5276512\",\n \"0.5251661\",\n \"0.52168566\",\n \"0.5215222\",\n \"0.5215091\",\n \"0.519419\",\n \"0.5177541\",\n \"0.51646894\",\n \"0.51382333\",\n \"0.51326233\",\n \"0.51307577\",\n \"0.5121972\",\n \"0.51205784\",\n \"0.50991315\",\n \"0.5092092\",\n \"0.5089654\",\n \"0.50879407\",\n \"0.5083061\",\n \"0.5081658\",\n \"0.50802225\",\n \"0.5063257\",\n \"0.50510687\",\n \"0.50437534\",\n \"0.5041081\",\n \"0.5040269\",\n \"0.50372285\",\n \"0.5033088\",\n \"0.5026242\",\n \"0.5018926\",\n \"0.50151\",\n \"0.50132805\",\n \"0.50117785\",\n \"0.5011202\",\n \"0.5008567\",\n \"0.50082326\",\n \"0.50037014\",\n \"0.4992751\",\n \"0.4989968\",\n \"0.49868098\",\n \"0.49868098\",\n \"0.4984998\",\n \"0.4979844\",\n \"0.49750414\",\n \"0.4971317\",\n \"0.4970092\",\n \"0.4969535\",\n \"0.49662217\",\n \"0.4961967\",\n \"0.49591482\",\n \"0.4946072\",\n \"0.49437115\",\n \"0.49434003\",\n \"0.49398217\",\n \"0.4915655\"\n]","isConversational":false},"document_score":{"kind":"string","value":"0.0"},"document_rank":{"kind":"string","value":"-1"}}},{"rowIdx":93291,"cells":{"query":{"kind":"string","value":"Delete the user from the newsletter."},"document":{"kind":"string","value":"def user_pre_delete(sender, instance: User, **kwargs):\n\n if not hasattr(settings, 'ANYMAIL'):\n log.info(\"Mailgun not configured, skipping mailing list subscription update\")\n return\n\n if not hasattr(settings, 'MAILING_LIST_NEWSLETTER_EMAIL'):\n log.debug(\"Newsletter not configured, skipping mailing list subscription update\")\n return\n\n api_url_base = f\"https://api.mailgun.net/v3\"\n api_url_newsletter = f\"{api_url_base}/lists/{settings.MAILING_LIST_NEWSLETTER_EMAIL}\"\n requests.delete(\n f\"{api_url_newsletter}/members/{instance.email}\",\n auth=('api', settings.ANYMAIL['MAILGUN_API_KEY']),\n )\n log.info(\"Deleted user %i from mailing list\" % instance.id)"},"metadata":{"kind":"string","value":"{\n \"objective\": {\n \"self\": [],\n \"paired\": [],\n \"triplet\": [\n [\n \"query\",\n \"document\",\n \"negatives\"\n ]\n ]\n }\n}"},"negatives":{"kind":"list like","value":["def delete_user():","def delete_user():\n #TODO user delete\n pass","def delete_user(self, user):\n self.delete(user)","def delete_user(id):\n pass","def delete_user(self):\n User.user_list.remove(self)","def delete_user(self):\n User.user_list.remove(self)","def delete_user(self):\n User.user_list.remove(self)","def delete(self, user_id):\r\n return delete_user(request, user_id)","def delete_user(self):\n\n User.user_list.remove(self)","def delete_user(self, _id):\n return self.make_request(\"DELETE\", \"users/\"+_id, {})","def delete_user(self, user):\n # noinspection PyUnresolvedReferences\n self.delete(user)","def del_user(self, username):\n pass","def delete_user(self, user):\n self.execute(TABELLE['id_users'][\"delete\"], user[\"id\"])","def delete(self):\n\n user_id = get_jwt_identity()\n user = user_crud.get(user_id)\n if not user:\n abort(404, message=\"User not Found\")\n all_tokens = auth_crud.get_user_tokens(user_id)\n tokens = [token.to_dict() for token in all_tokens]\n for token in tokens:\n auth_crud.revoke_token(token['id'], user_id)\n user = user_crud.remove(user_id)\n\n return {'msg': 'User Removed'}","def delete(self):\n self.deleted = True\n # Deactivate the user to disallow authentication and also\n # to let the user verify the email again after recovery.\n self.is_active = False\n self.save()\n self.history.create(user_id=self.pk, action=user_history.DELETION)","def delete_user(cls, user_email):\n\n User.query.filter_by(email=user_email).delete()\n\n db.session.commit()\n\n print \"Successfully deleted user with the email: %s!\" % user_email","def delete(self, user_id):\n\n user = User.objects.get_or_404(public_id=user_id)\n return user.delete()","def delete_user(self):\n db_acces = DbMethods()\n username = self.result_table.currentItem().text()\n response = db_acces.delete_user(username)\n\n if response == True:\n self.populate_table()\n else:\n message = Message(\n self.language[\"error\"], self.language[\"inf_error\"])\n warning_message = message.create_iw_message(\n self.language[\"ok\"], \"warning\")\n warning_message.exec()","def delete_user(self) -> None:\n table_dictionary = {\n 'Apple': {\n 'table': 'AppleReceipts',\n 'user_id': 'User_id'\n },\n 'ESL': {\n 'table': 'ESLReceipts',\n 'user_id': 'User_id'\n },\n 'Transactions': {\n 'table': 'Transactions',\n 'user_id': 'User_id'\n },\n 'Users': {\n 'table': 'Users',\n 'user_id': 'id'\n },\n }\n\n # delete the current user's information from the db.\n for key in table_dictionary:\n query = f\"\"\"\n DELETE\n FROM {table_dictionary[key]['table']}\n WHERE {table_dictionary[key]['user_id']}=?;\n \"\"\"\n self.db.commit(query, values=(self.id,))\n\n # perform a sign out\n self.sign_out()\n\n log(f\"User:{self.id} has deleted their account.\")","def delete_user(self, instance, name):\n return instance.delete_user(name)","def deleteUser(user):\n delete_user(user)\n return redirect(url_for('login'))","def delete(self, user_id):\n return delete_user(user_id)","def delete_user(self):\n raise NotImplementedError(\"Function not yet implemented contact package creator\")","def delete_user(id):\n user = Users.query.filter_by(id=id).first()\n user.delete()\n if not user:\n return send_msg(404, 'Not Found')\n return send_msg(204, \"No data\")","def delete_user(user_id):\n user = User.query.get_or_404(user_id)\n db.session.delete(user)\n db.session.commit()\n\n return redirect('/')","def delete_user(self) -> 'outputs.ActingUserResponse':\n return pulumi.get(self, \"delete_user\")","def delete_user(self, user_id):\n return self._delete('/users/{0}'.format(user_id))","def delete(self, request, *args, **kwargs):\n self.object = self.get_object()\n user = request.user\n success_url = reverse_lazy('muxic:user', kwargs={'username': user.username})\n self.object.delete()\n return HttpResponseRedirect(success_url)","def delete_user(BrokerId=None, Username=None):\n pass","def delete(user_id: int):\n usr = get_by_id(user_id)\n if not usr:\n raise UserNotFound\n\n db.session.delete(usr)\n db.session.commit()","def delete_user():\r\n raise NotImplementedError()","def delete_user(self, user_id):\n\n # ask the model to delete the user\n um = User(self.settings)\n status = um.delete(user_id)\n\n # return\n return status","def delete(self, id):\n\t\ttry:\n\t\t\tuser_service.delete(id)\n\t\texcept AssertionError as e:\n\t\t\tuser_space.abort(400, e.args[0], status = \"Could not delete user\", statusCode = \"400\")\n\t\texcept Exception as e:\n\t\t\tuser_space.abort(500, e.args[0], status = \"Could not delete user\", statusCode = \"500\")","def delete_user(self, user_name):\n user = self.get_user(user_name)\n return self.client.delete_resource(user.get('href'))","def delete_user(user_id):\n\n user = User.query.get(user_id)\n db.session.delete(user)\n db.session.commit()\n return","def delete_user(user_id):\n user = User.query.get_or_404(user_id)\n db.session.delete(user)\n db.session.commit()\n\n return redirect(\"/users\")","def delete_user_with_email_confirmation(self, userid, deleteuserlink, template=''):\n payload = {'appkey': self._lr_object._get_api_key(), 'appsecret': self._lr_object._get_api_secret(),\n 'userid':userid, 'deleteuserlink':deleteuserlink,'template':template}\n url = SECURE_API_URL + \"raas/v1/user/deleteuseremail\"\n return self._lr_object._get_json(url, payload)","def delete_user(user_id):\n\n user = User.query.get_or_404(user_id)\n db.session.delete(user)\n db.session.commit()\n\n return redirect(\"/users\")","def delete_user():\n del globalopts.appdata[request.user]\n del globalopts.users[request.user]\n return \"\", 200","def delete(request):\n if request.method != 'POST':\n return HttpResponseNotAllowed(['POST'])\n username = request.user.get_username()\n context = RequestContext(request, {\n 'username': username})\n try:\n DataHubManager.remove_user(username=username, remove_db=True)\n django_logout(request)\n return render(request, 'delete-done.html', context)\n except User.DoesNotExist:\n return HttpResponseNotFound('User {0} not found.'.format(username))","def delete_user(UserName=None, AuthenticationType=None):\n pass","def delete(user_id):\n assert isinstance(user_id, ObjectId)\n\n User.objects(id=user_id).delete()","def delete_user(self, user):\n name = utils.get_name(user)\n self._user_manager.delete(name)","async def delete_user(user_id):\n \n user = User.select().where(User.id == user_id).first()\n\n if not user:\n return HTTPException(404, 'User not found')\n else:\n user.delete_instance()\n\n return f\"User {user.username} deleted successfully\"","def delete_user(request):\n user_id = request.POST.get('user_id')\n User.objects.filter(id=user_id).delete()\n response = {'status': 1, 'status_message': 'Success'}\n return HttpResponse(json.dumps(response))","def delete(self):\n return self._router_request(\n self._make_request_data(\n 'deleteUserCommand',\n data=dict(\n uid=self.parent,\n id=self.id\n )\n )\n )","def delete_by_email(\n user_to_delete: schemas.UserDelete,\n db_session: Session = Depends(get_db),\n current_user: models.User = Depends(get_current_admin_user)\n):\n db_user = crud.get_by_email(db_session, user_to_delete.email)\n\n if not db_user:\n raise HTTPException(\n status_code=status.HTTP_404_NOT_FOUND,\n detail=f'User with email \"{user_to_delete.email}\" not found.'\n )\n\n crud.remove(db_session, db_user)","def delete(self):\n data = UserRegister.parser.parse_args()\n user = UserModel.find_by_username(data['username'])\n\n if user:\n user.delete_from_db()\n else :\n return {'message': 'User not found!'} , 204\n\n return {'message': 'User deleted'},202","def delete(self, new_data, user_id):\n print(new_data)\n request_id = get_jwt_identity()\n user = user_crud.get(request_id)\n if not user.is_superuser:\n abort(401,\n message=\"You do not have permission to view this endpoint\")\n all_tokens = auth_crud.get_user_tokens(user_id)\n tokens = [token.to_dict() for token in all_tokens]\n for token in tokens:\n auth_crud.revoke_token(token['id'], user_id)\n user = user_crud.remove(user_id)\n\n return {'msg': 'User Removed'}","def delete(bot, update):\n chatID = update.message.chat_id\n username = get_user_info(chatID)['PID']\n logger.info(\"Deleting user credentials for {}!\".format(username))\n Chat.query.filter(Chat.chatID == chatID).delete() # Delete the user's record referenced by their ChatID\n Misc.query.filter(Misc.chatID == chatID).delete()\n db_session.commit()\n messageContent = \"Your credentials have been deleted, {}\\nHope to see you back soon!\".format(username[3:-4].title())\n bot.sendMessage(chat_id=update.message.chat_id, text=messageContent)\n \n mp.track(username, 'User Left')\n mp.people_set(username, {'active': False })","def user_delete(user_id):\n user = storage.get('User', user_id)\n if user is None:\n abort(404)\n user.delete()\n storage.save()\n return jsonify({}), 200","def user_id_delete(user_id):\n user = storage.get(\"User\", user_id)\n\n if user is None:\n abort(404)\n user.delete()\n del user\n return make_response(jsonify({}), 200)","def delete_user_process(user_id):\n\n db_user = User.query.get_or_404(user_id)\n\n db.session.delete(db_user)\n db.session.commit()\n\n return redirect(\"/users\")","def deleteUser(self,name):\n raise BorkedDeleteUser","async def del_user(conn: LDAPConnection, user: dict, mailman: Client) -> None:\n await conn.delete(user[\"dn\"])\n uid = user[\"attributes\"][\"uid\"][0]\n rmtree(user[\"attributes\"][\"homeDirectory\"][0])\n rmtree(f\"/webtree/{uid[:1]}/{uid}\")\n mailing_list = mailman.get_list(\"announce-redbrick\")\n mailing_list.unsubscribe(f\"{uid}@redbrick.dcu.ie\")","def delete_user(user_id):\n usr = storage.get(User, user_id)\n if usr:\n usr.delete(), storage.save()\n return {}\n else:\n abort(404)","def delete(self, id):\n userDao.delete(id)\n return \"\", 204","def deleteUser(self):\r\n #Find name and ID column\r\n userData = self.getCurrentUserData()\r\n\r\n #Prompt for confirmation\r\n deleteChoice = QMessageBox.question(self.view, 'Confirm user deletion', \r\n 'Are you sure you want to delete user ' \r\n + userData['Name'] + \" with ID \" + userData['User_ID'] + \r\n \" from database permanently?\", \r\n QMessageBox.Yes | QMessageBox.No)\r\n \r\n if (deleteChoice == QMessageBox.Yes):\r\n DBController().deleteUser(userData['User_ID'] )\r\n self.updateUserTable() #Re-fill table\r","def delete_user_by_email(tx: Transaction, user_email: str) -> BoltStatementResult:\n query = f\"\"\"\n MATCH(n: Person {{email: '{user_email}'}})\n OPTIONAL MATCH(n)--(p: Post)\n DETACH DELETE n, p\"\"\"\n return tx.run(query)","def delete_user(id_user: int):\n mycursor.execute(f\"\"\"DELETE FROM User\n WHERE id_user = {id_user}\"\"\")\n mydb.commit()\n return f\"L'utilisateur {id_user} a été supprimé\"","def delete_user(self, username):\n Log.info(\"Start to delete user.\")\n if self.check_if_user_exists(username):\n self.click((\"xpath\", self.user_checkbox_xpath.format(username)))\n self.click(self.user_del_btn)\n self.click(self.dialog_del_btn)\n self.wait_unit_el_present(self.user_table)\n Log.info(\"Use is deleted.\")\n else:\n Log.info(\"User \" + username + \" is not in the user list, not delete.\")","async def delete(self, request, uid):\n return await super(User, self).delete_item(request.app.pool, 'user',\n uid)","def delete_user(user_id=None):\n\n user = storage.get(\"User\", user_id)\n if user is None:\n abort(404)\n else:\n storage.delete(user)\n storage.save()\n return jsonify({}), 200","def delete(self):\n return self.collection().delete_one({'username': self.username})","def delete_user(user_id):\n user = storage.get(User, user_id)\n if user is None:\n abort(404)\n storage.delete(user)\n storage.save()\n return jsonify({}), 200","def delete(self, request, user_id=None):\n data = json.loads(request.body.decode())\n authenticated = Account.check_credentials(request, data['email'], data['password'])\n user = {}\n user['account_id'] = authenticated.id\n\n if authenticated.check_admin(request, user):\n NLTKOutput.remove(request=request, pk=user_id)\n Account.remove(request=request, pk=user_id)\n return Response(json='Account and content deleted', status=204)\n\n return Response(json='Not Authorized', status=401)","def user_delete(user_id=None):\n obj = storage.get(\"User\", user_id)\n if obj is None:\n abort(404)\n storage.delete(obj)\n storage.save()\n return jsonify({}), 200","def delete_user(id):\n user_repo = UserRepository(db)\n base_repo = BaseRepository(db, User)\n u = base_repo.get_by_id(id)\n if not u:\n click.echo(\"User with specified id does not exists.\")\n return ERROR_USER_DOES_NOT_EXIST\n user_repo.delete_user(u)\n click.echo(\"User with id \" + str(id) + \" has been deleted.\")","def del_user(item, username, passw):\n user = User.load_user_by_username(item, username)\n if not user:\n print(\"User does not exist!\")\n elif check_password(passw, user.hashed_password):\n user.delete(item)\n print(\"User deleted.\")\n else:\n print(\"Incorrect password!\")","def delete(self, userinformation):\n self.db.remove(userinformation)","def delete_user(self, user_name):\r\n params = {'UserName' : user_name}\r\n return self.get_response('DeleteUser', params)","def delete_user(user_id):\n user_obj = storage.get(\"User\", user_id)\n if user_obj:\n storage.delete(user_obj)\n storage.save()\n return jsonify({}), 200\n else:\n abort(404)","def delete(self, user_id):\n user = User.query.get(user_id)\n \n if user is None:\n return abort(422, message=\"User does not exist\")\n \n # check if the user is an admin and is the only one\n admins = User.query.filter_by(admin=True).all()\n if user.id == get_jwt_identity() and len(admins) == 1:\n return abort(422, message=\"User is the only admin, there must be at least one admin in the system\")\n \n user.delete()\n \n return { 'message': \"User '{}' has been deleted\".format(user.id) }","def delete(cls):\n user = user_schema.load(request.get_json(), partial=(\"email\",))\n\n current_identity = get_jwt_identity()\n db_user = UserModel.find_by_id(current_identity)\n logging.info(\n f\"Delete called by {db_user.id}: {db_user.username} with data: {user['username']}\"\n )\n if db_user.username == user['username']:\n if is_correct_password(db_user.pw_salt, db_user.pw_hash, user['password']):\n db_user.delete_from_db()\n return {\"message\": msgs.DELETED.format(db_user.username)}, 200\n else:\n return {\"error\": msgs.INVALID_PASSWORD}, 401\n return {\"error\": msgs.OWN_RECORD_ONLY}, 401","def delete(user_id):\n # Get the user requested\n user = User.query.filter(User.user_id == user_id).one_or_none()\n\n if user is not None:\n db.session.delete(user)\n db.session.commit()\n return (\n \"User {user_id} deleted\".format(user_id=user_id), 200\n )\n\n else:\n abort(\n 404,\n \"Person not found for Id: {user_id}\".format(user_id=user_id),\n )","def delete_user(user_id):\n netAdminToolDB = app.config['DATABASE']\n user = netAdminToolDB.get_user(user_id)\n\n if user == None:\n return jsonify({'error': 'User_id not found'}), 404\n\n netAdminToolDB.delete_user(user_id)\n return jsonify({'result': True})","def delete_user(self, user_id):\n if self.database is None:\n raise Exception(\"No database.\")\n if user_id is None or len(user_id) == 0:\n raise Exception(\"Bad parameter.\")\n return self.database.delete_user(user_id)","def del_user(request):\r\n mdict = request.matchdict\r\n\r\n # Submit a username.\r\n del_username = mdict.get('username', None)\r\n\r\n if del_username is None:\r\n LOG.error('No username to remove.')\r\n request.response.status_int = 400\r\n return _api_response(request, {\r\n 'error': 'Bad Request: No username to remove.',\r\n })\r\n\r\n u = UserMgr.get(username=del_username)\r\n\r\n if not u:\r\n LOG.error('Username not found.')\r\n request.response.status_int = 404\r\n return _api_response(request, {\r\n 'error': 'User not found.',\r\n })\r\n\r\n try:\r\n # First delete all the tag references for this user's bookmarks.\r\n res = DBSession.query(Bmark.bid).filter(Bmark.username == u.username)\r\n bids = [b[0] for b in res]\r\n\r\n qry = bmarks_tags.delete(bmarks_tags.c.bmark_id.in_(bids))\r\n qry.execute()\r\n\r\n # Delete all of the bmarks for this year.\r\n Bmark.query.filter(Bmark.username == u.username).delete()\r\n DBSession.delete(u)\r\n return _api_response(request, {\r\n 'success': True,\r\n 'message': 'Removed user: ' + del_username\r\n })\r\n except Exception, exc:\r\n # There might be cascade issues or something that causes us to fail in\r\n # removing.\r\n LOG.error(exc)\r\n request.response.status_int = 500\r\n return _api_response(request, {\r\n 'error': 'Bad Request: ' + str(exc)\r\n })","def delete_user(self, email):\n\n try:\n return self.client.admin_delete_user(\n UserPoolId=self.user_pool_id,\n Username=email\n )\n except self.client.exceptions.UserNotFoundException:\n raise Exception('An account with the given email does not exist.')","def delete_admin(request, username, d):\n a = AdminUser.objects.get(username__exact=d)\n a.delete()\n print('deleted')\n return redirect('list_users', username)","def kick_user(msg: telebot.types.Message):\n user_id = int(msg.text.split()[1])\n if User.delete().where(User.user_id == user_id).execute():\n bot.send_message(\n msg.from_user.id,\n 'OK.'\n )\n else:\n bot.send_message(\n msg.from_user.id,\n 'Cannot fo that.'\n )","def delete_user(self,userid, cursor):\n sql=\"DELETE FROM users WHERE userid = %s\"\n cursor.execute(sql,(userid))","def deleteUser(self, request, context):\n context.set_code(grpc.StatusCode.UNIMPLEMENTED)\n context.set_details('Method not implemented!')\n raise NotImplementedError('Method not implemented!')","def delete_user(user_id=None):\n obj = storage.get('User', user_id)\n if obj is None:\n abort(404)\n else:\n storage.delete(obj)\n storage.save()\n return jsonify({}), 200","def delete_user_account(connection,user):\r\n with connection:\r\n connection.execute(DELETE_SPECIFIC_USER,(user,))","def onUserDeletion(event):\n request = getRequest()\n if not IProductLayer.providedBy(request):\n return\n\n client = getUtility(IAdminClient)\n xmpp_users = getUtility(IXMPPUsers)\n\n principal_id = event.principal\n principal_jid = xmpp_users.getUserJID(principal_id)\n\n pass_storage = getUtility(IXMPPPasswordStorage)\n pass_storage.remove(principal_id)\n\n d = users.deletePrincipal(client, principal_jid)\n return d","def delete(self, *args, **kwargs):\n # Delete the User and UserProfile objects associated with the\n # Member.\n user_profile = self.userprofile\n user = user_profile.user\n user_profile.delete()\n user.delete()\n # Delete the member itself\n super(Member, self).delete(*args, **kwargs)","def delete_user_by_id(user_id):\n return woo_request_helper().delete_details(wc_endpoint='customers/{}'.format(user_id))","def deleted_user(request):\n auth.logout(request)\n messages.success(request, \"Your profile has been deleted. Please contact us if you want to undo this.\")\n return redirect(reverse('index'))","def delete_user(user_id):\n temp = models.storage.get('User', user_id)\n if temp is None:\n abort(404)\n temp.delete()\n models.storage.save()\n return jsonify({})","def delete_user(self, uname):\r\n\r\n if not uname:\r\n return _('Must provide username')\r\n if '@' in uname:\r\n try:\r\n user = User.objects.get(email=uname)\r\n except User.DoesNotExist, err:\r\n msg = _('Cannot find user with email address {0}').format(uname)\r\n return msg\r\n else:\r\n try:\r\n user = User.objects.get(username=uname)\r\n except User.DoesNotExist, err:\r\n msg = _('Cannot find user with username {0} - {1}'\r\n ).format(uname, str(err))\r\n return msg\r\n user.delete()\r\n return _('Deleted user {0}').format(uname)","def _delete_user(self, user):\n if User.delete_user(user):\n self.session.output({'deleted': 'user {} and their related accounts'.format(user)})\n return True\n else:\n self.session.output({'invalid_user': 'please enter valid user ID!\\n'}, '[ Fail to delete user ]')\n return False","def deleteUser(self, password, feedback=\"\"):\n\t\turl = \"https://habitica.com/api/v3/user\"\n\t\tpayload = {'password': password, 'feedback': feedback}\n\t\treturn(deleteUrl(url, self.credentials, payload))","def delete_student(request, s_id):\n user = CustomUser.objects.get(id=s_id)\n user.delete()\n messages.success(request, \"The student has been deleted successfully\")\n return redirect('all_students')","def delete_personal_message(request, pk=None):\n user = User.objects.get(email=request.user.email)\n contactuserposts = ContactUser.objects.all()\n contactuserpost = get_object_or_404(ContactUser, pk=pk)\n if request.method == \"POST\":\n contactuserpost.delete()\n messages.success(request, 'This message has been successfully deleted.')\n return redirect(user_profile)\n return render(request, \"personalmessagedelete.html\", {'contactuserposts': contactuserposts})","def unsubscribe(self, update, context):\n # remove or update to the sqlite table.\n chat = update.message.chat\n self.db_manager.remove_user(chat.id)\n self.logger.info(\n 'Username: %s and chat_id: %s unsubscribed to the list.' % (chat.username, chat.id)\n )\n update.message.reply_text('You have successfully unsubscribed the notifications forever.')","def remove(self, user_id):\n pass","def delete_user(self, userId):\n\n try:\n query = \"delete from user where userId = {}\".format(userId)\n print(query)\n cur = self.con.cursor()\n cur.execute(query)\n self.con.commit()\n\n logger.info(\"Deleted\")\n except Exception as e:\n logger.error(\"Error occured at data deletion..\", e)","def delete(self, user_id):\n user = User.query.get(user_id)\n\n if user is None:\n return mk_response(\"User does not exist\", 422)\n\n # check if the user is an admin and is the only one\n admins = User.query.filter_by(admin=True).all()\n if user.id == get_jwt_identity() and len(admins) == 1:\n return mk_response(\"User is the only admin, there must \" +\n \"be at least one admin in the system\", 422)\n\n user.delete()\n\n return mk_response(\"User '{}' has been deleted\".format(user.id))","def del_user(self, name):\n del self.users[irc.strings.IRCFoldedCase(modules.trim_nick(name))]"],"string":"[\n \"def delete_user():\",\n \"def delete_user():\\n #TODO user delete\\n pass\",\n \"def delete_user(self, user):\\n self.delete(user)\",\n \"def delete_user(id):\\n pass\",\n \"def delete_user(self):\\n User.user_list.remove(self)\",\n \"def delete_user(self):\\n User.user_list.remove(self)\",\n \"def delete_user(self):\\n User.user_list.remove(self)\",\n \"def delete(self, user_id):\\r\\n return delete_user(request, user_id)\",\n \"def delete_user(self):\\n\\n User.user_list.remove(self)\",\n \"def delete_user(self, _id):\\n return self.make_request(\\\"DELETE\\\", \\\"users/\\\"+_id, {})\",\n \"def delete_user(self, user):\\n # noinspection PyUnresolvedReferences\\n self.delete(user)\",\n \"def del_user(self, username):\\n pass\",\n \"def delete_user(self, user):\\n self.execute(TABELLE['id_users'][\\\"delete\\\"], user[\\\"id\\\"])\",\n \"def delete(self):\\n\\n user_id = get_jwt_identity()\\n user = user_crud.get(user_id)\\n if not user:\\n abort(404, message=\\\"User not Found\\\")\\n all_tokens = auth_crud.get_user_tokens(user_id)\\n tokens = [token.to_dict() for token in all_tokens]\\n for token in tokens:\\n auth_crud.revoke_token(token['id'], user_id)\\n user = user_crud.remove(user_id)\\n\\n return {'msg': 'User Removed'}\",\n \"def delete(self):\\n self.deleted = True\\n # Deactivate the user to disallow authentication and also\\n # to let the user verify the email again after recovery.\\n self.is_active = False\\n self.save()\\n self.history.create(user_id=self.pk, action=user_history.DELETION)\",\n \"def delete_user(cls, user_email):\\n\\n User.query.filter_by(email=user_email).delete()\\n\\n db.session.commit()\\n\\n print \\\"Successfully deleted user with the email: %s!\\\" % user_email\",\n \"def delete(self, user_id):\\n\\n user = User.objects.get_or_404(public_id=user_id)\\n return user.delete()\",\n \"def delete_user(self):\\n db_acces = DbMethods()\\n username = self.result_table.currentItem().text()\\n response = db_acces.delete_user(username)\\n\\n if response == True:\\n self.populate_table()\\n else:\\n message = Message(\\n self.language[\\\"error\\\"], self.language[\\\"inf_error\\\"])\\n warning_message = message.create_iw_message(\\n self.language[\\\"ok\\\"], \\\"warning\\\")\\n warning_message.exec()\",\n \"def delete_user(self) -> None:\\n table_dictionary = {\\n 'Apple': {\\n 'table': 'AppleReceipts',\\n 'user_id': 'User_id'\\n },\\n 'ESL': {\\n 'table': 'ESLReceipts',\\n 'user_id': 'User_id'\\n },\\n 'Transactions': {\\n 'table': 'Transactions',\\n 'user_id': 'User_id'\\n },\\n 'Users': {\\n 'table': 'Users',\\n 'user_id': 'id'\\n },\\n }\\n\\n # delete the current user's information from the db.\\n for key in table_dictionary:\\n query = f\\\"\\\"\\\"\\n DELETE\\n FROM {table_dictionary[key]['table']}\\n WHERE {table_dictionary[key]['user_id']}=?;\\n \\\"\\\"\\\"\\n self.db.commit(query, values=(self.id,))\\n\\n # perform a sign out\\n self.sign_out()\\n\\n log(f\\\"User:{self.id} has deleted their account.\\\")\",\n \"def delete_user(self, instance, name):\\n return instance.delete_user(name)\",\n \"def deleteUser(user):\\n delete_user(user)\\n return redirect(url_for('login'))\",\n \"def delete(self, user_id):\\n return delete_user(user_id)\",\n \"def delete_user(self):\\n raise NotImplementedError(\\\"Function not yet implemented contact package creator\\\")\",\n \"def delete_user(id):\\n user = Users.query.filter_by(id=id).first()\\n user.delete()\\n if not user:\\n return send_msg(404, 'Not Found')\\n return send_msg(204, \\\"No data\\\")\",\n \"def delete_user(user_id):\\n user = User.query.get_or_404(user_id)\\n db.session.delete(user)\\n db.session.commit()\\n\\n return redirect('/')\",\n \"def delete_user(self) -> 'outputs.ActingUserResponse':\\n return pulumi.get(self, \\\"delete_user\\\")\",\n \"def delete_user(self, user_id):\\n return self._delete('/users/{0}'.format(user_id))\",\n \"def delete(self, request, *args, **kwargs):\\n self.object = self.get_object()\\n user = request.user\\n success_url = reverse_lazy('muxic:user', kwargs={'username': user.username})\\n self.object.delete()\\n return HttpResponseRedirect(success_url)\",\n \"def delete_user(BrokerId=None, Username=None):\\n pass\",\n \"def delete(user_id: int):\\n usr = get_by_id(user_id)\\n if not usr:\\n raise UserNotFound\\n\\n db.session.delete(usr)\\n db.session.commit()\",\n \"def delete_user():\\r\\n raise NotImplementedError()\",\n \"def delete_user(self, user_id):\\n\\n # ask the model to delete the user\\n um = User(self.settings)\\n status = um.delete(user_id)\\n\\n # return\\n return status\",\n \"def delete(self, id):\\n\\t\\ttry:\\n\\t\\t\\tuser_service.delete(id)\\n\\t\\texcept AssertionError as e:\\n\\t\\t\\tuser_space.abort(400, e.args[0], status = \\\"Could not delete user\\\", statusCode = \\\"400\\\")\\n\\t\\texcept Exception as e:\\n\\t\\t\\tuser_space.abort(500, e.args[0], status = \\\"Could not delete user\\\", statusCode = \\\"500\\\")\",\n \"def delete_user(self, user_name):\\n user = self.get_user(user_name)\\n return self.client.delete_resource(user.get('href'))\",\n \"def delete_user(user_id):\\n\\n user = User.query.get(user_id)\\n db.session.delete(user)\\n db.session.commit()\\n return\",\n \"def delete_user(user_id):\\n user = User.query.get_or_404(user_id)\\n db.session.delete(user)\\n db.session.commit()\\n\\n return redirect(\\\"/users\\\")\",\n \"def delete_user_with_email_confirmation(self, userid, deleteuserlink, template=''):\\n payload = {'appkey': self._lr_object._get_api_key(), 'appsecret': self._lr_object._get_api_secret(),\\n 'userid':userid, 'deleteuserlink':deleteuserlink,'template':template}\\n url = SECURE_API_URL + \\\"raas/v1/user/deleteuseremail\\\"\\n return self._lr_object._get_json(url, payload)\",\n \"def delete_user(user_id):\\n\\n user = User.query.get_or_404(user_id)\\n db.session.delete(user)\\n db.session.commit()\\n\\n return redirect(\\\"/users\\\")\",\n \"def delete_user():\\n del globalopts.appdata[request.user]\\n del globalopts.users[request.user]\\n return \\\"\\\", 200\",\n \"def delete(request):\\n if request.method != 'POST':\\n return HttpResponseNotAllowed(['POST'])\\n username = request.user.get_username()\\n context = RequestContext(request, {\\n 'username': username})\\n try:\\n DataHubManager.remove_user(username=username, remove_db=True)\\n django_logout(request)\\n return render(request, 'delete-done.html', context)\\n except User.DoesNotExist:\\n return HttpResponseNotFound('User {0} not found.'.format(username))\",\n \"def delete_user(UserName=None, AuthenticationType=None):\\n pass\",\n \"def delete(user_id):\\n assert isinstance(user_id, ObjectId)\\n\\n User.objects(id=user_id).delete()\",\n \"def delete_user(self, user):\\n name = utils.get_name(user)\\n self._user_manager.delete(name)\",\n \"async def delete_user(user_id):\\n \\n user = User.select().where(User.id == user_id).first()\\n\\n if not user:\\n return HTTPException(404, 'User not found')\\n else:\\n user.delete_instance()\\n\\n return f\\\"User {user.username} deleted successfully\\\"\",\n \"def delete_user(request):\\n user_id = request.POST.get('user_id')\\n User.objects.filter(id=user_id).delete()\\n response = {'status': 1, 'status_message': 'Success'}\\n return HttpResponse(json.dumps(response))\",\n \"def delete(self):\\n return self._router_request(\\n self._make_request_data(\\n 'deleteUserCommand',\\n data=dict(\\n uid=self.parent,\\n id=self.id\\n )\\n )\\n )\",\n \"def delete_by_email(\\n user_to_delete: schemas.UserDelete,\\n db_session: Session = Depends(get_db),\\n current_user: models.User = Depends(get_current_admin_user)\\n):\\n db_user = crud.get_by_email(db_session, user_to_delete.email)\\n\\n if not db_user:\\n raise HTTPException(\\n status_code=status.HTTP_404_NOT_FOUND,\\n detail=f'User with email \\\"{user_to_delete.email}\\\" not found.'\\n )\\n\\n crud.remove(db_session, db_user)\",\n \"def delete(self):\\n data = UserRegister.parser.parse_args()\\n user = UserModel.find_by_username(data['username'])\\n\\n if user:\\n user.delete_from_db()\\n else :\\n return {'message': 'User not found!'} , 204\\n\\n return {'message': 'User deleted'},202\",\n \"def delete(self, new_data, user_id):\\n print(new_data)\\n request_id = get_jwt_identity()\\n user = user_crud.get(request_id)\\n if not user.is_superuser:\\n abort(401,\\n message=\\\"You do not have permission to view this endpoint\\\")\\n all_tokens = auth_crud.get_user_tokens(user_id)\\n tokens = [token.to_dict() for token in all_tokens]\\n for token in tokens:\\n auth_crud.revoke_token(token['id'], user_id)\\n user = user_crud.remove(user_id)\\n\\n return {'msg': 'User Removed'}\",\n \"def delete(bot, update):\\n chatID = update.message.chat_id\\n username = get_user_info(chatID)['PID']\\n logger.info(\\\"Deleting user credentials for {}!\\\".format(username))\\n Chat.query.filter(Chat.chatID == chatID).delete() # Delete the user's record referenced by their ChatID\\n Misc.query.filter(Misc.chatID == chatID).delete()\\n db_session.commit()\\n messageContent = \\\"Your credentials have been deleted, {}\\\\nHope to see you back soon!\\\".format(username[3:-4].title())\\n bot.sendMessage(chat_id=update.message.chat_id, text=messageContent)\\n \\n mp.track(username, 'User Left')\\n mp.people_set(username, {'active': False })\",\n \"def user_delete(user_id):\\n user = storage.get('User', user_id)\\n if user is None:\\n abort(404)\\n user.delete()\\n storage.save()\\n return jsonify({}), 200\",\n \"def user_id_delete(user_id):\\n user = storage.get(\\\"User\\\", user_id)\\n\\n if user is None:\\n abort(404)\\n user.delete()\\n del user\\n return make_response(jsonify({}), 200)\",\n \"def delete_user_process(user_id):\\n\\n db_user = User.query.get_or_404(user_id)\\n\\n db.session.delete(db_user)\\n db.session.commit()\\n\\n return redirect(\\\"/users\\\")\",\n \"def deleteUser(self,name):\\n raise BorkedDeleteUser\",\n \"async def del_user(conn: LDAPConnection, user: dict, mailman: Client) -> None:\\n await conn.delete(user[\\\"dn\\\"])\\n uid = user[\\\"attributes\\\"][\\\"uid\\\"][0]\\n rmtree(user[\\\"attributes\\\"][\\\"homeDirectory\\\"][0])\\n rmtree(f\\\"/webtree/{uid[:1]}/{uid}\\\")\\n mailing_list = mailman.get_list(\\\"announce-redbrick\\\")\\n mailing_list.unsubscribe(f\\\"{uid}@redbrick.dcu.ie\\\")\",\n \"def delete_user(user_id):\\n usr = storage.get(User, user_id)\\n if usr:\\n usr.delete(), storage.save()\\n return {}\\n else:\\n abort(404)\",\n \"def delete(self, id):\\n userDao.delete(id)\\n return \\\"\\\", 204\",\n \"def deleteUser(self):\\r\\n #Find name and ID column\\r\\n userData = self.getCurrentUserData()\\r\\n\\r\\n #Prompt for confirmation\\r\\n deleteChoice = QMessageBox.question(self.view, 'Confirm user deletion', \\r\\n 'Are you sure you want to delete user ' \\r\\n + userData['Name'] + \\\" with ID \\\" + userData['User_ID'] + \\r\\n \\\" from database permanently?\\\", \\r\\n QMessageBox.Yes | QMessageBox.No)\\r\\n \\r\\n if (deleteChoice == QMessageBox.Yes):\\r\\n DBController().deleteUser(userData['User_ID'] )\\r\\n self.updateUserTable() #Re-fill table\\r\",\n \"def delete_user_by_email(tx: Transaction, user_email: str) -> BoltStatementResult:\\n query = f\\\"\\\"\\\"\\n MATCH(n: Person {{email: '{user_email}'}})\\n OPTIONAL MATCH(n)--(p: Post)\\n DETACH DELETE n, p\\\"\\\"\\\"\\n return tx.run(query)\",\n \"def delete_user(id_user: int):\\n mycursor.execute(f\\\"\\\"\\\"DELETE FROM User\\n WHERE id_user = {id_user}\\\"\\\"\\\")\\n mydb.commit()\\n return f\\\"L'utilisateur {id_user} a été supprimé\\\"\",\n \"def delete_user(self, username):\\n Log.info(\\\"Start to delete user.\\\")\\n if self.check_if_user_exists(username):\\n self.click((\\\"xpath\\\", self.user_checkbox_xpath.format(username)))\\n self.click(self.user_del_btn)\\n self.click(self.dialog_del_btn)\\n self.wait_unit_el_present(self.user_table)\\n Log.info(\\\"Use is deleted.\\\")\\n else:\\n Log.info(\\\"User \\\" + username + \\\" is not in the user list, not delete.\\\")\",\n \"async def delete(self, request, uid):\\n return await super(User, self).delete_item(request.app.pool, 'user',\\n uid)\",\n \"def delete_user(user_id=None):\\n\\n user = storage.get(\\\"User\\\", user_id)\\n if user is None:\\n abort(404)\\n else:\\n storage.delete(user)\\n storage.save()\\n return jsonify({}), 200\",\n \"def delete(self):\\n return self.collection().delete_one({'username': self.username})\",\n \"def delete_user(user_id):\\n user = storage.get(User, user_id)\\n if user is None:\\n abort(404)\\n storage.delete(user)\\n storage.save()\\n return jsonify({}), 200\",\n \"def delete(self, request, user_id=None):\\n data = json.loads(request.body.decode())\\n authenticated = Account.check_credentials(request, data['email'], data['password'])\\n user = {}\\n user['account_id'] = authenticated.id\\n\\n if authenticated.check_admin(request, user):\\n NLTKOutput.remove(request=request, pk=user_id)\\n Account.remove(request=request, pk=user_id)\\n return Response(json='Account and content deleted', status=204)\\n\\n return Response(json='Not Authorized', status=401)\",\n \"def user_delete(user_id=None):\\n obj = storage.get(\\\"User\\\", user_id)\\n if obj is None:\\n abort(404)\\n storage.delete(obj)\\n storage.save()\\n return jsonify({}), 200\",\n \"def delete_user(id):\\n user_repo = UserRepository(db)\\n base_repo = BaseRepository(db, User)\\n u = base_repo.get_by_id(id)\\n if not u:\\n click.echo(\\\"User with specified id does not exists.\\\")\\n return ERROR_USER_DOES_NOT_EXIST\\n user_repo.delete_user(u)\\n click.echo(\\\"User with id \\\" + str(id) + \\\" has been deleted.\\\")\",\n \"def del_user(item, username, passw):\\n user = User.load_user_by_username(item, username)\\n if not user:\\n print(\\\"User does not exist!\\\")\\n elif check_password(passw, user.hashed_password):\\n user.delete(item)\\n print(\\\"User deleted.\\\")\\n else:\\n print(\\\"Incorrect password!\\\")\",\n \"def delete(self, userinformation):\\n self.db.remove(userinformation)\",\n \"def delete_user(self, user_name):\\r\\n params = {'UserName' : user_name}\\r\\n return self.get_response('DeleteUser', params)\",\n \"def delete_user(user_id):\\n user_obj = storage.get(\\\"User\\\", user_id)\\n if user_obj:\\n storage.delete(user_obj)\\n storage.save()\\n return jsonify({}), 200\\n else:\\n abort(404)\",\n \"def delete(self, user_id):\\n user = User.query.get(user_id)\\n \\n if user is None:\\n return abort(422, message=\\\"User does not exist\\\")\\n \\n # check if the user is an admin and is the only one\\n admins = User.query.filter_by(admin=True).all()\\n if user.id == get_jwt_identity() and len(admins) == 1:\\n return abort(422, message=\\\"User is the only admin, there must be at least one admin in the system\\\")\\n \\n user.delete()\\n \\n return { 'message': \\\"User '{}' has been deleted\\\".format(user.id) }\",\n \"def delete(cls):\\n user = user_schema.load(request.get_json(), partial=(\\\"email\\\",))\\n\\n current_identity = get_jwt_identity()\\n db_user = UserModel.find_by_id(current_identity)\\n logging.info(\\n f\\\"Delete called by {db_user.id}: {db_user.username} with data: {user['username']}\\\"\\n )\\n if db_user.username == user['username']:\\n if is_correct_password(db_user.pw_salt, db_user.pw_hash, user['password']):\\n db_user.delete_from_db()\\n return {\\\"message\\\": msgs.DELETED.format(db_user.username)}, 200\\n else:\\n return {\\\"error\\\": msgs.INVALID_PASSWORD}, 401\\n return {\\\"error\\\": msgs.OWN_RECORD_ONLY}, 401\",\n \"def delete(user_id):\\n # Get the user requested\\n user = User.query.filter(User.user_id == user_id).one_or_none()\\n\\n if user is not None:\\n db.session.delete(user)\\n db.session.commit()\\n return (\\n \\\"User {user_id} deleted\\\".format(user_id=user_id), 200\\n )\\n\\n else:\\n abort(\\n 404,\\n \\\"Person not found for Id: {user_id}\\\".format(user_id=user_id),\\n )\",\n \"def delete_user(user_id):\\n netAdminToolDB = app.config['DATABASE']\\n user = netAdminToolDB.get_user(user_id)\\n\\n if user == None:\\n return jsonify({'error': 'User_id not found'}), 404\\n\\n netAdminToolDB.delete_user(user_id)\\n return jsonify({'result': True})\",\n \"def delete_user(self, user_id):\\n if self.database is None:\\n raise Exception(\\\"No database.\\\")\\n if user_id is None or len(user_id) == 0:\\n raise Exception(\\\"Bad parameter.\\\")\\n return self.database.delete_user(user_id)\",\n \"def del_user(request):\\r\\n mdict = request.matchdict\\r\\n\\r\\n # Submit a username.\\r\\n del_username = mdict.get('username', None)\\r\\n\\r\\n if del_username is None:\\r\\n LOG.error('No username to remove.')\\r\\n request.response.status_int = 400\\r\\n return _api_response(request, {\\r\\n 'error': 'Bad Request: No username to remove.',\\r\\n })\\r\\n\\r\\n u = UserMgr.get(username=del_username)\\r\\n\\r\\n if not u:\\r\\n LOG.error('Username not found.')\\r\\n request.response.status_int = 404\\r\\n return _api_response(request, {\\r\\n 'error': 'User not found.',\\r\\n })\\r\\n\\r\\n try:\\r\\n # First delete all the tag references for this user's bookmarks.\\r\\n res = DBSession.query(Bmark.bid).filter(Bmark.username == u.username)\\r\\n bids = [b[0] for b in res]\\r\\n\\r\\n qry = bmarks_tags.delete(bmarks_tags.c.bmark_id.in_(bids))\\r\\n qry.execute()\\r\\n\\r\\n # Delete all of the bmarks for this year.\\r\\n Bmark.query.filter(Bmark.username == u.username).delete()\\r\\n DBSession.delete(u)\\r\\n return _api_response(request, {\\r\\n 'success': True,\\r\\n 'message': 'Removed user: ' + del_username\\r\\n })\\r\\n except Exception, exc:\\r\\n # There might be cascade issues or something that causes us to fail in\\r\\n # removing.\\r\\n LOG.error(exc)\\r\\n request.response.status_int = 500\\r\\n return _api_response(request, {\\r\\n 'error': 'Bad Request: ' + str(exc)\\r\\n })\",\n \"def delete_user(self, email):\\n\\n try:\\n return self.client.admin_delete_user(\\n UserPoolId=self.user_pool_id,\\n Username=email\\n )\\n except self.client.exceptions.UserNotFoundException:\\n raise Exception('An account with the given email does not exist.')\",\n \"def delete_admin(request, username, d):\\n a = AdminUser.objects.get(username__exact=d)\\n a.delete()\\n print('deleted')\\n return redirect('list_users', username)\",\n \"def kick_user(msg: telebot.types.Message):\\n user_id = int(msg.text.split()[1])\\n if User.delete().where(User.user_id == user_id).execute():\\n bot.send_message(\\n msg.from_user.id,\\n 'OK.'\\n )\\n else:\\n bot.send_message(\\n msg.from_user.id,\\n 'Cannot fo that.'\\n )\",\n \"def delete_user(self,userid, cursor):\\n sql=\\\"DELETE FROM users WHERE userid = %s\\\"\\n cursor.execute(sql,(userid))\",\n \"def deleteUser(self, request, context):\\n context.set_code(grpc.StatusCode.UNIMPLEMENTED)\\n context.set_details('Method not implemented!')\\n raise NotImplementedError('Method not implemented!')\",\n \"def delete_user(user_id=None):\\n obj = storage.get('User', user_id)\\n if obj is None:\\n abort(404)\\n else:\\n storage.delete(obj)\\n storage.save()\\n return jsonify({}), 200\",\n \"def delete_user_account(connection,user):\\r\\n with connection:\\r\\n connection.execute(DELETE_SPECIFIC_USER,(user,))\",\n \"def onUserDeletion(event):\\n request = getRequest()\\n if not IProductLayer.providedBy(request):\\n return\\n\\n client = getUtility(IAdminClient)\\n xmpp_users = getUtility(IXMPPUsers)\\n\\n principal_id = event.principal\\n principal_jid = xmpp_users.getUserJID(principal_id)\\n\\n pass_storage = getUtility(IXMPPPasswordStorage)\\n pass_storage.remove(principal_id)\\n\\n d = users.deletePrincipal(client, principal_jid)\\n return d\",\n \"def delete(self, *args, **kwargs):\\n # Delete the User and UserProfile objects associated with the\\n # Member.\\n user_profile = self.userprofile\\n user = user_profile.user\\n user_profile.delete()\\n user.delete()\\n # Delete the member itself\\n super(Member, self).delete(*args, **kwargs)\",\n \"def delete_user_by_id(user_id):\\n return woo_request_helper().delete_details(wc_endpoint='customers/{}'.format(user_id))\",\n \"def deleted_user(request):\\n auth.logout(request)\\n messages.success(request, \\\"Your profile has been deleted. Please contact us if you want to undo this.\\\")\\n return redirect(reverse('index'))\",\n \"def delete_user(user_id):\\n temp = models.storage.get('User', user_id)\\n if temp is None:\\n abort(404)\\n temp.delete()\\n models.storage.save()\\n return jsonify({})\",\n \"def delete_user(self, uname):\\r\\n\\r\\n if not uname:\\r\\n return _('Must provide username')\\r\\n if '@' in uname:\\r\\n try:\\r\\n user = User.objects.get(email=uname)\\r\\n except User.DoesNotExist, err:\\r\\n msg = _('Cannot find user with email address {0}').format(uname)\\r\\n return msg\\r\\n else:\\r\\n try:\\r\\n user = User.objects.get(username=uname)\\r\\n except User.DoesNotExist, err:\\r\\n msg = _('Cannot find user with username {0} - {1}'\\r\\n ).format(uname, str(err))\\r\\n return msg\\r\\n user.delete()\\r\\n return _('Deleted user {0}').format(uname)\",\n \"def _delete_user(self, user):\\n if User.delete_user(user):\\n self.session.output({'deleted': 'user {} and their related accounts'.format(user)})\\n return True\\n else:\\n self.session.output({'invalid_user': 'please enter valid user ID!\\\\n'}, '[ Fail to delete user ]')\\n return False\",\n \"def deleteUser(self, password, feedback=\\\"\\\"):\\n\\t\\turl = \\\"https://habitica.com/api/v3/user\\\"\\n\\t\\tpayload = {'password': password, 'feedback': feedback}\\n\\t\\treturn(deleteUrl(url, self.credentials, payload))\",\n \"def delete_student(request, s_id):\\n user = CustomUser.objects.get(id=s_id)\\n user.delete()\\n messages.success(request, \\\"The student has been deleted successfully\\\")\\n return redirect('all_students')\",\n \"def delete_personal_message(request, pk=None):\\n user = User.objects.get(email=request.user.email)\\n contactuserposts = ContactUser.objects.all()\\n contactuserpost = get_object_or_404(ContactUser, pk=pk)\\n if request.method == \\\"POST\\\":\\n contactuserpost.delete()\\n messages.success(request, 'This message has been successfully deleted.')\\n return redirect(user_profile)\\n return render(request, \\\"personalmessagedelete.html\\\", {'contactuserposts': contactuserposts})\",\n \"def unsubscribe(self, update, context):\\n # remove or update to the sqlite table.\\n chat = update.message.chat\\n self.db_manager.remove_user(chat.id)\\n self.logger.info(\\n 'Username: %s and chat_id: %s unsubscribed to the list.' % (chat.username, chat.id)\\n )\\n update.message.reply_text('You have successfully unsubscribed the notifications forever.')\",\n \"def remove(self, user_id):\\n pass\",\n \"def delete_user(self, userId):\\n\\n try:\\n query = \\\"delete from user where userId = {}\\\".format(userId)\\n print(query)\\n cur = self.con.cursor()\\n cur.execute(query)\\n self.con.commit()\\n\\n logger.info(\\\"Deleted\\\")\\n except Exception as e:\\n logger.error(\\\"Error occured at data deletion..\\\", e)\",\n \"def delete(self, user_id):\\n user = User.query.get(user_id)\\n\\n if user is None:\\n return mk_response(\\\"User does not exist\\\", 422)\\n\\n # check if the user is an admin and is the only one\\n admins = User.query.filter_by(admin=True).all()\\n if user.id == get_jwt_identity() and len(admins) == 1:\\n return mk_response(\\\"User is the only admin, there must \\\" +\\n \\\"be at least one admin in the system\\\", 422)\\n\\n user.delete()\\n\\n return mk_response(\\\"User '{}' has been deleted\\\".format(user.id))\",\n \"def del_user(self, name):\\n del self.users[irc.strings.IRCFoldedCase(modules.trim_nick(name))]\"\n]","isConversational":false},"negative_scores":{"kind":"list like","value":["0.7976897","0.7772087","0.770037","0.7479432","0.746371","0.746371","0.746371","0.7442444","0.7436067","0.7431538","0.73823726","0.7339518","0.7288484","0.7285742","0.7276345","0.725602","0.72481054","0.72303545","0.72023326","0.7187994","0.71587986","0.715419","0.70862156","0.7080586","0.7078395","0.7056995","0.70511746","0.7044443","0.70407254","0.70397425","0.7026253","0.7017078","0.7007012","0.70062476","0.69873387","0.697222","0.69706154","0.69596577","0.69583553","0.6952527","0.69321126","0.69251925","0.69092786","0.68972814","0.6880645","0.68753177","0.6874243","0.68577576","0.685436","0.6837124","0.68361646","0.6832816","0.68253016","0.68189454","0.67905515","0.67877454","0.6771872","0.6762962","0.6748143","0.6735661","0.67311746","0.6719457","0.6686512","0.66838914","0.6683194","0.667591","0.66733074","0.6670615","0.6662184","0.6659396","0.6653305","0.6638963","0.66380465","0.6629472","0.66256803","0.6622302","0.6620556","0.66169167","0.66159326","0.6615582","0.6604264","0.6603646","0.65988564","0.65838444","0.6577579","0.6576885","0.6570169","0.6569275","0.6565057","0.65518194","0.6547423","0.6545083","0.65208524","0.6520612","0.6515669","0.6506336","0.650091","0.650085","0.64980674","0.64937365"],"string":"[\n \"0.7976897\",\n \"0.7772087\",\n \"0.770037\",\n \"0.7479432\",\n \"0.746371\",\n \"0.746371\",\n \"0.746371\",\n \"0.7442444\",\n \"0.7436067\",\n \"0.7431538\",\n \"0.73823726\",\n \"0.7339518\",\n \"0.7288484\",\n \"0.7285742\",\n \"0.7276345\",\n \"0.725602\",\n \"0.72481054\",\n \"0.72303545\",\n \"0.72023326\",\n \"0.7187994\",\n \"0.71587986\",\n \"0.715419\",\n \"0.70862156\",\n \"0.7080586\",\n \"0.7078395\",\n \"0.7056995\",\n \"0.70511746\",\n \"0.7044443\",\n \"0.70407254\",\n \"0.70397425\",\n \"0.7026253\",\n \"0.7017078\",\n \"0.7007012\",\n \"0.70062476\",\n \"0.69873387\",\n \"0.697222\",\n \"0.69706154\",\n \"0.69596577\",\n \"0.69583553\",\n \"0.6952527\",\n \"0.69321126\",\n \"0.69251925\",\n \"0.69092786\",\n \"0.68972814\",\n \"0.6880645\",\n \"0.68753177\",\n \"0.6874243\",\n \"0.68577576\",\n \"0.685436\",\n \"0.6837124\",\n \"0.68361646\",\n \"0.6832816\",\n \"0.68253016\",\n \"0.68189454\",\n \"0.67905515\",\n \"0.67877454\",\n \"0.6771872\",\n \"0.6762962\",\n \"0.6748143\",\n \"0.6735661\",\n \"0.67311746\",\n \"0.6719457\",\n \"0.6686512\",\n \"0.66838914\",\n \"0.6683194\",\n \"0.667591\",\n \"0.66733074\",\n \"0.6670615\",\n \"0.6662184\",\n \"0.6659396\",\n \"0.6653305\",\n \"0.6638963\",\n \"0.66380465\",\n \"0.6629472\",\n \"0.66256803\",\n \"0.6622302\",\n \"0.6620556\",\n \"0.66169167\",\n \"0.66159326\",\n \"0.6615582\",\n \"0.6604264\",\n \"0.6603646\",\n \"0.65988564\",\n \"0.65838444\",\n \"0.6577579\",\n \"0.6576885\",\n \"0.6570169\",\n \"0.6569275\",\n \"0.6565057\",\n \"0.65518194\",\n \"0.6547423\",\n \"0.6545083\",\n \"0.65208524\",\n \"0.6520612\",\n \"0.6515669\",\n \"0.6506336\",\n \"0.650091\",\n \"0.650085\",\n \"0.64980674\",\n \"0.64937365\"\n]","isConversational":false},"document_score":{"kind":"string","value":"0.71425474"},"document_rank":{"kind":"string","value":"22"}}},{"rowIdx":93292,"cells":{"query":{"kind":"string","value":"runs multiple experiments that calculates the precision and recall of the comparison algorithms' results"},"document":{"kind":"string","value":"def run_multiple_simple_experiments(self, path_to_csv_files, algorithms, thresholds):\n starting_time = time.time()\n print(\"started simple precision recall image_matching_module\")\n for algorithm in algorithms:\n start_file_time = time.time()\n print(\"started algorithm:\", algorithm.get_algorithm_name())\n self.run_single_simple_experiment(path_to_csv_files, algorithm, thresholds)\n finish_file_time = time.time()\n print(\"finished algorithm:\", algorithm.get_algorithm_name())\n print(\"time took:\", (finish_file_time - start_file_time) / 60, \"minutes\")\n finish_time = time.time()\n hours = int((finish_time - starting_time) // 3600)\n minutes = (finish_time - starting_time) / 60\n print(\"finished simple precision recall image_matching_module\")\n print(\"time image_matching_module took:\", hours, \"hours and\", minutes, \"minutes\")"},"metadata":{"kind":"string","value":"{\n \"objective\": {\n \"self\": [],\n \"paired\": [],\n \"triplet\": [\n [\n \"query\",\n \"document\",\n \"negatives\"\n ]\n ]\n }\n}"},"negatives":{"kind":"list like","value":["def run_experiments():\n if True: # Change to False when done finding max_scoring_num_rolls\n six_sided_max = max_scoring_num_rolls(six_sided)\n print('Max scoring num rolls for six-sided dice:', six_sided_max)\n rerolled_max = max_scoring_num_rolls(reroll(six_sided))\n print('Max scoring num rolls for re-rolled dice:', rerolled_max)\n\n if False: # Change to True to test always_roll(8)\n print('always_roll(8) win rate:', average_win_rate(always_roll(8)))\n\n if False: # Change to True to test bacon_strategy\n print('bacon_strategy win rate:', average_win_rate(bacon_strategy))\n\n if False: # Change to True to test swap_strategy\n print('swap_strategy win rate:', average_win_rate(swap_strategy))\n\n \"*** You may add additional experiments as you wish ***\"","def run_experiments():\n if True: # Change to False when done finding max_scoring_num_rolls\n six_sided_max = max_scoring_num_rolls(six_sided)\n print('Max scoring num rolls for six-sided dice:', six_sided_max)\n four_sided_max = max_scoring_num_rolls(four_sided)\n print('Max scoring num rolls for four-sided dice:', four_sided_max)\n\n if False: # Change to True to test always_roll(8)\n print('always_roll(8) win rate:', average_win_rate(always_roll(8)))\n\n if True: # Change to True to test bacon_strategy\n print('bacon_strategy win rate:', average_win_rate(bacon_strategy))\n\n if True: # Change to True to test swap_strategy\n print('swap_strategy win rate:', average_win_rate(swap_strategy))\n\n if True:\n print('final_strategy win rate:', average_win_rate(final_strategy))\n\n \"*** You may add additional experiments as you wish ***\"","def run_experiments():\n if False: # Change to False when done finding max_scoring_num_rolls\n six_sided_max = max_scoring_num_rolls(six_sided)\n print('Max scoring num rolls for six-sided dice:', six_sided_max)\n\n if False: # Change to True to test always_roll(8)\n print('always_roll(8) win rate:', average_win_rate(always_roll(8)))\n\n if False: # Change to True to test bacon_strategy\n print('bacon_strategy win rate:', average_win_rate(bacon_strategy))\n\n if False: # Change to True to test swap_strategy\n print('swap_strategy win rate:', average_win_rate(swap_strategy))\n\n if True: # Change to True to test final_strategy\n print('final_strategy win rate:', average_win_rate(final_strategy))\n\n \"*** You may add additional experiments as you wish ***\"","def main():\n run_test_summary1a()\n run_test_summary1c()\n run_test_summary1c()","def run_experiments():\r\n result = eval_strategy_range(always_roll, 1, 10)\r\n print('Best always_roll strategy:', result)\r\n\r\n if True: # Change to True when ready to test make_comeback_strategy\r\n result = eval_strategy_range(make_comeback_strategy, 5, 15)\r\n print('Best comeback strategy:', result)\r\n\r\n if True: # Change to True when ready to test make_mean_strategy\r\n result = eval_strategy_range(make_mean_strategy, 1, 10)\r\n print('Best mean strategy:', result)\r\n\r\n \"*** You may add additional experiments here if you wish ***\"","def all_experiments():\n elo_explain_experiments()\n alpha_beta_experiments()\n mtcs_experiments()","def run_experiments():\n nets = []\n nets.extend(shallow(n=1))\n nets.extend(basic_conv(n=1))\n nets.extend(omit_FC(n=1))\n nets.extend(dbl_conv(n=1, activation_fn=sigmoid))\n # omitted, but still interesting: regularized_dbl_conv()\n nets.extend(dbl_conv_relu(n=1))\n nets.extend(expanded_data(n=1, FC_n=100))\n nets.extend(expanded_data(n=1, FC_n=300))\n nets.extend(expanded_data(n=1, FC_n=1000))\n nets.extend(expanded_data_double_fc(n=1, FC_n=100)) \n nets.extend(expanded_data_double_fc(n=1, FC_n=300))\n nets.extend(expanded_data_double_fc(n=1, FC_n=1000))\n nets.extend(double_fc_dropout(0.5, 0.5, 0.5, n=5))\n # plot the erroneous digits in the ensemble of nets just trained\n error_locations, erroneous_predictions = ensemble(nets)\n plt = plot_errors(error_locations, erroneous_predictions)\n plt.savefig(\"ensemble_errors.png\")\n # plot the filters learned by the first of the nets just trained\n plt = plot_filters(nets[0], 0, 5, 4)\n plt.savefig(\"net_full_layer_0.png\")\n plt = plot_filters(nets[0], 1, 8, 5)\n plt.savefig(\"net_full_layer_1.png\")","def run_experiments(size, approach):\n\n # Create folder (if needed) where to store query evaluation logs and raw results\n if not os.path.isdir(f\"{path_reports}/eval-query-{current_run}\"):\n shell(f\"mkdir -p {path_reports}/eval-query-{current_run}\")\n\n # Create folder (if needed) where to store store evaluation logs and raw results\n if not os.path.isdir(f\"{path_reports}/eval-store-{current_run}\"):\n shell(f\"mkdir -p {path_reports}/eval-store-{current_run}\")\n\n # Helper function computing the path of a file/folder for the given query evaluation iteration \n def query_path(iteration, filename=None):\n folder = size[\"id\"] + \"_\" + str(iteration)\n return path_reports + \"/eval-query-\" + current_run + \"/\" + folder + (\"/\" + filename if filename != None else \"\") \n\n # Helper function computing the path of a file/folder for the given store evaluation test / cpu setting\n def store_path(test, cpu, filename=None):\n folder = size[\"id\"] + \"_\" + approach[\"id\"] + \"_\" + test + (\"_\" + cpu[\"id\"] if cpu != None else \"\")\n return path_reports + \"/eval-store-\" + current_run + \"/\" + folder + (\"/\" + filename if filename != None else \"\")\n \n # Determine whether partial traces and named graphs are supported\n partial = approach[\"supports_partial\"]\n graphs = approach[\"supports_graphs\"]\n \n # Skip setting if all data is available (check for presence of log files - delete them to repeat test)\n may_skip = (not sp_enable or not partial or not graphs or os.path.isfile(store_path(\"sp\", None, \"eval.log\")))\n may_skip = may_skip and (not sf_enable or os.path.isfile(store_path(\"sf\", None, \"eval.log\")))\n if query_enable and approach[\"id\"] == query_approach_id:\n for i in range(0, query_num_iterations):\n may_skip = may_skip and os.path.isfile(query_path(i, \"eval.log\"))\n for cpu in cpus:\n may_skip = may_skip and (not pp_enable or not partial or not graphs or os.path.isfile(store_path(\"pp\", cpu, \"eval.log\")))\n may_skip = may_skip and (not pf_enable or os.path.isfile(store_path(\"pf\", cpu, \"eval.log\")))\n if may_skip:\n return\n\n # Delete (if needed) and extract again the repository from its .tar.xz file, so to work on a clean repository (at the end of this test, the repository is no more clean)\n path_dump = path_data + \"/\" + size[\"id\"] + \"_\" + approach[\"id\"] + \".tar.lz\"\n path_repo = path_repos + \"/\" + size[\"id\"] + \"_\" + approach[\"id\"]\n if not os.path.isfile(path_dump):\n log(f\"Missing required file {path_dump}\")\n sys.exit()\n if os.path.isdir(path_repo):\n shell(f\"rm -Rf {path_repo}\")\n shell(f\"{cmd_plzip} -kdc {path_dump} | tar xf - -C {path_repos}\")\n\n # Locate the repository URL\n repo_url = f\"http://localhost:{server_port}/repositories/promo\"\n \n # Query test (if enabled)\n if query_enable and approach[\"id\"] == query_approach_id:\n for i in range(0, query_num_iterations):\n if not os.path.isfile(query_path(i, \"eval.log\")):\n shell(f\"mkdir -p {query_path(i)}\")\n shell(f\"taskset -a {query_taskset} {cmd_graphdb} start {path_repo}\")\n shell(f\"taskset -a {query_taskset} {cmd_mokip} eval-query -w -u {repo_url} -q {path_queries} -r {query_path(i)} \" +\n f\"| tee {query_path(i, 'eval.log')}\")\n shell(f\"taskset -a {query_taskset} {cmd_graphdb} stop {path_repo}\")\n\n # Sequential Partial test (to assess store times per trace and their components)\n if sp_enable and partial and graphs and not os.path.isfile(store_path(\"sp\", None, \"eval.log\")):\n shell(f\"mkdir -p {store_path('sp', None)}\")\n shell(f\"taskset -a {sp_taskset} {cmd_graphdb} start {path_repo}\")\n shell(f\"taskset -a {sp_taskset} {cmd_mokip} eval-store -d {path_data}/traces_pp.jsonl.gz \" + \n f\"-u {repo_url} -i {approach['inference']} -U REPLACE_GRAPH_PROTOCOL \" + \n f\"-o {path_ontology} --trace-namespace '{namespace_trace}' --graph-namespace '{namespace_graph}' \" +\n f\"-T {timeout} -r {store_path('sp', None)} -t 1 -w 50 -p -D \" + \n f\"| tee {store_path('sp', None, 'eval.log')}\")\n shell(f\"taskset -a {sp_taskset} {cmd_graphdb} stop {path_repo}\")\n\n # Parallel Partial (to assess throughput, varying # of CPU cores)\n for cpu in cpus:\n if pp_enable and partial and graphs and not os.path.isfile(store_path(\"pp\", cpu, \"eval.log\")):\n shell(f\"mkdir -p {store_path('pp', cpu)}\")\n shell(f\"taskset -a {cpu['taskset']} {cmd_graphdb} start {path_repo}\")\n shell(f\"taskset -a {cpu['taskset']} {cmd_mokip} eval-store -d {path_data}/traces_pp.jsonl.gz \" +\n f\"-u {repo_url} -i {approach['inference']} -U REPLACE_GRAPH_PROTOCOL \" + \n f\"-o {path_ontology} --trace-namespace '{namespace_trace}' --graph-namespace '{namespace_graph}' \" + \n f\"-T {timeout} -r {store_path('pp', cpu)} -t {max(2, cpu['num_threads'])} -w 50 -p -D \" + \n f\"| tee {store_path('pp', cpu, 'eval.log')}\")\n shell(f\"taskset -a {cpu['taskset']} {cmd_graphdb} stop {path_repo}\")\n\n # Sequential Full test (to assess store times per trace and their components)\n if sf_enable and not os.path.isfile(store_path(\"sf\", None, \"eval.log\")):\n shell(f\"mkdir -p {store_path('sf', None)}\")\n shell(f\"taskset -a {sf_taskset} {cmd_graphdb} start {path_repo}\")\n shell(f\"taskset -a {sf_taskset} {cmd_mokip} eval-store -d {path_data}/traces_sf.jsonl.gz \" + \n f\"-u {repo_url} -i {approach['inference']} -U {'APPEND' if graphs else 'APPEND_DEFAULT_GRAPH'} \" +\n f\"-o {path_ontology} --trace-namespace '{namespace_trace}' --graph-namespace '{namespace_graph}' \" +\n f\"-T {timeout} -r {store_path('sf', None)} -t 1 -w 50 \" + \n f\"| tee {store_path('sf', None, 'eval.log')}\")\n shell(f\"taskset -a {sf_taskset} {cmd_graphdb} stop {path_repo}\")\n\n # Parallel Full (to assess throughput where data is also deleted, varying # of CPU cores)\n for cpu in cpus:\n if pf_enable and not os.path.isfile(store_path(\"pf\", cpu, \"eval.log\")):\n update = \"APPEND\" if graphs else \"APPEND_DEFAULT_GRAPH\"\n shell(f\"mkdir -p {store_path('pf', cpu)}\")\n shell(f\"taskset -a {cpu['taskset']} {cmd_graphdb} start {path_repo}\")\n shell(f\"taskset -a {cpu['taskset']} {cmd_mokip} eval-store -d {path_data}/traces_pf_{cpu['id']}.jsonl.gz \" +\n f\"-u {repo_url} -i {approach['inference']} -U {'APPEND' if graphs else 'APPEND_DEFAULT_GRAPH'} \" + \n f\"-o {path_ontology} --trace-namespace '{namespace_trace}' --graph-namespace '{namespace_graph}' \" +\n f\"-T {timeout} -r {store_path('pf', cpu)} -t {max(2, cpu['num_threads'])} -w 50 \" + \n f\"| tee {store_path('pf', cpu, 'eval.log')}\")\n shell(f\"taskset -a {cpu['taskset']} {cmd_graphdb} stop {path_repo}\")\n\n # Drop the repository (both to save space and since it is not clean anymore)\n shell(f\"rm -Rf {path_repo}\")","def experiment1_outliers():\n\tdata_folder = \"ckan_subset/prepared_learnset/\"\n\ttest_folder = 'ckan_subset/testset/xml_csv/'\n\tgm = Graph_Maker()\n\tgm.store()\n\trounds = 5\n\tx = [\"Fingerprint\", \"Syntax Feature Model\", \"Word2Vec Matcher\"]\n\t\n\tnumber_of_classes = 15\n\texamples_per_class = 0\n\taccuracies = []\n\tprecisions = []\n\trecalls = []\n\tfmeasures = []\n\tsf_main = Storage_Files(data_folder, classes)\n\ttmp_acc = []\n\ttmp_prec = []\n\ttmp_rec = []\n\ttmp_fmeasure = []\n\ttotal_actual = []\n\ttotal_predicted = []\n\n\tfor i in range(0, rounds):\n\t\tprint(\"Fingerprint\")\n\t\t# --- Fingerprint\n\t\tccc = Column_Classification_Config()\n\t\tccc.add_feature('feature_main', 'Fingerprint', [sf_main, number_of_classes, examples_per_class, False, False])\n\n\t\tccc.add_matcher('matcher', 'Fingerprint_Matcher', {'feature_main': 'fingerprint'}) # main classifier\n\t\tsm = Schema_Matcher(ccc)\n\t\tactual, predicted = execute_test_ckan(sm, test_folder, False)\n\t\ttotal_actual += actual\n\t\ttotal_predicted += predicted\n\t\taccuracy = accuracy_score(actual, predicted)\n\t\ttmp_acc.append(accuracy)\n\t\ttmp_prec.append(precision(actual, predicted))\n\t\ttmp_rec.append(recall(actual, predicted))\n\t\ttmp_fmeasure.append(f_measure(actual, predicted))\n\n\taccuracies.append( round(sum(tmp_acc) / float(rounds), 2) )\n\tprecisions.append( round(sum(tmp_prec) / float(rounds), 2) )\n\trecalls.append( round(sum(tmp_rec) / float(rounds), 2) )\n\tfmeasures.append(round(sum(tmp_fmeasure) / float(rounds), 2))\n\tclassnames = list(set(get_class_names(total_actual) + get_class_names(total_predicted)))\n\tcm = confusion_matrix(total_actual, total_predicted, labels=classnames)\n\t#gm.plot_confusion_matrix(cm, classnames, normalize=True)\n\t\n\ttmp_acc = []\n\ttmp_prec = []\n\ttmp_rec = []\n\ttmp_fmeasure = []\n\ttotal_actual = []\n\ttotal_predicted = []\n\tfor i in range(0, rounds):\n\t\tprint(\"SFM\")\n\t\t# --- Syntax Feature Model\n\t\tccc = Column_Classification_Config()\n\t\tccc.add_feature('feature_main', 'Syntax_Feature_Model', [sf_main, 1, 0, False, False])\n\n\t\tccc.add_matcher('matcher', 'Syntax_Matcher', {'feature_main': 'syntax'}) # main classifier\n\t\tsm = Schema_Matcher(ccc)\n\t\tactual, predicted = execute_test_ckan(sm, test_folder, False)\n\t\ttotal_actual += actual\n\t\ttotal_predicted += predicted\n\t\taccuracy = accuracy_score(actual, predicted)\n\t\ttmp_acc.append(accuracy)\n\t\ttmp_prec.append(precision(actual, predicted))\n\t\ttmp_rec.append(recall(actual, predicted))\n\t\ttmp_fmeasure.append(f_measure(actual, predicted))\n\n\taccuracies.append( round(sum(tmp_acc) / float(rounds), 2) )\n\tprecisions.append( round(sum(tmp_prec) / float(rounds), 2) )\n\trecalls.append( round(sum(tmp_rec) / float(rounds), 2) )\n\tfmeasures.append(round(sum(tmp_fmeasure) / float(rounds), 2))\n\tclassnames = list(set(get_class_names(total_actual) + get_class_names(total_predicted)))\n\tcm = confusion_matrix(total_actual, total_predicted, labels=classnames)\n\t#gm.plot_confusion_matrix(cm, classnames, normalize=True)\n\n\ttmp_acc = []\n\ttmp_prec = []\n\ttmp_rec = []\n\ttmp_fmeasure = []\n\ttotal_actual = []\n\ttotal_predicted = []\n\tfor i in range(0, rounds):\n\t\tprint(\"W2V\")\n\t\t# --- Word2Vec Matcher\n\t\tccc = Column_Classification_Config()\n\t\tccc.add_feature('feature_main', 'Corpus', [sf_main, number_of_classes, examples_per_class, False, False])\n\n\t\tccc.add_matcher('matcher', 'Word2Vec_Matcher', {'feature_main': 'corpus'}) # main classifier\n\t\tsm = Schema_Matcher(ccc)\n\t\tactual, predicted = execute_test_ckan(sm, test_folder, False)\n\t\ttotal_actual += actual\n\t\ttotal_predicted += predicted\n\t\taccuracy = accuracy_score(actual, predicted)\n\t\ttmp_acc.append(accuracy)\n\t\ttmp_prec.append(precision(actual, predicted))\n\t\ttmp_rec.append(recall(actual, predicted))\n\t\ttmp_fmeasure.append(f_measure(actual, predicted))\n\n\taccuracies.append( round(sum(tmp_acc) / float(rounds), 2) )\n\tprecisions.append( round(sum(tmp_prec) / float(rounds), 2) )\n\trecalls.append( round(sum(tmp_rec) / float(rounds), 2) )\n\tfmeasures.append(round(sum(tmp_fmeasure) / float(rounds), 2))\n\tclassnames = list(set(get_class_names(total_actual) + get_class_names(total_predicted)))\n\tcm = confusion_matrix(total_actual, total_predicted, labels=classnames)\n\t#gm.plot_confusion_matrix(cm, classnames, normalize=True)\n\n\tgm.add_x(x)\n\t# accuracies = [0.4, 0.4, 0.4]\n\t# precisions = [0.5, 0.5, 0.5]\n\t# recalls = [0.62, 0.62, 0.62]\n\t# fmeasures = [0.23, 0.23, 0.28]\n\tgm.append_y(accuracies)\n\tgm.append_y(precisions)\n\tgm.append_y(recalls)\n\tgm.append_y(fmeasures)\n\tgm.store()\n\tsubtitle = \"Scores were averaged over \" + str(rounds) + \" tests with \" + str(len(classes)) + \" classes. \" + \\\n\t\"Number of simulated columns per class: \" + str(number_of_classes)\n\tlabels = [\"Accuracy\", \"Precision\", \"Recall\", \"F-Measure\"]\n\tgm.plot_bar_n(\"Matcher Type\", \"Score\", \"Accuracy of Matchers\", labels, subtitle=subtitle)","def run_Experiment(DP = None, QL = None):\n\n # Path information\n output_path, exp_num = create_new_dir() #dirs Exp/1, Exp/2, ...\n DP_path = join(output_path,'DP') #dirs Exp/1/DP\n QL_path = join(output_path,'QL') #dirs Exp/1/QL\n print(\"************ Exp \", exp_num, \"************ \\n\")\n\n # Exp_summary_data\n method = get_method_str(DP, QL)\n exp_summary = [str(exp_num), method]\n\n\n # Run DP\n if DP != None:\n print(\"In Runner: Executing DP !!\")\n\n prob_file = DP[0]\n createFolder(DP_path)\n # output_params = [V_so, mean, variance, bad_count]\n output_params = run_DP(setup_grid_params, prob_file, output_file, DP_path, threshold = threshold)\n\n \"\"\"CHANGE ARGUMENT if return order of setup_grid() is changed\"\"\"\n input_params = setup_grid_params[9].copy()\n input_params.append(prob_file)\n\n exp_summary = append_params_to_summary(exp_summary, input_params, output_params)\n append_summary_to_summaryFile('Experiments/Exp_summary.csv', exp_summary)\n print(\"In Runner: Executing DP Finished!!\")\n\n # Run QL\n if QL != None:\n print(\"In Runner: Executing QL !!\")\n\n QL_params = QL\n createFolder(QL_path)\n output_parameters_all_cases = run_QL(setup_grid_params, QL_params, QL_path, exp_num)\n # run_QL(setup_grid_params, QL_params, QL_path)\n\n print(\"In Runner: Executing QL Finished !!\")","def evaluate_model(args, eval_runs, warm_runs, metrics=['psnr', 'ssim', 'fps']):\n upsampler = Upsampler(args)\n if warm_runs > 0:\n print(\"Warming up for evaluation\")\n for i in range(warm_runs):\n print(\"Performing warm-up run\", str(i+1))\n for sequence in ['foliage', 'walk', 'calendar', 'city']:\n bix_dir = os.path.join(VID4_DIR, 'BIx4', sequence)\n upsampler.run_dir(bix_dir, reset=False)\n \n time = 0.\n psnrs = []\n ssims = []\n for i in range(eval_runs):\n run_psnrs = []\n run_ssims = []\n print(\"Performing evaluation run\", str(i+1))\n for sequence in ['foliage', 'walk', 'calendar', 'city']:\n bix_dir = os.path.join(VID4_DIR, 'BIx4', sequence)\n gt_dir = os.path.join(VID4_DIR, 'GT', sequence)\n print(\"Evaluating on\", bix_dir)\n time += upsampler.run_dir(bix_dir, reset=False)\n vid_psnrs, vid_ssims = _eval_sr_perf(os.path.join(bix_dir, 'up'), gt_dir)\n run_psnrs += vid_psnrs\n run_ssims += vid_ssims\n if i == eval_runs-1:\n with open(os.path.join(upsampler.get_model_dir(), \"psnr.txt\"), \"w\") as f:\n f.writelines(str(psnr) + '\\n' for psnr in run_psnrs)\n with open(os.path.join(upsampler.get_model_dir(), \"ssim.txt\"), \"w\") as f:\n f.writelines(str(ssim) + '\\n' for ssim in run_ssims)\n psnrs += run_psnrs\n ssims += run_ssims\n\n fps = VID4_LENGTH/ (time/eval_runs)\n return Performance(psnr=psnrs, ssim=ssims, fps=fps)","def basic_experiments(k, p, num_iter, transforms, train_path, test_path):\n start_time = time.time()\n print(\"Generating advice\")\n with open(train_path, \"r\") as f:\n train_data = f.readlines()\n advice = generate_advice(train_data, transforms)\n print(\"Done: \" + str(time.time() - start_time) + \"\\n\")\n\n print(\"Generating real count for test data\")\n with open(test_path, \"r\") as f:\n test_data = f.readlines()\n real_count = generate_advice(test_data, transforms)\n print(\"Done: \" + str(time.time() - start_time))\n actual = [x.moment(p) for x in real_count]\n print(\"Actual moments:\")\n print(actual)\n\n print(\"Estimating using noisy advice:\")\n est_with_advice = []\n for i in range(num_iter):\n seed = str(random.randint(1, 2**64 + 1)) + str(random.randint(1, 2**64 + 1))\n this_hash_exp = lambda x: hash_exp(x, seed)\n est_with_advice.append(\n estimate_using_advice(test_data, transforms, k, p, this_hash_exp,\n advice))\n print(\"Done: \" + str(time.time() - start_time))\n print(est_with_advice)\n\n print(\"Estimating using perfect advice:\")\n est_perfect_advice = []\n for i in range(num_iter):\n seed = str(random.randint(1, 2**64 + 1)) + str(random.randint(1, 2**64 + 1))\n this_hash_exp = lambda x: hash_exp(x, seed)\n est_perfect_advice.append(\n estimate_using_advice(test_data, transforms, k, p, this_hash_exp,\n real_count))\n print(\"Done: \" + str(time.time() - start_time))\n print(est_perfect_advice)\n\n print(\"Estimating using PPSWOR:\")\n est_ppswor = []\n for i in range(num_iter):\n cur = []\n for x in real_count:\n cur.append(ppswor_estimate_moment(x.counts.items(), k, p))\n est_ppswor.append(cur)\n print(\"Done: \" + str(time.time() - start_time))\n print(est_ppswor)\n\n print(\"Estimating using ell_2 PPSWOR:\")\n est_ppswor_l2 = []\n for i in range(num_iter):\n cur = []\n for x in real_count:\n cur.append(ppswor_estimate_moment(x.counts.items(), k, p, 2))\n est_ppswor_l2.append(cur)\n print(\"Done: \" + str(time.time() - start_time))\n print(est_ppswor_l2)","def run_all_tests(self) -> None:\n self.run_trt_precision_tests()\n logging.info(\"Check analysis result at: %s\", self._output_dir)","def run_and_evaluate():\n tsp_problems = read_all_problems()\n # Empty list of metrics\n results = []\n for problem in tqdm.tqdm(tsp_problems):\n # As random factors are involved repeat experiments a couple of times\n best_routes_base = []\n best_routes_af = []\n best_routes_ms = []\n base_times = []\n af_times = []\n ms_times = []\n for i in range(10):\n # Base solution\n start_time = timeit.default_timer()\n best_route_base = solve_tsp_basic(problem)\n base_time = timeit.default_timer() - start_time\n best_routes_base.append(Fitness(route=best_route_base).route_distance())\n base_times.append(base_time)\n\n # AF clustering solution\n start_time = timeit.default_timer()\n best_route_af = solve_tsp_affinity_propagation(problem)\n af_time = timeit.default_timer() - start_time\n best_routes_af.append(Fitness(route=best_route_af).route_distance())\n af_times.append(af_time)\n\n # MS solution\n start_time = timeit.default_timer()\n best_route_ms = solve_mean_shift(problem)\n ms_time = timeit.default_timer() - start_time\n best_routes_ms.append(Fitness(route=best_route_ms).route_distance())\n ms_times.append(ms_time)\n\n results.append(\n {\n \"problem name\": problem.name,\n \"optimal solution\": find_route_optimal_route_length(problem),\n \"baseline tour length\": mean(best_routes_base),\n \"af clustering tour length\": mean(best_routes_af),\n \"ms clustering tour length\": mean(best_routes_ms),\n \"baseline algorithm time\": mean(base_times),\n \"af clustering algorithm time\": mean(af_times),\n \"ms clustering algorithm time\": mean(ms_times),\n }\n )\n # Create dataframe and safe results\n df = pd.DataFrame(results)\n df.to_csv(\"results.csv\", index=False)\n return df","def run_analyses(y_predict_train, y_train, y_predict, y_test):\n # calculate metrics\n _, training_error = output_error(y_predict_train, y_train)\n (precision, recall, f1, _), testing_error = output_error(y_predict, y_test)\n \n # print out metrics\n print 'Average Precision:', np.average(precision)\n print 'Average Recall:', np.average(recall)\n print 'Average F1:', np.average(f1)\n print 'Training Error:', training_error\n print 'Testing Error:', testing_error","def test_compare_classifiers_performance_changing_k_vis_api(experiment_to_use):\n experiment = experiment_to_use\n probabilities = experiment.probabilities\n viz_outputs = (\"pdf\", \"png\")\n with TemporaryDirectory() as tmpvizdir:\n for viz_output in viz_outputs:\n vis_output_pattern_pdf = tmpvizdir + f\"/*.{viz_output}\"\n visualize.compare_classifiers_performance_changing_k(\n [probabilities, probabilities],\n experiment.ground_truth,\n experiment.ground_truth_metadata,\n experiment.output_feature_name,\n top_k=3,\n labels_limit=0,\n model_namess=[\"Model1\", \"Model2\"],\n output_directory=tmpvizdir,\n file_format=viz_output,\n )\n figure_cnt = glob.glob(vis_output_pattern_pdf)\n assert 1 == len(figure_cnt)","def main():\n run_simulation(spectral=False, ml=False, num_procs=1)\n run_simulation(spectral=True, ml=False, num_procs=1)\n run_simulation(spectral=False, ml=True, num_procs=1)\n run_simulation(spectral=True, ml=True, num_procs=1)\n run_simulation(spectral=False, ml=True, num_procs=10)\n run_simulation(spectral=True, ml=True, num_procs=10)","def evaluate(cfg: DictConfig):\n\n experiments = cfg.get('experiment_type', f'{cfg.model.name}_only')\n fixed_t0 = cfg.get('fixed_t0', False)\n ext = '_fixedT0' if fixed_t0 else ''\n\n base_dir = cfg.device.root\n datasource = cfg.datasource.name\n\n if experiments == 'ablations':\n models = {\n 'FluxRGNN': ['final',\n 'final_without_encoder',\n 'final_without_boundary'],\n 'LocalLSTM': ['final']\n }\n elif experiments == 'final':\n models = {\n 'FluxRGNN': ['final'],\n 'GAM': ['final'],\n 'HA': ['final'],\n 'GBT': ['final']\n }\n else:\n m = cfg.model.name\n year = cfg.datasource.test_year\n\n # find all experiments available for this model, datasource and test year\n result_dir = osp.join(base_dir, 'results', datasource, m, f'test_{year}')\n models = {\n m : [ f.name for f in os.scandir(result_dir) if f.is_dir() ]\n }\n\n\n # thresholds for binary classification metrics\n if cfg.datasource.name == 'abm':\n thresholds = [0.0019, 0.0207]\n else:\n thresholds = [0, 10, 20]\n\n rmse_per_hour = []\n mae_per_hour = []\n pcc_per_hour = []\n bin_per_hour = []\n\n rmse_per_night = []\n mae_per_night = []\n\n output_dir = osp.join(base_dir, 'results', datasource, f'performance_evaluation{ext}', experiments)\n os.makedirs(output_dir, exist_ok=True)\n\n counter = 0\n\n for m, dirs in models.items():\n print(f'evaluate {m}')\n\n for d in dirs:\n result_dir = osp.join(base_dir, 'results', datasource, m, f'test_{cfg.datasource.test_year}', d)\n\n # check if directory exists\n if os.path.isdir(result_dir):\n results, model_cfg = load_cv_results(result_dir, trials=cfg.task.repeats, ext=ext)\n\n df_prep = pd.read_csv(osp.join(base_dir, 'data', 'preprocessed',\n f'{model_cfg[\"t_unit\"]}_{model_cfg[\"model\"][\"edge_type\"]}_ndummy={model_cfg[\"datasource\"][\"n_dummy_radars\"]}',\n datasource, cfg.season, str(cfg.datasource.test_year), 'dynamic_features.csv'))\n tidx2night = dict(zip(df_prep.tidx, df_prep.nightID))\n\n rmse_per_hour.append(compute_rmse(m, d, results, tidx2night, groupby=['horizon', 'trial'],\n threshold=0, km2=True, fixed_t0=fixed_t0))\n mae_per_hour.append(compute_mae(m, d, results, tidx2night, groupby=['horizon', 'trial'],\n threshold=0, km2=True, fixed_t0=fixed_t0))\n pcc_per_hour.append(compute_pcc(m, d, results, tidx2night, groupby=['horizon', 'trial'],\n threshold=0, km2=True, fixed_t0=fixed_t0))\n\n if fixed_t0:\n rmse_per_night.append(compute_rmse_per_night(m, d, results, tidx2night, groupby=['night_horizon', 'trial']))\n mae_per_night.append(compute_mae_per_night(m, d, results, tidx2night, groupby=['night_horizon', 'trial']))\n\n # compute binary classification measures\n for thr in thresholds:\n bin_per_hour.append(compute_bin(m, d, results, groupby=['horizon', 'trial'], threshold=thr, km2=True))\n\n counter += 1\n\n else:\n print(f'Experiment \"{d}\" for model \"{m}\" and datasource \"{datasource}\" is not available. '\n f'Use \"run_experiments.py model={m} datasource={datasource} +experiment={d}\" to run this experiment.')\n\n if counter > 0:\n rmse_per_hour = pd.concat(rmse_per_hour)\n rmse_per_hour.to_csv(osp.join(output_dir, f'rmse_per_hour.csv'))\n\n mae_per_hour = pd.concat(mae_per_hour)\n mae_per_hour.to_csv(osp.join(output_dir, f'mae_per_hour.csv'))\n\n pcc_per_hour = pd.concat(pcc_per_hour)\n pcc_per_hour.to_csv(osp.join(output_dir, f'pcc_per_hour.csv'))\n\n bin_per_hour = pd.concat(bin_per_hour)\n bin_per_hour.to_csv(osp.join(output_dir, f'bin_per_hour.csv'))\n\n if fixed_t0:\n rmse_per_night = pd.concat(rmse_per_night)\n rmse_per_night.to_csv(osp.join(output_dir, f'rmse_per_night.csv'))\n\n mae_per_night = pd.concat(mae_per_night)\n mae_per_night.to_csv(osp.join(output_dir, f'mae_per_night.csv'))","def three_experiments(two_experiments, one_experiment):","def main():\r\n algos = [merge_sort, quick_sort, heap_sort, radix_sort, bucket_sort_general]\r\n array_sizes = [5000, 10000, 15000, 20000, 50000, 75000, 100000, 150000]\r\n results = {algo.__name__: [] for algo in algos}\r\n for algo in algos:\r\n result = []\r\n for size in array_sizes:\r\n time = test(algo, size)\r\n result.append(time)\r\n results[algo.__name__] = result\r\n\r\n display_results(results, array_sizes)","def test_compare_classifier_performance_from_pred_vis_api(experiment_to_use):\n experiment = experiment_to_use\n prediction = experiment.predictions\n viz_outputs = (\"pdf\", \"png\")\n with TemporaryDirectory() as tmpvizdir:\n for viz_output in viz_outputs:\n vis_output_pattern_pdf = tmpvizdir + f\"/*.{viz_output}\"\n visualize.compare_classifiers_performance_from_pred(\n [prediction, prediction],\n experiment.ground_truth,\n experiment.ground_truth_metadata,\n experiment.output_feature_name,\n labels_limit=0,\n model_namess=[\"Model1\", \"Model2\"],\n output_directory=tmpvizdir,\n file_format=viz_output,\n )\n figure_cnt = glob.glob(vis_output_pattern_pdf)\n assert 1 == len(figure_cnt)","def run_tests(augmenter, images, nb_runs):\n results = np.zeros((nb_runs,))\n for i in range(nb_runs):\n start = time.time()\n augmenter.augment_batch(images)\n results[i] = time.time() - start\n print(\"Run %d: %.8fs\" % (i, results[i]))\n print(\"Mean: %.8fs\" % (results.mean(),))\n print(\"Sum: %.8fs\" % (results.sum(),))","def run_oracle_experiments(model, query_iterator, N, recalls=[1]):\n times_dict = dict()\n calls_dict = dict()\n probs = dict()\n keys = list()\n recall_break = max(recalls)\n for query, relevant_set in query_iterator:\n candidates_iterator = iter_candidates(model, query)\n times, calls = candidates_time_oracle_calls_count(\n candidates_iterator, relevant_set, recall_break)\n k = len(relevant_set)\n if k in times_dict:\n times_dict[k].append(times)\n calls_dict[k].append(calls)\n probs[k] += 1\n else:\n keys.append(k)\n times_dict[k] = [times]\n calls_dict[k] = [calls]\n probs[k] = 1\n \n keys = np.array(keys)\n probs = np.array([probs[k] for k in keys])/sum(probs.values())\n times_dict = {k:np.array(v).mean(axis=0) for k,v in times_dict.items()}\n calls_dict = {k:np.array(v).mean(axis=0) for k,v in calls_dict.items()}\n \n expected_times = oracle_expectation_by_recalls(recalls, keys, probs, times_dict)\n expected_calls = oracle_expectation_by_recalls(recalls, keys, probs, calls_dict)\n expected_calls_baseline = oracle_random_calls_expectation_by_recall(recalls, keys, probs, N)\n \n return expected_times, expected_calls, expected_calls_baseline","def run_experiments() :\n #%%\n target_size=(32,32)\n g_specs = {\n \"batch_size\" : [ 30 , 60, 100 ],\n \"learning_rate\" : [ 0.0002, 0.0003, 0.0005 ],\n \"drop_out_rate\" : [ 0.2, 0.25, 0.3 ],\n \"rescale_mode\" : [ \"max_q\" , \"max\", \"\" ]\n }\n\n model_traits = MODEL_TRAITS[\"model2\"].copy()\n tt_obj = model_traits[\"trainer_tester_class\"]( model_traits )\n del model_traits[\"trainer_tester_class\"]\n\n cnt = 0\n for batchs, lrate, do_rate, resc_mode in product( g_specs[\"batch_size\"],\n g_specs[\"learning_rate\"],\n g_specs[\"drop_out_rate\"],\n g_specs[\"rescale_mode\"] ) :\n\n tt_obj.model_traits.update( {\"batch_size\" : batchs,\n \"learning_rate\" : lrate,\n \"rescale_mode\" : resc_mode,\n \"drop_out_rate\" : do_rate } )\n\n train_4d, train_gt = tu.make_4d_arrays( images_dir=\"images/train\",\n target_size=target_size )\n\n test_4d, test_gt = tu.make_4d_arrays( images_dir=\"images/test\",\n target_size=target_size )\n\n data = {\"train_4d\" : train_4d,\n \"test_4d\" : test_4d,\n \"train_y\" : train_gt,\n \"test_y\" : test_gt}\n\n valid_accu_log, train_accu_log = tt_obj.train( model_traits, data,\n logl=100 )\n idx_v = int(np.argmax( valid_accu_log))\n idx_t = int(np.argmax( train_accu_log))\n\n model_traits.update({\"valid_accu_log\" : valid_accu_log,\n \"train_accu_log\" : train_accu_log,\n \"best_valid\" : max(valid_accu_log),\n \"best_valid_at\" : idx_v,\n \"train_at_best_valid\" : train_accu_log[idx_v],\n \"best_train\" : max(train_accu_log),\n \"best_train_at\": idx_t })\n\n #print(cnt, pformat(model_traits) )\n print( \"%d : best_train = %.4f, best_valid = %.4f\" % \\\n (cnt, max(train_accu_log), max(valid_accu_log) ))\n\n with open( \"exp_results_%d.json\" % cnt,\n \"wt\" , encoding=\"utf8\" ) as f_out :\n print( json.dumps( model_traits ), file=f_out)\n\n\n cnt += 1\n #%%","def run_multiple_advanced_experiments(self, path_to_csv_files, algorithms_weights_list, thresholds):\n starting_time = time.time()\n print(\"started advanced precision recall image_matching_module\")\n for algorithms_weights in algorithms_weights_list:\n start_file_time = time.time()\n\n first_run = True\n for algorithm, weight in algorithms_weights:\n if first_run:\n algorithm_name = algorithm.get_algorithm_name() + weight\n first_run = False\n else:\n algorithm_name = algorithm_name + \"_\" + algorithm.get_algorithm_name() + weight\n\n print(\"started algorithm:\", algorithm_name)\n self.run_single_advanced_experiment(path_to_csv_files, algorithms_weights, thresholds)\n finish_file_time = time.time()\n print(\"finished algorithm:\", algorithm_name)\n print(\"time took:\", (finish_file_time - start_file_time) / 60, \"minutes\")\n finish_time = time.time()\n hours = int((finish_time - starting_time) // 3600)\n minutes = (finish_time - starting_time) / 60\n print(\"finished advanced precision recall image_matching_module\")\n print(\"time image_matching_module took:\", hours, \"hours and\", minutes, \"minutes\")","def test_any(alg, repNum=1):\n\tperformance = alg.predict_perf(repNum=repNum)\n\tprint performance\n\talg.plot()\n\n\treturn performance","def test_compare_classifier_performance_from_prob_vis_api(experiment_to_use):\n experiment = experiment_to_use\n probability = experiment.probabilities\n viz_outputs = (\"pdf\", \"png\")\n with TemporaryDirectory() as tmpvizdir:\n for viz_output in viz_outputs:\n vis_output_pattern_pdf = tmpvizdir + f\"/*.{viz_output}\"\n visualize.compare_classifiers_performance_from_prob(\n [probability, probability],\n experiment.ground_truth,\n experiment.ground_truth_metadata,\n experiment.output_feature_name,\n top_n_classes=[0],\n labels_limit=0,\n model_namess=[\"Model1\", \"Model2\"],\n output_directory=tmpvizdir,\n file_format=viz_output,\n )\n figure_cnt = glob.glob(vis_output_pattern_pdf)\n assert 1 == len(figure_cnt)","def test_run(id_, args, kwargs, expected):\n # Remove results for test.\n results_path = expected['results_path']\n shutil.rmtree(results_path, ignore_errors=True)\n # Start code.\n # [future] attempts to run classification with n_jobs>1\n # global cnfg_default\n # sys.modules['cnfg_default'] = get_params(args[0], 'cnfg_default')\n # import cnfg_default\n # #from cnfg_default import custom_score_metric\n objects = pycnfg.run(oid='default', *args, **kwargs)\n tmp = {k: type(v).__name__ for k, v in objects.items()}\n print('OBJECTS:')\n print(tmp)\n # Compare results:\n # * Compare objects (keys and str of values).\n objects_ = expected['objects']\n objects = {k: type(v).__name__ for k, v in objects.items()}\n if objects != objects_:\n print(set(objects.items()) ^ set(objects_.items()))\n assert objects == objects_\n # for k, v in objects.items():\n # assert k in objects_\n # assert type(v).__name__ == objects_[k]\n # * Compare predictions csv(all available).\n pred_path = glob.glob(f\"{results_path}/models/*_pred.csv\")\n pred_path_ = glob.glob(expected['pred_path'])\n assert len(pred_path) == len(pred_path_)\n for act, exp in zip(sorted(pred_path), sorted(pred_path_)):\n file_diff(act, exp)\n assert filecmp.cmp(act, exp)\n # * Compare test logs.\n logs_path = glob.glob(f\"{results_path}/logs*/*_test.log\")[0]\n logs_path_ = expected['logs_path']\n file_diff(logs_path, logs_path_)\n assert filecmp.cmp(logs_path, logs_path_)\n # * Compare runs dataframe, non-universe columns.\n runs_path = f\"{results_path}/runs\"\n runs_path_ = expected['runs_path']\n df = runs_loader(runs_path)\n df_ = runs_loader(runs_path_)\n # First False/True for each element, then check all by columns.\n # col1 True\n # col2 False\n # dtype: bool\n df_diff = df.eq(df_).all()\n # Column names that are not equal.\n columns = sorted(list(df_diff[df_diff==False].dropna().index))\n # columns_eq = sorted(list(df_diff[df_diff==True].dropna().index))\n columns_ = expected['columns_diff']\n print('DIFF:\\n', columns)\n time.sleep(1)\n # assert columns == columns_\n # * Compare model.\n model_path = glob.glob(f\"{results_path}/models/*.model\")\n model_path_ = glob.glob(expected['model_path'])\n assert len(model_path) == len(model_path_)\n for act, exp in zip(sorted(model_path), sorted(model_path_)):\n assert filecmp.cmp(act, exp)\n return","def one_experiment():\n\n # set the name of the experiment\n now = datetime.datetime.now()\n experiment_id = str(now.day) + \"_\" + str(now.month) + \"_\" + str(now.hour) + \".\" + str(now.minute)\n experiment_name = 'overfit_' + str(experiment_id)\n\n # define if you want to use preprocessed data from file\n use_prep_data = False\n if use_prep_data:\n set_params(preproc_data_id='16_5_10.16.47')\n\n # define the changing parameter and its value\n changing_param_name = 'class_weights'\n changing_param_value = [{0: 15, 1: 85}]\n # {0:15, 1:85}]#, {0:4, 1:100}, {0:3, 1:100}, {0:2, 1:100}, {0:1, 1:100}] #[{0:1, 1:1}, {0:15, 1:85}]#\n\n features_to_use = ['user', 'countries', 'session', 'format', 'token']\n # set constant parameters\n set_params(use_word_emb=1)\n set_params(epochs=40)\n set_params(features_to_use=features_to_use)\n\n # save constant parameters to a new \"experiment_..\" filgithx+P@2ub\n save_constant_parameters(experiment_name, changing_param_name)\n\n # run experiment for every parameter value\n for value in changing_param_value:\n process = psutil.Process(os.getpid())\n print(\"-----MEMORY before starting experiment ------\", int(process.memory_info().rss/(8*10**3)), \"KB\")\n\n # update the parameter value\n set_params(class_weights_1=value)\n\n # update the model_id for this new model\n now = datetime.datetime.now()\n new_model_id = str(now.day) + \"_\" + str(now.month) + \"_\" + str(now.hour) + \".\" + str(now.minute) + \".\" + str(now.second)\n\n set_params(model_id=new_model_id)\n\n # evaluate the new model and save the results in the experiment file\n oneExperiment = Process(target=run_experiment, args=(experiment_name,\n new_model_id, changing_param_name, value,))\n oneExperiment.start()\n oneExperiment.join()","def compute_and_visualize_coupling_experiments(experiments, results, seed,\n **kwargs):\n (logits_1,\n logits_2), couplings = get_coupling_estimates(experiments, results, seed,\n **kwargs)\n test_losses = compute_coupling_losses(experiments, logits_1, logits_2,\n couplings)\n visualize_coupling_experiments(test_losses, logits_1, logits_2, couplings)","def main():\n\tresults = []\n\n\tconfig = configparser.ConfigParser()\n\tconfig.read(\"simulation.ini\")\n\tsettings = config['sim']\n\n\tcompleted_obj_hw = int(settings[\"ClientsPerCampaign\"]) * float(settings[\"CompletedPctgHW\"])\n\texceeded_obj_hw = float(settings[\"ExceededPctgHW\"])\n\tsignificance_level = float(settings[\"SignificanceLevel\"])\n\tz_val_two_tails = scipy.stats.norm.ppf(1 - (significance_level / 2))\n\n\tprint(\"Completed Target HW: \" + str(completed_obj_hw))\n\tprint(\"Exceeded Target HW: \" + str(exceeded_obj_hw))\n\n\tcompleted_vals = []\n\texceeded_vals = []\n\tdone = False\n\n\tcompleted_avg = 0\n\texceeded_avg = 0\n\tcompleted_hw = 0\n\texceeded_hw = 0\n\n\ti = 0\n\twhile not done:\n\t\tprint(\"RUN: \" + str(i + 1))\n\t\tenv = simpy.Environment()\n\t\tsim = Simulation(env, settings, i == 0)\n\t\tsim.run()\n\t\tresults.append(sim.results)\n\t\ti += 1\n\n\t\tif settings['RunOnce'] == 'yes':\n\t\t\tprint(\"RUN ONCE\")\n\t\t\tsys.exit()\n\n\t\tcompleted_vals.append(sim.results['completed_count'])\n\t\texceeded_vals.append(sim.results['exceeded_proportion'])\n\n\t\tif i < 2:\n\t\t\tprint(\"---------------\")\n\t\t\tcontinue\n\n\t\tcompleted_avg = sum(completed_vals) / len(completed_vals)\n\t\tcompleted_S = sum([(v - completed_avg) ** 2 for v in completed_vals]) / (i - 1)\n\t\tcompleted_S = math.sqrt(completed_S)\n\t\tcompleted_hw = (z_val_two_tails * completed_S) / math.sqrt(i)\n\t\tprint(\"runs: \" + str(i) + \" completed HW: \" + str(completed_hw))\n\n\t\texceeded_avg = sum(exceeded_vals) / len(exceeded_vals)\n\t\texceeded_S = math.sqrt(exceeded_avg * (1 - exceeded_avg))\n\t\texceeded_hw = (z_val_two_tails * exceeded_S) / math.sqrt(i)\n\t\tprint(\"runs: \" + str(i) + \" exceeded HW: \" + str(exceeded_hw))\n\n\t\tif completed_hw < completed_obj_hw and exceeded_hw < exceeded_obj_hw:\n\t\t\tprint(\"END ITERATIONS\")\n\t\t\tdone = True\n\n\t\tprint(\"---------------\")\n\n\n\tfilename = 'results/Results_' + settings['FileSizeGB'] + '_' + settings['TorrentThreshold'] + '_' + settings['HTTPDownThreshold'] \\\n\t\t+ '_' + settings['HTTPUp'] + '_' + str(random.randint(0,10000)) + '.xlsx'\n\n\tprint(\"Saving XLSX to: \" + filename)\n\twb = xs.Workbook(filename)\n\n\tws = wb.add_worksheet()\n\n\tws.write(0, 1, 'Exceded')\n\tws.write(0, 2, 'Completed')\n\n\ti = 1\n\tfor result in results:\n\t\tws.write(i, 0, i)\n\t\tws.write(i, 1, result['exceeded_proportion'])\n\t\tws.write(i, 2, result['completed_count'])\n\t\ti += 1\n\n\tws.write(i, 0, 'average')\n\tws.write(i, 1, exceeded_avg)\n\tws.write(i, 2, completed_avg)\n\ti += 1\n\tws.write(i, 0, 'half width')\n\tws.write(i, 1, exceeded_hw)\n\tws.write(i, 2, completed_hw)\n\n\twb.close()","def test_compare_classifiers_performance_subset_vis_api(experiment_to_use):\n experiment = experiment_to_use\n probabilities = experiment.probabilities\n viz_outputs = (\"pdf\", \"png\")\n with TemporaryDirectory() as tmpvizdir:\n for viz_output in viz_outputs:\n vis_output_pattern_pdf = tmpvizdir + f\"/*.{viz_output}\"\n visualize.compare_classifiers_performance_subset(\n [probabilities, probabilities],\n experiment.ground_truth,\n experiment.ground_truth_metadata,\n experiment.output_feature_name,\n top_n_classes=[6],\n labels_limit=0,\n subset=\"ground_truth\",\n model_namess=[\"Model1\", \"Model2\"],\n output_directory=tmpvizdir,\n file_format=viz_output,\n )\n figure_cnt = glob.glob(vis_output_pattern_pdf)\n assert 1 == len(figure_cnt)","def test_calculate_all_metrics_precision():\n pass","def run():\n trials = 100\n\n multipliers = [0.25, 0.3, 0.35, 0.5, 0.75, 1, 1.25, 1.45, 1.5, 1.55, 1.6] # Coefficients for learning rate\n\n mean_penalty = []\n median_penalty = []\n std_penalty = []\n\n mean_trial_time = []\n median_trial_time = []\n std_trial_time = []\n\n mean_success_rate = []\n median_success_rate = []\n std_success_rate = []\n\n for m in multipliers:\n all_penalties = [] # All penalties from trail sets\n all_average_trial_time = []\n all_success_rates = []\n\n for i in range(0, 20):\n # print \"Trial set:\", i\n # Set up environment and agent\n e = Environment() # create environment (also adds some dummy traffic)\n agent = e.create_agent(LearnerAgent) # create agent\n agent.mult = m\n e.set_primary_agent(agent, enforce_deadline=True) # specify agent to track\n\n # Now simulate it\n sim = Simulator(e, update_delay=0.0, display=False) # create simulator (uses pygame when display=True, if available)\n\n sim.run(n_trials=trials) # run for a specified number of trials\n\n all_penalties.append(agent.all_trails_penalties)\n all_average_trial_time.append(agent.time/float(trials))\n all_success_rates.append(float(trials-agent.aborted_trials)/trials)\n\n mean_penalty.append(np.mean(all_penalties))\n median_penalty.append(np.median(all_penalties))\n std_penalty.append(np.std(all_penalties))\n\n mean_trial_time.append(np.mean(all_average_trial_time))\n median_trial_time.append(np.median(all_average_trial_time))\n std_trial_time.append(np.std(all_average_trial_time))\n\n mean_success_rate.append(np.mean(all_success_rates))\n median_success_rate.append(np.median(all_success_rates))\n std_success_rate.append(np.std(all_success_rates))\n\n for i in range(0, len(multipliers)):\n print \"\"\n print \"Multiplier:\", multipliers[i]\n print \"\"\n print \"Mean penalty per {} trials:\".format(trials), mean_penalty[i]\n print \"Median penalty per {} trials:\".format(trials), median_penalty[i]\n print \"Std.Dev. penalty per {} trials:\".format(trials), std_penalty[i]\n\n print \"\"\n print \"Mean trial time:\", mean_trial_time[i]\n print \"Median trial time:\", median_trial_time[i]\n print \"Std.Dev. trial time:\", std_trial_time[i]\n\n print \"\"\n print \"Mean success rate per {} trials:\".format(trials), mean_success_rate[i]\n print \"Median success rate per {} trials:\".format(trials), median_success_rate[i]\n print \"Std.Dev. success rate per {} trials:\".format(trials), std_success_rate[i]","def print_results(precision, recall, f1):\n print \"Precision:\", precision\n print \"Recall:\", recall\n print \"F1 score:\", f1","def three_experiments_with_trials(family_with_trials, single_with_trials):","def runTests():\r\n\r\n print(\"running a few tests\")\r\n\r\n average = compute .gpsAverage (4, 5)\r\n print(\"average = \", average)\r\n \r\n print (\"hello!\")","def evaluate(self, runs=100):\n score_record = []\n \n print('Evaluation in progress...')\n for i in range(runs):\n score = self.run_evaluation_episode()\n score_record.append(score)\n \n ave_score = np.mean(score_record)\n \n print('System evaluated with an average score of {} in {} runs'.format(ave_score, runs))","def run_tests():\n with open(FILENAME) as file:\n # Loads testing parameters from the yaml file.\n tests = yaml.safe_load(file)\n\n # create a dataframe to keep the results\n test_dict = tests['Tests']\n results = pd.DataFrame(test_dict)\n results['Last Average Score'] = \"\"\n results['No of Q-Learning episodes'] = \"\"\n\n # run experiments:\n for i, test in enumerate(test_dict):\n grid = Rooms(test[\"env_size\"], testing=True)\n learning = QLearning(grid, test[\"gamma\"], test[\"alpha\"], test[\"agent_start_pos\"])\n e_greedy = Policy(\"e-greedy\", test[\"epsilon\"], test[\"decay\"])\n greedy = Policy(policy_type=\"greedy\")\n experiment = Experiments(grid, learning, greedy, test[\"iters\"],\n test[\"agent_start_pos\"], test[\"test_no\"])\n\n for session in range(test[\"iters\"]):\n learning.run_multiple_episodes(test[\"batch_episodes\"], e_greedy)\n mean_reward = experiment.run_experiments(test[\"exp_per_batch\"])\n\n results.loc[i,'Last Average Score'] = mean_reward\n results.loc[i,'No of Q-Learning episodes'] = (session + 1) * test[\"batch_episodes\"]\n\n # save results to csv file\n filename = 'results/' + 'test_table.csv'\n results.to_csv(filename)\n\n # plot & save graphs\n experiment.generate_results(test[\"test_no\"], test)\n\n return results","def test_compare_performance_vis_api(experiment_to_use):\n experiment = experiment_to_use\n # extract test stats only\n test_stats = experiment.test_stats_full\n viz_outputs = (\"pdf\", \"png\")\n with TemporaryDirectory() as tmpvizdir:\n for viz_output in viz_outputs:\n vis_output_pattern_pdf = tmpvizdir + f\"/*.{viz_output}\"\n visualize.compare_performance(\n [test_stats, test_stats],\n output_feature_name=None,\n model_names=[\"Model1\", \"Model2\"],\n output_directory=tmpvizdir,\n file_format=viz_output,\n )\n figure_cnt = glob.glob(vis_output_pattern_pdf)\n assert 1 == len(figure_cnt)","def emPerformance(filesAndDirectories='None', resultsFileName='None', iterationCount='3', modes='None', testTypes='None', viewports='None', verbose='False'):\n\n pass","def evaluate_all(experiments,\n results,\n seed,\n num_pairs,\n samples_per_pair,\n loop_size=None):\n eval_results = evaluate_baselines(\n experiments[0], seed, num_pairs, samples_per_pair, loop_size=None)\n for ex, res in zip(experiments, results):\n eval_results.update(\n evaluate_experiment(ex, res, seed, num_pairs, samples_per_pair,\n loop_size))\n return eval_results","def _evaluate_performance__static_winners(self):\n # | - _evaluate_performance__\n\n # | - class attributes #################################################\n AL = self\n al_gen = self.al_gen\n verbose = self.verbose\n seed_ids = self.seed_ids\n acquisition_bin = self.acquisition_bin\n completed_ids = self.completed_ids\n CandidateSpace = self.CandidateSpace\n RegressionModel = self.RegressionModel\n DuplicateFinder = self.DuplicateFinder\n al_gen_dict = self.al_gen_dict\n\n stop_mode = self.stop_mode\n stop_num_generations = self.stop_num_generations\n\n index_acq_gen_dict = self.index_acq_gen_dict\n #__| #################################################################\n\n # #####################################################################\n mode = \"lowest_N\" # 'lowest_N' or 'lowest_perc'\n\n N_ids = 10\n lowest_perc = 5\n\n # Number of consecutive generations that the Nth best systems must\n # remain static\n M_gens = 3\n # #####################################################################\n\n if mode == \"lowest_perc\":\n num_candidates = CandidateSpace.FingerPrints.df_pre.shape[0]\n N_ids = int(num_candidates * (lowest_perc * 0.01))\n\n gen_keys = list(AL.al_gen_dict.keys())\n\n if len(gen_keys) > M_gens:\n latest_M_keys = gen_keys[-(M_gens + 1):]\n last_gen_key = gen_keys[-1]\n\n al_gen_dict_subset_i = dict(zip(\n latest_M_keys,\n [AL.al_gen_dict.get(i, None) for i in latest_M_keys]))\n\n indices_list = []\n iterator = enumerate(al_gen_dict_subset_i.items())\n for i_cnt, (gen_i, AL_i) in iterator:\n model_i = AL_i.model\n\n model_i = AL.add_main_Y_to_model(\n model_i, plot_dft_instead_of_pred=True)\n model_i = model_i[(model_i[\"duplicate\"] == False)]\n model_i = model_i.sort_values(\"Y_main\")\n\n indices_i = model_i.index.tolist()\n\n indices_list.append(indices_i)\n\n if i_cnt >= M_gens:\n indices_i = indices_list[i_cnt][0:N_ids]\n ids_static_list = []\n for j in range(M_gens):\n indices_j = indices_list[i_cnt - (j + 1)][0:N_ids]\n ids_static = indices_j == indices_i\n ids_static_list.append(ids_static)\n\n ids_are_static = all(ids_static_list)\n\n self.performance__static_winners[last_gen_key] = ids_are_static\n #__|","def evaluate_benchmarks(self):\n\n # iterate over replicates\n results = {}\n for replicate_id, replicate in self.replicates:\n\n # evaluate benchmark for current replicate\n bmark = SimulationBenchmark(replicate.copy(),\n graph=self.graphs[replicate_id],\n **self.params)\n\n # store results\n results[replicate_id] = dict(\n\n labels_MAE=bmark.scores['labels'].MAE,\n level_only_MAE=bmark.scores['level_only'].MAE,\n spatial_only_MAE=bmark.scores['spatial_only'].MAE,\n community_MAE=bmark.scores['labels_comm'].MAE,\n\n labels_PCT=bmark.scores['labels'].percent_correct,\n level_only_PCT=bmark.scores['level_only'].percent_correct,\n spatial_only_PCT=bmark.scores['spatial_only'].percent_correct,\n community_PCT=bmark.scores['labels_comm'].percent_correct)\n\n # compile dataframe\n results = pd.DataFrame.from_dict(results, orient='index')\n results.index.set_names(self.multiindex, inplace=True)\n\n return results","def run_experiment():\n pass","def evaluate_individual(predictions, test_files, models):\n\n print(\"\\nAccuracy for individual models\\n\")\n \n # Fix Location\n correct_predictions = [0, 0, 0]\n total_predictions = [0, 0, 0]\n num_failed_predictions = 0\n\n for prediction in predictions:\n if prediction[\"correct_data\"][\"correct_location\"] == prediction[\"predicted_location\"]:\n correct_predictions[FixType[prediction[\"correct_data\"][\"correct_type\"]].value] = correct_predictions[FixType[\n prediction[\"correct_data\"][\"correct_type\"]].value] + 1\n if prediction[\"predicted_location\"] is None:\n num_failed_predictions = num_failed_predictions + 1\n total_predictions[FixType[prediction[\"correct_data\"][\"correct_type\"]].value] = total_predictions[FixType[\n prediction[\"correct_data\"][\"correct_type\"]].value] + 1\n\n for i in range(3):\n if total_predictions[i] == 0: # If the type was never predicted\n accuracy = 0\n else:\n accuracy = correct_predictions[i] / total_predictions[i]\n print(f\"Fix Location accuracy for class {FixType(i).name}: {accuracy * 100} %\")\n\n accuracy = sum(correct_predictions) / (len(predictions) - num_failed_predictions)\n print(f\"Fix Location accuracy overall is {accuracy * 100} %\")\n \n # Fix type\n correct_predictions = [0, 0, 0]\n total_predictions = [0, 0, 0]\n num_failed_predictions = 0\n\n for prediction in predictions:\n if prediction[\"correct_data\"][\"correct_type\"] == prediction[\"predicted_type\"]:\n correct_predictions[FixType[prediction[\"predicted_type\"]].value] = correct_predictions[FixType[\n prediction[\"predicted_type\"]].value] + 1\n if prediction[\"predicted_type\"] is None:\n num_failed_predictions = num_failed_predictions + 1\n total_predictions[FixType[prediction[\"predicted_type\"]].value] = total_predictions[FixType[\n prediction[\"predicted_type\"]].value] + 1\n\n for i in range(3):\n if total_predictions[i] == 0: # If the type was never predicted\n accuracy = 0\n else:\n accuracy = correct_predictions[i] / total_predictions[i]\n print(f\"Fix Type accuracy for class {FixType(i).name}: {accuracy * 100} %\")\n\n accuracy = sum(correct_predictions) / (len(predictions) - num_failed_predictions)\n print(f\"Fix Type accuracy overall is {accuracy * 100} %\")\n \n # We repeat the predictions to evaluate the insert and modify models individually, regardless of the predicted fix type \n\n raw_training_samples = []\n\n if test_files.endswith(\".json\"): # Single JSON file\n with open(test_files) as file:\n logging.info(\"Source ending in .json. Predicting on single JSON file.\")\n raw_training_samples = json.load(file)\n else: # Folder path\n for filename in listdir(test_files):\n with open(test_files + filename) as file:\n raw_training_samples.extend(json.load(file))\n \n correct_predictions_insert = 0\n total_predictions_insert = 0\n correct_predictions_modify = 0\n total_predictions_modify = 0\n insert_tokens = []\n modify_tokens = []\n\n for sample in raw_training_samples:\n # Insert\n if sample[\"metadata\"][\"fix_type\"] == \"insert\":\n actual_sample, tokens = IOProcessor.preprocess(sample[\"wrong_code\"])\n pred = predict_single(actual_sample, models[2])\n token = IOProcessor.postprocess(pred, 2)\n if token == sample[\"metadata\"][\"fix_token\"]: # Correct Prediction\n correct_predictions_insert = correct_predictions_insert + 1\n else: # Incorrect prediction\n insert_tokens.append([token, sample[\"metadata\"][\"fix_token\"]])\n total_predictions_insert = total_predictions_insert + 1\n # Modify\n if sample[\"metadata\"][\"fix_type\"] == \"modify\":\n actual_sample, tokens = IOProcessor.preprocess(sample[\"wrong_code\"])\n pred = predict_single(actual_sample, models[3])\n token = IOProcessor.postprocess(pred, 3)\n if token == sample[\"metadata\"][\"fix_token\"]: # Correct Prediction\n correct_predictions_modify = correct_predictions_modify + 1\n else: # Incorrect prediction\n modify_tokens.append([token, sample[\"metadata\"][\"fix_token\"]])\n total_predictions_modify = total_predictions_modify + 1\n\n insert_accuracy = correct_predictions_insert / total_predictions_insert\n modify_accuracy = correct_predictions_modify / total_predictions_modify\n print(f\"Fix Token accuracy for insert is {insert_accuracy * 100} %\")\n print(f\"Fix Token accuracy for modify is {modify_accuracy * 100} %\")\n\n # The following code may be used to create a swarm plot of the erroneous predictions for fix locations\n # This does, however, require the installation of the pandas, seaborn, and matplotlib libraries.\n \n # import seaborn as sns\n # import matplotlib.pyplot as plt\n # import pandas as pd\n # location_distance_array = []\n # for prediction in predictions:\n # actual_sample, tokens = IOProcessor.preprocess(prediction[\"correct_data\"][\"wrong_code\"])\n # label = get_token_index(prediction[\"correct_data\"][\"wrong_code\"], tokens, prediction[\"correct_data\"][\"correct_location\"])\n # if prediction[\"predicted_token_location\"] - label == 0:\n # pass\n # else:\n # location_distance_array.append([prediction[\"predicted_token_location\"] - label, prediction[\"correct_data\"][\"correct_type\"]])\n \n # df = pd.DataFrame(data=location_distance_array)\n # sns.set_theme(style=\"whitegrid\")\n # f, ax = plt.subplots(figsize=(6, 4))\n # sns.despine(bottom=True, left=True)\n # sns.swarmplot(y=0, x=1, data=df, palette=\"dark\", size=6)\n # ax.set_xlabel('')\n # ax.set_ylabel('')\n # plt.ylim([-15, 16])\n \n # plt.savefig('line_plot.pdf', bbox_inches='tight', pad_inches=0)","def batch_simulation(self, iters=10000):\n power_cnt = 0\n correct_sign_cnt = 0\n\n for i in range(iters):\n if (self.verbose) and (i>0) and (i % (iters/10) == 0):\n print(i, \" / \", iters)\n f_stat, p_value, effect_point_estimates = self.simulate()\n power_cnt += (p_value < self.alpha)\n correct_sign_cnt += (effect_point_estimates * self.absolute_effect > 0)\n\n if self.verbose: print(iters, \" / \", iters)\n power = round(power_cnt / float(iters), 5)\n pct_correct_sign = round(correct_sign_cnt / float(iters), 5)\n return power, pct_correct_sign","def run(self):\n if self.pb.xvalEN and not self.isXvalMain:\n dt = np.ones(self.pb.total_exam_no,\n dtype=\"float32\") / self.train_exam_no\n dt[self.val_indices] = 0.0\n else:\n dt = np.ones(self.pb.total_exam_no,\n dtype=\"float32\") / self.pb.total_exam_no\n\n val = np.zeros(8,dtype=\"float32\")-1\n boosting = None\n wl = None\n if self.pb.algorithm == 'conf-rated':\n boosting = ConfidenceRated(self)\n wl = ConfidenceRatedWL(self)\n elif self.pb.algorithm == 'adaboost':\n boosting = AdaBoost(self)\n wl = AdaBoostWL(self)\n elif self.pb.algorithm == 'adaboost-fast':\n boosting = AdaBoostFast(self)\n wl = AdaBoostFastWLMPI(self)\n elif self.pb.algorithm == 'rankboost':\n boosting = RankBoost(self)\n wl = ConfidenceRatedWL(self)\n elif self.pb.algorithm == 'rankboost-fast':\n boosting = RankBoost(self)\n wl = AdaBoostFastWL(self)\n else:\n raise Exception(\"Unknown Boosting Algorithm\")\n \n for r in range(self.pb.rounds):\n tree = wl.run(dt)\n dt = boosting.run(dt = dt,\n r = r,\n tree = tree)\n \n if self.isXvalMain:\n boosting.finalize()\n \n \"\"\"Sync the predictions and save them to a file\"\"\"\n if self.pb.isLeader:\n if self.pb.xvalEN and not self.isXvalMain:\n val_predictions = boosting.get_val_predictions()\n hypotheses = boosting.get_hypotheses()\n np.savez(self.out_fp,\n val_predictions = val_predictions,\n hypotheses = hypotheses,\n )\n if self.pb.testEN and self.isXvalMain:\n train_predictions = np.zeros([self.pb.total_exam_no,self.pb.rounds],\n dtype=\"float32\")\n test_predictions = np.zeros([self.pb.test_exam_no,self.pb.rounds],\n dtype=\"float32\")\n for slv in np.arange(self.pb.comm_size):\n tr_i1 = self.pb.train_partition[slv]\n tr_i2 = self.pb.train_partition[slv+1]\n te_i1 = self.pb.test_partition[slv]\n te_i2 = self.pb.test_partition[slv+1]\n if slv == 0:\n train_predictions[tr_i1:tr_i2,:] = boosting.get_train_predictions()\n test_predictions[te_i1:te_i2,:] = boosting.get_test_predictions()\n else:\n train_predictions[tr_i1:tr_i2,:] = self.pb.comm.recv(source=slv,tag=11)\n test_predictions[te_i1:te_i2,:] = self.pb.comm.recv(source=slv,tag=12)\n hypotheses = boosting.get_hypotheses()\n np.savez(self.out_fp,\n train_predictions = train_predictions,\n test_predictions = test_predictions,\n hypotheses = hypotheses,\n )\n if not self.pb.testEN and self.isXvalMain:\n train_predictions = np.zeros([self.pb.total_exam_no,self.pb.rounds],\n dtype=\"float32\")\n for slv in np.arange(self.pb.comm_size):\n tr_i1 = self.pb.train_partition[slv]\n tr_i2 = self.pb.train_partition[slv+1]\n if slv == 0:\n train_predictions[tr_i1:tr_i2,:] = boosting.get_train_predictions()\n else:\n train_predictions[tr_i1:tr_i2,:] = self.pb.comm.recv(source=slv,tag=11)\n hypotheses = boosting.get_hypotheses()\n np.savez(self.out_fp,\n train_predictions = train_predictions,\n hypotheses = hypotheses,\n )\n else:\n if self.pb.testEN and self.isXvalMain:\n train_predictions = boosting.get_train_predictions()\n self.pb.comm.send(train_predictions,dest = 0,tag=11)\n test_predictions = boosting.get_test_predictions()\n self.pb.comm.send(test_predictions,dest = 0,tag=12)\n if not self.pb.testEN and self.isXvalMain:\n train_predictions = boosting.get_train_predictions()\n self.pb.comm.send(train_predictions,dest = 0,tag=11)","def _train_and_evaluate(estimator, output_dir):\n \n \"\"\"X_train, y_train =utils._feature_label_split(df_train,\"is_churn\",\"msno\")\n df_val = utils.read_from_bigquery(\"amiable-octane-267022.kkbox.output_val_1\",\"amiable-octane-267022\")\n X_val, y_val =utils._feature_label_split(df_val,\"is_churn\",\"msno\")\"\"\"\n \n df_train=utils.over_sample(\"amiable-octane-267022.kkbox.output_train_1\",\"amiable-octane-267022\")\n X_train, y_train =utils._feature_label_split(df_train,\"is_churn\",\"msno\")\n df_val=utils.over_sample(\"amiable-octane-267022.kkbox.output_val_1\",\"amiable-octane-267022\")\n X_val, y_val =utils._feature_label_split(df_val,\"is_churn\",\"msno\")\n\n estimator.fit(X_train, y_train)\n f1_scorer = make_scorer(f1_score)\n accuracy_scorer =make_scorer(accuracy_score)\n\n if metadata.HYPERPARAMTER_TUNING:\n scores=model_selection.cross_val_score(estimator, X_val, y_val, cv=3,scoring=f1_scorer)\n #,scoring=f1_scorer\n\n logging.info('Score: %s', scores)\n\n #tune hyper\n hpt = hypertune.HyperTune()\n hpt.report_hyperparameter_tuning_metric(\n hyperparameter_metric_tag='F1_SCORE',\n metric_value=np.mean(scores),\n global_step=10000)\n \n#joblib.dump(estimator, 'model.joblib')\n\n # Write model and eval metrics to `output_dir`\n model_output_path = os.path.join(output_dir, 'model',metadata.MODEL_FILE_NAME)\n \n utils.dump_object(estimator, model_output_path)","def run(self):\n if self.pb.xvalEN and not self.isXvalMain:\n dt = np.ones(self.pb.total_exam_no,\n dtype=\"float32\") / self.train_exam_no\n dt[self.val_indices] = 0.0\n else:\n dt = np.ones(self.pb.total_exam_no,\n dtype=\"float32\") / self.pb.total_exam_no\n\n val = np.zeros(8,dtype=\"float32\")-1\n boosting = None\n wl = None\n if self.pb.algorithm == 'conf-rated':\n boosting = ConfidenceRated(self)\n wl = ConfidenceRatedWL(self)\n elif self.pb.algorithm == 'adaboost':\n boosting = AdaBoost(self)\n wl = AdaBoostWL(self)\n elif self.pb.algorithm == 'adaboost-fast':\n boosting = AdaBoostFast(self)\n wl = AdaBoostFastWL(self)\n elif self.pb.algorithm == 'rankboost':\n boosting = RankBoost(self)\n wl = ConfidenceRatedWL(self)\n elif self.pb.algorithm == 'rankboost-fast':\n boosting = RankBoost(self)\n wl = AdaBoostFastWL(self)\n else:\n raise Exception(\"Unknown Boosting Algorithm\")\n \n for r in range(self.pb.rounds):\n tree = wl.run(dt)\n dt = boosting.run(dt = dt,\n r = r,\n tree = tree)\n \n if self.isXvalMain:\n boosting.finalize()\n \n \"\"\"Sync the predictions and save them to a file\"\"\"\n if self.pb.isLeader:\n if self.pb.xvalEN and not self.isXvalMain:\n val_predictions = boosting.get_val_predictions()\n hypotheses = boosting.get_hypotheses()\n np.savez(self.out_fp,\n val_predictions = val_predictions,\n hypotheses = hypotheses,\n )\n if self.pb.testEN and self.isXvalMain:\n train_predictions = boosting.get_train_predictions()\n test_predictions = boosting.get_test_predictions()\n hypotheses = boosting.get_hypotheses()\n np.savez(self.out_fp,\n train_predictions = train_predictions,\n test_predictions = test_predictions,\n hypotheses = hypotheses,\n )\n if not self.pb.testEN and self.isXvalMain:\n train_predictions = boosting.get_train_predictions()\n hypotheses = boosting.get_hypotheses()\n np.savez(self.out_fp,\n train_predictions = train_predictions,\n hypotheses = hypotheses,\n )","def compute_and_print_eval_metrics(self):\n s = ('%20s' + '%12s' * 6) % ('Class', 'Images', 'Targets', 'P', 'R', 'mAP@.5', 'mAP@.5:.95')\n precision, recall, f1, mean_precision, mean_recall, map50, map = 0., 0., 0., 0., 0., 0., 0.\n ap = []\n eval_stats = [np.concatenate(x, 0) for x in zip(*self.eval_stats)]\n if len(eval_stats) and eval_stats[0].any():\n precision, recall, ap, f1, ap_class = ap_per_class(*eval_stats)\n precision, recall, ap50, ap = precision[:, 0], recall[:, 0], ap[:, 0], ap.mean(1)\n mean_precision, mean_recall, map50, map = precision.mean(), recall.mean(), ap50.mean(), ap.mean()\n nt = np.bincount(eval_stats[3].astype(np.int64), minlength=len(self.class_names)) # number of targets per class\n else:\n nt = np.zeros(1)\n\n pf = '%20s' + '%12.5g' * 6 # print format\n print(\"\\n EVALUTAION \\n\")\n print(s)\n print(pf % ('all', self.seen, nt.sum(), mean_precision, mean_recall, map50, map))\n if self.cfg.eval.verbose:\n for indx, cls in enumerate(ap_class):\n print(pf % (self.class_names[cls], self.seen, nt[cls], precision[indx], recall[indx], ap50[indx], ap[indx]))","def test_precision_recall_curves_vis_api(experiment_to_use):\n experiment = experiment_to_use\n probabilities = experiment.probabilities\n viz_outputs = (\"pdf\", \"png\")\n positive_label = 1\n with TemporaryDirectory() as tmpvizdir:\n for viz_output in viz_outputs:\n vis_output_pattern_pdf = tmpvizdir + f\"/*.{viz_output}\"\n visualize.precision_recall_curves(\n [probabilities, probabilities],\n experiment.ground_truth,\n experiment.ground_truth_metadata,\n experiment.output_feature_name,\n positive_label,\n model_names=[\"Model1\", \"Model2\"],\n output_directory=tmpvizdir,\n file_format=viz_output,\n )\n figure_cnt = glob.glob(vis_output_pattern_pdf)\n assert 1 == len(figure_cnt)","def test(self, plot=False):\n accuracy_list = []\n fobj_avg = self.load_stats()\n\n for ii in range(settings.PARS['maxIters']):\n model = self.load_model(ii)\n D1 = model['D']\n W1 = model['W']\n\n # classification\n tic = time.time()\n accuracy_list.append(self.classification(D1, W1)[1])\n toc = time.time()\n print(\n 'Final recognition rate for OnlineDL is : {} , objective function value: {}, time: {}'\n .format(accuracy_list[ii], fobj_avg[ii], toc-tic)\n )\n\n accuracy_list = np.asarray(accuracy_list)\n\n print('Best recognition rate for OnlineDL is {} at iteration {}'.format(\n accuracy_list.max(), accuracy_list.argmax()))\n\n if plot:\n # plot the objective function values for all iterations\n plt.clf()\n plt.plot(list(fobj_avg.keys()), list(fobj_avg.values()), 'mo--', linewidth=2)\n plt.xlabel('Iterations')\n plt.ylabel('Average objective function value')\n plt.xticks(list(range(0, 20)), list(range(1, 21)))\n plt.show()\n\n plt.clf()\n plt.plot(accuracy_list, 'rs--', linewidth=2)\n plt.xticks(list(range(0, 20)), list(range(1, 21)))\n plt.xlabel('Iterations')\n plt.ylabel('Accuracy')\n plt.show()","def testDriver():\n exam1=90\n exam2=85\n assignmentScores = [50, 60, 70, 80, ]\n computeGrades(exam1, exam2, assignmentScores)","def run_regression_experiments(self):\n assert self.patientIDs # assert not empty\n self.log.info(\"Running experiments for {} patients\".format(len(self.patientIDs)))\n\n results_batch = defaultdict(dict)\n results_single = dict()\n for patient in self.patientIDs:\n patient_experiment = PatientPredictionExperiment(\n patient_id=patient, algorithm=self.algorithm,\n is_batch=self.isBatch\n )\n # run results for single patients\n if not self.isBatch:\n results_single[patient] = patient_experiment.run_experiment()\n self.log.debug(\"Start processing single results: {}\".format(results_single[patient]))\n # super verbose debug information should be logged with debug option\n else:\n # TODO: think about consistent interfaces\n batch_results, self._allFeatureDesp = patient_experiment.run_experiment()\n self.log.info(\"Start processing batch results with {} batches\".format(len(batch_results)))\n batchNo = 0\n for results in batch_results:\n # here are the result for each feature subset for each patients\n results_batch[batchNo][patient] = results\n batchNo += 1\n if self.algorithm == \"fm\":\n return\n if not self.isBatch:\n results_overall = self.process_results(results_single)\n self.store_results(results_overall, results_single)\n self.log.info(\"Finished running experiment\")\n else:\n self.record_batch_experiment()\n for batchNo, patient_results in results_batch.iteritems():\n # patient_results = results_single\n results_overall = self.process_results(patient_results)\n self.store_results(results_overall, patient_results, self._allFeatureDesp[batchNo])\n self.log.info(\"Finished running experiment No. {}\".format(batchNo))","def _run_evaluation(\n sess, experiment, eval_config, output_dir, min_range, max_range, num_bins,\n torsion_bins):\n tf.io.gfile.makedirs(os.path.join(output_dir, 'pickle_files'))\n\n logging.info('Eval config is %s\\nnum_bins: %d', eval_config, num_bins)\n num_examples = 0\n num_crops = 0\n start_all_time = time.time()\n\n # Either do the whole test set, or up to a specified limit.\n max_examples = experiment.num_eval_examples\n if eval_config.max_num_examples > 0:\n max_examples = min(max_examples, eval_config.max_num_examples)\n\n while num_examples < max_examples:\n one_prediction = compute_one_prediction(\n num_examples, experiment, sess, eval_config, num_bins, torsion_bins)\n\n single_message = one_prediction.single_message\n num_crops_local = one_prediction.num_crops_local\n sequence = one_prediction.sequence\n filebase = one_prediction.filebase\n softmax_probs = one_prediction.softmax_probs\n ss = one_prediction.ss\n asa = one_prediction.asa\n torsions = one_prediction.torsions\n\n num_examples += 1\n num_crops += num_crops_local\n\n # Save the output files.\n filename = os.path.join(output_dir,\n 'pickle_files', '%s.pickle' % filebase)\n distogram_io.save_distance_histogram(\n filename, softmax_probs, filebase, sequence,\n min_range=min_range, max_range=max_range, num_bins=num_bins)\n\n if experiment.model.torsion_multiplier > 0:\n torsions_dir = os.path.join(output_dir, 'torsions')\n tf.io.gfile.makedirs(torsions_dir)\n distogram_io.save_torsions(torsions_dir, filebase, sequence, torsions)\n\n if experiment.model.secstruct_multiplier > 0:\n ss_dir = os.path.join(output_dir, 'secstruct')\n tf.io.gfile.makedirs(ss_dir)\n secstruct.save_secstructs(ss_dir, filebase, None, sequence, ss)\n\n if experiment.model.asa_multiplier > 0:\n asa_dir = os.path.join(output_dir, 'asa')\n tf.io.gfile.makedirs(asa_dir)\n secstruct.save_secstructs(asa_dir, filebase, None, sequence,\n np.expand_dims(asa, 1), label='Deepmind 2D ASA')\n\n time_spent = time.time() - start_all_time\n logging.info(\n 'Evaluate %d examples, %d crops %.1f crops/ex. '\n 'Took %.1fs, %.3f s/example %.3f crops/s\\n%s',\n num_examples, num_crops, num_crops / float(num_examples), time_spent,\n time_spent / num_examples, num_crops / time_spent, single_message)\n\n logging.info('Tested on %d', num_examples)","def print_results(self, f1_score, precision_score, recall_score):\n print(\"Algorithm: %s\" % self.name)\n self.print_score(f1_score, \"F1 Score\")\n self.print_score(precision_score, \"Precision Score\")\n self.print_score(recall_score, \"Recall Score\")","def test_run():\n # Only few steps for test\n timesteps = 128\n\n # Compute all sub testing conf\n envs = ['CartPole-v0']\n ml_platforms = ['torch', 'tf']\n agents = ['dqn', 'a2c']\n\n test_combinations = list(it.product(\n envs,\n ml_platforms,\n agents\n )\n )\n\n # Finally test them all\n for conf in test_combinations:\n env_str, ml_platform_str, agent_str = conf\n run(\n agent_str,\n ml_platform_str,\n env_str,\n 'dense',\n timesteps,\n './target/')","def eval_perf_total(model, X_train, y_train, X_test, y_test):\n\n y_hat_train = model.predict(X_train)\n y_hat_test = model.predict(X_test)\n \n train_mae = metrics.mean_absolute_error(y_train, y_hat_train)\n train_mse = metrics.mean_squared_error(y_train, y_hat_train)\n train_rmse = np.sqrt(metrics.mean_squared_error(y_train, y_hat_train))\n train_r = metrics.r2_score(y_train, y_hat_train)\n\n print('Evaluating Performance on Training Data:\\n')\n print(f' Train Mean Absolute Error: {train_mae:,.2f}')\n print(f' Train Mean Squared Error: {train_mse:,.2f}\\n')\n print(f'Train Root Mean Squared Error: {train_rmse:,.2f}')\n print(f'Train R-Square Value: {round(train_r,2)}')\n\n print('\\n'+'---'*25+'\\n')\n\n test_mae = metrics.mean_absolute_error(y_test, y_hat_test)\n test_mse = metrics.mean_squared_error(y_test, y_hat_test)\n test_rmse = np.sqrt(metrics.mean_squared_error(y_test, y_hat_test))\n test_r = metrics.r2_score(y_test, y_hat_test)\n\n print('Evaluating Performance on Testing Data:\\n')\n print(f' Test Mean Absolute Error: {test_mae:,.2f}')\n print(f' Test Mean Squared Error: {test_mse:,.2f}\\n')\n print(f'Test Root Mean Squared Error: {test_rmse:,.2f}')\n print(f'Test R-Square Value: {round(test_r,2)}')","def main():\n parser = argparse.ArgumentParser(description=\"Process the results of an experiment.\")\n parser.add_argument(\"experiment\")\n arguments = parser.parse_args()\n path = f\"experiments/{arguments.experiment}\"\n if not os.path.exists(path):\n raise SystemExit(f\"Path {path} does not exists.\")\n\n # For efficiency, one should generate the results from the parts without merging them.\n files = [file for file in os.listdir(path) if os.path.isfile(os.path.join(path, file))]\n frames = []\n for file in files:\n device, experiment, _ = file.split(\".\")\n frame = pandas.read_csv(\n os.path.join(path, file),\n index_col=\"variable\",\n usecols=[\"variable\", \"group_index\", \"value_i\"], dtype={\"value_i\": \"Int64\"}\n )\n frame[\"board\"] = device\n frame[\"experiment\"] = experiment\n frames.append(frame)\n dataframe = pandas.concat(frames)\n frames = None\n\n current_grouping = dataframe.groupby([\"group_index\", \"variable\"])\n \n data = current_grouping.agg([\n numpy.median,\n _percentile_factory(95),\n numpy.mean,\n numpy.std,\n \"count\"\n ])\n\n print(data)\n \n data = data.droplevel([0], axis=1)\n data = data.unstack()\n data.columns = data.columns.map('_'.join)\n data.to_csv(f\"{arguments.experiment}.csv\")","def test_compare(self):\n config = {\n 'num_components': 512,\n 'num_features': 128,\n 'covariance': 'spherical'\n }\n\n samples = self.generate_samples(config, 100_000)\n sklearn_time = np.mean([self.train_sklearn(config, samples) for _ in range(3)])\n ours_cpu_time = np.mean([self.train_ours(config, samples) for _ in range(3)])\n ours_gpu_time = np.mean([\n self.train_ours(config, samples.cuda(), gpu=True) for _ in range(3)\n ])\n\n print(f\"-------------------------------------\")\n print(f\"Speedup of CPU implementation: {sklearn_time / ours_cpu_time:.2f}\")\n print(f\"Speedup of GPU implementation: {sklearn_time / ours_gpu_time:.2f}\")\n print(f\"-------------------------------------\")","def log_evaluation(tester, name, description):\r\n\tfor dataset, output in tester.preds.items():\r\n\t\tresults = pandas.DataFrame.from_dict(output)\r\n\t\tpath = os.path.join(\r\n\t\t\tEXPERIMENT_PATH, tester.config[\"name\"] + '-' + dataset)\r\n\t\twith open(path + \".csv\", \"w\") as f:\r\n\t\t\tresults.to_csv(f, sep=\"\\t\", encoding='utf-8',\r\n\t\t\t\tfloat_format='%.3f', index=False)","def eval(self):\n\n # How many questions we get right at precision@1.\n correct = 0\n\n total = self._analogy_questions.shape[0]\n start = 0\n while start < total:\n limit = start + 2500\n sub = self._analogy_questions[start:limit, :]\n idx = self._predict(sub)\n start = limit\n for question in xrange(sub.shape[0]):\n if sub[question, 3] in idx[question]:\n # print(sub[question, 3], idx[question])\n correct += 1\n\n print()\n print(\"Eval %4d/%d accuracy @ top5= %4.1f%%\" % (correct, total,\n correct * 100. / total)\n )","def main():\n for opt in optimizations:\n compile_command = [\"g++\", \"main.cpp\", f\"-O{opt}\", \"-lpthread\"]\n run(compile_command, check=True)\n for threads in num_threads:\n print(f\"{opt=}, {threads=}\", end=\"\")\n stdout.flush()\n test_command = ['./a.out', str(iterations), str(threads)]\n total = 0\n for samples in range(1, repeats_for_average + 1):\n print(\".\", end=\"\")\n stdout.flush()\n output = run(test_command, check=True, capture_output=True).stdout\n total += int(output.split()[-2]) / 1000\n print(f\"\\t{total / samples:.03f}\")","def run_experiments(data_set=\"\",compact=2,exp_name=\"\",x=\"\"):\n cwd = os.getcwd()\n results_path = cwd+'/data/geometric/'+exp_name+'_'\n dataset_path = cwd+'/data/geometric/'+data_set+'_'\n compact = int(compact)\n if \"m\" in x:\n # make new pxy\n pxy, Xdata, groups = gen_geometric_pxy()\n np.save(dataset_path+'Xdata',Xdata)\n np.save(dataset_path+'groups',groups)\n np.save(dataset_path+'pxy',pxy)\n else:\n # load existing pxy\n pxy = np.load(dataset_path+'pxy.npy')\n if \"r\" in x: # regular experiments\n if compact>1:\n metrics_stepwise, distributions_stepwise,\\\n metrics_converged, distributions_converged,\\\n metrics_stepwise_allreps, distributions_stepwise_allreps,\\\n metrics_converged_allreps, distributions_converged_allreps = test_IB(pxy,compact)\n metrics_converged.to_csv(results_path+'metrics_converged.csv')\n metrics_stepwise.to_csv(results_path+'metrics_stepwise.csv')\n distributions_converged.to_pickle(results_path+'distributions_converged.pkl')\n elif compact>0:\n metrics_stepwise,\\\n metrics_converged, distributions_converged,\\\n metrics_stepwise_allreps,\\\n metrics_converged_allreps, distributions_converged_allreps = test_IB(pxy,compact)\n metrics_converged.to_csv(results_path+'metrics_converged.csv')\n metrics_stepwise.to_csv(results_path+'metrics_stepwise.csv')\n distributions_converged.to_pickle(results_path+'distributions_converged.pkl')\n else:\n metrics_stepwise,\\\n metrics_converged,\\\n metrics_stepwise_allreps,\\\n metrics_converged_allreps = test_IB(pxy,compact)\n metrics_converged.to_csv(results_path+'metrics_converged.csv')\n metrics_stepwise.to_csv(results_path+'metrics_stepwise.csv')\n if \"ip\" in x: # initialization experiments - positive p0\n if compact>1:\n metrics_stepwise, distributions_stepwise,\\\n metrics_converged, distributions_converged,\\\n metrics_stepwise_allreps, distributions_stepwise_allreps,\\\n metrics_converged_allreps, distributions_converged_allreps = test_p0_pos(pxy,compact)\n metrics_converged.to_csv(results_path+'metrics_converged_p0_pos.csv')\n metrics_stepwise.to_csv(results_path+'metrics_stepwise_p0_pos.csv')\n distributions_converged.to_pickle(results_path+'distributions_converged_p0_pos.pkl')\n elif compact>0:\n metrics_stepwise,\\\n metrics_converged, distributions_converged,\\\n metrics_stepwise_allreps,\\\n metrics_converged_allreps, distributions_converged_allreps = test_p0_pos(pxy,compact)\n metrics_converged.to_csv(results_path+'metrics_converged_p0_pos.csv')\n metrics_stepwise.to_csv(results_path+'metrics_stepwise_p0_pos.csv')\n distributions_converged.to_pickle(results_path+'distributions_converged_p0_pos.pkl')\n else:\n metrics_stepwise,\\\n metrics_converged,\\\n metrics_stepwise_allreps,\\\n metrics_converged_allreps = test_p0_pos(pxy,compact)\n metrics_converged.to_csv(results_path+'metrics_converged_p0_pos.csv')\n metrics_stepwise.to_csv(results_path+'metrics_stepwise_p0_pos.csv')\n if \"in\" in x: # initialization experiments - negative p0\n if compact>1:\n metrics_stepwise, distributions_stepwise,\\\n metrics_converged, distributions_converged,\\\n metrics_stepwise_allreps, distributions_stepwise_allreps,\\\n metrics_converged_allreps, distributions_converged_allreps = test_p0_neg(pxy,compact)\n metrics_converged.to_csv(results_path+'metrics_converged_p0_neg.csv')\n metrics_stepwise.to_csv(results_path+'metrics_stepwise_p0_neg.csv')\n distributions_converged.to_pickle(results_path+'distributions_converged_p0_neg.pkl')\n elif compact>0:\n metrics_stepwise,\\\n metrics_converged, distributions_converged,\\\n metrics_stepwise_allreps,\\\n metrics_converged_allreps, distributions_converged_allreps = test_p0_neg(pxy,compact)\n metrics_converged.to_csv(results_path+'metrics_converged_p0_neg.csv')\n metrics_stepwise.to_csv(results_path+'metrics_stepwise_p0_neg.csv')\n distributions_converged.to_pickle(results_path+'distributions_converged_p0_neg.pkl')\n else:\n metrics_stepwise,\\\n metrics_converged,\\\n metrics_stepwise_allreps,\\\n metrics_converged_allreps = test_p0_neg(pxy,compact)\n metrics_converged.to_csv(results_path+'metrics_converged_p0_neg.csv')\n metrics_stepwise.to_csv(results_path+'metrics_stepwise_p0_neg.csv')\n if \"c\" in x: # convergence tolerance experiments\n if compact>1:\n metrics_stepwise, distributions_stepwise,\\\n metrics_converged, distributions_converged,\\\n metrics_stepwise_allreps, distributions_stepwise_allreps,\\\n metrics_converged_allreps, distributions_converged_allreps = test_ctol(pxy,compact)\n metrics_converged.to_csv(results_path+'metrics_converged_ctol.csv')\n metrics_stepwise.to_csv(results_path+'metrics_stepwise_ctol.csv')\n distributions_converged.to_pickle(results_path+'distributions_converged_ctol.pkl')\n elif compact>0:\n metrics_stepwise,\\\n metrics_converged, distributions_converged,\\\n metrics_stepwise_allreps,\\\n metrics_converged_allreps, distributions_converged_allreps = test_ctol(pxy,compact)\n metrics_converged.to_csv(results_path+'metrics_converged_ctol.csv')\n metrics_stepwise.to_csv(results_path+'metrics_stepwise_ctol.csv')\n distributions_converged.to_pickle(results_path+'distributions_converged_ctol.pkl')\n else:\n metrics_stepwise,\\\n metrics_converged,\\\n metrics_stepwise_allreps,\\\n metrics_converged_allreps = test_ctol(pxy,compact)\n metrics_converged.to_csv(results_path+'metrics_converged_ctol.csv')\n metrics_stepwise.to_csv(results_path+'metrics_stepwise_ctol.csv')\n if \"z\" in x: # zeroL tolerance experiments\n if compact>1:\n metrics_stepwise, distributions_stepwise,\\\n metrics_converged, distributions_converged,\\\n metrics_stepwise_allreps, distributions_stepwise_allreps,\\\n metrics_converged_allreps, distributions_converged_allreps = test_zeroLtol(pxy,compact)\n metrics_converged.to_csv(results_path+'metrics_converged_zeroLtol.csv')\n metrics_stepwise.to_csv(results_path+'metrics_stepwise_zeroLtol.csv')\n distributions_converged.to_pickle(results_path+'distributions_converged_zeroLtol.pkl')\n elif compact>0:\n metrics_stepwise,\\\n metrics_converged, distributions_converged,\\\n metrics_stepwise_allreps,\\\n metrics_converged_allreps, distributions_converged_allreps = test_zeroLtol(pxy,compact)\n metrics_converged.to_csv(results_path+'metrics_converged_zeroLtol.csv')\n metrics_stepwise.to_csv(results_path+'metrics_stepwise_zeroLtol.csv')\n distributions_converged.to_pickle(results_path+'distributions_converged_zeroLtol.pkl')\n else:\n metrics_stepwise,\\\n metrics_converged,\\\n metrics_stepwise_allreps,\\\n metrics_converged_allreps = test_zeroLtol(pxy,compact)\n metrics_converged.to_csv(results_path+'metrics_converged_zeroLtol.csv')\n metrics_stepwise.to_csv(results_path+'metrics_stepwise_zeroLtol.csv')\n if \"b\" in x: # trying proposed optimal beta\n if compact>1:\n metrics_stepwise, distributions_stepwise,\\\n metrics_converged, distributions_converged,\\\n metrics_stepwise_allreps, distributions_stepwise_allreps,\\\n metrics_converged_allreps, distributions_converged_allreps = test_IB(pxy,compact)\n metrics_converged.to_csv(results_path+'metrics_converged_bestbeta.csv')\n metrics_stepwise.to_csv(results_path+'metrics_stepwise_bestbeta.csv')\n distributions_converged.to_pickle(results_path+'distributions_converged_bestbeta.pkl')\n elif compact>0:\n metrics_stepwise,\\\n metrics_converged, distributions_converged,\\\n metrics_stepwise_allreps,\\\n metrics_converged_allreps, distributions_converged_allreps = test_IB(pxy,compact)\n metrics_converged.to_csv(results_path+'metrics_converged_bestbeta.csv')\n metrics_stepwise.to_csv(results_path+'metrics_stepwise_bestbeta.csv')\n distributions_converged.to_pickle(results_path+'distributions_converged_bestbeta.pkl')\n else:\n metrics_stepwise,\\\n metrics_converged,\\\n metrics_stepwise_allreps,\\\n metrics_converged_allreps = test_IB(pxy,compact)\n metrics_converged.to_csv(results_path+'metrics_converged_bestbeta.csv')\n metrics_stepwise.to_csv(results_path+'metrics_stepwise_bestbeta.csv')\n return 0","def run(self):\n self.evaluate()\n self.accumulate()\n self.summarize()","def run(self, verbose=0):\n self.verbose = verbose\n self._preproc()\n self._lda()\n self._evaluate()","def main(titles, paragraphs, printDetailedStats = True):\n avg_precision = 0 # all the precision values will be added to this value. In the end, this value will be divided by\n # the number of comparisons analysed to report the average precision\n avg_recall = 0\n avg_fmeasure = 0\n\n if paragraphs:\n manual_directory = path.MANUAL_PARAGRAPHS\n auto_directory = path.OUTDIR_PARAGRAPHS\n else:\n manual_directory = path.MANUAL_SENTENCES\n auto_directory = path.OUTDIR_SENTENCES\n\n for title in titles:\n with open(manual_directory + title) as m:\n with open(auto_directory + title) as a:\n tPositive, fPositive, fNegative, precision, recall, fmeasure = \\\n compare(m.readlines(), a.readlines(), paragraphs)\n avg_precision += precision\n avg_recall += recall\n avg_fmeasure += fmeasure\n if printDetailedStats: # then report precision and recall for every individual article\n print(\"comparing \" + title)\n print(\"tp=\" + str(tPositive) + \" fn=\" + str(fNegative) + \" fp=\" + str(fPositive))\n print(\"precision=\" + str(round(precision, 5)) + \"\\t\\t recall=\" + str(\n round(recall, 5)) + \"\\t\\t fmeasure=\" + str(round(fmeasure, 5)) + \" \\n\\n\")\n print(\"AVERAGE_PRECISION=\" + str(round(avg_precision/len(titles), 5)) + \"\\t\\t AVERAGE_RECALL=\"\n + str(round(avg_recall/len(titles), 5)) + \"\\t\\t AVERAGE_F_MEASURE=\"\n + str(round(avg_fmeasure/len(titles), 5)))\n return avg_fmeasure/len(titles)","def run_test(models, data):\n num_correct = 0.0\n num_total = 0.0\n\n # initialize lists to put in predictions & results for confusion matrix \n predicted_labels = []\n actual_labels = []\n langlist = models.keys()\n for ai, actual_lang in enumerate(langlist):\n test_files = open(os.path.join('traintestsplit', actual_lang+'.testlist')).read().split()\n print 'Testing', len(test_files), 'files from', actual_lang\n for filename in test_files:\n logprobs = {} # dict: total log prob of this file under each model \n for test_lang in langlist:\n logprobs[test_lang] = apply_model(models[test_lang], data[actual_lang][filename+'.npytxt'])\n predicted_lang = max(logprobs.items(), key=lambda x:x[1])[0]\n # insert prediction (of lang index) into predicted list \n predicted_labels.append(langlist.index(predicted_lang))\n actual_labels.append(ai)\n if actual_lang == predicted_lang:\n num_correct += 1\n num_total += 1\n\n print len(filter(lambda x:x==ai, predicted_labels[-len(test_files):])), 'correct'\n\n print\n print 'Accuracy', num_correct*100/num_total\n\n #CONFUSION MATRIX (y_test, y_pred) -> (actual label, predictions) \n cm = confusion_matrix(actual_labels, predicted_labels)\n cm_normalized = cm.astype('float') / cm.sum(axis=1)[:, np.newaxis]\n # display confusion stats by lang (TODO: visualize with matplotlib) \n print '*'*20\n for ai, actual_lang in enumerate(langlist):\n print actual_lang, 'confusion:'\n for pi, predicted_lang in enumerate(langlist):\n print '{0}: {1:.2f}%'.format(predicted_lang, cm_normalized[ai, pi]*100)\n print '*'*20","def test_print_results(self):\n calculated = super().predict_and_print()\n self.assertEqual(calculated, EXP_PRINT_OUTPUT_BASE.format(.18, .1, 0.186, self.test_model.model.train_time) +\n \"Max tree max_depth: 1\\n\"\n \"Number of n_estimators: 1\\n\"\n \"Impurity method: entropy\\n\")","def run_tests(): \n \n\n nextdata = [[21, 61, 42, 30], [33,45, 18, 29]]\n\n for xval, yval, snum, expect in nextdata:\n\n pmachine = PMachine()\n pmachine.serial_number = snum\n pmachine.run2_completion()\n result = pmachine.calc_square_total(xval, yval, showsquare=True)\n assert result == expect\n print(\"Got value {}={} as expected\".format(result, expect))","def run(self):\n logging.info('running experiment...')\n self._prepare()\n self._load_data()\n self._run()\n self._evaluate()\n self._summarise()\n return True","def exe_tests(self):\n self.rank = mpicom.rank()\n self.size = mpicom.size()\n if mpicom.parallel():\n self.test(\"libname\",os.path.split(mpicom.__file__)[1],\"mpicom.so\")\n else:\n self.test(\"libname\",os.path.split(mpicom.__file__)[1],\"mpistub.pyc\")\n self.test_broadcast()\n self.test_reduce()\n self.test_p2p()\n self.test_gather()\n self.test_scatter()\n #self.test_alltoall()","def _compute_experiment_statistics(self):\n pass","def run(self):\n evaluateAllRuns = False\n while True:\n if self.host == \"\":\n # respond to clients\n self.respond2Clients()\n else:\n print(\"Next...\")\n # randomly choose experiment + run\n if not evaluateAllRuns:\n print(\"Randomly fetching run\")\n self.exp, self.runnum, self.detname = randExpRunDet()\n else:\n\t\t try:\n print(\"Fecthing next run in experiment\")\n self.exp, self.runnum, self.detname = nextExpRunDet(self.goodExp, self.runList[0])\n if self.exp is None:\n self.runList.pop(0)\n continue\n except:\n evaluateAllRuns = False\n continue\n if not self.checkStatus(self.exp, self.runnum, self.detname):\n print \"trying: exp %s, run %s, det %s\"%(self.exp,self.runnum,self.detname)\n try: #temp\n self.ds = safeDataSource(self.exp, self.runnum)\n except: #temp\n continue #temp\n self.run = self.ds.runs().next()\n self.times = self.run.times()\n #Start temp code\n if self.detname is None:\n continue\n #End temp code\n self.det = psana.Detector(self.detname)\n self.det.do_reshape_2d_to_3d(flag=True)\n try:\n self.iX = np.array(self.det.indexes_x(self.run), dtype=np.int64)\n self.iY = np.array(self.det.indexes_y(self.run), dtype=np.int64)\n self.ipx, self.ipy = self.det.point_indexes(self.run, pxy_um=(0, 0))\n self.alg = PyAlgos()\n self.alg.set_peak_selection_pars(npix_min=2, npix_max=30, amax_thr=300, atot_thr=600, son_min=10)\n mask = self.det.mask(self.runnum, calib=True, status=True, edges=True, central=True, unbond=True, unbondnbrs=True)\n\n samples = np.linspace(0, len(self.times), num=100, endpoint=False, retstep=False, dtype='int')\n offset = np.floor(np.random.uniform(0, len(self.times)-samples[-1])).astype('int')\n mysamples = samples + offset\n numCrystals = 0\n for self.eventNum in mysamples:\n self.evt = self.run.event(self.times[self.eventNum])\n calib = self.det.calib(self.evt)\n if calib is not None:\n peaks = self.alg.peak_finder_v3r3(calib, rank=3, r0=3, dr=2, nsigm=10, mask=mask.astype(np.uint16))\n if self.likelihood(peaks) >= self.goodLikelihood:\n numCrystals += 1\n if numCrystals >= self.minCrystals:\n self.numSaved +=1\n self.updateStatus(self.exp, self.runnum, self.detname, self.numSaved)\n self.lastGood = True\n break\n except:\n print \"Could not analyse this run\"\n #If an experiment has not had all of its runs evaluated yet\n # and if the last randomly selected run in this experiment was good\n # then all the runs in this experiment should be evaluated\n if (self.exp not in self.goodList) and self.lastGood:\n self.goodExp = self.exp #Save the name of this experiment\n self.goodRun = self.runnum #Save the run that has already been evaluated\n self.lastGood = False #Reset the condition that the last run was \"good\"\n self.goodList.append(self.goodExp) #Add this experiment name to the list of experiments that have had all runs evaluated\n self.runList = returnRunList(self.goodExp, self.goodRun) #save list of all runs in this good exp\n evaluateAllRuns = True #rerun loop with new algorithm that evaluates each run in an experiment\n continue\n if evaluateAllRuns: #If the loop is currently evaluating all of the runs in an experiment\n if(len(self.runList) > 1):\n self.runList.pop(0) #Remove runs from the list of runs each time they are evaluated\n else:\n self.runList.pop(0)#Remove runs until the list is completely empty\n evaluateAllRuns = False #Stop evaluated all the runs of an experiment, go back to random fetching","def getTestResults():","def accuracy(num_points, num_dims, p, surv, num_runs=100, num_test=10000):\n p1 = []\n p2 = []\n p3 = []\n p4 = []\n for t in range(num_runs):\n if surv:\n cloud = generate_survival_curves(num_points + num_test, num_dims)\n else:\n cloud = generate_time_series(num_points + num_test, num_dims)\n opt_cloud = cloud[:num_points]\n test_cloud = cloud[num_points:]\n p1.append(proportion(test_cloud, olshen(opt_cloud, p, surv)))\n p2.append(proportion(test_cloud, two_sided_olshen(opt_cloud, p, surv)))\n p3.append(proportion(test_cloud, gspie(opt_cloud, p, surv)))\n p4.append(\n proportion(test_cloud, pointwise_bonferroni(opt_cloud, p, surv)))\n plt.boxplot([p1, p2, p3, p4])\n plt.xticks([1, 2, 3, 4],\n ['Olshen', 'Two-Sided Olshen', 'GSPIE', 'Pointwise Bonferroni'])\n plt.ylabel('Observed Coverage')\n plt.xlabel(f\"{p} Prescribed Coverage\")\n plt.show()","def main():\n\n # choose number of data-points and sample a pair of vectors: the input\n # values and the corresponding target values\n N = 500\n inputs, targets = sample_data(N, arbitrary_function_2, seed=1)\n\n # specify the centres and scale of some rbf basis functions\n default_centres = np.linspace(0,1,21)\n default_scale = 0.03\n default_reg_param = 0.08\n\n # get the cross-validation folds\n num_folds = 4\n folds = create_cv_folds(N, num_folds)\n\n # evaluate then plot the performance of different reg params\n evaluate_reg_param(inputs, targets, folds, default_centres, default_scale)\n # evaluate then plot the performance of different scales\n evaluate_scale(inputs, targets, folds, default_centres, default_reg_param)\n # evaluate then plot the performance of different numbers of basis\n # function centres.\n evaluate_num_centres(\n inputs, targets, folds, default_scale, default_reg_param)\n\n plt.show()","def see_evaluation(epoch, training_acc, test_acc):\n print (\"Epoch \", epoch, \"Training acc: \", training_acc*100, \"Test acc: \", test_acc*100)","def evaluate(model, test_files):\n print(\"Running predictions.\")\n models = load_model(model)\n predictions = predict(models, test_files)\n\n # # write predictions to file\n # write_predictions(\"evaluate_out.json\",predictions)\n evaluate_individual(predictions, test_files, models)\n evaluate_overall(predictions)","def sbm_experiments():\n # The key experiment is to fix n, r, and p and vary the ratio q/p.\n n = 1000\n average_over = 10\n\n # We will run a few experiments\n rs = [5]\n ps = [1e-11]\n run_lps = [False]\n\n # Use the same ratios for each experiment\n ratios = [0.5 * x for x in range(4, 61)]\n\n # Whether to append results to the results files\n append_results = False\n\n for index in range(len(rs)):\n sbm_experiment_internal(n, rs[index], ps[index], average_over, run_lps[index], ratios, append_results)","def run_tests(self) -> float:\n print(f\"Grading {self._test_case.__name__}\", file=sys.stderr)\n print(file=sys.stderr)\n for test, weight in self.test_weights.items():\n print(f\"Running {test.__name__}\", file=sys.stderr)\n if test.__doc__:\n print(test.__doc__, file=sys.stderr)\n result = self.run(test)\n if result.wasSuccessful():\n print(f\"Points: {weight}/{weight}\", file=sys.stderr)\n else:\n print(file=sys.stderr)\n print(\n \"Test failed with the error below, displayed between lines of ---.\",\n file=sys.stderr,\n )\n print(\n \"The expected value is given first, followed by the actual result.\",\n file=sys.stderr,\n )\n print(\"-\" * 70, file=sys.stderr)\n # Get the error/failure\n try:\n print(result.errors[0][1], file=sys.stderr)\n except IndexError:\n pass\n try:\n print(result.failures[0][1], file=sys.stderr)\n except IndexError:\n pass\n print(\"-\" * 70, file=sys.stderr)\n print(f\"Points: {0.0}/{weight}\", file=sys.stderr)\n print(file=sys.stderr)\n print(\"=\" * 70, file=sys.stderr)\n return self.grade","def print_results(results, random_counterpart=None, random_concepts=None, num_random_exp=100,\n min_p_val=0.05):\n\n # helper function, returns if this is a random concept\n def is_random_concept(concept):\n if random_counterpart:\n return random_counterpart == concept\n\n elif random_concepts:\n return concept in random_concepts\n\n else:\n return 'random500_' in concept\n\n # print class, it will be the same for all\n print(\"Class =\", results[0]['target_class'])\n\n # prepare data\n # dict with keys of concepts containing dict with bottlenecks\n result_summary = {}\n\n # random\n random_i_ups = {}\n\n for result in results:\n if result['cav_concept'] not in result_summary:\n result_summary[result['cav_concept']] = {}\n\n if result['bottleneck'] not in result_summary[result['cav_concept']]:\n result_summary[result['cav_concept']][result['bottleneck']] = []\n\n result_summary[result['cav_concept']][result['bottleneck']].append(result)\n\n # store random\n if is_random_concept(result['cav_concept']):\n if result['bottleneck'] not in random_i_ups:\n random_i_ups[result['bottleneck']] = []\n\n random_i_ups[result['bottleneck']].append(result['i_up'])\n\n # print concepts and classes with indentation\n for concept in result_summary:\n\n # if not random\n if not is_random_concept(concept):\n print(\" \", \"Concept =\", concept)\n\n for bottleneck in result_summary[concept]:\n i_ups = [item['i_up'] for item in result_summary[concept][bottleneck]]\n\n # Calculate statistical significance\n _, p_val = ttest_ind(random_i_ups[bottleneck], i_ups)\n\n print(3 * \" \", \"Bottleneck =\", (\"%s. TCAV Score = %.2f (+- %.2f), \"\n \"random was %.2f (+- %.2f). p-val = %.3f (%s)\") % (\n bottleneck, np.mean(i_ups), np.std(i_ups),\n np.mean(random_i_ups[bottleneck]),\n np.std(random_i_ups[bottleneck]), p_val,\n \"not significant\" if p_val > min_p_val else \"significant\"))","def analyze_end_performance(self) -> None:\n self.logger.info(f\"###### FINAL RESULTS FOR EACH GAME###### \\n\")\n total_results = [0, 0, 0, 0]\n for game in self.instance_predictions:\n self.logger.info(f\"Simulated game {game.home_team.name} vs {game.away_team.name} \"\n f\"{len(self.instance_predictions[game])} times.\")\n confusion_matrix = [0, 0, 0, 0]\n for predicted_winner in self.instance_predictions[game]:\n actual_winner = game.home_team if game.home_team.scores.get(\n GamePeriod.TOTAL) > game.away_team.scores.get(GamePeriod.TOTAL) else game.away_team\n actual_home_team_won = True if actual_winner == game.home_team else False\n predicted_home_team_won = True if predicted_winner == game.home_team else False\n confusion_matrix = recursive_build_confusion_matrix(confusion_matrix, actual_home_team_won,\n predicted_home_team_won)\n total_results[0] += confusion_matrix[0]\n total_results[1] += confusion_matrix[1]\n total_results[2] += confusion_matrix[2]\n total_results[3] += confusion_matrix[3]\n\n for k, v in analyze_confusion_matrix(confusion_matrix, len(self.instance_predictions[game])).items():\n self.logger.info(f\"{k} : {v}\")\n self.logger.info(\"\\n###### FINAL RESULTS FOR ALL GAMES###### \\n\")\n final_stats = analyze_confusion_matrix(total_results, sum(total_results))\n for k, v in final_stats.items():\n self.logger.info(f\"{k} : {v}\")\n self.write_stats_to_json(final_stats, self.outfile)","def check_expectations(self):\n self.load_results()\n\n for (benchmark, producer), result in self.results.items():\n if not result.reports:\n print('No results found for ' + benchmark + ' ' + producer)\n result.test_passed = False\n else:\n for report in result.reports:\n if check_benchmark_result(report, result.expectation):\n print('Test passed: ' + result.directory)\n result.test_passed = True\n else:\n print('Test failed: ' + result.directory)\n result.test_passed = False","def polling_experiment_multiple(dataset_id, times, **kwargs):\n total_yeahs = 0\n total_noes = 0\n for i in range(0, times):\n yeah, noes = polling_experiment(dataset_id, **kwargs)\n total_yeahs += yeah\n total_noes += noes\n logging.info(\"After \"+str(times)+\" \"+ str(dataset_id) + \" experiments:\")\n logging.info(\"\"+ str(total_yeahs) + \" correct predictions\")\n logging.info(\"\"+ str(total_noes) + \" wrong predictions\")\n logging.info(\"\"+ str(total_yeahs/(total_yeahs + total_noes)) + \" accuracy\")","def main():\n\t# \"\"\"\n\t# \tMain function of test python module\n\t# \"\"\"\n\t# random.seed(os.urandom(345634)) # initialize random generator\n\t# t = np.linspace(0.0, 24.0, 96.0) # define the time axis of a day, here we use 96 values every quarter of an hour\n\t# # standard load profile -- input\n\t# q = extra.read_slp(t,\n\t# 'Profielen-Elektriciteit-2015-versie-1.00 Folder/profielen Elektriciteit 2015 versie 1.00.csv') # read the sample standard load profile, can be any length, can be resized given a low/high resolution time axis\n\t# q = q / np.sum(q) # normalization of standard load profile\n\t# # process duration\n\t# duration_axis = np.linspace(0.0, 24.0, 96.0)\n\t# (p_d, E_p) = extra.app_time(duration_axis, 10, 2, 0.0,\n\t# 24.0) # function that define the pdf of duration of a process\n\t# # process consumption\n\t# consumption_axis = np.linspace(0.0, 3.5, 96.0)\n\t# (p_k, E_k) = extra.app_consumption(consumption_axis, 10, 2, 0.0,\n\t# 3.5) # function that define the pdf of duration of a process\n\t# # pdf of starting time\n\t# p_t_0 = lpd.infer_t_0(q, p_d, E_k) # computes the pdf of starting time of processes\n\t# p_t_0 = p_t_0 / np.sum(p_t_0) # normalization of the pdf to sum up to zero\n #\n\t# \"\"\"\n\t# 1st Approach, starting time of processes is a discrete propapibility density function\n\t# \"\"\"\n\t# # synthetic profile of D processes\n\t# D = 2000\n\t# synthetic_profile = lpd.synthetic_profile(D, t, p_d, consumption_axis, p_k, p_t_0)\n\t# synthetic_profile_1 = lpd.synthetic_profile(D, t, p_d, consumption_axis, p_k, p_t_0)\n\t# # expected value of D processes\n\t# q_e_e = lpd.infer_q_e(t, p_t_0, p_d, E_k, D)\n\t# # plot\n\t# plt.step(t, synthetic_profile, \"g-\")\n\t# plt.step(t, q_e_e, \"b--\")\n #\n\t# \"\"\"\n\t# 2nd Approach, starting time of processes is a continuous propapibility density function\n\t# \"\"\"\n\t# # synthetic profile of D processes\n\t# ts, cs = lpd.continous_synthetic_profile(D, t, p_d, consumption_axis, p_k, p_t_0)\n\t# plt.step(ts / len(t) * t[-1], cs, where='post', c='r')\n\t# plt.xlim(0, 24.0)\n\t# plt.legend([\"synthetic\", \"expected\", \"continuous\"], loc=0)\n\t# plt.show()\n #\n\t# \"\"\"\n\t# Time discretization\n\t# \"\"\"\n\t# n_intervals = 24 * 1 # discretized in minutes\n\t# discrete_timeaxis = np.linspace(0.0, 24.0, n_intervals + 1)\n\t# discrete_consumption = lpd.signal_discretization(discrete_timeaxis, t, ts, cs)\n\t# plt.step(ts / len(t) * t[-1], cs, where='post', c='r')\n\t# plt.step(discrete_timeaxis, discrete_consumption, where='post', c='k', ls='--', lw=2)\n\t# plt.legend([\"continuous\", \"discretized\"], loc=0)\n\t# plt.show()\n #\n #\n\t# \"\"\"\n\t# Repeated day synthetic profile creation\n\t# \"\"\"\n\t# # synthetic profile of D processes\n\t# D = 2000\n\t# n = 10\n\t# slp = lpd.synthetic_profile_repeated(D, t, p_d, consumption_axis, p_k, p_t_0, n)\n\t# plt.step(range(len(slp)), slp, \"g-\")\n\t# plt.show()\n\tt = np.linspace(0.0, 24.0, 96.0)\n\tload_profile = extra.read_slp(t, 'Profielen-Elektriciteit-2015-versie-1.00 Folder/profielen Elektriciteit 2015 versie 1.00.csv')\n\tslp = synthetic.create_synthetic_load(load_profile, 5.0, 5)\n\tplt.step(range(len(slp)), slp)\n\tplt.show()","def run_tests(self):\n\n self.test_report = []\n\n #dict of unsorted lists\n dict_of_un_lists = self.dict_un_lists_intersection_test(self.data_dict)\n self.test_report.append(dict_of_un_lists)\n\n #dict of sets\n dict_of_sets = self.build_dict_of_sets(self.data_dict)\n self.test_report.append(self.dict_sets_intersection_test(dict_of_sets))\n\n #pandas - experimental and probably not the way to use pandas\n # dict_of_pandas = self.build_dict_of_panda_series(self.data_dict)\n # self.test_report.append(self.dicts_any_intersection_node_test(dict_of_pandas))\n\n # print results\n\n if self.verbose:\n self.print_tests_results()","def evaluate(clf, dataset, feature_list, features, labels, num_iter, params):\n\n features_train, features_test, labels_train, labels_test = \\\n train_test_split(features, labels, test_size=0.3, random_state=42)\n\n\n\n precision_values = []\n recall_values = []\n accuracy_values = []\n print clf\n for i in xrange(0, num_iter):\n #print params\n clf = GridSearchCV(clf, params)\n clf.fit(features_train, labels_train)\n print '*****************************'\n print clf.best_estimator_\n print clf.best_params_\n\n clf = clf.best_estimator_\n #test_classifier(clf, dataset, feature_list)\n pred = clf.predict(features_test)\n precision_values.append(precision_score(labels_test, pred))\n recall_values.append(recall_score(labels_test, pred))\n accuracy_values.append(accuracy_score(labels_test, pred))\n print 'Recall score: ', mean(recall_values)\n print 'Precision score: ', mean(precision_values)\n print 'Accuracy score: ' , mean(accuracy_values)","def test_compare_classifiers_predictions_distribution_vis_api(experiment_to_use):\n experiment = experiment_to_use\n predictions = experiment.predictions_num\n viz_outputs = (\"pdf\", \"png\")\n with TemporaryDirectory() as tmpvizdir:\n for viz_output in viz_outputs:\n vis_output_pattern_pdf = tmpvizdir + f\"/*.{viz_output}\"\n visualize.compare_classifiers_predictions_distribution(\n [predictions, predictions],\n experiment.ground_truth,\n experiment.ground_truth_metadata,\n experiment.output_feature_name,\n labels_limit=0,\n model_names=[\"Model1\", \"Model2\"],\n output_directory=tmpvizdir,\n file_format=viz_output,\n )\n figure_cnt = glob.glob(vis_output_pattern_pdf)\n assert 1 == len(figure_cnt)","def test(self):\n\n scores = list()\n for ep in range(10):\n current_state = self.env.reset()\n done = False\n score = 0\n while not done:\n action = self.choose_action(current_state, 0)\n obs, reward, done, _ = self.env.step(action)\n current_state = obs\n score += reward\n scores.append(score)\n print(\"Ep: {}, Score: {}\".format(ep, score))\n scores = np.array(scores)\n print(\"Eval => Std: {}, Mean: {}\".format(np.std(scores), np.mean(scores)))","def perform_experiments(problem_type, output_dir, load_experiments):\n\n experiment_file_path = output_dir + \"experiments.pickle\"\n\n # data structure where to store all the problems, and the experimental results for each algorithm: solutions, final values and times\n experiment_dict = dict()\n\n # if experiments should be loaded (not repeated), do it if possible\n if load_experiments:\n with open(experiment_file_path, \"rb\") as _file:\n experiment_dict = pickle.load(_file)\n\n # perform experiments if pre-existing results were not loaded\n if not experiment_dict:\n\n # given a problem type, create a set of problem instances that are solved (optimally) with manual placements (using actions available for all the algorithms)\n problems, problem_names, manual_solutions = create_problems(problem_type)\n\n if problems and problem_names and manual_solutions:\n\n # parameters for the experimentation; note: calculating internal times and value evolution can increase the overall time of algorithms (in a slight, almost neglectible way)\n execution_num = 10 # 1\n process_num = 10 # 1\n calculate_internal_times = True\n calculate_value_evolution = True\n\n start_time = time.time()\n\n # solve each problem with each algorithm\n for i, (problem, problem_name, solution) in enumerate(zip(problems, problem_names, manual_solutions)):\n\n experiment_dict[problem_name] = {\"problem\": problem, \"manual_solution\": solution, \"algorithms\": dict()}\n\n # solve the problem with different algorithms, executing each one multiple times to gain statistical significance\n for (algorithm_name, algorithm) in [(\"Greedy\", greedy.solve_problem), (\"Reversible\", reversible.solve_problem), (\"Evolutionary\", evolutionary.solve_problem)]:\n\n solutions, values, value_evolutions, times, time_divisions = execute_algorithm(algorithm=algorithm, algorithm_name=algorithm_name, problem=problem, execution_num=execution_num, process_num=process_num, calculate_times=calculate_internal_times, calculate_fitness_stats=calculate_value_evolution)\n experiment_dict[problem_name][\"algorithms\"][algorithm_name] = {\"solutions\": solutions, \"values\": values, \"value_evolutions\": value_evolutions, \"times\": times, \"time_divisions\": time_divisions}\n\n # show the total time spent doing experiments (note that significant overhead can be introduced beyond calculation time if plots are shown or saved to files; for strict time measurements, plotting should be avoided altogether)\n print(\"Total experimental calculation time: {} s\".format(round(get_time_since(start_time) / 1000.), 3))\n\n if experiment_dict:\n\n # experiment-saving parameter\n save_experiments = True\n\n # if possible, save the experiments to a binary file\n if not load_experiments and save_experiments:\n with open(experiment_file_path, \"wb\") as file:\n pickle.dump(experiment_dict, file)\n\n # visualization/saving parameters\n show_problem_stats = True\n save_problem_stats = True # False\n show_manual_solution_plots = True\n save_manual_solution_plots = True # False\n show_algorithm_solution_plots = True\n save_algorithm_solution_plots = True # False\n show_value_evolution_plots = False\n save_value_evolution_plots = False # True\n show_time_division_plots = False\n save_time_division_plots = False # True\n show_algorithm_comparison = True\n save_algorithm_comparison = True # False\n show_aggregated_result_tables = True\n save_aggregated_result_tables = True # False\n\n # show/save the results of the experiments\n visualize_and_save_experiments(experiment_dict, output_dir, can_plots_show_value_and_weight=problem_type == KNAPSACK_PACKING_PROBLEM_TYPE, show_problem_stats=show_problem_stats, save_problem_stats=save_problem_stats, show_manual_solution_plots=show_manual_solution_plots, save_manual_solution_plots=save_manual_solution_plots, show_algorithm_solution_plots=show_algorithm_solution_plots, save_algorithm_solution_plots=save_algorithm_solution_plots, show_value_evolution_plots=show_value_evolution_plots, save_value_evolution_plots=save_value_evolution_plots, show_time_division_plots=show_time_division_plots, save_time_division_plots=save_time_division_plots, show_algorithm_comparison=show_algorithm_comparison, save_algorithm_comparison=save_algorithm_comparison, show_aggregated_result_tables=show_aggregated_result_tables, save_aggregated_result_tables=save_aggregated_result_tables)\n\n else:\n print(\"The experiments cannot be performed (there are no problems available).\")","def benchmark(options):\n # Prepare experiments\n with open(options['<benchmark>']) as f:\n benchmark_config = json.loads(f.read())\n generate_agent_configs(benchmark_config)\n experiments = product(benchmark_config['environments'], benchmark_config['agents'], [options])\n\n # Run evaluations\n with Pool(processes=int(options['--processes'])) as pool:\n results = pool.starmap(evaluate, experiments)\n\n # Clean temporary config files\n generate_agent_configs(benchmark_config, clean=True)\n\n # Write evaluations summary\n benchmark_filename = os.path.join(Evaluation.OUTPUT_FOLDER, '{}_{}.{}.json'.format(\n BENCHMARK_FILE, datetime.datetime.now().strftime('%Y%m%d-%H%M%S'), os.getpid()))\n with open(benchmark_filename, 'w') as f:\n json.dump(results, f, sort_keys=True, indent=4)\n gym.logger.info('Benchmark done. Summary written in: {}'.format(benchmark_filename))","def test_make_results_verbose1(self):\n\t\ttest = sentiment.LibraryRun(self.text3, self.lib)\n\t\ttest.do_run()\n\t\ttest.make_results_verbose()\n\t\tobj_ut = test.results_verbose\n\t\tself.assertEqual(obj_ut, [['100', 'not good', 2, -1, 0],\n\t\t\t['100', 'not very good', 4, -1, 0]])","def main():\n\n tests = [500, 1000, 10000]\n results = {\n 'Insertion Sort': 0.0,\n 'Shell Sort': 0.0,\n 'Python Sort': 0.0\n }\n\n for test in tests:\n i = 0\n\n while i < 100:\n test_list = gen_random_list(test)\n results['Insertion Sort'] += insertion_sort(test_list)[0]\n results['Shell Sort'] += shell_sort(test_list)[0]\n results['Python Sort'] += python_sort(test_list)[0]\n i += 1\n\n print(\"Sort results for list of size %s items:\" % test)\n for key, value in results.items():\n print(\"%s took %10.7f seconds to run, on average.\" % (key, (value/100)))\n print(\"\\n\")","def evaluation(deploymentID, start_time,\n end_time, sites, validation, manOrLik):\n \n estScoreBound = 5\n falsePositive_dep = 0\n falseNegative_dep = 0\n truePositive_dep = 0\n trueNegative_dep = 0\n\n db_con = MySQLdb.connect(user=\"root\", db=\"qraat\")\n for i in sites:\n truePositive = 0\n trueNegative = 0\n falsePositive = 0\n falseNegative = 0\n \n cur = db_con.cursor()\n cur.execute(\"\"\"SELECT isPulse, band3, band10, frequency,\n ec, tnp, edsp, fdsp, edsnr, fdsnr\n FROM est INNER JOIN est_class%s\n ON ID = est_class%s.estID\n INNER JOIN estscore2\n ON ID = estscore2.estID\n WHERE timestamp > %s\n AND timestamp < %s\n AND deploymentID = %s\n AND siteID = %s\n AND setNum = %s;\n \"\"\"%(manOrLik, manOrLik, start_time, end_time,\n deploymentID, i, validation))\n for row in cur.fetchall():\n estData = {'band3':row[1], 'band10':row[2],\n 'frequency':row[3], 'ec':row[4], 'tnp':row[5],\n 'edsp':row[6], 'fdsp':row[7],\n 'edsnr':row[8], 'fdsnr':row[9]}\n \n #Classify data\n isPulse = modifiedBC(deploymentID, i, start_time,\n validation, estData, manOrLik)\n\n #Determine whether the classification results are correct or not\n if (row[0] == 1):\n if (isPulse == 1):\n truePositive += 1\n else:\n falseNegative += 1\n else:\n if (isPulse == 1):\n falsePositive += 1\n else:\n trueNegative += 1\n \n falsePositive_dep += falsePositive\n falseNegative_dep += falseNegative\n truePositive_dep += truePositive\n trueNegative_dep += trueNegative\n return [truePositive_dep, trueNegative_dep,\n falsePositive_dep, falseNegative_dep]","def test_quality(approach):\n tp, fn, fp, tn = 0, 0, 0, 0\n with open(\"test.txt\", \"r\") as f:\n test_data = json.load(f)\n for i in test_data:\n rez = approach(i[0][0], i[0][1])\n if rez == 1 and i[1] == 1:\n tp += 1\n elif rez == 0 and i[1] == 1:\n fn += 1\n elif rez == 1 and i[1] == 0:\n fp += 1\n else:\n tn += 1\n print \"Precision: {}%.\\nRecall: {}%.\".format(\n round(tp / (tp + fp), 2), round(tp / (tp + fn), 2))","def run_simulation(n, experiments, iterations, budget, recovery_count, performance_factor, current_top,\r\n precomputed=True, dataset=None):\r\n T = 0\r\n k = 5\r\n delta = 0.1\r\n range_m = 1*np.arange(1, budget+1)\r\n scores, true_top = init(n, precomputed=precomputed, dataset=dataset)\r\n true_ranks = get_ranks(scores)\r\n\r\n for exp in tqdm.tqdm(range(experiments), desc=\"experiments\"):\r\n for itr in tqdm.tqdm(range(iterations), desc=\"iterations\"):\r\n t = 1\r\n count = 0\r\n A = np.arange(n)\r\n P = np.zeros((n, n))\r\n S = np.zeros(n)\r\n r_count = 0\r\n data = []\r\n for b in range(budget):\r\n m = range_m[b]\r\n\r\n est, t, A, P, S, count, r_count, data = sparse_borda(n, t, T, k, A, P, S, delta, scores, m*n, count, r_count, data)\r\n\r\n ranking, ranks, top = get_ranking(n, est)\r\n if(top == true_top):\r\n recovery_count[b][exp] += 1\r\n performance_factor[b][exp] += ranks[true_top]\r\n current_top[b][exp] += true_ranks[top]\r\n\r\n current_top /= iterations\r\n performance_factor /= iterations\r\n\r\n return ranking, ranks, data, scores, true_top, est, recovery_count, performance_factor, current_top","def run():\n\n # Set up environment and agent\n e = Environment() # create environment (also adds some dummy traffic)\n a = e.create_agent(LearningAgent) # create agent\n e.set_primary_agent(a, enforce_deadline=False) # set agent to track\n\n # Now simulate it\n sim = Simulator(e, update_delay=0.0) # reduce update_delay to speed up simulation\n sim.run(n_trials=num_of_experiments) # press Esc or close pygame window to quit\n \n pd.Series(a.success).to_pickle('success_' + exp_id + '.pickle')\n a.Q_table.to_pickle('qtable_' + exp_id + '.pickle')\n pd.Series(a.q_delta_avg).to_pickle('convergence_' + exp_id + '.pickle')\n pd.Series(a.t_total).to_pickle('steps_' + exp_id + '.pickle')","def run_example(num_points_to_sample=1000, verbose=True, **kwargs):\n\n exp = Experiment([[1, 52], [0, 6], [1, 52]]) # 2D experiment, we build a tensor product domain\n # Bootstrap with some known or already sampled point(s)\n exp.historical_data.append_sample_points([\n SamplePoint([26, 2, 46], get_fitness([26, 2, 35]), 0.5), # Iterables of the form [point, f_val, f_var] are also allowed\n ])\n # Sample num_points_to_sample points\n for i in range(num_points_to_sample):\n # Use MOE to determine what is the point with highest Expected Improvement to use next\n next_point_to_sample = map(round, gp_next_points(exp, **kwargs)[0]) # in [A, X, B] form, rounded integers\n value_of_next_point = get_fitness(next_point_to_sample)\n\n if verbose:\n if in_results(next_point_to_sample):\n print '***', \"Sampled f({0:s}) = {1:.18E}\".format(str(next_point_to_sample), value_of_next_point), '***'\n else:\n print \"Sampled f({0:s}) = {1:.18E}\".format(str(next_point_to_sample), value_of_next_point)\n\n bank[i,0:3] = next_point_to_sample\n bank[i,3] = value_of_next_point\n # Add the information about the point to the experiment historical data to inform the GP\n exp.historical_data.append_sample_points([SamplePoint(next_point_to_sample, value_of_next_point, 0.01)]) # We can add some noise"],"string":"[\n \"def run_experiments():\\n if True: # Change to False when done finding max_scoring_num_rolls\\n six_sided_max = max_scoring_num_rolls(six_sided)\\n print('Max scoring num rolls for six-sided dice:', six_sided_max)\\n rerolled_max = max_scoring_num_rolls(reroll(six_sided))\\n print('Max scoring num rolls for re-rolled dice:', rerolled_max)\\n\\n if False: # Change to True to test always_roll(8)\\n print('always_roll(8) win rate:', average_win_rate(always_roll(8)))\\n\\n if False: # Change to True to test bacon_strategy\\n print('bacon_strategy win rate:', average_win_rate(bacon_strategy))\\n\\n if False: # Change to True to test swap_strategy\\n print('swap_strategy win rate:', average_win_rate(swap_strategy))\\n\\n \\\"*** You may add additional experiments as you wish ***\\\"\",\n \"def run_experiments():\\n if True: # Change to False when done finding max_scoring_num_rolls\\n six_sided_max = max_scoring_num_rolls(six_sided)\\n print('Max scoring num rolls for six-sided dice:', six_sided_max)\\n four_sided_max = max_scoring_num_rolls(four_sided)\\n print('Max scoring num rolls for four-sided dice:', four_sided_max)\\n\\n if False: # Change to True to test always_roll(8)\\n print('always_roll(8) win rate:', average_win_rate(always_roll(8)))\\n\\n if True: # Change to True to test bacon_strategy\\n print('bacon_strategy win rate:', average_win_rate(bacon_strategy))\\n\\n if True: # Change to True to test swap_strategy\\n print('swap_strategy win rate:', average_win_rate(swap_strategy))\\n\\n if True:\\n print('final_strategy win rate:', average_win_rate(final_strategy))\\n\\n \\\"*** You may add additional experiments as you wish ***\\\"\",\n \"def run_experiments():\\n if False: # Change to False when done finding max_scoring_num_rolls\\n six_sided_max = max_scoring_num_rolls(six_sided)\\n print('Max scoring num rolls for six-sided dice:', six_sided_max)\\n\\n if False: # Change to True to test always_roll(8)\\n print('always_roll(8) win rate:', average_win_rate(always_roll(8)))\\n\\n if False: # Change to True to test bacon_strategy\\n print('bacon_strategy win rate:', average_win_rate(bacon_strategy))\\n\\n if False: # Change to True to test swap_strategy\\n print('swap_strategy win rate:', average_win_rate(swap_strategy))\\n\\n if True: # Change to True to test final_strategy\\n print('final_strategy win rate:', average_win_rate(final_strategy))\\n\\n \\\"*** You may add additional experiments as you wish ***\\\"\",\n \"def main():\\n run_test_summary1a()\\n run_test_summary1c()\\n run_test_summary1c()\",\n \"def run_experiments():\\r\\n result = eval_strategy_range(always_roll, 1, 10)\\r\\n print('Best always_roll strategy:', result)\\r\\n\\r\\n if True: # Change to True when ready to test make_comeback_strategy\\r\\n result = eval_strategy_range(make_comeback_strategy, 5, 15)\\r\\n print('Best comeback strategy:', result)\\r\\n\\r\\n if True: # Change to True when ready to test make_mean_strategy\\r\\n result = eval_strategy_range(make_mean_strategy, 1, 10)\\r\\n print('Best mean strategy:', result)\\r\\n\\r\\n \\\"*** You may add additional experiments here if you wish ***\\\"\",\n \"def all_experiments():\\n elo_explain_experiments()\\n alpha_beta_experiments()\\n mtcs_experiments()\",\n \"def run_experiments():\\n nets = []\\n nets.extend(shallow(n=1))\\n nets.extend(basic_conv(n=1))\\n nets.extend(omit_FC(n=1))\\n nets.extend(dbl_conv(n=1, activation_fn=sigmoid))\\n # omitted, but still interesting: regularized_dbl_conv()\\n nets.extend(dbl_conv_relu(n=1))\\n nets.extend(expanded_data(n=1, FC_n=100))\\n nets.extend(expanded_data(n=1, FC_n=300))\\n nets.extend(expanded_data(n=1, FC_n=1000))\\n nets.extend(expanded_data_double_fc(n=1, FC_n=100)) \\n nets.extend(expanded_data_double_fc(n=1, FC_n=300))\\n nets.extend(expanded_data_double_fc(n=1, FC_n=1000))\\n nets.extend(double_fc_dropout(0.5, 0.5, 0.5, n=5))\\n # plot the erroneous digits in the ensemble of nets just trained\\n error_locations, erroneous_predictions = ensemble(nets)\\n plt = plot_errors(error_locations, erroneous_predictions)\\n plt.savefig(\\\"ensemble_errors.png\\\")\\n # plot the filters learned by the first of the nets just trained\\n plt = plot_filters(nets[0], 0, 5, 4)\\n plt.savefig(\\\"net_full_layer_0.png\\\")\\n plt = plot_filters(nets[0], 1, 8, 5)\\n plt.savefig(\\\"net_full_layer_1.png\\\")\",\n \"def run_experiments(size, approach):\\n\\n # Create folder (if needed) where to store query evaluation logs and raw results\\n if not os.path.isdir(f\\\"{path_reports}/eval-query-{current_run}\\\"):\\n shell(f\\\"mkdir -p {path_reports}/eval-query-{current_run}\\\")\\n\\n # Create folder (if needed) where to store store evaluation logs and raw results\\n if not os.path.isdir(f\\\"{path_reports}/eval-store-{current_run}\\\"):\\n shell(f\\\"mkdir -p {path_reports}/eval-store-{current_run}\\\")\\n\\n # Helper function computing the path of a file/folder for the given query evaluation iteration \\n def query_path(iteration, filename=None):\\n folder = size[\\\"id\\\"] + \\\"_\\\" + str(iteration)\\n return path_reports + \\\"/eval-query-\\\" + current_run + \\\"/\\\" + folder + (\\\"/\\\" + filename if filename != None else \\\"\\\") \\n\\n # Helper function computing the path of a file/folder for the given store evaluation test / cpu setting\\n def store_path(test, cpu, filename=None):\\n folder = size[\\\"id\\\"] + \\\"_\\\" + approach[\\\"id\\\"] + \\\"_\\\" + test + (\\\"_\\\" + cpu[\\\"id\\\"] if cpu != None else \\\"\\\")\\n return path_reports + \\\"/eval-store-\\\" + current_run + \\\"/\\\" + folder + (\\\"/\\\" + filename if filename != None else \\\"\\\")\\n \\n # Determine whether partial traces and named graphs are supported\\n partial = approach[\\\"supports_partial\\\"]\\n graphs = approach[\\\"supports_graphs\\\"]\\n \\n # Skip setting if all data is available (check for presence of log files - delete them to repeat test)\\n may_skip = (not sp_enable or not partial or not graphs or os.path.isfile(store_path(\\\"sp\\\", None, \\\"eval.log\\\")))\\n may_skip = may_skip and (not sf_enable or os.path.isfile(store_path(\\\"sf\\\", None, \\\"eval.log\\\")))\\n if query_enable and approach[\\\"id\\\"] == query_approach_id:\\n for i in range(0, query_num_iterations):\\n may_skip = may_skip and os.path.isfile(query_path(i, \\\"eval.log\\\"))\\n for cpu in cpus:\\n may_skip = may_skip and (not pp_enable or not partial or not graphs or os.path.isfile(store_path(\\\"pp\\\", cpu, \\\"eval.log\\\")))\\n may_skip = may_skip and (not pf_enable or os.path.isfile(store_path(\\\"pf\\\", cpu, \\\"eval.log\\\")))\\n if may_skip:\\n return\\n\\n # Delete (if needed) and extract again the repository from its .tar.xz file, so to work on a clean repository (at the end of this test, the repository is no more clean)\\n path_dump = path_data + \\\"/\\\" + size[\\\"id\\\"] + \\\"_\\\" + approach[\\\"id\\\"] + \\\".tar.lz\\\"\\n path_repo = path_repos + \\\"/\\\" + size[\\\"id\\\"] + \\\"_\\\" + approach[\\\"id\\\"]\\n if not os.path.isfile(path_dump):\\n log(f\\\"Missing required file {path_dump}\\\")\\n sys.exit()\\n if os.path.isdir(path_repo):\\n shell(f\\\"rm -Rf {path_repo}\\\")\\n shell(f\\\"{cmd_plzip} -kdc {path_dump} | tar xf - -C {path_repos}\\\")\\n\\n # Locate the repository URL\\n repo_url = f\\\"http://localhost:{server_port}/repositories/promo\\\"\\n \\n # Query test (if enabled)\\n if query_enable and approach[\\\"id\\\"] == query_approach_id:\\n for i in range(0, query_num_iterations):\\n if not os.path.isfile(query_path(i, \\\"eval.log\\\")):\\n shell(f\\\"mkdir -p {query_path(i)}\\\")\\n shell(f\\\"taskset -a {query_taskset} {cmd_graphdb} start {path_repo}\\\")\\n shell(f\\\"taskset -a {query_taskset} {cmd_mokip} eval-query -w -u {repo_url} -q {path_queries} -r {query_path(i)} \\\" +\\n f\\\"| tee {query_path(i, 'eval.log')}\\\")\\n shell(f\\\"taskset -a {query_taskset} {cmd_graphdb} stop {path_repo}\\\")\\n\\n # Sequential Partial test (to assess store times per trace and their components)\\n if sp_enable and partial and graphs and not os.path.isfile(store_path(\\\"sp\\\", None, \\\"eval.log\\\")):\\n shell(f\\\"mkdir -p {store_path('sp', None)}\\\")\\n shell(f\\\"taskset -a {sp_taskset} {cmd_graphdb} start {path_repo}\\\")\\n shell(f\\\"taskset -a {sp_taskset} {cmd_mokip} eval-store -d {path_data}/traces_pp.jsonl.gz \\\" + \\n f\\\"-u {repo_url} -i {approach['inference']} -U REPLACE_GRAPH_PROTOCOL \\\" + \\n f\\\"-o {path_ontology} --trace-namespace '{namespace_trace}' --graph-namespace '{namespace_graph}' \\\" +\\n f\\\"-T {timeout} -r {store_path('sp', None)} -t 1 -w 50 -p -D \\\" + \\n f\\\"| tee {store_path('sp', None, 'eval.log')}\\\")\\n shell(f\\\"taskset -a {sp_taskset} {cmd_graphdb} stop {path_repo}\\\")\\n\\n # Parallel Partial (to assess throughput, varying # of CPU cores)\\n for cpu in cpus:\\n if pp_enable and partial and graphs and not os.path.isfile(store_path(\\\"pp\\\", cpu, \\\"eval.log\\\")):\\n shell(f\\\"mkdir -p {store_path('pp', cpu)}\\\")\\n shell(f\\\"taskset -a {cpu['taskset']} {cmd_graphdb} start {path_repo}\\\")\\n shell(f\\\"taskset -a {cpu['taskset']} {cmd_mokip} eval-store -d {path_data}/traces_pp.jsonl.gz \\\" +\\n f\\\"-u {repo_url} -i {approach['inference']} -U REPLACE_GRAPH_PROTOCOL \\\" + \\n f\\\"-o {path_ontology} --trace-namespace '{namespace_trace}' --graph-namespace '{namespace_graph}' \\\" + \\n f\\\"-T {timeout} -r {store_path('pp', cpu)} -t {max(2, cpu['num_threads'])} -w 50 -p -D \\\" + \\n f\\\"| tee {store_path('pp', cpu, 'eval.log')}\\\")\\n shell(f\\\"taskset -a {cpu['taskset']} {cmd_graphdb} stop {path_repo}\\\")\\n\\n # Sequential Full test (to assess store times per trace and their components)\\n if sf_enable and not os.path.isfile(store_path(\\\"sf\\\", None, \\\"eval.log\\\")):\\n shell(f\\\"mkdir -p {store_path('sf', None)}\\\")\\n shell(f\\\"taskset -a {sf_taskset} {cmd_graphdb} start {path_repo}\\\")\\n shell(f\\\"taskset -a {sf_taskset} {cmd_mokip} eval-store -d {path_data}/traces_sf.jsonl.gz \\\" + \\n f\\\"-u {repo_url} -i {approach['inference']} -U {'APPEND' if graphs else 'APPEND_DEFAULT_GRAPH'} \\\" +\\n f\\\"-o {path_ontology} --trace-namespace '{namespace_trace}' --graph-namespace '{namespace_graph}' \\\" +\\n f\\\"-T {timeout} -r {store_path('sf', None)} -t 1 -w 50 \\\" + \\n f\\\"| tee {store_path('sf', None, 'eval.log')}\\\")\\n shell(f\\\"taskset -a {sf_taskset} {cmd_graphdb} stop {path_repo}\\\")\\n\\n # Parallel Full (to assess throughput where data is also deleted, varying # of CPU cores)\\n for cpu in cpus:\\n if pf_enable and not os.path.isfile(store_path(\\\"pf\\\", cpu, \\\"eval.log\\\")):\\n update = \\\"APPEND\\\" if graphs else \\\"APPEND_DEFAULT_GRAPH\\\"\\n shell(f\\\"mkdir -p {store_path('pf', cpu)}\\\")\\n shell(f\\\"taskset -a {cpu['taskset']} {cmd_graphdb} start {path_repo}\\\")\\n shell(f\\\"taskset -a {cpu['taskset']} {cmd_mokip} eval-store -d {path_data}/traces_pf_{cpu['id']}.jsonl.gz \\\" +\\n f\\\"-u {repo_url} -i {approach['inference']} -U {'APPEND' if graphs else 'APPEND_DEFAULT_GRAPH'} \\\" + \\n f\\\"-o {path_ontology} --trace-namespace '{namespace_trace}' --graph-namespace '{namespace_graph}' \\\" +\\n f\\\"-T {timeout} -r {store_path('pf', cpu)} -t {max(2, cpu['num_threads'])} -w 50 \\\" + \\n f\\\"| tee {store_path('pf', cpu, 'eval.log')}\\\")\\n shell(f\\\"taskset -a {cpu['taskset']} {cmd_graphdb} stop {path_repo}\\\")\\n\\n # Drop the repository (both to save space and since it is not clean anymore)\\n shell(f\\\"rm -Rf {path_repo}\\\")\",\n \"def experiment1_outliers():\\n\\tdata_folder = \\\"ckan_subset/prepared_learnset/\\\"\\n\\ttest_folder = 'ckan_subset/testset/xml_csv/'\\n\\tgm = Graph_Maker()\\n\\tgm.store()\\n\\trounds = 5\\n\\tx = [\\\"Fingerprint\\\", \\\"Syntax Feature Model\\\", \\\"Word2Vec Matcher\\\"]\\n\\t\\n\\tnumber_of_classes = 15\\n\\texamples_per_class = 0\\n\\taccuracies = []\\n\\tprecisions = []\\n\\trecalls = []\\n\\tfmeasures = []\\n\\tsf_main = Storage_Files(data_folder, classes)\\n\\ttmp_acc = []\\n\\ttmp_prec = []\\n\\ttmp_rec = []\\n\\ttmp_fmeasure = []\\n\\ttotal_actual = []\\n\\ttotal_predicted = []\\n\\n\\tfor i in range(0, rounds):\\n\\t\\tprint(\\\"Fingerprint\\\")\\n\\t\\t# --- Fingerprint\\n\\t\\tccc = Column_Classification_Config()\\n\\t\\tccc.add_feature('feature_main', 'Fingerprint', [sf_main, number_of_classes, examples_per_class, False, False])\\n\\n\\t\\tccc.add_matcher('matcher', 'Fingerprint_Matcher', {'feature_main': 'fingerprint'}) # main classifier\\n\\t\\tsm = Schema_Matcher(ccc)\\n\\t\\tactual, predicted = execute_test_ckan(sm, test_folder, False)\\n\\t\\ttotal_actual += actual\\n\\t\\ttotal_predicted += predicted\\n\\t\\taccuracy = accuracy_score(actual, predicted)\\n\\t\\ttmp_acc.append(accuracy)\\n\\t\\ttmp_prec.append(precision(actual, predicted))\\n\\t\\ttmp_rec.append(recall(actual, predicted))\\n\\t\\ttmp_fmeasure.append(f_measure(actual, predicted))\\n\\n\\taccuracies.append( round(sum(tmp_acc) / float(rounds), 2) )\\n\\tprecisions.append( round(sum(tmp_prec) / float(rounds), 2) )\\n\\trecalls.append( round(sum(tmp_rec) / float(rounds), 2) )\\n\\tfmeasures.append(round(sum(tmp_fmeasure) / float(rounds), 2))\\n\\tclassnames = list(set(get_class_names(total_actual) + get_class_names(total_predicted)))\\n\\tcm = confusion_matrix(total_actual, total_predicted, labels=classnames)\\n\\t#gm.plot_confusion_matrix(cm, classnames, normalize=True)\\n\\t\\n\\ttmp_acc = []\\n\\ttmp_prec = []\\n\\ttmp_rec = []\\n\\ttmp_fmeasure = []\\n\\ttotal_actual = []\\n\\ttotal_predicted = []\\n\\tfor i in range(0, rounds):\\n\\t\\tprint(\\\"SFM\\\")\\n\\t\\t# --- Syntax Feature Model\\n\\t\\tccc = Column_Classification_Config()\\n\\t\\tccc.add_feature('feature_main', 'Syntax_Feature_Model', [sf_main, 1, 0, False, False])\\n\\n\\t\\tccc.add_matcher('matcher', 'Syntax_Matcher', {'feature_main': 'syntax'}) # main classifier\\n\\t\\tsm = Schema_Matcher(ccc)\\n\\t\\tactual, predicted = execute_test_ckan(sm, test_folder, False)\\n\\t\\ttotal_actual += actual\\n\\t\\ttotal_predicted += predicted\\n\\t\\taccuracy = accuracy_score(actual, predicted)\\n\\t\\ttmp_acc.append(accuracy)\\n\\t\\ttmp_prec.append(precision(actual, predicted))\\n\\t\\ttmp_rec.append(recall(actual, predicted))\\n\\t\\ttmp_fmeasure.append(f_measure(actual, predicted))\\n\\n\\taccuracies.append( round(sum(tmp_acc) / float(rounds), 2) )\\n\\tprecisions.append( round(sum(tmp_prec) / float(rounds), 2) )\\n\\trecalls.append( round(sum(tmp_rec) / float(rounds), 2) )\\n\\tfmeasures.append(round(sum(tmp_fmeasure) / float(rounds), 2))\\n\\tclassnames = list(set(get_class_names(total_actual) + get_class_names(total_predicted)))\\n\\tcm = confusion_matrix(total_actual, total_predicted, labels=classnames)\\n\\t#gm.plot_confusion_matrix(cm, classnames, normalize=True)\\n\\n\\ttmp_acc = []\\n\\ttmp_prec = []\\n\\ttmp_rec = []\\n\\ttmp_fmeasure = []\\n\\ttotal_actual = []\\n\\ttotal_predicted = []\\n\\tfor i in range(0, rounds):\\n\\t\\tprint(\\\"W2V\\\")\\n\\t\\t# --- Word2Vec Matcher\\n\\t\\tccc = Column_Classification_Config()\\n\\t\\tccc.add_feature('feature_main', 'Corpus', [sf_main, number_of_classes, examples_per_class, False, False])\\n\\n\\t\\tccc.add_matcher('matcher', 'Word2Vec_Matcher', {'feature_main': 'corpus'}) # main classifier\\n\\t\\tsm = Schema_Matcher(ccc)\\n\\t\\tactual, predicted = execute_test_ckan(sm, test_folder, False)\\n\\t\\ttotal_actual += actual\\n\\t\\ttotal_predicted += predicted\\n\\t\\taccuracy = accuracy_score(actual, predicted)\\n\\t\\ttmp_acc.append(accuracy)\\n\\t\\ttmp_prec.append(precision(actual, predicted))\\n\\t\\ttmp_rec.append(recall(actual, predicted))\\n\\t\\ttmp_fmeasure.append(f_measure(actual, predicted))\\n\\n\\taccuracies.append( round(sum(tmp_acc) / float(rounds), 2) )\\n\\tprecisions.append( round(sum(tmp_prec) / float(rounds), 2) )\\n\\trecalls.append( round(sum(tmp_rec) / float(rounds), 2) )\\n\\tfmeasures.append(round(sum(tmp_fmeasure) / float(rounds), 2))\\n\\tclassnames = list(set(get_class_names(total_actual) + get_class_names(total_predicted)))\\n\\tcm = confusion_matrix(total_actual, total_predicted, labels=classnames)\\n\\t#gm.plot_confusion_matrix(cm, classnames, normalize=True)\\n\\n\\tgm.add_x(x)\\n\\t# accuracies = [0.4, 0.4, 0.4]\\n\\t# precisions = [0.5, 0.5, 0.5]\\n\\t# recalls = [0.62, 0.62, 0.62]\\n\\t# fmeasures = [0.23, 0.23, 0.28]\\n\\tgm.append_y(accuracies)\\n\\tgm.append_y(precisions)\\n\\tgm.append_y(recalls)\\n\\tgm.append_y(fmeasures)\\n\\tgm.store()\\n\\tsubtitle = \\\"Scores were averaged over \\\" + str(rounds) + \\\" tests with \\\" + str(len(classes)) + \\\" classes. \\\" + \\\\\\n\\t\\\"Number of simulated columns per class: \\\" + str(number_of_classes)\\n\\tlabels = [\\\"Accuracy\\\", \\\"Precision\\\", \\\"Recall\\\", \\\"F-Measure\\\"]\\n\\tgm.plot_bar_n(\\\"Matcher Type\\\", \\\"Score\\\", \\\"Accuracy of Matchers\\\", labels, subtitle=subtitle)\",\n \"def run_Experiment(DP = None, QL = None):\\n\\n # Path information\\n output_path, exp_num = create_new_dir() #dirs Exp/1, Exp/2, ...\\n DP_path = join(output_path,'DP') #dirs Exp/1/DP\\n QL_path = join(output_path,'QL') #dirs Exp/1/QL\\n print(\\\"************ Exp \\\", exp_num, \\\"************ \\\\n\\\")\\n\\n # Exp_summary_data\\n method = get_method_str(DP, QL)\\n exp_summary = [str(exp_num), method]\\n\\n\\n # Run DP\\n if DP != None:\\n print(\\\"In Runner: Executing DP !!\\\")\\n\\n prob_file = DP[0]\\n createFolder(DP_path)\\n # output_params = [V_so, mean, variance, bad_count]\\n output_params = run_DP(setup_grid_params, prob_file, output_file, DP_path, threshold = threshold)\\n\\n \\\"\\\"\\\"CHANGE ARGUMENT if return order of setup_grid() is changed\\\"\\\"\\\"\\n input_params = setup_grid_params[9].copy()\\n input_params.append(prob_file)\\n\\n exp_summary = append_params_to_summary(exp_summary, input_params, output_params)\\n append_summary_to_summaryFile('Experiments/Exp_summary.csv', exp_summary)\\n print(\\\"In Runner: Executing DP Finished!!\\\")\\n\\n # Run QL\\n if QL != None:\\n print(\\\"In Runner: Executing QL !!\\\")\\n\\n QL_params = QL\\n createFolder(QL_path)\\n output_parameters_all_cases = run_QL(setup_grid_params, QL_params, QL_path, exp_num)\\n # run_QL(setup_grid_params, QL_params, QL_path)\\n\\n print(\\\"In Runner: Executing QL Finished !!\\\")\",\n \"def evaluate_model(args, eval_runs, warm_runs, metrics=['psnr', 'ssim', 'fps']):\\n upsampler = Upsampler(args)\\n if warm_runs > 0:\\n print(\\\"Warming up for evaluation\\\")\\n for i in range(warm_runs):\\n print(\\\"Performing warm-up run\\\", str(i+1))\\n for sequence in ['foliage', 'walk', 'calendar', 'city']:\\n bix_dir = os.path.join(VID4_DIR, 'BIx4', sequence)\\n upsampler.run_dir(bix_dir, reset=False)\\n \\n time = 0.\\n psnrs = []\\n ssims = []\\n for i in range(eval_runs):\\n run_psnrs = []\\n run_ssims = []\\n print(\\\"Performing evaluation run\\\", str(i+1))\\n for sequence in ['foliage', 'walk', 'calendar', 'city']:\\n bix_dir = os.path.join(VID4_DIR, 'BIx4', sequence)\\n gt_dir = os.path.join(VID4_DIR, 'GT', sequence)\\n print(\\\"Evaluating on\\\", bix_dir)\\n time += upsampler.run_dir(bix_dir, reset=False)\\n vid_psnrs, vid_ssims = _eval_sr_perf(os.path.join(bix_dir, 'up'), gt_dir)\\n run_psnrs += vid_psnrs\\n run_ssims += vid_ssims\\n if i == eval_runs-1:\\n with open(os.path.join(upsampler.get_model_dir(), \\\"psnr.txt\\\"), \\\"w\\\") as f:\\n f.writelines(str(psnr) + '\\\\n' for psnr in run_psnrs)\\n with open(os.path.join(upsampler.get_model_dir(), \\\"ssim.txt\\\"), \\\"w\\\") as f:\\n f.writelines(str(ssim) + '\\\\n' for ssim in run_ssims)\\n psnrs += run_psnrs\\n ssims += run_ssims\\n\\n fps = VID4_LENGTH/ (time/eval_runs)\\n return Performance(psnr=psnrs, ssim=ssims, fps=fps)\",\n \"def basic_experiments(k, p, num_iter, transforms, train_path, test_path):\\n start_time = time.time()\\n print(\\\"Generating advice\\\")\\n with open(train_path, \\\"r\\\") as f:\\n train_data = f.readlines()\\n advice = generate_advice(train_data, transforms)\\n print(\\\"Done: \\\" + str(time.time() - start_time) + \\\"\\\\n\\\")\\n\\n print(\\\"Generating real count for test data\\\")\\n with open(test_path, \\\"r\\\") as f:\\n test_data = f.readlines()\\n real_count = generate_advice(test_data, transforms)\\n print(\\\"Done: \\\" + str(time.time() - start_time))\\n actual = [x.moment(p) for x in real_count]\\n print(\\\"Actual moments:\\\")\\n print(actual)\\n\\n print(\\\"Estimating using noisy advice:\\\")\\n est_with_advice = []\\n for i in range(num_iter):\\n seed = str(random.randint(1, 2**64 + 1)) + str(random.randint(1, 2**64 + 1))\\n this_hash_exp = lambda x: hash_exp(x, seed)\\n est_with_advice.append(\\n estimate_using_advice(test_data, transforms, k, p, this_hash_exp,\\n advice))\\n print(\\\"Done: \\\" + str(time.time() - start_time))\\n print(est_with_advice)\\n\\n print(\\\"Estimating using perfect advice:\\\")\\n est_perfect_advice = []\\n for i in range(num_iter):\\n seed = str(random.randint(1, 2**64 + 1)) + str(random.randint(1, 2**64 + 1))\\n this_hash_exp = lambda x: hash_exp(x, seed)\\n est_perfect_advice.append(\\n estimate_using_advice(test_data, transforms, k, p, this_hash_exp,\\n real_count))\\n print(\\\"Done: \\\" + str(time.time() - start_time))\\n print(est_perfect_advice)\\n\\n print(\\\"Estimating using PPSWOR:\\\")\\n est_ppswor = []\\n for i in range(num_iter):\\n cur = []\\n for x in real_count:\\n cur.append(ppswor_estimate_moment(x.counts.items(), k, p))\\n est_ppswor.append(cur)\\n print(\\\"Done: \\\" + str(time.time() - start_time))\\n print(est_ppswor)\\n\\n print(\\\"Estimating using ell_2 PPSWOR:\\\")\\n est_ppswor_l2 = []\\n for i in range(num_iter):\\n cur = []\\n for x in real_count:\\n cur.append(ppswor_estimate_moment(x.counts.items(), k, p, 2))\\n est_ppswor_l2.append(cur)\\n print(\\\"Done: \\\" + str(time.time() - start_time))\\n print(est_ppswor_l2)\",\n \"def run_all_tests(self) -> None:\\n self.run_trt_precision_tests()\\n logging.info(\\\"Check analysis result at: %s\\\", self._output_dir)\",\n \"def run_and_evaluate():\\n tsp_problems = read_all_problems()\\n # Empty list of metrics\\n results = []\\n for problem in tqdm.tqdm(tsp_problems):\\n # As random factors are involved repeat experiments a couple of times\\n best_routes_base = []\\n best_routes_af = []\\n best_routes_ms = []\\n base_times = []\\n af_times = []\\n ms_times = []\\n for i in range(10):\\n # Base solution\\n start_time = timeit.default_timer()\\n best_route_base = solve_tsp_basic(problem)\\n base_time = timeit.default_timer() - start_time\\n best_routes_base.append(Fitness(route=best_route_base).route_distance())\\n base_times.append(base_time)\\n\\n # AF clustering solution\\n start_time = timeit.default_timer()\\n best_route_af = solve_tsp_affinity_propagation(problem)\\n af_time = timeit.default_timer() - start_time\\n best_routes_af.append(Fitness(route=best_route_af).route_distance())\\n af_times.append(af_time)\\n\\n # MS solution\\n start_time = timeit.default_timer()\\n best_route_ms = solve_mean_shift(problem)\\n ms_time = timeit.default_timer() - start_time\\n best_routes_ms.append(Fitness(route=best_route_ms).route_distance())\\n ms_times.append(ms_time)\\n\\n results.append(\\n {\\n \\\"problem name\\\": problem.name,\\n \\\"optimal solution\\\": find_route_optimal_route_length(problem),\\n \\\"baseline tour length\\\": mean(best_routes_base),\\n \\\"af clustering tour length\\\": mean(best_routes_af),\\n \\\"ms clustering tour length\\\": mean(best_routes_ms),\\n \\\"baseline algorithm time\\\": mean(base_times),\\n \\\"af clustering algorithm time\\\": mean(af_times),\\n \\\"ms clustering algorithm time\\\": mean(ms_times),\\n }\\n )\\n # Create dataframe and safe results\\n df = pd.DataFrame(results)\\n df.to_csv(\\\"results.csv\\\", index=False)\\n return df\",\n \"def run_analyses(y_predict_train, y_train, y_predict, y_test):\\n # calculate metrics\\n _, training_error = output_error(y_predict_train, y_train)\\n (precision, recall, f1, _), testing_error = output_error(y_predict, y_test)\\n \\n # print out metrics\\n print 'Average Precision:', np.average(precision)\\n print 'Average Recall:', np.average(recall)\\n print 'Average F1:', np.average(f1)\\n print 'Training Error:', training_error\\n print 'Testing Error:', testing_error\",\n \"def test_compare_classifiers_performance_changing_k_vis_api(experiment_to_use):\\n experiment = experiment_to_use\\n probabilities = experiment.probabilities\\n viz_outputs = (\\\"pdf\\\", \\\"png\\\")\\n with TemporaryDirectory() as tmpvizdir:\\n for viz_output in viz_outputs:\\n vis_output_pattern_pdf = tmpvizdir + f\\\"/*.{viz_output}\\\"\\n visualize.compare_classifiers_performance_changing_k(\\n [probabilities, probabilities],\\n experiment.ground_truth,\\n experiment.ground_truth_metadata,\\n experiment.output_feature_name,\\n top_k=3,\\n labels_limit=0,\\n model_namess=[\\\"Model1\\\", \\\"Model2\\\"],\\n output_directory=tmpvizdir,\\n file_format=viz_output,\\n )\\n figure_cnt = glob.glob(vis_output_pattern_pdf)\\n assert 1 == len(figure_cnt)\",\n \"def main():\\n run_simulation(spectral=False, ml=False, num_procs=1)\\n run_simulation(spectral=True, ml=False, num_procs=1)\\n run_simulation(spectral=False, ml=True, num_procs=1)\\n run_simulation(spectral=True, ml=True, num_procs=1)\\n run_simulation(spectral=False, ml=True, num_procs=10)\\n run_simulation(spectral=True, ml=True, num_procs=10)\",\n \"def evaluate(cfg: DictConfig):\\n\\n experiments = cfg.get('experiment_type', f'{cfg.model.name}_only')\\n fixed_t0 = cfg.get('fixed_t0', False)\\n ext = '_fixedT0' if fixed_t0 else ''\\n\\n base_dir = cfg.device.root\\n datasource = cfg.datasource.name\\n\\n if experiments == 'ablations':\\n models = {\\n 'FluxRGNN': ['final',\\n 'final_without_encoder',\\n 'final_without_boundary'],\\n 'LocalLSTM': ['final']\\n }\\n elif experiments == 'final':\\n models = {\\n 'FluxRGNN': ['final'],\\n 'GAM': ['final'],\\n 'HA': ['final'],\\n 'GBT': ['final']\\n }\\n else:\\n m = cfg.model.name\\n year = cfg.datasource.test_year\\n\\n # find all experiments available for this model, datasource and test year\\n result_dir = osp.join(base_dir, 'results', datasource, m, f'test_{year}')\\n models = {\\n m : [ f.name for f in os.scandir(result_dir) if f.is_dir() ]\\n }\\n\\n\\n # thresholds for binary classification metrics\\n if cfg.datasource.name == 'abm':\\n thresholds = [0.0019, 0.0207]\\n else:\\n thresholds = [0, 10, 20]\\n\\n rmse_per_hour = []\\n mae_per_hour = []\\n pcc_per_hour = []\\n bin_per_hour = []\\n\\n rmse_per_night = []\\n mae_per_night = []\\n\\n output_dir = osp.join(base_dir, 'results', datasource, f'performance_evaluation{ext}', experiments)\\n os.makedirs(output_dir, exist_ok=True)\\n\\n counter = 0\\n\\n for m, dirs in models.items():\\n print(f'evaluate {m}')\\n\\n for d in dirs:\\n result_dir = osp.join(base_dir, 'results', datasource, m, f'test_{cfg.datasource.test_year}', d)\\n\\n # check if directory exists\\n if os.path.isdir(result_dir):\\n results, model_cfg = load_cv_results(result_dir, trials=cfg.task.repeats, ext=ext)\\n\\n df_prep = pd.read_csv(osp.join(base_dir, 'data', 'preprocessed',\\n f'{model_cfg[\\\"t_unit\\\"]}_{model_cfg[\\\"model\\\"][\\\"edge_type\\\"]}_ndummy={model_cfg[\\\"datasource\\\"][\\\"n_dummy_radars\\\"]}',\\n datasource, cfg.season, str(cfg.datasource.test_year), 'dynamic_features.csv'))\\n tidx2night = dict(zip(df_prep.tidx, df_prep.nightID))\\n\\n rmse_per_hour.append(compute_rmse(m, d, results, tidx2night, groupby=['horizon', 'trial'],\\n threshold=0, km2=True, fixed_t0=fixed_t0))\\n mae_per_hour.append(compute_mae(m, d, results, tidx2night, groupby=['horizon', 'trial'],\\n threshold=0, km2=True, fixed_t0=fixed_t0))\\n pcc_per_hour.append(compute_pcc(m, d, results, tidx2night, groupby=['horizon', 'trial'],\\n threshold=0, km2=True, fixed_t0=fixed_t0))\\n\\n if fixed_t0:\\n rmse_per_night.append(compute_rmse_per_night(m, d, results, tidx2night, groupby=['night_horizon', 'trial']))\\n mae_per_night.append(compute_mae_per_night(m, d, results, tidx2night, groupby=['night_horizon', 'trial']))\\n\\n # compute binary classification measures\\n for thr in thresholds:\\n bin_per_hour.append(compute_bin(m, d, results, groupby=['horizon', 'trial'], threshold=thr, km2=True))\\n\\n counter += 1\\n\\n else:\\n print(f'Experiment \\\"{d}\\\" for model \\\"{m}\\\" and datasource \\\"{datasource}\\\" is not available. '\\n f'Use \\\"run_experiments.py model={m} datasource={datasource} +experiment={d}\\\" to run this experiment.')\\n\\n if counter > 0:\\n rmse_per_hour = pd.concat(rmse_per_hour)\\n rmse_per_hour.to_csv(osp.join(output_dir, f'rmse_per_hour.csv'))\\n\\n mae_per_hour = pd.concat(mae_per_hour)\\n mae_per_hour.to_csv(osp.join(output_dir, f'mae_per_hour.csv'))\\n\\n pcc_per_hour = pd.concat(pcc_per_hour)\\n pcc_per_hour.to_csv(osp.join(output_dir, f'pcc_per_hour.csv'))\\n\\n bin_per_hour = pd.concat(bin_per_hour)\\n bin_per_hour.to_csv(osp.join(output_dir, f'bin_per_hour.csv'))\\n\\n if fixed_t0:\\n rmse_per_night = pd.concat(rmse_per_night)\\n rmse_per_night.to_csv(osp.join(output_dir, f'rmse_per_night.csv'))\\n\\n mae_per_night = pd.concat(mae_per_night)\\n mae_per_night.to_csv(osp.join(output_dir, f'mae_per_night.csv'))\",\n \"def three_experiments(two_experiments, one_experiment):\",\n \"def main():\\r\\n algos = [merge_sort, quick_sort, heap_sort, radix_sort, bucket_sort_general]\\r\\n array_sizes = [5000, 10000, 15000, 20000, 50000, 75000, 100000, 150000]\\r\\n results = {algo.__name__: [] for algo in algos}\\r\\n for algo in algos:\\r\\n result = []\\r\\n for size in array_sizes:\\r\\n time = test(algo, size)\\r\\n result.append(time)\\r\\n results[algo.__name__] = result\\r\\n\\r\\n display_results(results, array_sizes)\",\n \"def test_compare_classifier_performance_from_pred_vis_api(experiment_to_use):\\n experiment = experiment_to_use\\n prediction = experiment.predictions\\n viz_outputs = (\\\"pdf\\\", \\\"png\\\")\\n with TemporaryDirectory() as tmpvizdir:\\n for viz_output in viz_outputs:\\n vis_output_pattern_pdf = tmpvizdir + f\\\"/*.{viz_output}\\\"\\n visualize.compare_classifiers_performance_from_pred(\\n [prediction, prediction],\\n experiment.ground_truth,\\n experiment.ground_truth_metadata,\\n experiment.output_feature_name,\\n labels_limit=0,\\n model_namess=[\\\"Model1\\\", \\\"Model2\\\"],\\n output_directory=tmpvizdir,\\n file_format=viz_output,\\n )\\n figure_cnt = glob.glob(vis_output_pattern_pdf)\\n assert 1 == len(figure_cnt)\",\n \"def run_tests(augmenter, images, nb_runs):\\n results = np.zeros((nb_runs,))\\n for i in range(nb_runs):\\n start = time.time()\\n augmenter.augment_batch(images)\\n results[i] = time.time() - start\\n print(\\\"Run %d: %.8fs\\\" % (i, results[i]))\\n print(\\\"Mean: %.8fs\\\" % (results.mean(),))\\n print(\\\"Sum: %.8fs\\\" % (results.sum(),))\",\n \"def run_oracle_experiments(model, query_iterator, N, recalls=[1]):\\n times_dict = dict()\\n calls_dict = dict()\\n probs = dict()\\n keys = list()\\n recall_break = max(recalls)\\n for query, relevant_set in query_iterator:\\n candidates_iterator = iter_candidates(model, query)\\n times, calls = candidates_time_oracle_calls_count(\\n candidates_iterator, relevant_set, recall_break)\\n k = len(relevant_set)\\n if k in times_dict:\\n times_dict[k].append(times)\\n calls_dict[k].append(calls)\\n probs[k] += 1\\n else:\\n keys.append(k)\\n times_dict[k] = [times]\\n calls_dict[k] = [calls]\\n probs[k] = 1\\n \\n keys = np.array(keys)\\n probs = np.array([probs[k] for k in keys])/sum(probs.values())\\n times_dict = {k:np.array(v).mean(axis=0) for k,v in times_dict.items()}\\n calls_dict = {k:np.array(v).mean(axis=0) for k,v in calls_dict.items()}\\n \\n expected_times = oracle_expectation_by_recalls(recalls, keys, probs, times_dict)\\n expected_calls = oracle_expectation_by_recalls(recalls, keys, probs, calls_dict)\\n expected_calls_baseline = oracle_random_calls_expectation_by_recall(recalls, keys, probs, N)\\n \\n return expected_times, expected_calls, expected_calls_baseline\",\n \"def run_experiments() :\\n #%%\\n target_size=(32,32)\\n g_specs = {\\n \\\"batch_size\\\" : [ 30 , 60, 100 ],\\n \\\"learning_rate\\\" : [ 0.0002, 0.0003, 0.0005 ],\\n \\\"drop_out_rate\\\" : [ 0.2, 0.25, 0.3 ],\\n \\\"rescale_mode\\\" : [ \\\"max_q\\\" , \\\"max\\\", \\\"\\\" ]\\n }\\n\\n model_traits = MODEL_TRAITS[\\\"model2\\\"].copy()\\n tt_obj = model_traits[\\\"trainer_tester_class\\\"]( model_traits )\\n del model_traits[\\\"trainer_tester_class\\\"]\\n\\n cnt = 0\\n for batchs, lrate, do_rate, resc_mode in product( g_specs[\\\"batch_size\\\"],\\n g_specs[\\\"learning_rate\\\"],\\n g_specs[\\\"drop_out_rate\\\"],\\n g_specs[\\\"rescale_mode\\\"] ) :\\n\\n tt_obj.model_traits.update( {\\\"batch_size\\\" : batchs,\\n \\\"learning_rate\\\" : lrate,\\n \\\"rescale_mode\\\" : resc_mode,\\n \\\"drop_out_rate\\\" : do_rate } )\\n\\n train_4d, train_gt = tu.make_4d_arrays( images_dir=\\\"images/train\\\",\\n target_size=target_size )\\n\\n test_4d, test_gt = tu.make_4d_arrays( images_dir=\\\"images/test\\\",\\n target_size=target_size )\\n\\n data = {\\\"train_4d\\\" : train_4d,\\n \\\"test_4d\\\" : test_4d,\\n \\\"train_y\\\" : train_gt,\\n \\\"test_y\\\" : test_gt}\\n\\n valid_accu_log, train_accu_log = tt_obj.train( model_traits, data,\\n logl=100 )\\n idx_v = int(np.argmax( valid_accu_log))\\n idx_t = int(np.argmax( train_accu_log))\\n\\n model_traits.update({\\\"valid_accu_log\\\" : valid_accu_log,\\n \\\"train_accu_log\\\" : train_accu_log,\\n \\\"best_valid\\\" : max(valid_accu_log),\\n \\\"best_valid_at\\\" : idx_v,\\n \\\"train_at_best_valid\\\" : train_accu_log[idx_v],\\n \\\"best_train\\\" : max(train_accu_log),\\n \\\"best_train_at\\\": idx_t })\\n\\n #print(cnt, pformat(model_traits) )\\n print( \\\"%d : best_train = %.4f, best_valid = %.4f\\\" % \\\\\\n (cnt, max(train_accu_log), max(valid_accu_log) ))\\n\\n with open( \\\"exp_results_%d.json\\\" % cnt,\\n \\\"wt\\\" , encoding=\\\"utf8\\\" ) as f_out :\\n print( json.dumps( model_traits ), file=f_out)\\n\\n\\n cnt += 1\\n #%%\",\n \"def run_multiple_advanced_experiments(self, path_to_csv_files, algorithms_weights_list, thresholds):\\n starting_time = time.time()\\n print(\\\"started advanced precision recall image_matching_module\\\")\\n for algorithms_weights in algorithms_weights_list:\\n start_file_time = time.time()\\n\\n first_run = True\\n for algorithm, weight in algorithms_weights:\\n if first_run:\\n algorithm_name = algorithm.get_algorithm_name() + weight\\n first_run = False\\n else:\\n algorithm_name = algorithm_name + \\\"_\\\" + algorithm.get_algorithm_name() + weight\\n\\n print(\\\"started algorithm:\\\", algorithm_name)\\n self.run_single_advanced_experiment(path_to_csv_files, algorithms_weights, thresholds)\\n finish_file_time = time.time()\\n print(\\\"finished algorithm:\\\", algorithm_name)\\n print(\\\"time took:\\\", (finish_file_time - start_file_time) / 60, \\\"minutes\\\")\\n finish_time = time.time()\\n hours = int((finish_time - starting_time) // 3600)\\n minutes = (finish_time - starting_time) / 60\\n print(\\\"finished advanced precision recall image_matching_module\\\")\\n print(\\\"time image_matching_module took:\\\", hours, \\\"hours and\\\", minutes, \\\"minutes\\\")\",\n \"def test_any(alg, repNum=1):\\n\\tperformance = alg.predict_perf(repNum=repNum)\\n\\tprint performance\\n\\talg.plot()\\n\\n\\treturn performance\",\n \"def test_compare_classifier_performance_from_prob_vis_api(experiment_to_use):\\n experiment = experiment_to_use\\n probability = experiment.probabilities\\n viz_outputs = (\\\"pdf\\\", \\\"png\\\")\\n with TemporaryDirectory() as tmpvizdir:\\n for viz_output in viz_outputs:\\n vis_output_pattern_pdf = tmpvizdir + f\\\"/*.{viz_output}\\\"\\n visualize.compare_classifiers_performance_from_prob(\\n [probability, probability],\\n experiment.ground_truth,\\n experiment.ground_truth_metadata,\\n experiment.output_feature_name,\\n top_n_classes=[0],\\n labels_limit=0,\\n model_namess=[\\\"Model1\\\", \\\"Model2\\\"],\\n output_directory=tmpvizdir,\\n file_format=viz_output,\\n )\\n figure_cnt = glob.glob(vis_output_pattern_pdf)\\n assert 1 == len(figure_cnt)\",\n \"def test_run(id_, args, kwargs, expected):\\n # Remove results for test.\\n results_path = expected['results_path']\\n shutil.rmtree(results_path, ignore_errors=True)\\n # Start code.\\n # [future] attempts to run classification with n_jobs>1\\n # global cnfg_default\\n # sys.modules['cnfg_default'] = get_params(args[0], 'cnfg_default')\\n # import cnfg_default\\n # #from cnfg_default import custom_score_metric\\n objects = pycnfg.run(oid='default', *args, **kwargs)\\n tmp = {k: type(v).__name__ for k, v in objects.items()}\\n print('OBJECTS:')\\n print(tmp)\\n # Compare results:\\n # * Compare objects (keys and str of values).\\n objects_ = expected['objects']\\n objects = {k: type(v).__name__ for k, v in objects.items()}\\n if objects != objects_:\\n print(set(objects.items()) ^ set(objects_.items()))\\n assert objects == objects_\\n # for k, v in objects.items():\\n # assert k in objects_\\n # assert type(v).__name__ == objects_[k]\\n # * Compare predictions csv(all available).\\n pred_path = glob.glob(f\\\"{results_path}/models/*_pred.csv\\\")\\n pred_path_ = glob.glob(expected['pred_path'])\\n assert len(pred_path) == len(pred_path_)\\n for act, exp in zip(sorted(pred_path), sorted(pred_path_)):\\n file_diff(act, exp)\\n assert filecmp.cmp(act, exp)\\n # * Compare test logs.\\n logs_path = glob.glob(f\\\"{results_path}/logs*/*_test.log\\\")[0]\\n logs_path_ = expected['logs_path']\\n file_diff(logs_path, logs_path_)\\n assert filecmp.cmp(logs_path, logs_path_)\\n # * Compare runs dataframe, non-universe columns.\\n runs_path = f\\\"{results_path}/runs\\\"\\n runs_path_ = expected['runs_path']\\n df = runs_loader(runs_path)\\n df_ = runs_loader(runs_path_)\\n # First False/True for each element, then check all by columns.\\n # col1 True\\n # col2 False\\n # dtype: bool\\n df_diff = df.eq(df_).all()\\n # Column names that are not equal.\\n columns = sorted(list(df_diff[df_diff==False].dropna().index))\\n # columns_eq = sorted(list(df_diff[df_diff==True].dropna().index))\\n columns_ = expected['columns_diff']\\n print('DIFF:\\\\n', columns)\\n time.sleep(1)\\n # assert columns == columns_\\n # * Compare model.\\n model_path = glob.glob(f\\\"{results_path}/models/*.model\\\")\\n model_path_ = glob.glob(expected['model_path'])\\n assert len(model_path) == len(model_path_)\\n for act, exp in zip(sorted(model_path), sorted(model_path_)):\\n assert filecmp.cmp(act, exp)\\n return\",\n \"def one_experiment():\\n\\n # set the name of the experiment\\n now = datetime.datetime.now()\\n experiment_id = str(now.day) + \\\"_\\\" + str(now.month) + \\\"_\\\" + str(now.hour) + \\\".\\\" + str(now.minute)\\n experiment_name = 'overfit_' + str(experiment_id)\\n\\n # define if you want to use preprocessed data from file\\n use_prep_data = False\\n if use_prep_data:\\n set_params(preproc_data_id='16_5_10.16.47')\\n\\n # define the changing parameter and its value\\n changing_param_name = 'class_weights'\\n changing_param_value = [{0: 15, 1: 85}]\\n # {0:15, 1:85}]#, {0:4, 1:100}, {0:3, 1:100}, {0:2, 1:100}, {0:1, 1:100}] #[{0:1, 1:1}, {0:15, 1:85}]#\\n\\n features_to_use = ['user', 'countries', 'session', 'format', 'token']\\n # set constant parameters\\n set_params(use_word_emb=1)\\n set_params(epochs=40)\\n set_params(features_to_use=features_to_use)\\n\\n # save constant parameters to a new \\\"experiment_..\\\" filgithx+P@2ub\\n save_constant_parameters(experiment_name, changing_param_name)\\n\\n # run experiment for every parameter value\\n for value in changing_param_value:\\n process = psutil.Process(os.getpid())\\n print(\\\"-----MEMORY before starting experiment ------\\\", int(process.memory_info().rss/(8*10**3)), \\\"KB\\\")\\n\\n # update the parameter value\\n set_params(class_weights_1=value)\\n\\n # update the model_id for this new model\\n now = datetime.datetime.now()\\n new_model_id = str(now.day) + \\\"_\\\" + str(now.month) + \\\"_\\\" + str(now.hour) + \\\".\\\" + str(now.minute) + \\\".\\\" + str(now.second)\\n\\n set_params(model_id=new_model_id)\\n\\n # evaluate the new model and save the results in the experiment file\\n oneExperiment = Process(target=run_experiment, args=(experiment_name,\\n new_model_id, changing_param_name, value,))\\n oneExperiment.start()\\n oneExperiment.join()\",\n \"def compute_and_visualize_coupling_experiments(experiments, results, seed,\\n **kwargs):\\n (logits_1,\\n logits_2), couplings = get_coupling_estimates(experiments, results, seed,\\n **kwargs)\\n test_losses = compute_coupling_losses(experiments, logits_1, logits_2,\\n couplings)\\n visualize_coupling_experiments(test_losses, logits_1, logits_2, couplings)\",\n \"def main():\\n\\tresults = []\\n\\n\\tconfig = configparser.ConfigParser()\\n\\tconfig.read(\\\"simulation.ini\\\")\\n\\tsettings = config['sim']\\n\\n\\tcompleted_obj_hw = int(settings[\\\"ClientsPerCampaign\\\"]) * float(settings[\\\"CompletedPctgHW\\\"])\\n\\texceeded_obj_hw = float(settings[\\\"ExceededPctgHW\\\"])\\n\\tsignificance_level = float(settings[\\\"SignificanceLevel\\\"])\\n\\tz_val_two_tails = scipy.stats.norm.ppf(1 - (significance_level / 2))\\n\\n\\tprint(\\\"Completed Target HW: \\\" + str(completed_obj_hw))\\n\\tprint(\\\"Exceeded Target HW: \\\" + str(exceeded_obj_hw))\\n\\n\\tcompleted_vals = []\\n\\texceeded_vals = []\\n\\tdone = False\\n\\n\\tcompleted_avg = 0\\n\\texceeded_avg = 0\\n\\tcompleted_hw = 0\\n\\texceeded_hw = 0\\n\\n\\ti = 0\\n\\twhile not done:\\n\\t\\tprint(\\\"RUN: \\\" + str(i + 1))\\n\\t\\tenv = simpy.Environment()\\n\\t\\tsim = Simulation(env, settings, i == 0)\\n\\t\\tsim.run()\\n\\t\\tresults.append(sim.results)\\n\\t\\ti += 1\\n\\n\\t\\tif settings['RunOnce'] == 'yes':\\n\\t\\t\\tprint(\\\"RUN ONCE\\\")\\n\\t\\t\\tsys.exit()\\n\\n\\t\\tcompleted_vals.append(sim.results['completed_count'])\\n\\t\\texceeded_vals.append(sim.results['exceeded_proportion'])\\n\\n\\t\\tif i < 2:\\n\\t\\t\\tprint(\\\"---------------\\\")\\n\\t\\t\\tcontinue\\n\\n\\t\\tcompleted_avg = sum(completed_vals) / len(completed_vals)\\n\\t\\tcompleted_S = sum([(v - completed_avg) ** 2 for v in completed_vals]) / (i - 1)\\n\\t\\tcompleted_S = math.sqrt(completed_S)\\n\\t\\tcompleted_hw = (z_val_two_tails * completed_S) / math.sqrt(i)\\n\\t\\tprint(\\\"runs: \\\" + str(i) + \\\" completed HW: \\\" + str(completed_hw))\\n\\n\\t\\texceeded_avg = sum(exceeded_vals) / len(exceeded_vals)\\n\\t\\texceeded_S = math.sqrt(exceeded_avg * (1 - exceeded_avg))\\n\\t\\texceeded_hw = (z_val_two_tails * exceeded_S) / math.sqrt(i)\\n\\t\\tprint(\\\"runs: \\\" + str(i) + \\\" exceeded HW: \\\" + str(exceeded_hw))\\n\\n\\t\\tif completed_hw < completed_obj_hw and exceeded_hw < exceeded_obj_hw:\\n\\t\\t\\tprint(\\\"END ITERATIONS\\\")\\n\\t\\t\\tdone = True\\n\\n\\t\\tprint(\\\"---------------\\\")\\n\\n\\n\\tfilename = 'results/Results_' + settings['FileSizeGB'] + '_' + settings['TorrentThreshold'] + '_' + settings['HTTPDownThreshold'] \\\\\\n\\t\\t+ '_' + settings['HTTPUp'] + '_' + str(random.randint(0,10000)) + '.xlsx'\\n\\n\\tprint(\\\"Saving XLSX to: \\\" + filename)\\n\\twb = xs.Workbook(filename)\\n\\n\\tws = wb.add_worksheet()\\n\\n\\tws.write(0, 1, 'Exceded')\\n\\tws.write(0, 2, 'Completed')\\n\\n\\ti = 1\\n\\tfor result in results:\\n\\t\\tws.write(i, 0, i)\\n\\t\\tws.write(i, 1, result['exceeded_proportion'])\\n\\t\\tws.write(i, 2, result['completed_count'])\\n\\t\\ti += 1\\n\\n\\tws.write(i, 0, 'average')\\n\\tws.write(i, 1, exceeded_avg)\\n\\tws.write(i, 2, completed_avg)\\n\\ti += 1\\n\\tws.write(i, 0, 'half width')\\n\\tws.write(i, 1, exceeded_hw)\\n\\tws.write(i, 2, completed_hw)\\n\\n\\twb.close()\",\n \"def test_compare_classifiers_performance_subset_vis_api(experiment_to_use):\\n experiment = experiment_to_use\\n probabilities = experiment.probabilities\\n viz_outputs = (\\\"pdf\\\", \\\"png\\\")\\n with TemporaryDirectory() as tmpvizdir:\\n for viz_output in viz_outputs:\\n vis_output_pattern_pdf = tmpvizdir + f\\\"/*.{viz_output}\\\"\\n visualize.compare_classifiers_performance_subset(\\n [probabilities, probabilities],\\n experiment.ground_truth,\\n experiment.ground_truth_metadata,\\n experiment.output_feature_name,\\n top_n_classes=[6],\\n labels_limit=0,\\n subset=\\\"ground_truth\\\",\\n model_namess=[\\\"Model1\\\", \\\"Model2\\\"],\\n output_directory=tmpvizdir,\\n file_format=viz_output,\\n )\\n figure_cnt = glob.glob(vis_output_pattern_pdf)\\n assert 1 == len(figure_cnt)\",\n \"def test_calculate_all_metrics_precision():\\n pass\",\n \"def run():\\n trials = 100\\n\\n multipliers = [0.25, 0.3, 0.35, 0.5, 0.75, 1, 1.25, 1.45, 1.5, 1.55, 1.6] # Coefficients for learning rate\\n\\n mean_penalty = []\\n median_penalty = []\\n std_penalty = []\\n\\n mean_trial_time = []\\n median_trial_time = []\\n std_trial_time = []\\n\\n mean_success_rate = []\\n median_success_rate = []\\n std_success_rate = []\\n\\n for m in multipliers:\\n all_penalties = [] # All penalties from trail sets\\n all_average_trial_time = []\\n all_success_rates = []\\n\\n for i in range(0, 20):\\n # print \\\"Trial set:\\\", i\\n # Set up environment and agent\\n e = Environment() # create environment (also adds some dummy traffic)\\n agent = e.create_agent(LearnerAgent) # create agent\\n agent.mult = m\\n e.set_primary_agent(agent, enforce_deadline=True) # specify agent to track\\n\\n # Now simulate it\\n sim = Simulator(e, update_delay=0.0, display=False) # create simulator (uses pygame when display=True, if available)\\n\\n sim.run(n_trials=trials) # run for a specified number of trials\\n\\n all_penalties.append(agent.all_trails_penalties)\\n all_average_trial_time.append(agent.time/float(trials))\\n all_success_rates.append(float(trials-agent.aborted_trials)/trials)\\n\\n mean_penalty.append(np.mean(all_penalties))\\n median_penalty.append(np.median(all_penalties))\\n std_penalty.append(np.std(all_penalties))\\n\\n mean_trial_time.append(np.mean(all_average_trial_time))\\n median_trial_time.append(np.median(all_average_trial_time))\\n std_trial_time.append(np.std(all_average_trial_time))\\n\\n mean_success_rate.append(np.mean(all_success_rates))\\n median_success_rate.append(np.median(all_success_rates))\\n std_success_rate.append(np.std(all_success_rates))\\n\\n for i in range(0, len(multipliers)):\\n print \\\"\\\"\\n print \\\"Multiplier:\\\", multipliers[i]\\n print \\\"\\\"\\n print \\\"Mean penalty per {} trials:\\\".format(trials), mean_penalty[i]\\n print \\\"Median penalty per {} trials:\\\".format(trials), median_penalty[i]\\n print \\\"Std.Dev. penalty per {} trials:\\\".format(trials), std_penalty[i]\\n\\n print \\\"\\\"\\n print \\\"Mean trial time:\\\", mean_trial_time[i]\\n print \\\"Median trial time:\\\", median_trial_time[i]\\n print \\\"Std.Dev. trial time:\\\", std_trial_time[i]\\n\\n print \\\"\\\"\\n print \\\"Mean success rate per {} trials:\\\".format(trials), mean_success_rate[i]\\n print \\\"Median success rate per {} trials:\\\".format(trials), median_success_rate[i]\\n print \\\"Std.Dev. success rate per {} trials:\\\".format(trials), std_success_rate[i]\",\n \"def print_results(precision, recall, f1):\\n print \\\"Precision:\\\", precision\\n print \\\"Recall:\\\", recall\\n print \\\"F1 score:\\\", f1\",\n \"def three_experiments_with_trials(family_with_trials, single_with_trials):\",\n \"def runTests():\\r\\n\\r\\n print(\\\"running a few tests\\\")\\r\\n\\r\\n average = compute .gpsAverage (4, 5)\\r\\n print(\\\"average = \\\", average)\\r\\n \\r\\n print (\\\"hello!\\\")\",\n \"def evaluate(self, runs=100):\\n score_record = []\\n \\n print('Evaluation in progress...')\\n for i in range(runs):\\n score = self.run_evaluation_episode()\\n score_record.append(score)\\n \\n ave_score = np.mean(score_record)\\n \\n print('System evaluated with an average score of {} in {} runs'.format(ave_score, runs))\",\n \"def run_tests():\\n with open(FILENAME) as file:\\n # Loads testing parameters from the yaml file.\\n tests = yaml.safe_load(file)\\n\\n # create a dataframe to keep the results\\n test_dict = tests['Tests']\\n results = pd.DataFrame(test_dict)\\n results['Last Average Score'] = \\\"\\\"\\n results['No of Q-Learning episodes'] = \\\"\\\"\\n\\n # run experiments:\\n for i, test in enumerate(test_dict):\\n grid = Rooms(test[\\\"env_size\\\"], testing=True)\\n learning = QLearning(grid, test[\\\"gamma\\\"], test[\\\"alpha\\\"], test[\\\"agent_start_pos\\\"])\\n e_greedy = Policy(\\\"e-greedy\\\", test[\\\"epsilon\\\"], test[\\\"decay\\\"])\\n greedy = Policy(policy_type=\\\"greedy\\\")\\n experiment = Experiments(grid, learning, greedy, test[\\\"iters\\\"],\\n test[\\\"agent_start_pos\\\"], test[\\\"test_no\\\"])\\n\\n for session in range(test[\\\"iters\\\"]):\\n learning.run_multiple_episodes(test[\\\"batch_episodes\\\"], e_greedy)\\n mean_reward = experiment.run_experiments(test[\\\"exp_per_batch\\\"])\\n\\n results.loc[i,'Last Average Score'] = mean_reward\\n results.loc[i,'No of Q-Learning episodes'] = (session + 1) * test[\\\"batch_episodes\\\"]\\n\\n # save results to csv file\\n filename = 'results/' + 'test_table.csv'\\n results.to_csv(filename)\\n\\n # plot & save graphs\\n experiment.generate_results(test[\\\"test_no\\\"], test)\\n\\n return results\",\n \"def test_compare_performance_vis_api(experiment_to_use):\\n experiment = experiment_to_use\\n # extract test stats only\\n test_stats = experiment.test_stats_full\\n viz_outputs = (\\\"pdf\\\", \\\"png\\\")\\n with TemporaryDirectory() as tmpvizdir:\\n for viz_output in viz_outputs:\\n vis_output_pattern_pdf = tmpvizdir + f\\\"/*.{viz_output}\\\"\\n visualize.compare_performance(\\n [test_stats, test_stats],\\n output_feature_name=None,\\n model_names=[\\\"Model1\\\", \\\"Model2\\\"],\\n output_directory=tmpvizdir,\\n file_format=viz_output,\\n )\\n figure_cnt = glob.glob(vis_output_pattern_pdf)\\n assert 1 == len(figure_cnt)\",\n \"def emPerformance(filesAndDirectories='None', resultsFileName='None', iterationCount='3', modes='None', testTypes='None', viewports='None', verbose='False'):\\n\\n pass\",\n \"def evaluate_all(experiments,\\n results,\\n seed,\\n num_pairs,\\n samples_per_pair,\\n loop_size=None):\\n eval_results = evaluate_baselines(\\n experiments[0], seed, num_pairs, samples_per_pair, loop_size=None)\\n for ex, res in zip(experiments, results):\\n eval_results.update(\\n evaluate_experiment(ex, res, seed, num_pairs, samples_per_pair,\\n loop_size))\\n return eval_results\",\n \"def _evaluate_performance__static_winners(self):\\n # | - _evaluate_performance__\\n\\n # | - class attributes #################################################\\n AL = self\\n al_gen = self.al_gen\\n verbose = self.verbose\\n seed_ids = self.seed_ids\\n acquisition_bin = self.acquisition_bin\\n completed_ids = self.completed_ids\\n CandidateSpace = self.CandidateSpace\\n RegressionModel = self.RegressionModel\\n DuplicateFinder = self.DuplicateFinder\\n al_gen_dict = self.al_gen_dict\\n\\n stop_mode = self.stop_mode\\n stop_num_generations = self.stop_num_generations\\n\\n index_acq_gen_dict = self.index_acq_gen_dict\\n #__| #################################################################\\n\\n # #####################################################################\\n mode = \\\"lowest_N\\\" # 'lowest_N' or 'lowest_perc'\\n\\n N_ids = 10\\n lowest_perc = 5\\n\\n # Number of consecutive generations that the Nth best systems must\\n # remain static\\n M_gens = 3\\n # #####################################################################\\n\\n if mode == \\\"lowest_perc\\\":\\n num_candidates = CandidateSpace.FingerPrints.df_pre.shape[0]\\n N_ids = int(num_candidates * (lowest_perc * 0.01))\\n\\n gen_keys = list(AL.al_gen_dict.keys())\\n\\n if len(gen_keys) > M_gens:\\n latest_M_keys = gen_keys[-(M_gens + 1):]\\n last_gen_key = gen_keys[-1]\\n\\n al_gen_dict_subset_i = dict(zip(\\n latest_M_keys,\\n [AL.al_gen_dict.get(i, None) for i in latest_M_keys]))\\n\\n indices_list = []\\n iterator = enumerate(al_gen_dict_subset_i.items())\\n for i_cnt, (gen_i, AL_i) in iterator:\\n model_i = AL_i.model\\n\\n model_i = AL.add_main_Y_to_model(\\n model_i, plot_dft_instead_of_pred=True)\\n model_i = model_i[(model_i[\\\"duplicate\\\"] == False)]\\n model_i = model_i.sort_values(\\\"Y_main\\\")\\n\\n indices_i = model_i.index.tolist()\\n\\n indices_list.append(indices_i)\\n\\n if i_cnt >= M_gens:\\n indices_i = indices_list[i_cnt][0:N_ids]\\n ids_static_list = []\\n for j in range(M_gens):\\n indices_j = indices_list[i_cnt - (j + 1)][0:N_ids]\\n ids_static = indices_j == indices_i\\n ids_static_list.append(ids_static)\\n\\n ids_are_static = all(ids_static_list)\\n\\n self.performance__static_winners[last_gen_key] = ids_are_static\\n #__|\",\n \"def evaluate_benchmarks(self):\\n\\n # iterate over replicates\\n results = {}\\n for replicate_id, replicate in self.replicates:\\n\\n # evaluate benchmark for current replicate\\n bmark = SimulationBenchmark(replicate.copy(),\\n graph=self.graphs[replicate_id],\\n **self.params)\\n\\n # store results\\n results[replicate_id] = dict(\\n\\n labels_MAE=bmark.scores['labels'].MAE,\\n level_only_MAE=bmark.scores['level_only'].MAE,\\n spatial_only_MAE=bmark.scores['spatial_only'].MAE,\\n community_MAE=bmark.scores['labels_comm'].MAE,\\n\\n labels_PCT=bmark.scores['labels'].percent_correct,\\n level_only_PCT=bmark.scores['level_only'].percent_correct,\\n spatial_only_PCT=bmark.scores['spatial_only'].percent_correct,\\n community_PCT=bmark.scores['labels_comm'].percent_correct)\\n\\n # compile dataframe\\n results = pd.DataFrame.from_dict(results, orient='index')\\n results.index.set_names(self.multiindex, inplace=True)\\n\\n return results\",\n \"def run_experiment():\\n pass\",\n \"def evaluate_individual(predictions, test_files, models):\\n\\n print(\\\"\\\\nAccuracy for individual models\\\\n\\\")\\n \\n # Fix Location\\n correct_predictions = [0, 0, 0]\\n total_predictions = [0, 0, 0]\\n num_failed_predictions = 0\\n\\n for prediction in predictions:\\n if prediction[\\\"correct_data\\\"][\\\"correct_location\\\"] == prediction[\\\"predicted_location\\\"]:\\n correct_predictions[FixType[prediction[\\\"correct_data\\\"][\\\"correct_type\\\"]].value] = correct_predictions[FixType[\\n prediction[\\\"correct_data\\\"][\\\"correct_type\\\"]].value] + 1\\n if prediction[\\\"predicted_location\\\"] is None:\\n num_failed_predictions = num_failed_predictions + 1\\n total_predictions[FixType[prediction[\\\"correct_data\\\"][\\\"correct_type\\\"]].value] = total_predictions[FixType[\\n prediction[\\\"correct_data\\\"][\\\"correct_type\\\"]].value] + 1\\n\\n for i in range(3):\\n if total_predictions[i] == 0: # If the type was never predicted\\n accuracy = 0\\n else:\\n accuracy = correct_predictions[i] / total_predictions[i]\\n print(f\\\"Fix Location accuracy for class {FixType(i).name}: {accuracy * 100} %\\\")\\n\\n accuracy = sum(correct_predictions) / (len(predictions) - num_failed_predictions)\\n print(f\\\"Fix Location accuracy overall is {accuracy * 100} %\\\")\\n \\n # Fix type\\n correct_predictions = [0, 0, 0]\\n total_predictions = [0, 0, 0]\\n num_failed_predictions = 0\\n\\n for prediction in predictions:\\n if prediction[\\\"correct_data\\\"][\\\"correct_type\\\"] == prediction[\\\"predicted_type\\\"]:\\n correct_predictions[FixType[prediction[\\\"predicted_type\\\"]].value] = correct_predictions[FixType[\\n prediction[\\\"predicted_type\\\"]].value] + 1\\n if prediction[\\\"predicted_type\\\"] is None:\\n num_failed_predictions = num_failed_predictions + 1\\n total_predictions[FixType[prediction[\\\"predicted_type\\\"]].value] = total_predictions[FixType[\\n prediction[\\\"predicted_type\\\"]].value] + 1\\n\\n for i in range(3):\\n if total_predictions[i] == 0: # If the type was never predicted\\n accuracy = 0\\n else:\\n accuracy = correct_predictions[i] / total_predictions[i]\\n print(f\\\"Fix Type accuracy for class {FixType(i).name}: {accuracy * 100} %\\\")\\n\\n accuracy = sum(correct_predictions) / (len(predictions) - num_failed_predictions)\\n print(f\\\"Fix Type accuracy overall is {accuracy * 100} %\\\")\\n \\n # We repeat the predictions to evaluate the insert and modify models individually, regardless of the predicted fix type \\n\\n raw_training_samples = []\\n\\n if test_files.endswith(\\\".json\\\"): # Single JSON file\\n with open(test_files) as file:\\n logging.info(\\\"Source ending in .json. Predicting on single JSON file.\\\")\\n raw_training_samples = json.load(file)\\n else: # Folder path\\n for filename in listdir(test_files):\\n with open(test_files + filename) as file:\\n raw_training_samples.extend(json.load(file))\\n \\n correct_predictions_insert = 0\\n total_predictions_insert = 0\\n correct_predictions_modify = 0\\n total_predictions_modify = 0\\n insert_tokens = []\\n modify_tokens = []\\n\\n for sample in raw_training_samples:\\n # Insert\\n if sample[\\\"metadata\\\"][\\\"fix_type\\\"] == \\\"insert\\\":\\n actual_sample, tokens = IOProcessor.preprocess(sample[\\\"wrong_code\\\"])\\n pred = predict_single(actual_sample, models[2])\\n token = IOProcessor.postprocess(pred, 2)\\n if token == sample[\\\"metadata\\\"][\\\"fix_token\\\"]: # Correct Prediction\\n correct_predictions_insert = correct_predictions_insert + 1\\n else: # Incorrect prediction\\n insert_tokens.append([token, sample[\\\"metadata\\\"][\\\"fix_token\\\"]])\\n total_predictions_insert = total_predictions_insert + 1\\n # Modify\\n if sample[\\\"metadata\\\"][\\\"fix_type\\\"] == \\\"modify\\\":\\n actual_sample, tokens = IOProcessor.preprocess(sample[\\\"wrong_code\\\"])\\n pred = predict_single(actual_sample, models[3])\\n token = IOProcessor.postprocess(pred, 3)\\n if token == sample[\\\"metadata\\\"][\\\"fix_token\\\"]: # Correct Prediction\\n correct_predictions_modify = correct_predictions_modify + 1\\n else: # Incorrect prediction\\n modify_tokens.append([token, sample[\\\"metadata\\\"][\\\"fix_token\\\"]])\\n total_predictions_modify = total_predictions_modify + 1\\n\\n insert_accuracy = correct_predictions_insert / total_predictions_insert\\n modify_accuracy = correct_predictions_modify / total_predictions_modify\\n print(f\\\"Fix Token accuracy for insert is {insert_accuracy * 100} %\\\")\\n print(f\\\"Fix Token accuracy for modify is {modify_accuracy * 100} %\\\")\\n\\n # The following code may be used to create a swarm plot of the erroneous predictions for fix locations\\n # This does, however, require the installation of the pandas, seaborn, and matplotlib libraries.\\n \\n # import seaborn as sns\\n # import matplotlib.pyplot as plt\\n # import pandas as pd\\n # location_distance_array = []\\n # for prediction in predictions:\\n # actual_sample, tokens = IOProcessor.preprocess(prediction[\\\"correct_data\\\"][\\\"wrong_code\\\"])\\n # label = get_token_index(prediction[\\\"correct_data\\\"][\\\"wrong_code\\\"], tokens, prediction[\\\"correct_data\\\"][\\\"correct_location\\\"])\\n # if prediction[\\\"predicted_token_location\\\"] - label == 0:\\n # pass\\n # else:\\n # location_distance_array.append([prediction[\\\"predicted_token_location\\\"] - label, prediction[\\\"correct_data\\\"][\\\"correct_type\\\"]])\\n \\n # df = pd.DataFrame(data=location_distance_array)\\n # sns.set_theme(style=\\\"whitegrid\\\")\\n # f, ax = plt.subplots(figsize=(6, 4))\\n # sns.despine(bottom=True, left=True)\\n # sns.swarmplot(y=0, x=1, data=df, palette=\\\"dark\\\", size=6)\\n # ax.set_xlabel('')\\n # ax.set_ylabel('')\\n # plt.ylim([-15, 16])\\n \\n # plt.savefig('line_plot.pdf', bbox_inches='tight', pad_inches=0)\",\n \"def batch_simulation(self, iters=10000):\\n power_cnt = 0\\n correct_sign_cnt = 0\\n\\n for i in range(iters):\\n if (self.verbose) and (i>0) and (i % (iters/10) == 0):\\n print(i, \\\" / \\\", iters)\\n f_stat, p_value, effect_point_estimates = self.simulate()\\n power_cnt += (p_value < self.alpha)\\n correct_sign_cnt += (effect_point_estimates * self.absolute_effect > 0)\\n\\n if self.verbose: print(iters, \\\" / \\\", iters)\\n power = round(power_cnt / float(iters), 5)\\n pct_correct_sign = round(correct_sign_cnt / float(iters), 5)\\n return power, pct_correct_sign\",\n \"def run(self):\\n if self.pb.xvalEN and not self.isXvalMain:\\n dt = np.ones(self.pb.total_exam_no,\\n dtype=\\\"float32\\\") / self.train_exam_no\\n dt[self.val_indices] = 0.0\\n else:\\n dt = np.ones(self.pb.total_exam_no,\\n dtype=\\\"float32\\\") / self.pb.total_exam_no\\n\\n val = np.zeros(8,dtype=\\\"float32\\\")-1\\n boosting = None\\n wl = None\\n if self.pb.algorithm == 'conf-rated':\\n boosting = ConfidenceRated(self)\\n wl = ConfidenceRatedWL(self)\\n elif self.pb.algorithm == 'adaboost':\\n boosting = AdaBoost(self)\\n wl = AdaBoostWL(self)\\n elif self.pb.algorithm == 'adaboost-fast':\\n boosting = AdaBoostFast(self)\\n wl = AdaBoostFastWLMPI(self)\\n elif self.pb.algorithm == 'rankboost':\\n boosting = RankBoost(self)\\n wl = ConfidenceRatedWL(self)\\n elif self.pb.algorithm == 'rankboost-fast':\\n boosting = RankBoost(self)\\n wl = AdaBoostFastWL(self)\\n else:\\n raise Exception(\\\"Unknown Boosting Algorithm\\\")\\n \\n for r in range(self.pb.rounds):\\n tree = wl.run(dt)\\n dt = boosting.run(dt = dt,\\n r = r,\\n tree = tree)\\n \\n if self.isXvalMain:\\n boosting.finalize()\\n \\n \\\"\\\"\\\"Sync the predictions and save them to a file\\\"\\\"\\\"\\n if self.pb.isLeader:\\n if self.pb.xvalEN and not self.isXvalMain:\\n val_predictions = boosting.get_val_predictions()\\n hypotheses = boosting.get_hypotheses()\\n np.savez(self.out_fp,\\n val_predictions = val_predictions,\\n hypotheses = hypotheses,\\n )\\n if self.pb.testEN and self.isXvalMain:\\n train_predictions = np.zeros([self.pb.total_exam_no,self.pb.rounds],\\n dtype=\\\"float32\\\")\\n test_predictions = np.zeros([self.pb.test_exam_no,self.pb.rounds],\\n dtype=\\\"float32\\\")\\n for slv in np.arange(self.pb.comm_size):\\n tr_i1 = self.pb.train_partition[slv]\\n tr_i2 = self.pb.train_partition[slv+1]\\n te_i1 = self.pb.test_partition[slv]\\n te_i2 = self.pb.test_partition[slv+1]\\n if slv == 0:\\n train_predictions[tr_i1:tr_i2,:] = boosting.get_train_predictions()\\n test_predictions[te_i1:te_i2,:] = boosting.get_test_predictions()\\n else:\\n train_predictions[tr_i1:tr_i2,:] = self.pb.comm.recv(source=slv,tag=11)\\n test_predictions[te_i1:te_i2,:] = self.pb.comm.recv(source=slv,tag=12)\\n hypotheses = boosting.get_hypotheses()\\n np.savez(self.out_fp,\\n train_predictions = train_predictions,\\n test_predictions = test_predictions,\\n hypotheses = hypotheses,\\n )\\n if not self.pb.testEN and self.isXvalMain:\\n train_predictions = np.zeros([self.pb.total_exam_no,self.pb.rounds],\\n dtype=\\\"float32\\\")\\n for slv in np.arange(self.pb.comm_size):\\n tr_i1 = self.pb.train_partition[slv]\\n tr_i2 = self.pb.train_partition[slv+1]\\n if slv == 0:\\n train_predictions[tr_i1:tr_i2,:] = boosting.get_train_predictions()\\n else:\\n train_predictions[tr_i1:tr_i2,:] = self.pb.comm.recv(source=slv,tag=11)\\n hypotheses = boosting.get_hypotheses()\\n np.savez(self.out_fp,\\n train_predictions = train_predictions,\\n hypotheses = hypotheses,\\n )\\n else:\\n if self.pb.testEN and self.isXvalMain:\\n train_predictions = boosting.get_train_predictions()\\n self.pb.comm.send(train_predictions,dest = 0,tag=11)\\n test_predictions = boosting.get_test_predictions()\\n self.pb.comm.send(test_predictions,dest = 0,tag=12)\\n if not self.pb.testEN and self.isXvalMain:\\n train_predictions = boosting.get_train_predictions()\\n self.pb.comm.send(train_predictions,dest = 0,tag=11)\",\n \"def _train_and_evaluate(estimator, output_dir):\\n \\n \\\"\\\"\\\"X_train, y_train =utils._feature_label_split(df_train,\\\"is_churn\\\",\\\"msno\\\")\\n df_val = utils.read_from_bigquery(\\\"amiable-octane-267022.kkbox.output_val_1\\\",\\\"amiable-octane-267022\\\")\\n X_val, y_val =utils._feature_label_split(df_val,\\\"is_churn\\\",\\\"msno\\\")\\\"\\\"\\\"\\n \\n df_train=utils.over_sample(\\\"amiable-octane-267022.kkbox.output_train_1\\\",\\\"amiable-octane-267022\\\")\\n X_train, y_train =utils._feature_label_split(df_train,\\\"is_churn\\\",\\\"msno\\\")\\n df_val=utils.over_sample(\\\"amiable-octane-267022.kkbox.output_val_1\\\",\\\"amiable-octane-267022\\\")\\n X_val, y_val =utils._feature_label_split(df_val,\\\"is_churn\\\",\\\"msno\\\")\\n\\n estimator.fit(X_train, y_train)\\n f1_scorer = make_scorer(f1_score)\\n accuracy_scorer =make_scorer(accuracy_score)\\n\\n if metadata.HYPERPARAMTER_TUNING:\\n scores=model_selection.cross_val_score(estimator, X_val, y_val, cv=3,scoring=f1_scorer)\\n #,scoring=f1_scorer\\n\\n logging.info('Score: %s', scores)\\n\\n #tune hyper\\n hpt = hypertune.HyperTune()\\n hpt.report_hyperparameter_tuning_metric(\\n hyperparameter_metric_tag='F1_SCORE',\\n metric_value=np.mean(scores),\\n global_step=10000)\\n \\n#joblib.dump(estimator, 'model.joblib')\\n\\n # Write model and eval metrics to `output_dir`\\n model_output_path = os.path.join(output_dir, 'model',metadata.MODEL_FILE_NAME)\\n \\n utils.dump_object(estimator, model_output_path)\",\n \"def run(self):\\n if self.pb.xvalEN and not self.isXvalMain:\\n dt = np.ones(self.pb.total_exam_no,\\n dtype=\\\"float32\\\") / self.train_exam_no\\n dt[self.val_indices] = 0.0\\n else:\\n dt = np.ones(self.pb.total_exam_no,\\n dtype=\\\"float32\\\") / self.pb.total_exam_no\\n\\n val = np.zeros(8,dtype=\\\"float32\\\")-1\\n boosting = None\\n wl = None\\n if self.pb.algorithm == 'conf-rated':\\n boosting = ConfidenceRated(self)\\n wl = ConfidenceRatedWL(self)\\n elif self.pb.algorithm == 'adaboost':\\n boosting = AdaBoost(self)\\n wl = AdaBoostWL(self)\\n elif self.pb.algorithm == 'adaboost-fast':\\n boosting = AdaBoostFast(self)\\n wl = AdaBoostFastWL(self)\\n elif self.pb.algorithm == 'rankboost':\\n boosting = RankBoost(self)\\n wl = ConfidenceRatedWL(self)\\n elif self.pb.algorithm == 'rankboost-fast':\\n boosting = RankBoost(self)\\n wl = AdaBoostFastWL(self)\\n else:\\n raise Exception(\\\"Unknown Boosting Algorithm\\\")\\n \\n for r in range(self.pb.rounds):\\n tree = wl.run(dt)\\n dt = boosting.run(dt = dt,\\n r = r,\\n tree = tree)\\n \\n if self.isXvalMain:\\n boosting.finalize()\\n \\n \\\"\\\"\\\"Sync the predictions and save them to a file\\\"\\\"\\\"\\n if self.pb.isLeader:\\n if self.pb.xvalEN and not self.isXvalMain:\\n val_predictions = boosting.get_val_predictions()\\n hypotheses = boosting.get_hypotheses()\\n np.savez(self.out_fp,\\n val_predictions = val_predictions,\\n hypotheses = hypotheses,\\n )\\n if self.pb.testEN and self.isXvalMain:\\n train_predictions = boosting.get_train_predictions()\\n test_predictions = boosting.get_test_predictions()\\n hypotheses = boosting.get_hypotheses()\\n np.savez(self.out_fp,\\n train_predictions = train_predictions,\\n test_predictions = test_predictions,\\n hypotheses = hypotheses,\\n )\\n if not self.pb.testEN and self.isXvalMain:\\n train_predictions = boosting.get_train_predictions()\\n hypotheses = boosting.get_hypotheses()\\n np.savez(self.out_fp,\\n train_predictions = train_predictions,\\n hypotheses = hypotheses,\\n )\",\n \"def compute_and_print_eval_metrics(self):\\n s = ('%20s' + '%12s' * 6) % ('Class', 'Images', 'Targets', 'P', 'R', 'mAP@.5', 'mAP@.5:.95')\\n precision, recall, f1, mean_precision, mean_recall, map50, map = 0., 0., 0., 0., 0., 0., 0.\\n ap = []\\n eval_stats = [np.concatenate(x, 0) for x in zip(*self.eval_stats)]\\n if len(eval_stats) and eval_stats[0].any():\\n precision, recall, ap, f1, ap_class = ap_per_class(*eval_stats)\\n precision, recall, ap50, ap = precision[:, 0], recall[:, 0], ap[:, 0], ap.mean(1)\\n mean_precision, mean_recall, map50, map = precision.mean(), recall.mean(), ap50.mean(), ap.mean()\\n nt = np.bincount(eval_stats[3].astype(np.int64), minlength=len(self.class_names)) # number of targets per class\\n else:\\n nt = np.zeros(1)\\n\\n pf = '%20s' + '%12.5g' * 6 # print format\\n print(\\\"\\\\n EVALUTAION \\\\n\\\")\\n print(s)\\n print(pf % ('all', self.seen, nt.sum(), mean_precision, mean_recall, map50, map))\\n if self.cfg.eval.verbose:\\n for indx, cls in enumerate(ap_class):\\n print(pf % (self.class_names[cls], self.seen, nt[cls], precision[indx], recall[indx], ap50[indx], ap[indx]))\",\n \"def test_precision_recall_curves_vis_api(experiment_to_use):\\n experiment = experiment_to_use\\n probabilities = experiment.probabilities\\n viz_outputs = (\\\"pdf\\\", \\\"png\\\")\\n positive_label = 1\\n with TemporaryDirectory() as tmpvizdir:\\n for viz_output in viz_outputs:\\n vis_output_pattern_pdf = tmpvizdir + f\\\"/*.{viz_output}\\\"\\n visualize.precision_recall_curves(\\n [probabilities, probabilities],\\n experiment.ground_truth,\\n experiment.ground_truth_metadata,\\n experiment.output_feature_name,\\n positive_label,\\n model_names=[\\\"Model1\\\", \\\"Model2\\\"],\\n output_directory=tmpvizdir,\\n file_format=viz_output,\\n )\\n figure_cnt = glob.glob(vis_output_pattern_pdf)\\n assert 1 == len(figure_cnt)\",\n \"def test(self, plot=False):\\n accuracy_list = []\\n fobj_avg = self.load_stats()\\n\\n for ii in range(settings.PARS['maxIters']):\\n model = self.load_model(ii)\\n D1 = model['D']\\n W1 = model['W']\\n\\n # classification\\n tic = time.time()\\n accuracy_list.append(self.classification(D1, W1)[1])\\n toc = time.time()\\n print(\\n 'Final recognition rate for OnlineDL is : {} , objective function value: {}, time: {}'\\n .format(accuracy_list[ii], fobj_avg[ii], toc-tic)\\n )\\n\\n accuracy_list = np.asarray(accuracy_list)\\n\\n print('Best recognition rate for OnlineDL is {} at iteration {}'.format(\\n accuracy_list.max(), accuracy_list.argmax()))\\n\\n if plot:\\n # plot the objective function values for all iterations\\n plt.clf()\\n plt.plot(list(fobj_avg.keys()), list(fobj_avg.values()), 'mo--', linewidth=2)\\n plt.xlabel('Iterations')\\n plt.ylabel('Average objective function value')\\n plt.xticks(list(range(0, 20)), list(range(1, 21)))\\n plt.show()\\n\\n plt.clf()\\n plt.plot(accuracy_list, 'rs--', linewidth=2)\\n plt.xticks(list(range(0, 20)), list(range(1, 21)))\\n plt.xlabel('Iterations')\\n plt.ylabel('Accuracy')\\n plt.show()\",\n \"def testDriver():\\n exam1=90\\n exam2=85\\n assignmentScores = [50, 60, 70, 80, ]\\n computeGrades(exam1, exam2, assignmentScores)\",\n \"def run_regression_experiments(self):\\n assert self.patientIDs # assert not empty\\n self.log.info(\\\"Running experiments for {} patients\\\".format(len(self.patientIDs)))\\n\\n results_batch = defaultdict(dict)\\n results_single = dict()\\n for patient in self.patientIDs:\\n patient_experiment = PatientPredictionExperiment(\\n patient_id=patient, algorithm=self.algorithm,\\n is_batch=self.isBatch\\n )\\n # run results for single patients\\n if not self.isBatch:\\n results_single[patient] = patient_experiment.run_experiment()\\n self.log.debug(\\\"Start processing single results: {}\\\".format(results_single[patient]))\\n # super verbose debug information should be logged with debug option\\n else:\\n # TODO: think about consistent interfaces\\n batch_results, self._allFeatureDesp = patient_experiment.run_experiment()\\n self.log.info(\\\"Start processing batch results with {} batches\\\".format(len(batch_results)))\\n batchNo = 0\\n for results in batch_results:\\n # here are the result for each feature subset for each patients\\n results_batch[batchNo][patient] = results\\n batchNo += 1\\n if self.algorithm == \\\"fm\\\":\\n return\\n if not self.isBatch:\\n results_overall = self.process_results(results_single)\\n self.store_results(results_overall, results_single)\\n self.log.info(\\\"Finished running experiment\\\")\\n else:\\n self.record_batch_experiment()\\n for batchNo, patient_results in results_batch.iteritems():\\n # patient_results = results_single\\n results_overall = self.process_results(patient_results)\\n self.store_results(results_overall, patient_results, self._allFeatureDesp[batchNo])\\n self.log.info(\\\"Finished running experiment No. {}\\\".format(batchNo))\",\n \"def _run_evaluation(\\n sess, experiment, eval_config, output_dir, min_range, max_range, num_bins,\\n torsion_bins):\\n tf.io.gfile.makedirs(os.path.join(output_dir, 'pickle_files'))\\n\\n logging.info('Eval config is %s\\\\nnum_bins: %d', eval_config, num_bins)\\n num_examples = 0\\n num_crops = 0\\n start_all_time = time.time()\\n\\n # Either do the whole test set, or up to a specified limit.\\n max_examples = experiment.num_eval_examples\\n if eval_config.max_num_examples > 0:\\n max_examples = min(max_examples, eval_config.max_num_examples)\\n\\n while num_examples < max_examples:\\n one_prediction = compute_one_prediction(\\n num_examples, experiment, sess, eval_config, num_bins, torsion_bins)\\n\\n single_message = one_prediction.single_message\\n num_crops_local = one_prediction.num_crops_local\\n sequence = one_prediction.sequence\\n filebase = one_prediction.filebase\\n softmax_probs = one_prediction.softmax_probs\\n ss = one_prediction.ss\\n asa = one_prediction.asa\\n torsions = one_prediction.torsions\\n\\n num_examples += 1\\n num_crops += num_crops_local\\n\\n # Save the output files.\\n filename = os.path.join(output_dir,\\n 'pickle_files', '%s.pickle' % filebase)\\n distogram_io.save_distance_histogram(\\n filename, softmax_probs, filebase, sequence,\\n min_range=min_range, max_range=max_range, num_bins=num_bins)\\n\\n if experiment.model.torsion_multiplier > 0:\\n torsions_dir = os.path.join(output_dir, 'torsions')\\n tf.io.gfile.makedirs(torsions_dir)\\n distogram_io.save_torsions(torsions_dir, filebase, sequence, torsions)\\n\\n if experiment.model.secstruct_multiplier > 0:\\n ss_dir = os.path.join(output_dir, 'secstruct')\\n tf.io.gfile.makedirs(ss_dir)\\n secstruct.save_secstructs(ss_dir, filebase, None, sequence, ss)\\n\\n if experiment.model.asa_multiplier > 0:\\n asa_dir = os.path.join(output_dir, 'asa')\\n tf.io.gfile.makedirs(asa_dir)\\n secstruct.save_secstructs(asa_dir, filebase, None, sequence,\\n np.expand_dims(asa, 1), label='Deepmind 2D ASA')\\n\\n time_spent = time.time() - start_all_time\\n logging.info(\\n 'Evaluate %d examples, %d crops %.1f crops/ex. '\\n 'Took %.1fs, %.3f s/example %.3f crops/s\\\\n%s',\\n num_examples, num_crops, num_crops / float(num_examples), time_spent,\\n time_spent / num_examples, num_crops / time_spent, single_message)\\n\\n logging.info('Tested on %d', num_examples)\",\n \"def print_results(self, f1_score, precision_score, recall_score):\\n print(\\\"Algorithm: %s\\\" % self.name)\\n self.print_score(f1_score, \\\"F1 Score\\\")\\n self.print_score(precision_score, \\\"Precision Score\\\")\\n self.print_score(recall_score, \\\"Recall Score\\\")\",\n \"def test_run():\\n # Only few steps for test\\n timesteps = 128\\n\\n # Compute all sub testing conf\\n envs = ['CartPole-v0']\\n ml_platforms = ['torch', 'tf']\\n agents = ['dqn', 'a2c']\\n\\n test_combinations = list(it.product(\\n envs,\\n ml_platforms,\\n agents\\n )\\n )\\n\\n # Finally test them all\\n for conf in test_combinations:\\n env_str, ml_platform_str, agent_str = conf\\n run(\\n agent_str,\\n ml_platform_str,\\n env_str,\\n 'dense',\\n timesteps,\\n './target/')\",\n \"def eval_perf_total(model, X_train, y_train, X_test, y_test):\\n\\n y_hat_train = model.predict(X_train)\\n y_hat_test = model.predict(X_test)\\n \\n train_mae = metrics.mean_absolute_error(y_train, y_hat_train)\\n train_mse = metrics.mean_squared_error(y_train, y_hat_train)\\n train_rmse = np.sqrt(metrics.mean_squared_error(y_train, y_hat_train))\\n train_r = metrics.r2_score(y_train, y_hat_train)\\n\\n print('Evaluating Performance on Training Data:\\\\n')\\n print(f' Train Mean Absolute Error: {train_mae:,.2f}')\\n print(f' Train Mean Squared Error: {train_mse:,.2f}\\\\n')\\n print(f'Train Root Mean Squared Error: {train_rmse:,.2f}')\\n print(f'Train R-Square Value: {round(train_r,2)}')\\n\\n print('\\\\n'+'---'*25+'\\\\n')\\n\\n test_mae = metrics.mean_absolute_error(y_test, y_hat_test)\\n test_mse = metrics.mean_squared_error(y_test, y_hat_test)\\n test_rmse = np.sqrt(metrics.mean_squared_error(y_test, y_hat_test))\\n test_r = metrics.r2_score(y_test, y_hat_test)\\n\\n print('Evaluating Performance on Testing Data:\\\\n')\\n print(f' Test Mean Absolute Error: {test_mae:,.2f}')\\n print(f' Test Mean Squared Error: {test_mse:,.2f}\\\\n')\\n print(f'Test Root Mean Squared Error: {test_rmse:,.2f}')\\n print(f'Test R-Square Value: {round(test_r,2)}')\",\n \"def main():\\n parser = argparse.ArgumentParser(description=\\\"Process the results of an experiment.\\\")\\n parser.add_argument(\\\"experiment\\\")\\n arguments = parser.parse_args()\\n path = f\\\"experiments/{arguments.experiment}\\\"\\n if not os.path.exists(path):\\n raise SystemExit(f\\\"Path {path} does not exists.\\\")\\n\\n # For efficiency, one should generate the results from the parts without merging them.\\n files = [file for file in os.listdir(path) if os.path.isfile(os.path.join(path, file))]\\n frames = []\\n for file in files:\\n device, experiment, _ = file.split(\\\".\\\")\\n frame = pandas.read_csv(\\n os.path.join(path, file),\\n index_col=\\\"variable\\\",\\n usecols=[\\\"variable\\\", \\\"group_index\\\", \\\"value_i\\\"], dtype={\\\"value_i\\\": \\\"Int64\\\"}\\n )\\n frame[\\\"board\\\"] = device\\n frame[\\\"experiment\\\"] = experiment\\n frames.append(frame)\\n dataframe = pandas.concat(frames)\\n frames = None\\n\\n current_grouping = dataframe.groupby([\\\"group_index\\\", \\\"variable\\\"])\\n \\n data = current_grouping.agg([\\n numpy.median,\\n _percentile_factory(95),\\n numpy.mean,\\n numpy.std,\\n \\\"count\\\"\\n ])\\n\\n print(data)\\n \\n data = data.droplevel([0], axis=1)\\n data = data.unstack()\\n data.columns = data.columns.map('_'.join)\\n data.to_csv(f\\\"{arguments.experiment}.csv\\\")\",\n \"def test_compare(self):\\n config = {\\n 'num_components': 512,\\n 'num_features': 128,\\n 'covariance': 'spherical'\\n }\\n\\n samples = self.generate_samples(config, 100_000)\\n sklearn_time = np.mean([self.train_sklearn(config, samples) for _ in range(3)])\\n ours_cpu_time = np.mean([self.train_ours(config, samples) for _ in range(3)])\\n ours_gpu_time = np.mean([\\n self.train_ours(config, samples.cuda(), gpu=True) for _ in range(3)\\n ])\\n\\n print(f\\\"-------------------------------------\\\")\\n print(f\\\"Speedup of CPU implementation: {sklearn_time / ours_cpu_time:.2f}\\\")\\n print(f\\\"Speedup of GPU implementation: {sklearn_time / ours_gpu_time:.2f}\\\")\\n print(f\\\"-------------------------------------\\\")\",\n \"def log_evaluation(tester, name, description):\\r\\n\\tfor dataset, output in tester.preds.items():\\r\\n\\t\\tresults = pandas.DataFrame.from_dict(output)\\r\\n\\t\\tpath = os.path.join(\\r\\n\\t\\t\\tEXPERIMENT_PATH, tester.config[\\\"name\\\"] + '-' + dataset)\\r\\n\\t\\twith open(path + \\\".csv\\\", \\\"w\\\") as f:\\r\\n\\t\\t\\tresults.to_csv(f, sep=\\\"\\\\t\\\", encoding='utf-8',\\r\\n\\t\\t\\t\\tfloat_format='%.3f', index=False)\",\n \"def eval(self):\\n\\n # How many questions we get right at precision@1.\\n correct = 0\\n\\n total = self._analogy_questions.shape[0]\\n start = 0\\n while start < total:\\n limit = start + 2500\\n sub = self._analogy_questions[start:limit, :]\\n idx = self._predict(sub)\\n start = limit\\n for question in xrange(sub.shape[0]):\\n if sub[question, 3] in idx[question]:\\n # print(sub[question, 3], idx[question])\\n correct += 1\\n\\n print()\\n print(\\\"Eval %4d/%d accuracy @ top5= %4.1f%%\\\" % (correct, total,\\n correct * 100. / total)\\n )\",\n \"def main():\\n for opt in optimizations:\\n compile_command = [\\\"g++\\\", \\\"main.cpp\\\", f\\\"-O{opt}\\\", \\\"-lpthread\\\"]\\n run(compile_command, check=True)\\n for threads in num_threads:\\n print(f\\\"{opt=}, {threads=}\\\", end=\\\"\\\")\\n stdout.flush()\\n test_command = ['./a.out', str(iterations), str(threads)]\\n total = 0\\n for samples in range(1, repeats_for_average + 1):\\n print(\\\".\\\", end=\\\"\\\")\\n stdout.flush()\\n output = run(test_command, check=True, capture_output=True).stdout\\n total += int(output.split()[-2]) / 1000\\n print(f\\\"\\\\t{total / samples:.03f}\\\")\",\n \"def run_experiments(data_set=\\\"\\\",compact=2,exp_name=\\\"\\\",x=\\\"\\\"):\\n cwd = os.getcwd()\\n results_path = cwd+'/data/geometric/'+exp_name+'_'\\n dataset_path = cwd+'/data/geometric/'+data_set+'_'\\n compact = int(compact)\\n if \\\"m\\\" in x:\\n # make new pxy\\n pxy, Xdata, groups = gen_geometric_pxy()\\n np.save(dataset_path+'Xdata',Xdata)\\n np.save(dataset_path+'groups',groups)\\n np.save(dataset_path+'pxy',pxy)\\n else:\\n # load existing pxy\\n pxy = np.load(dataset_path+'pxy.npy')\\n if \\\"r\\\" in x: # regular experiments\\n if compact>1:\\n metrics_stepwise, distributions_stepwise,\\\\\\n metrics_converged, distributions_converged,\\\\\\n metrics_stepwise_allreps, distributions_stepwise_allreps,\\\\\\n metrics_converged_allreps, distributions_converged_allreps = test_IB(pxy,compact)\\n metrics_converged.to_csv(results_path+'metrics_converged.csv')\\n metrics_stepwise.to_csv(results_path+'metrics_stepwise.csv')\\n distributions_converged.to_pickle(results_path+'distributions_converged.pkl')\\n elif compact>0:\\n metrics_stepwise,\\\\\\n metrics_converged, distributions_converged,\\\\\\n metrics_stepwise_allreps,\\\\\\n metrics_converged_allreps, distributions_converged_allreps = test_IB(pxy,compact)\\n metrics_converged.to_csv(results_path+'metrics_converged.csv')\\n metrics_stepwise.to_csv(results_path+'metrics_stepwise.csv')\\n distributions_converged.to_pickle(results_path+'distributions_converged.pkl')\\n else:\\n metrics_stepwise,\\\\\\n metrics_converged,\\\\\\n metrics_stepwise_allreps,\\\\\\n metrics_converged_allreps = test_IB(pxy,compact)\\n metrics_converged.to_csv(results_path+'metrics_converged.csv')\\n metrics_stepwise.to_csv(results_path+'metrics_stepwise.csv')\\n if \\\"ip\\\" in x: # initialization experiments - positive p0\\n if compact>1:\\n metrics_stepwise, distributions_stepwise,\\\\\\n metrics_converged, distributions_converged,\\\\\\n metrics_stepwise_allreps, distributions_stepwise_allreps,\\\\\\n metrics_converged_allreps, distributions_converged_allreps = test_p0_pos(pxy,compact)\\n metrics_converged.to_csv(results_path+'metrics_converged_p0_pos.csv')\\n metrics_stepwise.to_csv(results_path+'metrics_stepwise_p0_pos.csv')\\n distributions_converged.to_pickle(results_path+'distributions_converged_p0_pos.pkl')\\n elif compact>0:\\n metrics_stepwise,\\\\\\n metrics_converged, distributions_converged,\\\\\\n metrics_stepwise_allreps,\\\\\\n metrics_converged_allreps, distributions_converged_allreps = test_p0_pos(pxy,compact)\\n metrics_converged.to_csv(results_path+'metrics_converged_p0_pos.csv')\\n metrics_stepwise.to_csv(results_path+'metrics_stepwise_p0_pos.csv')\\n distributions_converged.to_pickle(results_path+'distributions_converged_p0_pos.pkl')\\n else:\\n metrics_stepwise,\\\\\\n metrics_converged,\\\\\\n metrics_stepwise_allreps,\\\\\\n metrics_converged_allreps = test_p0_pos(pxy,compact)\\n metrics_converged.to_csv(results_path+'metrics_converged_p0_pos.csv')\\n metrics_stepwise.to_csv(results_path+'metrics_stepwise_p0_pos.csv')\\n if \\\"in\\\" in x: # initialization experiments - negative p0\\n if compact>1:\\n metrics_stepwise, distributions_stepwise,\\\\\\n metrics_converged, distributions_converged,\\\\\\n metrics_stepwise_allreps, distributions_stepwise_allreps,\\\\\\n metrics_converged_allreps, distributions_converged_allreps = test_p0_neg(pxy,compact)\\n metrics_converged.to_csv(results_path+'metrics_converged_p0_neg.csv')\\n metrics_stepwise.to_csv(results_path+'metrics_stepwise_p0_neg.csv')\\n distributions_converged.to_pickle(results_path+'distributions_converged_p0_neg.pkl')\\n elif compact>0:\\n metrics_stepwise,\\\\\\n metrics_converged, distributions_converged,\\\\\\n metrics_stepwise_allreps,\\\\\\n metrics_converged_allreps, distributions_converged_allreps = test_p0_neg(pxy,compact)\\n metrics_converged.to_csv(results_path+'metrics_converged_p0_neg.csv')\\n metrics_stepwise.to_csv(results_path+'metrics_stepwise_p0_neg.csv')\\n distributions_converged.to_pickle(results_path+'distributions_converged_p0_neg.pkl')\\n else:\\n metrics_stepwise,\\\\\\n metrics_converged,\\\\\\n metrics_stepwise_allreps,\\\\\\n metrics_converged_allreps = test_p0_neg(pxy,compact)\\n metrics_converged.to_csv(results_path+'metrics_converged_p0_neg.csv')\\n metrics_stepwise.to_csv(results_path+'metrics_stepwise_p0_neg.csv')\\n if \\\"c\\\" in x: # convergence tolerance experiments\\n if compact>1:\\n metrics_stepwise, distributions_stepwise,\\\\\\n metrics_converged, distributions_converged,\\\\\\n metrics_stepwise_allreps, distributions_stepwise_allreps,\\\\\\n metrics_converged_allreps, distributions_converged_allreps = test_ctol(pxy,compact)\\n metrics_converged.to_csv(results_path+'metrics_converged_ctol.csv')\\n metrics_stepwise.to_csv(results_path+'metrics_stepwise_ctol.csv')\\n distributions_converged.to_pickle(results_path+'distributions_converged_ctol.pkl')\\n elif compact>0:\\n metrics_stepwise,\\\\\\n metrics_converged, distributions_converged,\\\\\\n metrics_stepwise_allreps,\\\\\\n metrics_converged_allreps, distributions_converged_allreps = test_ctol(pxy,compact)\\n metrics_converged.to_csv(results_path+'metrics_converged_ctol.csv')\\n metrics_stepwise.to_csv(results_path+'metrics_stepwise_ctol.csv')\\n distributions_converged.to_pickle(results_path+'distributions_converged_ctol.pkl')\\n else:\\n metrics_stepwise,\\\\\\n metrics_converged,\\\\\\n metrics_stepwise_allreps,\\\\\\n metrics_converged_allreps = test_ctol(pxy,compact)\\n metrics_converged.to_csv(results_path+'metrics_converged_ctol.csv')\\n metrics_stepwise.to_csv(results_path+'metrics_stepwise_ctol.csv')\\n if \\\"z\\\" in x: # zeroL tolerance experiments\\n if compact>1:\\n metrics_stepwise, distributions_stepwise,\\\\\\n metrics_converged, distributions_converged,\\\\\\n metrics_stepwise_allreps, distributions_stepwise_allreps,\\\\\\n metrics_converged_allreps, distributions_converged_allreps = test_zeroLtol(pxy,compact)\\n metrics_converged.to_csv(results_path+'metrics_converged_zeroLtol.csv')\\n metrics_stepwise.to_csv(results_path+'metrics_stepwise_zeroLtol.csv')\\n distributions_converged.to_pickle(results_path+'distributions_converged_zeroLtol.pkl')\\n elif compact>0:\\n metrics_stepwise,\\\\\\n metrics_converged, distributions_converged,\\\\\\n metrics_stepwise_allreps,\\\\\\n metrics_converged_allreps, distributions_converged_allreps = test_zeroLtol(pxy,compact)\\n metrics_converged.to_csv(results_path+'metrics_converged_zeroLtol.csv')\\n metrics_stepwise.to_csv(results_path+'metrics_stepwise_zeroLtol.csv')\\n distributions_converged.to_pickle(results_path+'distributions_converged_zeroLtol.pkl')\\n else:\\n metrics_stepwise,\\\\\\n metrics_converged,\\\\\\n metrics_stepwise_allreps,\\\\\\n metrics_converged_allreps = test_zeroLtol(pxy,compact)\\n metrics_converged.to_csv(results_path+'metrics_converged_zeroLtol.csv')\\n metrics_stepwise.to_csv(results_path+'metrics_stepwise_zeroLtol.csv')\\n if \\\"b\\\" in x: # trying proposed optimal beta\\n if compact>1:\\n metrics_stepwise, distributions_stepwise,\\\\\\n metrics_converged, distributions_converged,\\\\\\n metrics_stepwise_allreps, distributions_stepwise_allreps,\\\\\\n metrics_converged_allreps, distributions_converged_allreps = test_IB(pxy,compact)\\n metrics_converged.to_csv(results_path+'metrics_converged_bestbeta.csv')\\n metrics_stepwise.to_csv(results_path+'metrics_stepwise_bestbeta.csv')\\n distributions_converged.to_pickle(results_path+'distributions_converged_bestbeta.pkl')\\n elif compact>0:\\n metrics_stepwise,\\\\\\n metrics_converged, distributions_converged,\\\\\\n metrics_stepwise_allreps,\\\\\\n metrics_converged_allreps, distributions_converged_allreps = test_IB(pxy,compact)\\n metrics_converged.to_csv(results_path+'metrics_converged_bestbeta.csv')\\n metrics_stepwise.to_csv(results_path+'metrics_stepwise_bestbeta.csv')\\n distributions_converged.to_pickle(results_path+'distributions_converged_bestbeta.pkl')\\n else:\\n metrics_stepwise,\\\\\\n metrics_converged,\\\\\\n metrics_stepwise_allreps,\\\\\\n metrics_converged_allreps = test_IB(pxy,compact)\\n metrics_converged.to_csv(results_path+'metrics_converged_bestbeta.csv')\\n metrics_stepwise.to_csv(results_path+'metrics_stepwise_bestbeta.csv')\\n return 0\",\n \"def run(self):\\n self.evaluate()\\n self.accumulate()\\n self.summarize()\",\n \"def run(self, verbose=0):\\n self.verbose = verbose\\n self._preproc()\\n self._lda()\\n self._evaluate()\",\n \"def main(titles, paragraphs, printDetailedStats = True):\\n avg_precision = 0 # all the precision values will be added to this value. In the end, this value will be divided by\\n # the number of comparisons analysed to report the average precision\\n avg_recall = 0\\n avg_fmeasure = 0\\n\\n if paragraphs:\\n manual_directory = path.MANUAL_PARAGRAPHS\\n auto_directory = path.OUTDIR_PARAGRAPHS\\n else:\\n manual_directory = path.MANUAL_SENTENCES\\n auto_directory = path.OUTDIR_SENTENCES\\n\\n for title in titles:\\n with open(manual_directory + title) as m:\\n with open(auto_directory + title) as a:\\n tPositive, fPositive, fNegative, precision, recall, fmeasure = \\\\\\n compare(m.readlines(), a.readlines(), paragraphs)\\n avg_precision += precision\\n avg_recall += recall\\n avg_fmeasure += fmeasure\\n if printDetailedStats: # then report precision and recall for every individual article\\n print(\\\"comparing \\\" + title)\\n print(\\\"tp=\\\" + str(tPositive) + \\\" fn=\\\" + str(fNegative) + \\\" fp=\\\" + str(fPositive))\\n print(\\\"precision=\\\" + str(round(precision, 5)) + \\\"\\\\t\\\\t recall=\\\" + str(\\n round(recall, 5)) + \\\"\\\\t\\\\t fmeasure=\\\" + str(round(fmeasure, 5)) + \\\" \\\\n\\\\n\\\")\\n print(\\\"AVERAGE_PRECISION=\\\" + str(round(avg_precision/len(titles), 5)) + \\\"\\\\t\\\\t AVERAGE_RECALL=\\\"\\n + str(round(avg_recall/len(titles), 5)) + \\\"\\\\t\\\\t AVERAGE_F_MEASURE=\\\"\\n + str(round(avg_fmeasure/len(titles), 5)))\\n return avg_fmeasure/len(titles)\",\n \"def run_test(models, data):\\n num_correct = 0.0\\n num_total = 0.0\\n\\n # initialize lists to put in predictions & results for confusion matrix \\n predicted_labels = []\\n actual_labels = []\\n langlist = models.keys()\\n for ai, actual_lang in enumerate(langlist):\\n test_files = open(os.path.join('traintestsplit', actual_lang+'.testlist')).read().split()\\n print 'Testing', len(test_files), 'files from', actual_lang\\n for filename in test_files:\\n logprobs = {} # dict: total log prob of this file under each model \\n for test_lang in langlist:\\n logprobs[test_lang] = apply_model(models[test_lang], data[actual_lang][filename+'.npytxt'])\\n predicted_lang = max(logprobs.items(), key=lambda x:x[1])[0]\\n # insert prediction (of lang index) into predicted list \\n predicted_labels.append(langlist.index(predicted_lang))\\n actual_labels.append(ai)\\n if actual_lang == predicted_lang:\\n num_correct += 1\\n num_total += 1\\n\\n print len(filter(lambda x:x==ai, predicted_labels[-len(test_files):])), 'correct'\\n\\n print\\n print 'Accuracy', num_correct*100/num_total\\n\\n #CONFUSION MATRIX (y_test, y_pred) -> (actual label, predictions) \\n cm = confusion_matrix(actual_labels, predicted_labels)\\n cm_normalized = cm.astype('float') / cm.sum(axis=1)[:, np.newaxis]\\n # display confusion stats by lang (TODO: visualize with matplotlib) \\n print '*'*20\\n for ai, actual_lang in enumerate(langlist):\\n print actual_lang, 'confusion:'\\n for pi, predicted_lang in enumerate(langlist):\\n print '{0}: {1:.2f}%'.format(predicted_lang, cm_normalized[ai, pi]*100)\\n print '*'*20\",\n \"def test_print_results(self):\\n calculated = super().predict_and_print()\\n self.assertEqual(calculated, EXP_PRINT_OUTPUT_BASE.format(.18, .1, 0.186, self.test_model.model.train_time) +\\n \\\"Max tree max_depth: 1\\\\n\\\"\\n \\\"Number of n_estimators: 1\\\\n\\\"\\n \\\"Impurity method: entropy\\\\n\\\")\",\n \"def run_tests(): \\n \\n\\n nextdata = [[21, 61, 42, 30], [33,45, 18, 29]]\\n\\n for xval, yval, snum, expect in nextdata:\\n\\n pmachine = PMachine()\\n pmachine.serial_number = snum\\n pmachine.run2_completion()\\n result = pmachine.calc_square_total(xval, yval, showsquare=True)\\n assert result == expect\\n print(\\\"Got value {}={} as expected\\\".format(result, expect))\",\n \"def run(self):\\n logging.info('running experiment...')\\n self._prepare()\\n self._load_data()\\n self._run()\\n self._evaluate()\\n self._summarise()\\n return True\",\n \"def exe_tests(self):\\n self.rank = mpicom.rank()\\n self.size = mpicom.size()\\n if mpicom.parallel():\\n self.test(\\\"libname\\\",os.path.split(mpicom.__file__)[1],\\\"mpicom.so\\\")\\n else:\\n self.test(\\\"libname\\\",os.path.split(mpicom.__file__)[1],\\\"mpistub.pyc\\\")\\n self.test_broadcast()\\n self.test_reduce()\\n self.test_p2p()\\n self.test_gather()\\n self.test_scatter()\\n #self.test_alltoall()\",\n \"def _compute_experiment_statistics(self):\\n pass\",\n \"def run(self):\\n evaluateAllRuns = False\\n while True:\\n if self.host == \\\"\\\":\\n # respond to clients\\n self.respond2Clients()\\n else:\\n print(\\\"Next...\\\")\\n # randomly choose experiment + run\\n if not evaluateAllRuns:\\n print(\\\"Randomly fetching run\\\")\\n self.exp, self.runnum, self.detname = randExpRunDet()\\n else:\\n\\t\\t try:\\n print(\\\"Fecthing next run in experiment\\\")\\n self.exp, self.runnum, self.detname = nextExpRunDet(self.goodExp, self.runList[0])\\n if self.exp is None:\\n self.runList.pop(0)\\n continue\\n except:\\n evaluateAllRuns = False\\n continue\\n if not self.checkStatus(self.exp, self.runnum, self.detname):\\n print \\\"trying: exp %s, run %s, det %s\\\"%(self.exp,self.runnum,self.detname)\\n try: #temp\\n self.ds = safeDataSource(self.exp, self.runnum)\\n except: #temp\\n continue #temp\\n self.run = self.ds.runs().next()\\n self.times = self.run.times()\\n #Start temp code\\n if self.detname is None:\\n continue\\n #End temp code\\n self.det = psana.Detector(self.detname)\\n self.det.do_reshape_2d_to_3d(flag=True)\\n try:\\n self.iX = np.array(self.det.indexes_x(self.run), dtype=np.int64)\\n self.iY = np.array(self.det.indexes_y(self.run), dtype=np.int64)\\n self.ipx, self.ipy = self.det.point_indexes(self.run, pxy_um=(0, 0))\\n self.alg = PyAlgos()\\n self.alg.set_peak_selection_pars(npix_min=2, npix_max=30, amax_thr=300, atot_thr=600, son_min=10)\\n mask = self.det.mask(self.runnum, calib=True, status=True, edges=True, central=True, unbond=True, unbondnbrs=True)\\n\\n samples = np.linspace(0, len(self.times), num=100, endpoint=False, retstep=False, dtype='int')\\n offset = np.floor(np.random.uniform(0, len(self.times)-samples[-1])).astype('int')\\n mysamples = samples + offset\\n numCrystals = 0\\n for self.eventNum in mysamples:\\n self.evt = self.run.event(self.times[self.eventNum])\\n calib = self.det.calib(self.evt)\\n if calib is not None:\\n peaks = self.alg.peak_finder_v3r3(calib, rank=3, r0=3, dr=2, nsigm=10, mask=mask.astype(np.uint16))\\n if self.likelihood(peaks) >= self.goodLikelihood:\\n numCrystals += 1\\n if numCrystals >= self.minCrystals:\\n self.numSaved +=1\\n self.updateStatus(self.exp, self.runnum, self.detname, self.numSaved)\\n self.lastGood = True\\n break\\n except:\\n print \\\"Could not analyse this run\\\"\\n #If an experiment has not had all of its runs evaluated yet\\n # and if the last randomly selected run in this experiment was good\\n # then all the runs in this experiment should be evaluated\\n if (self.exp not in self.goodList) and self.lastGood:\\n self.goodExp = self.exp #Save the name of this experiment\\n self.goodRun = self.runnum #Save the run that has already been evaluated\\n self.lastGood = False #Reset the condition that the last run was \\\"good\\\"\\n self.goodList.append(self.goodExp) #Add this experiment name to the list of experiments that have had all runs evaluated\\n self.runList = returnRunList(self.goodExp, self.goodRun) #save list of all runs in this good exp\\n evaluateAllRuns = True #rerun loop with new algorithm that evaluates each run in an experiment\\n continue\\n if evaluateAllRuns: #If the loop is currently evaluating all of the runs in an experiment\\n if(len(self.runList) > 1):\\n self.runList.pop(0) #Remove runs from the list of runs each time they are evaluated\\n else:\\n self.runList.pop(0)#Remove runs until the list is completely empty\\n evaluateAllRuns = False #Stop evaluated all the runs of an experiment, go back to random fetching\",\n \"def getTestResults():\",\n \"def accuracy(num_points, num_dims, p, surv, num_runs=100, num_test=10000):\\n p1 = []\\n p2 = []\\n p3 = []\\n p4 = []\\n for t in range(num_runs):\\n if surv:\\n cloud = generate_survival_curves(num_points + num_test, num_dims)\\n else:\\n cloud = generate_time_series(num_points + num_test, num_dims)\\n opt_cloud = cloud[:num_points]\\n test_cloud = cloud[num_points:]\\n p1.append(proportion(test_cloud, olshen(opt_cloud, p, surv)))\\n p2.append(proportion(test_cloud, two_sided_olshen(opt_cloud, p, surv)))\\n p3.append(proportion(test_cloud, gspie(opt_cloud, p, surv)))\\n p4.append(\\n proportion(test_cloud, pointwise_bonferroni(opt_cloud, p, surv)))\\n plt.boxplot([p1, p2, p3, p4])\\n plt.xticks([1, 2, 3, 4],\\n ['Olshen', 'Two-Sided Olshen', 'GSPIE', 'Pointwise Bonferroni'])\\n plt.ylabel('Observed Coverage')\\n plt.xlabel(f\\\"{p} Prescribed Coverage\\\")\\n plt.show()\",\n \"def main():\\n\\n # choose number of data-points and sample a pair of vectors: the input\\n # values and the corresponding target values\\n N = 500\\n inputs, targets = sample_data(N, arbitrary_function_2, seed=1)\\n\\n # specify the centres and scale of some rbf basis functions\\n default_centres = np.linspace(0,1,21)\\n default_scale = 0.03\\n default_reg_param = 0.08\\n\\n # get the cross-validation folds\\n num_folds = 4\\n folds = create_cv_folds(N, num_folds)\\n\\n # evaluate then plot the performance of different reg params\\n evaluate_reg_param(inputs, targets, folds, default_centres, default_scale)\\n # evaluate then plot the performance of different scales\\n evaluate_scale(inputs, targets, folds, default_centres, default_reg_param)\\n # evaluate then plot the performance of different numbers of basis\\n # function centres.\\n evaluate_num_centres(\\n inputs, targets, folds, default_scale, default_reg_param)\\n\\n plt.show()\",\n \"def see_evaluation(epoch, training_acc, test_acc):\\n print (\\\"Epoch \\\", epoch, \\\"Training acc: \\\", training_acc*100, \\\"Test acc: \\\", test_acc*100)\",\n \"def evaluate(model, test_files):\\n print(\\\"Running predictions.\\\")\\n models = load_model(model)\\n predictions = predict(models, test_files)\\n\\n # # write predictions to file\\n # write_predictions(\\\"evaluate_out.json\\\",predictions)\\n evaluate_individual(predictions, test_files, models)\\n evaluate_overall(predictions)\",\n \"def sbm_experiments():\\n # The key experiment is to fix n, r, and p and vary the ratio q/p.\\n n = 1000\\n average_over = 10\\n\\n # We will run a few experiments\\n rs = [5]\\n ps = [1e-11]\\n run_lps = [False]\\n\\n # Use the same ratios for each experiment\\n ratios = [0.5 * x for x in range(4, 61)]\\n\\n # Whether to append results to the results files\\n append_results = False\\n\\n for index in range(len(rs)):\\n sbm_experiment_internal(n, rs[index], ps[index], average_over, run_lps[index], ratios, append_results)\",\n \"def run_tests(self) -> float:\\n print(f\\\"Grading {self._test_case.__name__}\\\", file=sys.stderr)\\n print(file=sys.stderr)\\n for test, weight in self.test_weights.items():\\n print(f\\\"Running {test.__name__}\\\", file=sys.stderr)\\n if test.__doc__:\\n print(test.__doc__, file=sys.stderr)\\n result = self.run(test)\\n if result.wasSuccessful():\\n print(f\\\"Points: {weight}/{weight}\\\", file=sys.stderr)\\n else:\\n print(file=sys.stderr)\\n print(\\n \\\"Test failed with the error below, displayed between lines of ---.\\\",\\n file=sys.stderr,\\n )\\n print(\\n \\\"The expected value is given first, followed by the actual result.\\\",\\n file=sys.stderr,\\n )\\n print(\\\"-\\\" * 70, file=sys.stderr)\\n # Get the error/failure\\n try:\\n print(result.errors[0][1], file=sys.stderr)\\n except IndexError:\\n pass\\n try:\\n print(result.failures[0][1], file=sys.stderr)\\n except IndexError:\\n pass\\n print(\\\"-\\\" * 70, file=sys.stderr)\\n print(f\\\"Points: {0.0}/{weight}\\\", file=sys.stderr)\\n print(file=sys.stderr)\\n print(\\\"=\\\" * 70, file=sys.stderr)\\n return self.grade\",\n \"def print_results(results, random_counterpart=None, random_concepts=None, num_random_exp=100,\\n min_p_val=0.05):\\n\\n # helper function, returns if this is a random concept\\n def is_random_concept(concept):\\n if random_counterpart:\\n return random_counterpart == concept\\n\\n elif random_concepts:\\n return concept in random_concepts\\n\\n else:\\n return 'random500_' in concept\\n\\n # print class, it will be the same for all\\n print(\\\"Class =\\\", results[0]['target_class'])\\n\\n # prepare data\\n # dict with keys of concepts containing dict with bottlenecks\\n result_summary = {}\\n\\n # random\\n random_i_ups = {}\\n\\n for result in results:\\n if result['cav_concept'] not in result_summary:\\n result_summary[result['cav_concept']] = {}\\n\\n if result['bottleneck'] not in result_summary[result['cav_concept']]:\\n result_summary[result['cav_concept']][result['bottleneck']] = []\\n\\n result_summary[result['cav_concept']][result['bottleneck']].append(result)\\n\\n # store random\\n if is_random_concept(result['cav_concept']):\\n if result['bottleneck'] not in random_i_ups:\\n random_i_ups[result['bottleneck']] = []\\n\\n random_i_ups[result['bottleneck']].append(result['i_up'])\\n\\n # print concepts and classes with indentation\\n for concept in result_summary:\\n\\n # if not random\\n if not is_random_concept(concept):\\n print(\\\" \\\", \\\"Concept =\\\", concept)\\n\\n for bottleneck in result_summary[concept]:\\n i_ups = [item['i_up'] for item in result_summary[concept][bottleneck]]\\n\\n # Calculate statistical significance\\n _, p_val = ttest_ind(random_i_ups[bottleneck], i_ups)\\n\\n print(3 * \\\" \\\", \\\"Bottleneck =\\\", (\\\"%s. TCAV Score = %.2f (+- %.2f), \\\"\\n \\\"random was %.2f (+- %.2f). p-val = %.3f (%s)\\\") % (\\n bottleneck, np.mean(i_ups), np.std(i_ups),\\n np.mean(random_i_ups[bottleneck]),\\n np.std(random_i_ups[bottleneck]), p_val,\\n \\\"not significant\\\" if p_val > min_p_val else \\\"significant\\\"))\",\n \"def analyze_end_performance(self) -> None:\\n self.logger.info(f\\\"###### FINAL RESULTS FOR EACH GAME###### \\\\n\\\")\\n total_results = [0, 0, 0, 0]\\n for game in self.instance_predictions:\\n self.logger.info(f\\\"Simulated game {game.home_team.name} vs {game.away_team.name} \\\"\\n f\\\"{len(self.instance_predictions[game])} times.\\\")\\n confusion_matrix = [0, 0, 0, 0]\\n for predicted_winner in self.instance_predictions[game]:\\n actual_winner = game.home_team if game.home_team.scores.get(\\n GamePeriod.TOTAL) > game.away_team.scores.get(GamePeriod.TOTAL) else game.away_team\\n actual_home_team_won = True if actual_winner == game.home_team else False\\n predicted_home_team_won = True if predicted_winner == game.home_team else False\\n confusion_matrix = recursive_build_confusion_matrix(confusion_matrix, actual_home_team_won,\\n predicted_home_team_won)\\n total_results[0] += confusion_matrix[0]\\n total_results[1] += confusion_matrix[1]\\n total_results[2] += confusion_matrix[2]\\n total_results[3] += confusion_matrix[3]\\n\\n for k, v in analyze_confusion_matrix(confusion_matrix, len(self.instance_predictions[game])).items():\\n self.logger.info(f\\\"{k} : {v}\\\")\\n self.logger.info(\\\"\\\\n###### FINAL RESULTS FOR ALL GAMES###### \\\\n\\\")\\n final_stats = analyze_confusion_matrix(total_results, sum(total_results))\\n for k, v in final_stats.items():\\n self.logger.info(f\\\"{k} : {v}\\\")\\n self.write_stats_to_json(final_stats, self.outfile)\",\n \"def check_expectations(self):\\n self.load_results()\\n\\n for (benchmark, producer), result in self.results.items():\\n if not result.reports:\\n print('No results found for ' + benchmark + ' ' + producer)\\n result.test_passed = False\\n else:\\n for report in result.reports:\\n if check_benchmark_result(report, result.expectation):\\n print('Test passed: ' + result.directory)\\n result.test_passed = True\\n else:\\n print('Test failed: ' + result.directory)\\n result.test_passed = False\",\n \"def polling_experiment_multiple(dataset_id, times, **kwargs):\\n total_yeahs = 0\\n total_noes = 0\\n for i in range(0, times):\\n yeah, noes = polling_experiment(dataset_id, **kwargs)\\n total_yeahs += yeah\\n total_noes += noes\\n logging.info(\\\"After \\\"+str(times)+\\\" \\\"+ str(dataset_id) + \\\" experiments:\\\")\\n logging.info(\\\"\\\"+ str(total_yeahs) + \\\" correct predictions\\\")\\n logging.info(\\\"\\\"+ str(total_noes) + \\\" wrong predictions\\\")\\n logging.info(\\\"\\\"+ str(total_yeahs/(total_yeahs + total_noes)) + \\\" accuracy\\\")\",\n \"def main():\\n\\t# \\\"\\\"\\\"\\n\\t# \\tMain function of test python module\\n\\t# \\\"\\\"\\\"\\n\\t# random.seed(os.urandom(345634)) # initialize random generator\\n\\t# t = np.linspace(0.0, 24.0, 96.0) # define the time axis of a day, here we use 96 values every quarter of an hour\\n\\t# # standard load profile -- input\\n\\t# q = extra.read_slp(t,\\n\\t# 'Profielen-Elektriciteit-2015-versie-1.00 Folder/profielen Elektriciteit 2015 versie 1.00.csv') # read the sample standard load profile, can be any length, can be resized given a low/high resolution time axis\\n\\t# q = q / np.sum(q) # normalization of standard load profile\\n\\t# # process duration\\n\\t# duration_axis = np.linspace(0.0, 24.0, 96.0)\\n\\t# (p_d, E_p) = extra.app_time(duration_axis, 10, 2, 0.0,\\n\\t# 24.0) # function that define the pdf of duration of a process\\n\\t# # process consumption\\n\\t# consumption_axis = np.linspace(0.0, 3.5, 96.0)\\n\\t# (p_k, E_k) = extra.app_consumption(consumption_axis, 10, 2, 0.0,\\n\\t# 3.5) # function that define the pdf of duration of a process\\n\\t# # pdf of starting time\\n\\t# p_t_0 = lpd.infer_t_0(q, p_d, E_k) # computes the pdf of starting time of processes\\n\\t# p_t_0 = p_t_0 / np.sum(p_t_0) # normalization of the pdf to sum up to zero\\n #\\n\\t# \\\"\\\"\\\"\\n\\t# 1st Approach, starting time of processes is a discrete propapibility density function\\n\\t# \\\"\\\"\\\"\\n\\t# # synthetic profile of D processes\\n\\t# D = 2000\\n\\t# synthetic_profile = lpd.synthetic_profile(D, t, p_d, consumption_axis, p_k, p_t_0)\\n\\t# synthetic_profile_1 = lpd.synthetic_profile(D, t, p_d, consumption_axis, p_k, p_t_0)\\n\\t# # expected value of D processes\\n\\t# q_e_e = lpd.infer_q_e(t, p_t_0, p_d, E_k, D)\\n\\t# # plot\\n\\t# plt.step(t, synthetic_profile, \\\"g-\\\")\\n\\t# plt.step(t, q_e_e, \\\"b--\\\")\\n #\\n\\t# \\\"\\\"\\\"\\n\\t# 2nd Approach, starting time of processes is a continuous propapibility density function\\n\\t# \\\"\\\"\\\"\\n\\t# # synthetic profile of D processes\\n\\t# ts, cs = lpd.continous_synthetic_profile(D, t, p_d, consumption_axis, p_k, p_t_0)\\n\\t# plt.step(ts / len(t) * t[-1], cs, where='post', c='r')\\n\\t# plt.xlim(0, 24.0)\\n\\t# plt.legend([\\\"synthetic\\\", \\\"expected\\\", \\\"continuous\\\"], loc=0)\\n\\t# plt.show()\\n #\\n\\t# \\\"\\\"\\\"\\n\\t# Time discretization\\n\\t# \\\"\\\"\\\"\\n\\t# n_intervals = 24 * 1 # discretized in minutes\\n\\t# discrete_timeaxis = np.linspace(0.0, 24.0, n_intervals + 1)\\n\\t# discrete_consumption = lpd.signal_discretization(discrete_timeaxis, t, ts, cs)\\n\\t# plt.step(ts / len(t) * t[-1], cs, where='post', c='r')\\n\\t# plt.step(discrete_timeaxis, discrete_consumption, where='post', c='k', ls='--', lw=2)\\n\\t# plt.legend([\\\"continuous\\\", \\\"discretized\\\"], loc=0)\\n\\t# plt.show()\\n #\\n #\\n\\t# \\\"\\\"\\\"\\n\\t# Repeated day synthetic profile creation\\n\\t# \\\"\\\"\\\"\\n\\t# # synthetic profile of D processes\\n\\t# D = 2000\\n\\t# n = 10\\n\\t# slp = lpd.synthetic_profile_repeated(D, t, p_d, consumption_axis, p_k, p_t_0, n)\\n\\t# plt.step(range(len(slp)), slp, \\\"g-\\\")\\n\\t# plt.show()\\n\\tt = np.linspace(0.0, 24.0, 96.0)\\n\\tload_profile = extra.read_slp(t, 'Profielen-Elektriciteit-2015-versie-1.00 Folder/profielen Elektriciteit 2015 versie 1.00.csv')\\n\\tslp = synthetic.create_synthetic_load(load_profile, 5.0, 5)\\n\\tplt.step(range(len(slp)), slp)\\n\\tplt.show()\",\n \"def run_tests(self):\\n\\n self.test_report = []\\n\\n #dict of unsorted lists\\n dict_of_un_lists = self.dict_un_lists_intersection_test(self.data_dict)\\n self.test_report.append(dict_of_un_lists)\\n\\n #dict of sets\\n dict_of_sets = self.build_dict_of_sets(self.data_dict)\\n self.test_report.append(self.dict_sets_intersection_test(dict_of_sets))\\n\\n #pandas - experimental and probably not the way to use pandas\\n # dict_of_pandas = self.build_dict_of_panda_series(self.data_dict)\\n # self.test_report.append(self.dicts_any_intersection_node_test(dict_of_pandas))\\n\\n # print results\\n\\n if self.verbose:\\n self.print_tests_results()\",\n \"def evaluate(clf, dataset, feature_list, features, labels, num_iter, params):\\n\\n features_train, features_test, labels_train, labels_test = \\\\\\n train_test_split(features, labels, test_size=0.3, random_state=42)\\n\\n\\n\\n precision_values = []\\n recall_values = []\\n accuracy_values = []\\n print clf\\n for i in xrange(0, num_iter):\\n #print params\\n clf = GridSearchCV(clf, params)\\n clf.fit(features_train, labels_train)\\n print '*****************************'\\n print clf.best_estimator_\\n print clf.best_params_\\n\\n clf = clf.best_estimator_\\n #test_classifier(clf, dataset, feature_list)\\n pred = clf.predict(features_test)\\n precision_values.append(precision_score(labels_test, pred))\\n recall_values.append(recall_score(labels_test, pred))\\n accuracy_values.append(accuracy_score(labels_test, pred))\\n print 'Recall score: ', mean(recall_values)\\n print 'Precision score: ', mean(precision_values)\\n print 'Accuracy score: ' , mean(accuracy_values)\",\n \"def test_compare_classifiers_predictions_distribution_vis_api(experiment_to_use):\\n experiment = experiment_to_use\\n predictions = experiment.predictions_num\\n viz_outputs = (\\\"pdf\\\", \\\"png\\\")\\n with TemporaryDirectory() as tmpvizdir:\\n for viz_output in viz_outputs:\\n vis_output_pattern_pdf = tmpvizdir + f\\\"/*.{viz_output}\\\"\\n visualize.compare_classifiers_predictions_distribution(\\n [predictions, predictions],\\n experiment.ground_truth,\\n experiment.ground_truth_metadata,\\n experiment.output_feature_name,\\n labels_limit=0,\\n model_names=[\\\"Model1\\\", \\\"Model2\\\"],\\n output_directory=tmpvizdir,\\n file_format=viz_output,\\n )\\n figure_cnt = glob.glob(vis_output_pattern_pdf)\\n assert 1 == len(figure_cnt)\",\n \"def test(self):\\n\\n scores = list()\\n for ep in range(10):\\n current_state = self.env.reset()\\n done = False\\n score = 0\\n while not done:\\n action = self.choose_action(current_state, 0)\\n obs, reward, done, _ = self.env.step(action)\\n current_state = obs\\n score += reward\\n scores.append(score)\\n print(\\\"Ep: {}, Score: {}\\\".format(ep, score))\\n scores = np.array(scores)\\n print(\\\"Eval => Std: {}, Mean: {}\\\".format(np.std(scores), np.mean(scores)))\",\n \"def perform_experiments(problem_type, output_dir, load_experiments):\\n\\n experiment_file_path = output_dir + \\\"experiments.pickle\\\"\\n\\n # data structure where to store all the problems, and the experimental results for each algorithm: solutions, final values and times\\n experiment_dict = dict()\\n\\n # if experiments should be loaded (not repeated), do it if possible\\n if load_experiments:\\n with open(experiment_file_path, \\\"rb\\\") as _file:\\n experiment_dict = pickle.load(_file)\\n\\n # perform experiments if pre-existing results were not loaded\\n if not experiment_dict:\\n\\n # given a problem type, create a set of problem instances that are solved (optimally) with manual placements (using actions available for all the algorithms)\\n problems, problem_names, manual_solutions = create_problems(problem_type)\\n\\n if problems and problem_names and manual_solutions:\\n\\n # parameters for the experimentation; note: calculating internal times and value evolution can increase the overall time of algorithms (in a slight, almost neglectible way)\\n execution_num = 10 # 1\\n process_num = 10 # 1\\n calculate_internal_times = True\\n calculate_value_evolution = True\\n\\n start_time = time.time()\\n\\n # solve each problem with each algorithm\\n for i, (problem, problem_name, solution) in enumerate(zip(problems, problem_names, manual_solutions)):\\n\\n experiment_dict[problem_name] = {\\\"problem\\\": problem, \\\"manual_solution\\\": solution, \\\"algorithms\\\": dict()}\\n\\n # solve the problem with different algorithms, executing each one multiple times to gain statistical significance\\n for (algorithm_name, algorithm) in [(\\\"Greedy\\\", greedy.solve_problem), (\\\"Reversible\\\", reversible.solve_problem), (\\\"Evolutionary\\\", evolutionary.solve_problem)]:\\n\\n solutions, values, value_evolutions, times, time_divisions = execute_algorithm(algorithm=algorithm, algorithm_name=algorithm_name, problem=problem, execution_num=execution_num, process_num=process_num, calculate_times=calculate_internal_times, calculate_fitness_stats=calculate_value_evolution)\\n experiment_dict[problem_name][\\\"algorithms\\\"][algorithm_name] = {\\\"solutions\\\": solutions, \\\"values\\\": values, \\\"value_evolutions\\\": value_evolutions, \\\"times\\\": times, \\\"time_divisions\\\": time_divisions}\\n\\n # show the total time spent doing experiments (note that significant overhead can be introduced beyond calculation time if plots are shown or saved to files; for strict time measurements, plotting should be avoided altogether)\\n print(\\\"Total experimental calculation time: {} s\\\".format(round(get_time_since(start_time) / 1000.), 3))\\n\\n if experiment_dict:\\n\\n # experiment-saving parameter\\n save_experiments = True\\n\\n # if possible, save the experiments to a binary file\\n if not load_experiments and save_experiments:\\n with open(experiment_file_path, \\\"wb\\\") as file:\\n pickle.dump(experiment_dict, file)\\n\\n # visualization/saving parameters\\n show_problem_stats = True\\n save_problem_stats = True # False\\n show_manual_solution_plots = True\\n save_manual_solution_plots = True # False\\n show_algorithm_solution_plots = True\\n save_algorithm_solution_plots = True # False\\n show_value_evolution_plots = False\\n save_value_evolution_plots = False # True\\n show_time_division_plots = False\\n save_time_division_plots = False # True\\n show_algorithm_comparison = True\\n save_algorithm_comparison = True # False\\n show_aggregated_result_tables = True\\n save_aggregated_result_tables = True # False\\n\\n # show/save the results of the experiments\\n visualize_and_save_experiments(experiment_dict, output_dir, can_plots_show_value_and_weight=problem_type == KNAPSACK_PACKING_PROBLEM_TYPE, show_problem_stats=show_problem_stats, save_problem_stats=save_problem_stats, show_manual_solution_plots=show_manual_solution_plots, save_manual_solution_plots=save_manual_solution_plots, show_algorithm_solution_plots=show_algorithm_solution_plots, save_algorithm_solution_plots=save_algorithm_solution_plots, show_value_evolution_plots=show_value_evolution_plots, save_value_evolution_plots=save_value_evolution_plots, show_time_division_plots=show_time_division_plots, save_time_division_plots=save_time_division_plots, show_algorithm_comparison=show_algorithm_comparison, save_algorithm_comparison=save_algorithm_comparison, show_aggregated_result_tables=show_aggregated_result_tables, save_aggregated_result_tables=save_aggregated_result_tables)\\n\\n else:\\n print(\\\"The experiments cannot be performed (there are no problems available).\\\")\",\n \"def benchmark(options):\\n # Prepare experiments\\n with open(options['<benchmark>']) as f:\\n benchmark_config = json.loads(f.read())\\n generate_agent_configs(benchmark_config)\\n experiments = product(benchmark_config['environments'], benchmark_config['agents'], [options])\\n\\n # Run evaluations\\n with Pool(processes=int(options['--processes'])) as pool:\\n results = pool.starmap(evaluate, experiments)\\n\\n # Clean temporary config files\\n generate_agent_configs(benchmark_config, clean=True)\\n\\n # Write evaluations summary\\n benchmark_filename = os.path.join(Evaluation.OUTPUT_FOLDER, '{}_{}.{}.json'.format(\\n BENCHMARK_FILE, datetime.datetime.now().strftime('%Y%m%d-%H%M%S'), os.getpid()))\\n with open(benchmark_filename, 'w') as f:\\n json.dump(results, f, sort_keys=True, indent=4)\\n gym.logger.info('Benchmark done. Summary written in: {}'.format(benchmark_filename))\",\n \"def test_make_results_verbose1(self):\\n\\t\\ttest = sentiment.LibraryRun(self.text3, self.lib)\\n\\t\\ttest.do_run()\\n\\t\\ttest.make_results_verbose()\\n\\t\\tobj_ut = test.results_verbose\\n\\t\\tself.assertEqual(obj_ut, [['100', 'not good', 2, -1, 0],\\n\\t\\t\\t['100', 'not very good', 4, -1, 0]])\",\n \"def main():\\n\\n tests = [500, 1000, 10000]\\n results = {\\n 'Insertion Sort': 0.0,\\n 'Shell Sort': 0.0,\\n 'Python Sort': 0.0\\n }\\n\\n for test in tests:\\n i = 0\\n\\n while i < 100:\\n test_list = gen_random_list(test)\\n results['Insertion Sort'] += insertion_sort(test_list)[0]\\n results['Shell Sort'] += shell_sort(test_list)[0]\\n results['Python Sort'] += python_sort(test_list)[0]\\n i += 1\\n\\n print(\\\"Sort results for list of size %s items:\\\" % test)\\n for key, value in results.items():\\n print(\\\"%s took %10.7f seconds to run, on average.\\\" % (key, (value/100)))\\n print(\\\"\\\\n\\\")\",\n \"def evaluation(deploymentID, start_time,\\n end_time, sites, validation, manOrLik):\\n \\n estScoreBound = 5\\n falsePositive_dep = 0\\n falseNegative_dep = 0\\n truePositive_dep = 0\\n trueNegative_dep = 0\\n\\n db_con = MySQLdb.connect(user=\\\"root\\\", db=\\\"qraat\\\")\\n for i in sites:\\n truePositive = 0\\n trueNegative = 0\\n falsePositive = 0\\n falseNegative = 0\\n \\n cur = db_con.cursor()\\n cur.execute(\\\"\\\"\\\"SELECT isPulse, band3, band10, frequency,\\n ec, tnp, edsp, fdsp, edsnr, fdsnr\\n FROM est INNER JOIN est_class%s\\n ON ID = est_class%s.estID\\n INNER JOIN estscore2\\n ON ID = estscore2.estID\\n WHERE timestamp > %s\\n AND timestamp < %s\\n AND deploymentID = %s\\n AND siteID = %s\\n AND setNum = %s;\\n \\\"\\\"\\\"%(manOrLik, manOrLik, start_time, end_time,\\n deploymentID, i, validation))\\n for row in cur.fetchall():\\n estData = {'band3':row[1], 'band10':row[2],\\n 'frequency':row[3], 'ec':row[4], 'tnp':row[5],\\n 'edsp':row[6], 'fdsp':row[7],\\n 'edsnr':row[8], 'fdsnr':row[9]}\\n \\n #Classify data\\n isPulse = modifiedBC(deploymentID, i, start_time,\\n validation, estData, manOrLik)\\n\\n #Determine whether the classification results are correct or not\\n if (row[0] == 1):\\n if (isPulse == 1):\\n truePositive += 1\\n else:\\n falseNegative += 1\\n else:\\n if (isPulse == 1):\\n falsePositive += 1\\n else:\\n trueNegative += 1\\n \\n falsePositive_dep += falsePositive\\n falseNegative_dep += falseNegative\\n truePositive_dep += truePositive\\n trueNegative_dep += trueNegative\\n return [truePositive_dep, trueNegative_dep,\\n falsePositive_dep, falseNegative_dep]\",\n \"def test_quality(approach):\\n tp, fn, fp, tn = 0, 0, 0, 0\\n with open(\\\"test.txt\\\", \\\"r\\\") as f:\\n test_data = json.load(f)\\n for i in test_data:\\n rez = approach(i[0][0], i[0][1])\\n if rez == 1 and i[1] == 1:\\n tp += 1\\n elif rez == 0 and i[1] == 1:\\n fn += 1\\n elif rez == 1 and i[1] == 0:\\n fp += 1\\n else:\\n tn += 1\\n print \\\"Precision: {}%.\\\\nRecall: {}%.\\\".format(\\n round(tp / (tp + fp), 2), round(tp / (tp + fn), 2))\",\n \"def run_simulation(n, experiments, iterations, budget, recovery_count, performance_factor, current_top,\\r\\n precomputed=True, dataset=None):\\r\\n T = 0\\r\\n k = 5\\r\\n delta = 0.1\\r\\n range_m = 1*np.arange(1, budget+1)\\r\\n scores, true_top = init(n, precomputed=precomputed, dataset=dataset)\\r\\n true_ranks = get_ranks(scores)\\r\\n\\r\\n for exp in tqdm.tqdm(range(experiments), desc=\\\"experiments\\\"):\\r\\n for itr in tqdm.tqdm(range(iterations), desc=\\\"iterations\\\"):\\r\\n t = 1\\r\\n count = 0\\r\\n A = np.arange(n)\\r\\n P = np.zeros((n, n))\\r\\n S = np.zeros(n)\\r\\n r_count = 0\\r\\n data = []\\r\\n for b in range(budget):\\r\\n m = range_m[b]\\r\\n\\r\\n est, t, A, P, S, count, r_count, data = sparse_borda(n, t, T, k, A, P, S, delta, scores, m*n, count, r_count, data)\\r\\n\\r\\n ranking, ranks, top = get_ranking(n, est)\\r\\n if(top == true_top):\\r\\n recovery_count[b][exp] += 1\\r\\n performance_factor[b][exp] += ranks[true_top]\\r\\n current_top[b][exp] += true_ranks[top]\\r\\n\\r\\n current_top /= iterations\\r\\n performance_factor /= iterations\\r\\n\\r\\n return ranking, ranks, data, scores, true_top, est, recovery_count, performance_factor, current_top\",\n \"def run():\\n\\n # Set up environment and agent\\n e = Environment() # create environment (also adds some dummy traffic)\\n a = e.create_agent(LearningAgent) # create agent\\n e.set_primary_agent(a, enforce_deadline=False) # set agent to track\\n\\n # Now simulate it\\n sim = Simulator(e, update_delay=0.0) # reduce update_delay to speed up simulation\\n sim.run(n_trials=num_of_experiments) # press Esc or close pygame window to quit\\n \\n pd.Series(a.success).to_pickle('success_' + exp_id + '.pickle')\\n a.Q_table.to_pickle('qtable_' + exp_id + '.pickle')\\n pd.Series(a.q_delta_avg).to_pickle('convergence_' + exp_id + '.pickle')\\n pd.Series(a.t_total).to_pickle('steps_' + exp_id + '.pickle')\",\n \"def run_example(num_points_to_sample=1000, verbose=True, **kwargs):\\n\\n exp = Experiment([[1, 52], [0, 6], [1, 52]]) # 2D experiment, we build a tensor product domain\\n # Bootstrap with some known or already sampled point(s)\\n exp.historical_data.append_sample_points([\\n SamplePoint([26, 2, 46], get_fitness([26, 2, 35]), 0.5), # Iterables of the form [point, f_val, f_var] are also allowed\\n ])\\n # Sample num_points_to_sample points\\n for i in range(num_points_to_sample):\\n # Use MOE to determine what is the point with highest Expected Improvement to use next\\n next_point_to_sample = map(round, gp_next_points(exp, **kwargs)[0]) # in [A, X, B] form, rounded integers\\n value_of_next_point = get_fitness(next_point_to_sample)\\n\\n if verbose:\\n if in_results(next_point_to_sample):\\n print '***', \\\"Sampled f({0:s}) = {1:.18E}\\\".format(str(next_point_to_sample), value_of_next_point), '***'\\n else:\\n print \\\"Sampled f({0:s}) = {1:.18E}\\\".format(str(next_point_to_sample), value_of_next_point)\\n\\n bank[i,0:3] = next_point_to_sample\\n bank[i,3] = value_of_next_point\\n # Add the information about the point to the experiment historical data to inform the GP\\n exp.historical_data.append_sample_points([SamplePoint(next_point_to_sample, value_of_next_point, 0.01)]) # We can add some noise\"\n]","isConversational":false},"negative_scores":{"kind":"list like","value":["0.7036995","0.6971232","0.69174814","0.68236893","0.6810289","0.67737514","0.6682389","0.66659284","0.6623596","0.6608259","0.65498793","0.6528051","0.65186554","0.64956766","0.6495176","0.6481607","0.6480974","0.6465037","0.6449332","0.64473736","0.6442062","0.6422625","0.6418362","0.6407606","0.64022803","0.63838893","0.63786054","0.63609254","0.6346425","0.63344216","0.6333071","0.6324928","0.63155687","0.631426","0.6310214","0.62999463","0.6294989","0.62916553","0.6287622","0.62825644","0.62460804","0.62452036","0.6227405","0.62246394","0.6214109","0.61828005","0.61782116","0.61750317","0.61656106","0.6156095","0.61461425","0.61387825","0.61315477","0.61270756","0.61263156","0.6125021","0.61215174","0.611316","0.6109129","0.60980564","0.60962415","0.60875136","0.6086574","0.60818535","0.60784256","0.6076886","0.6075275","0.607442","0.6067918","0.6067572","0.60671026","0.60621977","0.6061239","0.6058888","0.60575265","0.6056491","0.60485274","0.6041622","0.602987","0.60279214","0.6020767","0.6017449","0.60094446","0.60059977","0.6005152","0.60021627","0.5996669","0.5988561","0.59819496","0.59807634","0.597913","0.597427","0.5972096","0.5969686","0.59655535","0.59651726","0.59649754","0.5962564","0.59618664","0.59598774"],"string":"[\n \"0.7036995\",\n \"0.6971232\",\n \"0.69174814\",\n \"0.68236893\",\n \"0.6810289\",\n \"0.67737514\",\n \"0.6682389\",\n \"0.66659284\",\n \"0.6623596\",\n \"0.6608259\",\n \"0.65498793\",\n \"0.6528051\",\n \"0.65186554\",\n \"0.64956766\",\n \"0.6495176\",\n \"0.6481607\",\n \"0.6480974\",\n \"0.6465037\",\n \"0.6449332\",\n \"0.64473736\",\n \"0.6442062\",\n \"0.6422625\",\n \"0.6418362\",\n \"0.6407606\",\n \"0.64022803\",\n \"0.63838893\",\n \"0.63786054\",\n \"0.63609254\",\n \"0.6346425\",\n \"0.63344216\",\n \"0.6333071\",\n \"0.6324928\",\n \"0.63155687\",\n \"0.631426\",\n \"0.6310214\",\n \"0.62999463\",\n \"0.6294989\",\n \"0.62916553\",\n \"0.6287622\",\n \"0.62825644\",\n \"0.62460804\",\n \"0.62452036\",\n \"0.6227405\",\n \"0.62246394\",\n \"0.6214109\",\n \"0.61828005\",\n \"0.61782116\",\n \"0.61750317\",\n \"0.61656106\",\n \"0.6156095\",\n \"0.61461425\",\n \"0.61387825\",\n \"0.61315477\",\n \"0.61270756\",\n \"0.61263156\",\n \"0.6125021\",\n \"0.61215174\",\n \"0.611316\",\n \"0.6109129\",\n \"0.60980564\",\n \"0.60962415\",\n \"0.60875136\",\n \"0.6086574\",\n \"0.60818535\",\n \"0.60784256\",\n \"0.6076886\",\n \"0.6075275\",\n \"0.607442\",\n \"0.6067918\",\n \"0.6067572\",\n \"0.60671026\",\n \"0.60621977\",\n \"0.6061239\",\n \"0.6058888\",\n \"0.60575265\",\n \"0.6056491\",\n \"0.60485274\",\n \"0.6041622\",\n \"0.602987\",\n \"0.60279214\",\n \"0.6020767\",\n \"0.6017449\",\n \"0.60094446\",\n \"0.60059977\",\n \"0.6005152\",\n \"0.60021627\",\n \"0.5996669\",\n \"0.5988561\",\n \"0.59819496\",\n \"0.59807634\",\n \"0.597913\",\n \"0.597427\",\n \"0.5972096\",\n \"0.5969686\",\n \"0.59655535\",\n \"0.59651726\",\n \"0.59649754\",\n \"0.5962564\",\n \"0.59618664\",\n \"0.59598774\"\n]","isConversational":false},"document_score":{"kind":"string","value":"0.66045517"},"document_rank":{"kind":"string","value":"10"}}},{"rowIdx":93293,"cells":{"query":{"kind":"string","value":"runs an image_matching_module that calculates the precision and recall of the comparison algorithm's results"},"document":{"kind":"string","value":"def run_single_simple_experiment(self, path_to_csv_files, algorithm, thresholds):\n\n splitted_path = os.path.abspath(__file__).split(\"/\")\n if not os.path.isdir(\"/\".join(splitted_path[:-1]) + \"/results\"):\n os.mkdir(\"/\".join(splitted_path[:-1]) + \"/results\")\n\n experiment_dict, list_dict = self.create_experiment_dict(algorithm.get_algorithm_name() + \"_actual\", thresholds)\n\n csv_paths_list = self.reader.read_all_csv_paths_from_path(path_to_csv_files)\n file_number = 1\n for csv_path in csv_paths_list:\n if 'total' in csv_path:\n continue\n df = pd.read_csv(csv_path)\n start_file_time = time.time()\n print(\"started file number:\", file_number)\n for index, row in df.iterrows():\n algorithm_column_name = algorithm.get_algorithm_name() + \"_actual\"\n actual_score = row['actual_score']\n for threshold in thresholds:\n algorithm_score = row[algorithm_column_name]\n if actual_score == 1:\n if algorithm_score >= threshold:\n experiment_dict[str(threshold)][algorithm_column_name][list_dict['TP']] += 1\n else:\n experiment_dict[str(threshold)][algorithm_column_name][list_dict['FN']] += 1\n else:\n if algorithm_score < threshold:\n experiment_dict[str(threshold)][algorithm_column_name][list_dict['TN']] += 1\n else:\n experiment_dict[str(threshold)][algorithm_column_name][list_dict['FP']] += 1\n finish_file_time = time.time()\n print(\"finished file number:\", file_number)\n print(\"file number:\", file_number, \"took:\", (finish_file_time - start_file_time) / 60, \"minutes\")\n file_number += 1\n\n # calculate the precision and recall for each threshold and algorithm\n experiment_dict = self.update_precision_and_recall(experiment_dict, list_dict,\n algorithm.get_algorithm_name() + \"_actual\", thresholds)\n\n # create a dictionary that can be ingested into a dataframe\n experiment_dict_for_csv = self.create_simple_experiment_dict_for_csv(experiment_dict, list_dict, algorithm,\n thresholds)\n\n # create a dataframe from the results\n experiment_dict_df = pd.DataFrame(experiment_dict_for_csv)\n\n # write the results into a csv file\n experiment_csv_file_name = os.path.dirname(\n \"/\".join(splitted_path)) + \"/results/precision_recall_results_\" + algorithm.get_algorithm_name() + \".csv\"\n experiment_dict_df.to_csv(experiment_csv_file_name, index=False, header=True)"},"metadata":{"kind":"string","value":"{\n \"objective\": {\n \"self\": [],\n \"paired\": [],\n \"triplet\": [\n [\n \"query\",\n \"document\",\n \"negatives\"\n ]\n ]\n }\n}"},"negatives":{"kind":"list like","value":["def patternMatchingTest(matchingAlgorithm,\n observations= defaultCOMPREFdump , \n modelsFolder= defaultWRFdumpsFolder,\n #key1='0828',\n key1='',\n key2='',\n #key3='',\n verbose=True,\n makeSmallImages=True,\n smallDpi=40,\n frameDpi = 600 , # added 2013-09-23\n loadFormerAnalysisResult=False, #added 2013-09-23\n #summary_folder=summary_folder, # needed if loadFormerAnalysisResult is True\n summary_folder=summary_folder+timeString + '/', # 2013-09-27\n matching_algorithm_name=\"\", # needed if loadFormerAnalysisResult is True\n panel_cols = 5,\n panel_rows = 5,\n useCV2 = False, # switch off opencv since it doesn't work yet\n useCV = False,\n averageScoreOrdering=True, # take an average of the ordering to get a total score for the entire model over the time series\n #kwargs = {}, # keyword arguments for the matchinging algorithm, just in case\n **kwargs\n ):\n if key1 != '' or key2 != '':\n summary_folder += '_'.join([key1, key2])\n if not averageScoreOrdering:\n summary_folder += '_nonaver'\n summary_folder += '%dx%d/' % (panel_rows, panel_cols)\n\n t0 = time.time()\n if loadFormerAnalysisResult:\n mv_folder = summary_folder + matching_algorithm_name + '/'\n result = pickle.load(open(mv_folder + 'result.pydump', 'r'))\n\n \"\"\"\n result= {'best matches' : bestMatchAtEachTime, \n 'best overall match' : bestMatchOverall, \n 'all scores' : scores,\n 'matching algorithm' : matching_algorithm_name,\n 'dataset1[obs]' : observations,\n 'dataset2[models]' : modelsFolder,\n 'time spent' : time.time() - t0,\n 'mv folder' : mv_folder, #added 2013-09-23\n 'ordering' : ordering, #added 2013-09-23\n }\n \"\"\"\n bestMatchAtEachTime = result['best matches'] \n bestMatchOverall = result['best overall match']\n scores = result['all scores']\n matching_algorithm_name = result['matching algorithm']\n observations = result['dataset1[obs]']\n modelsFolder = result['dataset2[models]']\n #mv_folder = result['mv folder']\n ordering = result['ordering']\n\n # THE FOLLOWING 30 LINES ARE COPIED FROM \"ELSE...\" BELOW\n ########### ******\n ##\n #\n ds1 = pickle.load(open(observations,'r'))\n ds1.setFloor(0) # added 2013-09-27\n ds1.list = [v for v in ds1.list if (key1 in v.dataTime or key1 in v.name) \\\n and (key2 in v.dataTime or key2 in v.name)]\n ds1.list.sort(key=lambda v:v.dataTime)\n ds1.setVmin(-40) #line added 2013-09-23\n # debug\n #print ds1.list\n print 'loading list with keywords [ %s %s ]' %(key1, key2) \n ds1.load(verbose=verbose)\n # end debug\n modelsList = [v for v in os.listdir(modelsFolder) if '.pydump' in v]\n modelsList.sort(key = lambda v: int(re.findall(r'\\d+',v)[0]))\n print \"modelsList\", modelsList\n scores = {} # scores = {model number: [('dataTime', score), ('dataTime', score),...], ...}\n print 'loading ds2 (model data)'\n wrfs = [] # line added 2013-09-23\n for path in modelsList:\n modelIndex = int(re.findall(r'\\d+', path)[0])\n ds2 = pickle.load(open(modelsFolder+path, 'r'))\n ds2.setFloor(0) # added 2013-09-27\n # debug\n #ds2.list = [v for v in ds2.list if key1 in v.dataTime] # line redundant, 2013-09-23\n ds1.list, ds2.list = ds1.intersect(ds2) #line added 2013-09-23 \n ds2.setVmin(-40) #line added 2013-09-23\n wrfs.append(ds2) # line added 2013-09-23\n # end debug\n\n ####################\n\n\n else:\n # THE FOLLOWING 30 LINES ARE COPIED TO \"IF...\" ABOVE\n ########### ******\n ##\n #\n ds1 = pickle.load(open(observations,'r'))\n ds1.list = [v for v in ds1.list if (key1 in v.dataTime or key1 in v.name) \\\n and (key2 in v.dataTime or key2 in v.name)]\n ds1.list.sort(key=lambda v:v.dataTime)\n ds1.setVmin(-40) #line added 2013-09-23\n # debug\n #print ds1.list\n print 'loading list with keywords [ %s %s ]' %(key1, key2) \n ds1.load(verbose=verbose)\n ds1.setFloor(0) # added 2013-09-27\n # end debug\n modelsList = [v for v in os.listdir(modelsFolder) if '.pydump' in v]\n modelsList.sort(key = lambda v: int(re.findall(r'\\d+',v)[0]))\n print \"modelsList\", modelsList\n scores = {} # scores = {model number: [('dataTime', score), ('dataTime', score),...], ...}\n print 'loading ds2 (model data)'\n wrfs = [] # line added 2013-09-23\n ds1.commonMaskSet=False # tempory attribute added 2013-09-27\n for path in modelsList:\n modelIndex = int(re.findall(r'\\d+', path)[0])\n ds2 = pickle.load(open(modelsFolder+path, 'r'))\n # debug\n #ds2.list = [v for v in ds2.list if key1 in v.dataTime] # line redundant, 2013-09-23\n ds1.list, ds2.list = ds1.intersect(ds2) #line added 2013-09-23 \n ds2.setFloor(0) # added 2013-09-27\n ds2[0].matrix.mask += ds1[0].matrix.mask # added 2013-09-27\n ds2.setCommonMask() # added 2013-09-27\n if not ds1.commonMaskSet: # added 2013-09-27\n ds1[0].matrix.mask +=ds2[0].matrix.mask # added 2013-09-27\n ds1.setCommonMask() # added 2013-09-27\n ds2.setVmin(-40) #line added 2013-09-23\n wrfs.append(ds2) # line added 2013-09-23\n # end debug\n\n #######################################################################################################################\n # key lines\n print 'computing scores between %s and %s ...' %(ds1.name, ds2.name)\n s, matching_algorithm_name = matchingAlgorithm(ds1, ds2, **kwargs) # list of the form [('20120612.0000', 0.9),...] \n scores[modelIndex] = dict(s) # { model index: {'dataTime': score,...}, ...}\n # debug\n if verbose:\n print scores\n print '..................................................'\n #\n ########################################################################################################################\n\n print 'sorting the results...'\n #print 'scores', '\\n', scores #debug\n # bestMatchAtEachTime\n T1 = [v.dataTime for v in ds1]\n T2 = [v.dataTime for v in ds2] #any ds2 will do - they are the same, in this case we pick the last one \n T = set(T1).intersection(T2)\n dataTimeList = sorted(list(T))\n\n #print 'dataTimeList:', dataTimeList #debug\n highestScores = []\n ordering = [] #added 2013-09-23\n\n if averageScoreOrdering: # take the average scores before ordering to make it uniform\n ave_scores = [ (v, np.mean( [scores[v].values() ] ) ) for v in scores.keys()]\n ave_scores = dict(ave_scores)\n print '..taking averages...............................'\n print ave_scores\n for u in scores.keys():\n for t in scores[u].keys():\n scores[u][t] = ave_scores[u]\n print '\\n'.join([str(scores[v]) for v in scores.keys()])\n\n for t in dataTimeList:\n print t\n highestScoreAt_t = max([scores[u][t] for u in scores.keys()]) # u = model index (integer 1-20), t = dataTime (string)\n new_entry = (t, highestScoreAt_t, [u for u in scores.keys() if scores[u][t]==highestScoreAt_t])\n print new_entry\n highestScores.append(new_entry)\n #added 2013-09-23\n ordering.append([0]+sorted(range(1, len(modelsList)+1), key=lambda v: scores[v][t], reverse=True))\n\n\n\n\n # bestMatchOverall\n print scores # debug\n averageScores = [(Model, np.mean(scores[Model].values())) for Model in scores.keys()] # list of [(Model, mean score),...]\n highestScore = max([v[1] for v in averageScores])\n bestMatch = [u[0] for u in averageScores if u[1]==highestScore] \n # debug\n print 'average scores , highest average, best match', averageScores, highestScore, bestMatch\n #xx= raw_input('press enter') \n\n\n bestMatchAtEachTime = highestScores\n bestMatchOverall = (highestScore, bestMatch)\n\n # the following lines do not work well since oftentimes we have multiple pointers (names) to a single object/function\n # and so they are replaced by returning the function name in the function itself\n #allvars = vars()\n #matching_algorithm_name = [v for v in allvars if allvars[v]==matchingAlgorithm][0]\n #mv_folder = summary_folder + matching_algorithm_name + '/'\n mv_folder = summary_folder + matching_algorithm_name + '/' # added 2013-09-27\n\n result= {'best matches' : bestMatchAtEachTime, \n 'best overall match' : bestMatchOverall, \n 'all scores' : scores,\n 'matching algorithm' : matching_algorithm_name,\n 'dataset1[obs]' : observations,\n 'dataset2[models]' : modelsFolder,\n 'time spent' : time.time() - t0,\n 'mv folder' : mv_folder, #added 2013-09-23\n 'ordering' : ordering, #added 2013-09-23\n }\n\n try:\n os.makedirs(mv_folder)\n print 'folder', mv_folder, \"created!\"\n except:\n print 'folder', mv_folder, 'exists!'\n pickle.dump(result, open(mv_folder+'result.pydump','w'))\n\n ############################################################################\n # to make the videos/collected images\n # calling armor.dataStreamTools.makeVideo.makeVideo()\n # 2013-09-23\n\n from . import makeVideo as mv\n \n try:\n os.makedirs(mv_folder)\n print 'folder', mv_folder, \"created!\"\n except:\n print 'folder', mv_folder, 'exists!'\n # make images\n if makeSmallImages:\n print 'making small images'\n for ds in [ds1] + wrfs:\n #ds.imageFolder= mv_folder + ds.name + '/'\n ds.imageFolder= summary_root + ds.name + '/' #added 2013-09-27 to replace the above line - too tired waiting the plots\n ds.saveImages(drawCoast=True, flipud=True, dpi=smallDpi)\n print ds.imageFolder\n else:\n # just set the paths\n for ds in [ds1]+wrfs:\n ds.imageFolder = summary_root + ds.name +'/' # /home/k/ARMOR/data/KONG-REY/summary/WRF01.DBZ/\n for dbzPattern in ds.list:\n dbzPattern.imagePath = ds.imageFolder + ds.name + dbzPattern.dataTime + '.png'\n print dbzPattern.imagePath\n #ds.setImagePaths() # doesn't for for old pickled objects???\n\n\n # make panels\n\n mv.makeVideo(DSS = [ds1]+ wrfs, # [ds0, ds1, ds2, ds3, ds4, ...], a list of armor.pattern.DBZstream objects\n panel_cols = panel_cols, # number of colums in the panel\n panel_rows = panel_rows, # no need to be filled\n #fourcc = cv.CV_FOURCC('F', 'L', 'V', '1'),\n #fps = defaultFps,\n extension= '.avi',\n #fourcc = cv.CV_FOURCC('P', 'I', 'M', '1'),\n outputFileName =\"\",\n outputFolder= mv_folder,\n sandbox = mv_folder ,\n saveFrames = True, # saving the frames as images\n frameDpi = frameDpi,\n useCV2 = useCV2,\n useCV = useCV,\n ordering = ordering, # ordering of the models\n )\n ############################################################################\n # dumping the results\n try:\n os.makedirs(summary_folder+ 'pydumps/')\n except:\n print 'Folder exists! -- ', summary_folder+ 'pydumps/' \n\n pickle.dump(result, open(summary_folder+ 'pydumps/results_' + matching_algorithm_name+ '+' +\\\n key1 + '+' + key2 + str(int(time.time())) + '.pydump', 'w'))\n\n \n\n print \"time spent:\", time.time()-t0\n return result","def painting_matching_ml(imgs, db_imgs, method_list, text_masks, author_text, gt_text, metrics, weights, splits=30, max_rank=10):\n descriptor_extractors = [get_descriptor_extractor(method_name) for method_name in method_list]\n tmp_img_format = []\n tmp_mask_format = []\n tmp_text_format = []\n for i in range(len(imgs)):\n for j in range(len(imgs[i])):\n tmp_img_format.append(imgs[i][j])\n tmp_mask_format.append(text_masks[i][j])\n tmp_text_format.append(author_text[i][j])\n\n #db_imgs = [img[0] for img in db_imgs]\n db_img_splits = [i*len(db_imgs)//splits for i in range(splits-1)]\n \n scores = []\n query_descriptors = extract_descriptors(tmp_img_format, descriptor_extractors, method_list, tmp_text_format, tmp_mask_format) \n #np.array([extract_descriptors(img, matching_methods, mask) for img, mask in zip(tmp_img_format, tmp_mask_format)])\n print(\"Starting db extraction + matching\")\n for split in tqdm(range(splits-2)):\n db_descriptors = extract_descriptors(db_imgs[db_img_splits[split]:db_img_splits[split+1]], descriptor_extractors, method_list, gt_text[db_img_splits[split]:db_img_splits[split+1]], None) #np.array([mrhm(db_img) for db_img in db_imgs[db_img_splits[split]:db_img_splits[split+1]]])\n scores.append(compare_descriptors(query_descriptors, db_descriptors, metrics, method_list, weights))\n # compare_descriptors(query_descriptors, db_descriptors, descriptor_comp_methods, descriptor_names, weights)\n db_descriptors = extract_descriptors(db_imgs[db_img_splits[-1]:], descriptor_extractors, method_list, gt_text[db_img_splits[-1]:], None)\n scores.append(compare_descriptors(query_descriptors, db_descriptors, metrics, method_list, weights))\n \n # concatenate all the results\n scores = np.concatenate(scores, 1)\n \n top_k_matches = np.argpartition(scores, list(range(max_rank)))[:, :max_rank]\n return top_k_matches","def compare(model, input1, input2): \n patch_image_time_start = time.time()\n input1_patches = get_patches_non_overlap(input1, 48, 48)\n input2_patches = get_patches_non_overlap(input2, 48, 48)\n patch_image_time_end = time.time()\n compare_image_time_start = time.time()\n pred = model.predict([input1_patches, input2_patches])\n compare_image_time_end = time.time()\n\n result[\"patch_retrieval_time\"].append(patch_image_time_end - patch_image_time_start)\n result[\"image_comparison_time\"].append(compare_image_time_end - compare_image_time_start)\n\n return np.sum(pred)","def ransac(matches, kp1, kp2, s=4, threshold=3, maxIterations=2000, returnMatches=False, inlierRatio=0.05, ransacRatio=0.6):\n\n sizes_kp1 = [kp1[dt[0].queryIdx].size for dt in matches]\n sizes_kp2 = [kp1[dt[0].trainIdx].size for dt in matches]\n tup_matches_kp1 = [kp1[dt[0].queryIdx].pt for dt in matches]\n tup_matches_kp2 = [kp2[dt[0].trainIdx].pt for dt in matches]\n matches_kp1 = np.array([[h for h in kp] + [1] for kp in tup_matches_kp1])\n matches_kp2 = np.array([[h for h in kp] + [1] for kp in tup_matches_kp2])\n\n cnt_matches = len(matches)\n\n max_matches = []\n max_p1, max_p2 = [], []\n max_p1_sizes, max_p2_sizes = [], []\n max_total = 0\n\n for iter in range(maxIterations):\n # Find Homography based on random sample\n data = random.sample(matches, s)\n data_p1 = np.array([matches_kp1[dt[0].queryIdx] for dt in data])\n data_p2 = np.array([matches_kp2[dt[0].trainIdx] for dt in data])\n homography = homomat(data_p1[:, :2], data_p2[:, :2])\n\n # Find P1 projection from the homography matrix\n projected_p2 = np.dot(homography, matches_kp1.transpose())\n projected_p2 = projected_p2[0:3] / projected_p2[2] # make sure w' is 1\n projected_p2 = projected_p2.transpose()\n\n # Initialize Current Matches\n current_matches = []\n current_p1, current_p2 = [], []\n current_p1_sizes, current_p2_sizes = [], []\n current_total = 0\n\n # Check for inliers and outliers for each matches\n for i, (match) in enumerate(matches):\n # normalize the error\n error = np.linalg.norm(matches_kp2[i] - projected_p2[i])\n\n # Check for inliers\n if error < threshold:\n current_matches.append([cv.DMatch(current_total, current_total, match[0].distance)])\n current_p1.append(matches_kp1[i][0:2])\n current_p2.append(matches_kp2[i][0:2])\n current_p1_sizes.append(sizes_kp1[i])\n current_p2_sizes.append(sizes_kp2[i])\n current_total += 1\n\n # If\n if current_total > max_total and current_total >= np.round(inlierRatio*cnt_matches):\n max_matches = current_matches\n max_p1 = current_p1\n max_p2 = current_p2\n max_p1_sizes = current_p1_sizes\n max_p2_sizes = current_p2_sizes\n max_total = current_total\n\n # # we are done in case we have enough inliers\n if current_total > cnt_matches * ransacRatio:\n break\n\n\n # Re-evaluate the Homography based on the best inliers\n max_homography = homomat(np.array(max_p1), np.array(max_p2))\n\n if returnMatches:\n max_kp1 = [cv.KeyPoint(d[0], d[1], max_p1_sizes[i]) for i, d in enumerate(max_p1)]\n max_kp2 = [cv.KeyPoint(d[0], d[1], max_p2_sizes[i]) for i, d in enumerate(max_p2)]\n return max_homography, max_matches, max_kp1, max_kp2\n\n return max_homography","def evaluate_matrices(self, precision_matrix, recall_matrix):\n\t\tground_truth_count = precision_matrix.shape[0]\n\t\tdetection_count = precision_matrix.shape[1]\n\t\tground_truth_sets_precision = defaultdict(set) # number of ground truth items that match a particular detection in the precision matrix\n\t\tdetection_sets_precision = defaultdict(set) # number of detection items that match a particular ground truth in the precision matrix\n\t\tground_truth_sets_recall = defaultdict(set) # number of ground truth items that match a particular detection in the recall matrix\n\t\tdetection_sets_recall = defaultdict(set) # number of detection items that match a particular ground truth in the recall matrix\n\n\t\tfor gt_index in range(ground_truth_count):\n\t\t\tfor det_index in range(detection_count):\n\t\t\t\tif precision_matrix[gt_index, det_index] >= self.precision_threshold:\n\t\t\t\t\tground_truth_sets_precision[det_index].add(gt_index)\n\t\t\t\t\tdetection_sets_precision[gt_index].add(det_index)\n\t\t\t\tif recall_matrix[gt_index, det_index] >= self.recall_threshold:\n\t\t\t\t\tground_truth_sets_recall[det_index].add(gt_index)\n\t\t\t\t\tdetection_sets_recall[gt_index].add(det_index)\n\n\t\tmatch_ground_truth = 0. # sum of MatchG\n\t\tmatch_detection = 0. # sum of MatchD\n\n\t\tone_to_one_precision = set()\n\t\tfor gt_index in detection_sets_precision:\n\t\t\tmatching_detections_precision = detection_sets_precision[gt_index]\n\t\t\tif len(matching_detections_precision) == 1:\n\t\t\t\t(detection_precision, ) = matching_detections_precision\n\t\t\t\tif len(ground_truth_sets_precision[detection_precision]) == 1:\n\t\t\t\t\tone_to_one_precision.add((gt_index, detection_precision))\n\t\t\telse:\n\t\t\t\t# one-to-many (one ground truth to many detections)\n\t\t\t\tgt_sum = 0.\n\t\t\t\tfor detection_precision in matching_detections_precision:\n\t\t\t\t\tgt_sum += recall_matrix[gt_index, detection_precision]\n\t\t\t\tif gt_sum >= self.recall_threshold:\n\t\t\t\t\t#print(\"1:N ~ GT {} : DT {}\".format(gt_index,matching_detections_precision))\n\t\t\t\t\tmatch_ground_truth += self.scatter_punishment(matching_detections_precision)\n\t\t\t\t\tmatch_detection += len(matching_detections_precision) * self.scatter_punishment(matching_detections_precision)\n\n\t\tone_to_one_recall = set()\n\t\tfor det_index in ground_truth_sets_recall:\n\t\t\tmatching_ground_truths_recall = ground_truth_sets_recall[det_index]\n\t\t\tif len(matching_ground_truths_recall) == 1:\n\t\t\t\t(ground_truth_recall, ) = matching_ground_truths_recall\n\t\t\t\tif len(detection_sets_recall[ground_truth_recall]) == 1:\n\t\t\t\t\tone_to_one_recall.add((ground_truth_recall, det_index))\n\t\t\telse:\n\t\t\t\t# many-to-one (many ground truths covered by one detection)\n\t\t\t\tdet_sum = 0\n\t\t\t\tfor ground_truth_recall in matching_ground_truths_recall:\n\t\t\t\t\tdet_sum += precision_matrix[ground_truth_recall, det_index]\n\t\t\t\tif det_sum >= self.precision_threshold:\n\t\t\t\t\t#print(\"N:1 ~ DT {} : GT {}\".format(det_index,matching_ground_truths_recall))\n\t\t\t\t\tmatch_detection += self.scatter_punishment(matching_ground_truths_recall)\n\t\t\t\t\tmatch_ground_truth += len(matching_ground_truths_recall) * self.scatter_punishment(matching_ground_truths_recall)\n\n\t\tone_to_one_matches = one_to_one_precision & one_to_one_recall\n\t\tmatch_ground_truth += len(one_to_one_matches)\n\t\tmatch_detection += len(one_to_one_matches)\n\n\t\trecall = match_ground_truth / float(ground_truth_count)\n\t\tprecision = match_detection / float(detection_count)\n\t\treturn (precision, recall, (match_detection, float(detection_count)), (match_ground_truth, float(ground_truth_count)))","def findMatchesBetweenImages(image_1, image_2):\n # matches - type: list of cv2.DMath\n matches = None\n # image_1_kp - type: list of cv2.KeyPoint items.\n image_1_kp = None\n # image_1_desc - type: numpy.ndarray of numpy.uint8 values.\n image_1_desc = None\n # image_2_kp - type: list of cv2.KeyPoint items.\n image_2_kp = None\n # image_2_desc - type: numpy.ndarray of numpy.uint8 values.\n image_2_desc = None\n # WRITE YOUR CODE HERE.\n\n sift = cv2.ORB_create()\n image_1_kp, image_1_desc = sift.detectAndCompute(image_1, None)\n image_2_kp, image_2_desc = sift.detectAndCompute(image_2, None)\n\n # bf = cv2.BFMatcher(cv2.NORM_HAMMING, crossCheck=True)\n # matches = bf.match(image_1_desc, image_2_desc)\n # matches = sorted(matches, key = lambda x:x.distance)\n # matches = matches[:10]\n\n bf = cv2.BFMatcher()\n matches = bf.knnMatch(image_1_desc, image_2_desc, k=2)\n\n # Apply ratio test\n good = []\n for m, n in matches:\n print(m.distance, n.distance, m.distance < .75*n.distance)\n if m.distance < (0.75 * n.distance):\n good.append([m])\n\n # We coded the return statement for you. You are free to modify it -- just\n # make sure the tests pass.\n print(len(good), good)\n return image_1_kp, image_2_kp, matches\n # END OF FUNCTION.","def match(image1,image2,threshold,useRansac=False,t_orientation=30,t_scale=0.5):\r\n im1, keypoints1, descriptors1 = ReadKeys(image1)\r\n im2, keypoints2, descriptors2 = ReadKeys(image2)\r\n #\r\n # REPLACE THIS CODE WITH YOUR SOLUTION (ASSIGNMENT 5, QUESTION 3)\r\n #\r\n # Generate five random matches (for testing purposes)\r\n # matched_pairs = []\r\n # num = 5\r\n # for i in range(num):\r\n # matched_pairs.append([keypoints1[i],keypoints2[i]])\r\n # return DisplayMatches(im1, im2, matched_pairs)\r\n\r\n # END OF SECTION OF CODE TO REPLACE\r\n #\r\n\r\n #q3\r\n matched_pairs = []\r\n between_angles = np.arccos(np.dot(descriptors1, np.transpose(descriptors2)))\r\n for i, row in enumerate(between_angles):\r\n \tratio = sorted(row)[0] / sorted(row)[1]\r\n \tif ratio <= threshold:\r\n\t \tmatched_pairs.append([keypoints1[i], keypoints2[np.where(row == sorted(row)[0])[0][0]]])\r\n # print(matched_pairs)\r\n if useRansac is False:\r\n return DisplayMatches(im1, im2, matched_pairs)\r\n\t# \r\n\r\n #q4\r\n repetition = 10\r\n subsets = [[]] * repetition\r\n for i in range(repetition):\r\n r = random.randint(0, len(matched_pairs))\r\n for match in matched_pairs:\r\n ds1, ds2 = matched_pairs[r][1][2]/matched_pairs[r][0][2], match[1][2]/match[0][2]\r\n do1, do2 = (matched_pairs[r][1][3]-matched_pairs[r][0][3]), (match[1][3]-match[0][3])\r\n if abs(ds2 - ds1) <= t_scale * ds1 and abs(do2 - do1) % (2 * math.pi) <= t_orientation:\r\n subsets[i].append(match)\r\n\r\n max_i, max_len = 0, subsets[0]\r\n for i in range(10):\r\n l = len(subsets[i])\r\n if l > max_len:\r\n max_len = l\r\n max_i = i\r\n\r\n im3 = DisplayMatches(im1, im2, subsets[max_i])\r\n return im3","def _match_keypoints(self, query_image: Image) -> typing.Tuple[typing.List[int], typing.List[typing.Any]]:\n query_image.get_keypoints_and_descriptors()\n\n # matches = self.matcher.match(query_image.descriptor)\n matches = self.matcher.knnMatch(query_image.descriptor, k=2)\n good = []\n for m_n in matches:\n if len(m_n) == 1:\n good.append(m_n[0])\n continue\n elif len(m_n) != 2:\n continue\n (m, n) = m_n\n if m.distance < 0.7 * n.distance:\n good.append(m)\n\n images_scores = [0] * len(self._candidate_images)\n images_matches = [None] * len(self._candidate_images)\n for image_index, image in enumerate(self._candidate_images):\n matches_scores = []\n matches = []\n for i, match in enumerate(good):\n if match.imgIdx != image_index:\n continue\n matches.append(match)\n matches_scores.append((256 - match.distance) / 256)\n\n match_cnt = len(matches_scores)\n if match_cnt <= 0:\n continue\n\n images_scores[image_index] = (\n 0.5 + ((math.tanh(match_cnt / 3 - 1)) / 2)) * (sum(matches_scores) / match_cnt)\n\n images_matches[image_index] = matches\n\n return images_scores, images_matches","def compute_precision_recall_bbox(per_image_detections, per_image_gts, num_gt_classes,\n matching_iou_threshold=0.5):\n per_image_eval = per_image_evaluation.PerImageEvaluation(\n num_groundtruth_classes=num_gt_classes,\n matching_iou_threshold=matching_iou_threshold,\n nms_iou_threshold=1.0,\n nms_max_output_boxes=10000\n )\n\n print('Running per-object analysis...')\n\n detection_tp_fp = defaultdict(list) # key is the category; in each list, 1 is tp, 0 is fp\n detection_scores = defaultdict(list)\n num_total_gt = defaultdict(int)\n\n for image_id, dets in tqdm(per_image_detections.items()):\n\n detected_boxes = np.array(dets['boxes'], dtype=np.float32)\n detected_scores = np.array(dets['scores'], dtype=np.float32)\n # labels input to compute_object_detection_metrics() needs to start at 0, not 1\n detected_labels = np.array(dets['labels'], dtype=np.int) - 1 # start at 0\n # num_detections = len(dets['boxes'])\n\n gts = per_image_gts[image_id]\n gt_boxes = np.array(gts['gt_boxes'], dtype=np.float32)\n gt_labels = np.array(gts['gt_labels'], dtype=np.int) - 1 # start at 0\n num_gts = len(gts['gt_boxes'])\n\n groundtruth_is_difficult_list = np.zeros(num_gts, dtype=bool) # place holders - we don't have these\n groundtruth_is_group_of_list = np.zeros(num_gts, dtype=bool)\n\n # to prevent 'Invalid dimensions for box data.' error\n if num_gts == 0:\n # this box will not match any detections\n gt_boxes = np.array([[0, 0, 0, 0]], dtype=np.float32)\n\n scores, tp_fp_labels, is_class_correctly_detected_in_image = (\n per_image_eval.compute_object_detection_metrics(\n detected_boxes=detected_boxes,\n detected_scores=detected_scores,\n detected_class_labels=detected_labels,\n groundtruth_boxes=gt_boxes,\n groundtruth_class_labels=gt_labels,\n groundtruth_is_difficult_list=groundtruth_is_difficult_list,\n groundtruth_is_group_of_list=groundtruth_is_group_of_list\n )\n )\n\n for i, tp_fp_labels_cat in enumerate(tp_fp_labels):\n assert sum(tp_fp_labels_cat) <= sum(gt_labels == i) # true positives < gt of that category\n cat = i + 1 # categories start at 1\n detection_tp_fp[cat].append(tp_fp_labels_cat)\n detection_scores[cat].append(scores[i])\n num_total_gt[cat] += sum(gt_labels == i) # gt_labels start at 0\n\n all_scores = []\n all_tp_fp = []\n\n print('Computing precision recall for each category...')\n per_cat_metrics = {}\n for i in range(num_gt_classes):\n cat = i + 1\n scores_cat = np.concatenate(detection_scores[cat])\n tp_fp_cat = np.concatenate(detection_tp_fp[cat]).astype(np.bool)\n all_scores.append(scores_cat)\n all_tp_fp.append(tp_fp_cat)\n\n precision, recall = metrics.compute_precision_recall(\n scores_cat, tp_fp_cat, num_total_gt[cat]\n )\n average_precision = metrics.compute_average_precision(precision, recall)\n\n per_cat_metrics[cat] = {\n 'category': cat,\n 'precision': precision,\n 'recall': recall,\n 'average_precision': average_precision,\n 'scores': scores_cat,\n 'tp_fp': tp_fp_cat,\n 'num_gt': num_total_gt[cat]\n }\n print('Number of ground truth in category {} is {}'.format(cat, num_total_gt[cat]))\n\n # compute one-class precision/recall/average precision (if every box is just of an object class)\n all_scores = np.concatenate(all_scores)\n all_tp_fp = np.concatenate(all_tp_fp)\n overall_gt_count = sum(num_total_gt.values())\n\n one_class_prec, one_class_recall = metrics.compute_precision_recall(\n all_scores, all_tp_fp, overall_gt_count\n )\n one_class_average_precision = metrics.compute_average_precision(one_class_prec, one_class_recall)\n\n per_cat_metrics['one_class'] = {\n 'category': 'one_class',\n 'precision': one_class_prec,\n 'recall': one_class_recall,\n 'average_precision': one_class_average_precision,\n 'scores': all_scores,\n 'tp_fp': all_tp_fp,\n 'num_gt': overall_gt_count\n }\n\n return per_cat_metrics","def _compute_prediction(self, k, img, kp, des):\n ###taken from example code:\n\n #find matches\n matches = self.matcher.knnMatch(des,self.descs[k],k=2)\n\n # make sure the distance to the closest match is sufficiently better than the second closest\n good_matches = []\n for m,n in matches: \n if (m.distance < 0.7*n.distance and\n kp[m.queryIdx].response > 0.0 and\n self.kps[k][m.trainIdx].response > 0.0):\n good_matches.append((m.queryIdx, m.trainIdx))\n \n if len(good_matches) < 4:\n visual_diff = None\n else:\n #create pt lists\n img_pts = np.zeros((len(good_matches),2))\n tem_pts = np.zeros((len(good_matches),2))\n for idx in range(len(good_matches)):\n match = good_matches[idx]\n img_pts[idx,:] = kp[match[0]].pt\n tem_pts[idx,:] = self.kps[k][match[1]].pt\n \n # Transform input image so that it matches the template image as well as possible\n M, mask = cv2.findHomography(img_pts, tem_pts, cv2.RANSAC, 8.0)\n img_T = cv2.warpPerspective(img, M, self.signs[k].shape[::-1])\n \t\n #calculate error\n visual_diff = compare_images(img_T, self.signs[k])\n return visual_diff\n #not functioning properly (currently does not return highest value for identical image/scene pair)","def ratio_test(fx2_match, fx1_match, fx1_norm, match_dist, norm_dist,\n ratio_thresh=.625, fm_dtype=np.int32, fs_dtype=np.float32):\n fx2_to_ratio = np.divide(match_dist, norm_dist).astype(fs_dtype)\n fx2_to_isvalid = np.less(fx2_to_ratio, ratio_thresh)\n fx2_match_RAT = fx2_match.compress(fx2_to_isvalid).astype(fm_dtype)\n fx1_match_RAT = fx1_match.compress(fx2_to_isvalid).astype(fm_dtype)\n fx1_norm_RAT = fx1_norm.compress(fx2_to_isvalid).astype(fm_dtype)\n # Turn the ratio into a score\n fs_RAT = np.subtract(1.0, fx2_to_ratio.compress(fx2_to_isvalid))\n fm_RAT = np.vstack((fx1_match_RAT, fx2_match_RAT)).T\n # return normalizer info as well\n fm_norm_RAT = np.vstack((fx1_norm_RAT, fx2_match_RAT)).T\n ratio_tup = MatchTup3(fm_RAT, fs_RAT, fm_norm_RAT)\n return ratio_tup","def evaluate_detections(ground_truth, predictions, class_name, overlap_threshold=0.5,\n allow_multiple_matches_per_ignored=True,\n verbose=True):\n\n Detection = namedtuple('Detection', ['image', 'bbox', 'score', 'gt_match'])\n GT = namedtuple('GroundTruth', ['bbox', 'is_matched', 'is_ignored'])\n detections = [Detection(image=img_pred.image_path,\n bbox=np.array(obj_pred[\"bbox\"]),\n score=obj_pred.get(\"score\", 0.0),\n gt_match=-1)\n for img_pred in predictions\n for obj_pred in img_pred\n if obj_pred[\"type\"] == class_name]\n\n scores = np.array([detection.score for detection in detections])\n sorted_ind = np.argsort(-scores)\n detections = [detections[i] for i in sorted_ind]\n\n gts = {}\n for img_gt in ground_truth:\n gts[img_gt.image_path] = GT(\n bbox=np.vstack([np.array(obj_gt[\"bbox\"]) for obj_gt in img_gt]) if img_gt else np.empty(\n (0, 4)),\n is_matched=np.zeros(len(img_gt), dtype=bool),\n is_ignored=np.array([obj_gt.get(\"is_ignored\", False) for obj_gt in img_gt], dtype=bool))\n\n detections_num = len(detections)\n true_pos = np.zeros(detections_num)\n false_pos = np.zeros(detections_num)\n\n for i, detection in tqdm(enumerate(detections), desc=\"Processing detections\",\n disable=not verbose):\n image_path = detection.image\n bboxes_gt = gts[image_path].bbox\n bbox = detection.bbox\n max_overlap = -np.inf\n\n if bboxes_gt is not None and bboxes_gt.shape[0] > 0:\n intersection_xmin = np.maximum(bboxes_gt[:, 0], bbox[0])\n intersection_ymin = np.maximum(bboxes_gt[:, 1], bbox[1])\n intersection_xmax = np.minimum(bboxes_gt[:, 0] + bboxes_gt[:, 2], bbox[0] + bbox[2])\n intersection_ymax = np.minimum(bboxes_gt[:, 1] + bboxes_gt[:, 3], bbox[1] + bbox[3])\n intersection_width = np.maximum(intersection_xmax - intersection_xmin, 0.)\n intersection_height = np.maximum(intersection_ymax - intersection_ymin, 0.)\n intersection = intersection_width * intersection_height\n\n det_area = bbox[2] * bbox[3]\n gt_area = bboxes_gt[:, 2] * bboxes_gt[:, 3]\n union = (det_area + gt_area - intersection)\n ignored_mask = gts[image_path].is_ignored\n if allow_multiple_matches_per_ignored:\n if np.any(ignored_mask):\n union[ignored_mask] = det_area\n\n overlaps = intersection / union\n # Match not ignored ground truths first.\n if np.any(~ignored_mask):\n overlaps_filtered = np.copy(overlaps)\n overlaps_filtered[ignored_mask] = 0.0\n max_overlap = np.max(overlaps_filtered)\n argmax_overlap = np.argmax(overlaps_filtered)\n # If match with non-ignored ground truth is not good enough,\n # try to match with ignored ones.\n if max_overlap < overlap_threshold and np.any(ignored_mask):\n overlaps_filtered = np.copy(overlaps)\n overlaps_filtered[~ignored_mask] = 0.0\n max_overlap = np.max(overlaps_filtered)\n argmax_overlap = np.argmax(overlaps_filtered)\n detections[i] = detection._replace(gt_match=argmax_overlap)\n\n if max_overlap >= overlap_threshold:\n if not gts[image_path].is_ignored[argmax_overlap]:\n if not gts[image_path].is_matched[argmax_overlap]:\n true_pos[i] = 1.\n gts[image_path].is_matched[argmax_overlap] = True\n else:\n false_pos[i] = 1.\n elif not allow_multiple_matches_per_ignored:\n gts[image_path].is_matched[argmax_overlap] = True\n else:\n false_pos[i] = 1.\n\n false_pos = np.cumsum(false_pos)\n true_pos = np.cumsum(true_pos)\n\n debug_visualization = False\n if debug_visualization:\n for image_path, bboxes_gt in gts.items():\n\n print(image_path)\n image = cv2.imread(image_path)\n image_gt = np.copy(image)\n for bbox in bboxes_gt.bbox:\n cv2.rectangle(image_gt, tuple(bbox[:2]), tuple(bbox[2:] + bbox[:2]),\n color=(255, 255, 0), thickness=2)\n cv2.imshow(\"gt\", image_gt)\n for detection in detections:\n if detection.image != image_path:\n continue\n bbox = detection.bbox\n cv2.rectangle(image, tuple(bbox[:2]), tuple(bbox[2:] + bbox[:2]), color=(0, 255, 0),\n thickness=2)\n if detection.gt_match is not None:\n bbox = bboxes_gt.bbox[detection.gt_match]\n cv2.rectangle(image, tuple(bbox[:2]), tuple(bbox[2:] + bbox[:2]),\n color=(0, 0, 255), thickness=1)\n cv2.imshow(\"image\", image)\n cv2.waitKey(0)\n\n # Handle equal-score detections.\n # Get index of the last occurrence of a score.\n ind = len(scores) - np.unique(scores[sorted_ind[::-1]], return_index=True)[1] - 1\n ind = ind[::-1]\n # Though away redundant points.\n false_pos = false_pos[ind]\n true_pos = true_pos[ind]\n\n total_positives_num = np.sum([np.count_nonzero(~gt.is_ignored) for gt in gts.values()])\n recall = true_pos / float(total_positives_num)\n # Avoid divide by zero in case the first detection matches an ignored ground truth.\n precision = true_pos / np.maximum(true_pos + false_pos, np.finfo(np.float64).eps)\n miss_rate = 1.0 - recall\n fppi = false_pos / float(len(gts))\n\n return recall, precision, miss_rate, fppi","def evaluate(pred_file, ref_file):\n ref_dict, pred_dict, query_dict, id_dict = build_pred_ref_dict(ref_file, pred_file, ref_file)\n total, acc, scores = res_eval_with_type_acc(query_dict, pred_dict, ref_dict, id_dict, save=False)\n em = calculate_exact_match(pred_dict, ref_dict)\n print('Comp Acc: {:.3f}%\\tBleu-4: {:.3f}\\tRouge-L: {:.3f}'.format(acc, scores['Bleu-4'], scores['Rouge-L']))\n print('EM: {:.3f}%'.format(em))\n # calculate_sketch_type_acc(ref_file, pred_file)\n # calculate_exact_match_for_each_q_type(ref_file, pred_file)\n return total, acc, scores, em","def ransac(matches, kp1, kp2, sample_points=4, iterations=5, inlier_tolerance=3, inlier_ratio=0.45, check=True, return_max_x=False):\n\n best_inlier_count = 0\n best_h = None\n best_inlier_indices = None\n\n # Get all the corresponing matching pairs for both the images\n pts1 = np.array([kp1[m.queryIdx].pt for m in matches])\n pts2 = np.array([kp2[m.trainIdx].pt for m in matches])\n\n # Re-usable variables for all iterations\n homogeneous_pts1 = np.hstack((pts1, np.ones((pts1.shape[0], 1)))).T\n indices = np.arange(len(pts1))\n num_pts = pts1.shape[0]\n required_inliers = inlier_ratio * num_pts\n\n # For number of iterations\n for _ in range(iterations):\n\n # Sample a small set of points from the point match pairs\n indices_to_sample = np.random.choice(indices, sample_points)\n pts1_sample = pts1[indices_to_sample]\n pts2_sample = pts2[indices_to_sample]\n\n # Get the homography matrix\n h = get_homography_matrix(pts1_sample, pts2_sample)\n\n # Find the new points using the homography matrix\n transformed_points = np.dot(h, homogeneous_pts1).T\n\n # Convert it to world coordinates\n last_col = np.copy(transformed_points[:, -1])\n last_col = last_col[:, np.newaxis]\n transformed_points /= last_col\n transformed_points = transformed_points[:, :-1]\n\n # Find the distance between the actual and the mapped points\n distance = np.linalg.norm(pts2 - transformed_points, axis=1)\n inlier_indices = distance < inlier_tolerance\n inlier_count = inlier_indices.sum()\n\n # Update the best_h if the current h has more inliers\n if inlier_count > best_inlier_count:\n best_h = h\n best_inlier_indices = inlier_indices\n best_inlier_count = inlier_count\n\n # If required inliers is reached break\n if inlier_count > required_inliers:\n break\n\n # Verbose mode - Print the number of inliers\n if check:\n transformed_points = np.dot(best_h, homogeneous_pts1).T\n # Convert it to world coordinates\n last_col = np.copy(transformed_points[:, -1])\n last_col = last_col[:, np.newaxis]\n transformed_points /= last_col\n transformed_points = transformed_points[:, :-1]\n distance = np.linalg.norm(pts2 - transformed_points, axis=1)\n inlier_count = len(distance[distance < inlier_tolerance])\n print('%2.2f of the points are inliers' %\n (inlier_count / num_pts * 100))\n\n # If x coordinates are needed\n if return_max_x:\n max_x_inlier_1 = ceil(pts1[best_inlier_indices].max(axis=0)[0])\n max_x_inlier_2 = ceil(pts2[best_inlier_indices].max(axis=0)[0])\n return best_h, max_x_inlier_1, max_x_inlier_2\n return best_h","def test_compare_classifier_performance_from_pred_vis_api(experiment_to_use):\n experiment = experiment_to_use\n prediction = experiment.predictions\n viz_outputs = (\"pdf\", \"png\")\n with TemporaryDirectory() as tmpvizdir:\n for viz_output in viz_outputs:\n vis_output_pattern_pdf = tmpvizdir + f\"/*.{viz_output}\"\n visualize.compare_classifiers_performance_from_pred(\n [prediction, prediction],\n experiment.ground_truth,\n experiment.ground_truth_metadata,\n experiment.output_feature_name,\n labels_limit=0,\n model_namess=[\"Model1\", \"Model2\"],\n output_directory=tmpvizdir,\n file_format=viz_output,\n )\n figure_cnt = glob.glob(vis_output_pattern_pdf)\n assert 1 == len(figure_cnt)","def calculates_results_stats(results_dic): \n # Creates empty dictionary for results_stats_dic\n results_stats_dic = dict()\n \n # Sets all counters to initial values of zero so that they can \n # be incremented while processing through the images in results_dic \n results_stats_dic['n_dogs_img'] = 0\n results_stats_dic['n_match'] = 0\n results_stats_dic['n_correct_dogs'] = 0\n results_stats_dic['n_correct_notdogs'] = 0\n results_stats_dic['n_correct_breed'] = 0\n \n # process through the results dictionary\n for key in results_dic:\n \n # Labels Match Exactly\n if results_dic[key][2] == 1:\n results_stats_dic['n_match'] += 1\n\n # TODO: 5a. REPLACE pass with CODE that counts how many pet images of\n # dogs had their breed correctly classified. This happens \n # when the pet image label indicates the image is-a-dog AND \n # the pet image label and the classifier label match. You \n # will need to write a conditional statement that determines\n # when the dog breed is correctly classified and then \n # increments 'n_correct_breed' by 1. Recall 'n_correct_breed' \n # is a key in the results_stats_dic dictionary with it's value \n # representing the number of correctly classified dog breeds.\n # \n # Pet Image Label is a Dog AND Labels match- counts Correct Breed\n if results_dic[key][3] == 1 and results_dic[key][2] == 1:\n results_stats_dic['n_correct_breed'] += 1\n \n # Pet Image Label is a Dog - counts number of dog images\n if results_dic[key][3] == 1:\n results_stats_dic['n_dogs_img'] += 1\n \n # Classifier classifies image as Dog (& pet image is a dog)\n # counts number of correct dog classifications\n if results_dic[key][4] == 1:\n results_stats_dic['n_correct_dogs'] += 1\n\n # TODO: 5b. REPLACE pass with CODE that counts how many pet images \n # that are NOT dogs were correctly classified. This happens \n # when the pet image label indicates the image is-NOT-a-dog \n # AND the classifier label indicates the images is-NOT-a-dog.\n # You will need to write a conditional statement that \n # determines when the classifier label indicates the image \n # is-NOT-a-dog and then increments 'n_correct_notdogs' by 1. \n # Recall the 'else:' above 'pass' already indicates that the \n # pet image label indicates the image is-NOT-a-dog and \n # 'n_correct_notdogs' is a key in the results_stats_dic dictionary \n # with it's value representing the number of correctly \n # classified NOT-a-dog images.\n # \n # Pet Image Label is NOT a Dog\n else:\n # Classifier classifies image as NOT a Dog(& pet image isn't a dog)\n # counts number of correct NOT dog clasifications.\n if results_dic[key][3] == 0 and results_dic[key][4] == 0:\n results_stats_dic['n_correct_notdogs'] += 1\n\n\n # Calculates run statistics (counts & percentages) below that are calculated\n # using the counters from above.\n\n # calculates number of total images\n results_stats_dic['n_images'] = len(results_dic)\n\n # calculates number of not-a-dog images using - images & dog images counts\n results_stats_dic['n_notdogs_img'] = (results_stats_dic['n_images'] - \n results_stats_dic['n_dogs_img']) \n\n # TODO: 5c. REPLACE zero(0.0) with CODE that calculates the % of correctly\n # matched images. Recall that this can be calculated by the\n # number of correctly matched images ('n_match') divided by the \n # number of images('n_images'). This result will need to be \n # multiplied by 100.0 to provide the percentage.\n # \n # Calculates % correct for matches\n results_stats_dic['pct_match'] = (results_stats_dic['n_match'] / results_stats_dic['n_images']) * 100\n\n # TODO: 5d. REPLACE zero(0.0) with CODE that calculates the % of correctly\n # classified dog images. Recall that this can be calculated by \n # the number of correctly classified dog images('n_correct_dogs')\n # divided by the number of dog images('n_dogs_img'). This result \n # will need to be multiplied by 100.0 to provide the percentage.\n # \n # Calculates % correct dogs\n results_stats_dic['pct_correct_dogs'] = (results_stats_dic['n_correct_dogs'] / results_stats_dic['n_dogs_img']) * 100\n\n # TODO: 5e. REPLACE zero(0.0) with CODE that calculates the % of correctly\n # classified breeds of dogs. Recall that this can be calculated \n # by the number of correctly classified breeds of dog('n_correct_breed') \n # divided by the number of dog images('n_dogs_img'). This result \n # will need to be multiplied by 100.0 to provide the percentage.\n # \n # Calculates % correct breed of dog\n results_stats_dic['pct_correct_breed'] = (results_stats_dic['n_correct_breed'] / results_stats_dic['n_dogs_img']) * 100\n\n # Calculates % correct not-a-dog images\n # Uses conditional statement for when no 'not a dog' images were submitted \n if results_stats_dic['n_notdogs_img'] > 0:\n results_stats_dic['pct_correct_notdogs'] = (results_stats_dic['n_correct_notdogs'] /\n results_stats_dic['n_notdogs_img']) * 100.0\n else:\n results_stats_dic['pct_correct_notdogs'] = 0.0\n\n \n # TODO 5f. REPLACE None with the results_stats_dic dictionary that you \n # created with this function \n return results_stats_dic","def _run():\n matching_terms = {'a', 'b'}\n source_counts = {'a': 10, 'b': 50, 'c': 25}\n target_counts = {'a': 4, 'b': 73, 'c': 15}\n source_chunk = ['a', 'b']\n target_chunk = ['a', 'c', 'b']\n source_distance = score.find_distance(\n matching_terms, source_chunk, source_counts)\n target_distance = score.find_distance(\n matching_terms, target_chunk, target_counts)\n match_score = score.vanilla(\n matching_terms, source_distance, target_distance, source_counts,\n target_counts)\n print('Calculated score:', match_score)","def analyze_similarities():\r\n print('Total number of candidate pairs:', len(pairs))\r\n print(f'\\nNumber of actual item pairs in the train set: {pairs[\"true_match\"].sum()}\\n')\r\n\r\n for feature in ['text_score', 'image_score', 'txt_img_score', 'words_ratio', 'txt_img_words']:\r\n\r\n # Check distribution of True and False predictions for various similarity scores\r\n print('-' * 50)\r\n print(f'\\nDistribution of True/False predictions for {feature}')\r\n for thr in (0.6, 0.65, 0.7, 0.75, 0.8, 0.85, 0.9, 0.95):\r\n print('-' * 50)\r\n print(f'Similarity score over {thr}')\r\n pairs_sample = pairs[pairs[feature] >= thr]\r\n print(f'Number of similar item pairs: {len(pairs_sample)}')\r\n print(pairs_sample['true_match'].value_counts(normalize=True))\r\n\r\n # Check if identical phash can be used to improve the accuracy\r\n same_phash = pairs[pairs['phash_match'] == 1]\r\n different_phash = pairs[pairs['phash_match'] == 0]\r\n\r\n print('\\nFor item pairs with the same phash:')\r\n print(same_phash['true_match'].value_counts(normalize=True))\r\n print('Number of item pairs in this subset:', len(same_phash))\r\n\r\n print('\\nFor item pairs with different phash:')\r\n print(different_phash['true_match'].value_counts(normalize=True))\r\n print('Number of item pairs in this subset:', len(different_phash))\r\n\r\n # Check if numbers in titles can be used to improve the accuracy\r\n same_numbers = pairs[pairs['nums_match'] == 1]\r\n different_numbers = pairs[pairs['nums_match'] == 0]\r\n\r\n print('\\nFor item pairs with the same numbers:')\r\n print(same_numbers['true_match'].value_counts(normalize=True))\r\n print('Number of item pairs in this subset:', len(same_numbers))\r\n\r\n print('\\nFor item pairs with different numbers:')\r\n print(different_numbers['true_match'].value_counts(normalize=True))\r\n print('Number of item pairs in this subset:', len(different_numbers))","def drawMatches(img1, kp1, img2, kp2, matches):\n\n # Create a new output image that concatenates the two images together\n # (a.k.a) a montage\n rows1 = img1.shape[0]\n cols1 = img1.shape[1]\n rows2 = img2.shape[0]\n cols2 = img2.shape[1]\n\n out = np.zeros((max([rows1,rows2]),cols1+cols2,3), dtype='uint8')\n\n # Place the first image to the left\n out[:rows1,:cols1] = np.dstack([img1, img1, img1])\n\n # Place the next image to the right of it\n out[:rows2,cols1:] = np.dstack([img2, img2, img2])\n\n # For each pair of points we have between both images\n # draw circles, then connect a line between them\n for mat in matches:\n\n # Get the matching keypoints for each of the images\n img1_idx = mat.queryIdx\n img2_idx = des_dict[mat.trainIdx][1]\n\n # x - columns\n # y - rows\n (x1,y1) = kp1[img1_idx].pt\n (x2,y2) = kp2[img2_idx].pt\n\n # Draw a small circle at both co-ordinates\n # radius 4\n # colour green\n # thickness = 1\n cv2.circle(out, (int(x1),int(y1)), 4, (0, 255, 0)) \n cv2.circle(out, (int(x2)+cols1,int(y2)), 4, (0, 255, 0))\n\n # Draw a line in between the two points\n # thickness = 1\n # colour green\n cv2.line(out, (int(x1),int(y1)), (int(x2)+cols1,int(y2)), (0, 255, 0), 1)\n\n\n # Show the image\n #cv2.imshow('Matched Features', out)\n #cv2.waitKey(0)\n #cv2.destroyWindow('Matched Features')\n\n # Also return the image if you'd like a copy\n return out","def my_evaluate(original_image, img_id, annotations, min_score=0.4, max_overlap=0.3, top_k=200):\n\n # Transform\n image = normalize(to_tensor(resize(original_image)))\n\n # Move to default device\n image = image.to(device)\n\n # Forward prop.\n predicted_locs, predicted_scores = model(image.unsqueeze(0))\n\n # Detect objects in SSD output\n det_boxes, det_labels, det_scores = model.detect_objects(predicted_locs, predicted_scores, min_score=min_score,\n max_overlap=max_overlap, top_k=top_k)\n\n # Move detections to the CPU\n det_boxes = det_boxes[0].to('cpu')\n\n # Transform to original image dimensions\n original_dims = torch.FloatTensor(\n [original_image.width, original_image.height, original_image.width, original_image.height]).unsqueeze(0)\n det_boxes = det_boxes * original_dims\n\n # Decode class integer labels\n det_labels = [rev_label_map[l] for l in det_labels[0].to('cpu').tolist()]\n det_scores = det_scores[0].to('cpu').tolist()\n\n if det_labels != ['background']:\n for i in range(det_boxes.size(0)):\n \n # For every detection, see whether it matches a ground truth\n match = False\n for j in range(len(annotations['boxes'])):\n if annotations['labels'][j] == -1: # this is being set when a ground truth already matched\n continue\n match = box_match(det_boxes[i], \n det_labels[i], \n torch.Tensor(annotations['boxes'][j]), \n rev_label_map[annotations['labels'][j]])\n if match:\n annotations['labels'][j] = -1\n break\n \n if match: # true positive if the detection is correct and matched a ground truth\n all_results['true_positive'] += 1\n else: # false positive if the detection did not match a ground truth\n all_results['false_positive'] += 1\n \n # After all detections were checked, let us see whether the detector missed something\n for label in annotations['labels']: \n if label == -1: # This is set after a detection matched this ground truth\n continue\n \n # false negative if we reach this line, since the ground truth object was not found\n all_results['false_negative'] += 1","def test_calculate_precision_happy_path(\n mock_engine: DicomImagePiiVerifyEngine,\n gt_labels: List[dict],\n all_pos: List[dict],\n expected_result: float,\n):\n # Act\n test_precision = mock_engine.calculate_precision(gt_labels, all_pos)\n\n # Assert\n assert test_precision == expected_result","def testdata_matcher(fname1='easy1.png', fname2='easy2.png'):\n import utool as ut\n #import vtool as vt\n from vtool import image as gtool\n from vtool import features as feattool\n fpath1 = ut.grab_test_imgpath(fname1)\n fpath2 = ut.grab_test_imgpath(fname2)\n featkw = dict(rotation_invariance=True)\n kpts1, vecs1 = feattool.extract_features(fpath1, **featkw)\n kpts2, vecs2 = feattool.extract_features(fpath2, **featkw)\n #if featkw['rotation_invariance']:\n # print('ori stats 1 ' + ut.get_stats_str(vt.get_oris(kpts2)))\n # print('ori stats 2 ' + ut.get_stats_str(vt.get_oris(kpts1)))\n rchip1 = gtool.imread(fpath1)\n rchip2 = gtool.imread(fpath2)\n #chip1_shape = vt.gtool.open_image_size(fpath1)\n chip2_shape = gtool.open_image_size(fpath2)\n dlen_sqrd2 = chip2_shape[0] ** 2 + chip2_shape[1]\n testtup = (rchip1, rchip2, kpts1, vecs1, kpts2, vecs2, dlen_sqrd2)\n return testtup","def run_matching(self):\n paradic = self.cfg['param']['paradic']\n print 'in run_matching() n_bins = ' +str(paradic['n_bins'])\n\n f = open(self.work_dir+'matches.txt','w')\n matching = self.run_proc(['match_cli', 'keys_im0.txt',\n 'keys_im1.txt',\n str(paradic['flag_match']),\n str(paradic['C_match']),\n str(paradic['n_hist']),\n str(paradic['n_ori']),\n str(paradic['n_bins'])],\n stdout=f)\n self.wait_proc(matching, timeout=self.timeout)\n return 1","def drawMatches(orbimg1, kp1, orbimg2, kp2, matches):\r\n\r\n # Create a new output image that concatenates the two images together\r\n # (a.k.a) a montage\r\n rows1 = orbimg1.shape[0]\r\n cols1 = orbimg1.shape[1]\r\n rows2 = orbimg2.shape[0]\r\n cols2 = orbimg2.shape[1]\r\n\r\n # Create the output image\r\n # The rows of the output are the largest between the two images\r\n # and the columns are simply the sum of the two together\r\n # The intent is to make this a colour image, so make this 3 channels\r\n out = np.zeros((max([rows1,rows2]),cols1+cols2,3), dtype='uint8')\r\n\r\n # Place the first image to the left\r\n out[:rows1,:cols1] = np.dstack([orbimg1, orbimg1, orbimg1])\r\n\r\n # Place the next image to the right of it\r\n out[:rows2,cols1:] = np.dstack([orbimg2, orbimg2, orbimg2])\r\n\r\n # For each pair of points we have between both images\r\n # draw circles, then connect a line between them\r\n for mat in matches:\r\n\r\n # Get the matching keypoints for each of the images\r\n orbimg1_idx = mat.queryIdx\r\n orbimg2_idx = mat.trainIdx\r\n\r\n # x - columns\r\n # y - rows\r\n (x1,y1) = kp1[orbimg1_idx].pt\r\n (x2,y2) = kp2[orbimg2_idx].pt\r\n\r\n # Draw a small circle at both co-ordinates\r\n # radius 4\r\n # colour blue\r\n # thickness = 1\r\n cv2.circle(out, (int(x1),int(y1)), 4, (255, 0, 0), 1) \r\n cv2.circle(out, (int(x2)+cols1,int(y2)), 4, (255, 0, 0), 1)\r\n\r\n # Draw a line in between the two points\r\n # thickness = 1\r\n # colour blue\r\n cv2.line(out, (int(x1),int(y1)), (int(x2)+cols1,int(y2)), (255,0,0), 1)\r\n\r\n\r\n # Show the image\r\n #cv2.imshow('Matched Features', out)\r\n #cv2.waitKey(0)\r\n #cv2.destroyWindow('Matched Features')\r\n\r\n # Also return the image if you'd like a copy\r\n return out","def calculate_precision_recall_all_images(\n all_prediction_boxes, all_gt_boxes, iou_threshold):\n # Find total true positives, false positives and false negatives\n # over all images\n\n tot_num_tp, tot_num_fp, tot_num_fn = 0, 0, 0\n\n for image in range(len(all_prediction_boxes)):\n prediction_boxes = all_prediction_boxes[image]\n gt_boxes = all_gt_boxes[image]\n\n image_results = calculate_individual_image_result(\n prediction_boxes,\n gt_boxes,\n iou_threshold)\n\n tot_num_tp += image_results[\"true_pos\"]\n tot_num_fp += image_results[\"false_pos\"]\n tot_num_fn += image_results[\"false_neg\"]\n\n # Compute precision, recall\n precision = calculate_precision(tot_num_tp, tot_num_fp, tot_num_fn)\n recall = calculate_recall(tot_num_tp, tot_num_fp, tot_num_fn)\n\n return (precision, recall)","def test_compare_classifier_performance_from_prob_vis_api(experiment_to_use):\n experiment = experiment_to_use\n probability = experiment.probabilities\n viz_outputs = (\"pdf\", \"png\")\n with TemporaryDirectory() as tmpvizdir:\n for viz_output in viz_outputs:\n vis_output_pattern_pdf = tmpvizdir + f\"/*.{viz_output}\"\n visualize.compare_classifiers_performance_from_prob(\n [probability, probability],\n experiment.ground_truth,\n experiment.ground_truth_metadata,\n experiment.output_feature_name,\n top_n_classes=[0],\n labels_limit=0,\n model_namess=[\"Model1\", \"Model2\"],\n output_directory=tmpvizdir,\n file_format=viz_output,\n )\n figure_cnt = glob.glob(vis_output_pattern_pdf)\n assert 1 == len(figure_cnt)","def find_matching_points(img1, img2, max_pix_movement=50, normalize=True, show=False):\n\n # Initiate ORB detector\n orb = cv2.ORB_create()\n\n # find the keypoints and descriptors with ORB\n kp1, des1 = orb.detectAndCompute(img1, None)\n kp2, des2 = orb.detectAndCompute(img2, None)\n\n # create BFMatcher object\n bf = cv2.BFMatcher(cv2.NORM_HAMMING, crossCheck=True)\n # Match descriptors.\n matches = bf.match(des1,des2)\n # Sort them in the order of their distance.\n matches = sorted(matches, key = lambda x:x.distance)\n # Draw first 10 matches.\n if show:\n img3 = cv2.drawMatches(img1,kp1,img2,kp2,matches[:500], None,flags=2)\n plt.imshow(img3),plt.show()\n # Get the matching keypoints for each of the images\n\n list_kp1 = []\n list_kp2 = []\n for mat in matches:\n img1_idx = mat.queryIdx\n img2_idx = mat.trainIdx\n\n # x - columns\n # y - rows\n list_kp1.append(kp1[img1_idx].pt)\n list_kp2.append(kp2[img2_idx].pt)\n\n n_kp1, n_kp2 = np.float32(list_kp1), np.float32(list_kp2)\n n_kp1 /= np.asarray([img1.shape[1], img1.shape[0]], np.float32)\n n_kp2 /= np.asarray([img2.shape[1], img2.shape[0]], np.float32)\n n_kp1 = n_kp1 * 2. - 1.\n n_kp2 = n_kp2 * 2. - 1.\n\n return np.int32(list_kp1), np.int32(list_kp2), n_kp1, n_kp2","def compute_average_precision_per_class(num_true_cases, gt_boxes,\n\t\t\t\t\t\t\t\t\t\tprediction_file, iou_threshold, use_2007_metric):\n\twith open(prediction_file) as f:\n\t\timage_ids = []\n\t\tboxes = []\n\t\tscores = []\n\t\tfor line in f:\n\t\t\tt = line.rstrip().split(\" \")\n\t\t\timage_ids.append(t[0])\n\t\t\tscores.append(float(t[1]))\n\t\t\tbox = torch.tensor([float(v) for v in t[2:]]).unsqueeze(0)\n\t\t\tbox -= 1.0 # convert to python format where indexes start from 0\n\t\t\tboxes.append(box)\n\t\tscores = np.array(scores)\n\t\tsorted_indexes = np.argsort(-scores)\n\t\tboxes = [boxes[i] for i in sorted_indexes]\n\t\timage_ids = [image_ids[i] for i in sorted_indexes]\n\t\ttrue_positive = np.zeros(len(image_ids))\n\t\tfalse_positive = np.zeros(len(image_ids))\n\t\tmatched = set()\n\t\tfor i, image_id in enumerate(image_ids):\n\t\t\tbox = boxes[i]\n\t\t\tif image_id not in gt_boxes:\n\t\t\t\tfalse_positive[i] = 1\n\t\t\t\tcontinue\n\n\t\t\tgt_box = gt_boxes[image_id]\n\t\t\tious = box_utils.iou_of(box, gt_box)\n\t\t\tmax_iou = torch.max(ious).item()\n\t\t\tmax_arg = torch.argmax(ious).item()\n\t\t\tif max_iou > iou_threshold:\n\t\t\t\tif (image_id, max_arg) not in matched:\n\t\t\t\t\ttrue_positive[i] = 1\n\t\t\t\t\tmatched.add((image_id, max_arg))\n\t\t\t\telse:\n\t\t\t\t\tfalse_positive[i] = 1\n\t\t\telse:\n\t\t\t\tfalse_positive[i] = 1\n\n\ttrue_positive = true_positive.cumsum()\n\tfalse_positive = false_positive.cumsum()\n\tprecision = true_positive / (true_positive + false_positive)\n\trecall = true_positive / num_true_cases\n\tif use_2007_metric:\n\t\treturn measurements.compute_voc2007_average_precision(precision, recall)\n\telse:\n\t\treturn measurements.compute_average_precision(precision, recall)","def matching_accuracy(pmat_pred, pmat_gt, ns):\n device = pmat_pred.device\n batch_num = pmat_pred.shape[0]\n\n pmat_gt = pmat_gt.to(device)\n\n assert torch.all((pmat_pred == 0) + (pmat_pred == 1)), 'pmat_pred can noly contain 0/1 elements.'\n assert torch.all((pmat_gt == 0) + (pmat_gt == 1)), 'pmat_gt should noly contain 0/1 elements.'\n assert torch.all(torch.sum(pmat_gt, dim=-1) <= 1) and torch.all(torch.sum(pmat_gt, dim=-2) <= 1)\n assert torch.all(torch.sum(pmat_pred, dim=-1) <= 1) and torch.all(torch.sum(pmat_pred, dim=-2) <= 1)\n\n #indices_pred = torch.argmax(pmat_pred, dim=-1)\n #indices_gt = torch.argmax(pmat_gt, dim=-1)\n\n #matched = (indices_gt == indices_pred).type(pmat_pred.dtype)\n match_num_list = []\n gt_num_list = []\n pred_num_list = []\n acc_gt = []\n acc_pred = []\n for b in range(batch_num): #acc_gt,acc_pred\n #match_num += torch.sum(matched[b, :ns[b]])\n #total_num += ns[b].item()\n match_num = torch.sum(pmat_pred[b, :] * pmat_gt[b, :]) + 1e-8 #:ns[b]\n gt_num = torch.sum(pmat_gt[b, :]) + 1e-8\n pred_num = torch.sum(pmat_pred[b,:]) + 1e-8\n match_num_list.append(match_num.cpu().numpy())\n gt_num_list.append(gt_num.cpu().numpy())\n pred_num_list.append(pred_num.cpu().numpy())\n acc_gt.append((match_num/gt_num).cpu().numpy())\n acc_pred.append((match_num/pred_num).cpu().numpy())\n\n return {'acc_gt': np.array(acc_gt),\n 'acc_pred': np.array(acc_pred),\n 'match_num': np.array(match_num_list),\n 'gt_num': np.array(gt_num_list),\n 'pred_num': np.array(pred_num_list)}","def evaluate(self, test_data):\n test_results = [(self.decide(self.feedforward(x)), y)\n for (x, y) in test_data]\n\n # Check for and keep track of TP's FP's and FN's\n # Write FP's and FN's to a special directories\n TP = FP = FN = count = 0\n TN = 0\n # what you got = x, what should be = y\n for ((x, y), (image, gt)) in zip(test_results, test_data):\n count += 1\n if x == 1 and y == 1:\n TP += 1\n elif x == 1 and y == 0:\n FP += 1\n if not self.validating: self.save_Image(FP_PATH, count, image)\n elif x == 0 and y == 1:\n FN += 1\n if not self.validating: self.save_Image(FN_PATH, count, image)\n \n if (TP + FP ) == 0:\n return TP, FP, FN, '---------', '---------','---------'\n else:\n false_rate = float(FP) / float(TP + FP)\n\n if (TP + FN ) == 0:\n return TP, FP, FN, '---------', '---------', false_rate\n else:\n detect_rate = float(TP) / float(TP + FN)\n\n if (TP + FP + FN ) == 0:\n return TP, FP, FN, '---------', detect_rate, false_rate\n else:\n quality_rate = float(TP) / float(TP+FP+FN)\n \n return TP, FP, FN, quality_rate, detect_rate, false_rate","def match(desc1,desc2):\n desc1 = array([d/linalg.norm(d) for d in desc1])\n desc2 = array([d/linalg.norm(d) for d in desc2])\n dist_ratio = 0.6\n desc1_size = desc1.shape\n matchscores = zeros((desc1_size[0],1),'int')\n desc2t = desc2.T # precompute matrix transpose\n for i in range(desc1_size[0]):\n dotprods = dot(desc1[i, :], desc2t) # vector of dot products\n dotprods *= 0.9999\n # inverse cosine and sort, return index for features in second image\n indx = argsort(arccos(dotprods))\n # check if nearest neighbor has angle less than dist_ratio times 2nd\n if arccos(dotprods)[indx[0]] < dist_ratio * arccos(dotprods)[indx[1]]:\n matchscores[i] = int(indx[0])\n return matchscores","def max_disp(left_img, right_img):\n\n # Max initial matches\n MAX_INI_MATCH = np.inf\n\n # ORB detector and brute-force matcher\n detector = cv2.ORB_create(nfeatures=100000, edgeThreshold=5)\n matcher = cv2.BFMatcher(cv2.NORM_HAMMING, crossCheck=True)\n\n # Find the keypoints and descriptors of both images\n keypoint_left, descriptor_left = detector.detectAndCompute(\n image=left_img, mask=None\n )\n keypoint_right, descriptor_right = detector.detectAndCompute(\n image=right_img, mask=None\n )\n\n # Find the best matches\n matches = matcher.match(\n queryDescriptors=descriptor_left, trainDescriptors=descriptor_right\n )\n matches = sorted(matches, key=lambda x: x.distance)\n\n if DEBUG:\n print(f\"Initial matches#\")\n print(f\"match_lr: {len(matches)}\")\n\n # Match refinement with RANSAC\n left_pts = np.float32([keypoint_left[m.queryIdx].pt for m in matches]).reshape(\n -1, 1, 2\n )\n right_pts = np.float32([keypoint_right[m.trainIdx].pt for m in matches]).reshape(\n -1, 1, 2\n )\n homography, mask = cv2.findHomography(left_pts, right_pts, cv2.RANSAC, 5.0)\n matchesMask = mask.ravel().tolist()\n\n ransac_matches = []\n for (m, keep) in zip(matches, matchesMask):\n if keep:\n ransac_matches.append(m)\n\n if DEBUG:\n print(f\"After RANSAC:\")\n print(f\"match: {len(ransac_matches)}\")\n ransac_img = cv2.drawMatches(\n left_img,\n keypoint_left,\n right_img,\n keypoint_right,\n ransac_matches,\n None,\n flags=2,\n )\n cv2.imwrite(\"./debug/RANSAC.png\", ransac_img)\n\n # Get the range of disparity\n disp_min = np.inf\n disp_max = 0\n\n for m in ransac_matches:\n # Keypoints coordinates\n pt_left = np.float32(keypoint_left[m.queryIdx].pt)\n pt_right = np.float32(keypoint_right[m.trainIdx].pt)\n\n # Compute the distance\n dist = int(abs(pt_left - pt_right)[0])\n\n if disp_max < dist:\n disp_max = dist\n\n # Ceiling the number\n disp_max += 1\n\n return disp_max","def calc_TP_FP(groundtruth, eval_boxes, matches_predicate, confidence_threshold=-9e999):\n TP = 0\n FP = 0\n FN = 0\n for image_key in groundtruth.keys() + eval_boxes.keys():\n gt_bboxes = set(groundtruth.get(image_key, []))\n other_bboxes = set(box\n for box in eval_boxes.get(image_key, [])\n if box.confidence >= confidence_threshold\n or box.confidence is None\n )\n for gt_bbox in gt_bboxes:\n # Do any of their boxes match?\n match = None\n for other_box in other_bboxes:\n if matches_predicate(gt_bbox, other_box):\n match = other_box\n if match:\n other_bboxes.remove(match)\n TP += 1\n else:\n FN += 1\n FP += len(other_bboxes)\n # if len(other_bboxes):\n # print \"False positive on\", image_key\n return {\"TP\": TP, \"FP\": FP, \"FN\": FN, \"TN\": 0}","def illustrate_matching(self):\n work_dir = self.work_dir\n draw_keys_oriented(work_dir+'matching_keys_im0.txt',\n work_dir+'input_0.orig.png',\n work_dir+'matching_keys_im0.png')\n draw_keys_oriented(work_dir+'matching_keys_im1.txt',\n work_dir+'input_1.orig.png',\n work_dir+'matching_keys_im1.png')\n draw_matches(work_dir+'matches.txt',\n work_dir+'input_0.orig.png',\n work_dir+'input_1.orig.png',\n work_dir+'OUTmatches.png')\n return 1","def evaluate_match_quality(matches, thresh=.7):\n marr = np.zeros((len(matches), 2))\n quality = np.zeros((len(matches), 1))\n for i in range(len(matches)):\n marr[i,0] = matches[i][0].distance\n marr[i,1] = matches[i][1].distance\n quality[i] = marr[i, 0] < thresh*marr[i, 1]","def main():\n\n parser = argparse.ArgumentParser(description='codec_compare')\n parser.add_argument('path', metavar='DIR',\n help='path to images folder')\n args = parser.parse_args()\n classpath = args.path\n classname = classpath.split('/')[1]\n\n images = set(listdir_full_path(classpath))\n if len(images) <= 0:\n print \"\\033[91m[ERROR]\\033[0m\" + \" no source files in ./images.\"\n sys.exit(1)\n\n codeclist_full = set(['aom', 'deepcoder', 'deepcoder-lite', 'fuif', 'fvdo', 'hevc', 'kakadu', 'jpeg',\n 'pik', 'tat', 'xavs', 'xavs-fast', 'xavs-median', 'webp'])\n\n bpp_targets = set([0.06, 0.12, 0.25, 0.50, 0.75, 1.00, 1.50, 2.00])\n for image in images:\n width, height, depth = get_dimensions(image, classname)\n name, imgfmt = os.path.splitext(image)\n imgfmt = os.path.basename(image).split(\".\")[-1]\n derivative_images = []\n if classname[:6] == 'classB':\n derivative_images = create_derivatives(image, classname)\n else:\n derivative_images.append((image, imgfmt))\n\n for derivative_image, pix_fmt in derivative_images:\n json_dir = 'metrics'\n mkdir_p(json_dir)\n json_file = os.path.join(json_dir,\n os.path.splitext(os.path.basename(derivative_image))[0] + \".\" + pix_fmt + \".json\")\n # if os.path.isfile(json_file):\n # print \"\\033[92m[JSON OK]\\033[0m \" + json_file\n # continue\n main_dict = dict()\n derivative_image_metrics = dict()\n for codecname in codeclist_full:\n convertflag = 1\n caseflag = pix_fmt\n if (codecname == 'webp' or codecname == 'tat' or 'deepcoder' in codecname) and depth != '8':\n continue\n if 'xavs' in codecname and depth != '8' and depth != '10':\n continue\n if 'classE' in classname and ('tat' in codecname or 'xavs' in codecname or 'deepcoder' in codecname):\n continue\n if codecname == 'kakadu' and classname[:6] == 'classB':\n convertflag = 0\n caseflag = imgfmt\n bpp_target_metrics = dict()\n for bpp_target in bpp_targets:\n print(codecname)\n if codecname == 'aom' and classname[:6] == 'classB':\n # ('AERIAL2' in image or 'CATS' in image or 'XRAY' in image or 'GOLD' in image or 'TEXTURE1' in image):\n encoded_image_name = os.path.splitext(os.path.basename(derivative_image))[\n 0] + '_' + str(bpp_target) + '_' + imgfmt + '.' + 'av1'\n encoded_image = os.path.join('outputs', codecname, encoded_image_name)\n decoded_image = os.path.join('outputs', codecname, 'decoded', encoded_image_name + '.' + imgfmt)\n original_image = image\n elif codecname == 'kakadu' and classname[:6] == 'classB':\n encoded_image_name = os.path.splitext(os.path.basename(derivative_image))[\n 0] + '_' + str(bpp_target) + '_' + imgfmt + '.' + codecname\n encoded_image = os.path.join('outputs', codecname, encoded_image_name)\n decoded_image = os.path.join('outputs', codecname, 'decoded', encoded_image_name + '.' + imgfmt)\n original_image = image\n elif 'xavs' in codecname and classname[:6] == 'classB':\n encoded_image_name = os.path.splitext(os.path.basename(derivative_image))[\n 0] + '_' + str(bpp_target) + '_' + imgfmt + '.' + codecname\n encoded_image = os.path.join('outputs', codecname, encoded_image_name)\n decoded_image = os.path.join('outputs', codecname, 'decoded', encoded_image_name + '.' + imgfmt)\n original_image = image\n elif codecname == 'fvdo' and classname[:6] == 'classB':\n encoded_image_name = os.path.splitext(os.path.basename(derivative_image))[\n 0] + '_' + str(bpp_target) + '_pgm' + '.' + codecname\n encoded_image = os.path.join('outputs', codecname, encoded_image_name)\n decoded_image = os.path.join('outputs', codecname, 'decoded', encoded_image_name + '.pgm')\n original_image = image\n else:\n if codecname == 'fuif' and 'tif' in imgfmt:\n encoded_image_name = os.path.splitext(os.path.basename(derivative_image))[\n 0] + '.tif_' + str(bpp_target) + '_' + pix_fmt + '.' + codecname\n elif codecname == 'webp' or codecname == 'tat':\n encoded_image_name = os.path.splitext(os.path.basename(derivative_image))[\n 0] + '_' + str(bpp_target) + '_yuv420p.' + codecname\n else:\n encoded_image_name = os.path.splitext(os.path.basename(derivative_image))[\n 0] + '_' + str(bpp_target) + '_' + pix_fmt + '.' + codecname\n encoded_image = os.path.join('outputs', codecname, encoded_image_name)\n decoded_image_path = os.path.join('outputs', codecname, 'decoded')\n decoded_image = ''\n for decodedfile in os.listdir(decoded_image_path):\n encoderoot = '_'.join(os.path.splitext(os.path.basename(encoded_image_name))[0].split('_')[:-1])\n if encoderoot in decodedfile:\n if ('tat' in codecname or 'webp' in codecname) and os.path.splitext(os.path.basename(decodedfile))[1] == '.yuv':\n decoded_image = os.path.join('outputs', codecname, 'decoded', decodedfile)\n print(decoded_image)\n if ('tat' not in codecname or 'webp' not in codecname) and os.path.splitext(os.path.basename(decodedfile))[1] != '.yuv':\n decoded_image = os.path.join('outputs', codecname, 'decoded', decodedfile)\n if 'classE' not in classname and 'classB' not in classname and os.path.isfile(decoded_image):\n decoded_image = convert_decoded(decoded_image, width, height, depth, codecname)\n original_image = convert_decoded(derivative_image, width, height, depth, 'reference')\n else:\n original_image = derivative_image\n\n print('Reference:' + original_image)\n print('Encoded:' + encoded_image)\n print('Decoded:' + decoded_image)\n if (os.path.isfile(original_image) and os.path.isfile(decoded_image) and os.path.isfile(encoded_image)):\n if 'classE' in classname:\n metrics = compute_metrics_HDR(original_image, decoded_image, encoded_image, bpp_target,\n codecname, width, height, pix_fmt, depth)\n\n elif 'classB' in classname:\n metrics = compute_metrics(original_image, decoded_image, encoded_image, bpp_target, codecname,\n width, height, pix_fmt)\n else:\n metrics = compute_metrics_SDR(original_image, decoded_image, encoded_image, bpp_target,\n codecname, width,\n height, imgfmt, depth)\n measured_bpp = (os.path.getsize(encoded_image) * 1.024 * 8) / (float((int(width) * int(height))))\n bpp_target_metrics[measured_bpp] = metrics\n else:\n continue\n \n derivative_image_metrics[codecname] = bpp_target_metrics\n main_dict[derivative_image] = derivative_image_metrics\n\n mkdir_p(json_dir)\n with open(json_file, 'w') as f:\n f.write(json.dumps(main_dict, indent=2))","def evaluation(eval_dict, faster_rcnn, test_num=Config.eval_num):\n bboxes, labels, scores = list(), list(), list()\n gt_bboxes, gt_labels, gt_difficult = list(), list(), list()\n for i, [img_dir, img_info] in tqdm(enumerate(eval_dict.items())):\n img, img_info = rescale_image(img_dir, img_info, flip=False)\n img_tensor = create_img_tensor(img)\n box, score, label = faster_rcnn.predict(img_tensor)\n gt_bbox = np.array(img_info['objects'])[:, 1:5].astype(np.float32)\n gt_label = np.array(img_info['objects'])[:, 0]\n difficult = np.array(img_info['difficult'])\n\n # label from text to number\n gt_label = text_to_num(gt_label)\n\n bboxes.append(box)\n labels.append(label)\n scores.append(score)\n gt_bboxes.append(gt_bbox)\n gt_labels.append(gt_label)\n gt_difficult.append(difficult)\n if i == test_num:\n break\n\n AP, mAP = calc_map(bboxes, labels, scores, gt_bboxes, gt_labels, gt_difficult, use_07_metric=True)\n\n return mAP","def test_calculate_recall_happy_path(\n mock_engine: DicomImagePiiVerifyEngine,\n gt_labels: dict,\n all_pos: dict,\n expected_result: float,\n):\n # Act\n test_recall = mock_engine.calculate_recall(gt_labels, all_pos)\n\n # Assert\n assert test_recall == expected_result","def compute_transform(self, image_pair, figure):\t\n\t\timage_pair = [int(image_pair[1]), int(image_pair[3])]\n\t\tindices = np.where(np.all(np.abs(self.Matches[\"image_pairs\"][0] - image_pair) == 0, axis=1)) \n\t\tfeature_pairs = self.Matches[\"image_match_pairs\"][0][indices[0][0]]\n\t\tfeature_pair_scores = self.Matches[\"image_feature_scores\"][0][indices[0][0]] \n\n\t\ti1_loc = self.Database[\"image_locs\"][np.where(np.array(self.Database[\"image_idx\"]) == image_pair[0])[0][0]]\n\t\ti2_loc = self.Database[\"image_locs\"][np.where(np.array(self.Database[\"image_idx\"]) == image_pair[1])[0][0]]\n\n\t\tself.image1 = cv2.cvtColor(extract_image(figure, i1_loc), cv2.COLOR_BGR2GRAY)\n\t\tself.image2 = cv2.cvtColor(extract_image(figure, i2_loc), cv2.COLOR_BGR2GRAY)\n\t\t# self.image1 = extract_image(figure, i1_loc)\n\t\t# self.image2 = extract_image(figure, i2_loc)\n\n\t\tself.match_alignment = {}; self.match_alignment_scores = {}\n\t\tfor pair, score in zip(feature_pairs, feature_pair_scores):\n\t\t\ti1 = self.Database[\"image_idx\"][pair[0]]\n\t\t\ti2 = self.Database[\"image_idx\"][pair[1]]\n\n\t\t\tif i1 == image_pair[0]:\n\t\t\t\ttarget = pair[0] \n\t\t\t\tstart = pair[1]\n\t\t\t\tif target not in self.match_alignment.keys():\n\t\t\t\t\tself.init_match_alignment(target)\n\t\t\telse:\n\t\t\t\ttarget = pair[1]\n\t\t\t\tstart = pair[0]\t\t\t\t\n\t\t\t\tif target not in self.match_alignment.keys():\n\t\t\t\t\tself.init_match_alignment(target)\n\n\t\t\t# get ellipse and convert to three cardinal points\n\t\t\ttarget_ellipse = self.Database[\"orientation\"][target]\n\t\t\tstart_ellipse = self.Database[\"orientation\"][start]\n\t\t\ttarget_points, target_params = utils.ellipse2points(target_ellipse)\n\t\t\tstart_points, start_params = utils.ellipse2points(start_ellipse)\n\n\t\t\t# compute angle between Major axes\n\t\t\trotation_angle = int(np.abs(target_params[0] - start_params[0]))\n\t\t\treflection_truth = target_params[1] * start_params[1]\n\t\t\tif reflection_truth > 0:\n\t\t\t\treflection_truth = \"No\"\n\t\t\telse:\n\t\t\t\treflection_truth = \"Yes\"\n\t\t\tresolution_perc = round(((target_params[2] / start_params[2]) - 1), 1)\n\n\t\t\t# extract features for histogram matching\n\t\t\ttarget_bloc = self.Database[\"blot_locs\"][target]\n\t\t\tstart_bloc = self.Database[\"blot_locs\"][start]\n\t\t\ttarget_feature = extract_image(self.image1, target_bloc, ex=5)\n\t\t\tfinal_feature = extract_image(self.image2, start_bloc, ex=5) \n\n\t\t\t#find affine transform to match\n\t\t\tif self.AFFINE:\n\t\t\t\tT = utils.compute_affine(start_points, target_points)\n\t\t\t\timage2_warped = cv2.warpAffine(self.image2.copy(), T, (self.image1.shape[1], self.image1.shape[0]))\n\t\t\t\tfinal_feature = extract_image(image2_warped, target_bloc, ex=5)\n\t\t\t\n\t\t\t# match hist\n\t\t\tif self.CONTRAST:\n\t\t\t\tfinal_feature = utils.histogram_match(final_feature, target_feature)\n\n\t\t\tself.match_alignment[target].append((final_feature, start, rotation_angle, reflection_truth, \\\n\t\t\t\t\t\t\t\t\t\t\t\t resolution_perc, score))\n\t\t\tself.match_alignment_scores[target].append(score)","def evaluate_rouge_match(ref_data, pred_data, align=\"rouge\"):\n\n scorer = RougeScorer([\"rougeL\"])\n agg = scoring.BootstrapAggregator(confidence_interval=0.95)\n sample_evaluated = 0\n\n for tid in tqdm(pred_data):\n cur_sys = pred_data[tid][\"pred\"]\n if not tid in ref_data: continue\n cur_ref_lst = [x[\"tgt_words\"] for x in ref_data[tid]]\n if len(cur_ref_lst) > 0:\n if align == \"rouge\":\n best_rouge_scrs = __rouge_pairs(cur_sys, cur_ref_lst, scorer)\n agg.add_scores(best_rouge_scrs)\n else:\n best_paired_ref, _ = utils.pair_ref_sys_by_coverage(cur_sys, cur_ref_lst)\n agg.add_scores(scorer.score(best_paired_ref, cur_sys))\n sample_evaluated += 1\n\n print(sample_evaluated)\n result = agg.aggregate()\n for rouge_type in result:\n print(rouge_type)\n print(result[rouge_type].mid)\n\n print(agg.aggregate())\n return","def all_match():\n S1=Spectrum.Spectrum()\n S1.add_peak(50.7,234)\n S1.add_peak(54.6,585)\n S1.add_peak(60.7,773)\n S1.add_peak(65.6,387)\n S1.add_peak(87.7,546)\n S1.add_peak(104.6,598)\n S1.pep_mass=100\n S1.euclidean_scale()\n\n S2=Spectrum.Spectrum()\n S2.add_peak(50.5,234/2)\n S2.add_peak(54.8,585/2)\n S2.add_peak(61.0,773/2)\n S2.add_peak(65.4,387/2)\n S2.add_peak(88.0,546/2)\n S2.add_peak(104.3,598/2)\n S2.pep_mass=100\n S2.euclidean_scale()\n\n score,peaks=similarity.cosine_score_max(S1,S2)\n assert peaks==6, \"Incorrect number of peaks matched with greedy method\"\n assert math.isclose(score,1.0), \"Incorrect score with greedy method\"\n\n score,peaks=similarity.cosine_score_greedy(S1,S2)\n assert peaks==6, \"Incorrect number of peaks matched with maximum weighted method\"\n assert math.isclose(score,1.0), \"Incorrect score with maximum weighted method\"","def find(image):\n keypoint, description = describe(image)\n # load keypoints, descriptions from mongodb\n\n bf = cv2.BFMatcher(cv2.NORM_HAMMING, crossCheck=True)\n\n best_match_size = float(\"inf\")\n best_match_index = -1\n match_index = 0\n best_matches = 0\n\n for desc in descriptions:\n matches = bf.match(desc,description)\n matches = sorted(matches, key = lambda x:x.distance)\n if len(matches) > 0:\n match_size = sum(x.distance for x in matches[:10])\n\n print \"match size is \", match_size\n if match_size < best_match_size:\n best_match_size = match_size\n best_match_index = match_index\n best_matches = matches\n\n match_index += 1\n\n needle_color = cv2.imread('needle-stripped.png')[:,:,::-1] # needle\n best_match_image = cv2.imread(\"haystack/\"+files[best_match_index])\n print \"best match is \", files[best_match_index]\n\n # Draw first 10 matches.\n outImg = cv2.imread(\"output/outImg.png\")\n match = cv2.drawMatches(needle_color,keypoint,best_match_image[:,:,::-1],keypoints[best_match_index],best_matches[-20:],outImg, flags=6)\n\n plt.imshow(match),plt.show()\n return","def img_compare(file1, file2):\n # read image\n img1 = Image.open(file1)\n img2 = Image.open(file2)\n\n # resize \n size = 128, 128\n img1_res = img_resize(img1, size)\n img2_res = img_resize(img2, size)\n\n img1_res.save(\"img_1.thumbnail\", \"JPEG\")\n img2_res.save(\"img_2.thumbnail\", \"JPEG\")\n\n # convert to gray scale\n img1_grayscale = img1_res.convert('LA')\n img1_grayscale.save(\"img_1_grayscale.png\")\n\n img2_grayscale = img2_res.convert('LA')\n img2_grayscale.save(\"img_2_grayscale.png\")\n\n # normalise\n img1_norm = normalize(np.array(img1_grayscale.getdata()).astype(float))\n img2_norm = normalize(np.array(img2_grayscale.getdata()).astype(float))\n\n try:\n # compare two images\n diff = img1_norm - img2_norm\n m_norm = sum(abs(diff)) # Manhattan norm\n z_norm = norm(diff.ravel(), 0) # Zero norm\n\n # print(\"Manhattan norm:\", m_norm, \"/ per pixel:\", m_norm/img1_norm.size)\n # print(\"Zero norm:\", z_norm, \"/ per pixel:\", z_norm*1.0/img1_norm.size)\n\n return m_norm/img1_norm.size, float(z_norm) / img1_norm.size\n except:\n return 100, 100","def match(desc1,desc2):\n\t\n\tdesc1 = array([d/linalg.norm(d) for d in desc1])\n\tdesc2 = array([d/linalg.norm(d) for d in desc2])\n\t\n\tdist_ratio = 0.6\n\tdesc1_size = desc1.shape\n\t\n\tmatchscores = zeros((desc1_size[0],1))\n\tdesc2t = desc2.T #precompute matrix transpose\n\tfor i in range(desc1_size[0]):\n\t\tdotprods = dot(desc1[i,:],desc2t) #vector of dot products\n\t\tdotprods = 0.9999*dotprods\n\t\t#inverse cosine and sort, return index for features in second image\n\t\tindx = argsort(arccos(dotprods))\n\t\t\n\t\t#check if nearest neighbor has angle less than dist_ratio times 2nd\n#\t\tif arccos(dotprods)[indx[0]] < dist_ratio * arccos(dotprods)[indx[1]]:\n\t\tmatchscores[i] = int(indx[0])\n\t\n\treturn matchscores","def drawMatches(image_1, image_1_keypoints, image_2, image_2_keypoints, matches):\n # Compute number of channels.\n num_channels = 1\n if len(image_1.shape) == 3:\n num_channels = image_1.shape[2]\n # Separation between images.\n margin = 10\n # Create an array that will fit both images (with a margin of 10 to separate\n # the two images)\n joined_image = np.zeros((max(image_1.shape[0], image_2.shape[0]),\n image_1.shape[1] + image_2.shape[1] + margin,\n 3))\n if num_channels == 1:\n for channel_idx in range(3):\n joined_image[:image_1.shape[0],\n :image_1.shape[1],\n channel_idx] = image_1\n joined_image[:image_2.shape[0],\n image_1.shape[1] + margin:,\n channel_idx] = image_2\n else:\n joined_image[:image_1.shape[0], :image_1.shape[1]] = image_1\n joined_image[:image_2.shape[0], image_1.shape[1] + margin:] = image_2\n\n for match in matches:\n image_1_point = (int(image_1_keypoints[match.queryIdx].pt[0]),\n int(image_1_keypoints[match.queryIdx].pt[1]))\n image_2_point = (int(image_2_keypoints[match.trainIdx].pt[0] + \\\n image_1.shape[1] + margin),\n int(image_2_keypoints[match.trainIdx].pt[1]))\n\n cv2.circle(joined_image, image_1_point, 5, (0, 0, 255), thickness = -1)\n cv2.circle(joined_image, image_2_point, 5, (0, 255, 0), thickness = -1)\n cv2.line(joined_image, image_1_point, image_2_point, (255, 0, 0), \\\n thickness = 3)\n return joined_image","def compare_images(self, df):\n score_list = []\n elapsed_list = []\n for index,row in df.iterrows():\n start_time = time.time()\n image1 = imread(row['image1'],as_gray=\"True\")\n image2 = imread(row['image2'],as_gray=\"True\")\n if(image1.shape != image2.shape):\n image2 = resize(image2, (image1.shape)) # if images are of different dimensions, crop the second image based on first\n score = structural_similarity(image1, image2)\n new_score = self.__convert_similarity_score(score)\n elapsed_time = (time.time() - start_time)\n score_list.append(round(new_score,2))\n elapsed_list.append(elapsed_time)\n output_df = self.__create_output_csv(df, score_list, elapsed_list)\n return output_df","def test(label_dict, image_ids, args):\n\n log_header('Starting image retrieval preparations')\n\n queries = []\n\n if args.test_image is not None:\n queries.append(args.test_image)\n else:\n logging.info('Generating random test queries')\n\n number_of_images = len(image_ids)\n\n # Generate random queries to use in the test procedure\n for i in range(args.k):\n queries.append(image_ids[random.randint(0, number_of_images - 1)])\n\n # Calculate score for the retrieved images\n calculate_score(\n label_dict, queries, retrieve_similar_images(queries, args)\n )","def drawMatches(img1, kp1, img2, kp2, matches):\n\n # Create a new output image that concatenates the two images together\n # (a.k.a) a montage\n rows1 = img1.shape[0]\n cols1 = img1.shape[1]\n rows2 = img2.shape[0]\n cols2 = img2.shape[1]\n\n out = np.zeros((max([rows1,rows2]),cols1+cols2,3), dtype='uint8')\n\n # Place the first image to the left\n out[:rows1,:cols1] = np.dstack([img1, img1, img1])\n\n # Place the next image to the right of it\n out[:rows2,cols1:] = np.dstack([img2, img2, img2])\n\n # For each pair of points we have between both images\n # draw circles, then connect a line between them\n for mat in matches:\n\n # Get the matching keypoints for each of the images\n img1_idx = mat.queryIdx\n img2_idx = mat.trainIdx\n\n # x - columns\n # y - rows\n (x1,y1) = kp1[img1_idx].pt\n (x2,y2) = kp2[img2_idx].pt\n\n # Draw a small circle at both co-ordinates\n # radius 4\n # colour blue\n # thickness = 1\n cv2.circle(out, (int(x1),int(y1)), 4, (255, 0, 0), 1) \n cv2.circle(out, (int(x2)+cols1,int(y2)), 4, (255, 0, 0), 1)\n\n # Draw a line in between the two points\n # thickness = 1\n # colour blue\n cv2.line(out, (int(x1),int(y1)), (int(x2)+cols1,int(y2)), (255, 0, 0), 1)\n\n\n # Show the image\n\n # Also return the image if you'd like a copy\n return out","def evaluate(ground_truth, prediction):\n\n def prfs_to_dict(prfs):\n \"\"\"Returns a precision_recall_fscore_support() result as a dict.\"\"\"\n return {\"precision\": prfs[0], \"recall\": prfs[1], \"fscore\": prfs[2]}\n\n results = {}\n items_count = len(ground_truth)\n\n # accuracy\n accuracy = accuracy_score(ground_truth, prediction)\n results[\"accuracy\"] = accuracy\n\n # confusion matrix\n categories = set(ground_truth) | set(prediction)\n confusions = {\n gold: {pred: 0 for pred in categories} for gold in categories\n }\n for g, p in zip(ground_truth, prediction):\n confusions[g][p] += 1\n results[\"confusions\"] = confusions\n\n # class wise precision, recall & f1\n classwise = precision_recall_fscore_support(\n ground_truth, prediction, average=None, warn_for=()\n )\n results[\"true_cat_dist\"] = list(classwise[-1])\n results[\"classwise\"] = {\n str(cl): prfs_to_dict(\n [classwise[0][cl], classwise[1][cl], classwise[2][cl]]\n )\n for cl in categories\n }\n\n # average precision, recall & f1\n results[\"macro_avg\"] = prfs_to_dict(\n precision_recall_fscore_support(\n ground_truth,\n prediction,\n average=\"macro\",\n pos_label=None,\n warn_for=(),\n )\n )\n results[\"micro_avg\"] = prfs_to_dict(\n precision_recall_fscore_support(\n ground_truth,\n prediction,\n average=\"micro\",\n pos_label=None,\n warn_for=(),\n )\n )\n results[\"weigh_avg\"] = prfs_to_dict(\n precision_recall_fscore_support(\n ground_truth,\n prediction,\n average=\"weighted\",\n pos_label=None,\n warn_for=(),\n )\n )\n\n # marginals\n gold_category_distribution = {\n g: sum([confusions[g][p] for p in categories]) for g in categories\n }\n pred_category_distribution = {\n p: sum([confusions[g][p] for g in categories]) for p in categories\n }\n\n # kappa\n expected_agreement_fleiss = sum(\n [\n (\n (gold_category_distribution[c] + pred_category_distribution[c])\n / (2.0 * items_count)\n )\n ** 2\n for c in categories\n ]\n )\n expected_agreement_cohen = sum(\n [\n (float(gold_category_distribution[c]) / items_count)\n * (float(pred_category_distribution[c]) / items_count)\n for c in categories\n ]\n )\n kappa_fleiss = (\n 1.0\n * (accuracy - expected_agreement_fleiss)\n / (1 - expected_agreement_fleiss)\n )\n kappa_cohen = (\n 1.0\n * (accuracy - expected_agreement_cohen)\n / (1 - expected_agreement_cohen)\n )\n results[\"k_fleiss\"] = {\n \"k\": kappa_fleiss,\n \"AE\": expected_agreement_fleiss,\n \"AO\": accuracy,\n }\n results[\"k_cohen\"] = {\n \"k\": kappa_cohen,\n \"AE\": expected_agreement_cohen,\n \"AO\": accuracy,\n }\n\n return results","def get_recall_precision(imgs_truth, imgs_pred, is_hit):\n # Numbers of relevant and detect boxes, respectively.\n num_rel = sum([len(img.boxes) for img in imgs_truth])\n num_det = sum([len(img.boxes) for img in imgs_pred])\n\n boxes_for_fname_truth = {img.fname: img.boxes for img in imgs_truth}\n boxes_for_fname_pred = {img.fname: img.boxes for img in imgs_pred}\n\n num_hit = 0\n for fname, boxes in boxes_for_fname_truth.items():\n if fname not in boxes_for_fname_pred:\n continue\n\n num_hit += get_num_hit(boxes, boxes_for_fname_pred[fname], is_hit)\n\n return num_hit / num_rel, num_hit / num_det","def compute(self, result_file_dict):\r\n for part in self.parts:\r\n #=====================Need to change, temporal=========================\r\n if part == 'train':\r\n continue # because the train not have the label\r\n #=======================================================================\r\n gt = self.gt_dict[part]\r\n result_file = result_file_dict[part]\r\n # import ipdb; ipdb.set_trace()\r\n for key, item in result_file.items():\r\n self._result_name = item\r\n # score_records, num_videos = self.load_results(result_file)\r\n score_records, num_videos = self.load_results(item)\r\n logger.info(f'Compute Metric of {item}')\r\n assert num_videos == len(gt), f'the number of saved videos does not match the ground truth, {num_videos} != {len(gt)}'\r\n temp_result = self.eval_method(score_records, gt, str(key))\r\n if temp_result > self.optimal_resulst:\r\n self.optimal_resulst = temp_result\r\n \r\n return self.optimal_resulst","def perform_match(self, parser_manager):\n log = logging.getLogger(\".\".join([__name__, self.__class__.__name__]))\n log.addFilter(logconfig.ThreadContextFilter())\n extra = self._config_align.all_to_json()\n extra.update(self._config_crystal.all_to_json())\n log = logging.LoggerAdapter(log, extra)\n log.info(\"Matching Started\")\n log.debug(extra)\n\n input_poi = parser_manager.parse_selected_points_from_args()\n beamline_image = parser_manager.get_focused_image()\n parser_manager.save_focused_image(beamline_image)\n focused_image_path = parser_manager.get_focused_image_path()\n formulatrix_image_path = parser_manager.get_formulatrix_image_path()\n job_id = parser_manager.get_job_id()\n\n # Create the images\n image1 = Image.from_file(formulatrix_image_path)\n image2 = beamline_image\n\n # Create results object\n service_result = ServiceResult(job_id, formulatrix_image_path, focused_image_path)\n\n # Perform alignment\n try:\n\n aligned_images, scaled_poi = self._perform_alignment(image1, image2, input_poi)\n service_result.set_image_alignment_results(aligned_images)\n\n # Perform Crystal Matching - only proceed if we have a valid alignment\n if aligned_images.alignment_status_code() == ALIGNED_IMAGE_STATUS_OK:\n match_results = self._perform_matching(aligned_images, scaled_poi, parser_manager)\n\n service_result.append_crystal_matching_results(match_results)\n\n except Exception as e:\n if sys.version_info[0] < 3:\n log.error(\"ERROR: \" + e.message)\n else:\n log.error(\"ERROR: \" + str(e))\n service_result.set_err_state(e)\n\n return service_result","def ransac(keypoints1, keypoints2, matches, sampling_ratio=0.5, n_iters=500, threshold=20):\n N = matches.shape[0]\n n_samples = int(N * sampling_ratio)\n\n # Please note that coordinates are in the format (y, x)\n matched1 = pad(keypoints1[matches[:,0]])\n matched2 = pad(keypoints2[matches[:,1]])\n matched1_unpad = keypoints1[matches[:,0]]\n matched2_unpad = keypoints2[matches[:,1]]\n\n max_inliers = np.zeros(N)\n n_inliers = 0\n\n # RANSAC iteration start\n ### YOUR CODE HERE\n raise NotImplementedError() # Delete this line\n ### END YOUR CODE\n return H, matches[max_inliers]","def pixel_metrics(truth: Node, prediction: Node) -> Tuple[float, float, float]:\n recall, precision, fscore = 0, 0, 0\n\n intersection_truth = bounding_box_intersection(truth.bounding_box,\n prediction.bounding_box)\n if intersection_truth is None:\n logging.debug('No intersection for Nodes: t={0},'\n ' p={1}'.format(truth.bounding_box,\n prediction.bounding_box))\n return recall, precision, fscore\n\n intersection_pred = bounding_box_intersection(prediction.bounding_box,\n truth.bounding_box)\n\n logging.debug('Found intersection for Nodes: t={0},'\n ' p={1}'.format(truth.bounding_box,\n prediction.bounding_box))\n\n tt, tl, tb, tr = intersection_truth\n pt, pl, pb, pr = intersection_pred\n crop_truth = truth.mask[tt:tb, tl:tr]\n crop_pred = prediction.mask[pt:pb, pl:pr]\n\n # Assumes the mask values are 1...\n\n n_truth = float(truth.mask.sum())\n n_pred = float(prediction.mask.sum())\n n_common = float((crop_truth * crop_pred).sum())\n\n # There are no zero-pixel objects, but the overlap may be nonzero\n if n_truth == 0:\n recall = 0.0\n precision = 0.0\n elif n_pred == 0:\n recall = 0.0\n precision = 0.0\n else:\n recall = n_common / n_truth\n precision = n_common / n_pred\n\n if (recall == 0) or (precision == 0):\n fscore = 0\n else:\n fscore = 2 * recall * precision / (recall + precision)\n\n return recall, precision, fscore","def calc_accuracy(self, top_matches):\n # Calculate Translation accuracy\n # Accuracy at K\n for k in [1, 5, 10]:\n top_k_matches = top_matches[:, :k]\n # Checks for one match\n one_matching = (top_k_matches == self.tgt_ind[:, None]).sum(1)\n # Checks for multiple translations\n matching = {}\n for i, src_id in enumerate(self.src_ind):\n matching[src_id] = min(matching.get(src_id, 0) + one_matching[i], 1)\n precision_at_k = 100 * np.mean(list(matching.values()))\n print(\"%i source words - Precision at k = %i: %f\" % (len(matching), k, precision_at_k))","def run_compare(self):\n misses = []\n for row in range(self.data.get_number_of_rows()):\n ref_sentence = self.data.get_row(row)[self.data.get_gold()]\n results = {}\n for team, team_sentence in self.data.get_row_for_teams(self.evaluator.get_teams(row), row).iteritems():\n results[team] = self.get_sentence_score(ref_sentence, team_sentence)\n misses.append(self.evaluator.compare_all(results, row))\n print np.median(misses), np.mean(misses)\n return np.median(misses)","def compute_similarity(im1, im2, test_id):\n\n results = dict()\n results['id'] = test_id\n results['test_im'] = im2\n results['mse_value'], results['mse_map'] = compute_mse(im1, im2)\n results['ssim_value'], results['ssim_map'] = compute_ssim(im1, im2, 5)\n results['cw_ssim_value'], results['cw_ssim_map'] = compute_cw_ssim(im1, im2, 30)\n results['gms_value'], results['gms_map'] = compute_gms(im1, im2)\n results['fsim_value'], results['pc_max_map'] = compute_fsim(im1, im2)\n\n return results","def evaluate(self, test_dir='data/dev', target='real'):\n test_data = {c: os.path.join(test_dir, c) for c in self.classes}\n if not target in test_data:\n print('Error: target class does not exist in test data.')\n return\n outcomes = {'TP': 0, 'TN': 0, 'FP': 0, 'FN': 0}\n # >>> YOUR ANSWER HERE\n data = []\n for c in test_data:\n docs = open(test_data[c]).readlines()\n for doc in docs:\n preprocess_doc = doc.strip()\n data.append((c, preprocess_doc))\n for item in data:\n predict_ans = self.predict(item[1])\n if item[0] == 'real':\n if predict_ans == 'real':\n outcomes['TP'] += 1\n else:\n outcomes['FN'] += 1\n else:\n if predict_ans == 'real':\n outcomes['FP'] += 1\n else:\n outcomes['TN'] += 1\n precision = outcomes['TP'] / (outcomes['TP'] + outcomes['FP']) # replace with equation for precision\n recall = outcomes['TP'] / (outcomes['TP'] + outcomes['FN']) # replace with equation for recall\n f1_score = 2 * ((precision * recall) / (precision + recall)) # replace with equation for f1\n # >>> END YOUR ANSWER\n return precision, recall, f1_score","def _calculate_similarity(self):\n self._logger.info(\"Calculating the similarity between images.\")\n\n # Create a helper function to simplify the loops below:\n def get_wrapper(fdx, mdx):\n wrapper = pos_wrappers.image_similarity_wrapper(\n reference_image=self.f['src_gray'](idx=fdx),\n moving_image=self.f['src_gray'](idx=mdx),\n affine_transformation=self.f['part_transf'](mIdx=mdx, fIdx=fdx))\n return copy.copy(wrapper)\n\n commands = [] # Will hold commands for calculating the similarity\n\n # Will hold (moving, fixed) images partial_transforms basically: all\n # partial transformations array\n partial_transforms = []\n\n self._logger.debug(\"Generating similarity measure warppers.\")\n for moving_slice in self.options.slice_range:\n # Get all fixed images to which given moving slice will be aligned:\n tpair = list(flatten(self._get_slice_pair(moving_slice)))\n\n # Append partial transformations for given moving slice to the\n # global partial transformations array\n partial_transforms.append(tpair)\n\n # Generate wrapper for measuring similarity for a given partial\n # transformation.\n for mdx, fdx in tpair:\n commands.append(get_wrapper(fdx, mdx))\n\n # Execute and commands and workflow the similarity measurements.\n stdout, stderr = self.execute(commands)\n simmilarity = map(lambda x: float(x.strip()),\n stdout.strip().split(\"\\n\"))\n simmilarity = dict(zip(flatten(partial_transforms), simmilarity))\n\n self._logger.debug(\"Generating graph edges.\")\n graph_connections = []\n\n # Lambda defines slice skipping is preffered (lower l), or reluctant\n # to slice skipping (higher)\n l = self.options.graphEdgeLambda\n\n for (mdx, fdx), s in simmilarity.iteritems():\n w = (1.0 + s) * abs(mdx - fdx) * (1.0 + l) ** (abs(mdx - fdx))\n graph_connections.append((fdx, mdx, w))\n\n self._logger.info(\"Creating a graph based on image similarities.\")\n # Generate the graph basen on the weight of the edges\n self.G = nx.DiGraph()\n self.G.add_weighted_edges_from(graph_connections)\n\n self._logger.debug(\"Saving the graph to a file.\")\n # Save the edges for some further analysis.\n nx.write_weighted_edgelist(self.G,\n self.f['graph_edges'](sign=self.signature))\n\n # Also, save the individual similarity metrics:\n simm_fh = open(self.f['similarity'](sign=self.signature), 'w')\n for (mdx, fdx), s in sorted(simmilarity.iteritems()):\n simm_fh.write(\"%d %d %f\\n\" % (mdx, fdx, s))\n simm_fh.close()","def versus(bot1, bot2, num_matches, cpu_count, seed_list):\n W, L, D = 0, 0, 0\n tot_scores = [0, 0]\n avg_scores = [0, 0]\n all_scores = []\n show_match_results = 1\n\n pool = multiprocessing.Pool(cpu_count)\n\n try:\n bot1name = os.path.splitext(os.path.basename(bot1))[0]\n bot2name = os.path.splitext(os.path.basename(bot2))[0]\n start_time = time.time()\n results = [pool.apply_async(run_match, (bot1, bot2, seed_list[i]))\n for i in range(num_matches)]\n \n if show_match_results:\n sys.stdout.write('%20s - %3i: [' % (bot2name, num_matches))\n\n tenths_done = 0\n for r in results:\n scores = r.get(timeout=50)\n if scores[0] > scores[1]:\n W += 1\n elif scores[0] < scores[1]:\n L += 1\n elif scores[0] == scores[1]:\n D += 1\n tot_scores = [tot_scores[x]+scores[x] for x in range(2)]\n avg_scores = [tot_scores[x]//sum([W,L,D]) for x in range(2)]\n all_scores.append(scores[0:2])\n if (sum([W,L,D])*10//num_matches) > tenths_done:\n tenths_done += 1\n if show_match_results:\n sys.stdout.write ('.')\n\n pool.close()\n pool.join()\n \n if show_match_results:\n diff = [s[0] - s[1] for s in all_scores]\n avg, std = meanstd(diff)\n sys.stdout.write (' '*(10-tenths_done))\n print '] - %12s by %8s diff %5.1f std %4.1f in %5.1fs' % (str([W,L,D]), str(avg_scores), avg, std, time.time()-start_time)\n except multiprocessing.TimeoutError:\n if show_match_results:\n sys.stdout.write ('x'*(10-tenths_done))\n print '] - %12s by %8s in %5.1fs' % (str([W,L,D]), str(avg_scores), time.time()-start_time)\n pool.terminate()\n\n return [W,L,D], tot_scores, all_scores","def match_percentage(image1_pixels, image2_pixels):\n\n match, total = 0, 0\n for i in range(len(image1_pixels)):\n if image1_pixels[i] == image2_pixels[i]:\n match += 1\n total += 1\n else:\n total += 1\n return float(match) / float(total)","def compute_matches(self):\n\t\tself.local_database[\"figure_number\"] = [0] * len(self.local_database[\"feature_vectors\"])\n\t\tForensics = wbForensicsHOG(Database=self.local_database)\n\t\tForensics.KDTree_pairs(leaf_size = len(self.local_database)+1)\n\t\tForensics.d_rank(pairs=Forensics.pairs, distances=Forensics.dists, ratios=Forensics.ratios)\n\n\t\tself.local_matches = Forensics.Dist_Rank","def detection_analysis(y_pred, y_true):\n print(\"Precision: \", sm.precision_score(y_pred, y_true))\n print(\"Recall: \", sm.recall_score(y_pred, y_true))\n print(\"Accuracy: \", sm.accuracy_score(y_pred, y_true))\n print(\"\\n\")","def get_recall_and_precision(wdb, dets, th_conf, th_iou):\n hit_total, false_total, miss_total = 0, 0, 0\n\n for idx in range(wdb.get_image_count()):\n data = wdb.get_annos_by_image_index(idx)\n img_path = data['image_path']\n fname = os.path.splitext(os.path.basename(img_path))[0]\n\n dets4fname = []\n\n for d in dets:\n if d['filename'] == fname: # we have only ONE corresponding element\n dets4fname = d['data']\n break\n\n dets2compare = [e for e in dets4fname if e['c'] >= th_conf] # detections having confidence larger than threshold only\n\n hit, false, miss = count_hits(data['annos'], dets2compare, th_iou)\n hit_total += hit\n false_total += false\n miss_total += miss\n\n recall = hit_total/float(hit_total+miss_total+0.000001)\n precision = hit_total/float(hit_total+false_total+0.000001)\n\n return recall, precision","def compare_results(self, benchmark_result, reference_benchmark_result):\n pass","def compare(image_a, image_b, is_camera_image):\n\n # Generate a unique filename\n filename = uuid.uuid4().hex[:3]\n\n if is_camera_image:\n image_a = imutils.rotate_bound(image_a, 90)\n image_b = imutils.rotate_bound(image_b, 90)\n\n # Store original to show in future\n original = image_a\n\n # Convert to greyscale\n image_a = cv2.cvtColor(image_a, cv2.COLOR_BGR2GRAY)\n image_b = cv2.cvtColor(image_b, cv2.COLOR_BGR2GRAY)\n\n # Reduce size and blur to account for shaky handheld camera based images\n if is_camera_image:\n scale_multiplier = 0.03125\n image_a = cv2.resize(image_a, (0, 0), fx=scale_multiplier, fy=scale_multiplier)\n image_b = cv2.resize(image_b, (0, 0), fx=scale_multiplier, fy=scale_multiplier)\n image_a = cv2.GaussianBlur(image_a, (1001, 1001), cv2.BORDER_DEFAULT)\n image_b = cv2.GaussianBlur(image_b, (1001, 1001), cv2.BORDER_DEFAULT)\n\n # Obtain SSIM and determine differences\n try:\n _, differences = structural_similarity(image_a, image_b, full=True, gaussian_weights=True)\n except ValueError:\n print('Images are not the same size')\n return None\n\n # Convert to cv2 array\n differences = (differences * 255).astype('uint8')\n\n # Threshold and find contours (differences)\n thresh = cv2.threshold(differences, 0, 255, cv2.THRESH_BINARY_INV | cv2.THRESH_OTSU)[1]\n contours = cv2.findContours(thresh, cv2.RETR_EXTERNAL, cv2.CHAIN_APPROX_SIMPLE)\n contours = imutils.grab_contours(contours)\n\n # Draw contours (differences)\n for cont in contours:\n (x, y, w, h) = cv2.boundingRect(cont)\n if is_camera_image:\n multiplier = int(1 / scale_multiplier)\n y *= multiplier\n x *= multiplier\n h *= multiplier\n w *= multiplier\n cv2.rectangle(original, (x, y), (x + w, y + h), (255, 0, 0), 4)\n\n # TODO: Create GIF highlighting differences (instead of statuic image)\n cv2.imwrite('static/images/differences/' + filename + '.jpg', original)\n\n return filename","def run_multiple_simple_experiments(self, path_to_csv_files, algorithms, thresholds):\n starting_time = time.time()\n print(\"started simple precision recall image_matching_module\")\n for algorithm in algorithms:\n start_file_time = time.time()\n print(\"started algorithm:\", algorithm.get_algorithm_name())\n self.run_single_simple_experiment(path_to_csv_files, algorithm, thresholds)\n finish_file_time = time.time()\n print(\"finished algorithm:\", algorithm.get_algorithm_name())\n print(\"time took:\", (finish_file_time - start_file_time) / 60, \"minutes\")\n finish_time = time.time()\n hours = int((finish_time - starting_time) // 3600)\n minutes = (finish_time - starting_time) / 60\n print(\"finished simple precision recall image_matching_module\")\n print(\"time image_matching_module took:\", hours, \"hours and\", minutes, \"minutes\")","def return_MatchUpTest_r__():\n\n ####################################################################################################################\n # 1. Initialise test data\n ####################################################################################################################\n\n values = array([470.5, 720.56, 450.9, 295.6, 315.23,\n 70.5, 70.6, 70.3, 70.7, 70.5,\n 71.5, 71.6, 71.3, 71.7,\n 80.5, 80.6, 80.3, 80.7,\n 150.5, 151.1, 149.8, 150.2, 151.4,\n 140.5, 141.1, 139.8, 140.2,\n 160.5, 161.1, 169.8, 160.2,\n 30.2, 20.4, 28.2, 50.7, 45.6,\n 29.2, 37.4, 28.2, 50.7,\n 28.2, 32.4, 22.2, 53.7, ])\n unc = [Uncertainty(1, array([1.6, 1.5, 1.5, 1.3, 1.5])),\n Uncertainty(1, array([3.1, 3.2, 3.2, 3.1, 3.0])),\n Uncertainty(1, array([3.3, 3.4, 3.1, 3.2])),\n Uncertainty(1, array([2.1, 2.2, 2.2, 2.1])),\n Uncertainty(1, array([5.0, 4.7, 5.1, 5.2, 5.3])),\n Uncertainty(1, array([4.2, 4.3, 4.4, 4.3])),\n Uncertainty(1, array([4.0, 3.7, 4.4, 4.7])),\n Uncertainty(1, array([2.2, 1.7, 2.0, 4.3, 2.6])),\n Uncertainty(1, array([2.3, 1.2, 2.3, 4.4])),\n Uncertainty(1, array([3.2, 2.7, 3.0, 5.3]))]\n ks = array([1.2, 1.7, 1.3, 1.4, 1.3, 3.2, 3.7, 3.3, 3.4])\n unck = [Uncertainty(1, array([0.25, 0.25, 0.25, 0.25, 0.25])),\n Uncertainty(1, array([0.2644, 0.2644, 0.2644, 0.2644]))]\n idx = {\"Nm\": [5, 4],\n \"cNm\": [0, 5, 9],\n \"Im\": [[0, 1], [1, 2]],\n \"sensors\": [-1, 1, 2],\n \"sensor_ms\": [1, 3, 3],\n \"n_sensor\": [0, 1, 1, 2, 1, 1, 2, 1, 1, 2],\n \"n_mu\": [1, 1, 2, 2, 1, 2, 2, 1, 2, 2],\n \"n_cov\": [1, 1, 1, 1, 2, 2, 2, 3, 3, 3],\n \"N_var\": [5, 5, 4, 4, 5, 4, 4, 5, 4, 4],\n \"idx\": [0, 5, 10, 14, 18, 23, 27, 31, 36, 40, 44],\n \"Ia\": [1, 1, 1, 2, 2, 2]}\n a = array([1., 1.3, 0.002, 0.5, 1.1, 0.0005])\n\n ####################################################################################################################\n # 3. Initialise MatchUp object\n ####################################################################################################################\n\n MatchUpTest = MatchUp()\n MatchUpTest.values = values\n MatchUpTest.unc = unc\n MatchUpTest.ks = ks\n MatchUpTest.unck = unck\n MatchUpTest.idx = idx\n MatchUpTest.a = a\n\n return MatchUpTest","def test_compare_classifiers_predictions_vis_api(experiment_to_use):\n experiment = experiment_to_use\n predictions = experiment.predictions\n viz_outputs = (\"pdf\", \"png\")\n with TemporaryDirectory() as tmpvizdir:\n for viz_output in viz_outputs:\n vis_output_pattern_pdf = tmpvizdir + f\"/*.{viz_output}\"\n visualize.compare_classifiers_predictions(\n [predictions, predictions],\n experiment.ground_truth,\n experiment.ground_truth_metadata,\n experiment.output_feature_name,\n labels_limit=0,\n model_names=[\"Model1\", \"Model2\"],\n output_directory=tmpvizdir,\n file_format=viz_output,\n )\n figure_cnt = glob.glob(vis_output_pattern_pdf)\n assert 1 == len(figure_cnt)","def run_multiple_advanced_experiments(self, path_to_csv_files, algorithms_weights_list, thresholds):\n starting_time = time.time()\n print(\"started advanced precision recall image_matching_module\")\n for algorithms_weights in algorithms_weights_list:\n start_file_time = time.time()\n\n first_run = True\n for algorithm, weight in algorithms_weights:\n if first_run:\n algorithm_name = algorithm.get_algorithm_name() + weight\n first_run = False\n else:\n algorithm_name = algorithm_name + \"_\" + algorithm.get_algorithm_name() + weight\n\n print(\"started algorithm:\", algorithm_name)\n self.run_single_advanced_experiment(path_to_csv_files, algorithms_weights, thresholds)\n finish_file_time = time.time()\n print(\"finished algorithm:\", algorithm_name)\n print(\"time took:\", (finish_file_time - start_file_time) / 60, \"minutes\")\n finish_time = time.time()\n hours = int((finish_time - starting_time) // 3600)\n minutes = (finish_time - starting_time) / 60\n print(\"finished advanced precision recall image_matching_module\")\n print(\"time image_matching_module took:\", hours, \"hours and\", minutes, \"minutes\")","def plot_images(img_A, img_B, match_A, match_B, name=\"p1\", method=\"greedy\"):\n (hA, wA) = img_A.shape[:2]\n (hB, wB) = img_B.shape[:2]\n vis = np.zeros((max(hA, hB), wA + wB, 3), dtype=\"uint8\")\n vis[0:hA, 0:wA] = img_A\n vis[0:hB, wA:] = img_B\n\n # loop over the matches\n for ptA, ptB in zip(match_A, match_B):\n ptB = ptB + np.array([wA, 0])\n cv2.line(vis, tuple(ptA), tuple(ptB), (0, 255, 0), 1)\n\n cv2.imwrite(\"output/sift_\" + method + \"_\" + name + \".png\", vis)\n cv2.imshow(\"Keypoint matching\", vis)\n cv2.waitKey(0)","def compare_predictions():\n validation_labels = np.array(pd.read_csv(val_true_labels_dir + dataset_version + 'validation_labels.csv', index_col=0))\n validation_labels = np.reshape(validation_labels, (-1))\n\n diff_between_files = []\n also1s = []\n also2s = []\n for filename1 in os.listdir(val_predictions_dir):\n if filename1.endswith(\".csv\"):\n for filename2 in os.listdir(val_predictions_dir):\n if filename2.endswith(\".csv\"):\n if filename1 < filename2:\n wrong1 = 0\n wrong2 = 0\n diff_between = 0\n also1 = 0\n also2 = 0\n diff_corr1 = 0\n diff_corr2 = 0\n f1 = np.array(pd.read_csv(val_predictions_dir + filename1, index_col=0))\n f1 = np.reshape(f1, (-1))\n f2 = np.array(pd.read_csv(val_predictions_dir + filename2, index_col=0))\n f2 = np.reshape(f2, (-1))\n for line in range(f1.shape[0]):\n if f1[line] != validation_labels[line]:\n wrong1 += 1\n if f2[line] != validation_labels[line]:\n wrong2 += 1\n if f1[line] != f2[line]:\n diff_between += 1\n if f1[line] == validation_labels[line]:\n diff_corr1 += 1\n if f2[line] == validation_labels[line]:\n diff_corr2 += 1\n if f1[line] != validation_labels[line]:\n if f2[line] != validation_labels[line]:\n also2 += 1\n if f2[line] != validation_labels[line]:\n if f1[line] != validation_labels[line]:\n also1 += 1\n\n diff_between_files.append(diff_between)\n print(filename1)\n print('Wrongly predicted by 1: ' + str(100 * wrong1 / f1.shape[0]) + '%')\n print(filename2)\n print('Wrongly predicted by 2: ' + str(100 * wrong2 / f1.shape[0]) + '%')\n print()\n print('Differences between files: ' + str(100 * diff_between / f1.shape[0]) + '%')\n print(f'\\t of which correct by 1 {100 * diff_corr1 / diff_between}%, by 2 {100 * diff_corr2 / diff_between}%')\n also1s.append(also1 / wrong2)\n also2s.append(also2 / wrong1)\n print('Wrongly predicted by other among wrong ones: ' + str(100 * also2 / wrong1) + '%, ' + str(\n 100 * also1 / wrong2) + '%\\n\\n\\n')\n\n print('Max, min and avg differences between files:')\n print(str(100 * max(diff_between_files) / validation_labels.shape[0]) + '%')\n print(str(100 * min(diff_between_files) / validation_labels.shape[0]) + '%')\n print(str(100 * np.mean(diff_between_files) / validation_labels.shape[0]) + '%')\n\n print('\\nWrongly predicted by first that were also wrongly predicted by second:')\n print('Max: ' + str(100 * max(also2s)) + '%')\n print('Min: ' + str(100 * min(also2s)) + '%')\n print('Avg: ' + str(100 * np.mean(also2s)) + '%')\n\n print('\\nWrongly predicted by second that were also wrongly predicted by first:')\n print('Max: ' + str(100 * max(also1s)) + '%')\n print('Min: ' + str(100 * min(also1s)) + '%')\n print('Avg: ' + str(100 * np.mean(also1s)) + '%')","def test_compare_single_and_multiprocess_results(sidesweep_image_sequence):\n cc = Cwsim_container_from_ims(ims=sidesweep_image_sequence)\n test_idx = 20\n results_single_proc = cc.self_im_query_all(test_idx, plot_output=False)\n\n test_im = sidesweep_image_sequence[test_idx]\n cc.prepare_memory_bank_outside()\n results_multi_proc = cc.query_image_mp(test_im)\n\n plt.plot(results_single_proc, marker='D', label='single_proc')\n plt.plot(results_multi_proc, marker='D', label='multi_proc')\n plt.legend()\n plt.draw()\n plt.show(block=False)\n plt.pause(3)\n\n assert np.allclose(results_single_proc, results_multi_proc)","def evaluate(calculated_results, ground_truth, measure=''):\n score = 0\n truth_pairs = dict.fromkeys(pair[0] + ' ' + pair[1]\n for pair in ground_truth.edges())\n for pair in calculated_results:\n pair_reversed = get_reversed_pair(pair)\n if pair[0] in truth_pairs or pair_reversed in truth_pairs:\n score += 1\n accuracy = (score / len(truth_pairs)) * 100\n print(\"The accuracy is %f %% when using measure: %s\" % (accuracy, measure))","def evaluate(groundtruth, prediction, tiou_threshold=0.5):\n\n video_num = len(groundtruth)\n print('Computing grounding accuracy over {} videos...'.format(video_num))\n acc, acc_sub, acc_obj = 0.0, 0.0, 0.0\n\n gt_rnum = 0\n for qid, relation_gt in groundtruth.items():\n\n if qid not in prediction:\n continue\n relation_pred = prediction[qid]\n if len(relation_pred) == 0:\n continue\n\n video_acc, video_acc_sub, video_acc_obj, relation_num = eval_ground_scores(relation_gt, relation_pred, tiou_threshold)\n\n acc += video_acc\n acc_sub += video_acc_sub\n acc_obj += video_acc_obj\n gt_rnum += relation_num\n\n\n acc /= video_num\n acc_sub /= video_num\n acc_obj /= video_num\n\n print(\"Acc_S\\t Acc_O\\t Acc_R\")\n\n print('{:.2f}\\t {:.2f}\\t {:.2f}'.format(acc_sub*100, acc_obj*100, acc*100))","def test_quality(approach):\n tp, fn, fp, tn = 0, 0, 0, 0\n with open(\"test.txt\", \"r\") as f:\n test_data = json.load(f)\n for i in test_data:\n rez = approach(i[0][0], i[0][1])\n if rez == 1 and i[1] == 1:\n tp += 1\n elif rez == 0 and i[1] == 1:\n fn += 1\n elif rez == 1 and i[1] == 0:\n fp += 1\n else:\n tn += 1\n print \"Precision: {}%.\\nRecall: {}%.\".format(\n round(tp / (tp + fp), 2), round(tp / (tp + fn), 2))","def calculate_metrics(predictions, expected):\n # type: (np.ndarray, np.ndarray) -> (float, float, float)\n clients_count = predictions.shape[0]\n products_count = predictions.shape[1]\n\n true_positive = 0.0\n true_negative = 0.0\n false_positive = 0.0\n false_negative = 0.0\n\n total = float(clients_count * products_count)\n\n for c in range(0, clients_count):\n for p in range(0, products_count):\n if predictions[c, p] == expected[c, p]:\n if predictions[c, p] == 1:\n true_positive += 1\n else:\n true_negative += 1\n else:\n if predictions[c, p] == 1:\n false_positive += 1\n else:\n false_negative += 1\n\n accuracy = float(true_positive + true_negative) / total\n if true_positive + false_positive == 0:\n precision = 0\n else:\n precision = true_positive / float(true_positive + false_positive)\n\n if true_positive + false_negative == 0:\n recall = 0\n else:\n recall = true_positive / float(true_positive + false_negative)\n\n return accuracy, precision, recall","def evaluate(self):\n tic = time.time()\n print('Running per image evaluation...')\n p = self.params\n # add backward compatibility if useSegm is specified in params\n if p.useSegm is not None:\n p.iouType = 'segm' if p.useSegm == 1 else 'bbox'\n print(\n \"useSegm (deprecated) is not None.\"\n \"Running {} evaluation\".format(p.iouType)\n )\n print('Evaluate annotation type *{}*'.format(p.iouType))\n p.imgIds = list(np.unique(p.imgIds))\n if p.useCats:\n p.catIds = list(np.unique(p.catIds))\n p.maxDets = sorted(p.maxDets)\n self.params = p\n\n self._prepare()\n # loop through images, area range, max detection number\n catIds = p.catIds if p.useCats else [-1]\n\n self.ious = {\n (imgId, catId): self.computeIoU(imgId, catId)\n for imgId in p.imgIds\n for catId in catIds\n }\n # ignore any categories that is not having any attributes\n self.f1s = {\n (imgId, catId): self.computeF1(imgId, catId)\n for imgId in p.imgIds\n for catId in catIds\n if catId in self.FPParams.catsWithAttributes\n }\n\n # self.gt_attributes_ids = []\n\n # loop through images, area range, max detection number\n self.evalImgs = [\n self.evaluateImg(imgId, catId, areaRng, p.maxDets[-1])\n for catId in catIds\n for areaRng in p.areaRng\n for imgId in p.imgIds\n ]\n # self._paramsEval = copy.deepcopy(self.params) # seems do not need it\n toc = time.time()\n print('DONE (t={:0.2f}s).'.format(toc-tic))","def ransac_homography(keypoints1, keypoints2, matches, sampling_ratio=0.5, n_iters=500, delta=20):\n N = matches.shape[0]\n n_samples = int(N * sampling_ratio)\n\n matched1_unpad = keypoints1[matches[:,0]]\n matched2_unpad = keypoints2[matches[:,1]]\n\n\n max_inliers = np.zeros(N)\n n_inliers = 0\n\n # RANSAC iteration start\n ### YOUR CODE HERE\n for i in range(n_iters):\n random_sample = np.random.choice(N, n_samples)\n src = matched1_unpad[random_sample] # leave unpadded, is padded in compute homography\n dst = matched2_unpad[random_sample]\n\n # Get H using DLT\n H_matrix = compute_homography(src, dst)\n\n # Matched keypoints are considered inliers if the projected point from one image lies within distance δ to the matched point on the other point.\n\n projection = transform_homography(matched1_unpad, H_matrix) # x point, y point\n distances = np.sqrt(np.sum(np.square(projection-matched2_unpad), axis=1))\n if np.sum(distances < delta) > n_inliers:\n max_inliers = distances < delta\n n_inliers = np.sum(distances < delta)\n \n # Get final H using maximum inliers indices and src dst values\n src = matched1_unpad[max_inliers]\n dst = matched2_unpad[max_inliers]\n H = compute_homography(src, dst)\n \n ### END YOUR CODE\n return H, matches[max_inliers]","def PredictMatch(yourName, tgtName, teamModels, db):\n\n teamTgtPipe, teamYourPipe = QueryTeamData(tgtName, yourName, db)\n featureCoefTgt, probTgt = PredictOp(teamTgtPipe, teamYourPipe, tgtName, teamModels)\n featureCoefYour, probYour = PredictOp(teamYourPipe, teamTgtPipe, yourName, teamModels)\n odds = round(probYour / (probYour + probTgt), 2)\n\n # In featureCoefYour, you want INCREASE those with POSTIVE COEF, DECREASE those with NEGATIVE COEF\n # In featureCoefTgt, you want to do the opposite\n\n # reverse both the sign of the coef, and '_op' in features so as to be the same with featureCoefYour\n featureCoefTgt['coef'] = - featureCoefTgt['coef']\n featureCoefTgt.features = [ii[:-3] if \"_op\" in ii else ii + '_op' for ii in featureCoefTgt.features]\n\n # Combine only the most important 10 features\n # featureBoth = featureCoefTgt[11:].append(featureCoefYour[11:])\n\n # Combine only all the most important features\n featureBoth = featureCoefTgt.append(featureCoefYour)\n\n # get action recommendations\n # Somehow the pandas here uses a deprecated para cols, instaed of the new one subset\n #featureBoth.drop_duplicates(subset = 'features', take_last = True, inplace = True)\n featureBoth.drop_duplicates(cols = 'features', take_last = True, inplace = True)\n actions, featureImprove = GetActions(featureBoth)\n Imp = 0.1\n oddsNew = ImprovedScore(tgtName, yourName, teamModels, featureImprove, teamTgtPipe, teamYourPipe, Imp)\n\n return odds, oddsNew, actions","def test_evaluate(self):\n # test normalized by 'bbox_size'\n jhmdb_pck_metric = JhmdbPCKAccuracy(thr=0.5, norm_item='bbox')\n jhmdb_pck_metric.process(self.data_batch, self.data_samples)\n pck_results = jhmdb_pck_metric.evaluate(self.batch_size)\n target = {\n 'Head PCK': 1.0,\n 'Sho PCK': 1.0,\n 'Elb PCK': 1.0,\n 'Wri PCK': 1.0,\n 'Hip PCK': 1.0,\n 'Knee PCK': 1.0,\n 'Ank PCK': 1.0,\n 'PCK': 1.0,\n }\n self.assertDictEqual(pck_results, target)\n\n # test normalized by 'torso_size'\n jhmdb_tpck_metric = JhmdbPCKAccuracy(thr=0.2, norm_item='torso')\n jhmdb_tpck_metric.process(self.data_batch, self.data_samples)\n tpck_results = jhmdb_tpck_metric.evaluate(self.batch_size)\n target = {\n 'Head tPCK': 1.0,\n 'Sho tPCK': 1.0,\n 'Elb tPCK': 1.0,\n 'Wri tPCK': 1.0,\n 'Hip tPCK': 1.0,\n 'Knee tPCK': 1.0,\n 'Ank tPCK': 1.0,\n 'tPCK': 1.0,\n }\n self.assertDictEqual(tpck_results, target)","def test_compare_classifiers_performance_changing_k_vis_api(experiment_to_use):\n experiment = experiment_to_use\n probabilities = experiment.probabilities\n viz_outputs = (\"pdf\", \"png\")\n with TemporaryDirectory() as tmpvizdir:\n for viz_output in viz_outputs:\n vis_output_pattern_pdf = tmpvizdir + f\"/*.{viz_output}\"\n visualize.compare_classifiers_performance_changing_k(\n [probabilities, probabilities],\n experiment.ground_truth,\n experiment.ground_truth_metadata,\n experiment.output_feature_name,\n top_k=3,\n labels_limit=0,\n model_namess=[\"Model1\", \"Model2\"],\n output_directory=tmpvizdir,\n file_format=viz_output,\n )\n figure_cnt = glob.glob(vis_output_pattern_pdf)\n assert 1 == len(figure_cnt)","def calculate_score(label_dict, queries, results):\n\n total_score = 0.0\n\n log_header('Individual image scores')\n\n # Calculate score for all images\n for image_id in queries:\n if image_id in results.keys():\n # Run the score function\n image_score = score(\n label_dict=label_dict, target=image_id, selection=results[image_id]\n )\n else:\n logging.error('No result generated for %s' % image_id)\n\n image_score = 0.0\n\n total_score += image_score\n\n logging.info('%s: %8.6f' % (image_id, image_score))\n\n log_header('Average score over %d images: %10.8f' % (\n len(queries), total_score / len(queries)\n ))\n\n return total_score","def compute_pck(descs0, kps0, descs1, kps1, thresh=3):\n # matches points from image 0 to image 1, using NN\n if len(descs0.shape) == 3:\n idxs = nn_set2set_match(descs0, descs1)\n else:\n idxs = nn_match(descs0, descs1) # matched image 1 keypoint indices\n predicted = kps1[idxs] # matched image 1 keypoints\n\n correct = np.linalg.norm(predicted - kps1, 2, axis=1) <= thresh\n\n return np.sum(correct) / len(correct)","def try_template_matching(image,template):\n img2 = image.copy()\n w, h = template.shape[::-1]\n # All the 6 methods for comparison in a list\n methods = ['cv2.TM_CCOEFF', 'cv2.TM_CCOEFF_NORMED', 'cv2.TM_CCORR',\n 'cv2.TM_CCORR_NORMED', 'cv2.TM_SQDIFF', 'cv2.TM_SQDIFF_NORMED']\n for meth in methods:\n img = img2.copy()\n method = eval(meth)\n # Apply template Matching\n res = cv2.matchTemplate(img,template,method)\n res-=np.min(res)\n min_val, max_val, min_loc, max_loc = cv2.minMaxLoc(res)\n threshold=0.9*np.max(res)\n \n # If the method is TM_SQDIFF or TM_SQDIFF_NORMED, take minimum\n if method in [cv2.TM_SQDIFF, cv2.TM_SQDIFF_NORMED]:\n loc = np.where( res <=0.1*np.max(res))\n else:\n loc = np.where( res >= threshold)\n \n for pt in zip(*loc[::-1]):\n cv2.rectangle(img, pt, (pt[0] + w, pt[1] + h), 255, 2)\n plt.figure()\n plt.subplot(121),plt.imshow(image,cmap = 'gray')\n plt.title('Matching Result'), plt.xticks([]), plt.yticks([])\n plt.subplot(122),plt.imshow(img,cmap = 'gray')\n plt.title('Detected Point'), plt.xticks([]), plt.yticks([])\n plt.suptitle(meth)\n plt.show()","def compare_images(self):\r\n m = round(self.mse(self.image_a, self.image_b), 4)\r\n s = round(ssim(self.image_a, self.image_b) * 100, 5)\r\n return (\r\n m, s)","def test_run_multi_r__(self):\n\n # Test Description\n # ================\n #\n # 1. This test intialises an example *eopy.matchup.matchupIO.MatchUp* object\n #\n # 2. Compare transformed dataset to expected value\n\n ################################################################################################################\n # 1. Initialise Test Data Object\n ################################################################################################################\n\n MatchUpTest = return_MatchUpTest_r__()\n\n ################################################################################################################\n # 2. Define expected values\n ################################################################################################################\n\n # Original dataset values (should be unchanged)\n MatchUpOriginal_expected = return_MatchUpTest_r__()\n\n # Transformed dataset\n values_expected = array([294.0625, 480.3733333, 300.6, 227.3846154, 210.1533333,\n 22.74193548, 22.0625, 21.96875, 22.80645161, 23.5,\n 21.66666667, 21.05882353, 23, 22.40625,\n 38.33333333, 36.63636364, 36.5, 38.42857143,\n 30.1, 32.14893617, 29.37254902, 28.88461538, 28.56603774,\n 33.45238095, 32.81395349, 31.77272727, 32.60465116,\n 40.125, 43.54054054, 38.59090909, 34.08510638,\n 13.72727273, 12, 14.1, 11.79069767, 17.53846154,\n 12.69565217, 31.16666667, 12.26086957, 11.52272727,\n 8.8125, 12, 7.4, 10.13207547])\n unc_expected = [Uncertainty(1, array([1.6, 1.5, 1.5, 1.3, 1.5])),\n Uncertainty(1, array([3.1, 3.2, 3.2, 3.1, 3.0])),\n Uncertainty(1, array([3.3, 3.4, 3.1, 3.2])),\n Uncertainty(1, array([2.1, 2.2, 2.2, 2.1])),\n Uncertainty(1, array([5.0, 4.7, 5.1, 5.2, 5.3])),\n Uncertainty(1, array([4.2, 4.3, 4.4, 4.3])),\n Uncertainty(1, array([4.0, 3.7, 4.4, 4.7])),\n Uncertainty(1, array([2.2, 1.7, 2.0, 4.3, 2.6])),\n Uncertainty(1, array([2.3, 1.2, 2.3, 4.4])),\n Uncertainty(1, array([3.2, 2.7, 3.0, 5.3]))]\n ks_expected = array([4.8, 6.8, 5.2, 5.6, 5.2, 12.10287443, 13.99394856, 12.48108926, 12.85930408])\n unck_expected = [Uncertainty(1, array([0.25, 0.25, 0.25, 0.25, 0.25])),\n Uncertainty(1, array([0.2644, 0.2644, 0.2644, 0.2644]))]\n idx_expected = {\"Nm\": [5, 4],\n \"cNm\": [0, 5, 9],\n \"Im\": [[0, 1], [1, 2]],\n \"sensors\": [-1, 1, 2],\n \"sensor_ms\": [1, 3, 3],\n \"n_sensor\": [0, 1, 1, 2, 1, 1, 2, 1, 1, 2],\n \"n_mu\": [1, 1, 2, 2, 1, 2, 2, 1, 2, 2],\n \"n_cov\": [1, 1, 1, 1, 2, 2, 2, 3, 3, 3],\n \"N_var\": [5, 5, 4, 4, 5, 4, 4, 5, 4, 4],\n \"idx\": [0, 5, 10, 14, 18, 23, 27, 31, 36, 40, 44],\n \"Ia\": [1, 1, 1, 2, 2, 2]}\n a_expected = array([1., 1.3, 0.002, 0.5, 1.1, 0.0005])\n w_matrices_expected = []\n u_matrices_expected = []\n\n ################################################################################################################\n # 3. Run Transform2NormInd.run()\n ################################################################################################################\n\n Transform2NormIndOp = Transform2NormInd()\n MatchUpTransform = Transform2NormIndOp.run(MatchUpTest)\n\n values_test = MatchUpTransform.values\n unc_test = MatchUpTransform.unc\n w_matrices_test = MatchUpTransform.w_matrices\n u_matrices_test = MatchUpTransform.u_matrices\n ks_test = MatchUpTransform.ks\n unck_test = MatchUpTransform.unck\n idx_test = MatchUpTransform.idx\n\n ################################################################################################################\n # 4. Compare retrieve values to expect values\n ################################################################################################################\n\n # Test transformed data object attribute by attribute\n\n # a. values\n for i, (value_expected, value_test) in enumerate(zip(values_expected, values_test)):\n self.assertAlmostEqual(value_expected, value_test, places=5, msg=str(i))\n\n # b. unc\n for block_unc_test, block_unc_expected in zip(unc_test, unc_expected):\n self.assertEqual(block_unc_expected.typeID, block_unc_test.typeID)\n self.assertEqual(block_unc_expected.uR.tolist(), block_unc_test.uR.tolist())\n\n # c. w_matrices\n self.assertEqual(w_matrices_test, w_matrices_expected)\n\n # d. u_matrices\n self.assertEqual(u_matrices_test, u_matrices_expected)\n\n # e. ks\n for k_expected, k_test in zip(ks_expected, ks_test):\n self.assertAlmostEqual(k_test.tolist(), k_expected.tolist(), places=5)\n\n # f. unck\n for block_unck_test, block_unck_expected in zip(unck_test, unck_expected):\n self.assertEqual(block_unck_expected.typeID, block_unck_test.typeID)\n self.assertEqual(block_unck_expected.uR.tolist(), block_unck_test.uR.tolist())\n\n # h. idx\n self.assertEqual(set(idx_expected.keys()), set(idx_test.keys()))\n for key in idx_expected.keys():\n idx_i_test = idx_test[key]\n idx_i_expected = idx_expected[key]\n if isinstance(idx_i_expected, ndarray):\n self.assertEqual(idx_i_test.tolist(), idx_i_expected.tolist())\n else:\n self.assertEqual(idx_i_test, idx_i_expected)\n\n # Test original data object preserved attribute by attribute\n\n # a. values\n for i, (value_original_expected, value_original_test) in enumerate(zip(MatchUpOriginal_expected.values, MatchUpTest.values)):\n self.assertAlmostEqual(value_original_expected, value_original_test, places=5)\n\n # b. unc\n for block_unc_original_expected, block_unc_original_test in zip(MatchUpOriginal_expected.unc, MatchUpTest.unc):\n self.assertEqual(block_unc_original_expected.typeID, block_unc_original_test.typeID)\n self.assertEqual(block_unc_original_expected.uR.tolist(), block_unc_original_test.uR.tolist())\n\n # c. w_matrices\n self.assertEqual(MatchUpOriginal_expected.w_matrices, MatchUpTest.w_matrices)\n\n # d. u_matrices\n self.assertEqual(MatchUpOriginal_expected.u_matrices, MatchUpTest.u_matrices)\n\n # e. ks\n for k_original_expected, k_original_test in zip(MatchUpOriginal_expected.ks, MatchUpTest.ks):\n self.assertAlmostEqual(k_original_test.tolist(), k_original_expected.tolist(), places=5)\n\n # f. unck\n for block_unck_original_expected, block_unck_original_test in zip(MatchUpOriginal_expected.unck, MatchUpTest.unck):\n self.assertEqual(block_unck_original_expected.typeID, block_unck_original_test.typeID)\n self.assertEqual(block_unck_original_expected.uR.tolist(), block_unck_original_test.uR.tolist())\n\n # h. idx\n self.assertEqual(set(MatchUpOriginal_expected.idx), set(MatchUpTest.idx))\n for key in MatchUpOriginal_expected.idx.keys():\n idx_i_original_test = MatchUpTest.idx[key]\n idx_i_original_expected = MatchUpOriginal_expected.idx[key]\n if isinstance(idx_i_original_expected, ndarray):\n self.assertEqual(idx_i_original_test.tolist(), idx_i_original_expected.tolist())\n else:\n self.assertEqual(idx_i_original_test, idx_i_original_expected)","def detail_matching(self):\n paradic = self.cfg['param']['paradic']\n work_dir = self.work_dir\n \n x = float(self.cfg['param']['x']) # selected pixel in the first image\n y = float(self.cfg['param']['y'])\n \n # sift parameters\n # number of bins in the orientation histogram\n n_bins = int(paradic['n_bins']) \n n_hist = int(paradic['n_hist']) \n # descriptor of n_hist X n_hist weighted histograms with n_ori\n n_ori = int(paradic['n_ori']) \n delta_min = float(paradic['delta_min'])\n sigma_min = float(paradic['sigma_min'])\n sigma_in = float(paradic['sigma_in'])\n lambda_ori = float(paradic['lambda_ori'])\n lambda_descr = float(paradic['lambda_descr'])\n #threshold defining reference orientations\n n_spo = int(paradic['n_spo'])\n \n # Read feature vectors from output files\n if (os.path.getsize(work_dir+'OUTmatches.txt') > 0 ):\n pairdata = find_nearest_keypoint(work_dir+'OUTmatches.txt', y, x)\n \n illustrate_pair(pairdata, n_bins, n_hist, n_ori, work_dir)\n\n \n # Read keys coordinates.\n d = 6+n_bins+n_hist*n_hist*n_ori # size of keydata inside pairdata\n v = n_hist*n_hist*n_ori\n [x1, y1, sigma1, theta1] = [float(x) for x in pairdata[0:4]]\n [o1, s1] = [float(x) for x in pairdata[4+v:4+v+2]]\n [x2a, y2a, sigma2a, theta2a] = [float(x) for x in pairdata[d:d+4]]\n [o2a, s2a] = [float(x) for x in pairdata[d+4+v:d+4+v+2]]\n [x2b, y2b, sigma2b, theta2b] = \\\n [float(x) for x in pairdata[2*d:2*d+4]]\n [o2b, s2b] = [float(x) for x in pairdata[2*d+4+v:2*d+4+v+2]]\n \n draw_one_match(pairdata,\n work_dir+'input_0.png',\n work_dir+'input_1.png',\n d,\n lambda_ori,\n lambda_descr,\n n_hist,\n work_dir+'OUTonepair.png')\n \n \n # Extract thumbnails.\n # keypoint 1 (image 1)\n print ' '.join(['demo_extract_patch', work_dir+'input_0.png',\n str(x1), str(y1), str(sigma1), str(theta1), str(o1), str(s1),\n str(delta_min), str(sigma_min), str(sigma_in), str(n_spo),\n str(lambda_ori), str(lambda_descr), str(n_hist),\n work_dir+\"detail_im1\"])\n proc = self.run_proc(['demo_extract_patch', work_dir+'input_0.png',\n str(x1), str(y1), str(sigma1), str(theta1), str(o1), str(s1),\n str(delta_min), str(sigma_min), str(sigma_in), str(n_spo),\n str(lambda_ori), str(lambda_descr), str(n_hist),\n work_dir+\"detail_im1\"])\n self.wait_proc(proc, timeout=self.timeout)\n \n # keypoint 2a (nearest neighbor in image 2)\n print ' '.join(['demo_extract_patch', work_dir+'input_1.png',\n str(x2a), str(y2a), str(sigma2a), str(theta2a), str(o2a), str(s2a),\n str(delta_min), str(sigma_min), str(sigma_in), str(n_spo),\n str(lambda_ori), str(lambda_descr), str(n_hist),\n work_dir+\"detail_im2a\"])\n proc = self.run_proc(['demo_extract_patch', work_dir+'input_1.png',\n str(x2a), str(y2a), str(sigma2a), str(theta2a), str(o2a), str(s2a),\n str(delta_min), str(sigma_min), str(sigma_in), str(n_spo),\n str(lambda_ori), str(lambda_descr), str(n_hist),\n work_dir+\"detail_im2a\"])\n self.wait_proc(proc, timeout=self.timeout) \n \n # keypoint 2b (second nearest neighbor in image 2)\n proc = self.run_proc(['demo_extract_patch', work_dir+'input_1.png',\n str(x2b), str(y2b), str(sigma2b), str(theta2b), str(o2b), str(s2b),\n str(delta_min), str(sigma_min), str(sigma_in), str(n_spo),\n str(lambda_ori), str(lambda_descr), str(n_hist),\n work_dir+\"detail_im2b\"])\n self.wait_proc(proc, timeout=self.timeout) \n \n \n return 1","def eval_feature_verify(query_features, gallery_features, query_lst, gallery_lst, metric=\"cosine\"):\n if metric not in [\"euclidean\", \"cosine\"]:\n raise ValueError(\"Invalid metric! \")\n\n num_query = len(query_lst)\n num_gallery = len(gallery_lst)\n\n if metric == \"cosine\":\n gallery_features = normalize(gallery_features, axis=1)\n query_features = normalize(query_features, axis=1)\n\n if metric == \"euclidean\":\n dist = euclidean_distances(gallery_features, query_features).squeeze()\n else:\n dist = np.dot(gallery_features, query_features.T)\n print(dist[:].max())\n print(dist[:].min())\n\n gallery_cam_lst = np.array([x.cam_id for x in gallery_lst], dtype=np.int32)\n gallery_id_lst = np.array([x.class_id for x in gallery_lst], dtype=np.int32)\n query_cam_lst = np.array([x.cam_id for x in query_lst], dtype=np.int32)\n query_id_lst = np.array([x.class_id for x in query_lst], dtype=np.int32)\n\n # same id\n gallery_id_lst_tile = np.tile(gallery_id_lst, (num_query, 1))\n query_id_lst_tile = np.tile(query_id_lst, (num_gallery, 1))\n label_mat = query_id_lst_tile == gallery_id_lst_tile.T # gallery num * query num\n label_mat = label_mat.astype(np.int)\n\n # same camera and id\n gallery_cam_lst_tile = np.tile(gallery_cam_lst, (num_query, 1))\n query_cam_lst_tile = np.tile(query_cam_lst, (num_gallery, 1))\n # same camera, gallery num * query num\n label_cam_mat = query_cam_lst_tile == gallery_cam_lst_tile.T\n # filter by same id\n label_cam_mat = label_cam_mat * label_mat\n # same id but under same camera are labeled as -1,\n label_mat -= 2 * label_cam_mat.astype(np.int)\n\n ind = np.where(label_mat == 0)\n neg_dist = dist[ind]\n ind = np.where(label_mat == 1)\n pos_dist = dist[ind]\n\n new_dist = np.concatenate([pos_dist, neg_dist])\n new_label = np.concatenate([np.ones_like(pos_dist), np.zeros_like(neg_dist)])\n dist_label = zip(new_dist, new_label)\n dist_label_new = np.array(sorted(dist_label, key=lambda d: -d[0]))\n\n # ap = []\n n_good = np.sum(dist_label_new[:, 1])\n n_bad = np.sum(1 - dist_label_new[:, 1])\n print(n_good)\n for thresh in np.arange(0.90, 1, 0.01):\n # print(thresh)\n p = dist_label_new[:, 0] >= thresh\n if np.sum(p) == 0:\n recall, precision, f1 = 0, 0, 0\n else:\n tp = np.logical_and(dist_label_new[:, 1], p)\n recall = np.sum(tp) / n_good\n precision = np.float(np.sum(tp)) / np.sum(p)\n\n f1 = 2*recall*precision/(recall+precision)\n\n print('%.4f\\t%.4f\\t%.4f' % (recall, precision, f1))\n\n # ap.append(compute_ap(dist_label_new, thresh))\n # print(ap)","def apply_feature_matching(desc1: np.ndarray, desc2: np.ndarray,\n match_calculator: Callable[[list, list], float]) -> list:\n\n # Check descriptors dimensions are 2\n assert desc1.ndim == 2, \"Descriptor 1 shape is not 2\"\n assert desc2.ndim == 2, \"Descriptor 2 shape is not 2\"\n\n # Check that the two features have the same descriptor type\n assert desc1.shape[1] == desc2.shape[1], \"Descriptors shapes are not equal\"\n\n # If there is not key points in any of the images\n if desc1.shape[0] == 0 or desc2.shape[0] == 0:\n return []\n\n # number of key points in each image\n num_key_points1 = desc1.shape[0]\n num_key_points2 = desc2.shape[0]\n\n # List to store the matches scores\n matches = []\n\n # Loop over each key point in image1\n # We need to calculate similarity with each key point in image2\n for kp1 in range(num_key_points1):\n # Initial variables which will be updated in the loop\n distance = -np.inf\n y_index = -1\n\n # Loop over each key point in image2\n for kp2 in range(num_key_points2):\n\n # Match features between the 2 vectors\n value = match_calculator(desc1[kp1], desc2[kp2])\n\n # SSD values examples: (50, 200, 70), we need to minimize SSD (Choose 50)\n # In case of SSD matching: (value is returned as a \"negative\" number) (-50, -200, -70)\n # So we compare it with -np.inf. (The sorting will be reversed later)\n\n # NCC values examples: (0.58, 0.87, 0.42), we need to maximize NCC (Choose 0.87)\n # In case of NCC matching: (value is returned as a \"positive\" number) (-58, -0.87, -0.42)\n # So we compare it with -np.inf. (The sorting will be reversed later)\n\n if value > distance:\n distance = value\n y_index = kp2\n\n # Create a cv2.DMatch object for each match and set attributes:\n # queryIdx: The index of the feature in the first image\n # trainIdx: The index of the feature in the second image\n # distance: The distance between the two features\n cur = cv2.DMatch()\n cur.queryIdx = kp1\n cur.trainIdx = y_index\n cur.distance = distance\n matches.append(cur)\n\n return matches","def test_RawRun_process():\n for style in test_runs:\n test_runs[style].process()\n # now compare all images with ref\n ref = sorted(glob.glob('tests/data/processed_ref/*/*/*'))\n outputs = sorted(glob.glob('tests/data/processed/*/*/*'))\n for ref, out in zip(ref, outputs):\n print ref\n print out\n assert_image_equal(ref, out)","def match2(img1, img2, coordinates1, coordinates2, PATCH_SIZE):\n\n\tpossible_matches = pd.DataFrame(columns=['feature1', 'feature2', 'diff'])\n\n\t# iteration through all the possible pairs of features from img1 and img2\n\tfor (feature1, feature2) in product(coordinates1, coordinates2):\n\t\tpatch1 = make_patch(feature1, PATCH_SIZE, img1)\n\t\tpatch2 = make_patch(feature2, PATCH_SIZE, img2)\n\t\tdiff = count_difference(patch1, patch2)\n\t\t# print(feature1, feature2, diff)\n\t\t# next line is weird, don't ask\n\t\tpossible_matches = possible_matches.append({'feature1': feature1, 'feature2': feature2, 'diff': diff}, ignore_index=True)\n\n\n\t# sorting the possible_matches according to their difference and resetin the index\n\t# inplace=True means that changes will be performed in-place, so they will change the variable and there is no need to reassign\n\tpossible_matches.sort_values(['diff'], inplace=True)\n\tpossible_matches.reset_index(inplace=True, drop=True)\n\t# print(\"SORTED\")\n\t# print(possible_matches)\n\n\t# adding only the best matches\n\tmatches = []\n\twhile not possible_matches.empty:\n\t\tbest_match = possible_matches.loc[0]\n\t\tmatches.append(list(best_match))\n\t\t# creating a list of indicies to be dropped\n\t\ttrash = []\n\n\t\t#finding and deleting rows with features that are already used and reseting the index\n\t\tfor i, (f1, f2, d) in enumerate(possible_matches.values):\n\t\t \n\t\t\tif np.all(f1 == best_match[0]) or np.all(f2 == best_match[1]):\n\t\t\t\ttrash.append(i)\n\t\n\t\tfor i in trash:\n\t\t\tpossible_matches.drop(i, inplace=True)\n\t\tpossible_matches.reset_index(inplace=True, drop=True)\n\t\t# print(possible_matches)\n\n\treturn matches","def _runmatch(self,base,mrad,xsh,ysh):\n cmd = base+'matchrad='+str(mrad)+' xsh=%.2f ysh=%.2f' %(xsh,ysh)\n sproc = popen2.Popen3(cmd,1)\n output = sproc.fromchild.readlines()\n errs = sproc.childerr.readlines()\n sproc.fromchild.close()\n sproc.childerr.close()\n \n if errs and not (len(errs) == 1 and errs[0][0:27] == 'atFindMedtf: RMS <= 0.0, so'):\n self.logfile.write('match produced the following message(s): ')\n for f in errs:\n self.logfile.write(string.strip(f))\n self.errorList.append(('match',string.strip(f)))\n matchline = ''\n medianline = ''\n for line in output:\n if line[0:6] == 'TRANS:':\n matchline = line\n elif line[0:6] == 'MEDTF:':\n medianline = line\n else:\n self.logfile.write(\"ERROR:. Unexpected output from match program:\\n\\t\\t\\t\"+line)\n return (-1,None)\n if (not matchline) or (not medianline):\n self.logfile.write(\"Match program failed.\")\n return (-1,None)\n \n mdict = self._process_median(medianline,mrad,xsh,ysh)\n if not mdict:\n return (-1,None)\n return (0,mdict)","def evaluate(self, training_scores, original_test_scores, imitation_test_scores):\n\n #finding a threshold: third to smallest training score\n sorted_scores = np.sort(training_scores)\n threshold = sorted_scores[2]\n\n #computing the number of errors\n errors = len(np.where(original_test_scores < threshold)[0])\n errors += len(np.where(imitation_test_scores > threshold)[0])\n\n #computing the local accuracy\n accuracy = 1 - errors/(len(original_test_scores)+len(imitation_test_scores))\n return accuracy, threshold","def draw_matches(img1, kp1, img2, kp2, matches, inliers, ignore_indexes, filter_by_dist=True, color=None):\n # We're drawing them side by side. Get dimensions accordingly.\n # Handle both color and grayscale images.\n\n\n\n img1 = cv2.cvtColor(img1, cv2.COLOR_GRAY2RGB)\n img2 = cv2.cvtColor(img2, cv2.COLOR_GRAY2RGB)\n if len(img1.shape) == 3:\n new_shape = (img1.shape[0] + img2.shape[0], max(img1.shape[1], img2.shape[1]), img1.shape[2])\n elif len(img1.shape) == 2:\n new_shape = (img1.shape[0] + img2.shape[0], max(img1.shape[1], img2.shape[1]))\n new_img = np.zeros(new_shape, type(img1.flat[0])) \n # Place images onto the new image.\n new_img[0:img1.shape[0],0:img1.shape[1]] = img1\n new_img[img1.shape[0]:img1.shape[0]+img2.shape[0],0:img1.shape[1]] = img2\n \n # Draw lines between matches. Make sure to offset kp coords in second image appropriately.\n r = 1\n thickness = 1\n if color:\n c = color\n\n # print(new_img.shape)\n distances = []\n for m in matches:\n distances.append(m.distance)\n \n dist_threshold = min(distances) * 2\n # print(dist_threshold)\n \n for i, m in enumerate(matches):\n if inliers:\n if not i in inliers:\n continue\n if ignore_indexes:\n if i in ignore_indexes:\n continue\n if filter_by_dist:\n if m.distance > 50:\n continue\n \n # Generate random color for RGB/BGR and grayscale images as needed.\n if not color: \n c = tuple(np.random.randint(0,256,3)) if len(img1.shape) == 3 else np.random.randint(0,256)\n c = ( int (c [ 0 ]), int (c [ 1 ]), int (c [ 2 ])) \n \n # So the keypoint locs are stored as a tuple of floats. cv2.line(), like most other things,\n # wants locs as a tuple of ints.\n\n try:\n end1 = tuple(np.round(kp1[m.queryIdx].pt).astype(int))\n end2 = tuple(np.round(kp2[m.trainIdx].pt).astype(int) + np.array([ 0, img1.shape[0]]))\n cv2.line(new_img, end1, end2, c, thickness)\n cv2.circle(new_img, end1, r, c, thickness)\n cv2.circle(new_img, end2, r, c, thickness)\n except:\n continue\n \n\n return new_img","def test_compare_performance_vis_api(experiment_to_use):\n experiment = experiment_to_use\n # extract test stats only\n test_stats = experiment.test_stats_full\n viz_outputs = (\"pdf\", \"png\")\n with TemporaryDirectory() as tmpvizdir:\n for viz_output in viz_outputs:\n vis_output_pattern_pdf = tmpvizdir + f\"/*.{viz_output}\"\n visualize.compare_performance(\n [test_stats, test_stats],\n output_feature_name=None,\n model_names=[\"Model1\", \"Model2\"],\n output_directory=tmpvizdir,\n file_format=viz_output,\n )\n figure_cnt = glob.glob(vis_output_pattern_pdf)\n assert 1 == len(figure_cnt)","def detection_precision(ground_df, detections_df, FWHMxy=1.337, pixel_size=0.16125, time_frame=0,\n verbose=True, display=False, debug=False):\n import numpy as np\n import scipy.spatial.distance as dist\n from matplotlib import pyplot as plt\n import pandas as pd\n\n check_distance = (FWHMxy / 2) / pixel_size\n\n if debug:\n print ('*** Check distance in voxels: ' + str(check_distance))\n print ('*** Check distance in microns: ' + str(check_distance * pixel_size))\n print ('')\n\n # Gets only the desired time frame from the ground truth\n ground_df = ground_df[ground_df['t'] == time_frame]\n\n nground = len(ground_df[ground_df['t'] == time_frame])\n ndetections = len(detections_df[detections_df['t'] == time_frame])\n\n if verbose:\n print ('Coordinates on ground truth: ' + str(nground))\n print ('Number of detections : ' + str(ndetections))\n\n # Gets matrix of coordinates\n x_ground = ground_df['x'].values\n y_ground = ground_df['y'].values\n z_ground = ground_df['z'].values\n ground_coords = [x_ground, y_ground, z_ground]\n ground_coords = np.transpose(ground_coords)\n\n x_detections = detections_df['x'].values\n y_detections = detections_df['y'].values\n z_detections = detections_df['z'].values\n detections_coords = [x_detections, y_detections, z_detections]\n detections_coords = np.transpose(detections_coords)\n\n if debug:\n print ('\\n*** Ground truth matrix:'),\n print np.shape(ground_coords),\n print (ground_coords.dtype)\n print ('*** Detections matrix:'),\n print np.shape(detections_coords),\n print (ground_coords.dtype)\n\n dist_matrix = dist.cdist(detections_coords, ground_coords)\n sorted_matrix_truth = np.sort(dist_matrix, axis=0)\n # print (sorted_matrix_truth)\n sorted_matrix_detection = np.sort(dist_matrix, axis=1)\n\n # Distance Matrix\n if display:\n plt.figure(figsize=(15, 10))\n plt.imshow(sorted_matrix_truth, cmap='RdYlBu_r')\n plt.title('Sorted distances based on truth')\n plt.colorbar()\n\n plt.figure(figsize=(15, 10))\n plt.imshow(sorted_matrix_detection, cmap='RdYlBu_r')\n plt.title('Sorted distances based on detections')\n plt.colorbar()\n\n # Get a vector with the closest points\n closest_truth = np.sort(sorted_matrix_truth[0, :])\n closest_detections = np.sort(sorted_matrix_detection[:, 0])\n\n # Who's smaller, detections or ground truth ?\n max_possible_hits = min(len(closest_detections), len(closest_truth))\n\n if display:\n plt.figure(figsize=(15, 5))\n plt.plot(closest_detections, color='#30AAFF', linewidth=2, label='detections')\n plt.plot(closest_truth, color='#30FFAA', linewidth=2, label='ground truth')\n plt.legend(frameon=False, loc='upper left')\n ax = plt.gca()\n ax.spines['top'].set_visible(False)\n ax.spines['right'].set_visible(False)\n ax.spines['bottom'].set_visible(False)\n ax.spines['left'].set_visible(True)\n ax.xaxis.set_ticks_position('none')\n ax.yaxis.set_ticks_position('left')\n\n # Distribution of closest\n if display:\n nbins = 50\n plt.figure(figsize=(15, 8))\n # plt.hist(minimum_distances_micron, 100, histtype='stepfilled', color='blue', label='blue')\n plt.hist(closest_truth, nbins, color='#00AAFF', alpha=0.5)\n plt.axvline(x=check_distance, linewidth=3, color='#FF3322')\n plt.title('Distribution of minimum distances (' + str(nbins) + ' bins)')\n plt.xlabel('Distance in voxels')\n\n hits_truth = (closest_truth < check_distance).sum()\n hits_detections = (closest_detections < check_distance).sum()\n hits = min(hits_truth, hits_detections)\n\n # Sanity check\n if hits > max_possible_hits:\n print ('WARNING: More hits than max possible hits. Something is wrong !')\n\n if debug:\n print ('*** Truth hits: ' + str(hits_truth))\n print ('*** Detection hits: ' + str(hits_detections))\n\n total_matches = (dist_matrix < check_distance).sum()\n nduplicates = total_matches - hits\n\n if verbose:\n print ('\\nNumber of hits: ' + str(hits) + ' (' + str((float(hits) / ndetections) * 100) + '%)')\n print ('Found ' + str(nduplicates) + ' duplicates,'),\n print ('total of ' + str(total_matches) + ' points closer than ' + str(FWHMxy / 2) + ' microns')\n\n # Calculation of Jaccard index\n # Defined as intersection over the union\n jaccard = float(hits) / ((ndetections - hits) + nground)\n\n # Average distance of detections\n avg_dist = np.average(closest_truth)\n\n if verbose:\n print ('\\nJaccard index: ' + str(jaccard))\n print ('Average distance: ' + str(avg_dist) + ' voxels')\n if verbose:\n print ('\\nAll done !')\n\n return ndetections, nground, hits, jaccard, avg_dist","def test_detector1pipeline4(_bigdata):\n step = Detector1Pipeline()\n step.save_calibrated_ramp = True\n step.ipc.skip = True\n step.persistence.skip = True\n step.jump.rejection_threshold = 4.0\n step.ramp_fit.save_opt = False\n step.output_file = 'jw84600007001_02101_00001_nrs1_rate.fits'\n step.run(_bigdata+'/pipelines/jw84600007001_02101_00001_nrs1_uncal.fits')\n\n # Compare ramp product\n n_ramp = 'jw84600007001_02101_00001_nrs1_ramp.fits'\n h = fits.open( n_ramp )\n n_ref = _bigdata+'/pipelines/jw84600007001_02101_00001_nrs1_ramp_ref.fits'\n href = fits.open( n_ref )\n newh = fits.HDUList([h['primary'],h['sci'],h['err'],h['groupdq'],h['pixeldq']])\n newhref = fits.HDUList([href['primary'],href['sci'],href['err'],href['groupdq'],h['pixeldq']])\n result = fits.diff.FITSDiff(newh,\n newhref,\n ignore_keywords = ['DATE','CAL_VER','CAL_VCS','CRDS_VER','CRDS_CTX'],\n rtol = 0.00001\n )\n assert result.identical, result.report()\n\n # Compare countrate image product\n n_cr = 'jw84600007001_02101_00001_nrs1_rate.fits'\n h = fits.open( n_cr )\n n_ref = _bigdata+'/pipelines/jw84600007001_02101_00001_nrs1_rate_ref.fits'\n href = fits.open( n_ref )\n newh = fits.HDUList([h['primary'],h['sci'],h['err'],h['dq']])\n newhref = fits.HDUList([href['primary'],href['sci'],href['err'],href['dq']])\n result = fits.diff.FITSDiff(newh,\n newhref,\n ignore_keywords = ['DATE','CAL_VER','CAL_VCS','CRDS_VER','CRDS_CTX'],\n rtol = 0.00001\n )\n assert result.identical, result.report()","def statistics(img1_blobs, img2_blobs, matches):\n\tstatistics = {}\n\n\tstatistics['#Interest Points in img1'] = len(img1_blobs)\n\tstatistics['#Interest Points in img2'] = len(img2_blobs)\n\tstatistics['Accepted Matches'] = len(matches)\n\tdissimilarity = [match[2] for match in matches]\n\tstatistics['Mean of accepted matches'] = sum(dissimilarity)/len(dissimilarity)\n\tstatistics['SD of accepted matches'] = np.std(dissimilarity)\n\treturn statistics","def evaluate_prediction(self):\n\n # ratio_train = self.evaluate_data(self.train_x, self.train_y)\n ratio_test = self.evaluate_data(self.test_x, self.test_y)\n\n print(\"\\n*NAIVE BAYES:\")\n # print(\"Test1: {}%\".format(ratio_dev*100))\n print(\"Test: {} %\".format(ratio_test*100))","def compare_group_images(directory, show_imgs=True, similarity=\"high\", compression=100):\n # list where the found duplicate/similar images are stored\n duplicates = []\n lower_res = []\n\n imgs_matrix = create_imgs_matrix(directory, compression)\n\n # search for similar images\n if similarity == \"low\":\n ref = 13000\n # search for 1:1 duplicate images\n else:\n ref = 21000\n\n main_img = 0\n compared_img = 1\n nrows, ncols = compression, compression\n srow_A = 0\n erow_A = compression\n srow_B = erow_A\n erow_B = srow_B + compression\n\n while erow_B <= imgs_matrix.shape[0]:\n while compared_img < (len(image_files)):\n # select two images from imgs_matrix\n imgA = imgs_matrix[srow_A: erow_A, # rows\n 0: ncols] # columns\n imgB = imgs_matrix[srow_B: erow_B, # rows\n 0: ncols] # columns\n # compare the images\n rotations = 0\n while image_files[main_img] not in duplicates and rotations <= 3:\n if rotations != 0:\n imgB = rotate_img(imgB)\n err = mse(imgA, imgB)\n print ( \"err:\", err)\n if err <= ref:\n if show_imgs == True:\n show_file_info(compared_img, main_img)\n add_to_list(image_files[main_img], duplicates)\n check_img_quality(directory, image_files[main_img], image_files[compared_img], lower_res)\n rotations += 1\n srow_B += compression\n erow_B += compression\n compared_img += 1\n\n srow_A += compression\n erow_A += compression\n srow_B = erow_A\n erow_B = srow_B + compression\n main_img += 1\n compared_img = main_img + 1\n\n msg = \"\\n***\\n DONE: found \" + str(len(duplicates)) + \" duplicate image pairs in \" + str(\n len(image_files)) + \" total images.\\n The following files have lower resolution:\"\n print(msg)\n return set(lower_res)"],"string":"[\n \"def patternMatchingTest(matchingAlgorithm,\\n observations= defaultCOMPREFdump , \\n modelsFolder= defaultWRFdumpsFolder,\\n #key1='0828',\\n key1='',\\n key2='',\\n #key3='',\\n verbose=True,\\n makeSmallImages=True,\\n smallDpi=40,\\n frameDpi = 600 , # added 2013-09-23\\n loadFormerAnalysisResult=False, #added 2013-09-23\\n #summary_folder=summary_folder, # needed if loadFormerAnalysisResult is True\\n summary_folder=summary_folder+timeString + '/', # 2013-09-27\\n matching_algorithm_name=\\\"\\\", # needed if loadFormerAnalysisResult is True\\n panel_cols = 5,\\n panel_rows = 5,\\n useCV2 = False, # switch off opencv since it doesn't work yet\\n useCV = False,\\n averageScoreOrdering=True, # take an average of the ordering to get a total score for the entire model over the time series\\n #kwargs = {}, # keyword arguments for the matchinging algorithm, just in case\\n **kwargs\\n ):\\n if key1 != '' or key2 != '':\\n summary_folder += '_'.join([key1, key2])\\n if not averageScoreOrdering:\\n summary_folder += '_nonaver'\\n summary_folder += '%dx%d/' % (panel_rows, panel_cols)\\n\\n t0 = time.time()\\n if loadFormerAnalysisResult:\\n mv_folder = summary_folder + matching_algorithm_name + '/'\\n result = pickle.load(open(mv_folder + 'result.pydump', 'r'))\\n\\n \\\"\\\"\\\"\\n result= {'best matches' : bestMatchAtEachTime, \\n 'best overall match' : bestMatchOverall, \\n 'all scores' : scores,\\n 'matching algorithm' : matching_algorithm_name,\\n 'dataset1[obs]' : observations,\\n 'dataset2[models]' : modelsFolder,\\n 'time spent' : time.time() - t0,\\n 'mv folder' : mv_folder, #added 2013-09-23\\n 'ordering' : ordering, #added 2013-09-23\\n }\\n \\\"\\\"\\\"\\n bestMatchAtEachTime = result['best matches'] \\n bestMatchOverall = result['best overall match']\\n scores = result['all scores']\\n matching_algorithm_name = result['matching algorithm']\\n observations = result['dataset1[obs]']\\n modelsFolder = result['dataset2[models]']\\n #mv_folder = result['mv folder']\\n ordering = result['ordering']\\n\\n # THE FOLLOWING 30 LINES ARE COPIED FROM \\\"ELSE...\\\" BELOW\\n ########### ******\\n ##\\n #\\n ds1 = pickle.load(open(observations,'r'))\\n ds1.setFloor(0) # added 2013-09-27\\n ds1.list = [v for v in ds1.list if (key1 in v.dataTime or key1 in v.name) \\\\\\n and (key2 in v.dataTime or key2 in v.name)]\\n ds1.list.sort(key=lambda v:v.dataTime)\\n ds1.setVmin(-40) #line added 2013-09-23\\n # debug\\n #print ds1.list\\n print 'loading list with keywords [ %s %s ]' %(key1, key2) \\n ds1.load(verbose=verbose)\\n # end debug\\n modelsList = [v for v in os.listdir(modelsFolder) if '.pydump' in v]\\n modelsList.sort(key = lambda v: int(re.findall(r'\\\\d+',v)[0]))\\n print \\\"modelsList\\\", modelsList\\n scores = {} # scores = {model number: [('dataTime', score), ('dataTime', score),...], ...}\\n print 'loading ds2 (model data)'\\n wrfs = [] # line added 2013-09-23\\n for path in modelsList:\\n modelIndex = int(re.findall(r'\\\\d+', path)[0])\\n ds2 = pickle.load(open(modelsFolder+path, 'r'))\\n ds2.setFloor(0) # added 2013-09-27\\n # debug\\n #ds2.list = [v for v in ds2.list if key1 in v.dataTime] # line redundant, 2013-09-23\\n ds1.list, ds2.list = ds1.intersect(ds2) #line added 2013-09-23 \\n ds2.setVmin(-40) #line added 2013-09-23\\n wrfs.append(ds2) # line added 2013-09-23\\n # end debug\\n\\n ####################\\n\\n\\n else:\\n # THE FOLLOWING 30 LINES ARE COPIED TO \\\"IF...\\\" ABOVE\\n ########### ******\\n ##\\n #\\n ds1 = pickle.load(open(observations,'r'))\\n ds1.list = [v for v in ds1.list if (key1 in v.dataTime or key1 in v.name) \\\\\\n and (key2 in v.dataTime or key2 in v.name)]\\n ds1.list.sort(key=lambda v:v.dataTime)\\n ds1.setVmin(-40) #line added 2013-09-23\\n # debug\\n #print ds1.list\\n print 'loading list with keywords [ %s %s ]' %(key1, key2) \\n ds1.load(verbose=verbose)\\n ds1.setFloor(0) # added 2013-09-27\\n # end debug\\n modelsList = [v for v in os.listdir(modelsFolder) if '.pydump' in v]\\n modelsList.sort(key = lambda v: int(re.findall(r'\\\\d+',v)[0]))\\n print \\\"modelsList\\\", modelsList\\n scores = {} # scores = {model number: [('dataTime', score), ('dataTime', score),...], ...}\\n print 'loading ds2 (model data)'\\n wrfs = [] # line added 2013-09-23\\n ds1.commonMaskSet=False # tempory attribute added 2013-09-27\\n for path in modelsList:\\n modelIndex = int(re.findall(r'\\\\d+', path)[0])\\n ds2 = pickle.load(open(modelsFolder+path, 'r'))\\n # debug\\n #ds2.list = [v for v in ds2.list if key1 in v.dataTime] # line redundant, 2013-09-23\\n ds1.list, ds2.list = ds1.intersect(ds2) #line added 2013-09-23 \\n ds2.setFloor(0) # added 2013-09-27\\n ds2[0].matrix.mask += ds1[0].matrix.mask # added 2013-09-27\\n ds2.setCommonMask() # added 2013-09-27\\n if not ds1.commonMaskSet: # added 2013-09-27\\n ds1[0].matrix.mask +=ds2[0].matrix.mask # added 2013-09-27\\n ds1.setCommonMask() # added 2013-09-27\\n ds2.setVmin(-40) #line added 2013-09-23\\n wrfs.append(ds2) # line added 2013-09-23\\n # end debug\\n\\n #######################################################################################################################\\n # key lines\\n print 'computing scores between %s and %s ...' %(ds1.name, ds2.name)\\n s, matching_algorithm_name = matchingAlgorithm(ds1, ds2, **kwargs) # list of the form [('20120612.0000', 0.9),...] \\n scores[modelIndex] = dict(s) # { model index: {'dataTime': score,...}, ...}\\n # debug\\n if verbose:\\n print scores\\n print '..................................................'\\n #\\n ########################################################################################################################\\n\\n print 'sorting the results...'\\n #print 'scores', '\\\\n', scores #debug\\n # bestMatchAtEachTime\\n T1 = [v.dataTime for v in ds1]\\n T2 = [v.dataTime for v in ds2] #any ds2 will do - they are the same, in this case we pick the last one \\n T = set(T1).intersection(T2)\\n dataTimeList = sorted(list(T))\\n\\n #print 'dataTimeList:', dataTimeList #debug\\n highestScores = []\\n ordering = [] #added 2013-09-23\\n\\n if averageScoreOrdering: # take the average scores before ordering to make it uniform\\n ave_scores = [ (v, np.mean( [scores[v].values() ] ) ) for v in scores.keys()]\\n ave_scores = dict(ave_scores)\\n print '..taking averages...............................'\\n print ave_scores\\n for u in scores.keys():\\n for t in scores[u].keys():\\n scores[u][t] = ave_scores[u]\\n print '\\\\n'.join([str(scores[v]) for v in scores.keys()])\\n\\n for t in dataTimeList:\\n print t\\n highestScoreAt_t = max([scores[u][t] for u in scores.keys()]) # u = model index (integer 1-20), t = dataTime (string)\\n new_entry = (t, highestScoreAt_t, [u for u in scores.keys() if scores[u][t]==highestScoreAt_t])\\n print new_entry\\n highestScores.append(new_entry)\\n #added 2013-09-23\\n ordering.append([0]+sorted(range(1, len(modelsList)+1), key=lambda v: scores[v][t], reverse=True))\\n\\n\\n\\n\\n # bestMatchOverall\\n print scores # debug\\n averageScores = [(Model, np.mean(scores[Model].values())) for Model in scores.keys()] # list of [(Model, mean score),...]\\n highestScore = max([v[1] for v in averageScores])\\n bestMatch = [u[0] for u in averageScores if u[1]==highestScore] \\n # debug\\n print 'average scores , highest average, best match', averageScores, highestScore, bestMatch\\n #xx= raw_input('press enter') \\n\\n\\n bestMatchAtEachTime = highestScores\\n bestMatchOverall = (highestScore, bestMatch)\\n\\n # the following lines do not work well since oftentimes we have multiple pointers (names) to a single object/function\\n # and so they are replaced by returning the function name in the function itself\\n #allvars = vars()\\n #matching_algorithm_name = [v for v in allvars if allvars[v]==matchingAlgorithm][0]\\n #mv_folder = summary_folder + matching_algorithm_name + '/'\\n mv_folder = summary_folder + matching_algorithm_name + '/' # added 2013-09-27\\n\\n result= {'best matches' : bestMatchAtEachTime, \\n 'best overall match' : bestMatchOverall, \\n 'all scores' : scores,\\n 'matching algorithm' : matching_algorithm_name,\\n 'dataset1[obs]' : observations,\\n 'dataset2[models]' : modelsFolder,\\n 'time spent' : time.time() - t0,\\n 'mv folder' : mv_folder, #added 2013-09-23\\n 'ordering' : ordering, #added 2013-09-23\\n }\\n\\n try:\\n os.makedirs(mv_folder)\\n print 'folder', mv_folder, \\\"created!\\\"\\n except:\\n print 'folder', mv_folder, 'exists!'\\n pickle.dump(result, open(mv_folder+'result.pydump','w'))\\n\\n ############################################################################\\n # to make the videos/collected images\\n # calling armor.dataStreamTools.makeVideo.makeVideo()\\n # 2013-09-23\\n\\n from . import makeVideo as mv\\n \\n try:\\n os.makedirs(mv_folder)\\n print 'folder', mv_folder, \\\"created!\\\"\\n except:\\n print 'folder', mv_folder, 'exists!'\\n # make images\\n if makeSmallImages:\\n print 'making small images'\\n for ds in [ds1] + wrfs:\\n #ds.imageFolder= mv_folder + ds.name + '/'\\n ds.imageFolder= summary_root + ds.name + '/' #added 2013-09-27 to replace the above line - too tired waiting the plots\\n ds.saveImages(drawCoast=True, flipud=True, dpi=smallDpi)\\n print ds.imageFolder\\n else:\\n # just set the paths\\n for ds in [ds1]+wrfs:\\n ds.imageFolder = summary_root + ds.name +'/' # /home/k/ARMOR/data/KONG-REY/summary/WRF01.DBZ/\\n for dbzPattern in ds.list:\\n dbzPattern.imagePath = ds.imageFolder + ds.name + dbzPattern.dataTime + '.png'\\n print dbzPattern.imagePath\\n #ds.setImagePaths() # doesn't for for old pickled objects???\\n\\n\\n # make panels\\n\\n mv.makeVideo(DSS = [ds1]+ wrfs, # [ds0, ds1, ds2, ds3, ds4, ...], a list of armor.pattern.DBZstream objects\\n panel_cols = panel_cols, # number of colums in the panel\\n panel_rows = panel_rows, # no need to be filled\\n #fourcc = cv.CV_FOURCC('F', 'L', 'V', '1'),\\n #fps = defaultFps,\\n extension= '.avi',\\n #fourcc = cv.CV_FOURCC('P', 'I', 'M', '1'),\\n outputFileName =\\\"\\\",\\n outputFolder= mv_folder,\\n sandbox = mv_folder ,\\n saveFrames = True, # saving the frames as images\\n frameDpi = frameDpi,\\n useCV2 = useCV2,\\n useCV = useCV,\\n ordering = ordering, # ordering of the models\\n )\\n ############################################################################\\n # dumping the results\\n try:\\n os.makedirs(summary_folder+ 'pydumps/')\\n except:\\n print 'Folder exists! -- ', summary_folder+ 'pydumps/' \\n\\n pickle.dump(result, open(summary_folder+ 'pydumps/results_' + matching_algorithm_name+ '+' +\\\\\\n key1 + '+' + key2 + str(int(time.time())) + '.pydump', 'w'))\\n\\n \\n\\n print \\\"time spent:\\\", time.time()-t0\\n return result\",\n \"def painting_matching_ml(imgs, db_imgs, method_list, text_masks, author_text, gt_text, metrics, weights, splits=30, max_rank=10):\\n descriptor_extractors = [get_descriptor_extractor(method_name) for method_name in method_list]\\n tmp_img_format = []\\n tmp_mask_format = []\\n tmp_text_format = []\\n for i in range(len(imgs)):\\n for j in range(len(imgs[i])):\\n tmp_img_format.append(imgs[i][j])\\n tmp_mask_format.append(text_masks[i][j])\\n tmp_text_format.append(author_text[i][j])\\n\\n #db_imgs = [img[0] for img in db_imgs]\\n db_img_splits = [i*len(db_imgs)//splits for i in range(splits-1)]\\n \\n scores = []\\n query_descriptors = extract_descriptors(tmp_img_format, descriptor_extractors, method_list, tmp_text_format, tmp_mask_format) \\n #np.array([extract_descriptors(img, matching_methods, mask) for img, mask in zip(tmp_img_format, tmp_mask_format)])\\n print(\\\"Starting db extraction + matching\\\")\\n for split in tqdm(range(splits-2)):\\n db_descriptors = extract_descriptors(db_imgs[db_img_splits[split]:db_img_splits[split+1]], descriptor_extractors, method_list, gt_text[db_img_splits[split]:db_img_splits[split+1]], None) #np.array([mrhm(db_img) for db_img in db_imgs[db_img_splits[split]:db_img_splits[split+1]]])\\n scores.append(compare_descriptors(query_descriptors, db_descriptors, metrics, method_list, weights))\\n # compare_descriptors(query_descriptors, db_descriptors, descriptor_comp_methods, descriptor_names, weights)\\n db_descriptors = extract_descriptors(db_imgs[db_img_splits[-1]:], descriptor_extractors, method_list, gt_text[db_img_splits[-1]:], None)\\n scores.append(compare_descriptors(query_descriptors, db_descriptors, metrics, method_list, weights))\\n \\n # concatenate all the results\\n scores = np.concatenate(scores, 1)\\n \\n top_k_matches = np.argpartition(scores, list(range(max_rank)))[:, :max_rank]\\n return top_k_matches\",\n \"def compare(model, input1, input2): \\n patch_image_time_start = time.time()\\n input1_patches = get_patches_non_overlap(input1, 48, 48)\\n input2_patches = get_patches_non_overlap(input2, 48, 48)\\n patch_image_time_end = time.time()\\n compare_image_time_start = time.time()\\n pred = model.predict([input1_patches, input2_patches])\\n compare_image_time_end = time.time()\\n\\n result[\\\"patch_retrieval_time\\\"].append(patch_image_time_end - patch_image_time_start)\\n result[\\\"image_comparison_time\\\"].append(compare_image_time_end - compare_image_time_start)\\n\\n return np.sum(pred)\",\n \"def ransac(matches, kp1, kp2, s=4, threshold=3, maxIterations=2000, returnMatches=False, inlierRatio=0.05, ransacRatio=0.6):\\n\\n sizes_kp1 = [kp1[dt[0].queryIdx].size for dt in matches]\\n sizes_kp2 = [kp1[dt[0].trainIdx].size for dt in matches]\\n tup_matches_kp1 = [kp1[dt[0].queryIdx].pt for dt in matches]\\n tup_matches_kp2 = [kp2[dt[0].trainIdx].pt for dt in matches]\\n matches_kp1 = np.array([[h for h in kp] + [1] for kp in tup_matches_kp1])\\n matches_kp2 = np.array([[h for h in kp] + [1] for kp in tup_matches_kp2])\\n\\n cnt_matches = len(matches)\\n\\n max_matches = []\\n max_p1, max_p2 = [], []\\n max_p1_sizes, max_p2_sizes = [], []\\n max_total = 0\\n\\n for iter in range(maxIterations):\\n # Find Homography based on random sample\\n data = random.sample(matches, s)\\n data_p1 = np.array([matches_kp1[dt[0].queryIdx] for dt in data])\\n data_p2 = np.array([matches_kp2[dt[0].trainIdx] for dt in data])\\n homography = homomat(data_p1[:, :2], data_p2[:, :2])\\n\\n # Find P1 projection from the homography matrix\\n projected_p2 = np.dot(homography, matches_kp1.transpose())\\n projected_p2 = projected_p2[0:3] / projected_p2[2] # make sure w' is 1\\n projected_p2 = projected_p2.transpose()\\n\\n # Initialize Current Matches\\n current_matches = []\\n current_p1, current_p2 = [], []\\n current_p1_sizes, current_p2_sizes = [], []\\n current_total = 0\\n\\n # Check for inliers and outliers for each matches\\n for i, (match) in enumerate(matches):\\n # normalize the error\\n error = np.linalg.norm(matches_kp2[i] - projected_p2[i])\\n\\n # Check for inliers\\n if error < threshold:\\n current_matches.append([cv.DMatch(current_total, current_total, match[0].distance)])\\n current_p1.append(matches_kp1[i][0:2])\\n current_p2.append(matches_kp2[i][0:2])\\n current_p1_sizes.append(sizes_kp1[i])\\n current_p2_sizes.append(sizes_kp2[i])\\n current_total += 1\\n\\n # If\\n if current_total > max_total and current_total >= np.round(inlierRatio*cnt_matches):\\n max_matches = current_matches\\n max_p1 = current_p1\\n max_p2 = current_p2\\n max_p1_sizes = current_p1_sizes\\n max_p2_sizes = current_p2_sizes\\n max_total = current_total\\n\\n # # we are done in case we have enough inliers\\n if current_total > cnt_matches * ransacRatio:\\n break\\n\\n\\n # Re-evaluate the Homography based on the best inliers\\n max_homography = homomat(np.array(max_p1), np.array(max_p2))\\n\\n if returnMatches:\\n max_kp1 = [cv.KeyPoint(d[0], d[1], max_p1_sizes[i]) for i, d in enumerate(max_p1)]\\n max_kp2 = [cv.KeyPoint(d[0], d[1], max_p2_sizes[i]) for i, d in enumerate(max_p2)]\\n return max_homography, max_matches, max_kp1, max_kp2\\n\\n return max_homography\",\n \"def evaluate_matrices(self, precision_matrix, recall_matrix):\\n\\t\\tground_truth_count = precision_matrix.shape[0]\\n\\t\\tdetection_count = precision_matrix.shape[1]\\n\\t\\tground_truth_sets_precision = defaultdict(set) # number of ground truth items that match a particular detection in the precision matrix\\n\\t\\tdetection_sets_precision = defaultdict(set) # number of detection items that match a particular ground truth in the precision matrix\\n\\t\\tground_truth_sets_recall = defaultdict(set) # number of ground truth items that match a particular detection in the recall matrix\\n\\t\\tdetection_sets_recall = defaultdict(set) # number of detection items that match a particular ground truth in the recall matrix\\n\\n\\t\\tfor gt_index in range(ground_truth_count):\\n\\t\\t\\tfor det_index in range(detection_count):\\n\\t\\t\\t\\tif precision_matrix[gt_index, det_index] >= self.precision_threshold:\\n\\t\\t\\t\\t\\tground_truth_sets_precision[det_index].add(gt_index)\\n\\t\\t\\t\\t\\tdetection_sets_precision[gt_index].add(det_index)\\n\\t\\t\\t\\tif recall_matrix[gt_index, det_index] >= self.recall_threshold:\\n\\t\\t\\t\\t\\tground_truth_sets_recall[det_index].add(gt_index)\\n\\t\\t\\t\\t\\tdetection_sets_recall[gt_index].add(det_index)\\n\\n\\t\\tmatch_ground_truth = 0. # sum of MatchG\\n\\t\\tmatch_detection = 0. # sum of MatchD\\n\\n\\t\\tone_to_one_precision = set()\\n\\t\\tfor gt_index in detection_sets_precision:\\n\\t\\t\\tmatching_detections_precision = detection_sets_precision[gt_index]\\n\\t\\t\\tif len(matching_detections_precision) == 1:\\n\\t\\t\\t\\t(detection_precision, ) = matching_detections_precision\\n\\t\\t\\t\\tif len(ground_truth_sets_precision[detection_precision]) == 1:\\n\\t\\t\\t\\t\\tone_to_one_precision.add((gt_index, detection_precision))\\n\\t\\t\\telse:\\n\\t\\t\\t\\t# one-to-many (one ground truth to many detections)\\n\\t\\t\\t\\tgt_sum = 0.\\n\\t\\t\\t\\tfor detection_precision in matching_detections_precision:\\n\\t\\t\\t\\t\\tgt_sum += recall_matrix[gt_index, detection_precision]\\n\\t\\t\\t\\tif gt_sum >= self.recall_threshold:\\n\\t\\t\\t\\t\\t#print(\\\"1:N ~ GT {} : DT {}\\\".format(gt_index,matching_detections_precision))\\n\\t\\t\\t\\t\\tmatch_ground_truth += self.scatter_punishment(matching_detections_precision)\\n\\t\\t\\t\\t\\tmatch_detection += len(matching_detections_precision) * self.scatter_punishment(matching_detections_precision)\\n\\n\\t\\tone_to_one_recall = set()\\n\\t\\tfor det_index in ground_truth_sets_recall:\\n\\t\\t\\tmatching_ground_truths_recall = ground_truth_sets_recall[det_index]\\n\\t\\t\\tif len(matching_ground_truths_recall) == 1:\\n\\t\\t\\t\\t(ground_truth_recall, ) = matching_ground_truths_recall\\n\\t\\t\\t\\tif len(detection_sets_recall[ground_truth_recall]) == 1:\\n\\t\\t\\t\\t\\tone_to_one_recall.add((ground_truth_recall, det_index))\\n\\t\\t\\telse:\\n\\t\\t\\t\\t# many-to-one (many ground truths covered by one detection)\\n\\t\\t\\t\\tdet_sum = 0\\n\\t\\t\\t\\tfor ground_truth_recall in matching_ground_truths_recall:\\n\\t\\t\\t\\t\\tdet_sum += precision_matrix[ground_truth_recall, det_index]\\n\\t\\t\\t\\tif det_sum >= self.precision_threshold:\\n\\t\\t\\t\\t\\t#print(\\\"N:1 ~ DT {} : GT {}\\\".format(det_index,matching_ground_truths_recall))\\n\\t\\t\\t\\t\\tmatch_detection += self.scatter_punishment(matching_ground_truths_recall)\\n\\t\\t\\t\\t\\tmatch_ground_truth += len(matching_ground_truths_recall) * self.scatter_punishment(matching_ground_truths_recall)\\n\\n\\t\\tone_to_one_matches = one_to_one_precision & one_to_one_recall\\n\\t\\tmatch_ground_truth += len(one_to_one_matches)\\n\\t\\tmatch_detection += len(one_to_one_matches)\\n\\n\\t\\trecall = match_ground_truth / float(ground_truth_count)\\n\\t\\tprecision = match_detection / float(detection_count)\\n\\t\\treturn (precision, recall, (match_detection, float(detection_count)), (match_ground_truth, float(ground_truth_count)))\",\n \"def findMatchesBetweenImages(image_1, image_2):\\n # matches - type: list of cv2.DMath\\n matches = None\\n # image_1_kp - type: list of cv2.KeyPoint items.\\n image_1_kp = None\\n # image_1_desc - type: numpy.ndarray of numpy.uint8 values.\\n image_1_desc = None\\n # image_2_kp - type: list of cv2.KeyPoint items.\\n image_2_kp = None\\n # image_2_desc - type: numpy.ndarray of numpy.uint8 values.\\n image_2_desc = None\\n # WRITE YOUR CODE HERE.\\n\\n sift = cv2.ORB_create()\\n image_1_kp, image_1_desc = sift.detectAndCompute(image_1, None)\\n image_2_kp, image_2_desc = sift.detectAndCompute(image_2, None)\\n\\n # bf = cv2.BFMatcher(cv2.NORM_HAMMING, crossCheck=True)\\n # matches = bf.match(image_1_desc, image_2_desc)\\n # matches = sorted(matches, key = lambda x:x.distance)\\n # matches = matches[:10]\\n\\n bf = cv2.BFMatcher()\\n matches = bf.knnMatch(image_1_desc, image_2_desc, k=2)\\n\\n # Apply ratio test\\n good = []\\n for m, n in matches:\\n print(m.distance, n.distance, m.distance < .75*n.distance)\\n if m.distance < (0.75 * n.distance):\\n good.append([m])\\n\\n # We coded the return statement for you. You are free to modify it -- just\\n # make sure the tests pass.\\n print(len(good), good)\\n return image_1_kp, image_2_kp, matches\\n # END OF FUNCTION.\",\n \"def match(image1,image2,threshold,useRansac=False,t_orientation=30,t_scale=0.5):\\r\\n im1, keypoints1, descriptors1 = ReadKeys(image1)\\r\\n im2, keypoints2, descriptors2 = ReadKeys(image2)\\r\\n #\\r\\n # REPLACE THIS CODE WITH YOUR SOLUTION (ASSIGNMENT 5, QUESTION 3)\\r\\n #\\r\\n # Generate five random matches (for testing purposes)\\r\\n # matched_pairs = []\\r\\n # num = 5\\r\\n # for i in range(num):\\r\\n # matched_pairs.append([keypoints1[i],keypoints2[i]])\\r\\n # return DisplayMatches(im1, im2, matched_pairs)\\r\\n\\r\\n # END OF SECTION OF CODE TO REPLACE\\r\\n #\\r\\n\\r\\n #q3\\r\\n matched_pairs = []\\r\\n between_angles = np.arccos(np.dot(descriptors1, np.transpose(descriptors2)))\\r\\n for i, row in enumerate(between_angles):\\r\\n \\tratio = sorted(row)[0] / sorted(row)[1]\\r\\n \\tif ratio <= threshold:\\r\\n\\t \\tmatched_pairs.append([keypoints1[i], keypoints2[np.where(row == sorted(row)[0])[0][0]]])\\r\\n # print(matched_pairs)\\r\\n if useRansac is False:\\r\\n return DisplayMatches(im1, im2, matched_pairs)\\r\\n\\t# \\r\\n\\r\\n #q4\\r\\n repetition = 10\\r\\n subsets = [[]] * repetition\\r\\n for i in range(repetition):\\r\\n r = random.randint(0, len(matched_pairs))\\r\\n for match in matched_pairs:\\r\\n ds1, ds2 = matched_pairs[r][1][2]/matched_pairs[r][0][2], match[1][2]/match[0][2]\\r\\n do1, do2 = (matched_pairs[r][1][3]-matched_pairs[r][0][3]), (match[1][3]-match[0][3])\\r\\n if abs(ds2 - ds1) <= t_scale * ds1 and abs(do2 - do1) % (2 * math.pi) <= t_orientation:\\r\\n subsets[i].append(match)\\r\\n\\r\\n max_i, max_len = 0, subsets[0]\\r\\n for i in range(10):\\r\\n l = len(subsets[i])\\r\\n if l > max_len:\\r\\n max_len = l\\r\\n max_i = i\\r\\n\\r\\n im3 = DisplayMatches(im1, im2, subsets[max_i])\\r\\n return im3\",\n \"def _match_keypoints(self, query_image: Image) -> typing.Tuple[typing.List[int], typing.List[typing.Any]]:\\n query_image.get_keypoints_and_descriptors()\\n\\n # matches = self.matcher.match(query_image.descriptor)\\n matches = self.matcher.knnMatch(query_image.descriptor, k=2)\\n good = []\\n for m_n in matches:\\n if len(m_n) == 1:\\n good.append(m_n[0])\\n continue\\n elif len(m_n) != 2:\\n continue\\n (m, n) = m_n\\n if m.distance < 0.7 * n.distance:\\n good.append(m)\\n\\n images_scores = [0] * len(self._candidate_images)\\n images_matches = [None] * len(self._candidate_images)\\n for image_index, image in enumerate(self._candidate_images):\\n matches_scores = []\\n matches = []\\n for i, match in enumerate(good):\\n if match.imgIdx != image_index:\\n continue\\n matches.append(match)\\n matches_scores.append((256 - match.distance) / 256)\\n\\n match_cnt = len(matches_scores)\\n if match_cnt <= 0:\\n continue\\n\\n images_scores[image_index] = (\\n 0.5 + ((math.tanh(match_cnt / 3 - 1)) / 2)) * (sum(matches_scores) / match_cnt)\\n\\n images_matches[image_index] = matches\\n\\n return images_scores, images_matches\",\n \"def compute_precision_recall_bbox(per_image_detections, per_image_gts, num_gt_classes,\\n matching_iou_threshold=0.5):\\n per_image_eval = per_image_evaluation.PerImageEvaluation(\\n num_groundtruth_classes=num_gt_classes,\\n matching_iou_threshold=matching_iou_threshold,\\n nms_iou_threshold=1.0,\\n nms_max_output_boxes=10000\\n )\\n\\n print('Running per-object analysis...')\\n\\n detection_tp_fp = defaultdict(list) # key is the category; in each list, 1 is tp, 0 is fp\\n detection_scores = defaultdict(list)\\n num_total_gt = defaultdict(int)\\n\\n for image_id, dets in tqdm(per_image_detections.items()):\\n\\n detected_boxes = np.array(dets['boxes'], dtype=np.float32)\\n detected_scores = np.array(dets['scores'], dtype=np.float32)\\n # labels input to compute_object_detection_metrics() needs to start at 0, not 1\\n detected_labels = np.array(dets['labels'], dtype=np.int) - 1 # start at 0\\n # num_detections = len(dets['boxes'])\\n\\n gts = per_image_gts[image_id]\\n gt_boxes = np.array(gts['gt_boxes'], dtype=np.float32)\\n gt_labels = np.array(gts['gt_labels'], dtype=np.int) - 1 # start at 0\\n num_gts = len(gts['gt_boxes'])\\n\\n groundtruth_is_difficult_list = np.zeros(num_gts, dtype=bool) # place holders - we don't have these\\n groundtruth_is_group_of_list = np.zeros(num_gts, dtype=bool)\\n\\n # to prevent 'Invalid dimensions for box data.' error\\n if num_gts == 0:\\n # this box will not match any detections\\n gt_boxes = np.array([[0, 0, 0, 0]], dtype=np.float32)\\n\\n scores, tp_fp_labels, is_class_correctly_detected_in_image = (\\n per_image_eval.compute_object_detection_metrics(\\n detected_boxes=detected_boxes,\\n detected_scores=detected_scores,\\n detected_class_labels=detected_labels,\\n groundtruth_boxes=gt_boxes,\\n groundtruth_class_labels=gt_labels,\\n groundtruth_is_difficult_list=groundtruth_is_difficult_list,\\n groundtruth_is_group_of_list=groundtruth_is_group_of_list\\n )\\n )\\n\\n for i, tp_fp_labels_cat in enumerate(tp_fp_labels):\\n assert sum(tp_fp_labels_cat) <= sum(gt_labels == i) # true positives < gt of that category\\n cat = i + 1 # categories start at 1\\n detection_tp_fp[cat].append(tp_fp_labels_cat)\\n detection_scores[cat].append(scores[i])\\n num_total_gt[cat] += sum(gt_labels == i) # gt_labels start at 0\\n\\n all_scores = []\\n all_tp_fp = []\\n\\n print('Computing precision recall for each category...')\\n per_cat_metrics = {}\\n for i in range(num_gt_classes):\\n cat = i + 1\\n scores_cat = np.concatenate(detection_scores[cat])\\n tp_fp_cat = np.concatenate(detection_tp_fp[cat]).astype(np.bool)\\n all_scores.append(scores_cat)\\n all_tp_fp.append(tp_fp_cat)\\n\\n precision, recall = metrics.compute_precision_recall(\\n scores_cat, tp_fp_cat, num_total_gt[cat]\\n )\\n average_precision = metrics.compute_average_precision(precision, recall)\\n\\n per_cat_metrics[cat] = {\\n 'category': cat,\\n 'precision': precision,\\n 'recall': recall,\\n 'average_precision': average_precision,\\n 'scores': scores_cat,\\n 'tp_fp': tp_fp_cat,\\n 'num_gt': num_total_gt[cat]\\n }\\n print('Number of ground truth in category {} is {}'.format(cat, num_total_gt[cat]))\\n\\n # compute one-class precision/recall/average precision (if every box is just of an object class)\\n all_scores = np.concatenate(all_scores)\\n all_tp_fp = np.concatenate(all_tp_fp)\\n overall_gt_count = sum(num_total_gt.values())\\n\\n one_class_prec, one_class_recall = metrics.compute_precision_recall(\\n all_scores, all_tp_fp, overall_gt_count\\n )\\n one_class_average_precision = metrics.compute_average_precision(one_class_prec, one_class_recall)\\n\\n per_cat_metrics['one_class'] = {\\n 'category': 'one_class',\\n 'precision': one_class_prec,\\n 'recall': one_class_recall,\\n 'average_precision': one_class_average_precision,\\n 'scores': all_scores,\\n 'tp_fp': all_tp_fp,\\n 'num_gt': overall_gt_count\\n }\\n\\n return per_cat_metrics\",\n \"def _compute_prediction(self, k, img, kp, des):\\n ###taken from example code:\\n\\n #find matches\\n matches = self.matcher.knnMatch(des,self.descs[k],k=2)\\n\\n # make sure the distance to the closest match is sufficiently better than the second closest\\n good_matches = []\\n for m,n in matches: \\n if (m.distance < 0.7*n.distance and\\n kp[m.queryIdx].response > 0.0 and\\n self.kps[k][m.trainIdx].response > 0.0):\\n good_matches.append((m.queryIdx, m.trainIdx))\\n \\n if len(good_matches) < 4:\\n visual_diff = None\\n else:\\n #create pt lists\\n img_pts = np.zeros((len(good_matches),2))\\n tem_pts = np.zeros((len(good_matches),2))\\n for idx in range(len(good_matches)):\\n match = good_matches[idx]\\n img_pts[idx,:] = kp[match[0]].pt\\n tem_pts[idx,:] = self.kps[k][match[1]].pt\\n \\n # Transform input image so that it matches the template image as well as possible\\n M, mask = cv2.findHomography(img_pts, tem_pts, cv2.RANSAC, 8.0)\\n img_T = cv2.warpPerspective(img, M, self.signs[k].shape[::-1])\\n \\t\\n #calculate error\\n visual_diff = compare_images(img_T, self.signs[k])\\n return visual_diff\\n #not functioning properly (currently does not return highest value for identical image/scene pair)\",\n \"def ratio_test(fx2_match, fx1_match, fx1_norm, match_dist, norm_dist,\\n ratio_thresh=.625, fm_dtype=np.int32, fs_dtype=np.float32):\\n fx2_to_ratio = np.divide(match_dist, norm_dist).astype(fs_dtype)\\n fx2_to_isvalid = np.less(fx2_to_ratio, ratio_thresh)\\n fx2_match_RAT = fx2_match.compress(fx2_to_isvalid).astype(fm_dtype)\\n fx1_match_RAT = fx1_match.compress(fx2_to_isvalid).astype(fm_dtype)\\n fx1_norm_RAT = fx1_norm.compress(fx2_to_isvalid).astype(fm_dtype)\\n # Turn the ratio into a score\\n fs_RAT = np.subtract(1.0, fx2_to_ratio.compress(fx2_to_isvalid))\\n fm_RAT = np.vstack((fx1_match_RAT, fx2_match_RAT)).T\\n # return normalizer info as well\\n fm_norm_RAT = np.vstack((fx1_norm_RAT, fx2_match_RAT)).T\\n ratio_tup = MatchTup3(fm_RAT, fs_RAT, fm_norm_RAT)\\n return ratio_tup\",\n \"def evaluate_detections(ground_truth, predictions, class_name, overlap_threshold=0.5,\\n allow_multiple_matches_per_ignored=True,\\n verbose=True):\\n\\n Detection = namedtuple('Detection', ['image', 'bbox', 'score', 'gt_match'])\\n GT = namedtuple('GroundTruth', ['bbox', 'is_matched', 'is_ignored'])\\n detections = [Detection(image=img_pred.image_path,\\n bbox=np.array(obj_pred[\\\"bbox\\\"]),\\n score=obj_pred.get(\\\"score\\\", 0.0),\\n gt_match=-1)\\n for img_pred in predictions\\n for obj_pred in img_pred\\n if obj_pred[\\\"type\\\"] == class_name]\\n\\n scores = np.array([detection.score for detection in detections])\\n sorted_ind = np.argsort(-scores)\\n detections = [detections[i] for i in sorted_ind]\\n\\n gts = {}\\n for img_gt in ground_truth:\\n gts[img_gt.image_path] = GT(\\n bbox=np.vstack([np.array(obj_gt[\\\"bbox\\\"]) for obj_gt in img_gt]) if img_gt else np.empty(\\n (0, 4)),\\n is_matched=np.zeros(len(img_gt), dtype=bool),\\n is_ignored=np.array([obj_gt.get(\\\"is_ignored\\\", False) for obj_gt in img_gt], dtype=bool))\\n\\n detections_num = len(detections)\\n true_pos = np.zeros(detections_num)\\n false_pos = np.zeros(detections_num)\\n\\n for i, detection in tqdm(enumerate(detections), desc=\\\"Processing detections\\\",\\n disable=not verbose):\\n image_path = detection.image\\n bboxes_gt = gts[image_path].bbox\\n bbox = detection.bbox\\n max_overlap = -np.inf\\n\\n if bboxes_gt is not None and bboxes_gt.shape[0] > 0:\\n intersection_xmin = np.maximum(bboxes_gt[:, 0], bbox[0])\\n intersection_ymin = np.maximum(bboxes_gt[:, 1], bbox[1])\\n intersection_xmax = np.minimum(bboxes_gt[:, 0] + bboxes_gt[:, 2], bbox[0] + bbox[2])\\n intersection_ymax = np.minimum(bboxes_gt[:, 1] + bboxes_gt[:, 3], bbox[1] + bbox[3])\\n intersection_width = np.maximum(intersection_xmax - intersection_xmin, 0.)\\n intersection_height = np.maximum(intersection_ymax - intersection_ymin, 0.)\\n intersection = intersection_width * intersection_height\\n\\n det_area = bbox[2] * bbox[3]\\n gt_area = bboxes_gt[:, 2] * bboxes_gt[:, 3]\\n union = (det_area + gt_area - intersection)\\n ignored_mask = gts[image_path].is_ignored\\n if allow_multiple_matches_per_ignored:\\n if np.any(ignored_mask):\\n union[ignored_mask] = det_area\\n\\n overlaps = intersection / union\\n # Match not ignored ground truths first.\\n if np.any(~ignored_mask):\\n overlaps_filtered = np.copy(overlaps)\\n overlaps_filtered[ignored_mask] = 0.0\\n max_overlap = np.max(overlaps_filtered)\\n argmax_overlap = np.argmax(overlaps_filtered)\\n # If match with non-ignored ground truth is not good enough,\\n # try to match with ignored ones.\\n if max_overlap < overlap_threshold and np.any(ignored_mask):\\n overlaps_filtered = np.copy(overlaps)\\n overlaps_filtered[~ignored_mask] = 0.0\\n max_overlap = np.max(overlaps_filtered)\\n argmax_overlap = np.argmax(overlaps_filtered)\\n detections[i] = detection._replace(gt_match=argmax_overlap)\\n\\n if max_overlap >= overlap_threshold:\\n if not gts[image_path].is_ignored[argmax_overlap]:\\n if not gts[image_path].is_matched[argmax_overlap]:\\n true_pos[i] = 1.\\n gts[image_path].is_matched[argmax_overlap] = True\\n else:\\n false_pos[i] = 1.\\n elif not allow_multiple_matches_per_ignored:\\n gts[image_path].is_matched[argmax_overlap] = True\\n else:\\n false_pos[i] = 1.\\n\\n false_pos = np.cumsum(false_pos)\\n true_pos = np.cumsum(true_pos)\\n\\n debug_visualization = False\\n if debug_visualization:\\n for image_path, bboxes_gt in gts.items():\\n\\n print(image_path)\\n image = cv2.imread(image_path)\\n image_gt = np.copy(image)\\n for bbox in bboxes_gt.bbox:\\n cv2.rectangle(image_gt, tuple(bbox[:2]), tuple(bbox[2:] + bbox[:2]),\\n color=(255, 255, 0), thickness=2)\\n cv2.imshow(\\\"gt\\\", image_gt)\\n for detection in detections:\\n if detection.image != image_path:\\n continue\\n bbox = detection.bbox\\n cv2.rectangle(image, tuple(bbox[:2]), tuple(bbox[2:] + bbox[:2]), color=(0, 255, 0),\\n thickness=2)\\n if detection.gt_match is not None:\\n bbox = bboxes_gt.bbox[detection.gt_match]\\n cv2.rectangle(image, tuple(bbox[:2]), tuple(bbox[2:] + bbox[:2]),\\n color=(0, 0, 255), thickness=1)\\n cv2.imshow(\\\"image\\\", image)\\n cv2.waitKey(0)\\n\\n # Handle equal-score detections.\\n # Get index of the last occurrence of a score.\\n ind = len(scores) - np.unique(scores[sorted_ind[::-1]], return_index=True)[1] - 1\\n ind = ind[::-1]\\n # Though away redundant points.\\n false_pos = false_pos[ind]\\n true_pos = true_pos[ind]\\n\\n total_positives_num = np.sum([np.count_nonzero(~gt.is_ignored) for gt in gts.values()])\\n recall = true_pos / float(total_positives_num)\\n # Avoid divide by zero in case the first detection matches an ignored ground truth.\\n precision = true_pos / np.maximum(true_pos + false_pos, np.finfo(np.float64).eps)\\n miss_rate = 1.0 - recall\\n fppi = false_pos / float(len(gts))\\n\\n return recall, precision, miss_rate, fppi\",\n \"def evaluate(pred_file, ref_file):\\n ref_dict, pred_dict, query_dict, id_dict = build_pred_ref_dict(ref_file, pred_file, ref_file)\\n total, acc, scores = res_eval_with_type_acc(query_dict, pred_dict, ref_dict, id_dict, save=False)\\n em = calculate_exact_match(pred_dict, ref_dict)\\n print('Comp Acc: {:.3f}%\\\\tBleu-4: {:.3f}\\\\tRouge-L: {:.3f}'.format(acc, scores['Bleu-4'], scores['Rouge-L']))\\n print('EM: {:.3f}%'.format(em))\\n # calculate_sketch_type_acc(ref_file, pred_file)\\n # calculate_exact_match_for_each_q_type(ref_file, pred_file)\\n return total, acc, scores, em\",\n \"def ransac(matches, kp1, kp2, sample_points=4, iterations=5, inlier_tolerance=3, inlier_ratio=0.45, check=True, return_max_x=False):\\n\\n best_inlier_count = 0\\n best_h = None\\n best_inlier_indices = None\\n\\n # Get all the corresponing matching pairs for both the images\\n pts1 = np.array([kp1[m.queryIdx].pt for m in matches])\\n pts2 = np.array([kp2[m.trainIdx].pt for m in matches])\\n\\n # Re-usable variables for all iterations\\n homogeneous_pts1 = np.hstack((pts1, np.ones((pts1.shape[0], 1)))).T\\n indices = np.arange(len(pts1))\\n num_pts = pts1.shape[0]\\n required_inliers = inlier_ratio * num_pts\\n\\n # For number of iterations\\n for _ in range(iterations):\\n\\n # Sample a small set of points from the point match pairs\\n indices_to_sample = np.random.choice(indices, sample_points)\\n pts1_sample = pts1[indices_to_sample]\\n pts2_sample = pts2[indices_to_sample]\\n\\n # Get the homography matrix\\n h = get_homography_matrix(pts1_sample, pts2_sample)\\n\\n # Find the new points using the homography matrix\\n transformed_points = np.dot(h, homogeneous_pts1).T\\n\\n # Convert it to world coordinates\\n last_col = np.copy(transformed_points[:, -1])\\n last_col = last_col[:, np.newaxis]\\n transformed_points /= last_col\\n transformed_points = transformed_points[:, :-1]\\n\\n # Find the distance between the actual and the mapped points\\n distance = np.linalg.norm(pts2 - transformed_points, axis=1)\\n inlier_indices = distance < inlier_tolerance\\n inlier_count = inlier_indices.sum()\\n\\n # Update the best_h if the current h has more inliers\\n if inlier_count > best_inlier_count:\\n best_h = h\\n best_inlier_indices = inlier_indices\\n best_inlier_count = inlier_count\\n\\n # If required inliers is reached break\\n if inlier_count > required_inliers:\\n break\\n\\n # Verbose mode - Print the number of inliers\\n if check:\\n transformed_points = np.dot(best_h, homogeneous_pts1).T\\n # Convert it to world coordinates\\n last_col = np.copy(transformed_points[:, -1])\\n last_col = last_col[:, np.newaxis]\\n transformed_points /= last_col\\n transformed_points = transformed_points[:, :-1]\\n distance = np.linalg.norm(pts2 - transformed_points, axis=1)\\n inlier_count = len(distance[distance < inlier_tolerance])\\n print('%2.2f of the points are inliers' %\\n (inlier_count / num_pts * 100))\\n\\n # If x coordinates are needed\\n if return_max_x:\\n max_x_inlier_1 = ceil(pts1[best_inlier_indices].max(axis=0)[0])\\n max_x_inlier_2 = ceil(pts2[best_inlier_indices].max(axis=0)[0])\\n return best_h, max_x_inlier_1, max_x_inlier_2\\n return best_h\",\n \"def test_compare_classifier_performance_from_pred_vis_api(experiment_to_use):\\n experiment = experiment_to_use\\n prediction = experiment.predictions\\n viz_outputs = (\\\"pdf\\\", \\\"png\\\")\\n with TemporaryDirectory() as tmpvizdir:\\n for viz_output in viz_outputs:\\n vis_output_pattern_pdf = tmpvizdir + f\\\"/*.{viz_output}\\\"\\n visualize.compare_classifiers_performance_from_pred(\\n [prediction, prediction],\\n experiment.ground_truth,\\n experiment.ground_truth_metadata,\\n experiment.output_feature_name,\\n labels_limit=0,\\n model_namess=[\\\"Model1\\\", \\\"Model2\\\"],\\n output_directory=tmpvizdir,\\n file_format=viz_output,\\n )\\n figure_cnt = glob.glob(vis_output_pattern_pdf)\\n assert 1 == len(figure_cnt)\",\n \"def calculates_results_stats(results_dic): \\n # Creates empty dictionary for results_stats_dic\\n results_stats_dic = dict()\\n \\n # Sets all counters to initial values of zero so that they can \\n # be incremented while processing through the images in results_dic \\n results_stats_dic['n_dogs_img'] = 0\\n results_stats_dic['n_match'] = 0\\n results_stats_dic['n_correct_dogs'] = 0\\n results_stats_dic['n_correct_notdogs'] = 0\\n results_stats_dic['n_correct_breed'] = 0\\n \\n # process through the results dictionary\\n for key in results_dic:\\n \\n # Labels Match Exactly\\n if results_dic[key][2] == 1:\\n results_stats_dic['n_match'] += 1\\n\\n # TODO: 5a. REPLACE pass with CODE that counts how many pet images of\\n # dogs had their breed correctly classified. This happens \\n # when the pet image label indicates the image is-a-dog AND \\n # the pet image label and the classifier label match. You \\n # will need to write a conditional statement that determines\\n # when the dog breed is correctly classified and then \\n # increments 'n_correct_breed' by 1. Recall 'n_correct_breed' \\n # is a key in the results_stats_dic dictionary with it's value \\n # representing the number of correctly classified dog breeds.\\n # \\n # Pet Image Label is a Dog AND Labels match- counts Correct Breed\\n if results_dic[key][3] == 1 and results_dic[key][2] == 1:\\n results_stats_dic['n_correct_breed'] += 1\\n \\n # Pet Image Label is a Dog - counts number of dog images\\n if results_dic[key][3] == 1:\\n results_stats_dic['n_dogs_img'] += 1\\n \\n # Classifier classifies image as Dog (& pet image is a dog)\\n # counts number of correct dog classifications\\n if results_dic[key][4] == 1:\\n results_stats_dic['n_correct_dogs'] += 1\\n\\n # TODO: 5b. REPLACE pass with CODE that counts how many pet images \\n # that are NOT dogs were correctly classified. This happens \\n # when the pet image label indicates the image is-NOT-a-dog \\n # AND the classifier label indicates the images is-NOT-a-dog.\\n # You will need to write a conditional statement that \\n # determines when the classifier label indicates the image \\n # is-NOT-a-dog and then increments 'n_correct_notdogs' by 1. \\n # Recall the 'else:' above 'pass' already indicates that the \\n # pet image label indicates the image is-NOT-a-dog and \\n # 'n_correct_notdogs' is a key in the results_stats_dic dictionary \\n # with it's value representing the number of correctly \\n # classified NOT-a-dog images.\\n # \\n # Pet Image Label is NOT a Dog\\n else:\\n # Classifier classifies image as NOT a Dog(& pet image isn't a dog)\\n # counts number of correct NOT dog clasifications.\\n if results_dic[key][3] == 0 and results_dic[key][4] == 0:\\n results_stats_dic['n_correct_notdogs'] += 1\\n\\n\\n # Calculates run statistics (counts & percentages) below that are calculated\\n # using the counters from above.\\n\\n # calculates number of total images\\n results_stats_dic['n_images'] = len(results_dic)\\n\\n # calculates number of not-a-dog images using - images & dog images counts\\n results_stats_dic['n_notdogs_img'] = (results_stats_dic['n_images'] - \\n results_stats_dic['n_dogs_img']) \\n\\n # TODO: 5c. REPLACE zero(0.0) with CODE that calculates the % of correctly\\n # matched images. Recall that this can be calculated by the\\n # number of correctly matched images ('n_match') divided by the \\n # number of images('n_images'). This result will need to be \\n # multiplied by 100.0 to provide the percentage.\\n # \\n # Calculates % correct for matches\\n results_stats_dic['pct_match'] = (results_stats_dic['n_match'] / results_stats_dic['n_images']) * 100\\n\\n # TODO: 5d. REPLACE zero(0.0) with CODE that calculates the % of correctly\\n # classified dog images. Recall that this can be calculated by \\n # the number of correctly classified dog images('n_correct_dogs')\\n # divided by the number of dog images('n_dogs_img'). This result \\n # will need to be multiplied by 100.0 to provide the percentage.\\n # \\n # Calculates % correct dogs\\n results_stats_dic['pct_correct_dogs'] = (results_stats_dic['n_correct_dogs'] / results_stats_dic['n_dogs_img']) * 100\\n\\n # TODO: 5e. REPLACE zero(0.0) with CODE that calculates the % of correctly\\n # classified breeds of dogs. Recall that this can be calculated \\n # by the number of correctly classified breeds of dog('n_correct_breed') \\n # divided by the number of dog images('n_dogs_img'). This result \\n # will need to be multiplied by 100.0 to provide the percentage.\\n # \\n # Calculates % correct breed of dog\\n results_stats_dic['pct_correct_breed'] = (results_stats_dic['n_correct_breed'] / results_stats_dic['n_dogs_img']) * 100\\n\\n # Calculates % correct not-a-dog images\\n # Uses conditional statement for when no 'not a dog' images were submitted \\n if results_stats_dic['n_notdogs_img'] > 0:\\n results_stats_dic['pct_correct_notdogs'] = (results_stats_dic['n_correct_notdogs'] /\\n results_stats_dic['n_notdogs_img']) * 100.0\\n else:\\n results_stats_dic['pct_correct_notdogs'] = 0.0\\n\\n \\n # TODO 5f. REPLACE None with the results_stats_dic dictionary that you \\n # created with this function \\n return results_stats_dic\",\n \"def _run():\\n matching_terms = {'a', 'b'}\\n source_counts = {'a': 10, 'b': 50, 'c': 25}\\n target_counts = {'a': 4, 'b': 73, 'c': 15}\\n source_chunk = ['a', 'b']\\n target_chunk = ['a', 'c', 'b']\\n source_distance = score.find_distance(\\n matching_terms, source_chunk, source_counts)\\n target_distance = score.find_distance(\\n matching_terms, target_chunk, target_counts)\\n match_score = score.vanilla(\\n matching_terms, source_distance, target_distance, source_counts,\\n target_counts)\\n print('Calculated score:', match_score)\",\n \"def analyze_similarities():\\r\\n print('Total number of candidate pairs:', len(pairs))\\r\\n print(f'\\\\nNumber of actual item pairs in the train set: {pairs[\\\"true_match\\\"].sum()}\\\\n')\\r\\n\\r\\n for feature in ['text_score', 'image_score', 'txt_img_score', 'words_ratio', 'txt_img_words']:\\r\\n\\r\\n # Check distribution of True and False predictions for various similarity scores\\r\\n print('-' * 50)\\r\\n print(f'\\\\nDistribution of True/False predictions for {feature}')\\r\\n for thr in (0.6, 0.65, 0.7, 0.75, 0.8, 0.85, 0.9, 0.95):\\r\\n print('-' * 50)\\r\\n print(f'Similarity score over {thr}')\\r\\n pairs_sample = pairs[pairs[feature] >= thr]\\r\\n print(f'Number of similar item pairs: {len(pairs_sample)}')\\r\\n print(pairs_sample['true_match'].value_counts(normalize=True))\\r\\n\\r\\n # Check if identical phash can be used to improve the accuracy\\r\\n same_phash = pairs[pairs['phash_match'] == 1]\\r\\n different_phash = pairs[pairs['phash_match'] == 0]\\r\\n\\r\\n print('\\\\nFor item pairs with the same phash:')\\r\\n print(same_phash['true_match'].value_counts(normalize=True))\\r\\n print('Number of item pairs in this subset:', len(same_phash))\\r\\n\\r\\n print('\\\\nFor item pairs with different phash:')\\r\\n print(different_phash['true_match'].value_counts(normalize=True))\\r\\n print('Number of item pairs in this subset:', len(different_phash))\\r\\n\\r\\n # Check if numbers in titles can be used to improve the accuracy\\r\\n same_numbers = pairs[pairs['nums_match'] == 1]\\r\\n different_numbers = pairs[pairs['nums_match'] == 0]\\r\\n\\r\\n print('\\\\nFor item pairs with the same numbers:')\\r\\n print(same_numbers['true_match'].value_counts(normalize=True))\\r\\n print('Number of item pairs in this subset:', len(same_numbers))\\r\\n\\r\\n print('\\\\nFor item pairs with different numbers:')\\r\\n print(different_numbers['true_match'].value_counts(normalize=True))\\r\\n print('Number of item pairs in this subset:', len(different_numbers))\",\n \"def drawMatches(img1, kp1, img2, kp2, matches):\\n\\n # Create a new output image that concatenates the two images together\\n # (a.k.a) a montage\\n rows1 = img1.shape[0]\\n cols1 = img1.shape[1]\\n rows2 = img2.shape[0]\\n cols2 = img2.shape[1]\\n\\n out = np.zeros((max([rows1,rows2]),cols1+cols2,3), dtype='uint8')\\n\\n # Place the first image to the left\\n out[:rows1,:cols1] = np.dstack([img1, img1, img1])\\n\\n # Place the next image to the right of it\\n out[:rows2,cols1:] = np.dstack([img2, img2, img2])\\n\\n # For each pair of points we have between both images\\n # draw circles, then connect a line between them\\n for mat in matches:\\n\\n # Get the matching keypoints for each of the images\\n img1_idx = mat.queryIdx\\n img2_idx = des_dict[mat.trainIdx][1]\\n\\n # x - columns\\n # y - rows\\n (x1,y1) = kp1[img1_idx].pt\\n (x2,y2) = kp2[img2_idx].pt\\n\\n # Draw a small circle at both co-ordinates\\n # radius 4\\n # colour green\\n # thickness = 1\\n cv2.circle(out, (int(x1),int(y1)), 4, (0, 255, 0)) \\n cv2.circle(out, (int(x2)+cols1,int(y2)), 4, (0, 255, 0))\\n\\n # Draw a line in between the two points\\n # thickness = 1\\n # colour green\\n cv2.line(out, (int(x1),int(y1)), (int(x2)+cols1,int(y2)), (0, 255, 0), 1)\\n\\n\\n # Show the image\\n #cv2.imshow('Matched Features', out)\\n #cv2.waitKey(0)\\n #cv2.destroyWindow('Matched Features')\\n\\n # Also return the image if you'd like a copy\\n return out\",\n \"def my_evaluate(original_image, img_id, annotations, min_score=0.4, max_overlap=0.3, top_k=200):\\n\\n # Transform\\n image = normalize(to_tensor(resize(original_image)))\\n\\n # Move to default device\\n image = image.to(device)\\n\\n # Forward prop.\\n predicted_locs, predicted_scores = model(image.unsqueeze(0))\\n\\n # Detect objects in SSD output\\n det_boxes, det_labels, det_scores = model.detect_objects(predicted_locs, predicted_scores, min_score=min_score,\\n max_overlap=max_overlap, top_k=top_k)\\n\\n # Move detections to the CPU\\n det_boxes = det_boxes[0].to('cpu')\\n\\n # Transform to original image dimensions\\n original_dims = torch.FloatTensor(\\n [original_image.width, original_image.height, original_image.width, original_image.height]).unsqueeze(0)\\n det_boxes = det_boxes * original_dims\\n\\n # Decode class integer labels\\n det_labels = [rev_label_map[l] for l in det_labels[0].to('cpu').tolist()]\\n det_scores = det_scores[0].to('cpu').tolist()\\n\\n if det_labels != ['background']:\\n for i in range(det_boxes.size(0)):\\n \\n # For every detection, see whether it matches a ground truth\\n match = False\\n for j in range(len(annotations['boxes'])):\\n if annotations['labels'][j] == -1: # this is being set when a ground truth already matched\\n continue\\n match = box_match(det_boxes[i], \\n det_labels[i], \\n torch.Tensor(annotations['boxes'][j]), \\n rev_label_map[annotations['labels'][j]])\\n if match:\\n annotations['labels'][j] = -1\\n break\\n \\n if match: # true positive if the detection is correct and matched a ground truth\\n all_results['true_positive'] += 1\\n else: # false positive if the detection did not match a ground truth\\n all_results['false_positive'] += 1\\n \\n # After all detections were checked, let us see whether the detector missed something\\n for label in annotations['labels']: \\n if label == -1: # This is set after a detection matched this ground truth\\n continue\\n \\n # false negative if we reach this line, since the ground truth object was not found\\n all_results['false_negative'] += 1\",\n \"def test_calculate_precision_happy_path(\\n mock_engine: DicomImagePiiVerifyEngine,\\n gt_labels: List[dict],\\n all_pos: List[dict],\\n expected_result: float,\\n):\\n # Act\\n test_precision = mock_engine.calculate_precision(gt_labels, all_pos)\\n\\n # Assert\\n assert test_precision == expected_result\",\n \"def testdata_matcher(fname1='easy1.png', fname2='easy2.png'):\\n import utool as ut\\n #import vtool as vt\\n from vtool import image as gtool\\n from vtool import features as feattool\\n fpath1 = ut.grab_test_imgpath(fname1)\\n fpath2 = ut.grab_test_imgpath(fname2)\\n featkw = dict(rotation_invariance=True)\\n kpts1, vecs1 = feattool.extract_features(fpath1, **featkw)\\n kpts2, vecs2 = feattool.extract_features(fpath2, **featkw)\\n #if featkw['rotation_invariance']:\\n # print('ori stats 1 ' + ut.get_stats_str(vt.get_oris(kpts2)))\\n # print('ori stats 2 ' + ut.get_stats_str(vt.get_oris(kpts1)))\\n rchip1 = gtool.imread(fpath1)\\n rchip2 = gtool.imread(fpath2)\\n #chip1_shape = vt.gtool.open_image_size(fpath1)\\n chip2_shape = gtool.open_image_size(fpath2)\\n dlen_sqrd2 = chip2_shape[0] ** 2 + chip2_shape[1]\\n testtup = (rchip1, rchip2, kpts1, vecs1, kpts2, vecs2, dlen_sqrd2)\\n return testtup\",\n \"def run_matching(self):\\n paradic = self.cfg['param']['paradic']\\n print 'in run_matching() n_bins = ' +str(paradic['n_bins'])\\n\\n f = open(self.work_dir+'matches.txt','w')\\n matching = self.run_proc(['match_cli', 'keys_im0.txt',\\n 'keys_im1.txt',\\n str(paradic['flag_match']),\\n str(paradic['C_match']),\\n str(paradic['n_hist']),\\n str(paradic['n_ori']),\\n str(paradic['n_bins'])],\\n stdout=f)\\n self.wait_proc(matching, timeout=self.timeout)\\n return 1\",\n \"def drawMatches(orbimg1, kp1, orbimg2, kp2, matches):\\r\\n\\r\\n # Create a new output image that concatenates the two images together\\r\\n # (a.k.a) a montage\\r\\n rows1 = orbimg1.shape[0]\\r\\n cols1 = orbimg1.shape[1]\\r\\n rows2 = orbimg2.shape[0]\\r\\n cols2 = orbimg2.shape[1]\\r\\n\\r\\n # Create the output image\\r\\n # The rows of the output are the largest between the two images\\r\\n # and the columns are simply the sum of the two together\\r\\n # The intent is to make this a colour image, so make this 3 channels\\r\\n out = np.zeros((max([rows1,rows2]),cols1+cols2,3), dtype='uint8')\\r\\n\\r\\n # Place the first image to the left\\r\\n out[:rows1,:cols1] = np.dstack([orbimg1, orbimg1, orbimg1])\\r\\n\\r\\n # Place the next image to the right of it\\r\\n out[:rows2,cols1:] = np.dstack([orbimg2, orbimg2, orbimg2])\\r\\n\\r\\n # For each pair of points we have between both images\\r\\n # draw circles, then connect a line between them\\r\\n for mat in matches:\\r\\n\\r\\n # Get the matching keypoints for each of the images\\r\\n orbimg1_idx = mat.queryIdx\\r\\n orbimg2_idx = mat.trainIdx\\r\\n\\r\\n # x - columns\\r\\n # y - rows\\r\\n (x1,y1) = kp1[orbimg1_idx].pt\\r\\n (x2,y2) = kp2[orbimg2_idx].pt\\r\\n\\r\\n # Draw a small circle at both co-ordinates\\r\\n # radius 4\\r\\n # colour blue\\r\\n # thickness = 1\\r\\n cv2.circle(out, (int(x1),int(y1)), 4, (255, 0, 0), 1) \\r\\n cv2.circle(out, (int(x2)+cols1,int(y2)), 4, (255, 0, 0), 1)\\r\\n\\r\\n # Draw a line in between the two points\\r\\n # thickness = 1\\r\\n # colour blue\\r\\n cv2.line(out, (int(x1),int(y1)), (int(x2)+cols1,int(y2)), (255,0,0), 1)\\r\\n\\r\\n\\r\\n # Show the image\\r\\n #cv2.imshow('Matched Features', out)\\r\\n #cv2.waitKey(0)\\r\\n #cv2.destroyWindow('Matched Features')\\r\\n\\r\\n # Also return the image if you'd like a copy\\r\\n return out\",\n \"def calculate_precision_recall_all_images(\\n all_prediction_boxes, all_gt_boxes, iou_threshold):\\n # Find total true positives, false positives and false negatives\\n # over all images\\n\\n tot_num_tp, tot_num_fp, tot_num_fn = 0, 0, 0\\n\\n for image in range(len(all_prediction_boxes)):\\n prediction_boxes = all_prediction_boxes[image]\\n gt_boxes = all_gt_boxes[image]\\n\\n image_results = calculate_individual_image_result(\\n prediction_boxes,\\n gt_boxes,\\n iou_threshold)\\n\\n tot_num_tp += image_results[\\\"true_pos\\\"]\\n tot_num_fp += image_results[\\\"false_pos\\\"]\\n tot_num_fn += image_results[\\\"false_neg\\\"]\\n\\n # Compute precision, recall\\n precision = calculate_precision(tot_num_tp, tot_num_fp, tot_num_fn)\\n recall = calculate_recall(tot_num_tp, tot_num_fp, tot_num_fn)\\n\\n return (precision, recall)\",\n \"def test_compare_classifier_performance_from_prob_vis_api(experiment_to_use):\\n experiment = experiment_to_use\\n probability = experiment.probabilities\\n viz_outputs = (\\\"pdf\\\", \\\"png\\\")\\n with TemporaryDirectory() as tmpvizdir:\\n for viz_output in viz_outputs:\\n vis_output_pattern_pdf = tmpvizdir + f\\\"/*.{viz_output}\\\"\\n visualize.compare_classifiers_performance_from_prob(\\n [probability, probability],\\n experiment.ground_truth,\\n experiment.ground_truth_metadata,\\n experiment.output_feature_name,\\n top_n_classes=[0],\\n labels_limit=0,\\n model_namess=[\\\"Model1\\\", \\\"Model2\\\"],\\n output_directory=tmpvizdir,\\n file_format=viz_output,\\n )\\n figure_cnt = glob.glob(vis_output_pattern_pdf)\\n assert 1 == len(figure_cnt)\",\n \"def find_matching_points(img1, img2, max_pix_movement=50, normalize=True, show=False):\\n\\n # Initiate ORB detector\\n orb = cv2.ORB_create()\\n\\n # find the keypoints and descriptors with ORB\\n kp1, des1 = orb.detectAndCompute(img1, None)\\n kp2, des2 = orb.detectAndCompute(img2, None)\\n\\n # create BFMatcher object\\n bf = cv2.BFMatcher(cv2.NORM_HAMMING, crossCheck=True)\\n # Match descriptors.\\n matches = bf.match(des1,des2)\\n # Sort them in the order of their distance.\\n matches = sorted(matches, key = lambda x:x.distance)\\n # Draw first 10 matches.\\n if show:\\n img3 = cv2.drawMatches(img1,kp1,img2,kp2,matches[:500], None,flags=2)\\n plt.imshow(img3),plt.show()\\n # Get the matching keypoints for each of the images\\n\\n list_kp1 = []\\n list_kp2 = []\\n for mat in matches:\\n img1_idx = mat.queryIdx\\n img2_idx = mat.trainIdx\\n\\n # x - columns\\n # y - rows\\n list_kp1.append(kp1[img1_idx].pt)\\n list_kp2.append(kp2[img2_idx].pt)\\n\\n n_kp1, n_kp2 = np.float32(list_kp1), np.float32(list_kp2)\\n n_kp1 /= np.asarray([img1.shape[1], img1.shape[0]], np.float32)\\n n_kp2 /= np.asarray([img2.shape[1], img2.shape[0]], np.float32)\\n n_kp1 = n_kp1 * 2. - 1.\\n n_kp2 = n_kp2 * 2. - 1.\\n\\n return np.int32(list_kp1), np.int32(list_kp2), n_kp1, n_kp2\",\n \"def compute_average_precision_per_class(num_true_cases, gt_boxes,\\n\\t\\t\\t\\t\\t\\t\\t\\t\\t\\tprediction_file, iou_threshold, use_2007_metric):\\n\\twith open(prediction_file) as f:\\n\\t\\timage_ids = []\\n\\t\\tboxes = []\\n\\t\\tscores = []\\n\\t\\tfor line in f:\\n\\t\\t\\tt = line.rstrip().split(\\\" \\\")\\n\\t\\t\\timage_ids.append(t[0])\\n\\t\\t\\tscores.append(float(t[1]))\\n\\t\\t\\tbox = torch.tensor([float(v) for v in t[2:]]).unsqueeze(0)\\n\\t\\t\\tbox -= 1.0 # convert to python format where indexes start from 0\\n\\t\\t\\tboxes.append(box)\\n\\t\\tscores = np.array(scores)\\n\\t\\tsorted_indexes = np.argsort(-scores)\\n\\t\\tboxes = [boxes[i] for i in sorted_indexes]\\n\\t\\timage_ids = [image_ids[i] for i in sorted_indexes]\\n\\t\\ttrue_positive = np.zeros(len(image_ids))\\n\\t\\tfalse_positive = np.zeros(len(image_ids))\\n\\t\\tmatched = set()\\n\\t\\tfor i, image_id in enumerate(image_ids):\\n\\t\\t\\tbox = boxes[i]\\n\\t\\t\\tif image_id not in gt_boxes:\\n\\t\\t\\t\\tfalse_positive[i] = 1\\n\\t\\t\\t\\tcontinue\\n\\n\\t\\t\\tgt_box = gt_boxes[image_id]\\n\\t\\t\\tious = box_utils.iou_of(box, gt_box)\\n\\t\\t\\tmax_iou = torch.max(ious).item()\\n\\t\\t\\tmax_arg = torch.argmax(ious).item()\\n\\t\\t\\tif max_iou > iou_threshold:\\n\\t\\t\\t\\tif (image_id, max_arg) not in matched:\\n\\t\\t\\t\\t\\ttrue_positive[i] = 1\\n\\t\\t\\t\\t\\tmatched.add((image_id, max_arg))\\n\\t\\t\\t\\telse:\\n\\t\\t\\t\\t\\tfalse_positive[i] = 1\\n\\t\\t\\telse:\\n\\t\\t\\t\\tfalse_positive[i] = 1\\n\\n\\ttrue_positive = true_positive.cumsum()\\n\\tfalse_positive = false_positive.cumsum()\\n\\tprecision = true_positive / (true_positive + false_positive)\\n\\trecall = true_positive / num_true_cases\\n\\tif use_2007_metric:\\n\\t\\treturn measurements.compute_voc2007_average_precision(precision, recall)\\n\\telse:\\n\\t\\treturn measurements.compute_average_precision(precision, recall)\",\n \"def matching_accuracy(pmat_pred, pmat_gt, ns):\\n device = pmat_pred.device\\n batch_num = pmat_pred.shape[0]\\n\\n pmat_gt = pmat_gt.to(device)\\n\\n assert torch.all((pmat_pred == 0) + (pmat_pred == 1)), 'pmat_pred can noly contain 0/1 elements.'\\n assert torch.all((pmat_gt == 0) + (pmat_gt == 1)), 'pmat_gt should noly contain 0/1 elements.'\\n assert torch.all(torch.sum(pmat_gt, dim=-1) <= 1) and torch.all(torch.sum(pmat_gt, dim=-2) <= 1)\\n assert torch.all(torch.sum(pmat_pred, dim=-1) <= 1) and torch.all(torch.sum(pmat_pred, dim=-2) <= 1)\\n\\n #indices_pred = torch.argmax(pmat_pred, dim=-1)\\n #indices_gt = torch.argmax(pmat_gt, dim=-1)\\n\\n #matched = (indices_gt == indices_pred).type(pmat_pred.dtype)\\n match_num_list = []\\n gt_num_list = []\\n pred_num_list = []\\n acc_gt = []\\n acc_pred = []\\n for b in range(batch_num): #acc_gt,acc_pred\\n #match_num += torch.sum(matched[b, :ns[b]])\\n #total_num += ns[b].item()\\n match_num = torch.sum(pmat_pred[b, :] * pmat_gt[b, :]) + 1e-8 #:ns[b]\\n gt_num = torch.sum(pmat_gt[b, :]) + 1e-8\\n pred_num = torch.sum(pmat_pred[b,:]) + 1e-8\\n match_num_list.append(match_num.cpu().numpy())\\n gt_num_list.append(gt_num.cpu().numpy())\\n pred_num_list.append(pred_num.cpu().numpy())\\n acc_gt.append((match_num/gt_num).cpu().numpy())\\n acc_pred.append((match_num/pred_num).cpu().numpy())\\n\\n return {'acc_gt': np.array(acc_gt),\\n 'acc_pred': np.array(acc_pred),\\n 'match_num': np.array(match_num_list),\\n 'gt_num': np.array(gt_num_list),\\n 'pred_num': np.array(pred_num_list)}\",\n \"def evaluate(self, test_data):\\n test_results = [(self.decide(self.feedforward(x)), y)\\n for (x, y) in test_data]\\n\\n # Check for and keep track of TP's FP's and FN's\\n # Write FP's and FN's to a special directories\\n TP = FP = FN = count = 0\\n TN = 0\\n # what you got = x, what should be = y\\n for ((x, y), (image, gt)) in zip(test_results, test_data):\\n count += 1\\n if x == 1 and y == 1:\\n TP += 1\\n elif x == 1 and y == 0:\\n FP += 1\\n if not self.validating: self.save_Image(FP_PATH, count, image)\\n elif x == 0 and y == 1:\\n FN += 1\\n if not self.validating: self.save_Image(FN_PATH, count, image)\\n \\n if (TP + FP ) == 0:\\n return TP, FP, FN, '---------', '---------','---------'\\n else:\\n false_rate = float(FP) / float(TP + FP)\\n\\n if (TP + FN ) == 0:\\n return TP, FP, FN, '---------', '---------', false_rate\\n else:\\n detect_rate = float(TP) / float(TP + FN)\\n\\n if (TP + FP + FN ) == 0:\\n return TP, FP, FN, '---------', detect_rate, false_rate\\n else:\\n quality_rate = float(TP) / float(TP+FP+FN)\\n \\n return TP, FP, FN, quality_rate, detect_rate, false_rate\",\n \"def match(desc1,desc2):\\n desc1 = array([d/linalg.norm(d) for d in desc1])\\n desc2 = array([d/linalg.norm(d) for d in desc2])\\n dist_ratio = 0.6\\n desc1_size = desc1.shape\\n matchscores = zeros((desc1_size[0],1),'int')\\n desc2t = desc2.T # precompute matrix transpose\\n for i in range(desc1_size[0]):\\n dotprods = dot(desc1[i, :], desc2t) # vector of dot products\\n dotprods *= 0.9999\\n # inverse cosine and sort, return index for features in second image\\n indx = argsort(arccos(dotprods))\\n # check if nearest neighbor has angle less than dist_ratio times 2nd\\n if arccos(dotprods)[indx[0]] < dist_ratio * arccos(dotprods)[indx[1]]:\\n matchscores[i] = int(indx[0])\\n return matchscores\",\n \"def max_disp(left_img, right_img):\\n\\n # Max initial matches\\n MAX_INI_MATCH = np.inf\\n\\n # ORB detector and brute-force matcher\\n detector = cv2.ORB_create(nfeatures=100000, edgeThreshold=5)\\n matcher = cv2.BFMatcher(cv2.NORM_HAMMING, crossCheck=True)\\n\\n # Find the keypoints and descriptors of both images\\n keypoint_left, descriptor_left = detector.detectAndCompute(\\n image=left_img, mask=None\\n )\\n keypoint_right, descriptor_right = detector.detectAndCompute(\\n image=right_img, mask=None\\n )\\n\\n # Find the best matches\\n matches = matcher.match(\\n queryDescriptors=descriptor_left, trainDescriptors=descriptor_right\\n )\\n matches = sorted(matches, key=lambda x: x.distance)\\n\\n if DEBUG:\\n print(f\\\"Initial matches#\\\")\\n print(f\\\"match_lr: {len(matches)}\\\")\\n\\n # Match refinement with RANSAC\\n left_pts = np.float32([keypoint_left[m.queryIdx].pt for m in matches]).reshape(\\n -1, 1, 2\\n )\\n right_pts = np.float32([keypoint_right[m.trainIdx].pt for m in matches]).reshape(\\n -1, 1, 2\\n )\\n homography, mask = cv2.findHomography(left_pts, right_pts, cv2.RANSAC, 5.0)\\n matchesMask = mask.ravel().tolist()\\n\\n ransac_matches = []\\n for (m, keep) in zip(matches, matchesMask):\\n if keep:\\n ransac_matches.append(m)\\n\\n if DEBUG:\\n print(f\\\"After RANSAC:\\\")\\n print(f\\\"match: {len(ransac_matches)}\\\")\\n ransac_img = cv2.drawMatches(\\n left_img,\\n keypoint_left,\\n right_img,\\n keypoint_right,\\n ransac_matches,\\n None,\\n flags=2,\\n )\\n cv2.imwrite(\\\"./debug/RANSAC.png\\\", ransac_img)\\n\\n # Get the range of disparity\\n disp_min = np.inf\\n disp_max = 0\\n\\n for m in ransac_matches:\\n # Keypoints coordinates\\n pt_left = np.float32(keypoint_left[m.queryIdx].pt)\\n pt_right = np.float32(keypoint_right[m.trainIdx].pt)\\n\\n # Compute the distance\\n dist = int(abs(pt_left - pt_right)[0])\\n\\n if disp_max < dist:\\n disp_max = dist\\n\\n # Ceiling the number\\n disp_max += 1\\n\\n return disp_max\",\n \"def calc_TP_FP(groundtruth, eval_boxes, matches_predicate, confidence_threshold=-9e999):\\n TP = 0\\n FP = 0\\n FN = 0\\n for image_key in groundtruth.keys() + eval_boxes.keys():\\n gt_bboxes = set(groundtruth.get(image_key, []))\\n other_bboxes = set(box\\n for box in eval_boxes.get(image_key, [])\\n if box.confidence >= confidence_threshold\\n or box.confidence is None\\n )\\n for gt_bbox in gt_bboxes:\\n # Do any of their boxes match?\\n match = None\\n for other_box in other_bboxes:\\n if matches_predicate(gt_bbox, other_box):\\n match = other_box\\n if match:\\n other_bboxes.remove(match)\\n TP += 1\\n else:\\n FN += 1\\n FP += len(other_bboxes)\\n # if len(other_bboxes):\\n # print \\\"False positive on\\\", image_key\\n return {\\\"TP\\\": TP, \\\"FP\\\": FP, \\\"FN\\\": FN, \\\"TN\\\": 0}\",\n \"def illustrate_matching(self):\\n work_dir = self.work_dir\\n draw_keys_oriented(work_dir+'matching_keys_im0.txt',\\n work_dir+'input_0.orig.png',\\n work_dir+'matching_keys_im0.png')\\n draw_keys_oriented(work_dir+'matching_keys_im1.txt',\\n work_dir+'input_1.orig.png',\\n work_dir+'matching_keys_im1.png')\\n draw_matches(work_dir+'matches.txt',\\n work_dir+'input_0.orig.png',\\n work_dir+'input_1.orig.png',\\n work_dir+'OUTmatches.png')\\n return 1\",\n \"def evaluate_match_quality(matches, thresh=.7):\\n marr = np.zeros((len(matches), 2))\\n quality = np.zeros((len(matches), 1))\\n for i in range(len(matches)):\\n marr[i,0] = matches[i][0].distance\\n marr[i,1] = matches[i][1].distance\\n quality[i] = marr[i, 0] < thresh*marr[i, 1]\",\n \"def main():\\n\\n parser = argparse.ArgumentParser(description='codec_compare')\\n parser.add_argument('path', metavar='DIR',\\n help='path to images folder')\\n args = parser.parse_args()\\n classpath = args.path\\n classname = classpath.split('/')[1]\\n\\n images = set(listdir_full_path(classpath))\\n if len(images) <= 0:\\n print \\\"\\\\033[91m[ERROR]\\\\033[0m\\\" + \\\" no source files in ./images.\\\"\\n sys.exit(1)\\n\\n codeclist_full = set(['aom', 'deepcoder', 'deepcoder-lite', 'fuif', 'fvdo', 'hevc', 'kakadu', 'jpeg',\\n 'pik', 'tat', 'xavs', 'xavs-fast', 'xavs-median', 'webp'])\\n\\n bpp_targets = set([0.06, 0.12, 0.25, 0.50, 0.75, 1.00, 1.50, 2.00])\\n for image in images:\\n width, height, depth = get_dimensions(image, classname)\\n name, imgfmt = os.path.splitext(image)\\n imgfmt = os.path.basename(image).split(\\\".\\\")[-1]\\n derivative_images = []\\n if classname[:6] == 'classB':\\n derivative_images = create_derivatives(image, classname)\\n else:\\n derivative_images.append((image, imgfmt))\\n\\n for derivative_image, pix_fmt in derivative_images:\\n json_dir = 'metrics'\\n mkdir_p(json_dir)\\n json_file = os.path.join(json_dir,\\n os.path.splitext(os.path.basename(derivative_image))[0] + \\\".\\\" + pix_fmt + \\\".json\\\")\\n # if os.path.isfile(json_file):\\n # print \\\"\\\\033[92m[JSON OK]\\\\033[0m \\\" + json_file\\n # continue\\n main_dict = dict()\\n derivative_image_metrics = dict()\\n for codecname in codeclist_full:\\n convertflag = 1\\n caseflag = pix_fmt\\n if (codecname == 'webp' or codecname == 'tat' or 'deepcoder' in codecname) and depth != '8':\\n continue\\n if 'xavs' in codecname and depth != '8' and depth != '10':\\n continue\\n if 'classE' in classname and ('tat' in codecname or 'xavs' in codecname or 'deepcoder' in codecname):\\n continue\\n if codecname == 'kakadu' and classname[:6] == 'classB':\\n convertflag = 0\\n caseflag = imgfmt\\n bpp_target_metrics = dict()\\n for bpp_target in bpp_targets:\\n print(codecname)\\n if codecname == 'aom' and classname[:6] == 'classB':\\n # ('AERIAL2' in image or 'CATS' in image or 'XRAY' in image or 'GOLD' in image or 'TEXTURE1' in image):\\n encoded_image_name = os.path.splitext(os.path.basename(derivative_image))[\\n 0] + '_' + str(bpp_target) + '_' + imgfmt + '.' + 'av1'\\n encoded_image = os.path.join('outputs', codecname, encoded_image_name)\\n decoded_image = os.path.join('outputs', codecname, 'decoded', encoded_image_name + '.' + imgfmt)\\n original_image = image\\n elif codecname == 'kakadu' and classname[:6] == 'classB':\\n encoded_image_name = os.path.splitext(os.path.basename(derivative_image))[\\n 0] + '_' + str(bpp_target) + '_' + imgfmt + '.' + codecname\\n encoded_image = os.path.join('outputs', codecname, encoded_image_name)\\n decoded_image = os.path.join('outputs', codecname, 'decoded', encoded_image_name + '.' + imgfmt)\\n original_image = image\\n elif 'xavs' in codecname and classname[:6] == 'classB':\\n encoded_image_name = os.path.splitext(os.path.basename(derivative_image))[\\n 0] + '_' + str(bpp_target) + '_' + imgfmt + '.' + codecname\\n encoded_image = os.path.join('outputs', codecname, encoded_image_name)\\n decoded_image = os.path.join('outputs', codecname, 'decoded', encoded_image_name + '.' + imgfmt)\\n original_image = image\\n elif codecname == 'fvdo' and classname[:6] == 'classB':\\n encoded_image_name = os.path.splitext(os.path.basename(derivative_image))[\\n 0] + '_' + str(bpp_target) + '_pgm' + '.' + codecname\\n encoded_image = os.path.join('outputs', codecname, encoded_image_name)\\n decoded_image = os.path.join('outputs', codecname, 'decoded', encoded_image_name + '.pgm')\\n original_image = image\\n else:\\n if codecname == 'fuif' and 'tif' in imgfmt:\\n encoded_image_name = os.path.splitext(os.path.basename(derivative_image))[\\n 0] + '.tif_' + str(bpp_target) + '_' + pix_fmt + '.' + codecname\\n elif codecname == 'webp' or codecname == 'tat':\\n encoded_image_name = os.path.splitext(os.path.basename(derivative_image))[\\n 0] + '_' + str(bpp_target) + '_yuv420p.' + codecname\\n else:\\n encoded_image_name = os.path.splitext(os.path.basename(derivative_image))[\\n 0] + '_' + str(bpp_target) + '_' + pix_fmt + '.' + codecname\\n encoded_image = os.path.join('outputs', codecname, encoded_image_name)\\n decoded_image_path = os.path.join('outputs', codecname, 'decoded')\\n decoded_image = ''\\n for decodedfile in os.listdir(decoded_image_path):\\n encoderoot = '_'.join(os.path.splitext(os.path.basename(encoded_image_name))[0].split('_')[:-1])\\n if encoderoot in decodedfile:\\n if ('tat' in codecname or 'webp' in codecname) and os.path.splitext(os.path.basename(decodedfile))[1] == '.yuv':\\n decoded_image = os.path.join('outputs', codecname, 'decoded', decodedfile)\\n print(decoded_image)\\n if ('tat' not in codecname or 'webp' not in codecname) and os.path.splitext(os.path.basename(decodedfile))[1] != '.yuv':\\n decoded_image = os.path.join('outputs', codecname, 'decoded', decodedfile)\\n if 'classE' not in classname and 'classB' not in classname and os.path.isfile(decoded_image):\\n decoded_image = convert_decoded(decoded_image, width, height, depth, codecname)\\n original_image = convert_decoded(derivative_image, width, height, depth, 'reference')\\n else:\\n original_image = derivative_image\\n\\n print('Reference:' + original_image)\\n print('Encoded:' + encoded_image)\\n print('Decoded:' + decoded_image)\\n if (os.path.isfile(original_image) and os.path.isfile(decoded_image) and os.path.isfile(encoded_image)):\\n if 'classE' in classname:\\n metrics = compute_metrics_HDR(original_image, decoded_image, encoded_image, bpp_target,\\n codecname, width, height, pix_fmt, depth)\\n\\n elif 'classB' in classname:\\n metrics = compute_metrics(original_image, decoded_image, encoded_image, bpp_target, codecname,\\n width, height, pix_fmt)\\n else:\\n metrics = compute_metrics_SDR(original_image, decoded_image, encoded_image, bpp_target,\\n codecname, width,\\n height, imgfmt, depth)\\n measured_bpp = (os.path.getsize(encoded_image) * 1.024 * 8) / (float((int(width) * int(height))))\\n bpp_target_metrics[measured_bpp] = metrics\\n else:\\n continue\\n \\n derivative_image_metrics[codecname] = bpp_target_metrics\\n main_dict[derivative_image] = derivative_image_metrics\\n\\n mkdir_p(json_dir)\\n with open(json_file, 'w') as f:\\n f.write(json.dumps(main_dict, indent=2))\",\n \"def evaluation(eval_dict, faster_rcnn, test_num=Config.eval_num):\\n bboxes, labels, scores = list(), list(), list()\\n gt_bboxes, gt_labels, gt_difficult = list(), list(), list()\\n for i, [img_dir, img_info] in tqdm(enumerate(eval_dict.items())):\\n img, img_info = rescale_image(img_dir, img_info, flip=False)\\n img_tensor = create_img_tensor(img)\\n box, score, label = faster_rcnn.predict(img_tensor)\\n gt_bbox = np.array(img_info['objects'])[:, 1:5].astype(np.float32)\\n gt_label = np.array(img_info['objects'])[:, 0]\\n difficult = np.array(img_info['difficult'])\\n\\n # label from text to number\\n gt_label = text_to_num(gt_label)\\n\\n bboxes.append(box)\\n labels.append(label)\\n scores.append(score)\\n gt_bboxes.append(gt_bbox)\\n gt_labels.append(gt_label)\\n gt_difficult.append(difficult)\\n if i == test_num:\\n break\\n\\n AP, mAP = calc_map(bboxes, labels, scores, gt_bboxes, gt_labels, gt_difficult, use_07_metric=True)\\n\\n return mAP\",\n \"def test_calculate_recall_happy_path(\\n mock_engine: DicomImagePiiVerifyEngine,\\n gt_labels: dict,\\n all_pos: dict,\\n expected_result: float,\\n):\\n # Act\\n test_recall = mock_engine.calculate_recall(gt_labels, all_pos)\\n\\n # Assert\\n assert test_recall == expected_result\",\n \"def compute_transform(self, image_pair, figure):\\t\\n\\t\\timage_pair = [int(image_pair[1]), int(image_pair[3])]\\n\\t\\tindices = np.where(np.all(np.abs(self.Matches[\\\"image_pairs\\\"][0] - image_pair) == 0, axis=1)) \\n\\t\\tfeature_pairs = self.Matches[\\\"image_match_pairs\\\"][0][indices[0][0]]\\n\\t\\tfeature_pair_scores = self.Matches[\\\"image_feature_scores\\\"][0][indices[0][0]] \\n\\n\\t\\ti1_loc = self.Database[\\\"image_locs\\\"][np.where(np.array(self.Database[\\\"image_idx\\\"]) == image_pair[0])[0][0]]\\n\\t\\ti2_loc = self.Database[\\\"image_locs\\\"][np.where(np.array(self.Database[\\\"image_idx\\\"]) == image_pair[1])[0][0]]\\n\\n\\t\\tself.image1 = cv2.cvtColor(extract_image(figure, i1_loc), cv2.COLOR_BGR2GRAY)\\n\\t\\tself.image2 = cv2.cvtColor(extract_image(figure, i2_loc), cv2.COLOR_BGR2GRAY)\\n\\t\\t# self.image1 = extract_image(figure, i1_loc)\\n\\t\\t# self.image2 = extract_image(figure, i2_loc)\\n\\n\\t\\tself.match_alignment = {}; self.match_alignment_scores = {}\\n\\t\\tfor pair, score in zip(feature_pairs, feature_pair_scores):\\n\\t\\t\\ti1 = self.Database[\\\"image_idx\\\"][pair[0]]\\n\\t\\t\\ti2 = self.Database[\\\"image_idx\\\"][pair[1]]\\n\\n\\t\\t\\tif i1 == image_pair[0]:\\n\\t\\t\\t\\ttarget = pair[0] \\n\\t\\t\\t\\tstart = pair[1]\\n\\t\\t\\t\\tif target not in self.match_alignment.keys():\\n\\t\\t\\t\\t\\tself.init_match_alignment(target)\\n\\t\\t\\telse:\\n\\t\\t\\t\\ttarget = pair[1]\\n\\t\\t\\t\\tstart = pair[0]\\t\\t\\t\\t\\n\\t\\t\\t\\tif target not in self.match_alignment.keys():\\n\\t\\t\\t\\t\\tself.init_match_alignment(target)\\n\\n\\t\\t\\t# get ellipse and convert to three cardinal points\\n\\t\\t\\ttarget_ellipse = self.Database[\\\"orientation\\\"][target]\\n\\t\\t\\tstart_ellipse = self.Database[\\\"orientation\\\"][start]\\n\\t\\t\\ttarget_points, target_params = utils.ellipse2points(target_ellipse)\\n\\t\\t\\tstart_points, start_params = utils.ellipse2points(start_ellipse)\\n\\n\\t\\t\\t# compute angle between Major axes\\n\\t\\t\\trotation_angle = int(np.abs(target_params[0] - start_params[0]))\\n\\t\\t\\treflection_truth = target_params[1] * start_params[1]\\n\\t\\t\\tif reflection_truth > 0:\\n\\t\\t\\t\\treflection_truth = \\\"No\\\"\\n\\t\\t\\telse:\\n\\t\\t\\t\\treflection_truth = \\\"Yes\\\"\\n\\t\\t\\tresolution_perc = round(((target_params[2] / start_params[2]) - 1), 1)\\n\\n\\t\\t\\t# extract features for histogram matching\\n\\t\\t\\ttarget_bloc = self.Database[\\\"blot_locs\\\"][target]\\n\\t\\t\\tstart_bloc = self.Database[\\\"blot_locs\\\"][start]\\n\\t\\t\\ttarget_feature = extract_image(self.image1, target_bloc, ex=5)\\n\\t\\t\\tfinal_feature = extract_image(self.image2, start_bloc, ex=5) \\n\\n\\t\\t\\t#find affine transform to match\\n\\t\\t\\tif self.AFFINE:\\n\\t\\t\\t\\tT = utils.compute_affine(start_points, target_points)\\n\\t\\t\\t\\timage2_warped = cv2.warpAffine(self.image2.copy(), T, (self.image1.shape[1], self.image1.shape[0]))\\n\\t\\t\\t\\tfinal_feature = extract_image(image2_warped, target_bloc, ex=5)\\n\\t\\t\\t\\n\\t\\t\\t# match hist\\n\\t\\t\\tif self.CONTRAST:\\n\\t\\t\\t\\tfinal_feature = utils.histogram_match(final_feature, target_feature)\\n\\n\\t\\t\\tself.match_alignment[target].append((final_feature, start, rotation_angle, reflection_truth, \\\\\\n\\t\\t\\t\\t\\t\\t\\t\\t\\t\\t\\t\\t resolution_perc, score))\\n\\t\\t\\tself.match_alignment_scores[target].append(score)\",\n \"def evaluate_rouge_match(ref_data, pred_data, align=\\\"rouge\\\"):\\n\\n scorer = RougeScorer([\\\"rougeL\\\"])\\n agg = scoring.BootstrapAggregator(confidence_interval=0.95)\\n sample_evaluated = 0\\n\\n for tid in tqdm(pred_data):\\n cur_sys = pred_data[tid][\\\"pred\\\"]\\n if not tid in ref_data: continue\\n cur_ref_lst = [x[\\\"tgt_words\\\"] for x in ref_data[tid]]\\n if len(cur_ref_lst) > 0:\\n if align == \\\"rouge\\\":\\n best_rouge_scrs = __rouge_pairs(cur_sys, cur_ref_lst, scorer)\\n agg.add_scores(best_rouge_scrs)\\n else:\\n best_paired_ref, _ = utils.pair_ref_sys_by_coverage(cur_sys, cur_ref_lst)\\n agg.add_scores(scorer.score(best_paired_ref, cur_sys))\\n sample_evaluated += 1\\n\\n print(sample_evaluated)\\n result = agg.aggregate()\\n for rouge_type in result:\\n print(rouge_type)\\n print(result[rouge_type].mid)\\n\\n print(agg.aggregate())\\n return\",\n \"def all_match():\\n S1=Spectrum.Spectrum()\\n S1.add_peak(50.7,234)\\n S1.add_peak(54.6,585)\\n S1.add_peak(60.7,773)\\n S1.add_peak(65.6,387)\\n S1.add_peak(87.7,546)\\n S1.add_peak(104.6,598)\\n S1.pep_mass=100\\n S1.euclidean_scale()\\n\\n S2=Spectrum.Spectrum()\\n S2.add_peak(50.5,234/2)\\n S2.add_peak(54.8,585/2)\\n S2.add_peak(61.0,773/2)\\n S2.add_peak(65.4,387/2)\\n S2.add_peak(88.0,546/2)\\n S2.add_peak(104.3,598/2)\\n S2.pep_mass=100\\n S2.euclidean_scale()\\n\\n score,peaks=similarity.cosine_score_max(S1,S2)\\n assert peaks==6, \\\"Incorrect number of peaks matched with greedy method\\\"\\n assert math.isclose(score,1.0), \\\"Incorrect score with greedy method\\\"\\n\\n score,peaks=similarity.cosine_score_greedy(S1,S2)\\n assert peaks==6, \\\"Incorrect number of peaks matched with maximum weighted method\\\"\\n assert math.isclose(score,1.0), \\\"Incorrect score with maximum weighted method\\\"\",\n \"def find(image):\\n keypoint, description = describe(image)\\n # load keypoints, descriptions from mongodb\\n\\n bf = cv2.BFMatcher(cv2.NORM_HAMMING, crossCheck=True)\\n\\n best_match_size = float(\\\"inf\\\")\\n best_match_index = -1\\n match_index = 0\\n best_matches = 0\\n\\n for desc in descriptions:\\n matches = bf.match(desc,description)\\n matches = sorted(matches, key = lambda x:x.distance)\\n if len(matches) > 0:\\n match_size = sum(x.distance for x in matches[:10])\\n\\n print \\\"match size is \\\", match_size\\n if match_size < best_match_size:\\n best_match_size = match_size\\n best_match_index = match_index\\n best_matches = matches\\n\\n match_index += 1\\n\\n needle_color = cv2.imread('needle-stripped.png')[:,:,::-1] # needle\\n best_match_image = cv2.imread(\\\"haystack/\\\"+files[best_match_index])\\n print \\\"best match is \\\", files[best_match_index]\\n\\n # Draw first 10 matches.\\n outImg = cv2.imread(\\\"output/outImg.png\\\")\\n match = cv2.drawMatches(needle_color,keypoint,best_match_image[:,:,::-1],keypoints[best_match_index],best_matches[-20:],outImg, flags=6)\\n\\n plt.imshow(match),plt.show()\\n return\",\n \"def img_compare(file1, file2):\\n # read image\\n img1 = Image.open(file1)\\n img2 = Image.open(file2)\\n\\n # resize \\n size = 128, 128\\n img1_res = img_resize(img1, size)\\n img2_res = img_resize(img2, size)\\n\\n img1_res.save(\\\"img_1.thumbnail\\\", \\\"JPEG\\\")\\n img2_res.save(\\\"img_2.thumbnail\\\", \\\"JPEG\\\")\\n\\n # convert to gray scale\\n img1_grayscale = img1_res.convert('LA')\\n img1_grayscale.save(\\\"img_1_grayscale.png\\\")\\n\\n img2_grayscale = img2_res.convert('LA')\\n img2_grayscale.save(\\\"img_2_grayscale.png\\\")\\n\\n # normalise\\n img1_norm = normalize(np.array(img1_grayscale.getdata()).astype(float))\\n img2_norm = normalize(np.array(img2_grayscale.getdata()).astype(float))\\n\\n try:\\n # compare two images\\n diff = img1_norm - img2_norm\\n m_norm = sum(abs(diff)) # Manhattan norm\\n z_norm = norm(diff.ravel(), 0) # Zero norm\\n\\n # print(\\\"Manhattan norm:\\\", m_norm, \\\"/ per pixel:\\\", m_norm/img1_norm.size)\\n # print(\\\"Zero norm:\\\", z_norm, \\\"/ per pixel:\\\", z_norm*1.0/img1_norm.size)\\n\\n return m_norm/img1_norm.size, float(z_norm) / img1_norm.size\\n except:\\n return 100, 100\",\n \"def match(desc1,desc2):\\n\\t\\n\\tdesc1 = array([d/linalg.norm(d) for d in desc1])\\n\\tdesc2 = array([d/linalg.norm(d) for d in desc2])\\n\\t\\n\\tdist_ratio = 0.6\\n\\tdesc1_size = desc1.shape\\n\\t\\n\\tmatchscores = zeros((desc1_size[0],1))\\n\\tdesc2t = desc2.T #precompute matrix transpose\\n\\tfor i in range(desc1_size[0]):\\n\\t\\tdotprods = dot(desc1[i,:],desc2t) #vector of dot products\\n\\t\\tdotprods = 0.9999*dotprods\\n\\t\\t#inverse cosine and sort, return index for features in second image\\n\\t\\tindx = argsort(arccos(dotprods))\\n\\t\\t\\n\\t\\t#check if nearest neighbor has angle less than dist_ratio times 2nd\\n#\\t\\tif arccos(dotprods)[indx[0]] < dist_ratio * arccos(dotprods)[indx[1]]:\\n\\t\\tmatchscores[i] = int(indx[0])\\n\\t\\n\\treturn matchscores\",\n \"def drawMatches(image_1, image_1_keypoints, image_2, image_2_keypoints, matches):\\n # Compute number of channels.\\n num_channels = 1\\n if len(image_1.shape) == 3:\\n num_channels = image_1.shape[2]\\n # Separation between images.\\n margin = 10\\n # Create an array that will fit both images (with a margin of 10 to separate\\n # the two images)\\n joined_image = np.zeros((max(image_1.shape[0], image_2.shape[0]),\\n image_1.shape[1] + image_2.shape[1] + margin,\\n 3))\\n if num_channels == 1:\\n for channel_idx in range(3):\\n joined_image[:image_1.shape[0],\\n :image_1.shape[1],\\n channel_idx] = image_1\\n joined_image[:image_2.shape[0],\\n image_1.shape[1] + margin:,\\n channel_idx] = image_2\\n else:\\n joined_image[:image_1.shape[0], :image_1.shape[1]] = image_1\\n joined_image[:image_2.shape[0], image_1.shape[1] + margin:] = image_2\\n\\n for match in matches:\\n image_1_point = (int(image_1_keypoints[match.queryIdx].pt[0]),\\n int(image_1_keypoints[match.queryIdx].pt[1]))\\n image_2_point = (int(image_2_keypoints[match.trainIdx].pt[0] + \\\\\\n image_1.shape[1] + margin),\\n int(image_2_keypoints[match.trainIdx].pt[1]))\\n\\n cv2.circle(joined_image, image_1_point, 5, (0, 0, 255), thickness = -1)\\n cv2.circle(joined_image, image_2_point, 5, (0, 255, 0), thickness = -1)\\n cv2.line(joined_image, image_1_point, image_2_point, (255, 0, 0), \\\\\\n thickness = 3)\\n return joined_image\",\n \"def compare_images(self, df):\\n score_list = []\\n elapsed_list = []\\n for index,row in df.iterrows():\\n start_time = time.time()\\n image1 = imread(row['image1'],as_gray=\\\"True\\\")\\n image2 = imread(row['image2'],as_gray=\\\"True\\\")\\n if(image1.shape != image2.shape):\\n image2 = resize(image2, (image1.shape)) # if images are of different dimensions, crop the second image based on first\\n score = structural_similarity(image1, image2)\\n new_score = self.__convert_similarity_score(score)\\n elapsed_time = (time.time() - start_time)\\n score_list.append(round(new_score,2))\\n elapsed_list.append(elapsed_time)\\n output_df = self.__create_output_csv(df, score_list, elapsed_list)\\n return output_df\",\n \"def test(label_dict, image_ids, args):\\n\\n log_header('Starting image retrieval preparations')\\n\\n queries = []\\n\\n if args.test_image is not None:\\n queries.append(args.test_image)\\n else:\\n logging.info('Generating random test queries')\\n\\n number_of_images = len(image_ids)\\n\\n # Generate random queries to use in the test procedure\\n for i in range(args.k):\\n queries.append(image_ids[random.randint(0, number_of_images - 1)])\\n\\n # Calculate score for the retrieved images\\n calculate_score(\\n label_dict, queries, retrieve_similar_images(queries, args)\\n )\",\n \"def drawMatches(img1, kp1, img2, kp2, matches):\\n\\n # Create a new output image that concatenates the two images together\\n # (a.k.a) a montage\\n rows1 = img1.shape[0]\\n cols1 = img1.shape[1]\\n rows2 = img2.shape[0]\\n cols2 = img2.shape[1]\\n\\n out = np.zeros((max([rows1,rows2]),cols1+cols2,3), dtype='uint8')\\n\\n # Place the first image to the left\\n out[:rows1,:cols1] = np.dstack([img1, img1, img1])\\n\\n # Place the next image to the right of it\\n out[:rows2,cols1:] = np.dstack([img2, img2, img2])\\n\\n # For each pair of points we have between both images\\n # draw circles, then connect a line between them\\n for mat in matches:\\n\\n # Get the matching keypoints for each of the images\\n img1_idx = mat.queryIdx\\n img2_idx = mat.trainIdx\\n\\n # x - columns\\n # y - rows\\n (x1,y1) = kp1[img1_idx].pt\\n (x2,y2) = kp2[img2_idx].pt\\n\\n # Draw a small circle at both co-ordinates\\n # radius 4\\n # colour blue\\n # thickness = 1\\n cv2.circle(out, (int(x1),int(y1)), 4, (255, 0, 0), 1) \\n cv2.circle(out, (int(x2)+cols1,int(y2)), 4, (255, 0, 0), 1)\\n\\n # Draw a line in between the two points\\n # thickness = 1\\n # colour blue\\n cv2.line(out, (int(x1),int(y1)), (int(x2)+cols1,int(y2)), (255, 0, 0), 1)\\n\\n\\n # Show the image\\n\\n # Also return the image if you'd like a copy\\n return out\",\n \"def evaluate(ground_truth, prediction):\\n\\n def prfs_to_dict(prfs):\\n \\\"\\\"\\\"Returns a precision_recall_fscore_support() result as a dict.\\\"\\\"\\\"\\n return {\\\"precision\\\": prfs[0], \\\"recall\\\": prfs[1], \\\"fscore\\\": prfs[2]}\\n\\n results = {}\\n items_count = len(ground_truth)\\n\\n # accuracy\\n accuracy = accuracy_score(ground_truth, prediction)\\n results[\\\"accuracy\\\"] = accuracy\\n\\n # confusion matrix\\n categories = set(ground_truth) | set(prediction)\\n confusions = {\\n gold: {pred: 0 for pred in categories} for gold in categories\\n }\\n for g, p in zip(ground_truth, prediction):\\n confusions[g][p] += 1\\n results[\\\"confusions\\\"] = confusions\\n\\n # class wise precision, recall & f1\\n classwise = precision_recall_fscore_support(\\n ground_truth, prediction, average=None, warn_for=()\\n )\\n results[\\\"true_cat_dist\\\"] = list(classwise[-1])\\n results[\\\"classwise\\\"] = {\\n str(cl): prfs_to_dict(\\n [classwise[0][cl], classwise[1][cl], classwise[2][cl]]\\n )\\n for cl in categories\\n }\\n\\n # average precision, recall & f1\\n results[\\\"macro_avg\\\"] = prfs_to_dict(\\n precision_recall_fscore_support(\\n ground_truth,\\n prediction,\\n average=\\\"macro\\\",\\n pos_label=None,\\n warn_for=(),\\n )\\n )\\n results[\\\"micro_avg\\\"] = prfs_to_dict(\\n precision_recall_fscore_support(\\n ground_truth,\\n prediction,\\n average=\\\"micro\\\",\\n pos_label=None,\\n warn_for=(),\\n )\\n )\\n results[\\\"weigh_avg\\\"] = prfs_to_dict(\\n precision_recall_fscore_support(\\n ground_truth,\\n prediction,\\n average=\\\"weighted\\\",\\n pos_label=None,\\n warn_for=(),\\n )\\n )\\n\\n # marginals\\n gold_category_distribution = {\\n g: sum([confusions[g][p] for p in categories]) for g in categories\\n }\\n pred_category_distribution = {\\n p: sum([confusions[g][p] for g in categories]) for p in categories\\n }\\n\\n # kappa\\n expected_agreement_fleiss = sum(\\n [\\n (\\n (gold_category_distribution[c] + pred_category_distribution[c])\\n / (2.0 * items_count)\\n )\\n ** 2\\n for c in categories\\n ]\\n )\\n expected_agreement_cohen = sum(\\n [\\n (float(gold_category_distribution[c]) / items_count)\\n * (float(pred_category_distribution[c]) / items_count)\\n for c in categories\\n ]\\n )\\n kappa_fleiss = (\\n 1.0\\n * (accuracy - expected_agreement_fleiss)\\n / (1 - expected_agreement_fleiss)\\n )\\n kappa_cohen = (\\n 1.0\\n * (accuracy - expected_agreement_cohen)\\n / (1 - expected_agreement_cohen)\\n )\\n results[\\\"k_fleiss\\\"] = {\\n \\\"k\\\": kappa_fleiss,\\n \\\"AE\\\": expected_agreement_fleiss,\\n \\\"AO\\\": accuracy,\\n }\\n results[\\\"k_cohen\\\"] = {\\n \\\"k\\\": kappa_cohen,\\n \\\"AE\\\": expected_agreement_cohen,\\n \\\"AO\\\": accuracy,\\n }\\n\\n return results\",\n \"def get_recall_precision(imgs_truth, imgs_pred, is_hit):\\n # Numbers of relevant and detect boxes, respectively.\\n num_rel = sum([len(img.boxes) for img in imgs_truth])\\n num_det = sum([len(img.boxes) for img in imgs_pred])\\n\\n boxes_for_fname_truth = {img.fname: img.boxes for img in imgs_truth}\\n boxes_for_fname_pred = {img.fname: img.boxes for img in imgs_pred}\\n\\n num_hit = 0\\n for fname, boxes in boxes_for_fname_truth.items():\\n if fname not in boxes_for_fname_pred:\\n continue\\n\\n num_hit += get_num_hit(boxes, boxes_for_fname_pred[fname], is_hit)\\n\\n return num_hit / num_rel, num_hit / num_det\",\n \"def compute(self, result_file_dict):\\r\\n for part in self.parts:\\r\\n #=====================Need to change, temporal=========================\\r\\n if part == 'train':\\r\\n continue # because the train not have the label\\r\\n #=======================================================================\\r\\n gt = self.gt_dict[part]\\r\\n result_file = result_file_dict[part]\\r\\n # import ipdb; ipdb.set_trace()\\r\\n for key, item in result_file.items():\\r\\n self._result_name = item\\r\\n # score_records, num_videos = self.load_results(result_file)\\r\\n score_records, num_videos = self.load_results(item)\\r\\n logger.info(f'Compute Metric of {item}')\\r\\n assert num_videos == len(gt), f'the number of saved videos does not match the ground truth, {num_videos} != {len(gt)}'\\r\\n temp_result = self.eval_method(score_records, gt, str(key))\\r\\n if temp_result > self.optimal_resulst:\\r\\n self.optimal_resulst = temp_result\\r\\n \\r\\n return self.optimal_resulst\",\n \"def perform_match(self, parser_manager):\\n log = logging.getLogger(\\\".\\\".join([__name__, self.__class__.__name__]))\\n log.addFilter(logconfig.ThreadContextFilter())\\n extra = self._config_align.all_to_json()\\n extra.update(self._config_crystal.all_to_json())\\n log = logging.LoggerAdapter(log, extra)\\n log.info(\\\"Matching Started\\\")\\n log.debug(extra)\\n\\n input_poi = parser_manager.parse_selected_points_from_args()\\n beamline_image = parser_manager.get_focused_image()\\n parser_manager.save_focused_image(beamline_image)\\n focused_image_path = parser_manager.get_focused_image_path()\\n formulatrix_image_path = parser_manager.get_formulatrix_image_path()\\n job_id = parser_manager.get_job_id()\\n\\n # Create the images\\n image1 = Image.from_file(formulatrix_image_path)\\n image2 = beamline_image\\n\\n # Create results object\\n service_result = ServiceResult(job_id, formulatrix_image_path, focused_image_path)\\n\\n # Perform alignment\\n try:\\n\\n aligned_images, scaled_poi = self._perform_alignment(image1, image2, input_poi)\\n service_result.set_image_alignment_results(aligned_images)\\n\\n # Perform Crystal Matching - only proceed if we have a valid alignment\\n if aligned_images.alignment_status_code() == ALIGNED_IMAGE_STATUS_OK:\\n match_results = self._perform_matching(aligned_images, scaled_poi, parser_manager)\\n\\n service_result.append_crystal_matching_results(match_results)\\n\\n except Exception as e:\\n if sys.version_info[0] < 3:\\n log.error(\\\"ERROR: \\\" + e.message)\\n else:\\n log.error(\\\"ERROR: \\\" + str(e))\\n service_result.set_err_state(e)\\n\\n return service_result\",\n \"def ransac(keypoints1, keypoints2, matches, sampling_ratio=0.5, n_iters=500, threshold=20):\\n N = matches.shape[0]\\n n_samples = int(N * sampling_ratio)\\n\\n # Please note that coordinates are in the format (y, x)\\n matched1 = pad(keypoints1[matches[:,0]])\\n matched2 = pad(keypoints2[matches[:,1]])\\n matched1_unpad = keypoints1[matches[:,0]]\\n matched2_unpad = keypoints2[matches[:,1]]\\n\\n max_inliers = np.zeros(N)\\n n_inliers = 0\\n\\n # RANSAC iteration start\\n ### YOUR CODE HERE\\n raise NotImplementedError() # Delete this line\\n ### END YOUR CODE\\n return H, matches[max_inliers]\",\n \"def pixel_metrics(truth: Node, prediction: Node) -> Tuple[float, float, float]:\\n recall, precision, fscore = 0, 0, 0\\n\\n intersection_truth = bounding_box_intersection(truth.bounding_box,\\n prediction.bounding_box)\\n if intersection_truth is None:\\n logging.debug('No intersection for Nodes: t={0},'\\n ' p={1}'.format(truth.bounding_box,\\n prediction.bounding_box))\\n return recall, precision, fscore\\n\\n intersection_pred = bounding_box_intersection(prediction.bounding_box,\\n truth.bounding_box)\\n\\n logging.debug('Found intersection for Nodes: t={0},'\\n ' p={1}'.format(truth.bounding_box,\\n prediction.bounding_box))\\n\\n tt, tl, tb, tr = intersection_truth\\n pt, pl, pb, pr = intersection_pred\\n crop_truth = truth.mask[tt:tb, tl:tr]\\n crop_pred = prediction.mask[pt:pb, pl:pr]\\n\\n # Assumes the mask values are 1...\\n\\n n_truth = float(truth.mask.sum())\\n n_pred = float(prediction.mask.sum())\\n n_common = float((crop_truth * crop_pred).sum())\\n\\n # There are no zero-pixel objects, but the overlap may be nonzero\\n if n_truth == 0:\\n recall = 0.0\\n precision = 0.0\\n elif n_pred == 0:\\n recall = 0.0\\n precision = 0.0\\n else:\\n recall = n_common / n_truth\\n precision = n_common / n_pred\\n\\n if (recall == 0) or (precision == 0):\\n fscore = 0\\n else:\\n fscore = 2 * recall * precision / (recall + precision)\\n\\n return recall, precision, fscore\",\n \"def calc_accuracy(self, top_matches):\\n # Calculate Translation accuracy\\n # Accuracy at K\\n for k in [1, 5, 10]:\\n top_k_matches = top_matches[:, :k]\\n # Checks for one match\\n one_matching = (top_k_matches == self.tgt_ind[:, None]).sum(1)\\n # Checks for multiple translations\\n matching = {}\\n for i, src_id in enumerate(self.src_ind):\\n matching[src_id] = min(matching.get(src_id, 0) + one_matching[i], 1)\\n precision_at_k = 100 * np.mean(list(matching.values()))\\n print(\\\"%i source words - Precision at k = %i: %f\\\" % (len(matching), k, precision_at_k))\",\n \"def run_compare(self):\\n misses = []\\n for row in range(self.data.get_number_of_rows()):\\n ref_sentence = self.data.get_row(row)[self.data.get_gold()]\\n results = {}\\n for team, team_sentence in self.data.get_row_for_teams(self.evaluator.get_teams(row), row).iteritems():\\n results[team] = self.get_sentence_score(ref_sentence, team_sentence)\\n misses.append(self.evaluator.compare_all(results, row))\\n print np.median(misses), np.mean(misses)\\n return np.median(misses)\",\n \"def compute_similarity(im1, im2, test_id):\\n\\n results = dict()\\n results['id'] = test_id\\n results['test_im'] = im2\\n results['mse_value'], results['mse_map'] = compute_mse(im1, im2)\\n results['ssim_value'], results['ssim_map'] = compute_ssim(im1, im2, 5)\\n results['cw_ssim_value'], results['cw_ssim_map'] = compute_cw_ssim(im1, im2, 30)\\n results['gms_value'], results['gms_map'] = compute_gms(im1, im2)\\n results['fsim_value'], results['pc_max_map'] = compute_fsim(im1, im2)\\n\\n return results\",\n \"def evaluate(self, test_dir='data/dev', target='real'):\\n test_data = {c: os.path.join(test_dir, c) for c in self.classes}\\n if not target in test_data:\\n print('Error: target class does not exist in test data.')\\n return\\n outcomes = {'TP': 0, 'TN': 0, 'FP': 0, 'FN': 0}\\n # >>> YOUR ANSWER HERE\\n data = []\\n for c in test_data:\\n docs = open(test_data[c]).readlines()\\n for doc in docs:\\n preprocess_doc = doc.strip()\\n data.append((c, preprocess_doc))\\n for item in data:\\n predict_ans = self.predict(item[1])\\n if item[0] == 'real':\\n if predict_ans == 'real':\\n outcomes['TP'] += 1\\n else:\\n outcomes['FN'] += 1\\n else:\\n if predict_ans == 'real':\\n outcomes['FP'] += 1\\n else:\\n outcomes['TN'] += 1\\n precision = outcomes['TP'] / (outcomes['TP'] + outcomes['FP']) # replace with equation for precision\\n recall = outcomes['TP'] / (outcomes['TP'] + outcomes['FN']) # replace with equation for recall\\n f1_score = 2 * ((precision * recall) / (precision + recall)) # replace with equation for f1\\n # >>> END YOUR ANSWER\\n return precision, recall, f1_score\",\n \"def _calculate_similarity(self):\\n self._logger.info(\\\"Calculating the similarity between images.\\\")\\n\\n # Create a helper function to simplify the loops below:\\n def get_wrapper(fdx, mdx):\\n wrapper = pos_wrappers.image_similarity_wrapper(\\n reference_image=self.f['src_gray'](idx=fdx),\\n moving_image=self.f['src_gray'](idx=mdx),\\n affine_transformation=self.f['part_transf'](mIdx=mdx, fIdx=fdx))\\n return copy.copy(wrapper)\\n\\n commands = [] # Will hold commands for calculating the similarity\\n\\n # Will hold (moving, fixed) images partial_transforms basically: all\\n # partial transformations array\\n partial_transforms = []\\n\\n self._logger.debug(\\\"Generating similarity measure warppers.\\\")\\n for moving_slice in self.options.slice_range:\\n # Get all fixed images to which given moving slice will be aligned:\\n tpair = list(flatten(self._get_slice_pair(moving_slice)))\\n\\n # Append partial transformations for given moving slice to the\\n # global partial transformations array\\n partial_transforms.append(tpair)\\n\\n # Generate wrapper for measuring similarity for a given partial\\n # transformation.\\n for mdx, fdx in tpair:\\n commands.append(get_wrapper(fdx, mdx))\\n\\n # Execute and commands and workflow the similarity measurements.\\n stdout, stderr = self.execute(commands)\\n simmilarity = map(lambda x: float(x.strip()),\\n stdout.strip().split(\\\"\\\\n\\\"))\\n simmilarity = dict(zip(flatten(partial_transforms), simmilarity))\\n\\n self._logger.debug(\\\"Generating graph edges.\\\")\\n graph_connections = []\\n\\n # Lambda defines slice skipping is preffered (lower l), or reluctant\\n # to slice skipping (higher)\\n l = self.options.graphEdgeLambda\\n\\n for (mdx, fdx), s in simmilarity.iteritems():\\n w = (1.0 + s) * abs(mdx - fdx) * (1.0 + l) ** (abs(mdx - fdx))\\n graph_connections.append((fdx, mdx, w))\\n\\n self._logger.info(\\\"Creating a graph based on image similarities.\\\")\\n # Generate the graph basen on the weight of the edges\\n self.G = nx.DiGraph()\\n self.G.add_weighted_edges_from(graph_connections)\\n\\n self._logger.debug(\\\"Saving the graph to a file.\\\")\\n # Save the edges for some further analysis.\\n nx.write_weighted_edgelist(self.G,\\n self.f['graph_edges'](sign=self.signature))\\n\\n # Also, save the individual similarity metrics:\\n simm_fh = open(self.f['similarity'](sign=self.signature), 'w')\\n for (mdx, fdx), s in sorted(simmilarity.iteritems()):\\n simm_fh.write(\\\"%d %d %f\\\\n\\\" % (mdx, fdx, s))\\n simm_fh.close()\",\n \"def versus(bot1, bot2, num_matches, cpu_count, seed_list):\\n W, L, D = 0, 0, 0\\n tot_scores = [0, 0]\\n avg_scores = [0, 0]\\n all_scores = []\\n show_match_results = 1\\n\\n pool = multiprocessing.Pool(cpu_count)\\n\\n try:\\n bot1name = os.path.splitext(os.path.basename(bot1))[0]\\n bot2name = os.path.splitext(os.path.basename(bot2))[0]\\n start_time = time.time()\\n results = [pool.apply_async(run_match, (bot1, bot2, seed_list[i]))\\n for i in range(num_matches)]\\n \\n if show_match_results:\\n sys.stdout.write('%20s - %3i: [' % (bot2name, num_matches))\\n\\n tenths_done = 0\\n for r in results:\\n scores = r.get(timeout=50)\\n if scores[0] > scores[1]:\\n W += 1\\n elif scores[0] < scores[1]:\\n L += 1\\n elif scores[0] == scores[1]:\\n D += 1\\n tot_scores = [tot_scores[x]+scores[x] for x in range(2)]\\n avg_scores = [tot_scores[x]//sum([W,L,D]) for x in range(2)]\\n all_scores.append(scores[0:2])\\n if (sum([W,L,D])*10//num_matches) > tenths_done:\\n tenths_done += 1\\n if show_match_results:\\n sys.stdout.write ('.')\\n\\n pool.close()\\n pool.join()\\n \\n if show_match_results:\\n diff = [s[0] - s[1] for s in all_scores]\\n avg, std = meanstd(diff)\\n sys.stdout.write (' '*(10-tenths_done))\\n print '] - %12s by %8s diff %5.1f std %4.1f in %5.1fs' % (str([W,L,D]), str(avg_scores), avg, std, time.time()-start_time)\\n except multiprocessing.TimeoutError:\\n if show_match_results:\\n sys.stdout.write ('x'*(10-tenths_done))\\n print '] - %12s by %8s in %5.1fs' % (str([W,L,D]), str(avg_scores), time.time()-start_time)\\n pool.terminate()\\n\\n return [W,L,D], tot_scores, all_scores\",\n \"def match_percentage(image1_pixels, image2_pixels):\\n\\n match, total = 0, 0\\n for i in range(len(image1_pixels)):\\n if image1_pixels[i] == image2_pixels[i]:\\n match += 1\\n total += 1\\n else:\\n total += 1\\n return float(match) / float(total)\",\n \"def compute_matches(self):\\n\\t\\tself.local_database[\\\"figure_number\\\"] = [0] * len(self.local_database[\\\"feature_vectors\\\"])\\n\\t\\tForensics = wbForensicsHOG(Database=self.local_database)\\n\\t\\tForensics.KDTree_pairs(leaf_size = len(self.local_database)+1)\\n\\t\\tForensics.d_rank(pairs=Forensics.pairs, distances=Forensics.dists, ratios=Forensics.ratios)\\n\\n\\t\\tself.local_matches = Forensics.Dist_Rank\",\n \"def detection_analysis(y_pred, y_true):\\n print(\\\"Precision: \\\", sm.precision_score(y_pred, y_true))\\n print(\\\"Recall: \\\", sm.recall_score(y_pred, y_true))\\n print(\\\"Accuracy: \\\", sm.accuracy_score(y_pred, y_true))\\n print(\\\"\\\\n\\\")\",\n \"def get_recall_and_precision(wdb, dets, th_conf, th_iou):\\n hit_total, false_total, miss_total = 0, 0, 0\\n\\n for idx in range(wdb.get_image_count()):\\n data = wdb.get_annos_by_image_index(idx)\\n img_path = data['image_path']\\n fname = os.path.splitext(os.path.basename(img_path))[0]\\n\\n dets4fname = []\\n\\n for d in dets:\\n if d['filename'] == fname: # we have only ONE corresponding element\\n dets4fname = d['data']\\n break\\n\\n dets2compare = [e for e in dets4fname if e['c'] >= th_conf] # detections having confidence larger than threshold only\\n\\n hit, false, miss = count_hits(data['annos'], dets2compare, th_iou)\\n hit_total += hit\\n false_total += false\\n miss_total += miss\\n\\n recall = hit_total/float(hit_total+miss_total+0.000001)\\n precision = hit_total/float(hit_total+false_total+0.000001)\\n\\n return recall, precision\",\n \"def compare_results(self, benchmark_result, reference_benchmark_result):\\n pass\",\n \"def compare(image_a, image_b, is_camera_image):\\n\\n # Generate a unique filename\\n filename = uuid.uuid4().hex[:3]\\n\\n if is_camera_image:\\n image_a = imutils.rotate_bound(image_a, 90)\\n image_b = imutils.rotate_bound(image_b, 90)\\n\\n # Store original to show in future\\n original = image_a\\n\\n # Convert to greyscale\\n image_a = cv2.cvtColor(image_a, cv2.COLOR_BGR2GRAY)\\n image_b = cv2.cvtColor(image_b, cv2.COLOR_BGR2GRAY)\\n\\n # Reduce size and blur to account for shaky handheld camera based images\\n if is_camera_image:\\n scale_multiplier = 0.03125\\n image_a = cv2.resize(image_a, (0, 0), fx=scale_multiplier, fy=scale_multiplier)\\n image_b = cv2.resize(image_b, (0, 0), fx=scale_multiplier, fy=scale_multiplier)\\n image_a = cv2.GaussianBlur(image_a, (1001, 1001), cv2.BORDER_DEFAULT)\\n image_b = cv2.GaussianBlur(image_b, (1001, 1001), cv2.BORDER_DEFAULT)\\n\\n # Obtain SSIM and determine differences\\n try:\\n _, differences = structural_similarity(image_a, image_b, full=True, gaussian_weights=True)\\n except ValueError:\\n print('Images are not the same size')\\n return None\\n\\n # Convert to cv2 array\\n differences = (differences * 255).astype('uint8')\\n\\n # Threshold and find contours (differences)\\n thresh = cv2.threshold(differences, 0, 255, cv2.THRESH_BINARY_INV | cv2.THRESH_OTSU)[1]\\n contours = cv2.findContours(thresh, cv2.RETR_EXTERNAL, cv2.CHAIN_APPROX_SIMPLE)\\n contours = imutils.grab_contours(contours)\\n\\n # Draw contours (differences)\\n for cont in contours:\\n (x, y, w, h) = cv2.boundingRect(cont)\\n if is_camera_image:\\n multiplier = int(1 / scale_multiplier)\\n y *= multiplier\\n x *= multiplier\\n h *= multiplier\\n w *= multiplier\\n cv2.rectangle(original, (x, y), (x + w, y + h), (255, 0, 0), 4)\\n\\n # TODO: Create GIF highlighting differences (instead of statuic image)\\n cv2.imwrite('static/images/differences/' + filename + '.jpg', original)\\n\\n return filename\",\n \"def run_multiple_simple_experiments(self, path_to_csv_files, algorithms, thresholds):\\n starting_time = time.time()\\n print(\\\"started simple precision recall image_matching_module\\\")\\n for algorithm in algorithms:\\n start_file_time = time.time()\\n print(\\\"started algorithm:\\\", algorithm.get_algorithm_name())\\n self.run_single_simple_experiment(path_to_csv_files, algorithm, thresholds)\\n finish_file_time = time.time()\\n print(\\\"finished algorithm:\\\", algorithm.get_algorithm_name())\\n print(\\\"time took:\\\", (finish_file_time - start_file_time) / 60, \\\"minutes\\\")\\n finish_time = time.time()\\n hours = int((finish_time - starting_time) // 3600)\\n minutes = (finish_time - starting_time) / 60\\n print(\\\"finished simple precision recall image_matching_module\\\")\\n print(\\\"time image_matching_module took:\\\", hours, \\\"hours and\\\", minutes, \\\"minutes\\\")\",\n \"def return_MatchUpTest_r__():\\n\\n ####################################################################################################################\\n # 1. Initialise test data\\n ####################################################################################################################\\n\\n values = array([470.5, 720.56, 450.9, 295.6, 315.23,\\n 70.5, 70.6, 70.3, 70.7, 70.5,\\n 71.5, 71.6, 71.3, 71.7,\\n 80.5, 80.6, 80.3, 80.7,\\n 150.5, 151.1, 149.8, 150.2, 151.4,\\n 140.5, 141.1, 139.8, 140.2,\\n 160.5, 161.1, 169.8, 160.2,\\n 30.2, 20.4, 28.2, 50.7, 45.6,\\n 29.2, 37.4, 28.2, 50.7,\\n 28.2, 32.4, 22.2, 53.7, ])\\n unc = [Uncertainty(1, array([1.6, 1.5, 1.5, 1.3, 1.5])),\\n Uncertainty(1, array([3.1, 3.2, 3.2, 3.1, 3.0])),\\n Uncertainty(1, array([3.3, 3.4, 3.1, 3.2])),\\n Uncertainty(1, array([2.1, 2.2, 2.2, 2.1])),\\n Uncertainty(1, array([5.0, 4.7, 5.1, 5.2, 5.3])),\\n Uncertainty(1, array([4.2, 4.3, 4.4, 4.3])),\\n Uncertainty(1, array([4.0, 3.7, 4.4, 4.7])),\\n Uncertainty(1, array([2.2, 1.7, 2.0, 4.3, 2.6])),\\n Uncertainty(1, array([2.3, 1.2, 2.3, 4.4])),\\n Uncertainty(1, array([3.2, 2.7, 3.0, 5.3]))]\\n ks = array([1.2, 1.7, 1.3, 1.4, 1.3, 3.2, 3.7, 3.3, 3.4])\\n unck = [Uncertainty(1, array([0.25, 0.25, 0.25, 0.25, 0.25])),\\n Uncertainty(1, array([0.2644, 0.2644, 0.2644, 0.2644]))]\\n idx = {\\\"Nm\\\": [5, 4],\\n \\\"cNm\\\": [0, 5, 9],\\n \\\"Im\\\": [[0, 1], [1, 2]],\\n \\\"sensors\\\": [-1, 1, 2],\\n \\\"sensor_ms\\\": [1, 3, 3],\\n \\\"n_sensor\\\": [0, 1, 1, 2, 1, 1, 2, 1, 1, 2],\\n \\\"n_mu\\\": [1, 1, 2, 2, 1, 2, 2, 1, 2, 2],\\n \\\"n_cov\\\": [1, 1, 1, 1, 2, 2, 2, 3, 3, 3],\\n \\\"N_var\\\": [5, 5, 4, 4, 5, 4, 4, 5, 4, 4],\\n \\\"idx\\\": [0, 5, 10, 14, 18, 23, 27, 31, 36, 40, 44],\\n \\\"Ia\\\": [1, 1, 1, 2, 2, 2]}\\n a = array([1., 1.3, 0.002, 0.5, 1.1, 0.0005])\\n\\n ####################################################################################################################\\n # 3. Initialise MatchUp object\\n ####################################################################################################################\\n\\n MatchUpTest = MatchUp()\\n MatchUpTest.values = values\\n MatchUpTest.unc = unc\\n MatchUpTest.ks = ks\\n MatchUpTest.unck = unck\\n MatchUpTest.idx = idx\\n MatchUpTest.a = a\\n\\n return MatchUpTest\",\n \"def test_compare_classifiers_predictions_vis_api(experiment_to_use):\\n experiment = experiment_to_use\\n predictions = experiment.predictions\\n viz_outputs = (\\\"pdf\\\", \\\"png\\\")\\n with TemporaryDirectory() as tmpvizdir:\\n for viz_output in viz_outputs:\\n vis_output_pattern_pdf = tmpvizdir + f\\\"/*.{viz_output}\\\"\\n visualize.compare_classifiers_predictions(\\n [predictions, predictions],\\n experiment.ground_truth,\\n experiment.ground_truth_metadata,\\n experiment.output_feature_name,\\n labels_limit=0,\\n model_names=[\\\"Model1\\\", \\\"Model2\\\"],\\n output_directory=tmpvizdir,\\n file_format=viz_output,\\n )\\n figure_cnt = glob.glob(vis_output_pattern_pdf)\\n assert 1 == len(figure_cnt)\",\n \"def run_multiple_advanced_experiments(self, path_to_csv_files, algorithms_weights_list, thresholds):\\n starting_time = time.time()\\n print(\\\"started advanced precision recall image_matching_module\\\")\\n for algorithms_weights in algorithms_weights_list:\\n start_file_time = time.time()\\n\\n first_run = True\\n for algorithm, weight in algorithms_weights:\\n if first_run:\\n algorithm_name = algorithm.get_algorithm_name() + weight\\n first_run = False\\n else:\\n algorithm_name = algorithm_name + \\\"_\\\" + algorithm.get_algorithm_name() + weight\\n\\n print(\\\"started algorithm:\\\", algorithm_name)\\n self.run_single_advanced_experiment(path_to_csv_files, algorithms_weights, thresholds)\\n finish_file_time = time.time()\\n print(\\\"finished algorithm:\\\", algorithm_name)\\n print(\\\"time took:\\\", (finish_file_time - start_file_time) / 60, \\\"minutes\\\")\\n finish_time = time.time()\\n hours = int((finish_time - starting_time) // 3600)\\n minutes = (finish_time - starting_time) / 60\\n print(\\\"finished advanced precision recall image_matching_module\\\")\\n print(\\\"time image_matching_module took:\\\", hours, \\\"hours and\\\", minutes, \\\"minutes\\\")\",\n \"def plot_images(img_A, img_B, match_A, match_B, name=\\\"p1\\\", method=\\\"greedy\\\"):\\n (hA, wA) = img_A.shape[:2]\\n (hB, wB) = img_B.shape[:2]\\n vis = np.zeros((max(hA, hB), wA + wB, 3), dtype=\\\"uint8\\\")\\n vis[0:hA, 0:wA] = img_A\\n vis[0:hB, wA:] = img_B\\n\\n # loop over the matches\\n for ptA, ptB in zip(match_A, match_B):\\n ptB = ptB + np.array([wA, 0])\\n cv2.line(vis, tuple(ptA), tuple(ptB), (0, 255, 0), 1)\\n\\n cv2.imwrite(\\\"output/sift_\\\" + method + \\\"_\\\" + name + \\\".png\\\", vis)\\n cv2.imshow(\\\"Keypoint matching\\\", vis)\\n cv2.waitKey(0)\",\n \"def compare_predictions():\\n validation_labels = np.array(pd.read_csv(val_true_labels_dir + dataset_version + 'validation_labels.csv', index_col=0))\\n validation_labels = np.reshape(validation_labels, (-1))\\n\\n diff_between_files = []\\n also1s = []\\n also2s = []\\n for filename1 in os.listdir(val_predictions_dir):\\n if filename1.endswith(\\\".csv\\\"):\\n for filename2 in os.listdir(val_predictions_dir):\\n if filename2.endswith(\\\".csv\\\"):\\n if filename1 < filename2:\\n wrong1 = 0\\n wrong2 = 0\\n diff_between = 0\\n also1 = 0\\n also2 = 0\\n diff_corr1 = 0\\n diff_corr2 = 0\\n f1 = np.array(pd.read_csv(val_predictions_dir + filename1, index_col=0))\\n f1 = np.reshape(f1, (-1))\\n f2 = np.array(pd.read_csv(val_predictions_dir + filename2, index_col=0))\\n f2 = np.reshape(f2, (-1))\\n for line in range(f1.shape[0]):\\n if f1[line] != validation_labels[line]:\\n wrong1 += 1\\n if f2[line] != validation_labels[line]:\\n wrong2 += 1\\n if f1[line] != f2[line]:\\n diff_between += 1\\n if f1[line] == validation_labels[line]:\\n diff_corr1 += 1\\n if f2[line] == validation_labels[line]:\\n diff_corr2 += 1\\n if f1[line] != validation_labels[line]:\\n if f2[line] != validation_labels[line]:\\n also2 += 1\\n if f2[line] != validation_labels[line]:\\n if f1[line] != validation_labels[line]:\\n also1 += 1\\n\\n diff_between_files.append(diff_between)\\n print(filename1)\\n print('Wrongly predicted by 1: ' + str(100 * wrong1 / f1.shape[0]) + '%')\\n print(filename2)\\n print('Wrongly predicted by 2: ' + str(100 * wrong2 / f1.shape[0]) + '%')\\n print()\\n print('Differences between files: ' + str(100 * diff_between / f1.shape[0]) + '%')\\n print(f'\\\\t of which correct by 1 {100 * diff_corr1 / diff_between}%, by 2 {100 * diff_corr2 / diff_between}%')\\n also1s.append(also1 / wrong2)\\n also2s.append(also2 / wrong1)\\n print('Wrongly predicted by other among wrong ones: ' + str(100 * also2 / wrong1) + '%, ' + str(\\n 100 * also1 / wrong2) + '%\\\\n\\\\n\\\\n')\\n\\n print('Max, min and avg differences between files:')\\n print(str(100 * max(diff_between_files) / validation_labels.shape[0]) + '%')\\n print(str(100 * min(diff_between_files) / validation_labels.shape[0]) + '%')\\n print(str(100 * np.mean(diff_between_files) / validation_labels.shape[0]) + '%')\\n\\n print('\\\\nWrongly predicted by first that were also wrongly predicted by second:')\\n print('Max: ' + str(100 * max(also2s)) + '%')\\n print('Min: ' + str(100 * min(also2s)) + '%')\\n print('Avg: ' + str(100 * np.mean(also2s)) + '%')\\n\\n print('\\\\nWrongly predicted by second that were also wrongly predicted by first:')\\n print('Max: ' + str(100 * max(also1s)) + '%')\\n print('Min: ' + str(100 * min(also1s)) + '%')\\n print('Avg: ' + str(100 * np.mean(also1s)) + '%')\",\n \"def test_compare_single_and_multiprocess_results(sidesweep_image_sequence):\\n cc = Cwsim_container_from_ims(ims=sidesweep_image_sequence)\\n test_idx = 20\\n results_single_proc = cc.self_im_query_all(test_idx, plot_output=False)\\n\\n test_im = sidesweep_image_sequence[test_idx]\\n cc.prepare_memory_bank_outside()\\n results_multi_proc = cc.query_image_mp(test_im)\\n\\n plt.plot(results_single_proc, marker='D', label='single_proc')\\n plt.plot(results_multi_proc, marker='D', label='multi_proc')\\n plt.legend()\\n plt.draw()\\n plt.show(block=False)\\n plt.pause(3)\\n\\n assert np.allclose(results_single_proc, results_multi_proc)\",\n \"def evaluate(calculated_results, ground_truth, measure=''):\\n score = 0\\n truth_pairs = dict.fromkeys(pair[0] + ' ' + pair[1]\\n for pair in ground_truth.edges())\\n for pair in calculated_results:\\n pair_reversed = get_reversed_pair(pair)\\n if pair[0] in truth_pairs or pair_reversed in truth_pairs:\\n score += 1\\n accuracy = (score / len(truth_pairs)) * 100\\n print(\\\"The accuracy is %f %% when using measure: %s\\\" % (accuracy, measure))\",\n \"def evaluate(groundtruth, prediction, tiou_threshold=0.5):\\n\\n video_num = len(groundtruth)\\n print('Computing grounding accuracy over {} videos...'.format(video_num))\\n acc, acc_sub, acc_obj = 0.0, 0.0, 0.0\\n\\n gt_rnum = 0\\n for qid, relation_gt in groundtruth.items():\\n\\n if qid not in prediction:\\n continue\\n relation_pred = prediction[qid]\\n if len(relation_pred) == 0:\\n continue\\n\\n video_acc, video_acc_sub, video_acc_obj, relation_num = eval_ground_scores(relation_gt, relation_pred, tiou_threshold)\\n\\n acc += video_acc\\n acc_sub += video_acc_sub\\n acc_obj += video_acc_obj\\n gt_rnum += relation_num\\n\\n\\n acc /= video_num\\n acc_sub /= video_num\\n acc_obj /= video_num\\n\\n print(\\\"Acc_S\\\\t Acc_O\\\\t Acc_R\\\")\\n\\n print('{:.2f}\\\\t {:.2f}\\\\t {:.2f}'.format(acc_sub*100, acc_obj*100, acc*100))\",\n \"def test_quality(approach):\\n tp, fn, fp, tn = 0, 0, 0, 0\\n with open(\\\"test.txt\\\", \\\"r\\\") as f:\\n test_data = json.load(f)\\n for i in test_data:\\n rez = approach(i[0][0], i[0][1])\\n if rez == 1 and i[1] == 1:\\n tp += 1\\n elif rez == 0 and i[1] == 1:\\n fn += 1\\n elif rez == 1 and i[1] == 0:\\n fp += 1\\n else:\\n tn += 1\\n print \\\"Precision: {}%.\\\\nRecall: {}%.\\\".format(\\n round(tp / (tp + fp), 2), round(tp / (tp + fn), 2))\",\n \"def calculate_metrics(predictions, expected):\\n # type: (np.ndarray, np.ndarray) -> (float, float, float)\\n clients_count = predictions.shape[0]\\n products_count = predictions.shape[1]\\n\\n true_positive = 0.0\\n true_negative = 0.0\\n false_positive = 0.0\\n false_negative = 0.0\\n\\n total = float(clients_count * products_count)\\n\\n for c in range(0, clients_count):\\n for p in range(0, products_count):\\n if predictions[c, p] == expected[c, p]:\\n if predictions[c, p] == 1:\\n true_positive += 1\\n else:\\n true_negative += 1\\n else:\\n if predictions[c, p] == 1:\\n false_positive += 1\\n else:\\n false_negative += 1\\n\\n accuracy = float(true_positive + true_negative) / total\\n if true_positive + false_positive == 0:\\n precision = 0\\n else:\\n precision = true_positive / float(true_positive + false_positive)\\n\\n if true_positive + false_negative == 0:\\n recall = 0\\n else:\\n recall = true_positive / float(true_positive + false_negative)\\n\\n return accuracy, precision, recall\",\n \"def evaluate(self):\\n tic = time.time()\\n print('Running per image evaluation...')\\n p = self.params\\n # add backward compatibility if useSegm is specified in params\\n if p.useSegm is not None:\\n p.iouType = 'segm' if p.useSegm == 1 else 'bbox'\\n print(\\n \\\"useSegm (deprecated) is not None.\\\"\\n \\\"Running {} evaluation\\\".format(p.iouType)\\n )\\n print('Evaluate annotation type *{}*'.format(p.iouType))\\n p.imgIds = list(np.unique(p.imgIds))\\n if p.useCats:\\n p.catIds = list(np.unique(p.catIds))\\n p.maxDets = sorted(p.maxDets)\\n self.params = p\\n\\n self._prepare()\\n # loop through images, area range, max detection number\\n catIds = p.catIds if p.useCats else [-1]\\n\\n self.ious = {\\n (imgId, catId): self.computeIoU(imgId, catId)\\n for imgId in p.imgIds\\n for catId in catIds\\n }\\n # ignore any categories that is not having any attributes\\n self.f1s = {\\n (imgId, catId): self.computeF1(imgId, catId)\\n for imgId in p.imgIds\\n for catId in catIds\\n if catId in self.FPParams.catsWithAttributes\\n }\\n\\n # self.gt_attributes_ids = []\\n\\n # loop through images, area range, max detection number\\n self.evalImgs = [\\n self.evaluateImg(imgId, catId, areaRng, p.maxDets[-1])\\n for catId in catIds\\n for areaRng in p.areaRng\\n for imgId in p.imgIds\\n ]\\n # self._paramsEval = copy.deepcopy(self.params) # seems do not need it\\n toc = time.time()\\n print('DONE (t={:0.2f}s).'.format(toc-tic))\",\n \"def ransac_homography(keypoints1, keypoints2, matches, sampling_ratio=0.5, n_iters=500, delta=20):\\n N = matches.shape[0]\\n n_samples = int(N * sampling_ratio)\\n\\n matched1_unpad = keypoints1[matches[:,0]]\\n matched2_unpad = keypoints2[matches[:,1]]\\n\\n\\n max_inliers = np.zeros(N)\\n n_inliers = 0\\n\\n # RANSAC iteration start\\n ### YOUR CODE HERE\\n for i in range(n_iters):\\n random_sample = np.random.choice(N, n_samples)\\n src = matched1_unpad[random_sample] # leave unpadded, is padded in compute homography\\n dst = matched2_unpad[random_sample]\\n\\n # Get H using DLT\\n H_matrix = compute_homography(src, dst)\\n\\n # Matched keypoints are considered inliers if the projected point from one image lies within distance δ to the matched point on the other point.\\n\\n projection = transform_homography(matched1_unpad, H_matrix) # x point, y point\\n distances = np.sqrt(np.sum(np.square(projection-matched2_unpad), axis=1))\\n if np.sum(distances < delta) > n_inliers:\\n max_inliers = distances < delta\\n n_inliers = np.sum(distances < delta)\\n \\n # Get final H using maximum inliers indices and src dst values\\n src = matched1_unpad[max_inliers]\\n dst = matched2_unpad[max_inliers]\\n H = compute_homography(src, dst)\\n \\n ### END YOUR CODE\\n return H, matches[max_inliers]\",\n \"def PredictMatch(yourName, tgtName, teamModels, db):\\n\\n teamTgtPipe, teamYourPipe = QueryTeamData(tgtName, yourName, db)\\n featureCoefTgt, probTgt = PredictOp(teamTgtPipe, teamYourPipe, tgtName, teamModels)\\n featureCoefYour, probYour = PredictOp(teamYourPipe, teamTgtPipe, yourName, teamModels)\\n odds = round(probYour / (probYour + probTgt), 2)\\n\\n # In featureCoefYour, you want INCREASE those with POSTIVE COEF, DECREASE those with NEGATIVE COEF\\n # In featureCoefTgt, you want to do the opposite\\n\\n # reverse both the sign of the coef, and '_op' in features so as to be the same with featureCoefYour\\n featureCoefTgt['coef'] = - featureCoefTgt['coef']\\n featureCoefTgt.features = [ii[:-3] if \\\"_op\\\" in ii else ii + '_op' for ii in featureCoefTgt.features]\\n\\n # Combine only the most important 10 features\\n # featureBoth = featureCoefTgt[11:].append(featureCoefYour[11:])\\n\\n # Combine only all the most important features\\n featureBoth = featureCoefTgt.append(featureCoefYour)\\n\\n # get action recommendations\\n # Somehow the pandas here uses a deprecated para cols, instaed of the new one subset\\n #featureBoth.drop_duplicates(subset = 'features', take_last = True, inplace = True)\\n featureBoth.drop_duplicates(cols = 'features', take_last = True, inplace = True)\\n actions, featureImprove = GetActions(featureBoth)\\n Imp = 0.1\\n oddsNew = ImprovedScore(tgtName, yourName, teamModels, featureImprove, teamTgtPipe, teamYourPipe, Imp)\\n\\n return odds, oddsNew, actions\",\n \"def test_evaluate(self):\\n # test normalized by 'bbox_size'\\n jhmdb_pck_metric = JhmdbPCKAccuracy(thr=0.5, norm_item='bbox')\\n jhmdb_pck_metric.process(self.data_batch, self.data_samples)\\n pck_results = jhmdb_pck_metric.evaluate(self.batch_size)\\n target = {\\n 'Head PCK': 1.0,\\n 'Sho PCK': 1.0,\\n 'Elb PCK': 1.0,\\n 'Wri PCK': 1.0,\\n 'Hip PCK': 1.0,\\n 'Knee PCK': 1.0,\\n 'Ank PCK': 1.0,\\n 'PCK': 1.0,\\n }\\n self.assertDictEqual(pck_results, target)\\n\\n # test normalized by 'torso_size'\\n jhmdb_tpck_metric = JhmdbPCKAccuracy(thr=0.2, norm_item='torso')\\n jhmdb_tpck_metric.process(self.data_batch, self.data_samples)\\n tpck_results = jhmdb_tpck_metric.evaluate(self.batch_size)\\n target = {\\n 'Head tPCK': 1.0,\\n 'Sho tPCK': 1.0,\\n 'Elb tPCK': 1.0,\\n 'Wri tPCK': 1.0,\\n 'Hip tPCK': 1.0,\\n 'Knee tPCK': 1.0,\\n 'Ank tPCK': 1.0,\\n 'tPCK': 1.0,\\n }\\n self.assertDictEqual(tpck_results, target)\",\n \"def test_compare_classifiers_performance_changing_k_vis_api(experiment_to_use):\\n experiment = experiment_to_use\\n probabilities = experiment.probabilities\\n viz_outputs = (\\\"pdf\\\", \\\"png\\\")\\n with TemporaryDirectory() as tmpvizdir:\\n for viz_output in viz_outputs:\\n vis_output_pattern_pdf = tmpvizdir + f\\\"/*.{viz_output}\\\"\\n visualize.compare_classifiers_performance_changing_k(\\n [probabilities, probabilities],\\n experiment.ground_truth,\\n experiment.ground_truth_metadata,\\n experiment.output_feature_name,\\n top_k=3,\\n labels_limit=0,\\n model_namess=[\\\"Model1\\\", \\\"Model2\\\"],\\n output_directory=tmpvizdir,\\n file_format=viz_output,\\n )\\n figure_cnt = glob.glob(vis_output_pattern_pdf)\\n assert 1 == len(figure_cnt)\",\n \"def calculate_score(label_dict, queries, results):\\n\\n total_score = 0.0\\n\\n log_header('Individual image scores')\\n\\n # Calculate score for all images\\n for image_id in queries:\\n if image_id in results.keys():\\n # Run the score function\\n image_score = score(\\n label_dict=label_dict, target=image_id, selection=results[image_id]\\n )\\n else:\\n logging.error('No result generated for %s' % image_id)\\n\\n image_score = 0.0\\n\\n total_score += image_score\\n\\n logging.info('%s: %8.6f' % (image_id, image_score))\\n\\n log_header('Average score over %d images: %10.8f' % (\\n len(queries), total_score / len(queries)\\n ))\\n\\n return total_score\",\n \"def compute_pck(descs0, kps0, descs1, kps1, thresh=3):\\n # matches points from image 0 to image 1, using NN\\n if len(descs0.shape) == 3:\\n idxs = nn_set2set_match(descs0, descs1)\\n else:\\n idxs = nn_match(descs0, descs1) # matched image 1 keypoint indices\\n predicted = kps1[idxs] # matched image 1 keypoints\\n\\n correct = np.linalg.norm(predicted - kps1, 2, axis=1) <= thresh\\n\\n return np.sum(correct) / len(correct)\",\n \"def try_template_matching(image,template):\\n img2 = image.copy()\\n w, h = template.shape[::-1]\\n # All the 6 methods for comparison in a list\\n methods = ['cv2.TM_CCOEFF', 'cv2.TM_CCOEFF_NORMED', 'cv2.TM_CCORR',\\n 'cv2.TM_CCORR_NORMED', 'cv2.TM_SQDIFF', 'cv2.TM_SQDIFF_NORMED']\\n for meth in methods:\\n img = img2.copy()\\n method = eval(meth)\\n # Apply template Matching\\n res = cv2.matchTemplate(img,template,method)\\n res-=np.min(res)\\n min_val, max_val, min_loc, max_loc = cv2.minMaxLoc(res)\\n threshold=0.9*np.max(res)\\n \\n # If the method is TM_SQDIFF or TM_SQDIFF_NORMED, take minimum\\n if method in [cv2.TM_SQDIFF, cv2.TM_SQDIFF_NORMED]:\\n loc = np.where( res <=0.1*np.max(res))\\n else:\\n loc = np.where( res >= threshold)\\n \\n for pt in zip(*loc[::-1]):\\n cv2.rectangle(img, pt, (pt[0] + w, pt[1] + h), 255, 2)\\n plt.figure()\\n plt.subplot(121),plt.imshow(image,cmap = 'gray')\\n plt.title('Matching Result'), plt.xticks([]), plt.yticks([])\\n plt.subplot(122),plt.imshow(img,cmap = 'gray')\\n plt.title('Detected Point'), plt.xticks([]), plt.yticks([])\\n plt.suptitle(meth)\\n plt.show()\",\n \"def compare_images(self):\\r\\n m = round(self.mse(self.image_a, self.image_b), 4)\\r\\n s = round(ssim(self.image_a, self.image_b) * 100, 5)\\r\\n return (\\r\\n m, s)\",\n \"def test_run_multi_r__(self):\\n\\n # Test Description\\n # ================\\n #\\n # 1. This test intialises an example *eopy.matchup.matchupIO.MatchUp* object\\n #\\n # 2. Compare transformed dataset to expected value\\n\\n ################################################################################################################\\n # 1. Initialise Test Data Object\\n ################################################################################################################\\n\\n MatchUpTest = return_MatchUpTest_r__()\\n\\n ################################################################################################################\\n # 2. Define expected values\\n ################################################################################################################\\n\\n # Original dataset values (should be unchanged)\\n MatchUpOriginal_expected = return_MatchUpTest_r__()\\n\\n # Transformed dataset\\n values_expected = array([294.0625, 480.3733333, 300.6, 227.3846154, 210.1533333,\\n 22.74193548, 22.0625, 21.96875, 22.80645161, 23.5,\\n 21.66666667, 21.05882353, 23, 22.40625,\\n 38.33333333, 36.63636364, 36.5, 38.42857143,\\n 30.1, 32.14893617, 29.37254902, 28.88461538, 28.56603774,\\n 33.45238095, 32.81395349, 31.77272727, 32.60465116,\\n 40.125, 43.54054054, 38.59090909, 34.08510638,\\n 13.72727273, 12, 14.1, 11.79069767, 17.53846154,\\n 12.69565217, 31.16666667, 12.26086957, 11.52272727,\\n 8.8125, 12, 7.4, 10.13207547])\\n unc_expected = [Uncertainty(1, array([1.6, 1.5, 1.5, 1.3, 1.5])),\\n Uncertainty(1, array([3.1, 3.2, 3.2, 3.1, 3.0])),\\n Uncertainty(1, array([3.3, 3.4, 3.1, 3.2])),\\n Uncertainty(1, array([2.1, 2.2, 2.2, 2.1])),\\n Uncertainty(1, array([5.0, 4.7, 5.1, 5.2, 5.3])),\\n Uncertainty(1, array([4.2, 4.3, 4.4, 4.3])),\\n Uncertainty(1, array([4.0, 3.7, 4.4, 4.7])),\\n Uncertainty(1, array([2.2, 1.7, 2.0, 4.3, 2.6])),\\n Uncertainty(1, array([2.3, 1.2, 2.3, 4.4])),\\n Uncertainty(1, array([3.2, 2.7, 3.0, 5.3]))]\\n ks_expected = array([4.8, 6.8, 5.2, 5.6, 5.2, 12.10287443, 13.99394856, 12.48108926, 12.85930408])\\n unck_expected = [Uncertainty(1, array([0.25, 0.25, 0.25, 0.25, 0.25])),\\n Uncertainty(1, array([0.2644, 0.2644, 0.2644, 0.2644]))]\\n idx_expected = {\\\"Nm\\\": [5, 4],\\n \\\"cNm\\\": [0, 5, 9],\\n \\\"Im\\\": [[0, 1], [1, 2]],\\n \\\"sensors\\\": [-1, 1, 2],\\n \\\"sensor_ms\\\": [1, 3, 3],\\n \\\"n_sensor\\\": [0, 1, 1, 2, 1, 1, 2, 1, 1, 2],\\n \\\"n_mu\\\": [1, 1, 2, 2, 1, 2, 2, 1, 2, 2],\\n \\\"n_cov\\\": [1, 1, 1, 1, 2, 2, 2, 3, 3, 3],\\n \\\"N_var\\\": [5, 5, 4, 4, 5, 4, 4, 5, 4, 4],\\n \\\"idx\\\": [0, 5, 10, 14, 18, 23, 27, 31, 36, 40, 44],\\n \\\"Ia\\\": [1, 1, 1, 2, 2, 2]}\\n a_expected = array([1., 1.3, 0.002, 0.5, 1.1, 0.0005])\\n w_matrices_expected = []\\n u_matrices_expected = []\\n\\n ################################################################################################################\\n # 3. Run Transform2NormInd.run()\\n ################################################################################################################\\n\\n Transform2NormIndOp = Transform2NormInd()\\n MatchUpTransform = Transform2NormIndOp.run(MatchUpTest)\\n\\n values_test = MatchUpTransform.values\\n unc_test = MatchUpTransform.unc\\n w_matrices_test = MatchUpTransform.w_matrices\\n u_matrices_test = MatchUpTransform.u_matrices\\n ks_test = MatchUpTransform.ks\\n unck_test = MatchUpTransform.unck\\n idx_test = MatchUpTransform.idx\\n\\n ################################################################################################################\\n # 4. Compare retrieve values to expect values\\n ################################################################################################################\\n\\n # Test transformed data object attribute by attribute\\n\\n # a. values\\n for i, (value_expected, value_test) in enumerate(zip(values_expected, values_test)):\\n self.assertAlmostEqual(value_expected, value_test, places=5, msg=str(i))\\n\\n # b. unc\\n for block_unc_test, block_unc_expected in zip(unc_test, unc_expected):\\n self.assertEqual(block_unc_expected.typeID, block_unc_test.typeID)\\n self.assertEqual(block_unc_expected.uR.tolist(), block_unc_test.uR.tolist())\\n\\n # c. w_matrices\\n self.assertEqual(w_matrices_test, w_matrices_expected)\\n\\n # d. u_matrices\\n self.assertEqual(u_matrices_test, u_matrices_expected)\\n\\n # e. ks\\n for k_expected, k_test in zip(ks_expected, ks_test):\\n self.assertAlmostEqual(k_test.tolist(), k_expected.tolist(), places=5)\\n\\n # f. unck\\n for block_unck_test, block_unck_expected in zip(unck_test, unck_expected):\\n self.assertEqual(block_unck_expected.typeID, block_unck_test.typeID)\\n self.assertEqual(block_unck_expected.uR.tolist(), block_unck_test.uR.tolist())\\n\\n # h. idx\\n self.assertEqual(set(idx_expected.keys()), set(idx_test.keys()))\\n for key in idx_expected.keys():\\n idx_i_test = idx_test[key]\\n idx_i_expected = idx_expected[key]\\n if isinstance(idx_i_expected, ndarray):\\n self.assertEqual(idx_i_test.tolist(), idx_i_expected.tolist())\\n else:\\n self.assertEqual(idx_i_test, idx_i_expected)\\n\\n # Test original data object preserved attribute by attribute\\n\\n # a. values\\n for i, (value_original_expected, value_original_test) in enumerate(zip(MatchUpOriginal_expected.values, MatchUpTest.values)):\\n self.assertAlmostEqual(value_original_expected, value_original_test, places=5)\\n\\n # b. unc\\n for block_unc_original_expected, block_unc_original_test in zip(MatchUpOriginal_expected.unc, MatchUpTest.unc):\\n self.assertEqual(block_unc_original_expected.typeID, block_unc_original_test.typeID)\\n self.assertEqual(block_unc_original_expected.uR.tolist(), block_unc_original_test.uR.tolist())\\n\\n # c. w_matrices\\n self.assertEqual(MatchUpOriginal_expected.w_matrices, MatchUpTest.w_matrices)\\n\\n # d. u_matrices\\n self.assertEqual(MatchUpOriginal_expected.u_matrices, MatchUpTest.u_matrices)\\n\\n # e. ks\\n for k_original_expected, k_original_test in zip(MatchUpOriginal_expected.ks, MatchUpTest.ks):\\n self.assertAlmostEqual(k_original_test.tolist(), k_original_expected.tolist(), places=5)\\n\\n # f. unck\\n for block_unck_original_expected, block_unck_original_test in zip(MatchUpOriginal_expected.unck, MatchUpTest.unck):\\n self.assertEqual(block_unck_original_expected.typeID, block_unck_original_test.typeID)\\n self.assertEqual(block_unck_original_expected.uR.tolist(), block_unck_original_test.uR.tolist())\\n\\n # h. idx\\n self.assertEqual(set(MatchUpOriginal_expected.idx), set(MatchUpTest.idx))\\n for key in MatchUpOriginal_expected.idx.keys():\\n idx_i_original_test = MatchUpTest.idx[key]\\n idx_i_original_expected = MatchUpOriginal_expected.idx[key]\\n if isinstance(idx_i_original_expected, ndarray):\\n self.assertEqual(idx_i_original_test.tolist(), idx_i_original_expected.tolist())\\n else:\\n self.assertEqual(idx_i_original_test, idx_i_original_expected)\",\n \"def detail_matching(self):\\n paradic = self.cfg['param']['paradic']\\n work_dir = self.work_dir\\n \\n x = float(self.cfg['param']['x']) # selected pixel in the first image\\n y = float(self.cfg['param']['y'])\\n \\n # sift parameters\\n # number of bins in the orientation histogram\\n n_bins = int(paradic['n_bins']) \\n n_hist = int(paradic['n_hist']) \\n # descriptor of n_hist X n_hist weighted histograms with n_ori\\n n_ori = int(paradic['n_ori']) \\n delta_min = float(paradic['delta_min'])\\n sigma_min = float(paradic['sigma_min'])\\n sigma_in = float(paradic['sigma_in'])\\n lambda_ori = float(paradic['lambda_ori'])\\n lambda_descr = float(paradic['lambda_descr'])\\n #threshold defining reference orientations\\n n_spo = int(paradic['n_spo'])\\n \\n # Read feature vectors from output files\\n if (os.path.getsize(work_dir+'OUTmatches.txt') > 0 ):\\n pairdata = find_nearest_keypoint(work_dir+'OUTmatches.txt', y, x)\\n \\n illustrate_pair(pairdata, n_bins, n_hist, n_ori, work_dir)\\n\\n \\n # Read keys coordinates.\\n d = 6+n_bins+n_hist*n_hist*n_ori # size of keydata inside pairdata\\n v = n_hist*n_hist*n_ori\\n [x1, y1, sigma1, theta1] = [float(x) for x in pairdata[0:4]]\\n [o1, s1] = [float(x) for x in pairdata[4+v:4+v+2]]\\n [x2a, y2a, sigma2a, theta2a] = [float(x) for x in pairdata[d:d+4]]\\n [o2a, s2a] = [float(x) for x in pairdata[d+4+v:d+4+v+2]]\\n [x2b, y2b, sigma2b, theta2b] = \\\\\\n [float(x) for x in pairdata[2*d:2*d+4]]\\n [o2b, s2b] = [float(x) for x in pairdata[2*d+4+v:2*d+4+v+2]]\\n \\n draw_one_match(pairdata,\\n work_dir+'input_0.png',\\n work_dir+'input_1.png',\\n d,\\n lambda_ori,\\n lambda_descr,\\n n_hist,\\n work_dir+'OUTonepair.png')\\n \\n \\n # Extract thumbnails.\\n # keypoint 1 (image 1)\\n print ' '.join(['demo_extract_patch', work_dir+'input_0.png',\\n str(x1), str(y1), str(sigma1), str(theta1), str(o1), str(s1),\\n str(delta_min), str(sigma_min), str(sigma_in), str(n_spo),\\n str(lambda_ori), str(lambda_descr), str(n_hist),\\n work_dir+\\\"detail_im1\\\"])\\n proc = self.run_proc(['demo_extract_patch', work_dir+'input_0.png',\\n str(x1), str(y1), str(sigma1), str(theta1), str(o1), str(s1),\\n str(delta_min), str(sigma_min), str(sigma_in), str(n_spo),\\n str(lambda_ori), str(lambda_descr), str(n_hist),\\n work_dir+\\\"detail_im1\\\"])\\n self.wait_proc(proc, timeout=self.timeout)\\n \\n # keypoint 2a (nearest neighbor in image 2)\\n print ' '.join(['demo_extract_patch', work_dir+'input_1.png',\\n str(x2a), str(y2a), str(sigma2a), str(theta2a), str(o2a), str(s2a),\\n str(delta_min), str(sigma_min), str(sigma_in), str(n_spo),\\n str(lambda_ori), str(lambda_descr), str(n_hist),\\n work_dir+\\\"detail_im2a\\\"])\\n proc = self.run_proc(['demo_extract_patch', work_dir+'input_1.png',\\n str(x2a), str(y2a), str(sigma2a), str(theta2a), str(o2a), str(s2a),\\n str(delta_min), str(sigma_min), str(sigma_in), str(n_spo),\\n str(lambda_ori), str(lambda_descr), str(n_hist),\\n work_dir+\\\"detail_im2a\\\"])\\n self.wait_proc(proc, timeout=self.timeout) \\n \\n # keypoint 2b (second nearest neighbor in image 2)\\n proc = self.run_proc(['demo_extract_patch', work_dir+'input_1.png',\\n str(x2b), str(y2b), str(sigma2b), str(theta2b), str(o2b), str(s2b),\\n str(delta_min), str(sigma_min), str(sigma_in), str(n_spo),\\n str(lambda_ori), str(lambda_descr), str(n_hist),\\n work_dir+\\\"detail_im2b\\\"])\\n self.wait_proc(proc, timeout=self.timeout) \\n \\n \\n return 1\",\n \"def eval_feature_verify(query_features, gallery_features, query_lst, gallery_lst, metric=\\\"cosine\\\"):\\n if metric not in [\\\"euclidean\\\", \\\"cosine\\\"]:\\n raise ValueError(\\\"Invalid metric! \\\")\\n\\n num_query = len(query_lst)\\n num_gallery = len(gallery_lst)\\n\\n if metric == \\\"cosine\\\":\\n gallery_features = normalize(gallery_features, axis=1)\\n query_features = normalize(query_features, axis=1)\\n\\n if metric == \\\"euclidean\\\":\\n dist = euclidean_distances(gallery_features, query_features).squeeze()\\n else:\\n dist = np.dot(gallery_features, query_features.T)\\n print(dist[:].max())\\n print(dist[:].min())\\n\\n gallery_cam_lst = np.array([x.cam_id for x in gallery_lst], dtype=np.int32)\\n gallery_id_lst = np.array([x.class_id for x in gallery_lst], dtype=np.int32)\\n query_cam_lst = np.array([x.cam_id for x in query_lst], dtype=np.int32)\\n query_id_lst = np.array([x.class_id for x in query_lst], dtype=np.int32)\\n\\n # same id\\n gallery_id_lst_tile = np.tile(gallery_id_lst, (num_query, 1))\\n query_id_lst_tile = np.tile(query_id_lst, (num_gallery, 1))\\n label_mat = query_id_lst_tile == gallery_id_lst_tile.T # gallery num * query num\\n label_mat = label_mat.astype(np.int)\\n\\n # same camera and id\\n gallery_cam_lst_tile = np.tile(gallery_cam_lst, (num_query, 1))\\n query_cam_lst_tile = np.tile(query_cam_lst, (num_gallery, 1))\\n # same camera, gallery num * query num\\n label_cam_mat = query_cam_lst_tile == gallery_cam_lst_tile.T\\n # filter by same id\\n label_cam_mat = label_cam_mat * label_mat\\n # same id but under same camera are labeled as -1,\\n label_mat -= 2 * label_cam_mat.astype(np.int)\\n\\n ind = np.where(label_mat == 0)\\n neg_dist = dist[ind]\\n ind = np.where(label_mat == 1)\\n pos_dist = dist[ind]\\n\\n new_dist = np.concatenate([pos_dist, neg_dist])\\n new_label = np.concatenate([np.ones_like(pos_dist), np.zeros_like(neg_dist)])\\n dist_label = zip(new_dist, new_label)\\n dist_label_new = np.array(sorted(dist_label, key=lambda d: -d[0]))\\n\\n # ap = []\\n n_good = np.sum(dist_label_new[:, 1])\\n n_bad = np.sum(1 - dist_label_new[:, 1])\\n print(n_good)\\n for thresh in np.arange(0.90, 1, 0.01):\\n # print(thresh)\\n p = dist_label_new[:, 0] >= thresh\\n if np.sum(p) == 0:\\n recall, precision, f1 = 0, 0, 0\\n else:\\n tp = np.logical_and(dist_label_new[:, 1], p)\\n recall = np.sum(tp) / n_good\\n precision = np.float(np.sum(tp)) / np.sum(p)\\n\\n f1 = 2*recall*precision/(recall+precision)\\n\\n print('%.4f\\\\t%.4f\\\\t%.4f' % (recall, precision, f1))\\n\\n # ap.append(compute_ap(dist_label_new, thresh))\\n # print(ap)\",\n \"def apply_feature_matching(desc1: np.ndarray, desc2: np.ndarray,\\n match_calculator: Callable[[list, list], float]) -> list:\\n\\n # Check descriptors dimensions are 2\\n assert desc1.ndim == 2, \\\"Descriptor 1 shape is not 2\\\"\\n assert desc2.ndim == 2, \\\"Descriptor 2 shape is not 2\\\"\\n\\n # Check that the two features have the same descriptor type\\n assert desc1.shape[1] == desc2.shape[1], \\\"Descriptors shapes are not equal\\\"\\n\\n # If there is not key points in any of the images\\n if desc1.shape[0] == 0 or desc2.shape[0] == 0:\\n return []\\n\\n # number of key points in each image\\n num_key_points1 = desc1.shape[0]\\n num_key_points2 = desc2.shape[0]\\n\\n # List to store the matches scores\\n matches = []\\n\\n # Loop over each key point in image1\\n # We need to calculate similarity with each key point in image2\\n for kp1 in range(num_key_points1):\\n # Initial variables which will be updated in the loop\\n distance = -np.inf\\n y_index = -1\\n\\n # Loop over each key point in image2\\n for kp2 in range(num_key_points2):\\n\\n # Match features between the 2 vectors\\n value = match_calculator(desc1[kp1], desc2[kp2])\\n\\n # SSD values examples: (50, 200, 70), we need to minimize SSD (Choose 50)\\n # In case of SSD matching: (value is returned as a \\\"negative\\\" number) (-50, -200, -70)\\n # So we compare it with -np.inf. (The sorting will be reversed later)\\n\\n # NCC values examples: (0.58, 0.87, 0.42), we need to maximize NCC (Choose 0.87)\\n # In case of NCC matching: (value is returned as a \\\"positive\\\" number) (-58, -0.87, -0.42)\\n # So we compare it with -np.inf. (The sorting will be reversed later)\\n\\n if value > distance:\\n distance = value\\n y_index = kp2\\n\\n # Create a cv2.DMatch object for each match and set attributes:\\n # queryIdx: The index of the feature in the first image\\n # trainIdx: The index of the feature in the second image\\n # distance: The distance between the two features\\n cur = cv2.DMatch()\\n cur.queryIdx = kp1\\n cur.trainIdx = y_index\\n cur.distance = distance\\n matches.append(cur)\\n\\n return matches\",\n \"def test_RawRun_process():\\n for style in test_runs:\\n test_runs[style].process()\\n # now compare all images with ref\\n ref = sorted(glob.glob('tests/data/processed_ref/*/*/*'))\\n outputs = sorted(glob.glob('tests/data/processed/*/*/*'))\\n for ref, out in zip(ref, outputs):\\n print ref\\n print out\\n assert_image_equal(ref, out)\",\n \"def match2(img1, img2, coordinates1, coordinates2, PATCH_SIZE):\\n\\n\\tpossible_matches = pd.DataFrame(columns=['feature1', 'feature2', 'diff'])\\n\\n\\t# iteration through all the possible pairs of features from img1 and img2\\n\\tfor (feature1, feature2) in product(coordinates1, coordinates2):\\n\\t\\tpatch1 = make_patch(feature1, PATCH_SIZE, img1)\\n\\t\\tpatch2 = make_patch(feature2, PATCH_SIZE, img2)\\n\\t\\tdiff = count_difference(patch1, patch2)\\n\\t\\t# print(feature1, feature2, diff)\\n\\t\\t# next line is weird, don't ask\\n\\t\\tpossible_matches = possible_matches.append({'feature1': feature1, 'feature2': feature2, 'diff': diff}, ignore_index=True)\\n\\n\\n\\t# sorting the possible_matches according to their difference and resetin the index\\n\\t# inplace=True means that changes will be performed in-place, so they will change the variable and there is no need to reassign\\n\\tpossible_matches.sort_values(['diff'], inplace=True)\\n\\tpossible_matches.reset_index(inplace=True, drop=True)\\n\\t# print(\\\"SORTED\\\")\\n\\t# print(possible_matches)\\n\\n\\t# adding only the best matches\\n\\tmatches = []\\n\\twhile not possible_matches.empty:\\n\\t\\tbest_match = possible_matches.loc[0]\\n\\t\\tmatches.append(list(best_match))\\n\\t\\t# creating a list of indicies to be dropped\\n\\t\\ttrash = []\\n\\n\\t\\t#finding and deleting rows with features that are already used and reseting the index\\n\\t\\tfor i, (f1, f2, d) in enumerate(possible_matches.values):\\n\\t\\t \\n\\t\\t\\tif np.all(f1 == best_match[0]) or np.all(f2 == best_match[1]):\\n\\t\\t\\t\\ttrash.append(i)\\n\\t\\n\\t\\tfor i in trash:\\n\\t\\t\\tpossible_matches.drop(i, inplace=True)\\n\\t\\tpossible_matches.reset_index(inplace=True, drop=True)\\n\\t\\t# print(possible_matches)\\n\\n\\treturn matches\",\n \"def _runmatch(self,base,mrad,xsh,ysh):\\n cmd = base+'matchrad='+str(mrad)+' xsh=%.2f ysh=%.2f' %(xsh,ysh)\\n sproc = popen2.Popen3(cmd,1)\\n output = sproc.fromchild.readlines()\\n errs = sproc.childerr.readlines()\\n sproc.fromchild.close()\\n sproc.childerr.close()\\n \\n if errs and not (len(errs) == 1 and errs[0][0:27] == 'atFindMedtf: RMS <= 0.0, so'):\\n self.logfile.write('match produced the following message(s): ')\\n for f in errs:\\n self.logfile.write(string.strip(f))\\n self.errorList.append(('match',string.strip(f)))\\n matchline = ''\\n medianline = ''\\n for line in output:\\n if line[0:6] == 'TRANS:':\\n matchline = line\\n elif line[0:6] == 'MEDTF:':\\n medianline = line\\n else:\\n self.logfile.write(\\\"ERROR:. Unexpected output from match program:\\\\n\\\\t\\\\t\\\\t\\\"+line)\\n return (-1,None)\\n if (not matchline) or (not medianline):\\n self.logfile.write(\\\"Match program failed.\\\")\\n return (-1,None)\\n \\n mdict = self._process_median(medianline,mrad,xsh,ysh)\\n if not mdict:\\n return (-1,None)\\n return (0,mdict)\",\n \"def evaluate(self, training_scores, original_test_scores, imitation_test_scores):\\n\\n #finding a threshold: third to smallest training score\\n sorted_scores = np.sort(training_scores)\\n threshold = sorted_scores[2]\\n\\n #computing the number of errors\\n errors = len(np.where(original_test_scores < threshold)[0])\\n errors += len(np.where(imitation_test_scores > threshold)[0])\\n\\n #computing the local accuracy\\n accuracy = 1 - errors/(len(original_test_scores)+len(imitation_test_scores))\\n return accuracy, threshold\",\n \"def draw_matches(img1, kp1, img2, kp2, matches, inliers, ignore_indexes, filter_by_dist=True, color=None):\\n # We're drawing them side by side. Get dimensions accordingly.\\n # Handle both color and grayscale images.\\n\\n\\n\\n img1 = cv2.cvtColor(img1, cv2.COLOR_GRAY2RGB)\\n img2 = cv2.cvtColor(img2, cv2.COLOR_GRAY2RGB)\\n if len(img1.shape) == 3:\\n new_shape = (img1.shape[0] + img2.shape[0], max(img1.shape[1], img2.shape[1]), img1.shape[2])\\n elif len(img1.shape) == 2:\\n new_shape = (img1.shape[0] + img2.shape[0], max(img1.shape[1], img2.shape[1]))\\n new_img = np.zeros(new_shape, type(img1.flat[0])) \\n # Place images onto the new image.\\n new_img[0:img1.shape[0],0:img1.shape[1]] = img1\\n new_img[img1.shape[0]:img1.shape[0]+img2.shape[0],0:img1.shape[1]] = img2\\n \\n # Draw lines between matches. Make sure to offset kp coords in second image appropriately.\\n r = 1\\n thickness = 1\\n if color:\\n c = color\\n\\n # print(new_img.shape)\\n distances = []\\n for m in matches:\\n distances.append(m.distance)\\n \\n dist_threshold = min(distances) * 2\\n # print(dist_threshold)\\n \\n for i, m in enumerate(matches):\\n if inliers:\\n if not i in inliers:\\n continue\\n if ignore_indexes:\\n if i in ignore_indexes:\\n continue\\n if filter_by_dist:\\n if m.distance > 50:\\n continue\\n \\n # Generate random color for RGB/BGR and grayscale images as needed.\\n if not color: \\n c = tuple(np.random.randint(0,256,3)) if len(img1.shape) == 3 else np.random.randint(0,256)\\n c = ( int (c [ 0 ]), int (c [ 1 ]), int (c [ 2 ])) \\n \\n # So the keypoint locs are stored as a tuple of floats. cv2.line(), like most other things,\\n # wants locs as a tuple of ints.\\n\\n try:\\n end1 = tuple(np.round(kp1[m.queryIdx].pt).astype(int))\\n end2 = tuple(np.round(kp2[m.trainIdx].pt).astype(int) + np.array([ 0, img1.shape[0]]))\\n cv2.line(new_img, end1, end2, c, thickness)\\n cv2.circle(new_img, end1, r, c, thickness)\\n cv2.circle(new_img, end2, r, c, thickness)\\n except:\\n continue\\n \\n\\n return new_img\",\n \"def test_compare_performance_vis_api(experiment_to_use):\\n experiment = experiment_to_use\\n # extract test stats only\\n test_stats = experiment.test_stats_full\\n viz_outputs = (\\\"pdf\\\", \\\"png\\\")\\n with TemporaryDirectory() as tmpvizdir:\\n for viz_output in viz_outputs:\\n vis_output_pattern_pdf = tmpvizdir + f\\\"/*.{viz_output}\\\"\\n visualize.compare_performance(\\n [test_stats, test_stats],\\n output_feature_name=None,\\n model_names=[\\\"Model1\\\", \\\"Model2\\\"],\\n output_directory=tmpvizdir,\\n file_format=viz_output,\\n )\\n figure_cnt = glob.glob(vis_output_pattern_pdf)\\n assert 1 == len(figure_cnt)\",\n \"def detection_precision(ground_df, detections_df, FWHMxy=1.337, pixel_size=0.16125, time_frame=0,\\n verbose=True, display=False, debug=False):\\n import numpy as np\\n import scipy.spatial.distance as dist\\n from matplotlib import pyplot as plt\\n import pandas as pd\\n\\n check_distance = (FWHMxy / 2) / pixel_size\\n\\n if debug:\\n print ('*** Check distance in voxels: ' + str(check_distance))\\n print ('*** Check distance in microns: ' + str(check_distance * pixel_size))\\n print ('')\\n\\n # Gets only the desired time frame from the ground truth\\n ground_df = ground_df[ground_df['t'] == time_frame]\\n\\n nground = len(ground_df[ground_df['t'] == time_frame])\\n ndetections = len(detections_df[detections_df['t'] == time_frame])\\n\\n if verbose:\\n print ('Coordinates on ground truth: ' + str(nground))\\n print ('Number of detections : ' + str(ndetections))\\n\\n # Gets matrix of coordinates\\n x_ground = ground_df['x'].values\\n y_ground = ground_df['y'].values\\n z_ground = ground_df['z'].values\\n ground_coords = [x_ground, y_ground, z_ground]\\n ground_coords = np.transpose(ground_coords)\\n\\n x_detections = detections_df['x'].values\\n y_detections = detections_df['y'].values\\n z_detections = detections_df['z'].values\\n detections_coords = [x_detections, y_detections, z_detections]\\n detections_coords = np.transpose(detections_coords)\\n\\n if debug:\\n print ('\\\\n*** Ground truth matrix:'),\\n print np.shape(ground_coords),\\n print (ground_coords.dtype)\\n print ('*** Detections matrix:'),\\n print np.shape(detections_coords),\\n print (ground_coords.dtype)\\n\\n dist_matrix = dist.cdist(detections_coords, ground_coords)\\n sorted_matrix_truth = np.sort(dist_matrix, axis=0)\\n # print (sorted_matrix_truth)\\n sorted_matrix_detection = np.sort(dist_matrix, axis=1)\\n\\n # Distance Matrix\\n if display:\\n plt.figure(figsize=(15, 10))\\n plt.imshow(sorted_matrix_truth, cmap='RdYlBu_r')\\n plt.title('Sorted distances based on truth')\\n plt.colorbar()\\n\\n plt.figure(figsize=(15, 10))\\n plt.imshow(sorted_matrix_detection, cmap='RdYlBu_r')\\n plt.title('Sorted distances based on detections')\\n plt.colorbar()\\n\\n # Get a vector with the closest points\\n closest_truth = np.sort(sorted_matrix_truth[0, :])\\n closest_detections = np.sort(sorted_matrix_detection[:, 0])\\n\\n # Who's smaller, detections or ground truth ?\\n max_possible_hits = min(len(closest_detections), len(closest_truth))\\n\\n if display:\\n plt.figure(figsize=(15, 5))\\n plt.plot(closest_detections, color='#30AAFF', linewidth=2, label='detections')\\n plt.plot(closest_truth, color='#30FFAA', linewidth=2, label='ground truth')\\n plt.legend(frameon=False, loc='upper left')\\n ax = plt.gca()\\n ax.spines['top'].set_visible(False)\\n ax.spines['right'].set_visible(False)\\n ax.spines['bottom'].set_visible(False)\\n ax.spines['left'].set_visible(True)\\n ax.xaxis.set_ticks_position('none')\\n ax.yaxis.set_ticks_position('left')\\n\\n # Distribution of closest\\n if display:\\n nbins = 50\\n plt.figure(figsize=(15, 8))\\n # plt.hist(minimum_distances_micron, 100, histtype='stepfilled', color='blue', label='blue')\\n plt.hist(closest_truth, nbins, color='#00AAFF', alpha=0.5)\\n plt.axvline(x=check_distance, linewidth=3, color='#FF3322')\\n plt.title('Distribution of minimum distances (' + str(nbins) + ' bins)')\\n plt.xlabel('Distance in voxels')\\n\\n hits_truth = (closest_truth < check_distance).sum()\\n hits_detections = (closest_detections < check_distance).sum()\\n hits = min(hits_truth, hits_detections)\\n\\n # Sanity check\\n if hits > max_possible_hits:\\n print ('WARNING: More hits than max possible hits. Something is wrong !')\\n\\n if debug:\\n print ('*** Truth hits: ' + str(hits_truth))\\n print ('*** Detection hits: ' + str(hits_detections))\\n\\n total_matches = (dist_matrix < check_distance).sum()\\n nduplicates = total_matches - hits\\n\\n if verbose:\\n print ('\\\\nNumber of hits: ' + str(hits) + ' (' + str((float(hits) / ndetections) * 100) + '%)')\\n print ('Found ' + str(nduplicates) + ' duplicates,'),\\n print ('total of ' + str(total_matches) + ' points closer than ' + str(FWHMxy / 2) + ' microns')\\n\\n # Calculation of Jaccard index\\n # Defined as intersection over the union\\n jaccard = float(hits) / ((ndetections - hits) + nground)\\n\\n # Average distance of detections\\n avg_dist = np.average(closest_truth)\\n\\n if verbose:\\n print ('\\\\nJaccard index: ' + str(jaccard))\\n print ('Average distance: ' + str(avg_dist) + ' voxels')\\n if verbose:\\n print ('\\\\nAll done !')\\n\\n return ndetections, nground, hits, jaccard, avg_dist\",\n \"def test_detector1pipeline4(_bigdata):\\n step = Detector1Pipeline()\\n step.save_calibrated_ramp = True\\n step.ipc.skip = True\\n step.persistence.skip = True\\n step.jump.rejection_threshold = 4.0\\n step.ramp_fit.save_opt = False\\n step.output_file = 'jw84600007001_02101_00001_nrs1_rate.fits'\\n step.run(_bigdata+'/pipelines/jw84600007001_02101_00001_nrs1_uncal.fits')\\n\\n # Compare ramp product\\n n_ramp = 'jw84600007001_02101_00001_nrs1_ramp.fits'\\n h = fits.open( n_ramp )\\n n_ref = _bigdata+'/pipelines/jw84600007001_02101_00001_nrs1_ramp_ref.fits'\\n href = fits.open( n_ref )\\n newh = fits.HDUList([h['primary'],h['sci'],h['err'],h['groupdq'],h['pixeldq']])\\n newhref = fits.HDUList([href['primary'],href['sci'],href['err'],href['groupdq'],h['pixeldq']])\\n result = fits.diff.FITSDiff(newh,\\n newhref,\\n ignore_keywords = ['DATE','CAL_VER','CAL_VCS','CRDS_VER','CRDS_CTX'],\\n rtol = 0.00001\\n )\\n assert result.identical, result.report()\\n\\n # Compare countrate image product\\n n_cr = 'jw84600007001_02101_00001_nrs1_rate.fits'\\n h = fits.open( n_cr )\\n n_ref = _bigdata+'/pipelines/jw84600007001_02101_00001_nrs1_rate_ref.fits'\\n href = fits.open( n_ref )\\n newh = fits.HDUList([h['primary'],h['sci'],h['err'],h['dq']])\\n newhref = fits.HDUList([href['primary'],href['sci'],href['err'],href['dq']])\\n result = fits.diff.FITSDiff(newh,\\n newhref,\\n ignore_keywords = ['DATE','CAL_VER','CAL_VCS','CRDS_VER','CRDS_CTX'],\\n rtol = 0.00001\\n )\\n assert result.identical, result.report()\",\n \"def statistics(img1_blobs, img2_blobs, matches):\\n\\tstatistics = {}\\n\\n\\tstatistics['#Interest Points in img1'] = len(img1_blobs)\\n\\tstatistics['#Interest Points in img2'] = len(img2_blobs)\\n\\tstatistics['Accepted Matches'] = len(matches)\\n\\tdissimilarity = [match[2] for match in matches]\\n\\tstatistics['Mean of accepted matches'] = sum(dissimilarity)/len(dissimilarity)\\n\\tstatistics['SD of accepted matches'] = np.std(dissimilarity)\\n\\treturn statistics\",\n \"def evaluate_prediction(self):\\n\\n # ratio_train = self.evaluate_data(self.train_x, self.train_y)\\n ratio_test = self.evaluate_data(self.test_x, self.test_y)\\n\\n print(\\\"\\\\n*NAIVE BAYES:\\\")\\n # print(\\\"Test1: {}%\\\".format(ratio_dev*100))\\n print(\\\"Test: {} %\\\".format(ratio_test*100))\",\n \"def compare_group_images(directory, show_imgs=True, similarity=\\\"high\\\", compression=100):\\n # list where the found duplicate/similar images are stored\\n duplicates = []\\n lower_res = []\\n\\n imgs_matrix = create_imgs_matrix(directory, compression)\\n\\n # search for similar images\\n if similarity == \\\"low\\\":\\n ref = 13000\\n # search for 1:1 duplicate images\\n else:\\n ref = 21000\\n\\n main_img = 0\\n compared_img = 1\\n nrows, ncols = compression, compression\\n srow_A = 0\\n erow_A = compression\\n srow_B = erow_A\\n erow_B = srow_B + compression\\n\\n while erow_B <= imgs_matrix.shape[0]:\\n while compared_img < (len(image_files)):\\n # select two images from imgs_matrix\\n imgA = imgs_matrix[srow_A: erow_A, # rows\\n 0: ncols] # columns\\n imgB = imgs_matrix[srow_B: erow_B, # rows\\n 0: ncols] # columns\\n # compare the images\\n rotations = 0\\n while image_files[main_img] not in duplicates and rotations <= 3:\\n if rotations != 0:\\n imgB = rotate_img(imgB)\\n err = mse(imgA, imgB)\\n print ( \\\"err:\\\", err)\\n if err <= ref:\\n if show_imgs == True:\\n show_file_info(compared_img, main_img)\\n add_to_list(image_files[main_img], duplicates)\\n check_img_quality(directory, image_files[main_img], image_files[compared_img], lower_res)\\n rotations += 1\\n srow_B += compression\\n erow_B += compression\\n compared_img += 1\\n\\n srow_A += compression\\n erow_A += compression\\n srow_B = erow_A\\n erow_B = srow_B + compression\\n main_img += 1\\n compared_img = main_img + 1\\n\\n msg = \\\"\\\\n***\\\\n DONE: found \\\" + str(len(duplicates)) + \\\" duplicate image pairs in \\\" + str(\\n len(image_files)) + \\\" total images.\\\\n The following files have lower resolution:\\\"\\n print(msg)\\n return set(lower_res)\"\n]","isConversational":false},"negative_scores":{"kind":"list like","value":["0.686315","0.66412973","0.6606622","0.6439272","0.64282435","0.636913","0.63653","0.6287904","0.6282749","0.61645573","0.61143124","0.6112964","0.6099393","0.60614634","0.6056027","0.60436153","0.60412055","0.60172075","0.6009123","0.6001078","0.59953034","0.5930585","0.5918237","0.59124386","0.58996034","0.5891404","0.58743525","0.5869658","0.58696187","0.58491373","0.5847201","0.5825413","0.581902","0.58166116","0.5814364","0.5800021","0.5799361","0.579918","0.5794971","0.57817227","0.5770978","0.5770619","0.57700384","0.57690525","0.57671785","0.5763364","0.5759514","0.57578593","0.57518065","0.57433844","0.57409483","0.5734114","0.573133","0.57189715","0.5715005","0.5711862","0.5708938","0.57072455","0.5706794","0.57035613","0.56948674","0.568267","0.5670641","0.566545","0.5658778","0.5656565","0.56504637","0.56495345","0.56490767","0.56457835","0.56441766","0.56416214","0.5639337","0.5636603","0.5635879","0.5633442","0.56325656","0.56318444","0.5625502","0.5624584","0.5621428","0.56080663","0.56057537","0.5604201","0.5596297","0.55928326","0.55925345","0.5582626","0.5577914","0.55770147","0.5569992","0.55678135","0.55674356","0.5564452","0.5562736","0.5551233","0.5544595","0.55437493","0.55428886","0.5542736","0.5542403"],"string":"[\n \"0.686315\",\n \"0.66412973\",\n \"0.6606622\",\n \"0.6439272\",\n \"0.64282435\",\n \"0.636913\",\n \"0.63653\",\n \"0.6287904\",\n \"0.6282749\",\n \"0.61645573\",\n \"0.61143124\",\n \"0.6112964\",\n \"0.6099393\",\n \"0.60614634\",\n \"0.6056027\",\n \"0.60436153\",\n \"0.60412055\",\n \"0.60172075\",\n \"0.6009123\",\n \"0.6001078\",\n \"0.59953034\",\n \"0.5930585\",\n \"0.5918237\",\n \"0.59124386\",\n \"0.58996034\",\n \"0.5891404\",\n \"0.58743525\",\n \"0.5869658\",\n \"0.58696187\",\n \"0.58491373\",\n \"0.5847201\",\n \"0.5825413\",\n \"0.581902\",\n \"0.58166116\",\n \"0.5814364\",\n \"0.5800021\",\n \"0.5799361\",\n \"0.579918\",\n \"0.5794971\",\n \"0.57817227\",\n \"0.5770978\",\n \"0.5770619\",\n \"0.57700384\",\n \"0.57690525\",\n \"0.57671785\",\n \"0.5763364\",\n \"0.5759514\",\n \"0.57578593\",\n \"0.57518065\",\n \"0.57433844\",\n \"0.57409483\",\n \"0.5734114\",\n \"0.573133\",\n \"0.57189715\",\n \"0.5715005\",\n \"0.5711862\",\n \"0.5708938\",\n \"0.57072455\",\n \"0.5706794\",\n \"0.57035613\",\n \"0.56948674\",\n \"0.568267\",\n \"0.5670641\",\n \"0.566545\",\n \"0.5658778\",\n \"0.5656565\",\n \"0.56504637\",\n \"0.56495345\",\n \"0.56490767\",\n \"0.56457835\",\n \"0.56441766\",\n \"0.56416214\",\n \"0.5639337\",\n \"0.5636603\",\n \"0.5635879\",\n \"0.5633442\",\n \"0.56325656\",\n \"0.56318444\",\n \"0.5625502\",\n \"0.5624584\",\n \"0.5621428\",\n \"0.56080663\",\n \"0.56057537\",\n \"0.5604201\",\n \"0.5596297\",\n \"0.55928326\",\n \"0.55925345\",\n \"0.5582626\",\n \"0.5577914\",\n \"0.55770147\",\n \"0.5569992\",\n \"0.55678135\",\n \"0.55674356\",\n \"0.5564452\",\n \"0.5562736\",\n \"0.5551233\",\n \"0.5544595\",\n \"0.55437493\",\n \"0.55428886\",\n \"0.5542736\",\n \"0.5542403\"\n]","isConversational":false},"document_score":{"kind":"string","value":"0.0"},"document_rank":{"kind":"string","value":"-1"}}},{"rowIdx":93294,"cells":{"query":{"kind":"string","value":"runs multiple experiments that calculates the precision and recall of multiple comparison algorithms' results"},"document":{"kind":"string","value":"def run_multiple_advanced_experiments(self, path_to_csv_files, algorithms_weights_list, thresholds):\n starting_time = time.time()\n print(\"started advanced precision recall image_matching_module\")\n for algorithms_weights in algorithms_weights_list:\n start_file_time = time.time()\n\n first_run = True\n for algorithm, weight in algorithms_weights:\n if first_run:\n algorithm_name = algorithm.get_algorithm_name() + weight\n first_run = False\n else:\n algorithm_name = algorithm_name + \"_\" + algorithm.get_algorithm_name() + weight\n\n print(\"started algorithm:\", algorithm_name)\n self.run_single_advanced_experiment(path_to_csv_files, algorithms_weights, thresholds)\n finish_file_time = time.time()\n print(\"finished algorithm:\", algorithm_name)\n print(\"time took:\", (finish_file_time - start_file_time) / 60, \"minutes\")\n finish_time = time.time()\n hours = int((finish_time - starting_time) // 3600)\n minutes = (finish_time - starting_time) / 60\n print(\"finished advanced precision recall image_matching_module\")\n print(\"time image_matching_module took:\", hours, \"hours and\", minutes, \"minutes\")"},"metadata":{"kind":"string","value":"{\n \"objective\": {\n \"self\": [],\n \"paired\": [],\n \"triplet\": [\n [\n \"query\",\n \"document\",\n \"negatives\"\n ]\n ]\n }\n}"},"negatives":{"kind":"list like","value":["def run_experiments():\n if True: # Change to False when done finding max_scoring_num_rolls\n six_sided_max = max_scoring_num_rolls(six_sided)\n print('Max scoring num rolls for six-sided dice:', six_sided_max)\n rerolled_max = max_scoring_num_rolls(reroll(six_sided))\n print('Max scoring num rolls for re-rolled dice:', rerolled_max)\n\n if False: # Change to True to test always_roll(8)\n print('always_roll(8) win rate:', average_win_rate(always_roll(8)))\n\n if False: # Change to True to test bacon_strategy\n print('bacon_strategy win rate:', average_win_rate(bacon_strategy))\n\n if False: # Change to True to test swap_strategy\n print('swap_strategy win rate:', average_win_rate(swap_strategy))\n\n \"*** You may add additional experiments as you wish ***\"","def run_experiments():\n if True: # Change to False when done finding max_scoring_num_rolls\n six_sided_max = max_scoring_num_rolls(six_sided)\n print('Max scoring num rolls for six-sided dice:', six_sided_max)\n four_sided_max = max_scoring_num_rolls(four_sided)\n print('Max scoring num rolls for four-sided dice:', four_sided_max)\n\n if False: # Change to True to test always_roll(8)\n print('always_roll(8) win rate:', average_win_rate(always_roll(8)))\n\n if True: # Change to True to test bacon_strategy\n print('bacon_strategy win rate:', average_win_rate(bacon_strategy))\n\n if True: # Change to True to test swap_strategy\n print('swap_strategy win rate:', average_win_rate(swap_strategy))\n\n if True:\n print('final_strategy win rate:', average_win_rate(final_strategy))\n\n \"*** You may add additional experiments as you wish ***\"","def run_experiments():\n if False: # Change to False when done finding max_scoring_num_rolls\n six_sided_max = max_scoring_num_rolls(six_sided)\n print('Max scoring num rolls for six-sided dice:', six_sided_max)\n\n if False: # Change to True to test always_roll(8)\n print('always_roll(8) win rate:', average_win_rate(always_roll(8)))\n\n if False: # Change to True to test bacon_strategy\n print('bacon_strategy win rate:', average_win_rate(bacon_strategy))\n\n if False: # Change to True to test swap_strategy\n print('swap_strategy win rate:', average_win_rate(swap_strategy))\n\n if True: # Change to True to test final_strategy\n print('final_strategy win rate:', average_win_rate(final_strategy))\n\n \"*** You may add additional experiments as you wish ***\"","def all_experiments():\n elo_explain_experiments()\n alpha_beta_experiments()\n mtcs_experiments()","def run_experiments():\r\n result = eval_strategy_range(always_roll, 1, 10)\r\n print('Best always_roll strategy:', result)\r\n\r\n if True: # Change to True when ready to test make_comeback_strategy\r\n result = eval_strategy_range(make_comeback_strategy, 5, 15)\r\n print('Best comeback strategy:', result)\r\n\r\n if True: # Change to True when ready to test make_mean_strategy\r\n result = eval_strategy_range(make_mean_strategy, 1, 10)\r\n print('Best mean strategy:', result)\r\n\r\n \"*** You may add additional experiments here if you wish ***\"","def main():\n run_test_summary1a()\n run_test_summary1c()\n run_test_summary1c()","def run_experiments():\n nets = []\n nets.extend(shallow(n=1))\n nets.extend(basic_conv(n=1))\n nets.extend(omit_FC(n=1))\n nets.extend(dbl_conv(n=1, activation_fn=sigmoid))\n # omitted, but still interesting: regularized_dbl_conv()\n nets.extend(dbl_conv_relu(n=1))\n nets.extend(expanded_data(n=1, FC_n=100))\n nets.extend(expanded_data(n=1, FC_n=300))\n nets.extend(expanded_data(n=1, FC_n=1000))\n nets.extend(expanded_data_double_fc(n=1, FC_n=100)) \n nets.extend(expanded_data_double_fc(n=1, FC_n=300))\n nets.extend(expanded_data_double_fc(n=1, FC_n=1000))\n nets.extend(double_fc_dropout(0.5, 0.5, 0.5, n=5))\n # plot the erroneous digits in the ensemble of nets just trained\n error_locations, erroneous_predictions = ensemble(nets)\n plt = plot_errors(error_locations, erroneous_predictions)\n plt.savefig(\"ensemble_errors.png\")\n # plot the filters learned by the first of the nets just trained\n plt = plot_filters(nets[0], 0, 5, 4)\n plt.savefig(\"net_full_layer_0.png\")\n plt = plot_filters(nets[0], 1, 8, 5)\n plt.savefig(\"net_full_layer_1.png\")","def run_multiple_simple_experiments(self, path_to_csv_files, algorithms, thresholds):\n starting_time = time.time()\n print(\"started simple precision recall image_matching_module\")\n for algorithm in algorithms:\n start_file_time = time.time()\n print(\"started algorithm:\", algorithm.get_algorithm_name())\n self.run_single_simple_experiment(path_to_csv_files, algorithm, thresholds)\n finish_file_time = time.time()\n print(\"finished algorithm:\", algorithm.get_algorithm_name())\n print(\"time took:\", (finish_file_time - start_file_time) / 60, \"minutes\")\n finish_time = time.time()\n hours = int((finish_time - starting_time) // 3600)\n minutes = (finish_time - starting_time) / 60\n print(\"finished simple precision recall image_matching_module\")\n print(\"time image_matching_module took:\", hours, \"hours and\", minutes, \"minutes\")","def run_experiments(size, approach):\n\n # Create folder (if needed) where to store query evaluation logs and raw results\n if not os.path.isdir(f\"{path_reports}/eval-query-{current_run}\"):\n shell(f\"mkdir -p {path_reports}/eval-query-{current_run}\")\n\n # Create folder (if needed) where to store store evaluation logs and raw results\n if not os.path.isdir(f\"{path_reports}/eval-store-{current_run}\"):\n shell(f\"mkdir -p {path_reports}/eval-store-{current_run}\")\n\n # Helper function computing the path of a file/folder for the given query evaluation iteration \n def query_path(iteration, filename=None):\n folder = size[\"id\"] + \"_\" + str(iteration)\n return path_reports + \"/eval-query-\" + current_run + \"/\" + folder + (\"/\" + filename if filename != None else \"\") \n\n # Helper function computing the path of a file/folder for the given store evaluation test / cpu setting\n def store_path(test, cpu, filename=None):\n folder = size[\"id\"] + \"_\" + approach[\"id\"] + \"_\" + test + (\"_\" + cpu[\"id\"] if cpu != None else \"\")\n return path_reports + \"/eval-store-\" + current_run + \"/\" + folder + (\"/\" + filename if filename != None else \"\")\n \n # Determine whether partial traces and named graphs are supported\n partial = approach[\"supports_partial\"]\n graphs = approach[\"supports_graphs\"]\n \n # Skip setting if all data is available (check for presence of log files - delete them to repeat test)\n may_skip = (not sp_enable or not partial or not graphs or os.path.isfile(store_path(\"sp\", None, \"eval.log\")))\n may_skip = may_skip and (not sf_enable or os.path.isfile(store_path(\"sf\", None, \"eval.log\")))\n if query_enable and approach[\"id\"] == query_approach_id:\n for i in range(0, query_num_iterations):\n may_skip = may_skip and os.path.isfile(query_path(i, \"eval.log\"))\n for cpu in cpus:\n may_skip = may_skip and (not pp_enable or not partial or not graphs or os.path.isfile(store_path(\"pp\", cpu, \"eval.log\")))\n may_skip = may_skip and (not pf_enable or os.path.isfile(store_path(\"pf\", cpu, \"eval.log\")))\n if may_skip:\n return\n\n # Delete (if needed) and extract again the repository from its .tar.xz file, so to work on a clean repository (at the end of this test, the repository is no more clean)\n path_dump = path_data + \"/\" + size[\"id\"] + \"_\" + approach[\"id\"] + \".tar.lz\"\n path_repo = path_repos + \"/\" + size[\"id\"] + \"_\" + approach[\"id\"]\n if not os.path.isfile(path_dump):\n log(f\"Missing required file {path_dump}\")\n sys.exit()\n if os.path.isdir(path_repo):\n shell(f\"rm -Rf {path_repo}\")\n shell(f\"{cmd_plzip} -kdc {path_dump} | tar xf - -C {path_repos}\")\n\n # Locate the repository URL\n repo_url = f\"http://localhost:{server_port}/repositories/promo\"\n \n # Query test (if enabled)\n if query_enable and approach[\"id\"] == query_approach_id:\n for i in range(0, query_num_iterations):\n if not os.path.isfile(query_path(i, \"eval.log\")):\n shell(f\"mkdir -p {query_path(i)}\")\n shell(f\"taskset -a {query_taskset} {cmd_graphdb} start {path_repo}\")\n shell(f\"taskset -a {query_taskset} {cmd_mokip} eval-query -w -u {repo_url} -q {path_queries} -r {query_path(i)} \" +\n f\"| tee {query_path(i, 'eval.log')}\")\n shell(f\"taskset -a {query_taskset} {cmd_graphdb} stop {path_repo}\")\n\n # Sequential Partial test (to assess store times per trace and their components)\n if sp_enable and partial and graphs and not os.path.isfile(store_path(\"sp\", None, \"eval.log\")):\n shell(f\"mkdir -p {store_path('sp', None)}\")\n shell(f\"taskset -a {sp_taskset} {cmd_graphdb} start {path_repo}\")\n shell(f\"taskset -a {sp_taskset} {cmd_mokip} eval-store -d {path_data}/traces_pp.jsonl.gz \" + \n f\"-u {repo_url} -i {approach['inference']} -U REPLACE_GRAPH_PROTOCOL \" + \n f\"-o {path_ontology} --trace-namespace '{namespace_trace}' --graph-namespace '{namespace_graph}' \" +\n f\"-T {timeout} -r {store_path('sp', None)} -t 1 -w 50 -p -D \" + \n f\"| tee {store_path('sp', None, 'eval.log')}\")\n shell(f\"taskset -a {sp_taskset} {cmd_graphdb} stop {path_repo}\")\n\n # Parallel Partial (to assess throughput, varying # of CPU cores)\n for cpu in cpus:\n if pp_enable and partial and graphs and not os.path.isfile(store_path(\"pp\", cpu, \"eval.log\")):\n shell(f\"mkdir -p {store_path('pp', cpu)}\")\n shell(f\"taskset -a {cpu['taskset']} {cmd_graphdb} start {path_repo}\")\n shell(f\"taskset -a {cpu['taskset']} {cmd_mokip} eval-store -d {path_data}/traces_pp.jsonl.gz \" +\n f\"-u {repo_url} -i {approach['inference']} -U REPLACE_GRAPH_PROTOCOL \" + \n f\"-o {path_ontology} --trace-namespace '{namespace_trace}' --graph-namespace '{namespace_graph}' \" + \n f\"-T {timeout} -r {store_path('pp', cpu)} -t {max(2, cpu['num_threads'])} -w 50 -p -D \" + \n f\"| tee {store_path('pp', cpu, 'eval.log')}\")\n shell(f\"taskset -a {cpu['taskset']} {cmd_graphdb} stop {path_repo}\")\n\n # Sequential Full test (to assess store times per trace and their components)\n if sf_enable and not os.path.isfile(store_path(\"sf\", None, \"eval.log\")):\n shell(f\"mkdir -p {store_path('sf', None)}\")\n shell(f\"taskset -a {sf_taskset} {cmd_graphdb} start {path_repo}\")\n shell(f\"taskset -a {sf_taskset} {cmd_mokip} eval-store -d {path_data}/traces_sf.jsonl.gz \" + \n f\"-u {repo_url} -i {approach['inference']} -U {'APPEND' if graphs else 'APPEND_DEFAULT_GRAPH'} \" +\n f\"-o {path_ontology} --trace-namespace '{namespace_trace}' --graph-namespace '{namespace_graph}' \" +\n f\"-T {timeout} -r {store_path('sf', None)} -t 1 -w 50 \" + \n f\"| tee {store_path('sf', None, 'eval.log')}\")\n shell(f\"taskset -a {sf_taskset} {cmd_graphdb} stop {path_repo}\")\n\n # Parallel Full (to assess throughput where data is also deleted, varying # of CPU cores)\n for cpu in cpus:\n if pf_enable and not os.path.isfile(store_path(\"pf\", cpu, \"eval.log\")):\n update = \"APPEND\" if graphs else \"APPEND_DEFAULT_GRAPH\"\n shell(f\"mkdir -p {store_path('pf', cpu)}\")\n shell(f\"taskset -a {cpu['taskset']} {cmd_graphdb} start {path_repo}\")\n shell(f\"taskset -a {cpu['taskset']} {cmd_mokip} eval-store -d {path_data}/traces_pf_{cpu['id']}.jsonl.gz \" +\n f\"-u {repo_url} -i {approach['inference']} -U {'APPEND' if graphs else 'APPEND_DEFAULT_GRAPH'} \" + \n f\"-o {path_ontology} --trace-namespace '{namespace_trace}' --graph-namespace '{namespace_graph}' \" +\n f\"-T {timeout} -r {store_path('pf', cpu)} -t {max(2, cpu['num_threads'])} -w 50 \" + \n f\"| tee {store_path('pf', cpu, 'eval.log')}\")\n shell(f\"taskset -a {cpu['taskset']} {cmd_graphdb} stop {path_repo}\")\n\n # Drop the repository (both to save space and since it is not clean anymore)\n shell(f\"rm -Rf {path_repo}\")","def three_experiments(two_experiments, one_experiment):","def run_Experiment(DP = None, QL = None):\n\n # Path information\n output_path, exp_num = create_new_dir() #dirs Exp/1, Exp/2, ...\n DP_path = join(output_path,'DP') #dirs Exp/1/DP\n QL_path = join(output_path,'QL') #dirs Exp/1/QL\n print(\"************ Exp \", exp_num, \"************ \\n\")\n\n # Exp_summary_data\n method = get_method_str(DP, QL)\n exp_summary = [str(exp_num), method]\n\n\n # Run DP\n if DP != None:\n print(\"In Runner: Executing DP !!\")\n\n prob_file = DP[0]\n createFolder(DP_path)\n # output_params = [V_so, mean, variance, bad_count]\n output_params = run_DP(setup_grid_params, prob_file, output_file, DP_path, threshold = threshold)\n\n \"\"\"CHANGE ARGUMENT if return order of setup_grid() is changed\"\"\"\n input_params = setup_grid_params[9].copy()\n input_params.append(prob_file)\n\n exp_summary = append_params_to_summary(exp_summary, input_params, output_params)\n append_summary_to_summaryFile('Experiments/Exp_summary.csv', exp_summary)\n print(\"In Runner: Executing DP Finished!!\")\n\n # Run QL\n if QL != None:\n print(\"In Runner: Executing QL !!\")\n\n QL_params = QL\n createFolder(QL_path)\n output_parameters_all_cases = run_QL(setup_grid_params, QL_params, QL_path, exp_num)\n # run_QL(setup_grid_params, QL_params, QL_path)\n\n print(\"In Runner: Executing QL Finished !!\")","def experiment1_outliers():\n\tdata_folder = \"ckan_subset/prepared_learnset/\"\n\ttest_folder = 'ckan_subset/testset/xml_csv/'\n\tgm = Graph_Maker()\n\tgm.store()\n\trounds = 5\n\tx = [\"Fingerprint\", \"Syntax Feature Model\", \"Word2Vec Matcher\"]\n\t\n\tnumber_of_classes = 15\n\texamples_per_class = 0\n\taccuracies = []\n\tprecisions = []\n\trecalls = []\n\tfmeasures = []\n\tsf_main = Storage_Files(data_folder, classes)\n\ttmp_acc = []\n\ttmp_prec = []\n\ttmp_rec = []\n\ttmp_fmeasure = []\n\ttotal_actual = []\n\ttotal_predicted = []\n\n\tfor i in range(0, rounds):\n\t\tprint(\"Fingerprint\")\n\t\t# --- Fingerprint\n\t\tccc = Column_Classification_Config()\n\t\tccc.add_feature('feature_main', 'Fingerprint', [sf_main, number_of_classes, examples_per_class, False, False])\n\n\t\tccc.add_matcher('matcher', 'Fingerprint_Matcher', {'feature_main': 'fingerprint'}) # main classifier\n\t\tsm = Schema_Matcher(ccc)\n\t\tactual, predicted = execute_test_ckan(sm, test_folder, False)\n\t\ttotal_actual += actual\n\t\ttotal_predicted += predicted\n\t\taccuracy = accuracy_score(actual, predicted)\n\t\ttmp_acc.append(accuracy)\n\t\ttmp_prec.append(precision(actual, predicted))\n\t\ttmp_rec.append(recall(actual, predicted))\n\t\ttmp_fmeasure.append(f_measure(actual, predicted))\n\n\taccuracies.append( round(sum(tmp_acc) / float(rounds), 2) )\n\tprecisions.append( round(sum(tmp_prec) / float(rounds), 2) )\n\trecalls.append( round(sum(tmp_rec) / float(rounds), 2) )\n\tfmeasures.append(round(sum(tmp_fmeasure) / float(rounds), 2))\n\tclassnames = list(set(get_class_names(total_actual) + get_class_names(total_predicted)))\n\tcm = confusion_matrix(total_actual, total_predicted, labels=classnames)\n\t#gm.plot_confusion_matrix(cm, classnames, normalize=True)\n\t\n\ttmp_acc = []\n\ttmp_prec = []\n\ttmp_rec = []\n\ttmp_fmeasure = []\n\ttotal_actual = []\n\ttotal_predicted = []\n\tfor i in range(0, rounds):\n\t\tprint(\"SFM\")\n\t\t# --- Syntax Feature Model\n\t\tccc = Column_Classification_Config()\n\t\tccc.add_feature('feature_main', 'Syntax_Feature_Model', [sf_main, 1, 0, False, False])\n\n\t\tccc.add_matcher('matcher', 'Syntax_Matcher', {'feature_main': 'syntax'}) # main classifier\n\t\tsm = Schema_Matcher(ccc)\n\t\tactual, predicted = execute_test_ckan(sm, test_folder, False)\n\t\ttotal_actual += actual\n\t\ttotal_predicted += predicted\n\t\taccuracy = accuracy_score(actual, predicted)\n\t\ttmp_acc.append(accuracy)\n\t\ttmp_prec.append(precision(actual, predicted))\n\t\ttmp_rec.append(recall(actual, predicted))\n\t\ttmp_fmeasure.append(f_measure(actual, predicted))\n\n\taccuracies.append( round(sum(tmp_acc) / float(rounds), 2) )\n\tprecisions.append( round(sum(tmp_prec) / float(rounds), 2) )\n\trecalls.append( round(sum(tmp_rec) / float(rounds), 2) )\n\tfmeasures.append(round(sum(tmp_fmeasure) / float(rounds), 2))\n\tclassnames = list(set(get_class_names(total_actual) + get_class_names(total_predicted)))\n\tcm = confusion_matrix(total_actual, total_predicted, labels=classnames)\n\t#gm.plot_confusion_matrix(cm, classnames, normalize=True)\n\n\ttmp_acc = []\n\ttmp_prec = []\n\ttmp_rec = []\n\ttmp_fmeasure = []\n\ttotal_actual = []\n\ttotal_predicted = []\n\tfor i in range(0, rounds):\n\t\tprint(\"W2V\")\n\t\t# --- Word2Vec Matcher\n\t\tccc = Column_Classification_Config()\n\t\tccc.add_feature('feature_main', 'Corpus', [sf_main, number_of_classes, examples_per_class, False, False])\n\n\t\tccc.add_matcher('matcher', 'Word2Vec_Matcher', {'feature_main': 'corpus'}) # main classifier\n\t\tsm = Schema_Matcher(ccc)\n\t\tactual, predicted = execute_test_ckan(sm, test_folder, False)\n\t\ttotal_actual += actual\n\t\ttotal_predicted += predicted\n\t\taccuracy = accuracy_score(actual, predicted)\n\t\ttmp_acc.append(accuracy)\n\t\ttmp_prec.append(precision(actual, predicted))\n\t\ttmp_rec.append(recall(actual, predicted))\n\t\ttmp_fmeasure.append(f_measure(actual, predicted))\n\n\taccuracies.append( round(sum(tmp_acc) / float(rounds), 2) )\n\tprecisions.append( round(sum(tmp_prec) / float(rounds), 2) )\n\trecalls.append( round(sum(tmp_rec) / float(rounds), 2) )\n\tfmeasures.append(round(sum(tmp_fmeasure) / float(rounds), 2))\n\tclassnames = list(set(get_class_names(total_actual) + get_class_names(total_predicted)))\n\tcm = confusion_matrix(total_actual, total_predicted, labels=classnames)\n\t#gm.plot_confusion_matrix(cm, classnames, normalize=True)\n\n\tgm.add_x(x)\n\t# accuracies = [0.4, 0.4, 0.4]\n\t# precisions = [0.5, 0.5, 0.5]\n\t# recalls = [0.62, 0.62, 0.62]\n\t# fmeasures = [0.23, 0.23, 0.28]\n\tgm.append_y(accuracies)\n\tgm.append_y(precisions)\n\tgm.append_y(recalls)\n\tgm.append_y(fmeasures)\n\tgm.store()\n\tsubtitle = \"Scores were averaged over \" + str(rounds) + \" tests with \" + str(len(classes)) + \" classes. \" + \\\n\t\"Number of simulated columns per class: \" + str(number_of_classes)\n\tlabels = [\"Accuracy\", \"Precision\", \"Recall\", \"F-Measure\"]\n\tgm.plot_bar_n(\"Matcher Type\", \"Score\", \"Accuracy of Matchers\", labels, subtitle=subtitle)","def run_and_evaluate():\n tsp_problems = read_all_problems()\n # Empty list of metrics\n results = []\n for problem in tqdm.tqdm(tsp_problems):\n # As random factors are involved repeat experiments a couple of times\n best_routes_base = []\n best_routes_af = []\n best_routes_ms = []\n base_times = []\n af_times = []\n ms_times = []\n for i in range(10):\n # Base solution\n start_time = timeit.default_timer()\n best_route_base = solve_tsp_basic(problem)\n base_time = timeit.default_timer() - start_time\n best_routes_base.append(Fitness(route=best_route_base).route_distance())\n base_times.append(base_time)\n\n # AF clustering solution\n start_time = timeit.default_timer()\n best_route_af = solve_tsp_affinity_propagation(problem)\n af_time = timeit.default_timer() - start_time\n best_routes_af.append(Fitness(route=best_route_af).route_distance())\n af_times.append(af_time)\n\n # MS solution\n start_time = timeit.default_timer()\n best_route_ms = solve_mean_shift(problem)\n ms_time = timeit.default_timer() - start_time\n best_routes_ms.append(Fitness(route=best_route_ms).route_distance())\n ms_times.append(ms_time)\n\n results.append(\n {\n \"problem name\": problem.name,\n \"optimal solution\": find_route_optimal_route_length(problem),\n \"baseline tour length\": mean(best_routes_base),\n \"af clustering tour length\": mean(best_routes_af),\n \"ms clustering tour length\": mean(best_routes_ms),\n \"baseline algorithm time\": mean(base_times),\n \"af clustering algorithm time\": mean(af_times),\n \"ms clustering algorithm time\": mean(ms_times),\n }\n )\n # Create dataframe and safe results\n df = pd.DataFrame(results)\n df.to_csv(\"results.csv\", index=False)\n return df","def main():\r\n algos = [merge_sort, quick_sort, heap_sort, radix_sort, bucket_sort_general]\r\n array_sizes = [5000, 10000, 15000, 20000, 50000, 75000, 100000, 150000]\r\n results = {algo.__name__: [] for algo in algos}\r\n for algo in algos:\r\n result = []\r\n for size in array_sizes:\r\n time = test(algo, size)\r\n result.append(time)\r\n results[algo.__name__] = result\r\n\r\n display_results(results, array_sizes)","def three_experiments_with_trials(family_with_trials, single_with_trials):","def evaluate_model(args, eval_runs, warm_runs, metrics=['psnr', 'ssim', 'fps']):\n upsampler = Upsampler(args)\n if warm_runs > 0:\n print(\"Warming up for evaluation\")\n for i in range(warm_runs):\n print(\"Performing warm-up run\", str(i+1))\n for sequence in ['foliage', 'walk', 'calendar', 'city']:\n bix_dir = os.path.join(VID4_DIR, 'BIx4', sequence)\n upsampler.run_dir(bix_dir, reset=False)\n \n time = 0.\n psnrs = []\n ssims = []\n for i in range(eval_runs):\n run_psnrs = []\n run_ssims = []\n print(\"Performing evaluation run\", str(i+1))\n for sequence in ['foliage', 'walk', 'calendar', 'city']:\n bix_dir = os.path.join(VID4_DIR, 'BIx4', sequence)\n gt_dir = os.path.join(VID4_DIR, 'GT', sequence)\n print(\"Evaluating on\", bix_dir)\n time += upsampler.run_dir(bix_dir, reset=False)\n vid_psnrs, vid_ssims = _eval_sr_perf(os.path.join(bix_dir, 'up'), gt_dir)\n run_psnrs += vid_psnrs\n run_ssims += vid_ssims\n if i == eval_runs-1:\n with open(os.path.join(upsampler.get_model_dir(), \"psnr.txt\"), \"w\") as f:\n f.writelines(str(psnr) + '\\n' for psnr in run_psnrs)\n with open(os.path.join(upsampler.get_model_dir(), \"ssim.txt\"), \"w\") as f:\n f.writelines(str(ssim) + '\\n' for ssim in run_ssims)\n psnrs += run_psnrs\n ssims += run_ssims\n\n fps = VID4_LENGTH/ (time/eval_runs)\n return Performance(psnr=psnrs, ssim=ssims, fps=fps)","def run_analyses(y_predict_train, y_train, y_predict, y_test):\n # calculate metrics\n _, training_error = output_error(y_predict_train, y_train)\n (precision, recall, f1, _), testing_error = output_error(y_predict, y_test)\n \n # print out metrics\n print 'Average Precision:', np.average(precision)\n print 'Average Recall:', np.average(recall)\n print 'Average F1:', np.average(f1)\n print 'Training Error:', training_error\n print 'Testing Error:', testing_error","def main():\n run_simulation(spectral=False, ml=False, num_procs=1)\n run_simulation(spectral=True, ml=False, num_procs=1)\n run_simulation(spectral=False, ml=True, num_procs=1)\n run_simulation(spectral=True, ml=True, num_procs=1)\n run_simulation(spectral=False, ml=True, num_procs=10)\n run_simulation(spectral=True, ml=True, num_procs=10)","def evaluate(cfg: DictConfig):\n\n experiments = cfg.get('experiment_type', f'{cfg.model.name}_only')\n fixed_t0 = cfg.get('fixed_t0', False)\n ext = '_fixedT0' if fixed_t0 else ''\n\n base_dir = cfg.device.root\n datasource = cfg.datasource.name\n\n if experiments == 'ablations':\n models = {\n 'FluxRGNN': ['final',\n 'final_without_encoder',\n 'final_without_boundary'],\n 'LocalLSTM': ['final']\n }\n elif experiments == 'final':\n models = {\n 'FluxRGNN': ['final'],\n 'GAM': ['final'],\n 'HA': ['final'],\n 'GBT': ['final']\n }\n else:\n m = cfg.model.name\n year = cfg.datasource.test_year\n\n # find all experiments available for this model, datasource and test year\n result_dir = osp.join(base_dir, 'results', datasource, m, f'test_{year}')\n models = {\n m : [ f.name for f in os.scandir(result_dir) if f.is_dir() ]\n }\n\n\n # thresholds for binary classification metrics\n if cfg.datasource.name == 'abm':\n thresholds = [0.0019, 0.0207]\n else:\n thresholds = [0, 10, 20]\n\n rmse_per_hour = []\n mae_per_hour = []\n pcc_per_hour = []\n bin_per_hour = []\n\n rmse_per_night = []\n mae_per_night = []\n\n output_dir = osp.join(base_dir, 'results', datasource, f'performance_evaluation{ext}', experiments)\n os.makedirs(output_dir, exist_ok=True)\n\n counter = 0\n\n for m, dirs in models.items():\n print(f'evaluate {m}')\n\n for d in dirs:\n result_dir = osp.join(base_dir, 'results', datasource, m, f'test_{cfg.datasource.test_year}', d)\n\n # check if directory exists\n if os.path.isdir(result_dir):\n results, model_cfg = load_cv_results(result_dir, trials=cfg.task.repeats, ext=ext)\n\n df_prep = pd.read_csv(osp.join(base_dir, 'data', 'preprocessed',\n f'{model_cfg[\"t_unit\"]}_{model_cfg[\"model\"][\"edge_type\"]}_ndummy={model_cfg[\"datasource\"][\"n_dummy_radars\"]}',\n datasource, cfg.season, str(cfg.datasource.test_year), 'dynamic_features.csv'))\n tidx2night = dict(zip(df_prep.tidx, df_prep.nightID))\n\n rmse_per_hour.append(compute_rmse(m, d, results, tidx2night, groupby=['horizon', 'trial'],\n threshold=0, km2=True, fixed_t0=fixed_t0))\n mae_per_hour.append(compute_mae(m, d, results, tidx2night, groupby=['horizon', 'trial'],\n threshold=0, km2=True, fixed_t0=fixed_t0))\n pcc_per_hour.append(compute_pcc(m, d, results, tidx2night, groupby=['horizon', 'trial'],\n threshold=0, km2=True, fixed_t0=fixed_t0))\n\n if fixed_t0:\n rmse_per_night.append(compute_rmse_per_night(m, d, results, tidx2night, groupby=['night_horizon', 'trial']))\n mae_per_night.append(compute_mae_per_night(m, d, results, tidx2night, groupby=['night_horizon', 'trial']))\n\n # compute binary classification measures\n for thr in thresholds:\n bin_per_hour.append(compute_bin(m, d, results, groupby=['horizon', 'trial'], threshold=thr, km2=True))\n\n counter += 1\n\n else:\n print(f'Experiment \"{d}\" for model \"{m}\" and datasource \"{datasource}\" is not available. '\n f'Use \"run_experiments.py model={m} datasource={datasource} +experiment={d}\" to run this experiment.')\n\n if counter > 0:\n rmse_per_hour = pd.concat(rmse_per_hour)\n rmse_per_hour.to_csv(osp.join(output_dir, f'rmse_per_hour.csv'))\n\n mae_per_hour = pd.concat(mae_per_hour)\n mae_per_hour.to_csv(osp.join(output_dir, f'mae_per_hour.csv'))\n\n pcc_per_hour = pd.concat(pcc_per_hour)\n pcc_per_hour.to_csv(osp.join(output_dir, f'pcc_per_hour.csv'))\n\n bin_per_hour = pd.concat(bin_per_hour)\n bin_per_hour.to_csv(osp.join(output_dir, f'bin_per_hour.csv'))\n\n if fixed_t0:\n rmse_per_night = pd.concat(rmse_per_night)\n rmse_per_night.to_csv(osp.join(output_dir, f'rmse_per_night.csv'))\n\n mae_per_night = pd.concat(mae_per_night)\n mae_per_night.to_csv(osp.join(output_dir, f'mae_per_night.csv'))","def basic_experiments(k, p, num_iter, transforms, train_path, test_path):\n start_time = time.time()\n print(\"Generating advice\")\n with open(train_path, \"r\") as f:\n train_data = f.readlines()\n advice = generate_advice(train_data, transforms)\n print(\"Done: \" + str(time.time() - start_time) + \"\\n\")\n\n print(\"Generating real count for test data\")\n with open(test_path, \"r\") as f:\n test_data = f.readlines()\n real_count = generate_advice(test_data, transforms)\n print(\"Done: \" + str(time.time() - start_time))\n actual = [x.moment(p) for x in real_count]\n print(\"Actual moments:\")\n print(actual)\n\n print(\"Estimating using noisy advice:\")\n est_with_advice = []\n for i in range(num_iter):\n seed = str(random.randint(1, 2**64 + 1)) + str(random.randint(1, 2**64 + 1))\n this_hash_exp = lambda x: hash_exp(x, seed)\n est_with_advice.append(\n estimate_using_advice(test_data, transforms, k, p, this_hash_exp,\n advice))\n print(\"Done: \" + str(time.time() - start_time))\n print(est_with_advice)\n\n print(\"Estimating using perfect advice:\")\n est_perfect_advice = []\n for i in range(num_iter):\n seed = str(random.randint(1, 2**64 + 1)) + str(random.randint(1, 2**64 + 1))\n this_hash_exp = lambda x: hash_exp(x, seed)\n est_perfect_advice.append(\n estimate_using_advice(test_data, transforms, k, p, this_hash_exp,\n real_count))\n print(\"Done: \" + str(time.time() - start_time))\n print(est_perfect_advice)\n\n print(\"Estimating using PPSWOR:\")\n est_ppswor = []\n for i in range(num_iter):\n cur = []\n for x in real_count:\n cur.append(ppswor_estimate_moment(x.counts.items(), k, p))\n est_ppswor.append(cur)\n print(\"Done: \" + str(time.time() - start_time))\n print(est_ppswor)\n\n print(\"Estimating using ell_2 PPSWOR:\")\n est_ppswor_l2 = []\n for i in range(num_iter):\n cur = []\n for x in real_count:\n cur.append(ppswor_estimate_moment(x.counts.items(), k, p, 2))\n est_ppswor_l2.append(cur)\n print(\"Done: \" + str(time.time() - start_time))\n print(est_ppswor_l2)","def evaluate_all(experiments,\n results,\n seed,\n num_pairs,\n samples_per_pair,\n loop_size=None):\n eval_results = evaluate_baselines(\n experiments[0], seed, num_pairs, samples_per_pair, loop_size=None)\n for ex, res in zip(experiments, results):\n eval_results.update(\n evaluate_experiment(ex, res, seed, num_pairs, samples_per_pair,\n loop_size))\n return eval_results","def test_compare_classifier_performance_from_pred_vis_api(experiment_to_use):\n experiment = experiment_to_use\n prediction = experiment.predictions\n viz_outputs = (\"pdf\", \"png\")\n with TemporaryDirectory() as tmpvizdir:\n for viz_output in viz_outputs:\n vis_output_pattern_pdf = tmpvizdir + f\"/*.{viz_output}\"\n visualize.compare_classifiers_performance_from_pred(\n [prediction, prediction],\n experiment.ground_truth,\n experiment.ground_truth_metadata,\n experiment.output_feature_name,\n labels_limit=0,\n model_namess=[\"Model1\", \"Model2\"],\n output_directory=tmpvizdir,\n file_format=viz_output,\n )\n figure_cnt = glob.glob(vis_output_pattern_pdf)\n assert 1 == len(figure_cnt)","def run_all_tests(self) -> None:\n self.run_trt_precision_tests()\n logging.info(\"Check analysis result at: %s\", self._output_dir)","def run_oracle_experiments(model, query_iterator, N, recalls=[1]):\n times_dict = dict()\n calls_dict = dict()\n probs = dict()\n keys = list()\n recall_break = max(recalls)\n for query, relevant_set in query_iterator:\n candidates_iterator = iter_candidates(model, query)\n times, calls = candidates_time_oracle_calls_count(\n candidates_iterator, relevant_set, recall_break)\n k = len(relevant_set)\n if k in times_dict:\n times_dict[k].append(times)\n calls_dict[k].append(calls)\n probs[k] += 1\n else:\n keys.append(k)\n times_dict[k] = [times]\n calls_dict[k] = [calls]\n probs[k] = 1\n \n keys = np.array(keys)\n probs = np.array([probs[k] for k in keys])/sum(probs.values())\n times_dict = {k:np.array(v).mean(axis=0) for k,v in times_dict.items()}\n calls_dict = {k:np.array(v).mean(axis=0) for k,v in calls_dict.items()}\n \n expected_times = oracle_expectation_by_recalls(recalls, keys, probs, times_dict)\n expected_calls = oracle_expectation_by_recalls(recalls, keys, probs, calls_dict)\n expected_calls_baseline = oracle_random_calls_expectation_by_recall(recalls, keys, probs, N)\n \n return expected_times, expected_calls, expected_calls_baseline","def run_tests(augmenter, images, nb_runs):\n results = np.zeros((nb_runs,))\n for i in range(nb_runs):\n start = time.time()\n augmenter.augment_batch(images)\n results[i] = time.time() - start\n print(\"Run %d: %.8fs\" % (i, results[i]))\n print(\"Mean: %.8fs\" % (results.mean(),))\n print(\"Sum: %.8fs\" % (results.sum(),))","def test_compare_classifiers_performance_changing_k_vis_api(experiment_to_use):\n experiment = experiment_to_use\n probabilities = experiment.probabilities\n viz_outputs = (\"pdf\", \"png\")\n with TemporaryDirectory() as tmpvizdir:\n for viz_output in viz_outputs:\n vis_output_pattern_pdf = tmpvizdir + f\"/*.{viz_output}\"\n visualize.compare_classifiers_performance_changing_k(\n [probabilities, probabilities],\n experiment.ground_truth,\n experiment.ground_truth_metadata,\n experiment.output_feature_name,\n top_k=3,\n labels_limit=0,\n model_namess=[\"Model1\", \"Model2\"],\n output_directory=tmpvizdir,\n file_format=viz_output,\n )\n figure_cnt = glob.glob(vis_output_pattern_pdf)\n assert 1 == len(figure_cnt)","def run_experiments() :\n #%%\n target_size=(32,32)\n g_specs = {\n \"batch_size\" : [ 30 , 60, 100 ],\n \"learning_rate\" : [ 0.0002, 0.0003, 0.0005 ],\n \"drop_out_rate\" : [ 0.2, 0.25, 0.3 ],\n \"rescale_mode\" : [ \"max_q\" , \"max\", \"\" ]\n }\n\n model_traits = MODEL_TRAITS[\"model2\"].copy()\n tt_obj = model_traits[\"trainer_tester_class\"]( model_traits )\n del model_traits[\"trainer_tester_class\"]\n\n cnt = 0\n for batchs, lrate, do_rate, resc_mode in product( g_specs[\"batch_size\"],\n g_specs[\"learning_rate\"],\n g_specs[\"drop_out_rate\"],\n g_specs[\"rescale_mode\"] ) :\n\n tt_obj.model_traits.update( {\"batch_size\" : batchs,\n \"learning_rate\" : lrate,\n \"rescale_mode\" : resc_mode,\n \"drop_out_rate\" : do_rate } )\n\n train_4d, train_gt = tu.make_4d_arrays( images_dir=\"images/train\",\n target_size=target_size )\n\n test_4d, test_gt = tu.make_4d_arrays( images_dir=\"images/test\",\n target_size=target_size )\n\n data = {\"train_4d\" : train_4d,\n \"test_4d\" : test_4d,\n \"train_y\" : train_gt,\n \"test_y\" : test_gt}\n\n valid_accu_log, train_accu_log = tt_obj.train( model_traits, data,\n logl=100 )\n idx_v = int(np.argmax( valid_accu_log))\n idx_t = int(np.argmax( train_accu_log))\n\n model_traits.update({\"valid_accu_log\" : valid_accu_log,\n \"train_accu_log\" : train_accu_log,\n \"best_valid\" : max(valid_accu_log),\n \"best_valid_at\" : idx_v,\n \"train_at_best_valid\" : train_accu_log[idx_v],\n \"best_train\" : max(train_accu_log),\n \"best_train_at\": idx_t })\n\n #print(cnt, pformat(model_traits) )\n print( \"%d : best_train = %.4f, best_valid = %.4f\" % \\\n (cnt, max(train_accu_log), max(valid_accu_log) ))\n\n with open( \"exp_results_%d.json\" % cnt,\n \"wt\" , encoding=\"utf8\" ) as f_out :\n print( json.dumps( model_traits ), file=f_out)\n\n\n cnt += 1\n #%%","def test_any(alg, repNum=1):\n\tperformance = alg.predict_perf(repNum=repNum)\n\tprint performance\n\talg.plot()\n\n\treturn performance","def test_compare_classifiers_performance_subset_vis_api(experiment_to_use):\n experiment = experiment_to_use\n probabilities = experiment.probabilities\n viz_outputs = (\"pdf\", \"png\")\n with TemporaryDirectory() as tmpvizdir:\n for viz_output in viz_outputs:\n vis_output_pattern_pdf = tmpvizdir + f\"/*.{viz_output}\"\n visualize.compare_classifiers_performance_subset(\n [probabilities, probabilities],\n experiment.ground_truth,\n experiment.ground_truth_metadata,\n experiment.output_feature_name,\n top_n_classes=[6],\n labels_limit=0,\n subset=\"ground_truth\",\n model_namess=[\"Model1\", \"Model2\"],\n output_directory=tmpvizdir,\n file_format=viz_output,\n )\n figure_cnt = glob.glob(vis_output_pattern_pdf)\n assert 1 == len(figure_cnt)","def compute_and_visualize_coupling_experiments(experiments, results, seed,\n **kwargs):\n (logits_1,\n logits_2), couplings = get_coupling_estimates(experiments, results, seed,\n **kwargs)\n test_losses = compute_coupling_losses(experiments, logits_1, logits_2,\n couplings)\n visualize_coupling_experiments(test_losses, logits_1, logits_2, couplings)","def test_compare_classifier_performance_from_prob_vis_api(experiment_to_use):\n experiment = experiment_to_use\n probability = experiment.probabilities\n viz_outputs = (\"pdf\", \"png\")\n with TemporaryDirectory() as tmpvizdir:\n for viz_output in viz_outputs:\n vis_output_pattern_pdf = tmpvizdir + f\"/*.{viz_output}\"\n visualize.compare_classifiers_performance_from_prob(\n [probability, probability],\n experiment.ground_truth,\n experiment.ground_truth_metadata,\n experiment.output_feature_name,\n top_n_classes=[0],\n labels_limit=0,\n model_namess=[\"Model1\", \"Model2\"],\n output_directory=tmpvizdir,\n file_format=viz_output,\n )\n figure_cnt = glob.glob(vis_output_pattern_pdf)\n assert 1 == len(figure_cnt)","def run():\n trials = 100\n\n multipliers = [0.25, 0.3, 0.35, 0.5, 0.75, 1, 1.25, 1.45, 1.5, 1.55, 1.6] # Coefficients for learning rate\n\n mean_penalty = []\n median_penalty = []\n std_penalty = []\n\n mean_trial_time = []\n median_trial_time = []\n std_trial_time = []\n\n mean_success_rate = []\n median_success_rate = []\n std_success_rate = []\n\n for m in multipliers:\n all_penalties = [] # All penalties from trail sets\n all_average_trial_time = []\n all_success_rates = []\n\n for i in range(0, 20):\n # print \"Trial set:\", i\n # Set up environment and agent\n e = Environment() # create environment (also adds some dummy traffic)\n agent = e.create_agent(LearnerAgent) # create agent\n agent.mult = m\n e.set_primary_agent(agent, enforce_deadline=True) # specify agent to track\n\n # Now simulate it\n sim = Simulator(e, update_delay=0.0, display=False) # create simulator (uses pygame when display=True, if available)\n\n sim.run(n_trials=trials) # run for a specified number of trials\n\n all_penalties.append(agent.all_trails_penalties)\n all_average_trial_time.append(agent.time/float(trials))\n all_success_rates.append(float(trials-agent.aborted_trials)/trials)\n\n mean_penalty.append(np.mean(all_penalties))\n median_penalty.append(np.median(all_penalties))\n std_penalty.append(np.std(all_penalties))\n\n mean_trial_time.append(np.mean(all_average_trial_time))\n median_trial_time.append(np.median(all_average_trial_time))\n std_trial_time.append(np.std(all_average_trial_time))\n\n mean_success_rate.append(np.mean(all_success_rates))\n median_success_rate.append(np.median(all_success_rates))\n std_success_rate.append(np.std(all_success_rates))\n\n for i in range(0, len(multipliers)):\n print \"\"\n print \"Multiplier:\", multipliers[i]\n print \"\"\n print \"Mean penalty per {} trials:\".format(trials), mean_penalty[i]\n print \"Median penalty per {} trials:\".format(trials), median_penalty[i]\n print \"Std.Dev. penalty per {} trials:\".format(trials), std_penalty[i]\n\n print \"\"\n print \"Mean trial time:\", mean_trial_time[i]\n print \"Median trial time:\", median_trial_time[i]\n print \"Std.Dev. trial time:\", std_trial_time[i]\n\n print \"\"\n print \"Mean success rate per {} trials:\".format(trials), mean_success_rate[i]\n print \"Median success rate per {} trials:\".format(trials), median_success_rate[i]\n print \"Std.Dev. success rate per {} trials:\".format(trials), std_success_rate[i]","def test_run(id_, args, kwargs, expected):\n # Remove results for test.\n results_path = expected['results_path']\n shutil.rmtree(results_path, ignore_errors=True)\n # Start code.\n # [future] attempts to run classification with n_jobs>1\n # global cnfg_default\n # sys.modules['cnfg_default'] = get_params(args[0], 'cnfg_default')\n # import cnfg_default\n # #from cnfg_default import custom_score_metric\n objects = pycnfg.run(oid='default', *args, **kwargs)\n tmp = {k: type(v).__name__ for k, v in objects.items()}\n print('OBJECTS:')\n print(tmp)\n # Compare results:\n # * Compare objects (keys and str of values).\n objects_ = expected['objects']\n objects = {k: type(v).__name__ for k, v in objects.items()}\n if objects != objects_:\n print(set(objects.items()) ^ set(objects_.items()))\n assert objects == objects_\n # for k, v in objects.items():\n # assert k in objects_\n # assert type(v).__name__ == objects_[k]\n # * Compare predictions csv(all available).\n pred_path = glob.glob(f\"{results_path}/models/*_pred.csv\")\n pred_path_ = glob.glob(expected['pred_path'])\n assert len(pred_path) == len(pred_path_)\n for act, exp in zip(sorted(pred_path), sorted(pred_path_)):\n file_diff(act, exp)\n assert filecmp.cmp(act, exp)\n # * Compare test logs.\n logs_path = glob.glob(f\"{results_path}/logs*/*_test.log\")[0]\n logs_path_ = expected['logs_path']\n file_diff(logs_path, logs_path_)\n assert filecmp.cmp(logs_path, logs_path_)\n # * Compare runs dataframe, non-universe columns.\n runs_path = f\"{results_path}/runs\"\n runs_path_ = expected['runs_path']\n df = runs_loader(runs_path)\n df_ = runs_loader(runs_path_)\n # First False/True for each element, then check all by columns.\n # col1 True\n # col2 False\n # dtype: bool\n df_diff = df.eq(df_).all()\n # Column names that are not equal.\n columns = sorted(list(df_diff[df_diff==False].dropna().index))\n # columns_eq = sorted(list(df_diff[df_diff==True].dropna().index))\n columns_ = expected['columns_diff']\n print('DIFF:\\n', columns)\n time.sleep(1)\n # assert columns == columns_\n # * Compare model.\n model_path = glob.glob(f\"{results_path}/models/*.model\")\n model_path_ = glob.glob(expected['model_path'])\n assert len(model_path) == len(model_path_)\n for act, exp in zip(sorted(model_path), sorted(model_path_)):\n assert filecmp.cmp(act, exp)\n return","def run_tests():\n with open(FILENAME) as file:\n # Loads testing parameters from the yaml file.\n tests = yaml.safe_load(file)\n\n # create a dataframe to keep the results\n test_dict = tests['Tests']\n results = pd.DataFrame(test_dict)\n results['Last Average Score'] = \"\"\n results['No of Q-Learning episodes'] = \"\"\n\n # run experiments:\n for i, test in enumerate(test_dict):\n grid = Rooms(test[\"env_size\"], testing=True)\n learning = QLearning(grid, test[\"gamma\"], test[\"alpha\"], test[\"agent_start_pos\"])\n e_greedy = Policy(\"e-greedy\", test[\"epsilon\"], test[\"decay\"])\n greedy = Policy(policy_type=\"greedy\")\n experiment = Experiments(grid, learning, greedy, test[\"iters\"],\n test[\"agent_start_pos\"], test[\"test_no\"])\n\n for session in range(test[\"iters\"]):\n learning.run_multiple_episodes(test[\"batch_episodes\"], e_greedy)\n mean_reward = experiment.run_experiments(test[\"exp_per_batch\"])\n\n results.loc[i,'Last Average Score'] = mean_reward\n results.loc[i,'No of Q-Learning episodes'] = (session + 1) * test[\"batch_episodes\"]\n\n # save results to csv file\n filename = 'results/' + 'test_table.csv'\n results.to_csv(filename)\n\n # plot & save graphs\n experiment.generate_results(test[\"test_no\"], test)\n\n return results","def evaluate_individual(predictions, test_files, models):\n\n print(\"\\nAccuracy for individual models\\n\")\n \n # Fix Location\n correct_predictions = [0, 0, 0]\n total_predictions = [0, 0, 0]\n num_failed_predictions = 0\n\n for prediction in predictions:\n if prediction[\"correct_data\"][\"correct_location\"] == prediction[\"predicted_location\"]:\n correct_predictions[FixType[prediction[\"correct_data\"][\"correct_type\"]].value] = correct_predictions[FixType[\n prediction[\"correct_data\"][\"correct_type\"]].value] + 1\n if prediction[\"predicted_location\"] is None:\n num_failed_predictions = num_failed_predictions + 1\n total_predictions[FixType[prediction[\"correct_data\"][\"correct_type\"]].value] = total_predictions[FixType[\n prediction[\"correct_data\"][\"correct_type\"]].value] + 1\n\n for i in range(3):\n if total_predictions[i] == 0: # If the type was never predicted\n accuracy = 0\n else:\n accuracy = correct_predictions[i] / total_predictions[i]\n print(f\"Fix Location accuracy for class {FixType(i).name}: {accuracy * 100} %\")\n\n accuracy = sum(correct_predictions) / (len(predictions) - num_failed_predictions)\n print(f\"Fix Location accuracy overall is {accuracy * 100} %\")\n \n # Fix type\n correct_predictions = [0, 0, 0]\n total_predictions = [0, 0, 0]\n num_failed_predictions = 0\n\n for prediction in predictions:\n if prediction[\"correct_data\"][\"correct_type\"] == prediction[\"predicted_type\"]:\n correct_predictions[FixType[prediction[\"predicted_type\"]].value] = correct_predictions[FixType[\n prediction[\"predicted_type\"]].value] + 1\n if prediction[\"predicted_type\"] is None:\n num_failed_predictions = num_failed_predictions + 1\n total_predictions[FixType[prediction[\"predicted_type\"]].value] = total_predictions[FixType[\n prediction[\"predicted_type\"]].value] + 1\n\n for i in range(3):\n if total_predictions[i] == 0: # If the type was never predicted\n accuracy = 0\n else:\n accuracy = correct_predictions[i] / total_predictions[i]\n print(f\"Fix Type accuracy for class {FixType(i).name}: {accuracy * 100} %\")\n\n accuracy = sum(correct_predictions) / (len(predictions) - num_failed_predictions)\n print(f\"Fix Type accuracy overall is {accuracy * 100} %\")\n \n # We repeat the predictions to evaluate the insert and modify models individually, regardless of the predicted fix type \n\n raw_training_samples = []\n\n if test_files.endswith(\".json\"): # Single JSON file\n with open(test_files) as file:\n logging.info(\"Source ending in .json. Predicting on single JSON file.\")\n raw_training_samples = json.load(file)\n else: # Folder path\n for filename in listdir(test_files):\n with open(test_files + filename) as file:\n raw_training_samples.extend(json.load(file))\n \n correct_predictions_insert = 0\n total_predictions_insert = 0\n correct_predictions_modify = 0\n total_predictions_modify = 0\n insert_tokens = []\n modify_tokens = []\n\n for sample in raw_training_samples:\n # Insert\n if sample[\"metadata\"][\"fix_type\"] == \"insert\":\n actual_sample, tokens = IOProcessor.preprocess(sample[\"wrong_code\"])\n pred = predict_single(actual_sample, models[2])\n token = IOProcessor.postprocess(pred, 2)\n if token == sample[\"metadata\"][\"fix_token\"]: # Correct Prediction\n correct_predictions_insert = correct_predictions_insert + 1\n else: # Incorrect prediction\n insert_tokens.append([token, sample[\"metadata\"][\"fix_token\"]])\n total_predictions_insert = total_predictions_insert + 1\n # Modify\n if sample[\"metadata\"][\"fix_type\"] == \"modify\":\n actual_sample, tokens = IOProcessor.preprocess(sample[\"wrong_code\"])\n pred = predict_single(actual_sample, models[3])\n token = IOProcessor.postprocess(pred, 3)\n if token == sample[\"metadata\"][\"fix_token\"]: # Correct Prediction\n correct_predictions_modify = correct_predictions_modify + 1\n else: # Incorrect prediction\n modify_tokens.append([token, sample[\"metadata\"][\"fix_token\"]])\n total_predictions_modify = total_predictions_modify + 1\n\n insert_accuracy = correct_predictions_insert / total_predictions_insert\n modify_accuracy = correct_predictions_modify / total_predictions_modify\n print(f\"Fix Token accuracy for insert is {insert_accuracy * 100} %\")\n print(f\"Fix Token accuracy for modify is {modify_accuracy * 100} %\")\n\n # The following code may be used to create a swarm plot of the erroneous predictions for fix locations\n # This does, however, require the installation of the pandas, seaborn, and matplotlib libraries.\n \n # import seaborn as sns\n # import matplotlib.pyplot as plt\n # import pandas as pd\n # location_distance_array = []\n # for prediction in predictions:\n # actual_sample, tokens = IOProcessor.preprocess(prediction[\"correct_data\"][\"wrong_code\"])\n # label = get_token_index(prediction[\"correct_data\"][\"wrong_code\"], tokens, prediction[\"correct_data\"][\"correct_location\"])\n # if prediction[\"predicted_token_location\"] - label == 0:\n # pass\n # else:\n # location_distance_array.append([prediction[\"predicted_token_location\"] - label, prediction[\"correct_data\"][\"correct_type\"]])\n \n # df = pd.DataFrame(data=location_distance_array)\n # sns.set_theme(style=\"whitegrid\")\n # f, ax = plt.subplots(figsize=(6, 4))\n # sns.despine(bottom=True, left=True)\n # sns.swarmplot(y=0, x=1, data=df, palette=\"dark\", size=6)\n # ax.set_xlabel('')\n # ax.set_ylabel('')\n # plt.ylim([-15, 16])\n \n # plt.savefig('line_plot.pdf', bbox_inches='tight', pad_inches=0)","def main():\n\tresults = []\n\n\tconfig = configparser.ConfigParser()\n\tconfig.read(\"simulation.ini\")\n\tsettings = config['sim']\n\n\tcompleted_obj_hw = int(settings[\"ClientsPerCampaign\"]) * float(settings[\"CompletedPctgHW\"])\n\texceeded_obj_hw = float(settings[\"ExceededPctgHW\"])\n\tsignificance_level = float(settings[\"SignificanceLevel\"])\n\tz_val_two_tails = scipy.stats.norm.ppf(1 - (significance_level / 2))\n\n\tprint(\"Completed Target HW: \" + str(completed_obj_hw))\n\tprint(\"Exceeded Target HW: \" + str(exceeded_obj_hw))\n\n\tcompleted_vals = []\n\texceeded_vals = []\n\tdone = False\n\n\tcompleted_avg = 0\n\texceeded_avg = 0\n\tcompleted_hw = 0\n\texceeded_hw = 0\n\n\ti = 0\n\twhile not done:\n\t\tprint(\"RUN: \" + str(i + 1))\n\t\tenv = simpy.Environment()\n\t\tsim = Simulation(env, settings, i == 0)\n\t\tsim.run()\n\t\tresults.append(sim.results)\n\t\ti += 1\n\n\t\tif settings['RunOnce'] == 'yes':\n\t\t\tprint(\"RUN ONCE\")\n\t\t\tsys.exit()\n\n\t\tcompleted_vals.append(sim.results['completed_count'])\n\t\texceeded_vals.append(sim.results['exceeded_proportion'])\n\n\t\tif i < 2:\n\t\t\tprint(\"---------------\")\n\t\t\tcontinue\n\n\t\tcompleted_avg = sum(completed_vals) / len(completed_vals)\n\t\tcompleted_S = sum([(v - completed_avg) ** 2 for v in completed_vals]) / (i - 1)\n\t\tcompleted_S = math.sqrt(completed_S)\n\t\tcompleted_hw = (z_val_two_tails * completed_S) / math.sqrt(i)\n\t\tprint(\"runs: \" + str(i) + \" completed HW: \" + str(completed_hw))\n\n\t\texceeded_avg = sum(exceeded_vals) / len(exceeded_vals)\n\t\texceeded_S = math.sqrt(exceeded_avg * (1 - exceeded_avg))\n\t\texceeded_hw = (z_val_two_tails * exceeded_S) / math.sqrt(i)\n\t\tprint(\"runs: \" + str(i) + \" exceeded HW: \" + str(exceeded_hw))\n\n\t\tif completed_hw < completed_obj_hw and exceeded_hw < exceeded_obj_hw:\n\t\t\tprint(\"END ITERATIONS\")\n\t\t\tdone = True\n\n\t\tprint(\"---------------\")\n\n\n\tfilename = 'results/Results_' + settings['FileSizeGB'] + '_' + settings['TorrentThreshold'] + '_' + settings['HTTPDownThreshold'] \\\n\t\t+ '_' + settings['HTTPUp'] + '_' + str(random.randint(0,10000)) + '.xlsx'\n\n\tprint(\"Saving XLSX to: \" + filename)\n\twb = xs.Workbook(filename)\n\n\tws = wb.add_worksheet()\n\n\tws.write(0, 1, 'Exceded')\n\tws.write(0, 2, 'Completed')\n\n\ti = 1\n\tfor result in results:\n\t\tws.write(i, 0, i)\n\t\tws.write(i, 1, result['exceeded_proportion'])\n\t\tws.write(i, 2, result['completed_count'])\n\t\ti += 1\n\n\tws.write(i, 0, 'average')\n\tws.write(i, 1, exceeded_avg)\n\tws.write(i, 2, completed_avg)\n\ti += 1\n\tws.write(i, 0, 'half width')\n\tws.write(i, 1, exceeded_hw)\n\tws.write(i, 2, completed_hw)\n\n\twb.close()","def evaluate_benchmarks(self):\n\n # iterate over replicates\n results = {}\n for replicate_id, replicate in self.replicates:\n\n # evaluate benchmark for current replicate\n bmark = SimulationBenchmark(replicate.copy(),\n graph=self.graphs[replicate_id],\n **self.params)\n\n # store results\n results[replicate_id] = dict(\n\n labels_MAE=bmark.scores['labels'].MAE,\n level_only_MAE=bmark.scores['level_only'].MAE,\n spatial_only_MAE=bmark.scores['spatial_only'].MAE,\n community_MAE=bmark.scores['labels_comm'].MAE,\n\n labels_PCT=bmark.scores['labels'].percent_correct,\n level_only_PCT=bmark.scores['level_only'].percent_correct,\n spatial_only_PCT=bmark.scores['spatial_only'].percent_correct,\n community_PCT=bmark.scores['labels_comm'].percent_correct)\n\n # compile dataframe\n results = pd.DataFrame.from_dict(results, orient='index')\n results.index.set_names(self.multiindex, inplace=True)\n\n return results","def run_regression_experiments(self):\n assert self.patientIDs # assert not empty\n self.log.info(\"Running experiments for {} patients\".format(len(self.patientIDs)))\n\n results_batch = defaultdict(dict)\n results_single = dict()\n for patient in self.patientIDs:\n patient_experiment = PatientPredictionExperiment(\n patient_id=patient, algorithm=self.algorithm,\n is_batch=self.isBatch\n )\n # run results for single patients\n if not self.isBatch:\n results_single[patient] = patient_experiment.run_experiment()\n self.log.debug(\"Start processing single results: {}\".format(results_single[patient]))\n # super verbose debug information should be logged with debug option\n else:\n # TODO: think about consistent interfaces\n batch_results, self._allFeatureDesp = patient_experiment.run_experiment()\n self.log.info(\"Start processing batch results with {} batches\".format(len(batch_results)))\n batchNo = 0\n for results in batch_results:\n # here are the result for each feature subset for each patients\n results_batch[batchNo][patient] = results\n batchNo += 1\n if self.algorithm == \"fm\":\n return\n if not self.isBatch:\n results_overall = self.process_results(results_single)\n self.store_results(results_overall, results_single)\n self.log.info(\"Finished running experiment\")\n else:\n self.record_batch_experiment()\n for batchNo, patient_results in results_batch.iteritems():\n # patient_results = results_single\n results_overall = self.process_results(patient_results)\n self.store_results(results_overall, patient_results, self._allFeatureDesp[batchNo])\n self.log.info(\"Finished running experiment No. {}\".format(batchNo))","def one_experiment():\n\n # set the name of the experiment\n now = datetime.datetime.now()\n experiment_id = str(now.day) + \"_\" + str(now.month) + \"_\" + str(now.hour) + \".\" + str(now.minute)\n experiment_name = 'overfit_' + str(experiment_id)\n\n # define if you want to use preprocessed data from file\n use_prep_data = False\n if use_prep_data:\n set_params(preproc_data_id='16_5_10.16.47')\n\n # define the changing parameter and its value\n changing_param_name = 'class_weights'\n changing_param_value = [{0: 15, 1: 85}]\n # {0:15, 1:85}]#, {0:4, 1:100}, {0:3, 1:100}, {0:2, 1:100}, {0:1, 1:100}] #[{0:1, 1:1}, {0:15, 1:85}]#\n\n features_to_use = ['user', 'countries', 'session', 'format', 'token']\n # set constant parameters\n set_params(use_word_emb=1)\n set_params(epochs=40)\n set_params(features_to_use=features_to_use)\n\n # save constant parameters to a new \"experiment_..\" filgithx+P@2ub\n save_constant_parameters(experiment_name, changing_param_name)\n\n # run experiment for every parameter value\n for value in changing_param_value:\n process = psutil.Process(os.getpid())\n print(\"-----MEMORY before starting experiment ------\", int(process.memory_info().rss/(8*10**3)), \"KB\")\n\n # update the parameter value\n set_params(class_weights_1=value)\n\n # update the model_id for this new model\n now = datetime.datetime.now()\n new_model_id = str(now.day) + \"_\" + str(now.month) + \"_\" + str(now.hour) + \".\" + str(now.minute) + \".\" + str(now.second)\n\n set_params(model_id=new_model_id)\n\n # evaluate the new model and save the results in the experiment file\n oneExperiment = Process(target=run_experiment, args=(experiment_name,\n new_model_id, changing_param_name, value,))\n oneExperiment.start()\n oneExperiment.join()","def test_compare_performance_vis_api(experiment_to_use):\n experiment = experiment_to_use\n # extract test stats only\n test_stats = experiment.test_stats_full\n viz_outputs = (\"pdf\", \"png\")\n with TemporaryDirectory() as tmpvizdir:\n for viz_output in viz_outputs:\n vis_output_pattern_pdf = tmpvizdir + f\"/*.{viz_output}\"\n visualize.compare_performance(\n [test_stats, test_stats],\n output_feature_name=None,\n model_names=[\"Model1\", \"Model2\"],\n output_directory=tmpvizdir,\n file_format=viz_output,\n )\n figure_cnt = glob.glob(vis_output_pattern_pdf)\n assert 1 == len(figure_cnt)","def emPerformance(filesAndDirectories='None', resultsFileName='None', iterationCount='3', modes='None', testTypes='None', viewports='None', verbose='False'):\n\n pass","def _evaluate_performance__static_winners(self):\n # | - _evaluate_performance__\n\n # | - class attributes #################################################\n AL = self\n al_gen = self.al_gen\n verbose = self.verbose\n seed_ids = self.seed_ids\n acquisition_bin = self.acquisition_bin\n completed_ids = self.completed_ids\n CandidateSpace = self.CandidateSpace\n RegressionModel = self.RegressionModel\n DuplicateFinder = self.DuplicateFinder\n al_gen_dict = self.al_gen_dict\n\n stop_mode = self.stop_mode\n stop_num_generations = self.stop_num_generations\n\n index_acq_gen_dict = self.index_acq_gen_dict\n #__| #################################################################\n\n # #####################################################################\n mode = \"lowest_N\" # 'lowest_N' or 'lowest_perc'\n\n N_ids = 10\n lowest_perc = 5\n\n # Number of consecutive generations that the Nth best systems must\n # remain static\n M_gens = 3\n # #####################################################################\n\n if mode == \"lowest_perc\":\n num_candidates = CandidateSpace.FingerPrints.df_pre.shape[0]\n N_ids = int(num_candidates * (lowest_perc * 0.01))\n\n gen_keys = list(AL.al_gen_dict.keys())\n\n if len(gen_keys) > M_gens:\n latest_M_keys = gen_keys[-(M_gens + 1):]\n last_gen_key = gen_keys[-1]\n\n al_gen_dict_subset_i = dict(zip(\n latest_M_keys,\n [AL.al_gen_dict.get(i, None) for i in latest_M_keys]))\n\n indices_list = []\n iterator = enumerate(al_gen_dict_subset_i.items())\n for i_cnt, (gen_i, AL_i) in iterator:\n model_i = AL_i.model\n\n model_i = AL.add_main_Y_to_model(\n model_i, plot_dft_instead_of_pred=True)\n model_i = model_i[(model_i[\"duplicate\"] == False)]\n model_i = model_i.sort_values(\"Y_main\")\n\n indices_i = model_i.index.tolist()\n\n indices_list.append(indices_i)\n\n if i_cnt >= M_gens:\n indices_i = indices_list[i_cnt][0:N_ids]\n ids_static_list = []\n for j in range(M_gens):\n indices_j = indices_list[i_cnt - (j + 1)][0:N_ids]\n ids_static = indices_j == indices_i\n ids_static_list.append(ids_static)\n\n ids_are_static = all(ids_static_list)\n\n self.performance__static_winners[last_gen_key] = ids_are_static\n #__|","def print_results(precision, recall, f1):\n print \"Precision:\", precision\n print \"Recall:\", recall\n print \"F1 score:\", f1","def evaluate(self, runs=100):\n score_record = []\n \n print('Evaluation in progress...')\n for i in range(runs):\n score = self.run_evaluation_episode()\n score_record.append(score)\n \n ave_score = np.mean(score_record)\n \n print('System evaluated with an average score of {} in {} runs'.format(ave_score, runs))","def runTests():\r\n\r\n print(\"running a few tests\")\r\n\r\n average = compute .gpsAverage (4, 5)\r\n print(\"average = \", average)\r\n \r\n print (\"hello!\")","def main():\n parser = argparse.ArgumentParser(description=\"Process the results of an experiment.\")\n parser.add_argument(\"experiment\")\n arguments = parser.parse_args()\n path = f\"experiments/{arguments.experiment}\"\n if not os.path.exists(path):\n raise SystemExit(f\"Path {path} does not exists.\")\n\n # For efficiency, one should generate the results from the parts without merging them.\n files = [file for file in os.listdir(path) if os.path.isfile(os.path.join(path, file))]\n frames = []\n for file in files:\n device, experiment, _ = file.split(\".\")\n frame = pandas.read_csv(\n os.path.join(path, file),\n index_col=\"variable\",\n usecols=[\"variable\", \"group_index\", \"value_i\"], dtype={\"value_i\": \"Int64\"}\n )\n frame[\"board\"] = device\n frame[\"experiment\"] = experiment\n frames.append(frame)\n dataframe = pandas.concat(frames)\n frames = None\n\n current_grouping = dataframe.groupby([\"group_index\", \"variable\"])\n \n data = current_grouping.agg([\n numpy.median,\n _percentile_factory(95),\n numpy.mean,\n numpy.std,\n \"count\"\n ])\n\n print(data)\n \n data = data.droplevel([0], axis=1)\n data = data.unstack()\n data.columns = data.columns.map('_'.join)\n data.to_csv(f\"{arguments.experiment}.csv\")","def test_run():\n # Only few steps for test\n timesteps = 128\n\n # Compute all sub testing conf\n envs = ['CartPole-v0']\n ml_platforms = ['torch', 'tf']\n agents = ['dqn', 'a2c']\n\n test_combinations = list(it.product(\n envs,\n ml_platforms,\n agents\n )\n )\n\n # Finally test them all\n for conf in test_combinations:\n env_str, ml_platform_str, agent_str = conf\n run(\n agent_str,\n ml_platform_str,\n env_str,\n 'dense',\n timesteps,\n './target/')","def main():\n for opt in optimizations:\n compile_command = [\"g++\", \"main.cpp\", f\"-O{opt}\", \"-lpthread\"]\n run(compile_command, check=True)\n for threads in num_threads:\n print(f\"{opt=}, {threads=}\", end=\"\")\n stdout.flush()\n test_command = ['./a.out', str(iterations), str(threads)]\n total = 0\n for samples in range(1, repeats_for_average + 1):\n print(\".\", end=\"\")\n stdout.flush()\n output = run(test_command, check=True, capture_output=True).stdout\n total += int(output.split()[-2]) / 1000\n print(f\"\\t{total / samples:.03f}\")","def polling_experiment_multiple(dataset_id, times, **kwargs):\n total_yeahs = 0\n total_noes = 0\n for i in range(0, times):\n yeah, noes = polling_experiment(dataset_id, **kwargs)\n total_yeahs += yeah\n total_noes += noes\n logging.info(\"After \"+str(times)+\" \"+ str(dataset_id) + \" experiments:\")\n logging.info(\"\"+ str(total_yeahs) + \" correct predictions\")\n logging.info(\"\"+ str(total_noes) + \" wrong predictions\")\n logging.info(\"\"+ str(total_yeahs/(total_yeahs + total_noes)) + \" accuracy\")","def _train_and_evaluate(estimator, output_dir):\n \n \"\"\"X_train, y_train =utils._feature_label_split(df_train,\"is_churn\",\"msno\")\n df_val = utils.read_from_bigquery(\"amiable-octane-267022.kkbox.output_val_1\",\"amiable-octane-267022\")\n X_val, y_val =utils._feature_label_split(df_val,\"is_churn\",\"msno\")\"\"\"\n \n df_train=utils.over_sample(\"amiable-octane-267022.kkbox.output_train_1\",\"amiable-octane-267022\")\n X_train, y_train =utils._feature_label_split(df_train,\"is_churn\",\"msno\")\n df_val=utils.over_sample(\"amiable-octane-267022.kkbox.output_val_1\",\"amiable-octane-267022\")\n X_val, y_val =utils._feature_label_split(df_val,\"is_churn\",\"msno\")\n\n estimator.fit(X_train, y_train)\n f1_scorer = make_scorer(f1_score)\n accuracy_scorer =make_scorer(accuracy_score)\n\n if metadata.HYPERPARAMTER_TUNING:\n scores=model_selection.cross_val_score(estimator, X_val, y_val, cv=3,scoring=f1_scorer)\n #,scoring=f1_scorer\n\n logging.info('Score: %s', scores)\n\n #tune hyper\n hpt = hypertune.HyperTune()\n hpt.report_hyperparameter_tuning_metric(\n hyperparameter_metric_tag='F1_SCORE',\n metric_value=np.mean(scores),\n global_step=10000)\n \n#joblib.dump(estimator, 'model.joblib')\n\n # Write model and eval metrics to `output_dir`\n model_output_path = os.path.join(output_dir, 'model',metadata.MODEL_FILE_NAME)\n \n utils.dump_object(estimator, model_output_path)","def batch_simulation(self, iters=10000):\n power_cnt = 0\n correct_sign_cnt = 0\n\n for i in range(iters):\n if (self.verbose) and (i>0) and (i % (iters/10) == 0):\n print(i, \" / \", iters)\n f_stat, p_value, effect_point_estimates = self.simulate()\n power_cnt += (p_value < self.alpha)\n correct_sign_cnt += (effect_point_estimates * self.absolute_effect > 0)\n\n if self.verbose: print(iters, \" / \", iters)\n power = round(power_cnt / float(iters), 5)\n pct_correct_sign = round(correct_sign_cnt / float(iters), 5)\n return power, pct_correct_sign","def test_calculate_all_metrics_precision():\n pass","def exe_tests(self):\n self.rank = mpicom.rank()\n self.size = mpicom.size()\n if mpicom.parallel():\n self.test(\"libname\",os.path.split(mpicom.__file__)[1],\"mpicom.so\")\n else:\n self.test(\"libname\",os.path.split(mpicom.__file__)[1],\"mpistub.pyc\")\n self.test_broadcast()\n self.test_reduce()\n self.test_p2p()\n self.test_gather()\n self.test_scatter()\n #self.test_alltoall()","def compute_and_print_eval_metrics(self):\n s = ('%20s' + '%12s' * 6) % ('Class', 'Images', 'Targets', 'P', 'R', 'mAP@.5', 'mAP@.5:.95')\n precision, recall, f1, mean_precision, mean_recall, map50, map = 0., 0., 0., 0., 0., 0., 0.\n ap = []\n eval_stats = [np.concatenate(x, 0) for x in zip(*self.eval_stats)]\n if len(eval_stats) and eval_stats[0].any():\n precision, recall, ap, f1, ap_class = ap_per_class(*eval_stats)\n precision, recall, ap50, ap = precision[:, 0], recall[:, 0], ap[:, 0], ap.mean(1)\n mean_precision, mean_recall, map50, map = precision.mean(), recall.mean(), ap50.mean(), ap.mean()\n nt = np.bincount(eval_stats[3].astype(np.int64), minlength=len(self.class_names)) # number of targets per class\n else:\n nt = np.zeros(1)\n\n pf = '%20s' + '%12.5g' * 6 # print format\n print(\"\\n EVALUTAION \\n\")\n print(s)\n print(pf % ('all', self.seen, nt.sum(), mean_precision, mean_recall, map50, map))\n if self.cfg.eval.verbose:\n for indx, cls in enumerate(ap_class):\n print(pf % (self.class_names[cls], self.seen, nt[cls], precision[indx], recall[indx], ap50[indx], ap[indx]))","def run_experiments(data_set=\"\",compact=2,exp_name=\"\",x=\"\"):\n cwd = os.getcwd()\n results_path = cwd+'/data/geometric/'+exp_name+'_'\n dataset_path = cwd+'/data/geometric/'+data_set+'_'\n compact = int(compact)\n if \"m\" in x:\n # make new pxy\n pxy, Xdata, groups = gen_geometric_pxy()\n np.save(dataset_path+'Xdata',Xdata)\n np.save(dataset_path+'groups',groups)\n np.save(dataset_path+'pxy',pxy)\n else:\n # load existing pxy\n pxy = np.load(dataset_path+'pxy.npy')\n if \"r\" in x: # regular experiments\n if compact>1:\n metrics_stepwise, distributions_stepwise,\\\n metrics_converged, distributions_converged,\\\n metrics_stepwise_allreps, distributions_stepwise_allreps,\\\n metrics_converged_allreps, distributions_converged_allreps = test_IB(pxy,compact)\n metrics_converged.to_csv(results_path+'metrics_converged.csv')\n metrics_stepwise.to_csv(results_path+'metrics_stepwise.csv')\n distributions_converged.to_pickle(results_path+'distributions_converged.pkl')\n elif compact>0:\n metrics_stepwise,\\\n metrics_converged, distributions_converged,\\\n metrics_stepwise_allreps,\\\n metrics_converged_allreps, distributions_converged_allreps = test_IB(pxy,compact)\n metrics_converged.to_csv(results_path+'metrics_converged.csv')\n metrics_stepwise.to_csv(results_path+'metrics_stepwise.csv')\n distributions_converged.to_pickle(results_path+'distributions_converged.pkl')\n else:\n metrics_stepwise,\\\n metrics_converged,\\\n metrics_stepwise_allreps,\\\n metrics_converged_allreps = test_IB(pxy,compact)\n metrics_converged.to_csv(results_path+'metrics_converged.csv')\n metrics_stepwise.to_csv(results_path+'metrics_stepwise.csv')\n if \"ip\" in x: # initialization experiments - positive p0\n if compact>1:\n metrics_stepwise, distributions_stepwise,\\\n metrics_converged, distributions_converged,\\\n metrics_stepwise_allreps, distributions_stepwise_allreps,\\\n metrics_converged_allreps, distributions_converged_allreps = test_p0_pos(pxy,compact)\n metrics_converged.to_csv(results_path+'metrics_converged_p0_pos.csv')\n metrics_stepwise.to_csv(results_path+'metrics_stepwise_p0_pos.csv')\n distributions_converged.to_pickle(results_path+'distributions_converged_p0_pos.pkl')\n elif compact>0:\n metrics_stepwise,\\\n metrics_converged, distributions_converged,\\\n metrics_stepwise_allreps,\\\n metrics_converged_allreps, distributions_converged_allreps = test_p0_pos(pxy,compact)\n metrics_converged.to_csv(results_path+'metrics_converged_p0_pos.csv')\n metrics_stepwise.to_csv(results_path+'metrics_stepwise_p0_pos.csv')\n distributions_converged.to_pickle(results_path+'distributions_converged_p0_pos.pkl')\n else:\n metrics_stepwise,\\\n metrics_converged,\\\n metrics_stepwise_allreps,\\\n metrics_converged_allreps = test_p0_pos(pxy,compact)\n metrics_converged.to_csv(results_path+'metrics_converged_p0_pos.csv')\n metrics_stepwise.to_csv(results_path+'metrics_stepwise_p0_pos.csv')\n if \"in\" in x: # initialization experiments - negative p0\n if compact>1:\n metrics_stepwise, distributions_stepwise,\\\n metrics_converged, distributions_converged,\\\n metrics_stepwise_allreps, distributions_stepwise_allreps,\\\n metrics_converged_allreps, distributions_converged_allreps = test_p0_neg(pxy,compact)\n metrics_converged.to_csv(results_path+'metrics_converged_p0_neg.csv')\n metrics_stepwise.to_csv(results_path+'metrics_stepwise_p0_neg.csv')\n distributions_converged.to_pickle(results_path+'distributions_converged_p0_neg.pkl')\n elif compact>0:\n metrics_stepwise,\\\n metrics_converged, distributions_converged,\\\n metrics_stepwise_allreps,\\\n metrics_converged_allreps, distributions_converged_allreps = test_p0_neg(pxy,compact)\n metrics_converged.to_csv(results_path+'metrics_converged_p0_neg.csv')\n metrics_stepwise.to_csv(results_path+'metrics_stepwise_p0_neg.csv')\n distributions_converged.to_pickle(results_path+'distributions_converged_p0_neg.pkl')\n else:\n metrics_stepwise,\\\n metrics_converged,\\\n metrics_stepwise_allreps,\\\n metrics_converged_allreps = test_p0_neg(pxy,compact)\n metrics_converged.to_csv(results_path+'metrics_converged_p0_neg.csv')\n metrics_stepwise.to_csv(results_path+'metrics_stepwise_p0_neg.csv')\n if \"c\" in x: # convergence tolerance experiments\n if compact>1:\n metrics_stepwise, distributions_stepwise,\\\n metrics_converged, distributions_converged,\\\n metrics_stepwise_allreps, distributions_stepwise_allreps,\\\n metrics_converged_allreps, distributions_converged_allreps = test_ctol(pxy,compact)\n metrics_converged.to_csv(results_path+'metrics_converged_ctol.csv')\n metrics_stepwise.to_csv(results_path+'metrics_stepwise_ctol.csv')\n distributions_converged.to_pickle(results_path+'distributions_converged_ctol.pkl')\n elif compact>0:\n metrics_stepwise,\\\n metrics_converged, distributions_converged,\\\n metrics_stepwise_allreps,\\\n metrics_converged_allreps, distributions_converged_allreps = test_ctol(pxy,compact)\n metrics_converged.to_csv(results_path+'metrics_converged_ctol.csv')\n metrics_stepwise.to_csv(results_path+'metrics_stepwise_ctol.csv')\n distributions_converged.to_pickle(results_path+'distributions_converged_ctol.pkl')\n else:\n metrics_stepwise,\\\n metrics_converged,\\\n metrics_stepwise_allreps,\\\n metrics_converged_allreps = test_ctol(pxy,compact)\n metrics_converged.to_csv(results_path+'metrics_converged_ctol.csv')\n metrics_stepwise.to_csv(results_path+'metrics_stepwise_ctol.csv')\n if \"z\" in x: # zeroL tolerance experiments\n if compact>1:\n metrics_stepwise, distributions_stepwise,\\\n metrics_converged, distributions_converged,\\\n metrics_stepwise_allreps, distributions_stepwise_allreps,\\\n metrics_converged_allreps, distributions_converged_allreps = test_zeroLtol(pxy,compact)\n metrics_converged.to_csv(results_path+'metrics_converged_zeroLtol.csv')\n metrics_stepwise.to_csv(results_path+'metrics_stepwise_zeroLtol.csv')\n distributions_converged.to_pickle(results_path+'distributions_converged_zeroLtol.pkl')\n elif compact>0:\n metrics_stepwise,\\\n metrics_converged, distributions_converged,\\\n metrics_stepwise_allreps,\\\n metrics_converged_allreps, distributions_converged_allreps = test_zeroLtol(pxy,compact)\n metrics_converged.to_csv(results_path+'metrics_converged_zeroLtol.csv')\n metrics_stepwise.to_csv(results_path+'metrics_stepwise_zeroLtol.csv')\n distributions_converged.to_pickle(results_path+'distributions_converged_zeroLtol.pkl')\n else:\n metrics_stepwise,\\\n metrics_converged,\\\n metrics_stepwise_allreps,\\\n metrics_converged_allreps = test_zeroLtol(pxy,compact)\n metrics_converged.to_csv(results_path+'metrics_converged_zeroLtol.csv')\n metrics_stepwise.to_csv(results_path+'metrics_stepwise_zeroLtol.csv')\n if \"b\" in x: # trying proposed optimal beta\n if compact>1:\n metrics_stepwise, distributions_stepwise,\\\n metrics_converged, distributions_converged,\\\n metrics_stepwise_allreps, distributions_stepwise_allreps,\\\n metrics_converged_allreps, distributions_converged_allreps = test_IB(pxy,compact)\n metrics_converged.to_csv(results_path+'metrics_converged_bestbeta.csv')\n metrics_stepwise.to_csv(results_path+'metrics_stepwise_bestbeta.csv')\n distributions_converged.to_pickle(results_path+'distributions_converged_bestbeta.pkl')\n elif compact>0:\n metrics_stepwise,\\\n metrics_converged, distributions_converged,\\\n metrics_stepwise_allreps,\\\n metrics_converged_allreps, distributions_converged_allreps = test_IB(pxy,compact)\n metrics_converged.to_csv(results_path+'metrics_converged_bestbeta.csv')\n metrics_stepwise.to_csv(results_path+'metrics_stepwise_bestbeta.csv')\n distributions_converged.to_pickle(results_path+'distributions_converged_bestbeta.pkl')\n else:\n metrics_stepwise,\\\n metrics_converged,\\\n metrics_stepwise_allreps,\\\n metrics_converged_allreps = test_IB(pxy,compact)\n metrics_converged.to_csv(results_path+'metrics_converged_bestbeta.csv')\n metrics_stepwise.to_csv(results_path+'metrics_stepwise_bestbeta.csv')\n return 0","def run_tests(): \n \n\n nextdata = [[21, 61, 42, 30], [33,45, 18, 29]]\n\n for xval, yval, snum, expect in nextdata:\n\n pmachine = PMachine()\n pmachine.serial_number = snum\n pmachine.run2_completion()\n result = pmachine.calc_square_total(xval, yval, showsquare=True)\n assert result == expect\n print(\"Got value {}={} as expected\".format(result, expect))","def run_test(models, data):\n num_correct = 0.0\n num_total = 0.0\n\n # initialize lists to put in predictions & results for confusion matrix \n predicted_labels = []\n actual_labels = []\n langlist = models.keys()\n for ai, actual_lang in enumerate(langlist):\n test_files = open(os.path.join('traintestsplit', actual_lang+'.testlist')).read().split()\n print 'Testing', len(test_files), 'files from', actual_lang\n for filename in test_files:\n logprobs = {} # dict: total log prob of this file under each model \n for test_lang in langlist:\n logprobs[test_lang] = apply_model(models[test_lang], data[actual_lang][filename+'.npytxt'])\n predicted_lang = max(logprobs.items(), key=lambda x:x[1])[0]\n # insert prediction (of lang index) into predicted list \n predicted_labels.append(langlist.index(predicted_lang))\n actual_labels.append(ai)\n if actual_lang == predicted_lang:\n num_correct += 1\n num_total += 1\n\n print len(filter(lambda x:x==ai, predicted_labels[-len(test_files):])), 'correct'\n\n print\n print 'Accuracy', num_correct*100/num_total\n\n #CONFUSION MATRIX (y_test, y_pred) -> (actual label, predictions) \n cm = confusion_matrix(actual_labels, predicted_labels)\n cm_normalized = cm.astype('float') / cm.sum(axis=1)[:, np.newaxis]\n # display confusion stats by lang (TODO: visualize with matplotlib) \n print '*'*20\n for ai, actual_lang in enumerate(langlist):\n print actual_lang, 'confusion:'\n for pi, predicted_lang in enumerate(langlist):\n print '{0}: {1:.2f}%'.format(predicted_lang, cm_normalized[ai, pi]*100)\n print '*'*20","def _run_evaluation(\n sess, experiment, eval_config, output_dir, min_range, max_range, num_bins,\n torsion_bins):\n tf.io.gfile.makedirs(os.path.join(output_dir, 'pickle_files'))\n\n logging.info('Eval config is %s\\nnum_bins: %d', eval_config, num_bins)\n num_examples = 0\n num_crops = 0\n start_all_time = time.time()\n\n # Either do the whole test set, or up to a specified limit.\n max_examples = experiment.num_eval_examples\n if eval_config.max_num_examples > 0:\n max_examples = min(max_examples, eval_config.max_num_examples)\n\n while num_examples < max_examples:\n one_prediction = compute_one_prediction(\n num_examples, experiment, sess, eval_config, num_bins, torsion_bins)\n\n single_message = one_prediction.single_message\n num_crops_local = one_prediction.num_crops_local\n sequence = one_prediction.sequence\n filebase = one_prediction.filebase\n softmax_probs = one_prediction.softmax_probs\n ss = one_prediction.ss\n asa = one_prediction.asa\n torsions = one_prediction.torsions\n\n num_examples += 1\n num_crops += num_crops_local\n\n # Save the output files.\n filename = os.path.join(output_dir,\n 'pickle_files', '%s.pickle' % filebase)\n distogram_io.save_distance_histogram(\n filename, softmax_probs, filebase, sequence,\n min_range=min_range, max_range=max_range, num_bins=num_bins)\n\n if experiment.model.torsion_multiplier > 0:\n torsions_dir = os.path.join(output_dir, 'torsions')\n tf.io.gfile.makedirs(torsions_dir)\n distogram_io.save_torsions(torsions_dir, filebase, sequence, torsions)\n\n if experiment.model.secstruct_multiplier > 0:\n ss_dir = os.path.join(output_dir, 'secstruct')\n tf.io.gfile.makedirs(ss_dir)\n secstruct.save_secstructs(ss_dir, filebase, None, sequence, ss)\n\n if experiment.model.asa_multiplier > 0:\n asa_dir = os.path.join(output_dir, 'asa')\n tf.io.gfile.makedirs(asa_dir)\n secstruct.save_secstructs(asa_dir, filebase, None, sequence,\n np.expand_dims(asa, 1), label='Deepmind 2D ASA')\n\n time_spent = time.time() - start_all_time\n logging.info(\n 'Evaluate %d examples, %d crops %.1f crops/ex. '\n 'Took %.1fs, %.3f s/example %.3f crops/s\\n%s',\n num_examples, num_crops, num_crops / float(num_examples), time_spent,\n time_spent / num_examples, num_crops / time_spent, single_message)\n\n logging.info('Tested on %d', num_examples)","def run(self):\n if self.pb.xvalEN and not self.isXvalMain:\n dt = np.ones(self.pb.total_exam_no,\n dtype=\"float32\") / self.train_exam_no\n dt[self.val_indices] = 0.0\n else:\n dt = np.ones(self.pb.total_exam_no,\n dtype=\"float32\") / self.pb.total_exam_no\n\n val = np.zeros(8,dtype=\"float32\")-1\n boosting = None\n wl = None\n if self.pb.algorithm == 'conf-rated':\n boosting = ConfidenceRated(self)\n wl = ConfidenceRatedWL(self)\n elif self.pb.algorithm == 'adaboost':\n boosting = AdaBoost(self)\n wl = AdaBoostWL(self)\n elif self.pb.algorithm == 'adaboost-fast':\n boosting = AdaBoostFast(self)\n wl = AdaBoostFastWLMPI(self)\n elif self.pb.algorithm == 'rankboost':\n boosting = RankBoost(self)\n wl = ConfidenceRatedWL(self)\n elif self.pb.algorithm == 'rankboost-fast':\n boosting = RankBoost(self)\n wl = AdaBoostFastWL(self)\n else:\n raise Exception(\"Unknown Boosting Algorithm\")\n \n for r in range(self.pb.rounds):\n tree = wl.run(dt)\n dt = boosting.run(dt = dt,\n r = r,\n tree = tree)\n \n if self.isXvalMain:\n boosting.finalize()\n \n \"\"\"Sync the predictions and save them to a file\"\"\"\n if self.pb.isLeader:\n if self.pb.xvalEN and not self.isXvalMain:\n val_predictions = boosting.get_val_predictions()\n hypotheses = boosting.get_hypotheses()\n np.savez(self.out_fp,\n val_predictions = val_predictions,\n hypotheses = hypotheses,\n )\n if self.pb.testEN and self.isXvalMain:\n train_predictions = np.zeros([self.pb.total_exam_no,self.pb.rounds],\n dtype=\"float32\")\n test_predictions = np.zeros([self.pb.test_exam_no,self.pb.rounds],\n dtype=\"float32\")\n for slv in np.arange(self.pb.comm_size):\n tr_i1 = self.pb.train_partition[slv]\n tr_i2 = self.pb.train_partition[slv+1]\n te_i1 = self.pb.test_partition[slv]\n te_i2 = self.pb.test_partition[slv+1]\n if slv == 0:\n train_predictions[tr_i1:tr_i2,:] = boosting.get_train_predictions()\n test_predictions[te_i1:te_i2,:] = boosting.get_test_predictions()\n else:\n train_predictions[tr_i1:tr_i2,:] = self.pb.comm.recv(source=slv,tag=11)\n test_predictions[te_i1:te_i2,:] = self.pb.comm.recv(source=slv,tag=12)\n hypotheses = boosting.get_hypotheses()\n np.savez(self.out_fp,\n train_predictions = train_predictions,\n test_predictions = test_predictions,\n hypotheses = hypotheses,\n )\n if not self.pb.testEN and self.isXvalMain:\n train_predictions = np.zeros([self.pb.total_exam_no,self.pb.rounds],\n dtype=\"float32\")\n for slv in np.arange(self.pb.comm_size):\n tr_i1 = self.pb.train_partition[slv]\n tr_i2 = self.pb.train_partition[slv+1]\n if slv == 0:\n train_predictions[tr_i1:tr_i2,:] = boosting.get_train_predictions()\n else:\n train_predictions[tr_i1:tr_i2,:] = self.pb.comm.recv(source=slv,tag=11)\n hypotheses = boosting.get_hypotheses()\n np.savez(self.out_fp,\n train_predictions = train_predictions,\n hypotheses = hypotheses,\n )\n else:\n if self.pb.testEN and self.isXvalMain:\n train_predictions = boosting.get_train_predictions()\n self.pb.comm.send(train_predictions,dest = 0,tag=11)\n test_predictions = boosting.get_test_predictions()\n self.pb.comm.send(test_predictions,dest = 0,tag=12)\n if not self.pb.testEN and self.isXvalMain:\n train_predictions = boosting.get_train_predictions()\n self.pb.comm.send(train_predictions,dest = 0,tag=11)","def print_results(self, f1_score, precision_score, recall_score):\n print(\"Algorithm: %s\" % self.name)\n self.print_score(f1_score, \"F1 Score\")\n self.print_score(precision_score, \"Precision Score\")\n self.print_score(recall_score, \"Recall Score\")","def run(self):\n evaluateAllRuns = False\n while True:\n if self.host == \"\":\n # respond to clients\n self.respond2Clients()\n else:\n print(\"Next...\")\n # randomly choose experiment + run\n if not evaluateAllRuns:\n print(\"Randomly fetching run\")\n self.exp, self.runnum, self.detname = randExpRunDet()\n else:\n\t\t try:\n print(\"Fecthing next run in experiment\")\n self.exp, self.runnum, self.detname = nextExpRunDet(self.goodExp, self.runList[0])\n if self.exp is None:\n self.runList.pop(0)\n continue\n except:\n evaluateAllRuns = False\n continue\n if not self.checkStatus(self.exp, self.runnum, self.detname):\n print \"trying: exp %s, run %s, det %s\"%(self.exp,self.runnum,self.detname)\n try: #temp\n self.ds = safeDataSource(self.exp, self.runnum)\n except: #temp\n continue #temp\n self.run = self.ds.runs().next()\n self.times = self.run.times()\n #Start temp code\n if self.detname is None:\n continue\n #End temp code\n self.det = psana.Detector(self.detname)\n self.det.do_reshape_2d_to_3d(flag=True)\n try:\n self.iX = np.array(self.det.indexes_x(self.run), dtype=np.int64)\n self.iY = np.array(self.det.indexes_y(self.run), dtype=np.int64)\n self.ipx, self.ipy = self.det.point_indexes(self.run, pxy_um=(0, 0))\n self.alg = PyAlgos()\n self.alg.set_peak_selection_pars(npix_min=2, npix_max=30, amax_thr=300, atot_thr=600, son_min=10)\n mask = self.det.mask(self.runnum, calib=True, status=True, edges=True, central=True, unbond=True, unbondnbrs=True)\n\n samples = np.linspace(0, len(self.times), num=100, endpoint=False, retstep=False, dtype='int')\n offset = np.floor(np.random.uniform(0, len(self.times)-samples[-1])).astype('int')\n mysamples = samples + offset\n numCrystals = 0\n for self.eventNum in mysamples:\n self.evt = self.run.event(self.times[self.eventNum])\n calib = self.det.calib(self.evt)\n if calib is not None:\n peaks = self.alg.peak_finder_v3r3(calib, rank=3, r0=3, dr=2, nsigm=10, mask=mask.astype(np.uint16))\n if self.likelihood(peaks) >= self.goodLikelihood:\n numCrystals += 1\n if numCrystals >= self.minCrystals:\n self.numSaved +=1\n self.updateStatus(self.exp, self.runnum, self.detname, self.numSaved)\n self.lastGood = True\n break\n except:\n print \"Could not analyse this run\"\n #If an experiment has not had all of its runs evaluated yet\n # and if the last randomly selected run in this experiment was good\n # then all the runs in this experiment should be evaluated\n if (self.exp not in self.goodList) and self.lastGood:\n self.goodExp = self.exp #Save the name of this experiment\n self.goodRun = self.runnum #Save the run that has already been evaluated\n self.lastGood = False #Reset the condition that the last run was \"good\"\n self.goodList.append(self.goodExp) #Add this experiment name to the list of experiments that have had all runs evaluated\n self.runList = returnRunList(self.goodExp, self.goodRun) #save list of all runs in this good exp\n evaluateAllRuns = True #rerun loop with new algorithm that evaluates each run in an experiment\n continue\n if evaluateAllRuns: #If the loop is currently evaluating all of the runs in an experiment\n if(len(self.runList) > 1):\n self.runList.pop(0) #Remove runs from the list of runs each time they are evaluated\n else:\n self.runList.pop(0)#Remove runs until the list is completely empty\n evaluateAllRuns = False #Stop evaluated all the runs of an experiment, go back to random fetching","def testDriver():\n exam1=90\n exam2=85\n assignmentScores = [50, 60, 70, 80, ]\n computeGrades(exam1, exam2, assignmentScores)","def sbm_experiments():\n # The key experiment is to fix n, r, and p and vary the ratio q/p.\n n = 1000\n average_over = 10\n\n # We will run a few experiments\n rs = [5]\n ps = [1e-11]\n run_lps = [False]\n\n # Use the same ratios for each experiment\n ratios = [0.5 * x for x in range(4, 61)]\n\n # Whether to append results to the results files\n append_results = False\n\n for index in range(len(rs)):\n sbm_experiment_internal(n, rs[index], ps[index], average_over, run_lps[index], ratios, append_results)","def run(self):\n if self.pb.xvalEN and not self.isXvalMain:\n dt = np.ones(self.pb.total_exam_no,\n dtype=\"float32\") / self.train_exam_no\n dt[self.val_indices] = 0.0\n else:\n dt = np.ones(self.pb.total_exam_no,\n dtype=\"float32\") / self.pb.total_exam_no\n\n val = np.zeros(8,dtype=\"float32\")-1\n boosting = None\n wl = None\n if self.pb.algorithm == 'conf-rated':\n boosting = ConfidenceRated(self)\n wl = ConfidenceRatedWL(self)\n elif self.pb.algorithm == 'adaboost':\n boosting = AdaBoost(self)\n wl = AdaBoostWL(self)\n elif self.pb.algorithm == 'adaboost-fast':\n boosting = AdaBoostFast(self)\n wl = AdaBoostFastWL(self)\n elif self.pb.algorithm == 'rankboost':\n boosting = RankBoost(self)\n wl = ConfidenceRatedWL(self)\n elif self.pb.algorithm == 'rankboost-fast':\n boosting = RankBoost(self)\n wl = AdaBoostFastWL(self)\n else:\n raise Exception(\"Unknown Boosting Algorithm\")\n \n for r in range(self.pb.rounds):\n tree = wl.run(dt)\n dt = boosting.run(dt = dt,\n r = r,\n tree = tree)\n \n if self.isXvalMain:\n boosting.finalize()\n \n \"\"\"Sync the predictions and save them to a file\"\"\"\n if self.pb.isLeader:\n if self.pb.xvalEN and not self.isXvalMain:\n val_predictions = boosting.get_val_predictions()\n hypotheses = boosting.get_hypotheses()\n np.savez(self.out_fp,\n val_predictions = val_predictions,\n hypotheses = hypotheses,\n )\n if self.pb.testEN and self.isXvalMain:\n train_predictions = boosting.get_train_predictions()\n test_predictions = boosting.get_test_predictions()\n hypotheses = boosting.get_hypotheses()\n np.savez(self.out_fp,\n train_predictions = train_predictions,\n test_predictions = test_predictions,\n hypotheses = hypotheses,\n )\n if not self.pb.testEN and self.isXvalMain:\n train_predictions = boosting.get_train_predictions()\n hypotheses = boosting.get_hypotheses()\n np.savez(self.out_fp,\n train_predictions = train_predictions,\n hypotheses = hypotheses,\n )","def run_experiment():\n pass","def test_compare(self):\n config = {\n 'num_components': 512,\n 'num_features': 128,\n 'covariance': 'spherical'\n }\n\n samples = self.generate_samples(config, 100_000)\n sklearn_time = np.mean([self.train_sklearn(config, samples) for _ in range(3)])\n ours_cpu_time = np.mean([self.train_ours(config, samples) for _ in range(3)])\n ours_gpu_time = np.mean([\n self.train_ours(config, samples.cuda(), gpu=True) for _ in range(3)\n ])\n\n print(f\"-------------------------------------\")\n print(f\"Speedup of CPU implementation: {sklearn_time / ours_cpu_time:.2f}\")\n print(f\"Speedup of GPU implementation: {sklearn_time / ours_gpu_time:.2f}\")\n print(f\"-------------------------------------\")","def test(self, plot=False):\n accuracy_list = []\n fobj_avg = self.load_stats()\n\n for ii in range(settings.PARS['maxIters']):\n model = self.load_model(ii)\n D1 = model['D']\n W1 = model['W']\n\n # classification\n tic = time.time()\n accuracy_list.append(self.classification(D1, W1)[1])\n toc = time.time()\n print(\n 'Final recognition rate for OnlineDL is : {} , objective function value: {}, time: {}'\n .format(accuracy_list[ii], fobj_avg[ii], toc-tic)\n )\n\n accuracy_list = np.asarray(accuracy_list)\n\n print('Best recognition rate for OnlineDL is {} at iteration {}'.format(\n accuracy_list.max(), accuracy_list.argmax()))\n\n if plot:\n # plot the objective function values for all iterations\n plt.clf()\n plt.plot(list(fobj_avg.keys()), list(fobj_avg.values()), 'mo--', linewidth=2)\n plt.xlabel('Iterations')\n plt.ylabel('Average objective function value')\n plt.xticks(list(range(0, 20)), list(range(1, 21)))\n plt.show()\n\n plt.clf()\n plt.plot(accuracy_list, 'rs--', linewidth=2)\n plt.xticks(list(range(0, 20)), list(range(1, 21)))\n plt.xlabel('Iterations')\n plt.ylabel('Accuracy')\n plt.show()","def run_tests(self):\n\n self.test_report = []\n\n #dict of unsorted lists\n dict_of_un_lists = self.dict_un_lists_intersection_test(self.data_dict)\n self.test_report.append(dict_of_un_lists)\n\n #dict of sets\n dict_of_sets = self.build_dict_of_sets(self.data_dict)\n self.test_report.append(self.dict_sets_intersection_test(dict_of_sets))\n\n #pandas - experimental and probably not the way to use pandas\n # dict_of_pandas = self.build_dict_of_panda_series(self.data_dict)\n # self.test_report.append(self.dicts_any_intersection_node_test(dict_of_pandas))\n\n # print results\n\n if self.verbose:\n self.print_tests_results()","def main(titles, paragraphs, printDetailedStats = True):\n avg_precision = 0 # all the precision values will be added to this value. In the end, this value will be divided by\n # the number of comparisons analysed to report the average precision\n avg_recall = 0\n avg_fmeasure = 0\n\n if paragraphs:\n manual_directory = path.MANUAL_PARAGRAPHS\n auto_directory = path.OUTDIR_PARAGRAPHS\n else:\n manual_directory = path.MANUAL_SENTENCES\n auto_directory = path.OUTDIR_SENTENCES\n\n for title in titles:\n with open(manual_directory + title) as m:\n with open(auto_directory + title) as a:\n tPositive, fPositive, fNegative, precision, recall, fmeasure = \\\n compare(m.readlines(), a.readlines(), paragraphs)\n avg_precision += precision\n avg_recall += recall\n avg_fmeasure += fmeasure\n if printDetailedStats: # then report precision and recall for every individual article\n print(\"comparing \" + title)\n print(\"tp=\" + str(tPositive) + \" fn=\" + str(fNegative) + \" fp=\" + str(fPositive))\n print(\"precision=\" + str(round(precision, 5)) + \"\\t\\t recall=\" + str(\n round(recall, 5)) + \"\\t\\t fmeasure=\" + str(round(fmeasure, 5)) + \" \\n\\n\")\n print(\"AVERAGE_PRECISION=\" + str(round(avg_precision/len(titles), 5)) + \"\\t\\t AVERAGE_RECALL=\"\n + str(round(avg_recall/len(titles), 5)) + \"\\t\\t AVERAGE_F_MEASURE=\"\n + str(round(avg_fmeasure/len(titles), 5)))\n return avg_fmeasure/len(titles)","def test_run_multi_r__(self):\n\n # Test Description\n # ================\n #\n # 1. This test intialises an example *eopy.matchup.matchupIO.MatchUp* object\n #\n # 2. Compare transformed dataset to expected value\n\n ################################################################################################################\n # 1. Initialise Test Data Object\n ################################################################################################################\n\n MatchUpTest = return_MatchUpTest_r__()\n\n ################################################################################################################\n # 2. Define expected values\n ################################################################################################################\n\n # Original dataset values (should be unchanged)\n MatchUpOriginal_expected = return_MatchUpTest_r__()\n\n # Transformed dataset\n values_expected = array([294.0625, 480.3733333, 300.6, 227.3846154, 210.1533333,\n 22.74193548, 22.0625, 21.96875, 22.80645161, 23.5,\n 21.66666667, 21.05882353, 23, 22.40625,\n 38.33333333, 36.63636364, 36.5, 38.42857143,\n 30.1, 32.14893617, 29.37254902, 28.88461538, 28.56603774,\n 33.45238095, 32.81395349, 31.77272727, 32.60465116,\n 40.125, 43.54054054, 38.59090909, 34.08510638,\n 13.72727273, 12, 14.1, 11.79069767, 17.53846154,\n 12.69565217, 31.16666667, 12.26086957, 11.52272727,\n 8.8125, 12, 7.4, 10.13207547])\n unc_expected = [Uncertainty(1, array([1.6, 1.5, 1.5, 1.3, 1.5])),\n Uncertainty(1, array([3.1, 3.2, 3.2, 3.1, 3.0])),\n Uncertainty(1, array([3.3, 3.4, 3.1, 3.2])),\n Uncertainty(1, array([2.1, 2.2, 2.2, 2.1])),\n Uncertainty(1, array([5.0, 4.7, 5.1, 5.2, 5.3])),\n Uncertainty(1, array([4.2, 4.3, 4.4, 4.3])),\n Uncertainty(1, array([4.0, 3.7, 4.4, 4.7])),\n Uncertainty(1, array([2.2, 1.7, 2.0, 4.3, 2.6])),\n Uncertainty(1, array([2.3, 1.2, 2.3, 4.4])),\n Uncertainty(1, array([3.2, 2.7, 3.0, 5.3]))]\n ks_expected = array([4.8, 6.8, 5.2, 5.6, 5.2, 12.10287443, 13.99394856, 12.48108926, 12.85930408])\n unck_expected = [Uncertainty(1, array([0.25, 0.25, 0.25, 0.25, 0.25])),\n Uncertainty(1, array([0.2644, 0.2644, 0.2644, 0.2644]))]\n idx_expected = {\"Nm\": [5, 4],\n \"cNm\": [0, 5, 9],\n \"Im\": [[0, 1], [1, 2]],\n \"sensors\": [-1, 1, 2],\n \"sensor_ms\": [1, 3, 3],\n \"n_sensor\": [0, 1, 1, 2, 1, 1, 2, 1, 1, 2],\n \"n_mu\": [1, 1, 2, 2, 1, 2, 2, 1, 2, 2],\n \"n_cov\": [1, 1, 1, 1, 2, 2, 2, 3, 3, 3],\n \"N_var\": [5, 5, 4, 4, 5, 4, 4, 5, 4, 4],\n \"idx\": [0, 5, 10, 14, 18, 23, 27, 31, 36, 40, 44],\n \"Ia\": [1, 1, 1, 2, 2, 2]}\n a_expected = array([1., 1.3, 0.002, 0.5, 1.1, 0.0005])\n w_matrices_expected = []\n u_matrices_expected = []\n\n ################################################################################################################\n # 3. Run Transform2NormInd.run()\n ################################################################################################################\n\n Transform2NormIndOp = Transform2NormInd()\n MatchUpTransform = Transform2NormIndOp.run(MatchUpTest)\n\n values_test = MatchUpTransform.values\n unc_test = MatchUpTransform.unc\n w_matrices_test = MatchUpTransform.w_matrices\n u_matrices_test = MatchUpTransform.u_matrices\n ks_test = MatchUpTransform.ks\n unck_test = MatchUpTransform.unck\n idx_test = MatchUpTransform.idx\n\n ################################################################################################################\n # 4. Compare retrieve values to expect values\n ################################################################################################################\n\n # Test transformed data object attribute by attribute\n\n # a. values\n for i, (value_expected, value_test) in enumerate(zip(values_expected, values_test)):\n self.assertAlmostEqual(value_expected, value_test, places=5, msg=str(i))\n\n # b. unc\n for block_unc_test, block_unc_expected in zip(unc_test, unc_expected):\n self.assertEqual(block_unc_expected.typeID, block_unc_test.typeID)\n self.assertEqual(block_unc_expected.uR.tolist(), block_unc_test.uR.tolist())\n\n # c. w_matrices\n self.assertEqual(w_matrices_test, w_matrices_expected)\n\n # d. u_matrices\n self.assertEqual(u_matrices_test, u_matrices_expected)\n\n # e. ks\n for k_expected, k_test in zip(ks_expected, ks_test):\n self.assertAlmostEqual(k_test.tolist(), k_expected.tolist(), places=5)\n\n # f. unck\n for block_unck_test, block_unck_expected in zip(unck_test, unck_expected):\n self.assertEqual(block_unck_expected.typeID, block_unck_test.typeID)\n self.assertEqual(block_unck_expected.uR.tolist(), block_unck_test.uR.tolist())\n\n # h. idx\n self.assertEqual(set(idx_expected.keys()), set(idx_test.keys()))\n for key in idx_expected.keys():\n idx_i_test = idx_test[key]\n idx_i_expected = idx_expected[key]\n if isinstance(idx_i_expected, ndarray):\n self.assertEqual(idx_i_test.tolist(), idx_i_expected.tolist())\n else:\n self.assertEqual(idx_i_test, idx_i_expected)\n\n # Test original data object preserved attribute by attribute\n\n # a. values\n for i, (value_original_expected, value_original_test) in enumerate(zip(MatchUpOriginal_expected.values, MatchUpTest.values)):\n self.assertAlmostEqual(value_original_expected, value_original_test, places=5)\n\n # b. unc\n for block_unc_original_expected, block_unc_original_test in zip(MatchUpOriginal_expected.unc, MatchUpTest.unc):\n self.assertEqual(block_unc_original_expected.typeID, block_unc_original_test.typeID)\n self.assertEqual(block_unc_original_expected.uR.tolist(), block_unc_original_test.uR.tolist())\n\n # c. w_matrices\n self.assertEqual(MatchUpOriginal_expected.w_matrices, MatchUpTest.w_matrices)\n\n # d. u_matrices\n self.assertEqual(MatchUpOriginal_expected.u_matrices, MatchUpTest.u_matrices)\n\n # e. ks\n for k_original_expected, k_original_test in zip(MatchUpOriginal_expected.ks, MatchUpTest.ks):\n self.assertAlmostEqual(k_original_test.tolist(), k_original_expected.tolist(), places=5)\n\n # f. unck\n for block_unck_original_expected, block_unck_original_test in zip(MatchUpOriginal_expected.unck, MatchUpTest.unck):\n self.assertEqual(block_unck_original_expected.typeID, block_unck_original_test.typeID)\n self.assertEqual(block_unck_original_expected.uR.tolist(), block_unck_original_test.uR.tolist())\n\n # h. idx\n self.assertEqual(set(MatchUpOriginal_expected.idx), set(MatchUpTest.idx))\n for key in MatchUpOriginal_expected.idx.keys():\n idx_i_original_test = MatchUpTest.idx[key]\n idx_i_original_expected = MatchUpOriginal_expected.idx[key]\n if isinstance(idx_i_original_expected, ndarray):\n self.assertEqual(idx_i_original_test.tolist(), idx_i_original_expected.tolist())\n else:\n self.assertEqual(idx_i_original_test, idx_i_original_expected)","def run():\n tests = [\n # [\n # [\"\", \"sumRange\", \"sumRange\", \"sumRange\", \"update\", \"sumRange\", \"sumRange\", \"sumRange\"],\n # [[1, 2, 3, 4, 5], [0, 2], [1, 2], [1, 3], [1, 3], [0, 2], [1, 2], [0, 1]]\n # ],\n [\n [\"NumArray\", \"sumRange\", \"update\", \"sumRange\"],\n [[-1], [0, 0], [0, 1], [0, 0]]\n ]\n ]\n\n results = [\n # [None, 3, 2, 5, None, 4, 2, 2],\n [None, -1, None, 1],\n\n ]\n for test, res in zip(tests, results):\n print('-----'*20)\n cmds, params = test\n cmds.pop(0)\n p = params.pop(0)\n res.pop(0)\n obj = BIT(p)\n for cmd, param, exp in zip(cmds, params, res):\n v = getattr(obj, cmd)(*param)\n print(obj)\n print(f'------\\n{cmd}, {param} --> {v} : {exp}')\n assert exp == v","def log_evaluation(tester, name, description):\r\n\tfor dataset, output in tester.preds.items():\r\n\t\tresults = pandas.DataFrame.from_dict(output)\r\n\t\tpath = os.path.join(\r\n\t\t\tEXPERIMENT_PATH, tester.config[\"name\"] + '-' + dataset)\r\n\t\twith open(path + \".csv\", \"w\") as f:\r\n\t\t\tresults.to_csv(f, sep=\"\\t\", encoding='utf-8',\r\n\t\t\t\tfloat_format='%.3f', index=False)","def eval(self):\n\n # How many questions we get right at precision@1.\n correct = 0\n\n total = self._analogy_questions.shape[0]\n start = 0\n while start < total:\n limit = start + 2500\n sub = self._analogy_questions[start:limit, :]\n idx = self._predict(sub)\n start = limit\n for question in xrange(sub.shape[0]):\n if sub[question, 3] in idx[question]:\n # print(sub[question, 3], idx[question])\n correct += 1\n\n print()\n print(\"Eval %4d/%d accuracy @ top5= %4.1f%%\" % (correct, total,\n correct * 100. / total)\n )","def perform_experiments(problem_type, output_dir, load_experiments):\n\n experiment_file_path = output_dir + \"experiments.pickle\"\n\n # data structure where to store all the problems, and the experimental results for each algorithm: solutions, final values and times\n experiment_dict = dict()\n\n # if experiments should be loaded (not repeated), do it if possible\n if load_experiments:\n with open(experiment_file_path, \"rb\") as _file:\n experiment_dict = pickle.load(_file)\n\n # perform experiments if pre-existing results were not loaded\n if not experiment_dict:\n\n # given a problem type, create a set of problem instances that are solved (optimally) with manual placements (using actions available for all the algorithms)\n problems, problem_names, manual_solutions = create_problems(problem_type)\n\n if problems and problem_names and manual_solutions:\n\n # parameters for the experimentation; note: calculating internal times and value evolution can increase the overall time of algorithms (in a slight, almost neglectible way)\n execution_num = 10 # 1\n process_num = 10 # 1\n calculate_internal_times = True\n calculate_value_evolution = True\n\n start_time = time.time()\n\n # solve each problem with each algorithm\n for i, (problem, problem_name, solution) in enumerate(zip(problems, problem_names, manual_solutions)):\n\n experiment_dict[problem_name] = {\"problem\": problem, \"manual_solution\": solution, \"algorithms\": dict()}\n\n # solve the problem with different algorithms, executing each one multiple times to gain statistical significance\n for (algorithm_name, algorithm) in [(\"Greedy\", greedy.solve_problem), (\"Reversible\", reversible.solve_problem), (\"Evolutionary\", evolutionary.solve_problem)]:\n\n solutions, values, value_evolutions, times, time_divisions = execute_algorithm(algorithm=algorithm, algorithm_name=algorithm_name, problem=problem, execution_num=execution_num, process_num=process_num, calculate_times=calculate_internal_times, calculate_fitness_stats=calculate_value_evolution)\n experiment_dict[problem_name][\"algorithms\"][algorithm_name] = {\"solutions\": solutions, \"values\": values, \"value_evolutions\": value_evolutions, \"times\": times, \"time_divisions\": time_divisions}\n\n # show the total time spent doing experiments (note that significant overhead can be introduced beyond calculation time if plots are shown or saved to files; for strict time measurements, plotting should be avoided altogether)\n print(\"Total experimental calculation time: {} s\".format(round(get_time_since(start_time) / 1000.), 3))\n\n if experiment_dict:\n\n # experiment-saving parameter\n save_experiments = True\n\n # if possible, save the experiments to a binary file\n if not load_experiments and save_experiments:\n with open(experiment_file_path, \"wb\") as file:\n pickle.dump(experiment_dict, file)\n\n # visualization/saving parameters\n show_problem_stats = True\n save_problem_stats = True # False\n show_manual_solution_plots = True\n save_manual_solution_plots = True # False\n show_algorithm_solution_plots = True\n save_algorithm_solution_plots = True # False\n show_value_evolution_plots = False\n save_value_evolution_plots = False # True\n show_time_division_plots = False\n save_time_division_plots = False # True\n show_algorithm_comparison = True\n save_algorithm_comparison = True # False\n show_aggregated_result_tables = True\n save_aggregated_result_tables = True # False\n\n # show/save the results of the experiments\n visualize_and_save_experiments(experiment_dict, output_dir, can_plots_show_value_and_weight=problem_type == KNAPSACK_PACKING_PROBLEM_TYPE, show_problem_stats=show_problem_stats, save_problem_stats=save_problem_stats, show_manual_solution_plots=show_manual_solution_plots, save_manual_solution_plots=save_manual_solution_plots, show_algorithm_solution_plots=show_algorithm_solution_plots, save_algorithm_solution_plots=save_algorithm_solution_plots, show_value_evolution_plots=show_value_evolution_plots, save_value_evolution_plots=save_value_evolution_plots, show_time_division_plots=show_time_division_plots, save_time_division_plots=save_time_division_plots, show_algorithm_comparison=show_algorithm_comparison, save_algorithm_comparison=save_algorithm_comparison, show_aggregated_result_tables=show_aggregated_result_tables, save_aggregated_result_tables=save_aggregated_result_tables)\n\n else:\n print(\"The experiments cannot be performed (there are no problems available).\")","def evaluate(model, test_files):\n print(\"Running predictions.\")\n models = load_model(model)\n predictions = predict(models, test_files)\n\n # # write predictions to file\n # write_predictions(\"evaluate_out.json\",predictions)\n evaluate_individual(predictions, test_files, models)\n evaluate_overall(predictions)","def main():\n\n convergence_rates_e1 = pd.DataFrame(index=['Mean', 'St.D'])\n convergence_rates_e4 = pd.DataFrame(index=['Mean', 'St.D'])\n success_rates = pd.DataFrame(index=['1e-1', '1e-2', '1e-3', '1e-4', '1e-5'])\n peak_ratios = pd.DataFrame(index=['1e-1', '1e-2', '1e-3', '1e-4', '1e-5'])\n\n for function in range(1, 21):\n results = pickle.load(open('results/benchmarking_result_{}.pkl'.format(function), 'rb'))\n\n col_name = 'F{}'.format(function)\n index = function-1\n\n convergence_rates_e1.insert(index, col_name, results.ConvergenceRates[0])\n convergence_rates_e4.insert(index, col_name, results.ConvergenceRates[3])\n\n success_rates.insert(index, col_name, results.SuccessRate)\n peak_ratios.insert(index, col_name, results.PeakRatio)\n\n for i in results.SimulationSwarms:\n\n x_axis = [k for k, _ in i.items()]\n y_axis = [v for _, v in i.items()]\n\n plt.subplot(4, 5, function) # subplot indexes from 1\n plt.plot(x_axis, y_axis, 'k-')\n\n plt.savefig('results/nmmso_benchmark.png')\n plt.show()\n\n pd.set_option('display.max_columns', None) # make sure all columns are printed below\n print(convergence_rates_e1)\n print(convergence_rates_e4)\n print(success_rates)\n print(peak_ratios)\n\n # Table V from Fieldsend et al.\n table4 = pd.Series([\n peak_ratios.stack().median(),\n peak_ratios.stack().mean(),\n peak_ratios.stack().std()\n ])\n print(table4)\n\n # do we want to reproduce Fig. 3","def getTestResults():","def eval_perf_total(model, X_train, y_train, X_test, y_test):\n\n y_hat_train = model.predict(X_train)\n y_hat_test = model.predict(X_test)\n \n train_mae = metrics.mean_absolute_error(y_train, y_hat_train)\n train_mse = metrics.mean_squared_error(y_train, y_hat_train)\n train_rmse = np.sqrt(metrics.mean_squared_error(y_train, y_hat_train))\n train_r = metrics.r2_score(y_train, y_hat_train)\n\n print('Evaluating Performance on Training Data:\\n')\n print(f' Train Mean Absolute Error: {train_mae:,.2f}')\n print(f' Train Mean Squared Error: {train_mse:,.2f}\\n')\n print(f'Train Root Mean Squared Error: {train_rmse:,.2f}')\n print(f'Train R-Square Value: {round(train_r,2)}')\n\n print('\\n'+'---'*25+'\\n')\n\n test_mae = metrics.mean_absolute_error(y_test, y_hat_test)\n test_mse = metrics.mean_squared_error(y_test, y_hat_test)\n test_rmse = np.sqrt(metrics.mean_squared_error(y_test, y_hat_test))\n test_r = metrics.r2_score(y_test, y_hat_test)\n\n print('Evaluating Performance on Testing Data:\\n')\n print(f' Test Mean Absolute Error: {test_mae:,.2f}')\n print(f' Test Mean Squared Error: {test_mse:,.2f}\\n')\n print(f'Test Root Mean Squared Error: {test_rmse:,.2f}')\n print(f'Test R-Square Value: {round(test_r,2)}')","def test_precision_recall_curves_vis_api(experiment_to_use):\n experiment = experiment_to_use\n probabilities = experiment.probabilities\n viz_outputs = (\"pdf\", \"png\")\n positive_label = 1\n with TemporaryDirectory() as tmpvizdir:\n for viz_output in viz_outputs:\n vis_output_pattern_pdf = tmpvizdir + f\"/*.{viz_output}\"\n visualize.precision_recall_curves(\n [probabilities, probabilities],\n experiment.ground_truth,\n experiment.ground_truth_metadata,\n experiment.output_feature_name,\n positive_label,\n model_names=[\"Model1\", \"Model2\"],\n output_directory=tmpvizdir,\n file_format=viz_output,\n )\n figure_cnt = glob.glob(vis_output_pattern_pdf)\n assert 1 == len(figure_cnt)","def test_net_on_dataset(args, dataset_name, proposal_file, output_dir, multi_gpu=False, gpu_id=0, use_matlab = False, early_stop=False):\n\n \n # print(\"test_net_on_dataset\")\n dataset = JsonDataset(dataset_name)\n test_timer = Timer()\n \n test_timer.tic()\n \n all_boxes = test_net(args, dataset_name, proposal_file, output_dir, gpu_id=gpu_id, early_stop=early_stop)\n test_timer.toc()\n\n logger.info('Total inference time: {:.3f}s'.format(test_timer.average_time))\n\n roidb = dataset.get_roidb()\n num_images = len(roidb)\n num_classes = cfg.MODEL.NUM_CLASSES + 1\n final_boxes = empty_results(num_classes, num_images)\n test_corloc = 'train' in dataset_name\n \n\n all_cls_scores = {}\n\n for i, entry in enumerate(roidb):\n\n if early_stop and i > 10: break\n\n boxes = all_boxes[entry['image']]\n \n cls_key = entry['image'].replace('.jpg','').split('/')[-1]\n\n # print(cls_key)\n\n if boxes['scores'] is not None:\n if test_corloc:\n # print(\"corlooking\")\n _, _, cls_boxes_i = box_results_for_corloc(boxes['scores'], boxes['boxes'])\n else:\n _, _, cls_boxes_i = box_results_with_nms_and_limit(boxes['scores'], boxes['boxes'])\n\n extend_results(i, final_boxes, cls_boxes_i)\n else:\n final_boxes = None\n \n results = task_evaluation.evaluate_all(dataset, final_boxes, output_dir, test_corloc, use_matlab = use_matlab)\n return results","def print_results(results, random_counterpart=None, random_concepts=None, num_random_exp=100,\n min_p_val=0.05):\n\n # helper function, returns if this is a random concept\n def is_random_concept(concept):\n if random_counterpart:\n return random_counterpart == concept\n\n elif random_concepts:\n return concept in random_concepts\n\n else:\n return 'random500_' in concept\n\n # print class, it will be the same for all\n print(\"Class =\", results[0]['target_class'])\n\n # prepare data\n # dict with keys of concepts containing dict with bottlenecks\n result_summary = {}\n\n # random\n random_i_ups = {}\n\n for result in results:\n if result['cav_concept'] not in result_summary:\n result_summary[result['cav_concept']] = {}\n\n if result['bottleneck'] not in result_summary[result['cav_concept']]:\n result_summary[result['cav_concept']][result['bottleneck']] = []\n\n result_summary[result['cav_concept']][result['bottleneck']].append(result)\n\n # store random\n if is_random_concept(result['cav_concept']):\n if result['bottleneck'] not in random_i_ups:\n random_i_ups[result['bottleneck']] = []\n\n random_i_ups[result['bottleneck']].append(result['i_up'])\n\n # print concepts and classes with indentation\n for concept in result_summary:\n\n # if not random\n if not is_random_concept(concept):\n print(\" \", \"Concept =\", concept)\n\n for bottleneck in result_summary[concept]:\n i_ups = [item['i_up'] for item in result_summary[concept][bottleneck]]\n\n # Calculate statistical significance\n _, p_val = ttest_ind(random_i_ups[bottleneck], i_ups)\n\n print(3 * \" \", \"Bottleneck =\", (\"%s. TCAV Score = %.2f (+- %.2f), \"\n \"random was %.2f (+- %.2f). p-val = %.3f (%s)\") % (\n bottleneck, np.mean(i_ups), np.std(i_ups),\n np.mean(random_i_ups[bottleneck]),\n np.std(random_i_ups[bottleneck]), p_val,\n \"not significant\" if p_val > min_p_val else \"significant\"))","def analyze_end_performance(self) -> None:\n self.logger.info(f\"###### FINAL RESULTS FOR EACH GAME###### \\n\")\n total_results = [0, 0, 0, 0]\n for game in self.instance_predictions:\n self.logger.info(f\"Simulated game {game.home_team.name} vs {game.away_team.name} \"\n f\"{len(self.instance_predictions[game])} times.\")\n confusion_matrix = [0, 0, 0, 0]\n for predicted_winner in self.instance_predictions[game]:\n actual_winner = game.home_team if game.home_team.scores.get(\n GamePeriod.TOTAL) > game.away_team.scores.get(GamePeriod.TOTAL) else game.away_team\n actual_home_team_won = True if actual_winner == game.home_team else False\n predicted_home_team_won = True if predicted_winner == game.home_team else False\n confusion_matrix = recursive_build_confusion_matrix(confusion_matrix, actual_home_team_won,\n predicted_home_team_won)\n total_results[0] += confusion_matrix[0]\n total_results[1] += confusion_matrix[1]\n total_results[2] += confusion_matrix[2]\n total_results[3] += confusion_matrix[3]\n\n for k, v in analyze_confusion_matrix(confusion_matrix, len(self.instance_predictions[game])).items():\n self.logger.info(f\"{k} : {v}\")\n self.logger.info(\"\\n###### FINAL RESULTS FOR ALL GAMES###### \\n\")\n final_stats = analyze_confusion_matrix(total_results, sum(total_results))\n for k, v in final_stats.items():\n self.logger.info(f\"{k} : {v}\")\n self.write_stats_to_json(final_stats, self.outfile)","def accuracy(num_points, num_dims, p, surv, num_runs=100, num_test=10000):\n p1 = []\n p2 = []\n p3 = []\n p4 = []\n for t in range(num_runs):\n if surv:\n cloud = generate_survival_curves(num_points + num_test, num_dims)\n else:\n cloud = generate_time_series(num_points + num_test, num_dims)\n opt_cloud = cloud[:num_points]\n test_cloud = cloud[num_points:]\n p1.append(proportion(test_cloud, olshen(opt_cloud, p, surv)))\n p2.append(proportion(test_cloud, two_sided_olshen(opt_cloud, p, surv)))\n p3.append(proportion(test_cloud, gspie(opt_cloud, p, surv)))\n p4.append(\n proportion(test_cloud, pointwise_bonferroni(opt_cloud, p, surv)))\n plt.boxplot([p1, p2, p3, p4])\n plt.xticks([1, 2, 3, 4],\n ['Olshen', 'Two-Sided Olshen', 'GSPIE', 'Pointwise Bonferroni'])\n plt.ylabel('Observed Coverage')\n plt.xlabel(f\"{p} Prescribed Coverage\")\n plt.show()","def test_compare_classifiers_predictions_distribution_vis_api(experiment_to_use):\n experiment = experiment_to_use\n predictions = experiment.predictions_num\n viz_outputs = (\"pdf\", \"png\")\n with TemporaryDirectory() as tmpvizdir:\n for viz_output in viz_outputs:\n vis_output_pattern_pdf = tmpvizdir + f\"/*.{viz_output}\"\n visualize.compare_classifiers_predictions_distribution(\n [predictions, predictions],\n experiment.ground_truth,\n experiment.ground_truth_metadata,\n experiment.output_feature_name,\n labels_limit=0,\n model_names=[\"Model1\", \"Model2\"],\n output_directory=tmpvizdir,\n file_format=viz_output,\n )\n figure_cnt = glob.glob(vis_output_pattern_pdf)\n assert 1 == len(figure_cnt)","def evaluate_overall(predictions):\n print(\"\\nAccuracy for the the models combined (full prediction):\\n\")\n\n perfect_predictions = [0, 0, 0]\n exact_code_match = [0, 0, 0]\n exact_code_match_ignore_spaces = [0, 0, 0]\n correct_syntax = [0, 0, 0]\n total_predictions = [0, 0, 0]\n num_failed_predictions = 0\n \n for prediction in predictions:\n # Perfect Prediction\n if prediction[\"correct_data\"][\"correct_location\"] == prediction[\"predicted_location\"]:\n # Correct type\n if prediction[\"correct_data\"][\"correct_type\"] == prediction[\"predicted_type\"]:\n # Correct token\n if prediction[\"predicted_type\"] == \"insert\" or prediction[\"predicted_type\"] == \"modify\":\n if \"correct_token\" in prediction[\"correct_data\"] and prediction[\"correct_data\"][\"correct_token\"] == \\\n prediction[\"predicted_token\"]:\n perfect_predictions[FixType[prediction[\"predicted_type\"]].value] = perfect_predictions[FixType[\n prediction[\"predicted_type\"]].value] + 1\n else:\n perfect_predictions[FixType[prediction[\"predicted_type\"]].value] = perfect_predictions[FixType[\n prediction[\"predicted_type\"]].value] + 1\n \n # Exact Code Match\n if prediction[\"predicted_code\"] == prediction[\"correct_data\"][\"correct_code\"]:\n exact_code_match[FixType[prediction[\"predicted_type\"]].value] = exact_code_match[FixType[\n prediction[\"predicted_type\"]].value] + 1\n \n # Exact Code Match (Ignore Spaces)\n if prediction[\"predicted_code\"].replace(\" \", \"\") == prediction[\"correct_data\"][\"correct_code\"].replace(\" \", \"\"):\n exact_code_match_ignore_spaces[FixType[prediction[\"predicted_type\"]].value] = exact_code_match_ignore_spaces[FixType[\n prediction[\"predicted_type\"]].value] + 1\n \n # Correct Syntax Fix\n try:\n cst.parse_module(prediction[\"predicted_code\"])\n correct_syntax[FixType[prediction[\"predicted_type\"]].value] = correct_syntax[FixType[\n prediction[\"predicted_type\"]].value] + 1\n except Exception as e:\n logging.warning(f\"{e.__class__.__name__} occurred: {e}\")\n # Happens if parsing fails\n pass\n \n total_predictions[FixType[prediction[\"predicted_type\"]].value] = total_predictions[FixType[\n prediction[\"predicted_type\"]].value] + 1\n \n # Perfect Prediction\n for i in range(3):\n if total_predictions[i] == 0: # If the type was never predicted\n accuracy = 0\n else:\n accuracy = perfect_predictions[i] / total_predictions[i]\n print(f\"Perfect Prediction for class {FixType(i).name}: {accuracy * 100} %\")\n\n accuracy = sum(perfect_predictions) / (len(predictions) - num_failed_predictions)\n print(f\"Perfect Prediction accuracy overall is {accuracy * 100} %\")\n \n # Exact Code Match\n for i in range(3):\n if total_predictions[i] == 0: # If the type was never predicted\n accuracy = 0\n else:\n accuracy = exact_code_match[i] / total_predictions[i]\n print(f\"Exact Code Match for class {FixType(i).name}: {accuracy * 100} %\")\n\n accuracy = sum(exact_code_match) / (len(predictions) - num_failed_predictions)\n print(f\"Exact Code Match accuracy overall is {accuracy * 100} %\")\n \n # Exact Code Match (Ignore Spaces)\n for i in range(3):\n if total_predictions[i] == 0: # If the type was never predicted\n accuracy = 0\n else:\n accuracy = exact_code_match_ignore_spaces[i] / total_predictions[i]\n print(f\"Exact Code Match (Ignore Spaces) for class {FixType(i).name}: {accuracy * 100} %\")\n\n accuracy = sum(exact_code_match_ignore_spaces) / (len(predictions) - num_failed_predictions)\n print(f\"Exact Code Match (Ignore Spaces) accuracy overall is {accuracy * 100} %\")\n \n # Correct Syntax Fixes\n for i in range(3):\n if total_predictions[i] == 0: # If the type was never predicted\n accuracy = 0\n else:\n accuracy = correct_syntax[i] / total_predictions[i]\n print(f\"Correct Syntax Fixes for class {FixType(i).name}: {accuracy * 100} %\")\n\n accuracy = sum(correct_syntax) / (len(predictions) - num_failed_predictions)\n print(f\"Correct Syntax Fixes accuracy overall is {accuracy * 100} %\")","def test_calibration_multiclass_vis_api(experiment_to_use):\n experiment = experiment_to_use\n probabilities = experiment.probabilities\n viz_outputs = (\"pdf\", \"png\")\n with TemporaryDirectory() as tmpvizdir:\n for viz_output in viz_outputs:\n vis_output_pattern_pdf = tmpvizdir + f\"/*.{viz_output}\"\n visualize.calibration_multiclass(\n [probabilities, probabilities],\n experiment.ground_truth,\n experiment.ground_truth_metadata,\n experiment.output_feature_name,\n labels_limit=0,\n model_names=[\"Model1\", \"Model2\"],\n output_directory=tmpvizdir,\n file_format=viz_output,\n )\n figure_cnt = glob.glob(vis_output_pattern_pdf)\n assert 2 == len(figure_cnt)","def check_expectations(self):\n self.load_results()\n\n for (benchmark, producer), result in self.results.items():\n if not result.reports:\n print('No results found for ' + benchmark + ' ' + producer)\n result.test_passed = False\n else:\n for report in result.reports:\n if check_benchmark_result(report, result.expectation):\n print('Test passed: ' + result.directory)\n result.test_passed = True\n else:\n print('Test failed: ' + result.directory)\n result.test_passed = False","def evaluation(deploymentID, start_time,\n end_time, sites, validation, manOrLik):\n \n estScoreBound = 5\n falsePositive_dep = 0\n falseNegative_dep = 0\n truePositive_dep = 0\n trueNegative_dep = 0\n\n db_con = MySQLdb.connect(user=\"root\", db=\"qraat\")\n for i in sites:\n truePositive = 0\n trueNegative = 0\n falsePositive = 0\n falseNegative = 0\n \n cur = db_con.cursor()\n cur.execute(\"\"\"SELECT isPulse, band3, band10, frequency,\n ec, tnp, edsp, fdsp, edsnr, fdsnr\n FROM est INNER JOIN est_class%s\n ON ID = est_class%s.estID\n INNER JOIN estscore2\n ON ID = estscore2.estID\n WHERE timestamp > %s\n AND timestamp < %s\n AND deploymentID = %s\n AND siteID = %s\n AND setNum = %s;\n \"\"\"%(manOrLik, manOrLik, start_time, end_time,\n deploymentID, i, validation))\n for row in cur.fetchall():\n estData = {'band3':row[1], 'band10':row[2],\n 'frequency':row[3], 'ec':row[4], 'tnp':row[5],\n 'edsp':row[6], 'fdsp':row[7],\n 'edsnr':row[8], 'fdsnr':row[9]}\n \n #Classify data\n isPulse = modifiedBC(deploymentID, i, start_time,\n validation, estData, manOrLik)\n\n #Determine whether the classification results are correct or not\n if (row[0] == 1):\n if (isPulse == 1):\n truePositive += 1\n else:\n falseNegative += 1\n else:\n if (isPulse == 1):\n falsePositive += 1\n else:\n trueNegative += 1\n \n falsePositive_dep += falsePositive\n falseNegative_dep += falseNegative\n truePositive_dep += truePositive\n trueNegative_dep += trueNegative\n return [truePositive_dep, trueNegative_dep,\n falsePositive_dep, falseNegative_dep]","def main():\n\n # choose number of data-points and sample a pair of vectors: the input\n # values and the corresponding target values\n N = 500\n inputs, targets = sample_data(N, arbitrary_function_2, seed=1)\n\n # specify the centres and scale of some rbf basis functions\n default_centres = np.linspace(0,1,21)\n default_scale = 0.03\n default_reg_param = 0.08\n\n # get the cross-validation folds\n num_folds = 4\n folds = create_cv_folds(N, num_folds)\n\n # evaluate then plot the performance of different reg params\n evaluate_reg_param(inputs, targets, folds, default_centres, default_scale)\n # evaluate then plot the performance of different scales\n evaluate_scale(inputs, targets, folds, default_centres, default_reg_param)\n # evaluate then plot the performance of different numbers of basis\n # function centres.\n evaluate_num_centres(\n inputs, targets, folds, default_scale, default_reg_param)\n\n plt.show()","def benchmark_algos(algos: List[str], setup_script: str, number: int,\n *args) -> None:\n for algo_name in algos:\n print('\\nRAN {} {} times with args {}'.format(algo_name, number,\n *args))\n print(timeit('{}({})'.format(algo_name, *args),\n setup=setup_script, number=number))","def run(self):\n self.evaluate()\n self.accumulate()\n self.summarize()","def test_variational():\n # iris\n #pres = \"Test pour le data set Iris (facile, classique)\"\n #test_from_func_variational(pres, 15, 10, 3, True, Iris)\n\n # breast cancer\n pres = \"Test pour le data set Breast Cancer (facile, classique)\"\n test_from_func_variational(pres, 15, 10, 3, True, Breast_cancer)\n\n # digits\n # pres = \"Test pour le data set Digits (difficile, classique)\"\n # test_from_func(pres, 10, 10, 10, True, Digits, quantum_instance)\n\n # wine\n # pres = \"Test pour le data set Wine (moyen, classique)\"\n # test_from_func(pres, 15, 10, 5, True, Wine, quantum_instance)\n\n # gaussian\n pres = \"Test pour des données gaussiennes (moyen, classique)\"\n for _ in range(1):\n print(\"\\n\")\n print(\"New iteration\")\n test_from_func_variational(pres, 25, 10, 2, True, Gaussian)\n print(\"\\n\")\n\n # small adn strings\n pres = \"Test pour des séquences ADN courtes (difficile, classique)\"\n test_from_func_variational(pres, 10, 15, 14, True, Sequence)\n\n #Quantum data\n pres = \"Test pour des données générées par ordinateur quantique (facile, quantique)\"\n print(pres)\n _, samp_train, samp_test, labels = ad_hoc_data(15, 10, 2, 0.3, True)\n sample_m, sample_p = stock_get(20, 0.3)\n\n labels_me = [-1, 1]\n samp_train_me = {-1: np.array(sample_m[:15]), 1: np.array(sample_p[:15])}\n samp_test_me = {-1: np.array(sample_m[15:]), 1: np.array(sample_p[15:])}\n print(samp_train)\n print(samp_train_me)\n print(samp_test)\n print(samp_test_me)\n\n my_impl_variational(samp_train, samp_test, labels)\n print(\"Pour autres données quantiques\")\n my_impl_variational(samp_train_me, samp_test_me, labels_me)","def test_compare_classifiers_predictions_vis_api(experiment_to_use):\n experiment = experiment_to_use\n predictions = experiment.predictions\n viz_outputs = (\"pdf\", \"png\")\n with TemporaryDirectory() as tmpvizdir:\n for viz_output in viz_outputs:\n vis_output_pattern_pdf = tmpvizdir + f\"/*.{viz_output}\"\n visualize.compare_classifiers_predictions(\n [predictions, predictions],\n experiment.ground_truth,\n experiment.ground_truth_metadata,\n experiment.output_feature_name,\n labels_limit=0,\n model_names=[\"Model1\", \"Model2\"],\n output_directory=tmpvizdir,\n file_format=viz_output,\n )\n figure_cnt = glob.glob(vis_output_pattern_pdf)\n assert 1 == len(figure_cnt)","def evaluate(clf, dataset, feature_list, features, labels, num_iter, params):\n\n features_train, features_test, labels_train, labels_test = \\\n train_test_split(features, labels, test_size=0.3, random_state=42)\n\n\n\n precision_values = []\n recall_values = []\n accuracy_values = []\n print clf\n for i in xrange(0, num_iter):\n #print params\n clf = GridSearchCV(clf, params)\n clf.fit(features_train, labels_train)\n print '*****************************'\n print clf.best_estimator_\n print clf.best_params_\n\n clf = clf.best_estimator_\n #test_classifier(clf, dataset, feature_list)\n pred = clf.predict(features_test)\n precision_values.append(precision_score(labels_test, pred))\n recall_values.append(recall_score(labels_test, pred))\n accuracy_values.append(accuracy_score(labels_test, pred))\n print 'Recall score: ', mean(recall_values)\n print 'Precision score: ', mean(precision_values)\n print 'Accuracy score: ' , mean(accuracy_values)","def _compute_experiment_statistics(self):\n pass","def run_mcts(self, runs_per_round):\n for i in range(runs_per_round):\n self.select(self.env, 'r')\n self.env_reset()\n counts = [self.Nsa[('r', a)] for a in range(self.actions)]\n # print(\"counts \", counts)\n # print(\"Q-values\", [self.Qsa[('r', a)] for a in range(self.actions)])\n # print()\n return np.argmax(counts)","def benchmark(options):\n # Prepare experiments\n with open(options['<benchmark>']) as f:\n benchmark_config = json.loads(f.read())\n generate_agent_configs(benchmark_config)\n experiments = product(benchmark_config['environments'], benchmark_config['agents'], [options])\n\n # Run evaluations\n with Pool(processes=int(options['--processes'])) as pool:\n results = pool.starmap(evaluate, experiments)\n\n # Clean temporary config files\n generate_agent_configs(benchmark_config, clean=True)\n\n # Write evaluations summary\n benchmark_filename = os.path.join(Evaluation.OUTPUT_FOLDER, '{}_{}.{}.json'.format(\n BENCHMARK_FILE, datetime.datetime.now().strftime('%Y%m%d-%H%M%S'), os.getpid()))\n with open(benchmark_filename, 'w') as f:\n json.dump(results, f, sort_keys=True, indent=4)\n gym.logger.info('Benchmark done. Summary written in: {}'.format(benchmark_filename))","def plotDifferentSettings():\n\t#save location of the results\n\tresloc = 'Different_settings_results'\n\n\t#the numrows limits\n\tnrows_lims = [1e4, 1e7]\n\tnbits = 30\n\t\n\t#string for the file names of the to be saved files\n\tsettingsstr = 'nrows={:.0e}--{:.0e}_nbits={}'.format(nrows_lims[0], nrows_lims[1], nbits)\n\t\n\t#the relative approximation error for the different counting algorithms\n\tll_RAE = []\n\tprob_RAE = []\n\tcomb_RAE = []\n\t#the runtime for the different algorithms\n\ttc_runtime = []\n\tll_runtime = []\n\tprob_runtime = []\n\tcomb_runtime = []\n\n\t#the different settings we want to test\n\tnumrows = np.linspace(nrows_lims[0], nrows_lims[1], num = 15, dtype = int)\n\tnumbits = np.array([nbits])\n\t\n\tlooplength = len(numrows)\n\t\n\ttry:\n\t\t(ll_RAE, prob_RAE, comb_RAE, tc_runtime, ll_runtime, prob_runtime, comb_runtime) = np.loadtxt('./{0}/diffset_results_{1}.txt'.format(resloc, settingsstr))\n\texcept:\n\t\tfor i in np.arange(len(numrows)):\n\t\t\toF.progress(i, looplength)\n\t\t\tfor j in np.arange(len(numbits)):\n\t\t\t\tresults = runCounts(numrows[i], numbits[j], doprints = False)\n\t\t\t\n\t\t\t\tll_RAE = np.append(ll_RAE, results[0])\n\t\t\t\tprob_RAE = np.append(prob_RAE, results[1])\n\t\t\t\tcomb_RAE = np.append(comb_RAE, results[2])\n\t\t\t\n\t\t\t\ttc_runtime = np.append(tc_runtime, results[3])\n\t\t\t\tll_runtime = np.append(ll_runtime, results[4])\n\t\t\t\tprob_runtime = np.append(prob_runtime, results[5])\n\t\t\t\tcomb_runtime = np.append(comb_runtime, results[6])\n\t\t\t\n\t\tnp.savetxt('./{0}/diffset_results_{1}.txt'.format(resloc, settingsstr), \n\t\t\tnp.array([ll_RAE, prob_RAE, comb_RAE, tc_runtime, ll_runtime, prob_runtime, comb_runtime]), \n\t\t\theader = '#ll_RAE, prob_RAE, comb_RAE, tc_runtime, ll_runtime, prob_runtime, comb_runtime')\n\t\n\tplotTwoValues(numrows, ll_RAE, ll_runtime, 'Number of rows', 'RAE [\\%]', 'Runtime [s]', 'RAE and runtime of loglog count for different number of rows. \\nNumbits = {}'.format(nbits), 'RAEandRuntime_loglog_{0}.pdf'.format(settingsstr))\n\t\n\tplotTwoValues(numrows, prob_RAE, prob_runtime, 'Number of rows', 'RAE [\\%]', 'Runtime [s]', 'RAE and runtime of probabilisic count for different \\nnumber of rows. Numbits = {}'.format(nbits), 'RAEandRuntime_prob_{0}.pdf'.format(settingsstr))","def SkoptPaperStats(maxIters, numRuns):\n\n comm = MPI.COMM_WORLD\n rank = comm.Get_rank()\n\n assert comm.Get_size() == numRuns, \"Please ensure there is one process running per run i.e \" + str(numRuns) + \" processes.\"\n \n # Define the problem bounds.\n skoptBounds = [(10, 1300), (40, 230), (0, 90), (0, 90)]\n\n # Use the same seed list as previously.\n seedList = [572505, 357073, 584216, 604873, 854690, 573165, 298975, 650770, 243921, 191168]\n\n # The target for each algorithm. This was determined by using the values in the literature, so there is clearly some deviation either due to the detuning or computation.\n globalFoM = 1.033\n\n if rank == 0:\n timeList = []\n iterationList = []\n\n # Define which solver will be used.\n optimiser = skopt.Optimizer(skoptBounds, base_estimator = \"GP\", n_initial_points = int(np.ceil(maxIters/10)), random_state = seedList[rank])\n\n # Start timing.\n startTime = time.time()\n timeElapsed = None\n iterationSuccess = None\n\n # Start optimisation.\n for iteration in range(maxIters):\n\n # Make one suggestion.\n nextParams = optimiser.ask()\n\n # Check what FoM this gives. Go negative as this is a minimisation routine.\n fEval = FitnessSkopt(nextParams)\n\n # Update best FoM.\n if abs(fEval) >= globalFoM:\n # The algorithm has managed to surpass or equal the paper value.\n iterationSuccess = iteration\n timeElapsed = time.time() - startTime\n \n if rank == 0:\n iterationList.append(iterationSuccess)\n timeList.append(timeElapsed)\n\n break\n \n # Tell the optimiser about the result.\n optimiser.tell(nextParams, fEval)\n\n # Run complete. Send results to main process. Tags are unique identifiers.\n if rank != 0:\n comm.send(timeElapsed, dest = 0, tag = 1)\n comm.send(iterationSuccess, dest = 0, tag = 2)\n\n # Wait for all the processes to end.\n comm.Barrier()\n\n if rank == 0:\n # Aggregate the data.\n for process in range(comm.Get_size() - 1):\n # Get the data.\n individualTime = None\n individualTime = comm.recv(individualTime, source = process + 1, tag = 1)\n\n individualIter = None\n individualIter = comm.recv(individualIter, source = process + 1, tag = 2)\n\n if individualIter is not None:\n # Both values must therefore be non-null.\n iterationList.append(individualIter)\n timeList.append(individualTime)\n\n avgRuntime = np.average(timeList)\n avgIters = np.average(iterationList)\n try:\n\n fastestTime = np.min(timeList)\n\n except ValueError:\n \n # List is empty.\n fastestTime = float('NaN')\n\n numSuccess = len(iterationList)\n successRate = numSuccess/numRuns\n\n print(\"Bayesian optimisation paper testing complete! Here are the stats:\")\n print(\"Number of successful runs: \" + str(numSuccess) + \" (Success rate of \" + str(successRate) + \")\")\n print(\"Average iterations required for success: \" + str(avgIters))\n print(\"Average time required for success: \" + str(avgRuntime))\n print(\"Fastest convergence time: \" + str(fastestTime))\n print(\"------------------------------------------------------------------------------------------------------------------\")\n \n return","def run_training_and_tests(test_name, dataset, models, n_images = 1000, training_split = 0.7, \n n_training_images = None, n_test_images = None, \n n_iterations = 1, dimensions = (50, 50)):\n\n aggregate_metrics = {}\n\n # Run specified number of iterations\n for i in range(n_iterations):\n print(\"\\nTest iteration\", i+1)\n\n # Handle if specific training and test set size isn't given\n if (n_training_images is None):\n n_training_images = n_images * training_split\n if (n_test_images is None):\n n_test_images = n_images * (1-training_split)\n\n # Load training and test sets from single dataset\n train_data, train_labels, test_data, test_labels = image_utils.read_dataset(\n n_training_images, \n n_test_images, \n './datasets/' + dataset, \n dimensions[0], \n dimensions[1]\n )\n\n # Train and run tests for each model\n for model in models:\n print(\"Working with model '\" + model.label + \"'\")\n\n # Train model\n start = time.time()\n model.train(copy.deepcopy(train_data), train_labels)\n end = time.time()\n training_time = round(end - start, 3)\n\n # Run predictions on test set\n start = time.time()\n predicted = model.run(copy.deepcopy(test_data))\n end = time.time()\n test_time = round(end - start, 3)\n\n # Calculate metrics and store for aggregate calculations\n metrics = Metrics(test_labels, predicted, training_time, test_time)\n if model.label in aggregate_metrics:\n aggregate_metrics[model.label].append(metrics)\n else:\n aggregate_metrics[model.label] = [metrics]\n\n # Print results\n print(\"Results\\n\" + \"------\")\n print(str(metrics))\n\n # Save model\n filepath = \"./test/\" + test_name + \"/\" + model.label + \"/iteration\" + str(i+1) + \"/\"\n print(\"Saving model to '\" + filepath + model.label + \".joblib'\")\n os.makedirs(os.path.dirname(filepath), exist_ok = True)\n with open(filepath + model.label + '.joblib', 'wb') as file:\n dump(model, file)\n\n # Save results\n print(\"Saving results to '\" + filepath + \"results.txt'\\n\")\n with open(filepath + \"results.txt\", 'w') as file:\n file.write(str(metrics))\n\n # Calculate, print and write aggregate metrics\n print(\n 'Aggregate Results' + '\\n' +\n '-----------------'\n )\n for model in models:\n aggregate = combine_metrics(aggregate_metrics[model.label])\n print(model.label)\n print(aggregate)\n filepath = \"./test/\" + test_name + \"/\" + model.label + \"/\"\n print(\"Saving results to '\" + filepath + \"aggregate_results.txt'\" + \"\\n -- -\\n\")\n with open(filepath + \"aggregate_results.txt\", 'w') as file:\n file.write(str(aggregate))"],"string":"[\n \"def run_experiments():\\n if True: # Change to False when done finding max_scoring_num_rolls\\n six_sided_max = max_scoring_num_rolls(six_sided)\\n print('Max scoring num rolls for six-sided dice:', six_sided_max)\\n rerolled_max = max_scoring_num_rolls(reroll(six_sided))\\n print('Max scoring num rolls for re-rolled dice:', rerolled_max)\\n\\n if False: # Change to True to test always_roll(8)\\n print('always_roll(8) win rate:', average_win_rate(always_roll(8)))\\n\\n if False: # Change to True to test bacon_strategy\\n print('bacon_strategy win rate:', average_win_rate(bacon_strategy))\\n\\n if False: # Change to True to test swap_strategy\\n print('swap_strategy win rate:', average_win_rate(swap_strategy))\\n\\n \\\"*** You may add additional experiments as you wish ***\\\"\",\n \"def run_experiments():\\n if True: # Change to False when done finding max_scoring_num_rolls\\n six_sided_max = max_scoring_num_rolls(six_sided)\\n print('Max scoring num rolls for six-sided dice:', six_sided_max)\\n four_sided_max = max_scoring_num_rolls(four_sided)\\n print('Max scoring num rolls for four-sided dice:', four_sided_max)\\n\\n if False: # Change to True to test always_roll(8)\\n print('always_roll(8) win rate:', average_win_rate(always_roll(8)))\\n\\n if True: # Change to True to test bacon_strategy\\n print('bacon_strategy win rate:', average_win_rate(bacon_strategy))\\n\\n if True: # Change to True to test swap_strategy\\n print('swap_strategy win rate:', average_win_rate(swap_strategy))\\n\\n if True:\\n print('final_strategy win rate:', average_win_rate(final_strategy))\\n\\n \\\"*** You may add additional experiments as you wish ***\\\"\",\n \"def run_experiments():\\n if False: # Change to False when done finding max_scoring_num_rolls\\n six_sided_max = max_scoring_num_rolls(six_sided)\\n print('Max scoring num rolls for six-sided dice:', six_sided_max)\\n\\n if False: # Change to True to test always_roll(8)\\n print('always_roll(8) win rate:', average_win_rate(always_roll(8)))\\n\\n if False: # Change to True to test bacon_strategy\\n print('bacon_strategy win rate:', average_win_rate(bacon_strategy))\\n\\n if False: # Change to True to test swap_strategy\\n print('swap_strategy win rate:', average_win_rate(swap_strategy))\\n\\n if True: # Change to True to test final_strategy\\n print('final_strategy win rate:', average_win_rate(final_strategy))\\n\\n \\\"*** You may add additional experiments as you wish ***\\\"\",\n \"def all_experiments():\\n elo_explain_experiments()\\n alpha_beta_experiments()\\n mtcs_experiments()\",\n \"def run_experiments():\\r\\n result = eval_strategy_range(always_roll, 1, 10)\\r\\n print('Best always_roll strategy:', result)\\r\\n\\r\\n if True: # Change to True when ready to test make_comeback_strategy\\r\\n result = eval_strategy_range(make_comeback_strategy, 5, 15)\\r\\n print('Best comeback strategy:', result)\\r\\n\\r\\n if True: # Change to True when ready to test make_mean_strategy\\r\\n result = eval_strategy_range(make_mean_strategy, 1, 10)\\r\\n print('Best mean strategy:', result)\\r\\n\\r\\n \\\"*** You may add additional experiments here if you wish ***\\\"\",\n \"def main():\\n run_test_summary1a()\\n run_test_summary1c()\\n run_test_summary1c()\",\n \"def run_experiments():\\n nets = []\\n nets.extend(shallow(n=1))\\n nets.extend(basic_conv(n=1))\\n nets.extend(omit_FC(n=1))\\n nets.extend(dbl_conv(n=1, activation_fn=sigmoid))\\n # omitted, but still interesting: regularized_dbl_conv()\\n nets.extend(dbl_conv_relu(n=1))\\n nets.extend(expanded_data(n=1, FC_n=100))\\n nets.extend(expanded_data(n=1, FC_n=300))\\n nets.extend(expanded_data(n=1, FC_n=1000))\\n nets.extend(expanded_data_double_fc(n=1, FC_n=100)) \\n nets.extend(expanded_data_double_fc(n=1, FC_n=300))\\n nets.extend(expanded_data_double_fc(n=1, FC_n=1000))\\n nets.extend(double_fc_dropout(0.5, 0.5, 0.5, n=5))\\n # plot the erroneous digits in the ensemble of nets just trained\\n error_locations, erroneous_predictions = ensemble(nets)\\n plt = plot_errors(error_locations, erroneous_predictions)\\n plt.savefig(\\\"ensemble_errors.png\\\")\\n # plot the filters learned by the first of the nets just trained\\n plt = plot_filters(nets[0], 0, 5, 4)\\n plt.savefig(\\\"net_full_layer_0.png\\\")\\n plt = plot_filters(nets[0], 1, 8, 5)\\n plt.savefig(\\\"net_full_layer_1.png\\\")\",\n \"def run_multiple_simple_experiments(self, path_to_csv_files, algorithms, thresholds):\\n starting_time = time.time()\\n print(\\\"started simple precision recall image_matching_module\\\")\\n for algorithm in algorithms:\\n start_file_time = time.time()\\n print(\\\"started algorithm:\\\", algorithm.get_algorithm_name())\\n self.run_single_simple_experiment(path_to_csv_files, algorithm, thresholds)\\n finish_file_time = time.time()\\n print(\\\"finished algorithm:\\\", algorithm.get_algorithm_name())\\n print(\\\"time took:\\\", (finish_file_time - start_file_time) / 60, \\\"minutes\\\")\\n finish_time = time.time()\\n hours = int((finish_time - starting_time) // 3600)\\n minutes = (finish_time - starting_time) / 60\\n print(\\\"finished simple precision recall image_matching_module\\\")\\n print(\\\"time image_matching_module took:\\\", hours, \\\"hours and\\\", minutes, \\\"minutes\\\")\",\n \"def run_experiments(size, approach):\\n\\n # Create folder (if needed) where to store query evaluation logs and raw results\\n if not os.path.isdir(f\\\"{path_reports}/eval-query-{current_run}\\\"):\\n shell(f\\\"mkdir -p {path_reports}/eval-query-{current_run}\\\")\\n\\n # Create folder (if needed) where to store store evaluation logs and raw results\\n if not os.path.isdir(f\\\"{path_reports}/eval-store-{current_run}\\\"):\\n shell(f\\\"mkdir -p {path_reports}/eval-store-{current_run}\\\")\\n\\n # Helper function computing the path of a file/folder for the given query evaluation iteration \\n def query_path(iteration, filename=None):\\n folder = size[\\\"id\\\"] + \\\"_\\\" + str(iteration)\\n return path_reports + \\\"/eval-query-\\\" + current_run + \\\"/\\\" + folder + (\\\"/\\\" + filename if filename != None else \\\"\\\") \\n\\n # Helper function computing the path of a file/folder for the given store evaluation test / cpu setting\\n def store_path(test, cpu, filename=None):\\n folder = size[\\\"id\\\"] + \\\"_\\\" + approach[\\\"id\\\"] + \\\"_\\\" + test + (\\\"_\\\" + cpu[\\\"id\\\"] if cpu != None else \\\"\\\")\\n return path_reports + \\\"/eval-store-\\\" + current_run + \\\"/\\\" + folder + (\\\"/\\\" + filename if filename != None else \\\"\\\")\\n \\n # Determine whether partial traces and named graphs are supported\\n partial = approach[\\\"supports_partial\\\"]\\n graphs = approach[\\\"supports_graphs\\\"]\\n \\n # Skip setting if all data is available (check for presence of log files - delete them to repeat test)\\n may_skip = (not sp_enable or not partial or not graphs or os.path.isfile(store_path(\\\"sp\\\", None, \\\"eval.log\\\")))\\n may_skip = may_skip and (not sf_enable or os.path.isfile(store_path(\\\"sf\\\", None, \\\"eval.log\\\")))\\n if query_enable and approach[\\\"id\\\"] == query_approach_id:\\n for i in range(0, query_num_iterations):\\n may_skip = may_skip and os.path.isfile(query_path(i, \\\"eval.log\\\"))\\n for cpu in cpus:\\n may_skip = may_skip and (not pp_enable or not partial or not graphs or os.path.isfile(store_path(\\\"pp\\\", cpu, \\\"eval.log\\\")))\\n may_skip = may_skip and (not pf_enable or os.path.isfile(store_path(\\\"pf\\\", cpu, \\\"eval.log\\\")))\\n if may_skip:\\n return\\n\\n # Delete (if needed) and extract again the repository from its .tar.xz file, so to work on a clean repository (at the end of this test, the repository is no more clean)\\n path_dump = path_data + \\\"/\\\" + size[\\\"id\\\"] + \\\"_\\\" + approach[\\\"id\\\"] + \\\".tar.lz\\\"\\n path_repo = path_repos + \\\"/\\\" + size[\\\"id\\\"] + \\\"_\\\" + approach[\\\"id\\\"]\\n if not os.path.isfile(path_dump):\\n log(f\\\"Missing required file {path_dump}\\\")\\n sys.exit()\\n if os.path.isdir(path_repo):\\n shell(f\\\"rm -Rf {path_repo}\\\")\\n shell(f\\\"{cmd_plzip} -kdc {path_dump} | tar xf - -C {path_repos}\\\")\\n\\n # Locate the repository URL\\n repo_url = f\\\"http://localhost:{server_port}/repositories/promo\\\"\\n \\n # Query test (if enabled)\\n if query_enable and approach[\\\"id\\\"] == query_approach_id:\\n for i in range(0, query_num_iterations):\\n if not os.path.isfile(query_path(i, \\\"eval.log\\\")):\\n shell(f\\\"mkdir -p {query_path(i)}\\\")\\n shell(f\\\"taskset -a {query_taskset} {cmd_graphdb} start {path_repo}\\\")\\n shell(f\\\"taskset -a {query_taskset} {cmd_mokip} eval-query -w -u {repo_url} -q {path_queries} -r {query_path(i)} \\\" +\\n f\\\"| tee {query_path(i, 'eval.log')}\\\")\\n shell(f\\\"taskset -a {query_taskset} {cmd_graphdb} stop {path_repo}\\\")\\n\\n # Sequential Partial test (to assess store times per trace and their components)\\n if sp_enable and partial and graphs and not os.path.isfile(store_path(\\\"sp\\\", None, \\\"eval.log\\\")):\\n shell(f\\\"mkdir -p {store_path('sp', None)}\\\")\\n shell(f\\\"taskset -a {sp_taskset} {cmd_graphdb} start {path_repo}\\\")\\n shell(f\\\"taskset -a {sp_taskset} {cmd_mokip} eval-store -d {path_data}/traces_pp.jsonl.gz \\\" + \\n f\\\"-u {repo_url} -i {approach['inference']} -U REPLACE_GRAPH_PROTOCOL \\\" + \\n f\\\"-o {path_ontology} --trace-namespace '{namespace_trace}' --graph-namespace '{namespace_graph}' \\\" +\\n f\\\"-T {timeout} -r {store_path('sp', None)} -t 1 -w 50 -p -D \\\" + \\n f\\\"| tee {store_path('sp', None, 'eval.log')}\\\")\\n shell(f\\\"taskset -a {sp_taskset} {cmd_graphdb} stop {path_repo}\\\")\\n\\n # Parallel Partial (to assess throughput, varying # of CPU cores)\\n for cpu in cpus:\\n if pp_enable and partial and graphs and not os.path.isfile(store_path(\\\"pp\\\", cpu, \\\"eval.log\\\")):\\n shell(f\\\"mkdir -p {store_path('pp', cpu)}\\\")\\n shell(f\\\"taskset -a {cpu['taskset']} {cmd_graphdb} start {path_repo}\\\")\\n shell(f\\\"taskset -a {cpu['taskset']} {cmd_mokip} eval-store -d {path_data}/traces_pp.jsonl.gz \\\" +\\n f\\\"-u {repo_url} -i {approach['inference']} -U REPLACE_GRAPH_PROTOCOL \\\" + \\n f\\\"-o {path_ontology} --trace-namespace '{namespace_trace}' --graph-namespace '{namespace_graph}' \\\" + \\n f\\\"-T {timeout} -r {store_path('pp', cpu)} -t {max(2, cpu['num_threads'])} -w 50 -p -D \\\" + \\n f\\\"| tee {store_path('pp', cpu, 'eval.log')}\\\")\\n shell(f\\\"taskset -a {cpu['taskset']} {cmd_graphdb} stop {path_repo}\\\")\\n\\n # Sequential Full test (to assess store times per trace and their components)\\n if sf_enable and not os.path.isfile(store_path(\\\"sf\\\", None, \\\"eval.log\\\")):\\n shell(f\\\"mkdir -p {store_path('sf', None)}\\\")\\n shell(f\\\"taskset -a {sf_taskset} {cmd_graphdb} start {path_repo}\\\")\\n shell(f\\\"taskset -a {sf_taskset} {cmd_mokip} eval-store -d {path_data}/traces_sf.jsonl.gz \\\" + \\n f\\\"-u {repo_url} -i {approach['inference']} -U {'APPEND' if graphs else 'APPEND_DEFAULT_GRAPH'} \\\" +\\n f\\\"-o {path_ontology} --trace-namespace '{namespace_trace}' --graph-namespace '{namespace_graph}' \\\" +\\n f\\\"-T {timeout} -r {store_path('sf', None)} -t 1 -w 50 \\\" + \\n f\\\"| tee {store_path('sf', None, 'eval.log')}\\\")\\n shell(f\\\"taskset -a {sf_taskset} {cmd_graphdb} stop {path_repo}\\\")\\n\\n # Parallel Full (to assess throughput where data is also deleted, varying # of CPU cores)\\n for cpu in cpus:\\n if pf_enable and not os.path.isfile(store_path(\\\"pf\\\", cpu, \\\"eval.log\\\")):\\n update = \\\"APPEND\\\" if graphs else \\\"APPEND_DEFAULT_GRAPH\\\"\\n shell(f\\\"mkdir -p {store_path('pf', cpu)}\\\")\\n shell(f\\\"taskset -a {cpu['taskset']} {cmd_graphdb} start {path_repo}\\\")\\n shell(f\\\"taskset -a {cpu['taskset']} {cmd_mokip} eval-store -d {path_data}/traces_pf_{cpu['id']}.jsonl.gz \\\" +\\n f\\\"-u {repo_url} -i {approach['inference']} -U {'APPEND' if graphs else 'APPEND_DEFAULT_GRAPH'} \\\" + \\n f\\\"-o {path_ontology} --trace-namespace '{namespace_trace}' --graph-namespace '{namespace_graph}' \\\" +\\n f\\\"-T {timeout} -r {store_path('pf', cpu)} -t {max(2, cpu['num_threads'])} -w 50 \\\" + \\n f\\\"| tee {store_path('pf', cpu, 'eval.log')}\\\")\\n shell(f\\\"taskset -a {cpu['taskset']} {cmd_graphdb} stop {path_repo}\\\")\\n\\n # Drop the repository (both to save space and since it is not clean anymore)\\n shell(f\\\"rm -Rf {path_repo}\\\")\",\n \"def three_experiments(two_experiments, one_experiment):\",\n \"def run_Experiment(DP = None, QL = None):\\n\\n # Path information\\n output_path, exp_num = create_new_dir() #dirs Exp/1, Exp/2, ...\\n DP_path = join(output_path,'DP') #dirs Exp/1/DP\\n QL_path = join(output_path,'QL') #dirs Exp/1/QL\\n print(\\\"************ Exp \\\", exp_num, \\\"************ \\\\n\\\")\\n\\n # Exp_summary_data\\n method = get_method_str(DP, QL)\\n exp_summary = [str(exp_num), method]\\n\\n\\n # Run DP\\n if DP != None:\\n print(\\\"In Runner: Executing DP !!\\\")\\n\\n prob_file = DP[0]\\n createFolder(DP_path)\\n # output_params = [V_so, mean, variance, bad_count]\\n output_params = run_DP(setup_grid_params, prob_file, output_file, DP_path, threshold = threshold)\\n\\n \\\"\\\"\\\"CHANGE ARGUMENT if return order of setup_grid() is changed\\\"\\\"\\\"\\n input_params = setup_grid_params[9].copy()\\n input_params.append(prob_file)\\n\\n exp_summary = append_params_to_summary(exp_summary, input_params, output_params)\\n append_summary_to_summaryFile('Experiments/Exp_summary.csv', exp_summary)\\n print(\\\"In Runner: Executing DP Finished!!\\\")\\n\\n # Run QL\\n if QL != None:\\n print(\\\"In Runner: Executing QL !!\\\")\\n\\n QL_params = QL\\n createFolder(QL_path)\\n output_parameters_all_cases = run_QL(setup_grid_params, QL_params, QL_path, exp_num)\\n # run_QL(setup_grid_params, QL_params, QL_path)\\n\\n print(\\\"In Runner: Executing QL Finished !!\\\")\",\n \"def experiment1_outliers():\\n\\tdata_folder = \\\"ckan_subset/prepared_learnset/\\\"\\n\\ttest_folder = 'ckan_subset/testset/xml_csv/'\\n\\tgm = Graph_Maker()\\n\\tgm.store()\\n\\trounds = 5\\n\\tx = [\\\"Fingerprint\\\", \\\"Syntax Feature Model\\\", \\\"Word2Vec Matcher\\\"]\\n\\t\\n\\tnumber_of_classes = 15\\n\\texamples_per_class = 0\\n\\taccuracies = []\\n\\tprecisions = []\\n\\trecalls = []\\n\\tfmeasures = []\\n\\tsf_main = Storage_Files(data_folder, classes)\\n\\ttmp_acc = []\\n\\ttmp_prec = []\\n\\ttmp_rec = []\\n\\ttmp_fmeasure = []\\n\\ttotal_actual = []\\n\\ttotal_predicted = []\\n\\n\\tfor i in range(0, rounds):\\n\\t\\tprint(\\\"Fingerprint\\\")\\n\\t\\t# --- Fingerprint\\n\\t\\tccc = Column_Classification_Config()\\n\\t\\tccc.add_feature('feature_main', 'Fingerprint', [sf_main, number_of_classes, examples_per_class, False, False])\\n\\n\\t\\tccc.add_matcher('matcher', 'Fingerprint_Matcher', {'feature_main': 'fingerprint'}) # main classifier\\n\\t\\tsm = Schema_Matcher(ccc)\\n\\t\\tactual, predicted = execute_test_ckan(sm, test_folder, False)\\n\\t\\ttotal_actual += actual\\n\\t\\ttotal_predicted += predicted\\n\\t\\taccuracy = accuracy_score(actual, predicted)\\n\\t\\ttmp_acc.append(accuracy)\\n\\t\\ttmp_prec.append(precision(actual, predicted))\\n\\t\\ttmp_rec.append(recall(actual, predicted))\\n\\t\\ttmp_fmeasure.append(f_measure(actual, predicted))\\n\\n\\taccuracies.append( round(sum(tmp_acc) / float(rounds), 2) )\\n\\tprecisions.append( round(sum(tmp_prec) / float(rounds), 2) )\\n\\trecalls.append( round(sum(tmp_rec) / float(rounds), 2) )\\n\\tfmeasures.append(round(sum(tmp_fmeasure) / float(rounds), 2))\\n\\tclassnames = list(set(get_class_names(total_actual) + get_class_names(total_predicted)))\\n\\tcm = confusion_matrix(total_actual, total_predicted, labels=classnames)\\n\\t#gm.plot_confusion_matrix(cm, classnames, normalize=True)\\n\\t\\n\\ttmp_acc = []\\n\\ttmp_prec = []\\n\\ttmp_rec = []\\n\\ttmp_fmeasure = []\\n\\ttotal_actual = []\\n\\ttotal_predicted = []\\n\\tfor i in range(0, rounds):\\n\\t\\tprint(\\\"SFM\\\")\\n\\t\\t# --- Syntax Feature Model\\n\\t\\tccc = Column_Classification_Config()\\n\\t\\tccc.add_feature('feature_main', 'Syntax_Feature_Model', [sf_main, 1, 0, False, False])\\n\\n\\t\\tccc.add_matcher('matcher', 'Syntax_Matcher', {'feature_main': 'syntax'}) # main classifier\\n\\t\\tsm = Schema_Matcher(ccc)\\n\\t\\tactual, predicted = execute_test_ckan(sm, test_folder, False)\\n\\t\\ttotal_actual += actual\\n\\t\\ttotal_predicted += predicted\\n\\t\\taccuracy = accuracy_score(actual, predicted)\\n\\t\\ttmp_acc.append(accuracy)\\n\\t\\ttmp_prec.append(precision(actual, predicted))\\n\\t\\ttmp_rec.append(recall(actual, predicted))\\n\\t\\ttmp_fmeasure.append(f_measure(actual, predicted))\\n\\n\\taccuracies.append( round(sum(tmp_acc) / float(rounds), 2) )\\n\\tprecisions.append( round(sum(tmp_prec) / float(rounds), 2) )\\n\\trecalls.append( round(sum(tmp_rec) / float(rounds), 2) )\\n\\tfmeasures.append(round(sum(tmp_fmeasure) / float(rounds), 2))\\n\\tclassnames = list(set(get_class_names(total_actual) + get_class_names(total_predicted)))\\n\\tcm = confusion_matrix(total_actual, total_predicted, labels=classnames)\\n\\t#gm.plot_confusion_matrix(cm, classnames, normalize=True)\\n\\n\\ttmp_acc = []\\n\\ttmp_prec = []\\n\\ttmp_rec = []\\n\\ttmp_fmeasure = []\\n\\ttotal_actual = []\\n\\ttotal_predicted = []\\n\\tfor i in range(0, rounds):\\n\\t\\tprint(\\\"W2V\\\")\\n\\t\\t# --- Word2Vec Matcher\\n\\t\\tccc = Column_Classification_Config()\\n\\t\\tccc.add_feature('feature_main', 'Corpus', [sf_main, number_of_classes, examples_per_class, False, False])\\n\\n\\t\\tccc.add_matcher('matcher', 'Word2Vec_Matcher', {'feature_main': 'corpus'}) # main classifier\\n\\t\\tsm = Schema_Matcher(ccc)\\n\\t\\tactual, predicted = execute_test_ckan(sm, test_folder, False)\\n\\t\\ttotal_actual += actual\\n\\t\\ttotal_predicted += predicted\\n\\t\\taccuracy = accuracy_score(actual, predicted)\\n\\t\\ttmp_acc.append(accuracy)\\n\\t\\ttmp_prec.append(precision(actual, predicted))\\n\\t\\ttmp_rec.append(recall(actual, predicted))\\n\\t\\ttmp_fmeasure.append(f_measure(actual, predicted))\\n\\n\\taccuracies.append( round(sum(tmp_acc) / float(rounds), 2) )\\n\\tprecisions.append( round(sum(tmp_prec) / float(rounds), 2) )\\n\\trecalls.append( round(sum(tmp_rec) / float(rounds), 2) )\\n\\tfmeasures.append(round(sum(tmp_fmeasure) / float(rounds), 2))\\n\\tclassnames = list(set(get_class_names(total_actual) + get_class_names(total_predicted)))\\n\\tcm = confusion_matrix(total_actual, total_predicted, labels=classnames)\\n\\t#gm.plot_confusion_matrix(cm, classnames, normalize=True)\\n\\n\\tgm.add_x(x)\\n\\t# accuracies = [0.4, 0.4, 0.4]\\n\\t# precisions = [0.5, 0.5, 0.5]\\n\\t# recalls = [0.62, 0.62, 0.62]\\n\\t# fmeasures = [0.23, 0.23, 0.28]\\n\\tgm.append_y(accuracies)\\n\\tgm.append_y(precisions)\\n\\tgm.append_y(recalls)\\n\\tgm.append_y(fmeasures)\\n\\tgm.store()\\n\\tsubtitle = \\\"Scores were averaged over \\\" + str(rounds) + \\\" tests with \\\" + str(len(classes)) + \\\" classes. \\\" + \\\\\\n\\t\\\"Number of simulated columns per class: \\\" + str(number_of_classes)\\n\\tlabels = [\\\"Accuracy\\\", \\\"Precision\\\", \\\"Recall\\\", \\\"F-Measure\\\"]\\n\\tgm.plot_bar_n(\\\"Matcher Type\\\", \\\"Score\\\", \\\"Accuracy of Matchers\\\", labels, subtitle=subtitle)\",\n \"def run_and_evaluate():\\n tsp_problems = read_all_problems()\\n # Empty list of metrics\\n results = []\\n for problem in tqdm.tqdm(tsp_problems):\\n # As random factors are involved repeat experiments a couple of times\\n best_routes_base = []\\n best_routes_af = []\\n best_routes_ms = []\\n base_times = []\\n af_times = []\\n ms_times = []\\n for i in range(10):\\n # Base solution\\n start_time = timeit.default_timer()\\n best_route_base = solve_tsp_basic(problem)\\n base_time = timeit.default_timer() - start_time\\n best_routes_base.append(Fitness(route=best_route_base).route_distance())\\n base_times.append(base_time)\\n\\n # AF clustering solution\\n start_time = timeit.default_timer()\\n best_route_af = solve_tsp_affinity_propagation(problem)\\n af_time = timeit.default_timer() - start_time\\n best_routes_af.append(Fitness(route=best_route_af).route_distance())\\n af_times.append(af_time)\\n\\n # MS solution\\n start_time = timeit.default_timer()\\n best_route_ms = solve_mean_shift(problem)\\n ms_time = timeit.default_timer() - start_time\\n best_routes_ms.append(Fitness(route=best_route_ms).route_distance())\\n ms_times.append(ms_time)\\n\\n results.append(\\n {\\n \\\"problem name\\\": problem.name,\\n \\\"optimal solution\\\": find_route_optimal_route_length(problem),\\n \\\"baseline tour length\\\": mean(best_routes_base),\\n \\\"af clustering tour length\\\": mean(best_routes_af),\\n \\\"ms clustering tour length\\\": mean(best_routes_ms),\\n \\\"baseline algorithm time\\\": mean(base_times),\\n \\\"af clustering algorithm time\\\": mean(af_times),\\n \\\"ms clustering algorithm time\\\": mean(ms_times),\\n }\\n )\\n # Create dataframe and safe results\\n df = pd.DataFrame(results)\\n df.to_csv(\\\"results.csv\\\", index=False)\\n return df\",\n \"def main():\\r\\n algos = [merge_sort, quick_sort, heap_sort, radix_sort, bucket_sort_general]\\r\\n array_sizes = [5000, 10000, 15000, 20000, 50000, 75000, 100000, 150000]\\r\\n results = {algo.__name__: [] for algo in algos}\\r\\n for algo in algos:\\r\\n result = []\\r\\n for size in array_sizes:\\r\\n time = test(algo, size)\\r\\n result.append(time)\\r\\n results[algo.__name__] = result\\r\\n\\r\\n display_results(results, array_sizes)\",\n \"def three_experiments_with_trials(family_with_trials, single_with_trials):\",\n \"def evaluate_model(args, eval_runs, warm_runs, metrics=['psnr', 'ssim', 'fps']):\\n upsampler = Upsampler(args)\\n if warm_runs > 0:\\n print(\\\"Warming up for evaluation\\\")\\n for i in range(warm_runs):\\n print(\\\"Performing warm-up run\\\", str(i+1))\\n for sequence in ['foliage', 'walk', 'calendar', 'city']:\\n bix_dir = os.path.join(VID4_DIR, 'BIx4', sequence)\\n upsampler.run_dir(bix_dir, reset=False)\\n \\n time = 0.\\n psnrs = []\\n ssims = []\\n for i in range(eval_runs):\\n run_psnrs = []\\n run_ssims = []\\n print(\\\"Performing evaluation run\\\", str(i+1))\\n for sequence in ['foliage', 'walk', 'calendar', 'city']:\\n bix_dir = os.path.join(VID4_DIR, 'BIx4', sequence)\\n gt_dir = os.path.join(VID4_DIR, 'GT', sequence)\\n print(\\\"Evaluating on\\\", bix_dir)\\n time += upsampler.run_dir(bix_dir, reset=False)\\n vid_psnrs, vid_ssims = _eval_sr_perf(os.path.join(bix_dir, 'up'), gt_dir)\\n run_psnrs += vid_psnrs\\n run_ssims += vid_ssims\\n if i == eval_runs-1:\\n with open(os.path.join(upsampler.get_model_dir(), \\\"psnr.txt\\\"), \\\"w\\\") as f:\\n f.writelines(str(psnr) + '\\\\n' for psnr in run_psnrs)\\n with open(os.path.join(upsampler.get_model_dir(), \\\"ssim.txt\\\"), \\\"w\\\") as f:\\n f.writelines(str(ssim) + '\\\\n' for ssim in run_ssims)\\n psnrs += run_psnrs\\n ssims += run_ssims\\n\\n fps = VID4_LENGTH/ (time/eval_runs)\\n return Performance(psnr=psnrs, ssim=ssims, fps=fps)\",\n \"def run_analyses(y_predict_train, y_train, y_predict, y_test):\\n # calculate metrics\\n _, training_error = output_error(y_predict_train, y_train)\\n (precision, recall, f1, _), testing_error = output_error(y_predict, y_test)\\n \\n # print out metrics\\n print 'Average Precision:', np.average(precision)\\n print 'Average Recall:', np.average(recall)\\n print 'Average F1:', np.average(f1)\\n print 'Training Error:', training_error\\n print 'Testing Error:', testing_error\",\n \"def main():\\n run_simulation(spectral=False, ml=False, num_procs=1)\\n run_simulation(spectral=True, ml=False, num_procs=1)\\n run_simulation(spectral=False, ml=True, num_procs=1)\\n run_simulation(spectral=True, ml=True, num_procs=1)\\n run_simulation(spectral=False, ml=True, num_procs=10)\\n run_simulation(spectral=True, ml=True, num_procs=10)\",\n \"def evaluate(cfg: DictConfig):\\n\\n experiments = cfg.get('experiment_type', f'{cfg.model.name}_only')\\n fixed_t0 = cfg.get('fixed_t0', False)\\n ext = '_fixedT0' if fixed_t0 else ''\\n\\n base_dir = cfg.device.root\\n datasource = cfg.datasource.name\\n\\n if experiments == 'ablations':\\n models = {\\n 'FluxRGNN': ['final',\\n 'final_without_encoder',\\n 'final_without_boundary'],\\n 'LocalLSTM': ['final']\\n }\\n elif experiments == 'final':\\n models = {\\n 'FluxRGNN': ['final'],\\n 'GAM': ['final'],\\n 'HA': ['final'],\\n 'GBT': ['final']\\n }\\n else:\\n m = cfg.model.name\\n year = cfg.datasource.test_year\\n\\n # find all experiments available for this model, datasource and test year\\n result_dir = osp.join(base_dir, 'results', datasource, m, f'test_{year}')\\n models = {\\n m : [ f.name for f in os.scandir(result_dir) if f.is_dir() ]\\n }\\n\\n\\n # thresholds for binary classification metrics\\n if cfg.datasource.name == 'abm':\\n thresholds = [0.0019, 0.0207]\\n else:\\n thresholds = [0, 10, 20]\\n\\n rmse_per_hour = []\\n mae_per_hour = []\\n pcc_per_hour = []\\n bin_per_hour = []\\n\\n rmse_per_night = []\\n mae_per_night = []\\n\\n output_dir = osp.join(base_dir, 'results', datasource, f'performance_evaluation{ext}', experiments)\\n os.makedirs(output_dir, exist_ok=True)\\n\\n counter = 0\\n\\n for m, dirs in models.items():\\n print(f'evaluate {m}')\\n\\n for d in dirs:\\n result_dir = osp.join(base_dir, 'results', datasource, m, f'test_{cfg.datasource.test_year}', d)\\n\\n # check if directory exists\\n if os.path.isdir(result_dir):\\n results, model_cfg = load_cv_results(result_dir, trials=cfg.task.repeats, ext=ext)\\n\\n df_prep = pd.read_csv(osp.join(base_dir, 'data', 'preprocessed',\\n f'{model_cfg[\\\"t_unit\\\"]}_{model_cfg[\\\"model\\\"][\\\"edge_type\\\"]}_ndummy={model_cfg[\\\"datasource\\\"][\\\"n_dummy_radars\\\"]}',\\n datasource, cfg.season, str(cfg.datasource.test_year), 'dynamic_features.csv'))\\n tidx2night = dict(zip(df_prep.tidx, df_prep.nightID))\\n\\n rmse_per_hour.append(compute_rmse(m, d, results, tidx2night, groupby=['horizon', 'trial'],\\n threshold=0, km2=True, fixed_t0=fixed_t0))\\n mae_per_hour.append(compute_mae(m, d, results, tidx2night, groupby=['horizon', 'trial'],\\n threshold=0, km2=True, fixed_t0=fixed_t0))\\n pcc_per_hour.append(compute_pcc(m, d, results, tidx2night, groupby=['horizon', 'trial'],\\n threshold=0, km2=True, fixed_t0=fixed_t0))\\n\\n if fixed_t0:\\n rmse_per_night.append(compute_rmse_per_night(m, d, results, tidx2night, groupby=['night_horizon', 'trial']))\\n mae_per_night.append(compute_mae_per_night(m, d, results, tidx2night, groupby=['night_horizon', 'trial']))\\n\\n # compute binary classification measures\\n for thr in thresholds:\\n bin_per_hour.append(compute_bin(m, d, results, groupby=['horizon', 'trial'], threshold=thr, km2=True))\\n\\n counter += 1\\n\\n else:\\n print(f'Experiment \\\"{d}\\\" for model \\\"{m}\\\" and datasource \\\"{datasource}\\\" is not available. '\\n f'Use \\\"run_experiments.py model={m} datasource={datasource} +experiment={d}\\\" to run this experiment.')\\n\\n if counter > 0:\\n rmse_per_hour = pd.concat(rmse_per_hour)\\n rmse_per_hour.to_csv(osp.join(output_dir, f'rmse_per_hour.csv'))\\n\\n mae_per_hour = pd.concat(mae_per_hour)\\n mae_per_hour.to_csv(osp.join(output_dir, f'mae_per_hour.csv'))\\n\\n pcc_per_hour = pd.concat(pcc_per_hour)\\n pcc_per_hour.to_csv(osp.join(output_dir, f'pcc_per_hour.csv'))\\n\\n bin_per_hour = pd.concat(bin_per_hour)\\n bin_per_hour.to_csv(osp.join(output_dir, f'bin_per_hour.csv'))\\n\\n if fixed_t0:\\n rmse_per_night = pd.concat(rmse_per_night)\\n rmse_per_night.to_csv(osp.join(output_dir, f'rmse_per_night.csv'))\\n\\n mae_per_night = pd.concat(mae_per_night)\\n mae_per_night.to_csv(osp.join(output_dir, f'mae_per_night.csv'))\",\n \"def basic_experiments(k, p, num_iter, transforms, train_path, test_path):\\n start_time = time.time()\\n print(\\\"Generating advice\\\")\\n with open(train_path, \\\"r\\\") as f:\\n train_data = f.readlines()\\n advice = generate_advice(train_data, transforms)\\n print(\\\"Done: \\\" + str(time.time() - start_time) + \\\"\\\\n\\\")\\n\\n print(\\\"Generating real count for test data\\\")\\n with open(test_path, \\\"r\\\") as f:\\n test_data = f.readlines()\\n real_count = generate_advice(test_data, transforms)\\n print(\\\"Done: \\\" + str(time.time() - start_time))\\n actual = [x.moment(p) for x in real_count]\\n print(\\\"Actual moments:\\\")\\n print(actual)\\n\\n print(\\\"Estimating using noisy advice:\\\")\\n est_with_advice = []\\n for i in range(num_iter):\\n seed = str(random.randint(1, 2**64 + 1)) + str(random.randint(1, 2**64 + 1))\\n this_hash_exp = lambda x: hash_exp(x, seed)\\n est_with_advice.append(\\n estimate_using_advice(test_data, transforms, k, p, this_hash_exp,\\n advice))\\n print(\\\"Done: \\\" + str(time.time() - start_time))\\n print(est_with_advice)\\n\\n print(\\\"Estimating using perfect advice:\\\")\\n est_perfect_advice = []\\n for i in range(num_iter):\\n seed = str(random.randint(1, 2**64 + 1)) + str(random.randint(1, 2**64 + 1))\\n this_hash_exp = lambda x: hash_exp(x, seed)\\n est_perfect_advice.append(\\n estimate_using_advice(test_data, transforms, k, p, this_hash_exp,\\n real_count))\\n print(\\\"Done: \\\" + str(time.time() - start_time))\\n print(est_perfect_advice)\\n\\n print(\\\"Estimating using PPSWOR:\\\")\\n est_ppswor = []\\n for i in range(num_iter):\\n cur = []\\n for x in real_count:\\n cur.append(ppswor_estimate_moment(x.counts.items(), k, p))\\n est_ppswor.append(cur)\\n print(\\\"Done: \\\" + str(time.time() - start_time))\\n print(est_ppswor)\\n\\n print(\\\"Estimating using ell_2 PPSWOR:\\\")\\n est_ppswor_l2 = []\\n for i in range(num_iter):\\n cur = []\\n for x in real_count:\\n cur.append(ppswor_estimate_moment(x.counts.items(), k, p, 2))\\n est_ppswor_l2.append(cur)\\n print(\\\"Done: \\\" + str(time.time() - start_time))\\n print(est_ppswor_l2)\",\n \"def evaluate_all(experiments,\\n results,\\n seed,\\n num_pairs,\\n samples_per_pair,\\n loop_size=None):\\n eval_results = evaluate_baselines(\\n experiments[0], seed, num_pairs, samples_per_pair, loop_size=None)\\n for ex, res in zip(experiments, results):\\n eval_results.update(\\n evaluate_experiment(ex, res, seed, num_pairs, samples_per_pair,\\n loop_size))\\n return eval_results\",\n \"def test_compare_classifier_performance_from_pred_vis_api(experiment_to_use):\\n experiment = experiment_to_use\\n prediction = experiment.predictions\\n viz_outputs = (\\\"pdf\\\", \\\"png\\\")\\n with TemporaryDirectory() as tmpvizdir:\\n for viz_output in viz_outputs:\\n vis_output_pattern_pdf = tmpvizdir + f\\\"/*.{viz_output}\\\"\\n visualize.compare_classifiers_performance_from_pred(\\n [prediction, prediction],\\n experiment.ground_truth,\\n experiment.ground_truth_metadata,\\n experiment.output_feature_name,\\n labels_limit=0,\\n model_namess=[\\\"Model1\\\", \\\"Model2\\\"],\\n output_directory=tmpvizdir,\\n file_format=viz_output,\\n )\\n figure_cnt = glob.glob(vis_output_pattern_pdf)\\n assert 1 == len(figure_cnt)\",\n \"def run_all_tests(self) -> None:\\n self.run_trt_precision_tests()\\n logging.info(\\\"Check analysis result at: %s\\\", self._output_dir)\",\n \"def run_oracle_experiments(model, query_iterator, N, recalls=[1]):\\n times_dict = dict()\\n calls_dict = dict()\\n probs = dict()\\n keys = list()\\n recall_break = max(recalls)\\n for query, relevant_set in query_iterator:\\n candidates_iterator = iter_candidates(model, query)\\n times, calls = candidates_time_oracle_calls_count(\\n candidates_iterator, relevant_set, recall_break)\\n k = len(relevant_set)\\n if k in times_dict:\\n times_dict[k].append(times)\\n calls_dict[k].append(calls)\\n probs[k] += 1\\n else:\\n keys.append(k)\\n times_dict[k] = [times]\\n calls_dict[k] = [calls]\\n probs[k] = 1\\n \\n keys = np.array(keys)\\n probs = np.array([probs[k] for k in keys])/sum(probs.values())\\n times_dict = {k:np.array(v).mean(axis=0) for k,v in times_dict.items()}\\n calls_dict = {k:np.array(v).mean(axis=0) for k,v in calls_dict.items()}\\n \\n expected_times = oracle_expectation_by_recalls(recalls, keys, probs, times_dict)\\n expected_calls = oracle_expectation_by_recalls(recalls, keys, probs, calls_dict)\\n expected_calls_baseline = oracle_random_calls_expectation_by_recall(recalls, keys, probs, N)\\n \\n return expected_times, expected_calls, expected_calls_baseline\",\n \"def run_tests(augmenter, images, nb_runs):\\n results = np.zeros((nb_runs,))\\n for i in range(nb_runs):\\n start = time.time()\\n augmenter.augment_batch(images)\\n results[i] = time.time() - start\\n print(\\\"Run %d: %.8fs\\\" % (i, results[i]))\\n print(\\\"Mean: %.8fs\\\" % (results.mean(),))\\n print(\\\"Sum: %.8fs\\\" % (results.sum(),))\",\n \"def test_compare_classifiers_performance_changing_k_vis_api(experiment_to_use):\\n experiment = experiment_to_use\\n probabilities = experiment.probabilities\\n viz_outputs = (\\\"pdf\\\", \\\"png\\\")\\n with TemporaryDirectory() as tmpvizdir:\\n for viz_output in viz_outputs:\\n vis_output_pattern_pdf = tmpvizdir + f\\\"/*.{viz_output}\\\"\\n visualize.compare_classifiers_performance_changing_k(\\n [probabilities, probabilities],\\n experiment.ground_truth,\\n experiment.ground_truth_metadata,\\n experiment.output_feature_name,\\n top_k=3,\\n labels_limit=0,\\n model_namess=[\\\"Model1\\\", \\\"Model2\\\"],\\n output_directory=tmpvizdir,\\n file_format=viz_output,\\n )\\n figure_cnt = glob.glob(vis_output_pattern_pdf)\\n assert 1 == len(figure_cnt)\",\n \"def run_experiments() :\\n #%%\\n target_size=(32,32)\\n g_specs = {\\n \\\"batch_size\\\" : [ 30 , 60, 100 ],\\n \\\"learning_rate\\\" : [ 0.0002, 0.0003, 0.0005 ],\\n \\\"drop_out_rate\\\" : [ 0.2, 0.25, 0.3 ],\\n \\\"rescale_mode\\\" : [ \\\"max_q\\\" , \\\"max\\\", \\\"\\\" ]\\n }\\n\\n model_traits = MODEL_TRAITS[\\\"model2\\\"].copy()\\n tt_obj = model_traits[\\\"trainer_tester_class\\\"]( model_traits )\\n del model_traits[\\\"trainer_tester_class\\\"]\\n\\n cnt = 0\\n for batchs, lrate, do_rate, resc_mode in product( g_specs[\\\"batch_size\\\"],\\n g_specs[\\\"learning_rate\\\"],\\n g_specs[\\\"drop_out_rate\\\"],\\n g_specs[\\\"rescale_mode\\\"] ) :\\n\\n tt_obj.model_traits.update( {\\\"batch_size\\\" : batchs,\\n \\\"learning_rate\\\" : lrate,\\n \\\"rescale_mode\\\" : resc_mode,\\n \\\"drop_out_rate\\\" : do_rate } )\\n\\n train_4d, train_gt = tu.make_4d_arrays( images_dir=\\\"images/train\\\",\\n target_size=target_size )\\n\\n test_4d, test_gt = tu.make_4d_arrays( images_dir=\\\"images/test\\\",\\n target_size=target_size )\\n\\n data = {\\\"train_4d\\\" : train_4d,\\n \\\"test_4d\\\" : test_4d,\\n \\\"train_y\\\" : train_gt,\\n \\\"test_y\\\" : test_gt}\\n\\n valid_accu_log, train_accu_log = tt_obj.train( model_traits, data,\\n logl=100 )\\n idx_v = int(np.argmax( valid_accu_log))\\n idx_t = int(np.argmax( train_accu_log))\\n\\n model_traits.update({\\\"valid_accu_log\\\" : valid_accu_log,\\n \\\"train_accu_log\\\" : train_accu_log,\\n \\\"best_valid\\\" : max(valid_accu_log),\\n \\\"best_valid_at\\\" : idx_v,\\n \\\"train_at_best_valid\\\" : train_accu_log[idx_v],\\n \\\"best_train\\\" : max(train_accu_log),\\n \\\"best_train_at\\\": idx_t })\\n\\n #print(cnt, pformat(model_traits) )\\n print( \\\"%d : best_train = %.4f, best_valid = %.4f\\\" % \\\\\\n (cnt, max(train_accu_log), max(valid_accu_log) ))\\n\\n with open( \\\"exp_results_%d.json\\\" % cnt,\\n \\\"wt\\\" , encoding=\\\"utf8\\\" ) as f_out :\\n print( json.dumps( model_traits ), file=f_out)\\n\\n\\n cnt += 1\\n #%%\",\n \"def test_any(alg, repNum=1):\\n\\tperformance = alg.predict_perf(repNum=repNum)\\n\\tprint performance\\n\\talg.plot()\\n\\n\\treturn performance\",\n \"def test_compare_classifiers_performance_subset_vis_api(experiment_to_use):\\n experiment = experiment_to_use\\n probabilities = experiment.probabilities\\n viz_outputs = (\\\"pdf\\\", \\\"png\\\")\\n with TemporaryDirectory() as tmpvizdir:\\n for viz_output in viz_outputs:\\n vis_output_pattern_pdf = tmpvizdir + f\\\"/*.{viz_output}\\\"\\n visualize.compare_classifiers_performance_subset(\\n [probabilities, probabilities],\\n experiment.ground_truth,\\n experiment.ground_truth_metadata,\\n experiment.output_feature_name,\\n top_n_classes=[6],\\n labels_limit=0,\\n subset=\\\"ground_truth\\\",\\n model_namess=[\\\"Model1\\\", \\\"Model2\\\"],\\n output_directory=tmpvizdir,\\n file_format=viz_output,\\n )\\n figure_cnt = glob.glob(vis_output_pattern_pdf)\\n assert 1 == len(figure_cnt)\",\n \"def compute_and_visualize_coupling_experiments(experiments, results, seed,\\n **kwargs):\\n (logits_1,\\n logits_2), couplings = get_coupling_estimates(experiments, results, seed,\\n **kwargs)\\n test_losses = compute_coupling_losses(experiments, logits_1, logits_2,\\n couplings)\\n visualize_coupling_experiments(test_losses, logits_1, logits_2, couplings)\",\n \"def test_compare_classifier_performance_from_prob_vis_api(experiment_to_use):\\n experiment = experiment_to_use\\n probability = experiment.probabilities\\n viz_outputs = (\\\"pdf\\\", \\\"png\\\")\\n with TemporaryDirectory() as tmpvizdir:\\n for viz_output in viz_outputs:\\n vis_output_pattern_pdf = tmpvizdir + f\\\"/*.{viz_output}\\\"\\n visualize.compare_classifiers_performance_from_prob(\\n [probability, probability],\\n experiment.ground_truth,\\n experiment.ground_truth_metadata,\\n experiment.output_feature_name,\\n top_n_classes=[0],\\n labels_limit=0,\\n model_namess=[\\\"Model1\\\", \\\"Model2\\\"],\\n output_directory=tmpvizdir,\\n file_format=viz_output,\\n )\\n figure_cnt = glob.glob(vis_output_pattern_pdf)\\n assert 1 == len(figure_cnt)\",\n \"def run():\\n trials = 100\\n\\n multipliers = [0.25, 0.3, 0.35, 0.5, 0.75, 1, 1.25, 1.45, 1.5, 1.55, 1.6] # Coefficients for learning rate\\n\\n mean_penalty = []\\n median_penalty = []\\n std_penalty = []\\n\\n mean_trial_time = []\\n median_trial_time = []\\n std_trial_time = []\\n\\n mean_success_rate = []\\n median_success_rate = []\\n std_success_rate = []\\n\\n for m in multipliers:\\n all_penalties = [] # All penalties from trail sets\\n all_average_trial_time = []\\n all_success_rates = []\\n\\n for i in range(0, 20):\\n # print \\\"Trial set:\\\", i\\n # Set up environment and agent\\n e = Environment() # create environment (also adds some dummy traffic)\\n agent = e.create_agent(LearnerAgent) # create agent\\n agent.mult = m\\n e.set_primary_agent(agent, enforce_deadline=True) # specify agent to track\\n\\n # Now simulate it\\n sim = Simulator(e, update_delay=0.0, display=False) # create simulator (uses pygame when display=True, if available)\\n\\n sim.run(n_trials=trials) # run for a specified number of trials\\n\\n all_penalties.append(agent.all_trails_penalties)\\n all_average_trial_time.append(agent.time/float(trials))\\n all_success_rates.append(float(trials-agent.aborted_trials)/trials)\\n\\n mean_penalty.append(np.mean(all_penalties))\\n median_penalty.append(np.median(all_penalties))\\n std_penalty.append(np.std(all_penalties))\\n\\n mean_trial_time.append(np.mean(all_average_trial_time))\\n median_trial_time.append(np.median(all_average_trial_time))\\n std_trial_time.append(np.std(all_average_trial_time))\\n\\n mean_success_rate.append(np.mean(all_success_rates))\\n median_success_rate.append(np.median(all_success_rates))\\n std_success_rate.append(np.std(all_success_rates))\\n\\n for i in range(0, len(multipliers)):\\n print \\\"\\\"\\n print \\\"Multiplier:\\\", multipliers[i]\\n print \\\"\\\"\\n print \\\"Mean penalty per {} trials:\\\".format(trials), mean_penalty[i]\\n print \\\"Median penalty per {} trials:\\\".format(trials), median_penalty[i]\\n print \\\"Std.Dev. penalty per {} trials:\\\".format(trials), std_penalty[i]\\n\\n print \\\"\\\"\\n print \\\"Mean trial time:\\\", mean_trial_time[i]\\n print \\\"Median trial time:\\\", median_trial_time[i]\\n print \\\"Std.Dev. trial time:\\\", std_trial_time[i]\\n\\n print \\\"\\\"\\n print \\\"Mean success rate per {} trials:\\\".format(trials), mean_success_rate[i]\\n print \\\"Median success rate per {} trials:\\\".format(trials), median_success_rate[i]\\n print \\\"Std.Dev. success rate per {} trials:\\\".format(trials), std_success_rate[i]\",\n \"def test_run(id_, args, kwargs, expected):\\n # Remove results for test.\\n results_path = expected['results_path']\\n shutil.rmtree(results_path, ignore_errors=True)\\n # Start code.\\n # [future] attempts to run classification with n_jobs>1\\n # global cnfg_default\\n # sys.modules['cnfg_default'] = get_params(args[0], 'cnfg_default')\\n # import cnfg_default\\n # #from cnfg_default import custom_score_metric\\n objects = pycnfg.run(oid='default', *args, **kwargs)\\n tmp = {k: type(v).__name__ for k, v in objects.items()}\\n print('OBJECTS:')\\n print(tmp)\\n # Compare results:\\n # * Compare objects (keys and str of values).\\n objects_ = expected['objects']\\n objects = {k: type(v).__name__ for k, v in objects.items()}\\n if objects != objects_:\\n print(set(objects.items()) ^ set(objects_.items()))\\n assert objects == objects_\\n # for k, v in objects.items():\\n # assert k in objects_\\n # assert type(v).__name__ == objects_[k]\\n # * Compare predictions csv(all available).\\n pred_path = glob.glob(f\\\"{results_path}/models/*_pred.csv\\\")\\n pred_path_ = glob.glob(expected['pred_path'])\\n assert len(pred_path) == len(pred_path_)\\n for act, exp in zip(sorted(pred_path), sorted(pred_path_)):\\n file_diff(act, exp)\\n assert filecmp.cmp(act, exp)\\n # * Compare test logs.\\n logs_path = glob.glob(f\\\"{results_path}/logs*/*_test.log\\\")[0]\\n logs_path_ = expected['logs_path']\\n file_diff(logs_path, logs_path_)\\n assert filecmp.cmp(logs_path, logs_path_)\\n # * Compare runs dataframe, non-universe columns.\\n runs_path = f\\\"{results_path}/runs\\\"\\n runs_path_ = expected['runs_path']\\n df = runs_loader(runs_path)\\n df_ = runs_loader(runs_path_)\\n # First False/True for each element, then check all by columns.\\n # col1 True\\n # col2 False\\n # dtype: bool\\n df_diff = df.eq(df_).all()\\n # Column names that are not equal.\\n columns = sorted(list(df_diff[df_diff==False].dropna().index))\\n # columns_eq = sorted(list(df_diff[df_diff==True].dropna().index))\\n columns_ = expected['columns_diff']\\n print('DIFF:\\\\n', columns)\\n time.sleep(1)\\n # assert columns == columns_\\n # * Compare model.\\n model_path = glob.glob(f\\\"{results_path}/models/*.model\\\")\\n model_path_ = glob.glob(expected['model_path'])\\n assert len(model_path) == len(model_path_)\\n for act, exp in zip(sorted(model_path), sorted(model_path_)):\\n assert filecmp.cmp(act, exp)\\n return\",\n \"def run_tests():\\n with open(FILENAME) as file:\\n # Loads testing parameters from the yaml file.\\n tests = yaml.safe_load(file)\\n\\n # create a dataframe to keep the results\\n test_dict = tests['Tests']\\n results = pd.DataFrame(test_dict)\\n results['Last Average Score'] = \\\"\\\"\\n results['No of Q-Learning episodes'] = \\\"\\\"\\n\\n # run experiments:\\n for i, test in enumerate(test_dict):\\n grid = Rooms(test[\\\"env_size\\\"], testing=True)\\n learning = QLearning(grid, test[\\\"gamma\\\"], test[\\\"alpha\\\"], test[\\\"agent_start_pos\\\"])\\n e_greedy = Policy(\\\"e-greedy\\\", test[\\\"epsilon\\\"], test[\\\"decay\\\"])\\n greedy = Policy(policy_type=\\\"greedy\\\")\\n experiment = Experiments(grid, learning, greedy, test[\\\"iters\\\"],\\n test[\\\"agent_start_pos\\\"], test[\\\"test_no\\\"])\\n\\n for session in range(test[\\\"iters\\\"]):\\n learning.run_multiple_episodes(test[\\\"batch_episodes\\\"], e_greedy)\\n mean_reward = experiment.run_experiments(test[\\\"exp_per_batch\\\"])\\n\\n results.loc[i,'Last Average Score'] = mean_reward\\n results.loc[i,'No of Q-Learning episodes'] = (session + 1) * test[\\\"batch_episodes\\\"]\\n\\n # save results to csv file\\n filename = 'results/' + 'test_table.csv'\\n results.to_csv(filename)\\n\\n # plot & save graphs\\n experiment.generate_results(test[\\\"test_no\\\"], test)\\n\\n return results\",\n \"def evaluate_individual(predictions, test_files, models):\\n\\n print(\\\"\\\\nAccuracy for individual models\\\\n\\\")\\n \\n # Fix Location\\n correct_predictions = [0, 0, 0]\\n total_predictions = [0, 0, 0]\\n num_failed_predictions = 0\\n\\n for prediction in predictions:\\n if prediction[\\\"correct_data\\\"][\\\"correct_location\\\"] == prediction[\\\"predicted_location\\\"]:\\n correct_predictions[FixType[prediction[\\\"correct_data\\\"][\\\"correct_type\\\"]].value] = correct_predictions[FixType[\\n prediction[\\\"correct_data\\\"][\\\"correct_type\\\"]].value] + 1\\n if prediction[\\\"predicted_location\\\"] is None:\\n num_failed_predictions = num_failed_predictions + 1\\n total_predictions[FixType[prediction[\\\"correct_data\\\"][\\\"correct_type\\\"]].value] = total_predictions[FixType[\\n prediction[\\\"correct_data\\\"][\\\"correct_type\\\"]].value] + 1\\n\\n for i in range(3):\\n if total_predictions[i] == 0: # If the type was never predicted\\n accuracy = 0\\n else:\\n accuracy = correct_predictions[i] / total_predictions[i]\\n print(f\\\"Fix Location accuracy for class {FixType(i).name}: {accuracy * 100} %\\\")\\n\\n accuracy = sum(correct_predictions) / (len(predictions) - num_failed_predictions)\\n print(f\\\"Fix Location accuracy overall is {accuracy * 100} %\\\")\\n \\n # Fix type\\n correct_predictions = [0, 0, 0]\\n total_predictions = [0, 0, 0]\\n num_failed_predictions = 0\\n\\n for prediction in predictions:\\n if prediction[\\\"correct_data\\\"][\\\"correct_type\\\"] == prediction[\\\"predicted_type\\\"]:\\n correct_predictions[FixType[prediction[\\\"predicted_type\\\"]].value] = correct_predictions[FixType[\\n prediction[\\\"predicted_type\\\"]].value] + 1\\n if prediction[\\\"predicted_type\\\"] is None:\\n num_failed_predictions = num_failed_predictions + 1\\n total_predictions[FixType[prediction[\\\"predicted_type\\\"]].value] = total_predictions[FixType[\\n prediction[\\\"predicted_type\\\"]].value] + 1\\n\\n for i in range(3):\\n if total_predictions[i] == 0: # If the type was never predicted\\n accuracy = 0\\n else:\\n accuracy = correct_predictions[i] / total_predictions[i]\\n print(f\\\"Fix Type accuracy for class {FixType(i).name}: {accuracy * 100} %\\\")\\n\\n accuracy = sum(correct_predictions) / (len(predictions) - num_failed_predictions)\\n print(f\\\"Fix Type accuracy overall is {accuracy * 100} %\\\")\\n \\n # We repeat the predictions to evaluate the insert and modify models individually, regardless of the predicted fix type \\n\\n raw_training_samples = []\\n\\n if test_files.endswith(\\\".json\\\"): # Single JSON file\\n with open(test_files) as file:\\n logging.info(\\\"Source ending in .json. Predicting on single JSON file.\\\")\\n raw_training_samples = json.load(file)\\n else: # Folder path\\n for filename in listdir(test_files):\\n with open(test_files + filename) as file:\\n raw_training_samples.extend(json.load(file))\\n \\n correct_predictions_insert = 0\\n total_predictions_insert = 0\\n correct_predictions_modify = 0\\n total_predictions_modify = 0\\n insert_tokens = []\\n modify_tokens = []\\n\\n for sample in raw_training_samples:\\n # Insert\\n if sample[\\\"metadata\\\"][\\\"fix_type\\\"] == \\\"insert\\\":\\n actual_sample, tokens = IOProcessor.preprocess(sample[\\\"wrong_code\\\"])\\n pred = predict_single(actual_sample, models[2])\\n token = IOProcessor.postprocess(pred, 2)\\n if token == sample[\\\"metadata\\\"][\\\"fix_token\\\"]: # Correct Prediction\\n correct_predictions_insert = correct_predictions_insert + 1\\n else: # Incorrect prediction\\n insert_tokens.append([token, sample[\\\"metadata\\\"][\\\"fix_token\\\"]])\\n total_predictions_insert = total_predictions_insert + 1\\n # Modify\\n if sample[\\\"metadata\\\"][\\\"fix_type\\\"] == \\\"modify\\\":\\n actual_sample, tokens = IOProcessor.preprocess(sample[\\\"wrong_code\\\"])\\n pred = predict_single(actual_sample, models[3])\\n token = IOProcessor.postprocess(pred, 3)\\n if token == sample[\\\"metadata\\\"][\\\"fix_token\\\"]: # Correct Prediction\\n correct_predictions_modify = correct_predictions_modify + 1\\n else: # Incorrect prediction\\n modify_tokens.append([token, sample[\\\"metadata\\\"][\\\"fix_token\\\"]])\\n total_predictions_modify = total_predictions_modify + 1\\n\\n insert_accuracy = correct_predictions_insert / total_predictions_insert\\n modify_accuracy = correct_predictions_modify / total_predictions_modify\\n print(f\\\"Fix Token accuracy for insert is {insert_accuracy * 100} %\\\")\\n print(f\\\"Fix Token accuracy for modify is {modify_accuracy * 100} %\\\")\\n\\n # The following code may be used to create a swarm plot of the erroneous predictions for fix locations\\n # This does, however, require the installation of the pandas, seaborn, and matplotlib libraries.\\n \\n # import seaborn as sns\\n # import matplotlib.pyplot as plt\\n # import pandas as pd\\n # location_distance_array = []\\n # for prediction in predictions:\\n # actual_sample, tokens = IOProcessor.preprocess(prediction[\\\"correct_data\\\"][\\\"wrong_code\\\"])\\n # label = get_token_index(prediction[\\\"correct_data\\\"][\\\"wrong_code\\\"], tokens, prediction[\\\"correct_data\\\"][\\\"correct_location\\\"])\\n # if prediction[\\\"predicted_token_location\\\"] - label == 0:\\n # pass\\n # else:\\n # location_distance_array.append([prediction[\\\"predicted_token_location\\\"] - label, prediction[\\\"correct_data\\\"][\\\"correct_type\\\"]])\\n \\n # df = pd.DataFrame(data=location_distance_array)\\n # sns.set_theme(style=\\\"whitegrid\\\")\\n # f, ax = plt.subplots(figsize=(6, 4))\\n # sns.despine(bottom=True, left=True)\\n # sns.swarmplot(y=0, x=1, data=df, palette=\\\"dark\\\", size=6)\\n # ax.set_xlabel('')\\n # ax.set_ylabel('')\\n # plt.ylim([-15, 16])\\n \\n # plt.savefig('line_plot.pdf', bbox_inches='tight', pad_inches=0)\",\n \"def main():\\n\\tresults = []\\n\\n\\tconfig = configparser.ConfigParser()\\n\\tconfig.read(\\\"simulation.ini\\\")\\n\\tsettings = config['sim']\\n\\n\\tcompleted_obj_hw = int(settings[\\\"ClientsPerCampaign\\\"]) * float(settings[\\\"CompletedPctgHW\\\"])\\n\\texceeded_obj_hw = float(settings[\\\"ExceededPctgHW\\\"])\\n\\tsignificance_level = float(settings[\\\"SignificanceLevel\\\"])\\n\\tz_val_two_tails = scipy.stats.norm.ppf(1 - (significance_level / 2))\\n\\n\\tprint(\\\"Completed Target HW: \\\" + str(completed_obj_hw))\\n\\tprint(\\\"Exceeded Target HW: \\\" + str(exceeded_obj_hw))\\n\\n\\tcompleted_vals = []\\n\\texceeded_vals = []\\n\\tdone = False\\n\\n\\tcompleted_avg = 0\\n\\texceeded_avg = 0\\n\\tcompleted_hw = 0\\n\\texceeded_hw = 0\\n\\n\\ti = 0\\n\\twhile not done:\\n\\t\\tprint(\\\"RUN: \\\" + str(i + 1))\\n\\t\\tenv = simpy.Environment()\\n\\t\\tsim = Simulation(env, settings, i == 0)\\n\\t\\tsim.run()\\n\\t\\tresults.append(sim.results)\\n\\t\\ti += 1\\n\\n\\t\\tif settings['RunOnce'] == 'yes':\\n\\t\\t\\tprint(\\\"RUN ONCE\\\")\\n\\t\\t\\tsys.exit()\\n\\n\\t\\tcompleted_vals.append(sim.results['completed_count'])\\n\\t\\texceeded_vals.append(sim.results['exceeded_proportion'])\\n\\n\\t\\tif i < 2:\\n\\t\\t\\tprint(\\\"---------------\\\")\\n\\t\\t\\tcontinue\\n\\n\\t\\tcompleted_avg = sum(completed_vals) / len(completed_vals)\\n\\t\\tcompleted_S = sum([(v - completed_avg) ** 2 for v in completed_vals]) / (i - 1)\\n\\t\\tcompleted_S = math.sqrt(completed_S)\\n\\t\\tcompleted_hw = (z_val_two_tails * completed_S) / math.sqrt(i)\\n\\t\\tprint(\\\"runs: \\\" + str(i) + \\\" completed HW: \\\" + str(completed_hw))\\n\\n\\t\\texceeded_avg = sum(exceeded_vals) / len(exceeded_vals)\\n\\t\\texceeded_S = math.sqrt(exceeded_avg * (1 - exceeded_avg))\\n\\t\\texceeded_hw = (z_val_two_tails * exceeded_S) / math.sqrt(i)\\n\\t\\tprint(\\\"runs: \\\" + str(i) + \\\" exceeded HW: \\\" + str(exceeded_hw))\\n\\n\\t\\tif completed_hw < completed_obj_hw and exceeded_hw < exceeded_obj_hw:\\n\\t\\t\\tprint(\\\"END ITERATIONS\\\")\\n\\t\\t\\tdone = True\\n\\n\\t\\tprint(\\\"---------------\\\")\\n\\n\\n\\tfilename = 'results/Results_' + settings['FileSizeGB'] + '_' + settings['TorrentThreshold'] + '_' + settings['HTTPDownThreshold'] \\\\\\n\\t\\t+ '_' + settings['HTTPUp'] + '_' + str(random.randint(0,10000)) + '.xlsx'\\n\\n\\tprint(\\\"Saving XLSX to: \\\" + filename)\\n\\twb = xs.Workbook(filename)\\n\\n\\tws = wb.add_worksheet()\\n\\n\\tws.write(0, 1, 'Exceded')\\n\\tws.write(0, 2, 'Completed')\\n\\n\\ti = 1\\n\\tfor result in results:\\n\\t\\tws.write(i, 0, i)\\n\\t\\tws.write(i, 1, result['exceeded_proportion'])\\n\\t\\tws.write(i, 2, result['completed_count'])\\n\\t\\ti += 1\\n\\n\\tws.write(i, 0, 'average')\\n\\tws.write(i, 1, exceeded_avg)\\n\\tws.write(i, 2, completed_avg)\\n\\ti += 1\\n\\tws.write(i, 0, 'half width')\\n\\tws.write(i, 1, exceeded_hw)\\n\\tws.write(i, 2, completed_hw)\\n\\n\\twb.close()\",\n \"def evaluate_benchmarks(self):\\n\\n # iterate over replicates\\n results = {}\\n for replicate_id, replicate in self.replicates:\\n\\n # evaluate benchmark for current replicate\\n bmark = SimulationBenchmark(replicate.copy(),\\n graph=self.graphs[replicate_id],\\n **self.params)\\n\\n # store results\\n results[replicate_id] = dict(\\n\\n labels_MAE=bmark.scores['labels'].MAE,\\n level_only_MAE=bmark.scores['level_only'].MAE,\\n spatial_only_MAE=bmark.scores['spatial_only'].MAE,\\n community_MAE=bmark.scores['labels_comm'].MAE,\\n\\n labels_PCT=bmark.scores['labels'].percent_correct,\\n level_only_PCT=bmark.scores['level_only'].percent_correct,\\n spatial_only_PCT=bmark.scores['spatial_only'].percent_correct,\\n community_PCT=bmark.scores['labels_comm'].percent_correct)\\n\\n # compile dataframe\\n results = pd.DataFrame.from_dict(results, orient='index')\\n results.index.set_names(self.multiindex, inplace=True)\\n\\n return results\",\n \"def run_regression_experiments(self):\\n assert self.patientIDs # assert not empty\\n self.log.info(\\\"Running experiments for {} patients\\\".format(len(self.patientIDs)))\\n\\n results_batch = defaultdict(dict)\\n results_single = dict()\\n for patient in self.patientIDs:\\n patient_experiment = PatientPredictionExperiment(\\n patient_id=patient, algorithm=self.algorithm,\\n is_batch=self.isBatch\\n )\\n # run results for single patients\\n if not self.isBatch:\\n results_single[patient] = patient_experiment.run_experiment()\\n self.log.debug(\\\"Start processing single results: {}\\\".format(results_single[patient]))\\n # super verbose debug information should be logged with debug option\\n else:\\n # TODO: think about consistent interfaces\\n batch_results, self._allFeatureDesp = patient_experiment.run_experiment()\\n self.log.info(\\\"Start processing batch results with {} batches\\\".format(len(batch_results)))\\n batchNo = 0\\n for results in batch_results:\\n # here are the result for each feature subset for each patients\\n results_batch[batchNo][patient] = results\\n batchNo += 1\\n if self.algorithm == \\\"fm\\\":\\n return\\n if not self.isBatch:\\n results_overall = self.process_results(results_single)\\n self.store_results(results_overall, results_single)\\n self.log.info(\\\"Finished running experiment\\\")\\n else:\\n self.record_batch_experiment()\\n for batchNo, patient_results in results_batch.iteritems():\\n # patient_results = results_single\\n results_overall = self.process_results(patient_results)\\n self.store_results(results_overall, patient_results, self._allFeatureDesp[batchNo])\\n self.log.info(\\\"Finished running experiment No. {}\\\".format(batchNo))\",\n \"def one_experiment():\\n\\n # set the name of the experiment\\n now = datetime.datetime.now()\\n experiment_id = str(now.day) + \\\"_\\\" + str(now.month) + \\\"_\\\" + str(now.hour) + \\\".\\\" + str(now.minute)\\n experiment_name = 'overfit_' + str(experiment_id)\\n\\n # define if you want to use preprocessed data from file\\n use_prep_data = False\\n if use_prep_data:\\n set_params(preproc_data_id='16_5_10.16.47')\\n\\n # define the changing parameter and its value\\n changing_param_name = 'class_weights'\\n changing_param_value = [{0: 15, 1: 85}]\\n # {0:15, 1:85}]#, {0:4, 1:100}, {0:3, 1:100}, {0:2, 1:100}, {0:1, 1:100}] #[{0:1, 1:1}, {0:15, 1:85}]#\\n\\n features_to_use = ['user', 'countries', 'session', 'format', 'token']\\n # set constant parameters\\n set_params(use_word_emb=1)\\n set_params(epochs=40)\\n set_params(features_to_use=features_to_use)\\n\\n # save constant parameters to a new \\\"experiment_..\\\" filgithx+P@2ub\\n save_constant_parameters(experiment_name, changing_param_name)\\n\\n # run experiment for every parameter value\\n for value in changing_param_value:\\n process = psutil.Process(os.getpid())\\n print(\\\"-----MEMORY before starting experiment ------\\\", int(process.memory_info().rss/(8*10**3)), \\\"KB\\\")\\n\\n # update the parameter value\\n set_params(class_weights_1=value)\\n\\n # update the model_id for this new model\\n now = datetime.datetime.now()\\n new_model_id = str(now.day) + \\\"_\\\" + str(now.month) + \\\"_\\\" + str(now.hour) + \\\".\\\" + str(now.minute) + \\\".\\\" + str(now.second)\\n\\n set_params(model_id=new_model_id)\\n\\n # evaluate the new model and save the results in the experiment file\\n oneExperiment = Process(target=run_experiment, args=(experiment_name,\\n new_model_id, changing_param_name, value,))\\n oneExperiment.start()\\n oneExperiment.join()\",\n \"def test_compare_performance_vis_api(experiment_to_use):\\n experiment = experiment_to_use\\n # extract test stats only\\n test_stats = experiment.test_stats_full\\n viz_outputs = (\\\"pdf\\\", \\\"png\\\")\\n with TemporaryDirectory() as tmpvizdir:\\n for viz_output in viz_outputs:\\n vis_output_pattern_pdf = tmpvizdir + f\\\"/*.{viz_output}\\\"\\n visualize.compare_performance(\\n [test_stats, test_stats],\\n output_feature_name=None,\\n model_names=[\\\"Model1\\\", \\\"Model2\\\"],\\n output_directory=tmpvizdir,\\n file_format=viz_output,\\n )\\n figure_cnt = glob.glob(vis_output_pattern_pdf)\\n assert 1 == len(figure_cnt)\",\n \"def emPerformance(filesAndDirectories='None', resultsFileName='None', iterationCount='3', modes='None', testTypes='None', viewports='None', verbose='False'):\\n\\n pass\",\n \"def _evaluate_performance__static_winners(self):\\n # | - _evaluate_performance__\\n\\n # | - class attributes #################################################\\n AL = self\\n al_gen = self.al_gen\\n verbose = self.verbose\\n seed_ids = self.seed_ids\\n acquisition_bin = self.acquisition_bin\\n completed_ids = self.completed_ids\\n CandidateSpace = self.CandidateSpace\\n RegressionModel = self.RegressionModel\\n DuplicateFinder = self.DuplicateFinder\\n al_gen_dict = self.al_gen_dict\\n\\n stop_mode = self.stop_mode\\n stop_num_generations = self.stop_num_generations\\n\\n index_acq_gen_dict = self.index_acq_gen_dict\\n #__| #################################################################\\n\\n # #####################################################################\\n mode = \\\"lowest_N\\\" # 'lowest_N' or 'lowest_perc'\\n\\n N_ids = 10\\n lowest_perc = 5\\n\\n # Number of consecutive generations that the Nth best systems must\\n # remain static\\n M_gens = 3\\n # #####################################################################\\n\\n if mode == \\\"lowest_perc\\\":\\n num_candidates = CandidateSpace.FingerPrints.df_pre.shape[0]\\n N_ids = int(num_candidates * (lowest_perc * 0.01))\\n\\n gen_keys = list(AL.al_gen_dict.keys())\\n\\n if len(gen_keys) > M_gens:\\n latest_M_keys = gen_keys[-(M_gens + 1):]\\n last_gen_key = gen_keys[-1]\\n\\n al_gen_dict_subset_i = dict(zip(\\n latest_M_keys,\\n [AL.al_gen_dict.get(i, None) for i in latest_M_keys]))\\n\\n indices_list = []\\n iterator = enumerate(al_gen_dict_subset_i.items())\\n for i_cnt, (gen_i, AL_i) in iterator:\\n model_i = AL_i.model\\n\\n model_i = AL.add_main_Y_to_model(\\n model_i, plot_dft_instead_of_pred=True)\\n model_i = model_i[(model_i[\\\"duplicate\\\"] == False)]\\n model_i = model_i.sort_values(\\\"Y_main\\\")\\n\\n indices_i = model_i.index.tolist()\\n\\n indices_list.append(indices_i)\\n\\n if i_cnt >= M_gens:\\n indices_i = indices_list[i_cnt][0:N_ids]\\n ids_static_list = []\\n for j in range(M_gens):\\n indices_j = indices_list[i_cnt - (j + 1)][0:N_ids]\\n ids_static = indices_j == indices_i\\n ids_static_list.append(ids_static)\\n\\n ids_are_static = all(ids_static_list)\\n\\n self.performance__static_winners[last_gen_key] = ids_are_static\\n #__|\",\n \"def print_results(precision, recall, f1):\\n print \\\"Precision:\\\", precision\\n print \\\"Recall:\\\", recall\\n print \\\"F1 score:\\\", f1\",\n \"def evaluate(self, runs=100):\\n score_record = []\\n \\n print('Evaluation in progress...')\\n for i in range(runs):\\n score = self.run_evaluation_episode()\\n score_record.append(score)\\n \\n ave_score = np.mean(score_record)\\n \\n print('System evaluated with an average score of {} in {} runs'.format(ave_score, runs))\",\n \"def runTests():\\r\\n\\r\\n print(\\\"running a few tests\\\")\\r\\n\\r\\n average = compute .gpsAverage (4, 5)\\r\\n print(\\\"average = \\\", average)\\r\\n \\r\\n print (\\\"hello!\\\")\",\n \"def main():\\n parser = argparse.ArgumentParser(description=\\\"Process the results of an experiment.\\\")\\n parser.add_argument(\\\"experiment\\\")\\n arguments = parser.parse_args()\\n path = f\\\"experiments/{arguments.experiment}\\\"\\n if not os.path.exists(path):\\n raise SystemExit(f\\\"Path {path} does not exists.\\\")\\n\\n # For efficiency, one should generate the results from the parts without merging them.\\n files = [file for file in os.listdir(path) if os.path.isfile(os.path.join(path, file))]\\n frames = []\\n for file in files:\\n device, experiment, _ = file.split(\\\".\\\")\\n frame = pandas.read_csv(\\n os.path.join(path, file),\\n index_col=\\\"variable\\\",\\n usecols=[\\\"variable\\\", \\\"group_index\\\", \\\"value_i\\\"], dtype={\\\"value_i\\\": \\\"Int64\\\"}\\n )\\n frame[\\\"board\\\"] = device\\n frame[\\\"experiment\\\"] = experiment\\n frames.append(frame)\\n dataframe = pandas.concat(frames)\\n frames = None\\n\\n current_grouping = dataframe.groupby([\\\"group_index\\\", \\\"variable\\\"])\\n \\n data = current_grouping.agg([\\n numpy.median,\\n _percentile_factory(95),\\n numpy.mean,\\n numpy.std,\\n \\\"count\\\"\\n ])\\n\\n print(data)\\n \\n data = data.droplevel([0], axis=1)\\n data = data.unstack()\\n data.columns = data.columns.map('_'.join)\\n data.to_csv(f\\\"{arguments.experiment}.csv\\\")\",\n \"def test_run():\\n # Only few steps for test\\n timesteps = 128\\n\\n # Compute all sub testing conf\\n envs = ['CartPole-v0']\\n ml_platforms = ['torch', 'tf']\\n agents = ['dqn', 'a2c']\\n\\n test_combinations = list(it.product(\\n envs,\\n ml_platforms,\\n agents\\n )\\n )\\n\\n # Finally test them all\\n for conf in test_combinations:\\n env_str, ml_platform_str, agent_str = conf\\n run(\\n agent_str,\\n ml_platform_str,\\n env_str,\\n 'dense',\\n timesteps,\\n './target/')\",\n \"def main():\\n for opt in optimizations:\\n compile_command = [\\\"g++\\\", \\\"main.cpp\\\", f\\\"-O{opt}\\\", \\\"-lpthread\\\"]\\n run(compile_command, check=True)\\n for threads in num_threads:\\n print(f\\\"{opt=}, {threads=}\\\", end=\\\"\\\")\\n stdout.flush()\\n test_command = ['./a.out', str(iterations), str(threads)]\\n total = 0\\n for samples in range(1, repeats_for_average + 1):\\n print(\\\".\\\", end=\\\"\\\")\\n stdout.flush()\\n output = run(test_command, check=True, capture_output=True).stdout\\n total += int(output.split()[-2]) / 1000\\n print(f\\\"\\\\t{total / samples:.03f}\\\")\",\n \"def polling_experiment_multiple(dataset_id, times, **kwargs):\\n total_yeahs = 0\\n total_noes = 0\\n for i in range(0, times):\\n yeah, noes = polling_experiment(dataset_id, **kwargs)\\n total_yeahs += yeah\\n total_noes += noes\\n logging.info(\\\"After \\\"+str(times)+\\\" \\\"+ str(dataset_id) + \\\" experiments:\\\")\\n logging.info(\\\"\\\"+ str(total_yeahs) + \\\" correct predictions\\\")\\n logging.info(\\\"\\\"+ str(total_noes) + \\\" wrong predictions\\\")\\n logging.info(\\\"\\\"+ str(total_yeahs/(total_yeahs + total_noes)) + \\\" accuracy\\\")\",\n \"def _train_and_evaluate(estimator, output_dir):\\n \\n \\\"\\\"\\\"X_train, y_train =utils._feature_label_split(df_train,\\\"is_churn\\\",\\\"msno\\\")\\n df_val = utils.read_from_bigquery(\\\"amiable-octane-267022.kkbox.output_val_1\\\",\\\"amiable-octane-267022\\\")\\n X_val, y_val =utils._feature_label_split(df_val,\\\"is_churn\\\",\\\"msno\\\")\\\"\\\"\\\"\\n \\n df_train=utils.over_sample(\\\"amiable-octane-267022.kkbox.output_train_1\\\",\\\"amiable-octane-267022\\\")\\n X_train, y_train =utils._feature_label_split(df_train,\\\"is_churn\\\",\\\"msno\\\")\\n df_val=utils.over_sample(\\\"amiable-octane-267022.kkbox.output_val_1\\\",\\\"amiable-octane-267022\\\")\\n X_val, y_val =utils._feature_label_split(df_val,\\\"is_churn\\\",\\\"msno\\\")\\n\\n estimator.fit(X_train, y_train)\\n f1_scorer = make_scorer(f1_score)\\n accuracy_scorer =make_scorer(accuracy_score)\\n\\n if metadata.HYPERPARAMTER_TUNING:\\n scores=model_selection.cross_val_score(estimator, X_val, y_val, cv=3,scoring=f1_scorer)\\n #,scoring=f1_scorer\\n\\n logging.info('Score: %s', scores)\\n\\n #tune hyper\\n hpt = hypertune.HyperTune()\\n hpt.report_hyperparameter_tuning_metric(\\n hyperparameter_metric_tag='F1_SCORE',\\n metric_value=np.mean(scores),\\n global_step=10000)\\n \\n#joblib.dump(estimator, 'model.joblib')\\n\\n # Write model and eval metrics to `output_dir`\\n model_output_path = os.path.join(output_dir, 'model',metadata.MODEL_FILE_NAME)\\n \\n utils.dump_object(estimator, model_output_path)\",\n \"def batch_simulation(self, iters=10000):\\n power_cnt = 0\\n correct_sign_cnt = 0\\n\\n for i in range(iters):\\n if (self.verbose) and (i>0) and (i % (iters/10) == 0):\\n print(i, \\\" / \\\", iters)\\n f_stat, p_value, effect_point_estimates = self.simulate()\\n power_cnt += (p_value < self.alpha)\\n correct_sign_cnt += (effect_point_estimates * self.absolute_effect > 0)\\n\\n if self.verbose: print(iters, \\\" / \\\", iters)\\n power = round(power_cnt / float(iters), 5)\\n pct_correct_sign = round(correct_sign_cnt / float(iters), 5)\\n return power, pct_correct_sign\",\n \"def test_calculate_all_metrics_precision():\\n pass\",\n \"def exe_tests(self):\\n self.rank = mpicom.rank()\\n self.size = mpicom.size()\\n if mpicom.parallel():\\n self.test(\\\"libname\\\",os.path.split(mpicom.__file__)[1],\\\"mpicom.so\\\")\\n else:\\n self.test(\\\"libname\\\",os.path.split(mpicom.__file__)[1],\\\"mpistub.pyc\\\")\\n self.test_broadcast()\\n self.test_reduce()\\n self.test_p2p()\\n self.test_gather()\\n self.test_scatter()\\n #self.test_alltoall()\",\n \"def compute_and_print_eval_metrics(self):\\n s = ('%20s' + '%12s' * 6) % ('Class', 'Images', 'Targets', 'P', 'R', 'mAP@.5', 'mAP@.5:.95')\\n precision, recall, f1, mean_precision, mean_recall, map50, map = 0., 0., 0., 0., 0., 0., 0.\\n ap = []\\n eval_stats = [np.concatenate(x, 0) for x in zip(*self.eval_stats)]\\n if len(eval_stats) and eval_stats[0].any():\\n precision, recall, ap, f1, ap_class = ap_per_class(*eval_stats)\\n precision, recall, ap50, ap = precision[:, 0], recall[:, 0], ap[:, 0], ap.mean(1)\\n mean_precision, mean_recall, map50, map = precision.mean(), recall.mean(), ap50.mean(), ap.mean()\\n nt = np.bincount(eval_stats[3].astype(np.int64), minlength=len(self.class_names)) # number of targets per class\\n else:\\n nt = np.zeros(1)\\n\\n pf = '%20s' + '%12.5g' * 6 # print format\\n print(\\\"\\\\n EVALUTAION \\\\n\\\")\\n print(s)\\n print(pf % ('all', self.seen, nt.sum(), mean_precision, mean_recall, map50, map))\\n if self.cfg.eval.verbose:\\n for indx, cls in enumerate(ap_class):\\n print(pf % (self.class_names[cls], self.seen, nt[cls], precision[indx], recall[indx], ap50[indx], ap[indx]))\",\n \"def run_experiments(data_set=\\\"\\\",compact=2,exp_name=\\\"\\\",x=\\\"\\\"):\\n cwd = os.getcwd()\\n results_path = cwd+'/data/geometric/'+exp_name+'_'\\n dataset_path = cwd+'/data/geometric/'+data_set+'_'\\n compact = int(compact)\\n if \\\"m\\\" in x:\\n # make new pxy\\n pxy, Xdata, groups = gen_geometric_pxy()\\n np.save(dataset_path+'Xdata',Xdata)\\n np.save(dataset_path+'groups',groups)\\n np.save(dataset_path+'pxy',pxy)\\n else:\\n # load existing pxy\\n pxy = np.load(dataset_path+'pxy.npy')\\n if \\\"r\\\" in x: # regular experiments\\n if compact>1:\\n metrics_stepwise, distributions_stepwise,\\\\\\n metrics_converged, distributions_converged,\\\\\\n metrics_stepwise_allreps, distributions_stepwise_allreps,\\\\\\n metrics_converged_allreps, distributions_converged_allreps = test_IB(pxy,compact)\\n metrics_converged.to_csv(results_path+'metrics_converged.csv')\\n metrics_stepwise.to_csv(results_path+'metrics_stepwise.csv')\\n distributions_converged.to_pickle(results_path+'distributions_converged.pkl')\\n elif compact>0:\\n metrics_stepwise,\\\\\\n metrics_converged, distributions_converged,\\\\\\n metrics_stepwise_allreps,\\\\\\n metrics_converged_allreps, distributions_converged_allreps = test_IB(pxy,compact)\\n metrics_converged.to_csv(results_path+'metrics_converged.csv')\\n metrics_stepwise.to_csv(results_path+'metrics_stepwise.csv')\\n distributions_converged.to_pickle(results_path+'distributions_converged.pkl')\\n else:\\n metrics_stepwise,\\\\\\n metrics_converged,\\\\\\n metrics_stepwise_allreps,\\\\\\n metrics_converged_allreps = test_IB(pxy,compact)\\n metrics_converged.to_csv(results_path+'metrics_converged.csv')\\n metrics_stepwise.to_csv(results_path+'metrics_stepwise.csv')\\n if \\\"ip\\\" in x: # initialization experiments - positive p0\\n if compact>1:\\n metrics_stepwise, distributions_stepwise,\\\\\\n metrics_converged, distributions_converged,\\\\\\n metrics_stepwise_allreps, distributions_stepwise_allreps,\\\\\\n metrics_converged_allreps, distributions_converged_allreps = test_p0_pos(pxy,compact)\\n metrics_converged.to_csv(results_path+'metrics_converged_p0_pos.csv')\\n metrics_stepwise.to_csv(results_path+'metrics_stepwise_p0_pos.csv')\\n distributions_converged.to_pickle(results_path+'distributions_converged_p0_pos.pkl')\\n elif compact>0:\\n metrics_stepwise,\\\\\\n metrics_converged, distributions_converged,\\\\\\n metrics_stepwise_allreps,\\\\\\n metrics_converged_allreps, distributions_converged_allreps = test_p0_pos(pxy,compact)\\n metrics_converged.to_csv(results_path+'metrics_converged_p0_pos.csv')\\n metrics_stepwise.to_csv(results_path+'metrics_stepwise_p0_pos.csv')\\n distributions_converged.to_pickle(results_path+'distributions_converged_p0_pos.pkl')\\n else:\\n metrics_stepwise,\\\\\\n metrics_converged,\\\\\\n metrics_stepwise_allreps,\\\\\\n metrics_converged_allreps = test_p0_pos(pxy,compact)\\n metrics_converged.to_csv(results_path+'metrics_converged_p0_pos.csv')\\n metrics_stepwise.to_csv(results_path+'metrics_stepwise_p0_pos.csv')\\n if \\\"in\\\" in x: # initialization experiments - negative p0\\n if compact>1:\\n metrics_stepwise, distributions_stepwise,\\\\\\n metrics_converged, distributions_converged,\\\\\\n metrics_stepwise_allreps, distributions_stepwise_allreps,\\\\\\n metrics_converged_allreps, distributions_converged_allreps = test_p0_neg(pxy,compact)\\n metrics_converged.to_csv(results_path+'metrics_converged_p0_neg.csv')\\n metrics_stepwise.to_csv(results_path+'metrics_stepwise_p0_neg.csv')\\n distributions_converged.to_pickle(results_path+'distributions_converged_p0_neg.pkl')\\n elif compact>0:\\n metrics_stepwise,\\\\\\n metrics_converged, distributions_converged,\\\\\\n metrics_stepwise_allreps,\\\\\\n metrics_converged_allreps, distributions_converged_allreps = test_p0_neg(pxy,compact)\\n metrics_converged.to_csv(results_path+'metrics_converged_p0_neg.csv')\\n metrics_stepwise.to_csv(results_path+'metrics_stepwise_p0_neg.csv')\\n distributions_converged.to_pickle(results_path+'distributions_converged_p0_neg.pkl')\\n else:\\n metrics_stepwise,\\\\\\n metrics_converged,\\\\\\n metrics_stepwise_allreps,\\\\\\n metrics_converged_allreps = test_p0_neg(pxy,compact)\\n metrics_converged.to_csv(results_path+'metrics_converged_p0_neg.csv')\\n metrics_stepwise.to_csv(results_path+'metrics_stepwise_p0_neg.csv')\\n if \\\"c\\\" in x: # convergence tolerance experiments\\n if compact>1:\\n metrics_stepwise, distributions_stepwise,\\\\\\n metrics_converged, distributions_converged,\\\\\\n metrics_stepwise_allreps, distributions_stepwise_allreps,\\\\\\n metrics_converged_allreps, distributions_converged_allreps = test_ctol(pxy,compact)\\n metrics_converged.to_csv(results_path+'metrics_converged_ctol.csv')\\n metrics_stepwise.to_csv(results_path+'metrics_stepwise_ctol.csv')\\n distributions_converged.to_pickle(results_path+'distributions_converged_ctol.pkl')\\n elif compact>0:\\n metrics_stepwise,\\\\\\n metrics_converged, distributions_converged,\\\\\\n metrics_stepwise_allreps,\\\\\\n metrics_converged_allreps, distributions_converged_allreps = test_ctol(pxy,compact)\\n metrics_converged.to_csv(results_path+'metrics_converged_ctol.csv')\\n metrics_stepwise.to_csv(results_path+'metrics_stepwise_ctol.csv')\\n distributions_converged.to_pickle(results_path+'distributions_converged_ctol.pkl')\\n else:\\n metrics_stepwise,\\\\\\n metrics_converged,\\\\\\n metrics_stepwise_allreps,\\\\\\n metrics_converged_allreps = test_ctol(pxy,compact)\\n metrics_converged.to_csv(results_path+'metrics_converged_ctol.csv')\\n metrics_stepwise.to_csv(results_path+'metrics_stepwise_ctol.csv')\\n if \\\"z\\\" in x: # zeroL tolerance experiments\\n if compact>1:\\n metrics_stepwise, distributions_stepwise,\\\\\\n metrics_converged, distributions_converged,\\\\\\n metrics_stepwise_allreps, distributions_stepwise_allreps,\\\\\\n metrics_converged_allreps, distributions_converged_allreps = test_zeroLtol(pxy,compact)\\n metrics_converged.to_csv(results_path+'metrics_converged_zeroLtol.csv')\\n metrics_stepwise.to_csv(results_path+'metrics_stepwise_zeroLtol.csv')\\n distributions_converged.to_pickle(results_path+'distributions_converged_zeroLtol.pkl')\\n elif compact>0:\\n metrics_stepwise,\\\\\\n metrics_converged, distributions_converged,\\\\\\n metrics_stepwise_allreps,\\\\\\n metrics_converged_allreps, distributions_converged_allreps = test_zeroLtol(pxy,compact)\\n metrics_converged.to_csv(results_path+'metrics_converged_zeroLtol.csv')\\n metrics_stepwise.to_csv(results_path+'metrics_stepwise_zeroLtol.csv')\\n distributions_converged.to_pickle(results_path+'distributions_converged_zeroLtol.pkl')\\n else:\\n metrics_stepwise,\\\\\\n metrics_converged,\\\\\\n metrics_stepwise_allreps,\\\\\\n metrics_converged_allreps = test_zeroLtol(pxy,compact)\\n metrics_converged.to_csv(results_path+'metrics_converged_zeroLtol.csv')\\n metrics_stepwise.to_csv(results_path+'metrics_stepwise_zeroLtol.csv')\\n if \\\"b\\\" in x: # trying proposed optimal beta\\n if compact>1:\\n metrics_stepwise, distributions_stepwise,\\\\\\n metrics_converged, distributions_converged,\\\\\\n metrics_stepwise_allreps, distributions_stepwise_allreps,\\\\\\n metrics_converged_allreps, distributions_converged_allreps = test_IB(pxy,compact)\\n metrics_converged.to_csv(results_path+'metrics_converged_bestbeta.csv')\\n metrics_stepwise.to_csv(results_path+'metrics_stepwise_bestbeta.csv')\\n distributions_converged.to_pickle(results_path+'distributions_converged_bestbeta.pkl')\\n elif compact>0:\\n metrics_stepwise,\\\\\\n metrics_converged, distributions_converged,\\\\\\n metrics_stepwise_allreps,\\\\\\n metrics_converged_allreps, distributions_converged_allreps = test_IB(pxy,compact)\\n metrics_converged.to_csv(results_path+'metrics_converged_bestbeta.csv')\\n metrics_stepwise.to_csv(results_path+'metrics_stepwise_bestbeta.csv')\\n distributions_converged.to_pickle(results_path+'distributions_converged_bestbeta.pkl')\\n else:\\n metrics_stepwise,\\\\\\n metrics_converged,\\\\\\n metrics_stepwise_allreps,\\\\\\n metrics_converged_allreps = test_IB(pxy,compact)\\n metrics_converged.to_csv(results_path+'metrics_converged_bestbeta.csv')\\n metrics_stepwise.to_csv(results_path+'metrics_stepwise_bestbeta.csv')\\n return 0\",\n \"def run_tests(): \\n \\n\\n nextdata = [[21, 61, 42, 30], [33,45, 18, 29]]\\n\\n for xval, yval, snum, expect in nextdata:\\n\\n pmachine = PMachine()\\n pmachine.serial_number = snum\\n pmachine.run2_completion()\\n result = pmachine.calc_square_total(xval, yval, showsquare=True)\\n assert result == expect\\n print(\\\"Got value {}={} as expected\\\".format(result, expect))\",\n \"def run_test(models, data):\\n num_correct = 0.0\\n num_total = 0.0\\n\\n # initialize lists to put in predictions & results for confusion matrix \\n predicted_labels = []\\n actual_labels = []\\n langlist = models.keys()\\n for ai, actual_lang in enumerate(langlist):\\n test_files = open(os.path.join('traintestsplit', actual_lang+'.testlist')).read().split()\\n print 'Testing', len(test_files), 'files from', actual_lang\\n for filename in test_files:\\n logprobs = {} # dict: total log prob of this file under each model \\n for test_lang in langlist:\\n logprobs[test_lang] = apply_model(models[test_lang], data[actual_lang][filename+'.npytxt'])\\n predicted_lang = max(logprobs.items(), key=lambda x:x[1])[0]\\n # insert prediction (of lang index) into predicted list \\n predicted_labels.append(langlist.index(predicted_lang))\\n actual_labels.append(ai)\\n if actual_lang == predicted_lang:\\n num_correct += 1\\n num_total += 1\\n\\n print len(filter(lambda x:x==ai, predicted_labels[-len(test_files):])), 'correct'\\n\\n print\\n print 'Accuracy', num_correct*100/num_total\\n\\n #CONFUSION MATRIX (y_test, y_pred) -> (actual label, predictions) \\n cm = confusion_matrix(actual_labels, predicted_labels)\\n cm_normalized = cm.astype('float') / cm.sum(axis=1)[:, np.newaxis]\\n # display confusion stats by lang (TODO: visualize with matplotlib) \\n print '*'*20\\n for ai, actual_lang in enumerate(langlist):\\n print actual_lang, 'confusion:'\\n for pi, predicted_lang in enumerate(langlist):\\n print '{0}: {1:.2f}%'.format(predicted_lang, cm_normalized[ai, pi]*100)\\n print '*'*20\",\n \"def _run_evaluation(\\n sess, experiment, eval_config, output_dir, min_range, max_range, num_bins,\\n torsion_bins):\\n tf.io.gfile.makedirs(os.path.join(output_dir, 'pickle_files'))\\n\\n logging.info('Eval config is %s\\\\nnum_bins: %d', eval_config, num_bins)\\n num_examples = 0\\n num_crops = 0\\n start_all_time = time.time()\\n\\n # Either do the whole test set, or up to a specified limit.\\n max_examples = experiment.num_eval_examples\\n if eval_config.max_num_examples > 0:\\n max_examples = min(max_examples, eval_config.max_num_examples)\\n\\n while num_examples < max_examples:\\n one_prediction = compute_one_prediction(\\n num_examples, experiment, sess, eval_config, num_bins, torsion_bins)\\n\\n single_message = one_prediction.single_message\\n num_crops_local = one_prediction.num_crops_local\\n sequence = one_prediction.sequence\\n filebase = one_prediction.filebase\\n softmax_probs = one_prediction.softmax_probs\\n ss = one_prediction.ss\\n asa = one_prediction.asa\\n torsions = one_prediction.torsions\\n\\n num_examples += 1\\n num_crops += num_crops_local\\n\\n # Save the output files.\\n filename = os.path.join(output_dir,\\n 'pickle_files', '%s.pickle' % filebase)\\n distogram_io.save_distance_histogram(\\n filename, softmax_probs, filebase, sequence,\\n min_range=min_range, max_range=max_range, num_bins=num_bins)\\n\\n if experiment.model.torsion_multiplier > 0:\\n torsions_dir = os.path.join(output_dir, 'torsions')\\n tf.io.gfile.makedirs(torsions_dir)\\n distogram_io.save_torsions(torsions_dir, filebase, sequence, torsions)\\n\\n if experiment.model.secstruct_multiplier > 0:\\n ss_dir = os.path.join(output_dir, 'secstruct')\\n tf.io.gfile.makedirs(ss_dir)\\n secstruct.save_secstructs(ss_dir, filebase, None, sequence, ss)\\n\\n if experiment.model.asa_multiplier > 0:\\n asa_dir = os.path.join(output_dir, 'asa')\\n tf.io.gfile.makedirs(asa_dir)\\n secstruct.save_secstructs(asa_dir, filebase, None, sequence,\\n np.expand_dims(asa, 1), label='Deepmind 2D ASA')\\n\\n time_spent = time.time() - start_all_time\\n logging.info(\\n 'Evaluate %d examples, %d crops %.1f crops/ex. '\\n 'Took %.1fs, %.3f s/example %.3f crops/s\\\\n%s',\\n num_examples, num_crops, num_crops / float(num_examples), time_spent,\\n time_spent / num_examples, num_crops / time_spent, single_message)\\n\\n logging.info('Tested on %d', num_examples)\",\n \"def run(self):\\n if self.pb.xvalEN and not self.isXvalMain:\\n dt = np.ones(self.pb.total_exam_no,\\n dtype=\\\"float32\\\") / self.train_exam_no\\n dt[self.val_indices] = 0.0\\n else:\\n dt = np.ones(self.pb.total_exam_no,\\n dtype=\\\"float32\\\") / self.pb.total_exam_no\\n\\n val = np.zeros(8,dtype=\\\"float32\\\")-1\\n boosting = None\\n wl = None\\n if self.pb.algorithm == 'conf-rated':\\n boosting = ConfidenceRated(self)\\n wl = ConfidenceRatedWL(self)\\n elif self.pb.algorithm == 'adaboost':\\n boosting = AdaBoost(self)\\n wl = AdaBoostWL(self)\\n elif self.pb.algorithm == 'adaboost-fast':\\n boosting = AdaBoostFast(self)\\n wl = AdaBoostFastWLMPI(self)\\n elif self.pb.algorithm == 'rankboost':\\n boosting = RankBoost(self)\\n wl = ConfidenceRatedWL(self)\\n elif self.pb.algorithm == 'rankboost-fast':\\n boosting = RankBoost(self)\\n wl = AdaBoostFastWL(self)\\n else:\\n raise Exception(\\\"Unknown Boosting Algorithm\\\")\\n \\n for r in range(self.pb.rounds):\\n tree = wl.run(dt)\\n dt = boosting.run(dt = dt,\\n r = r,\\n tree = tree)\\n \\n if self.isXvalMain:\\n boosting.finalize()\\n \\n \\\"\\\"\\\"Sync the predictions and save them to a file\\\"\\\"\\\"\\n if self.pb.isLeader:\\n if self.pb.xvalEN and not self.isXvalMain:\\n val_predictions = boosting.get_val_predictions()\\n hypotheses = boosting.get_hypotheses()\\n np.savez(self.out_fp,\\n val_predictions = val_predictions,\\n hypotheses = hypotheses,\\n )\\n if self.pb.testEN and self.isXvalMain:\\n train_predictions = np.zeros([self.pb.total_exam_no,self.pb.rounds],\\n dtype=\\\"float32\\\")\\n test_predictions = np.zeros([self.pb.test_exam_no,self.pb.rounds],\\n dtype=\\\"float32\\\")\\n for slv in np.arange(self.pb.comm_size):\\n tr_i1 = self.pb.train_partition[slv]\\n tr_i2 = self.pb.train_partition[slv+1]\\n te_i1 = self.pb.test_partition[slv]\\n te_i2 = self.pb.test_partition[slv+1]\\n if slv == 0:\\n train_predictions[tr_i1:tr_i2,:] = boosting.get_train_predictions()\\n test_predictions[te_i1:te_i2,:] = boosting.get_test_predictions()\\n else:\\n train_predictions[tr_i1:tr_i2,:] = self.pb.comm.recv(source=slv,tag=11)\\n test_predictions[te_i1:te_i2,:] = self.pb.comm.recv(source=slv,tag=12)\\n hypotheses = boosting.get_hypotheses()\\n np.savez(self.out_fp,\\n train_predictions = train_predictions,\\n test_predictions = test_predictions,\\n hypotheses = hypotheses,\\n )\\n if not self.pb.testEN and self.isXvalMain:\\n train_predictions = np.zeros([self.pb.total_exam_no,self.pb.rounds],\\n dtype=\\\"float32\\\")\\n for slv in np.arange(self.pb.comm_size):\\n tr_i1 = self.pb.train_partition[slv]\\n tr_i2 = self.pb.train_partition[slv+1]\\n if slv == 0:\\n train_predictions[tr_i1:tr_i2,:] = boosting.get_train_predictions()\\n else:\\n train_predictions[tr_i1:tr_i2,:] = self.pb.comm.recv(source=slv,tag=11)\\n hypotheses = boosting.get_hypotheses()\\n np.savez(self.out_fp,\\n train_predictions = train_predictions,\\n hypotheses = hypotheses,\\n )\\n else:\\n if self.pb.testEN and self.isXvalMain:\\n train_predictions = boosting.get_train_predictions()\\n self.pb.comm.send(train_predictions,dest = 0,tag=11)\\n test_predictions = boosting.get_test_predictions()\\n self.pb.comm.send(test_predictions,dest = 0,tag=12)\\n if not self.pb.testEN and self.isXvalMain:\\n train_predictions = boosting.get_train_predictions()\\n self.pb.comm.send(train_predictions,dest = 0,tag=11)\",\n \"def print_results(self, f1_score, precision_score, recall_score):\\n print(\\\"Algorithm: %s\\\" % self.name)\\n self.print_score(f1_score, \\\"F1 Score\\\")\\n self.print_score(precision_score, \\\"Precision Score\\\")\\n self.print_score(recall_score, \\\"Recall Score\\\")\",\n \"def run(self):\\n evaluateAllRuns = False\\n while True:\\n if self.host == \\\"\\\":\\n # respond to clients\\n self.respond2Clients()\\n else:\\n print(\\\"Next...\\\")\\n # randomly choose experiment + run\\n if not evaluateAllRuns:\\n print(\\\"Randomly fetching run\\\")\\n self.exp, self.runnum, self.detname = randExpRunDet()\\n else:\\n\\t\\t try:\\n print(\\\"Fecthing next run in experiment\\\")\\n self.exp, self.runnum, self.detname = nextExpRunDet(self.goodExp, self.runList[0])\\n if self.exp is None:\\n self.runList.pop(0)\\n continue\\n except:\\n evaluateAllRuns = False\\n continue\\n if not self.checkStatus(self.exp, self.runnum, self.detname):\\n print \\\"trying: exp %s, run %s, det %s\\\"%(self.exp,self.runnum,self.detname)\\n try: #temp\\n self.ds = safeDataSource(self.exp, self.runnum)\\n except: #temp\\n continue #temp\\n self.run = self.ds.runs().next()\\n self.times = self.run.times()\\n #Start temp code\\n if self.detname is None:\\n continue\\n #End temp code\\n self.det = psana.Detector(self.detname)\\n self.det.do_reshape_2d_to_3d(flag=True)\\n try:\\n self.iX = np.array(self.det.indexes_x(self.run), dtype=np.int64)\\n self.iY = np.array(self.det.indexes_y(self.run), dtype=np.int64)\\n self.ipx, self.ipy = self.det.point_indexes(self.run, pxy_um=(0, 0))\\n self.alg = PyAlgos()\\n self.alg.set_peak_selection_pars(npix_min=2, npix_max=30, amax_thr=300, atot_thr=600, son_min=10)\\n mask = self.det.mask(self.runnum, calib=True, status=True, edges=True, central=True, unbond=True, unbondnbrs=True)\\n\\n samples = np.linspace(0, len(self.times), num=100, endpoint=False, retstep=False, dtype='int')\\n offset = np.floor(np.random.uniform(0, len(self.times)-samples[-1])).astype('int')\\n mysamples = samples + offset\\n numCrystals = 0\\n for self.eventNum in mysamples:\\n self.evt = self.run.event(self.times[self.eventNum])\\n calib = self.det.calib(self.evt)\\n if calib is not None:\\n peaks = self.alg.peak_finder_v3r3(calib, rank=3, r0=3, dr=2, nsigm=10, mask=mask.astype(np.uint16))\\n if self.likelihood(peaks) >= self.goodLikelihood:\\n numCrystals += 1\\n if numCrystals >= self.minCrystals:\\n self.numSaved +=1\\n self.updateStatus(self.exp, self.runnum, self.detname, self.numSaved)\\n self.lastGood = True\\n break\\n except:\\n print \\\"Could not analyse this run\\\"\\n #If an experiment has not had all of its runs evaluated yet\\n # and if the last randomly selected run in this experiment was good\\n # then all the runs in this experiment should be evaluated\\n if (self.exp not in self.goodList) and self.lastGood:\\n self.goodExp = self.exp #Save the name of this experiment\\n self.goodRun = self.runnum #Save the run that has already been evaluated\\n self.lastGood = False #Reset the condition that the last run was \\\"good\\\"\\n self.goodList.append(self.goodExp) #Add this experiment name to the list of experiments that have had all runs evaluated\\n self.runList = returnRunList(self.goodExp, self.goodRun) #save list of all runs in this good exp\\n evaluateAllRuns = True #rerun loop with new algorithm that evaluates each run in an experiment\\n continue\\n if evaluateAllRuns: #If the loop is currently evaluating all of the runs in an experiment\\n if(len(self.runList) > 1):\\n self.runList.pop(0) #Remove runs from the list of runs each time they are evaluated\\n else:\\n self.runList.pop(0)#Remove runs until the list is completely empty\\n evaluateAllRuns = False #Stop evaluated all the runs of an experiment, go back to random fetching\",\n \"def testDriver():\\n exam1=90\\n exam2=85\\n assignmentScores = [50, 60, 70, 80, ]\\n computeGrades(exam1, exam2, assignmentScores)\",\n \"def sbm_experiments():\\n # The key experiment is to fix n, r, and p and vary the ratio q/p.\\n n = 1000\\n average_over = 10\\n\\n # We will run a few experiments\\n rs = [5]\\n ps = [1e-11]\\n run_lps = [False]\\n\\n # Use the same ratios for each experiment\\n ratios = [0.5 * x for x in range(4, 61)]\\n\\n # Whether to append results to the results files\\n append_results = False\\n\\n for index in range(len(rs)):\\n sbm_experiment_internal(n, rs[index], ps[index], average_over, run_lps[index], ratios, append_results)\",\n \"def run(self):\\n if self.pb.xvalEN and not self.isXvalMain:\\n dt = np.ones(self.pb.total_exam_no,\\n dtype=\\\"float32\\\") / self.train_exam_no\\n dt[self.val_indices] = 0.0\\n else:\\n dt = np.ones(self.pb.total_exam_no,\\n dtype=\\\"float32\\\") / self.pb.total_exam_no\\n\\n val = np.zeros(8,dtype=\\\"float32\\\")-1\\n boosting = None\\n wl = None\\n if self.pb.algorithm == 'conf-rated':\\n boosting = ConfidenceRated(self)\\n wl = ConfidenceRatedWL(self)\\n elif self.pb.algorithm == 'adaboost':\\n boosting = AdaBoost(self)\\n wl = AdaBoostWL(self)\\n elif self.pb.algorithm == 'adaboost-fast':\\n boosting = AdaBoostFast(self)\\n wl = AdaBoostFastWL(self)\\n elif self.pb.algorithm == 'rankboost':\\n boosting = RankBoost(self)\\n wl = ConfidenceRatedWL(self)\\n elif self.pb.algorithm == 'rankboost-fast':\\n boosting = RankBoost(self)\\n wl = AdaBoostFastWL(self)\\n else:\\n raise Exception(\\\"Unknown Boosting Algorithm\\\")\\n \\n for r in range(self.pb.rounds):\\n tree = wl.run(dt)\\n dt = boosting.run(dt = dt,\\n r = r,\\n tree = tree)\\n \\n if self.isXvalMain:\\n boosting.finalize()\\n \\n \\\"\\\"\\\"Sync the predictions and save them to a file\\\"\\\"\\\"\\n if self.pb.isLeader:\\n if self.pb.xvalEN and not self.isXvalMain:\\n val_predictions = boosting.get_val_predictions()\\n hypotheses = boosting.get_hypotheses()\\n np.savez(self.out_fp,\\n val_predictions = val_predictions,\\n hypotheses = hypotheses,\\n )\\n if self.pb.testEN and self.isXvalMain:\\n train_predictions = boosting.get_train_predictions()\\n test_predictions = boosting.get_test_predictions()\\n hypotheses = boosting.get_hypotheses()\\n np.savez(self.out_fp,\\n train_predictions = train_predictions,\\n test_predictions = test_predictions,\\n hypotheses = hypotheses,\\n )\\n if not self.pb.testEN and self.isXvalMain:\\n train_predictions = boosting.get_train_predictions()\\n hypotheses = boosting.get_hypotheses()\\n np.savez(self.out_fp,\\n train_predictions = train_predictions,\\n hypotheses = hypotheses,\\n )\",\n \"def run_experiment():\\n pass\",\n \"def test_compare(self):\\n config = {\\n 'num_components': 512,\\n 'num_features': 128,\\n 'covariance': 'spherical'\\n }\\n\\n samples = self.generate_samples(config, 100_000)\\n sklearn_time = np.mean([self.train_sklearn(config, samples) for _ in range(3)])\\n ours_cpu_time = np.mean([self.train_ours(config, samples) for _ in range(3)])\\n ours_gpu_time = np.mean([\\n self.train_ours(config, samples.cuda(), gpu=True) for _ in range(3)\\n ])\\n\\n print(f\\\"-------------------------------------\\\")\\n print(f\\\"Speedup of CPU implementation: {sklearn_time / ours_cpu_time:.2f}\\\")\\n print(f\\\"Speedup of GPU implementation: {sklearn_time / ours_gpu_time:.2f}\\\")\\n print(f\\\"-------------------------------------\\\")\",\n \"def test(self, plot=False):\\n accuracy_list = []\\n fobj_avg = self.load_stats()\\n\\n for ii in range(settings.PARS['maxIters']):\\n model = self.load_model(ii)\\n D1 = model['D']\\n W1 = model['W']\\n\\n # classification\\n tic = time.time()\\n accuracy_list.append(self.classification(D1, W1)[1])\\n toc = time.time()\\n print(\\n 'Final recognition rate for OnlineDL is : {} , objective function value: {}, time: {}'\\n .format(accuracy_list[ii], fobj_avg[ii], toc-tic)\\n )\\n\\n accuracy_list = np.asarray(accuracy_list)\\n\\n print('Best recognition rate for OnlineDL is {} at iteration {}'.format(\\n accuracy_list.max(), accuracy_list.argmax()))\\n\\n if plot:\\n # plot the objective function values for all iterations\\n plt.clf()\\n plt.plot(list(fobj_avg.keys()), list(fobj_avg.values()), 'mo--', linewidth=2)\\n plt.xlabel('Iterations')\\n plt.ylabel('Average objective function value')\\n plt.xticks(list(range(0, 20)), list(range(1, 21)))\\n plt.show()\\n\\n plt.clf()\\n plt.plot(accuracy_list, 'rs--', linewidth=2)\\n plt.xticks(list(range(0, 20)), list(range(1, 21)))\\n plt.xlabel('Iterations')\\n plt.ylabel('Accuracy')\\n plt.show()\",\n \"def run_tests(self):\\n\\n self.test_report = []\\n\\n #dict of unsorted lists\\n dict_of_un_lists = self.dict_un_lists_intersection_test(self.data_dict)\\n self.test_report.append(dict_of_un_lists)\\n\\n #dict of sets\\n dict_of_sets = self.build_dict_of_sets(self.data_dict)\\n self.test_report.append(self.dict_sets_intersection_test(dict_of_sets))\\n\\n #pandas - experimental and probably not the way to use pandas\\n # dict_of_pandas = self.build_dict_of_panda_series(self.data_dict)\\n # self.test_report.append(self.dicts_any_intersection_node_test(dict_of_pandas))\\n\\n # print results\\n\\n if self.verbose:\\n self.print_tests_results()\",\n \"def main(titles, paragraphs, printDetailedStats = True):\\n avg_precision = 0 # all the precision values will be added to this value. In the end, this value will be divided by\\n # the number of comparisons analysed to report the average precision\\n avg_recall = 0\\n avg_fmeasure = 0\\n\\n if paragraphs:\\n manual_directory = path.MANUAL_PARAGRAPHS\\n auto_directory = path.OUTDIR_PARAGRAPHS\\n else:\\n manual_directory = path.MANUAL_SENTENCES\\n auto_directory = path.OUTDIR_SENTENCES\\n\\n for title in titles:\\n with open(manual_directory + title) as m:\\n with open(auto_directory + title) as a:\\n tPositive, fPositive, fNegative, precision, recall, fmeasure = \\\\\\n compare(m.readlines(), a.readlines(), paragraphs)\\n avg_precision += precision\\n avg_recall += recall\\n avg_fmeasure += fmeasure\\n if printDetailedStats: # then report precision and recall for every individual article\\n print(\\\"comparing \\\" + title)\\n print(\\\"tp=\\\" + str(tPositive) + \\\" fn=\\\" + str(fNegative) + \\\" fp=\\\" + str(fPositive))\\n print(\\\"precision=\\\" + str(round(precision, 5)) + \\\"\\\\t\\\\t recall=\\\" + str(\\n round(recall, 5)) + \\\"\\\\t\\\\t fmeasure=\\\" + str(round(fmeasure, 5)) + \\\" \\\\n\\\\n\\\")\\n print(\\\"AVERAGE_PRECISION=\\\" + str(round(avg_precision/len(titles), 5)) + \\\"\\\\t\\\\t AVERAGE_RECALL=\\\"\\n + str(round(avg_recall/len(titles), 5)) + \\\"\\\\t\\\\t AVERAGE_F_MEASURE=\\\"\\n + str(round(avg_fmeasure/len(titles), 5)))\\n return avg_fmeasure/len(titles)\",\n \"def test_run_multi_r__(self):\\n\\n # Test Description\\n # ================\\n #\\n # 1. This test intialises an example *eopy.matchup.matchupIO.MatchUp* object\\n #\\n # 2. Compare transformed dataset to expected value\\n\\n ################################################################################################################\\n # 1. Initialise Test Data Object\\n ################################################################################################################\\n\\n MatchUpTest = return_MatchUpTest_r__()\\n\\n ################################################################################################################\\n # 2. Define expected values\\n ################################################################################################################\\n\\n # Original dataset values (should be unchanged)\\n MatchUpOriginal_expected = return_MatchUpTest_r__()\\n\\n # Transformed dataset\\n values_expected = array([294.0625, 480.3733333, 300.6, 227.3846154, 210.1533333,\\n 22.74193548, 22.0625, 21.96875, 22.80645161, 23.5,\\n 21.66666667, 21.05882353, 23, 22.40625,\\n 38.33333333, 36.63636364, 36.5, 38.42857143,\\n 30.1, 32.14893617, 29.37254902, 28.88461538, 28.56603774,\\n 33.45238095, 32.81395349, 31.77272727, 32.60465116,\\n 40.125, 43.54054054, 38.59090909, 34.08510638,\\n 13.72727273, 12, 14.1, 11.79069767, 17.53846154,\\n 12.69565217, 31.16666667, 12.26086957, 11.52272727,\\n 8.8125, 12, 7.4, 10.13207547])\\n unc_expected = [Uncertainty(1, array([1.6, 1.5, 1.5, 1.3, 1.5])),\\n Uncertainty(1, array([3.1, 3.2, 3.2, 3.1, 3.0])),\\n Uncertainty(1, array([3.3, 3.4, 3.1, 3.2])),\\n Uncertainty(1, array([2.1, 2.2, 2.2, 2.1])),\\n Uncertainty(1, array([5.0, 4.7, 5.1, 5.2, 5.3])),\\n Uncertainty(1, array([4.2, 4.3, 4.4, 4.3])),\\n Uncertainty(1, array([4.0, 3.7, 4.4, 4.7])),\\n Uncertainty(1, array([2.2, 1.7, 2.0, 4.3, 2.6])),\\n Uncertainty(1, array([2.3, 1.2, 2.3, 4.4])),\\n Uncertainty(1, array([3.2, 2.7, 3.0, 5.3]))]\\n ks_expected = array([4.8, 6.8, 5.2, 5.6, 5.2, 12.10287443, 13.99394856, 12.48108926, 12.85930408])\\n unck_expected = [Uncertainty(1, array([0.25, 0.25, 0.25, 0.25, 0.25])),\\n Uncertainty(1, array([0.2644, 0.2644, 0.2644, 0.2644]))]\\n idx_expected = {\\\"Nm\\\": [5, 4],\\n \\\"cNm\\\": [0, 5, 9],\\n \\\"Im\\\": [[0, 1], [1, 2]],\\n \\\"sensors\\\": [-1, 1, 2],\\n \\\"sensor_ms\\\": [1, 3, 3],\\n \\\"n_sensor\\\": [0, 1, 1, 2, 1, 1, 2, 1, 1, 2],\\n \\\"n_mu\\\": [1, 1, 2, 2, 1, 2, 2, 1, 2, 2],\\n \\\"n_cov\\\": [1, 1, 1, 1, 2, 2, 2, 3, 3, 3],\\n \\\"N_var\\\": [5, 5, 4, 4, 5, 4, 4, 5, 4, 4],\\n \\\"idx\\\": [0, 5, 10, 14, 18, 23, 27, 31, 36, 40, 44],\\n \\\"Ia\\\": [1, 1, 1, 2, 2, 2]}\\n a_expected = array([1., 1.3, 0.002, 0.5, 1.1, 0.0005])\\n w_matrices_expected = []\\n u_matrices_expected = []\\n\\n ################################################################################################################\\n # 3. Run Transform2NormInd.run()\\n ################################################################################################################\\n\\n Transform2NormIndOp = Transform2NormInd()\\n MatchUpTransform = Transform2NormIndOp.run(MatchUpTest)\\n\\n values_test = MatchUpTransform.values\\n unc_test = MatchUpTransform.unc\\n w_matrices_test = MatchUpTransform.w_matrices\\n u_matrices_test = MatchUpTransform.u_matrices\\n ks_test = MatchUpTransform.ks\\n unck_test = MatchUpTransform.unck\\n idx_test = MatchUpTransform.idx\\n\\n ################################################################################################################\\n # 4. Compare retrieve values to expect values\\n ################################################################################################################\\n\\n # Test transformed data object attribute by attribute\\n\\n # a. values\\n for i, (value_expected, value_test) in enumerate(zip(values_expected, values_test)):\\n self.assertAlmostEqual(value_expected, value_test, places=5, msg=str(i))\\n\\n # b. unc\\n for block_unc_test, block_unc_expected in zip(unc_test, unc_expected):\\n self.assertEqual(block_unc_expected.typeID, block_unc_test.typeID)\\n self.assertEqual(block_unc_expected.uR.tolist(), block_unc_test.uR.tolist())\\n\\n # c. w_matrices\\n self.assertEqual(w_matrices_test, w_matrices_expected)\\n\\n # d. u_matrices\\n self.assertEqual(u_matrices_test, u_matrices_expected)\\n\\n # e. ks\\n for k_expected, k_test in zip(ks_expected, ks_test):\\n self.assertAlmostEqual(k_test.tolist(), k_expected.tolist(), places=5)\\n\\n # f. unck\\n for block_unck_test, block_unck_expected in zip(unck_test, unck_expected):\\n self.assertEqual(block_unck_expected.typeID, block_unck_test.typeID)\\n self.assertEqual(block_unck_expected.uR.tolist(), block_unck_test.uR.tolist())\\n\\n # h. idx\\n self.assertEqual(set(idx_expected.keys()), set(idx_test.keys()))\\n for key in idx_expected.keys():\\n idx_i_test = idx_test[key]\\n idx_i_expected = idx_expected[key]\\n if isinstance(idx_i_expected, ndarray):\\n self.assertEqual(idx_i_test.tolist(), idx_i_expected.tolist())\\n else:\\n self.assertEqual(idx_i_test, idx_i_expected)\\n\\n # Test original data object preserved attribute by attribute\\n\\n # a. values\\n for i, (value_original_expected, value_original_test) in enumerate(zip(MatchUpOriginal_expected.values, MatchUpTest.values)):\\n self.assertAlmostEqual(value_original_expected, value_original_test, places=5)\\n\\n # b. unc\\n for block_unc_original_expected, block_unc_original_test in zip(MatchUpOriginal_expected.unc, MatchUpTest.unc):\\n self.assertEqual(block_unc_original_expected.typeID, block_unc_original_test.typeID)\\n self.assertEqual(block_unc_original_expected.uR.tolist(), block_unc_original_test.uR.tolist())\\n\\n # c. w_matrices\\n self.assertEqual(MatchUpOriginal_expected.w_matrices, MatchUpTest.w_matrices)\\n\\n # d. u_matrices\\n self.assertEqual(MatchUpOriginal_expected.u_matrices, MatchUpTest.u_matrices)\\n\\n # e. ks\\n for k_original_expected, k_original_test in zip(MatchUpOriginal_expected.ks, MatchUpTest.ks):\\n self.assertAlmostEqual(k_original_test.tolist(), k_original_expected.tolist(), places=5)\\n\\n # f. unck\\n for block_unck_original_expected, block_unck_original_test in zip(MatchUpOriginal_expected.unck, MatchUpTest.unck):\\n self.assertEqual(block_unck_original_expected.typeID, block_unck_original_test.typeID)\\n self.assertEqual(block_unck_original_expected.uR.tolist(), block_unck_original_test.uR.tolist())\\n\\n # h. idx\\n self.assertEqual(set(MatchUpOriginal_expected.idx), set(MatchUpTest.idx))\\n for key in MatchUpOriginal_expected.idx.keys():\\n idx_i_original_test = MatchUpTest.idx[key]\\n idx_i_original_expected = MatchUpOriginal_expected.idx[key]\\n if isinstance(idx_i_original_expected, ndarray):\\n self.assertEqual(idx_i_original_test.tolist(), idx_i_original_expected.tolist())\\n else:\\n self.assertEqual(idx_i_original_test, idx_i_original_expected)\",\n \"def run():\\n tests = [\\n # [\\n # [\\\"\\\", \\\"sumRange\\\", \\\"sumRange\\\", \\\"sumRange\\\", \\\"update\\\", \\\"sumRange\\\", \\\"sumRange\\\", \\\"sumRange\\\"],\\n # [[1, 2, 3, 4, 5], [0, 2], [1, 2], [1, 3], [1, 3], [0, 2], [1, 2], [0, 1]]\\n # ],\\n [\\n [\\\"NumArray\\\", \\\"sumRange\\\", \\\"update\\\", \\\"sumRange\\\"],\\n [[-1], [0, 0], [0, 1], [0, 0]]\\n ]\\n ]\\n\\n results = [\\n # [None, 3, 2, 5, None, 4, 2, 2],\\n [None, -1, None, 1],\\n\\n ]\\n for test, res in zip(tests, results):\\n print('-----'*20)\\n cmds, params = test\\n cmds.pop(0)\\n p = params.pop(0)\\n res.pop(0)\\n obj = BIT(p)\\n for cmd, param, exp in zip(cmds, params, res):\\n v = getattr(obj, cmd)(*param)\\n print(obj)\\n print(f'------\\\\n{cmd}, {param} --> {v} : {exp}')\\n assert exp == v\",\n \"def log_evaluation(tester, name, description):\\r\\n\\tfor dataset, output in tester.preds.items():\\r\\n\\t\\tresults = pandas.DataFrame.from_dict(output)\\r\\n\\t\\tpath = os.path.join(\\r\\n\\t\\t\\tEXPERIMENT_PATH, tester.config[\\\"name\\\"] + '-' + dataset)\\r\\n\\t\\twith open(path + \\\".csv\\\", \\\"w\\\") as f:\\r\\n\\t\\t\\tresults.to_csv(f, sep=\\\"\\\\t\\\", encoding='utf-8',\\r\\n\\t\\t\\t\\tfloat_format='%.3f', index=False)\",\n \"def eval(self):\\n\\n # How many questions we get right at precision@1.\\n correct = 0\\n\\n total = self._analogy_questions.shape[0]\\n start = 0\\n while start < total:\\n limit = start + 2500\\n sub = self._analogy_questions[start:limit, :]\\n idx = self._predict(sub)\\n start = limit\\n for question in xrange(sub.shape[0]):\\n if sub[question, 3] in idx[question]:\\n # print(sub[question, 3], idx[question])\\n correct += 1\\n\\n print()\\n print(\\\"Eval %4d/%d accuracy @ top5= %4.1f%%\\\" % (correct, total,\\n correct * 100. / total)\\n )\",\n \"def perform_experiments(problem_type, output_dir, load_experiments):\\n\\n experiment_file_path = output_dir + \\\"experiments.pickle\\\"\\n\\n # data structure where to store all the problems, and the experimental results for each algorithm: solutions, final values and times\\n experiment_dict = dict()\\n\\n # if experiments should be loaded (not repeated), do it if possible\\n if load_experiments:\\n with open(experiment_file_path, \\\"rb\\\") as _file:\\n experiment_dict = pickle.load(_file)\\n\\n # perform experiments if pre-existing results were not loaded\\n if not experiment_dict:\\n\\n # given a problem type, create a set of problem instances that are solved (optimally) with manual placements (using actions available for all the algorithms)\\n problems, problem_names, manual_solutions = create_problems(problem_type)\\n\\n if problems and problem_names and manual_solutions:\\n\\n # parameters for the experimentation; note: calculating internal times and value evolution can increase the overall time of algorithms (in a slight, almost neglectible way)\\n execution_num = 10 # 1\\n process_num = 10 # 1\\n calculate_internal_times = True\\n calculate_value_evolution = True\\n\\n start_time = time.time()\\n\\n # solve each problem with each algorithm\\n for i, (problem, problem_name, solution) in enumerate(zip(problems, problem_names, manual_solutions)):\\n\\n experiment_dict[problem_name] = {\\\"problem\\\": problem, \\\"manual_solution\\\": solution, \\\"algorithms\\\": dict()}\\n\\n # solve the problem with different algorithms, executing each one multiple times to gain statistical significance\\n for (algorithm_name, algorithm) in [(\\\"Greedy\\\", greedy.solve_problem), (\\\"Reversible\\\", reversible.solve_problem), (\\\"Evolutionary\\\", evolutionary.solve_problem)]:\\n\\n solutions, values, value_evolutions, times, time_divisions = execute_algorithm(algorithm=algorithm, algorithm_name=algorithm_name, problem=problem, execution_num=execution_num, process_num=process_num, calculate_times=calculate_internal_times, calculate_fitness_stats=calculate_value_evolution)\\n experiment_dict[problem_name][\\\"algorithms\\\"][algorithm_name] = {\\\"solutions\\\": solutions, \\\"values\\\": values, \\\"value_evolutions\\\": value_evolutions, \\\"times\\\": times, \\\"time_divisions\\\": time_divisions}\\n\\n # show the total time spent doing experiments (note that significant overhead can be introduced beyond calculation time if plots are shown or saved to files; for strict time measurements, plotting should be avoided altogether)\\n print(\\\"Total experimental calculation time: {} s\\\".format(round(get_time_since(start_time) / 1000.), 3))\\n\\n if experiment_dict:\\n\\n # experiment-saving parameter\\n save_experiments = True\\n\\n # if possible, save the experiments to a binary file\\n if not load_experiments and save_experiments:\\n with open(experiment_file_path, \\\"wb\\\") as file:\\n pickle.dump(experiment_dict, file)\\n\\n # visualization/saving parameters\\n show_problem_stats = True\\n save_problem_stats = True # False\\n show_manual_solution_plots = True\\n save_manual_solution_plots = True # False\\n show_algorithm_solution_plots = True\\n save_algorithm_solution_plots = True # False\\n show_value_evolution_plots = False\\n save_value_evolution_plots = False # True\\n show_time_division_plots = False\\n save_time_division_plots = False # True\\n show_algorithm_comparison = True\\n save_algorithm_comparison = True # False\\n show_aggregated_result_tables = True\\n save_aggregated_result_tables = True # False\\n\\n # show/save the results of the experiments\\n visualize_and_save_experiments(experiment_dict, output_dir, can_plots_show_value_and_weight=problem_type == KNAPSACK_PACKING_PROBLEM_TYPE, show_problem_stats=show_problem_stats, save_problem_stats=save_problem_stats, show_manual_solution_plots=show_manual_solution_plots, save_manual_solution_plots=save_manual_solution_plots, show_algorithm_solution_plots=show_algorithm_solution_plots, save_algorithm_solution_plots=save_algorithm_solution_plots, show_value_evolution_plots=show_value_evolution_plots, save_value_evolution_plots=save_value_evolution_plots, show_time_division_plots=show_time_division_plots, save_time_division_plots=save_time_division_plots, show_algorithm_comparison=show_algorithm_comparison, save_algorithm_comparison=save_algorithm_comparison, show_aggregated_result_tables=show_aggregated_result_tables, save_aggregated_result_tables=save_aggregated_result_tables)\\n\\n else:\\n print(\\\"The experiments cannot be performed (there are no problems available).\\\")\",\n \"def evaluate(model, test_files):\\n print(\\\"Running predictions.\\\")\\n models = load_model(model)\\n predictions = predict(models, test_files)\\n\\n # # write predictions to file\\n # write_predictions(\\\"evaluate_out.json\\\",predictions)\\n evaluate_individual(predictions, test_files, models)\\n evaluate_overall(predictions)\",\n \"def main():\\n\\n convergence_rates_e1 = pd.DataFrame(index=['Mean', 'St.D'])\\n convergence_rates_e4 = pd.DataFrame(index=['Mean', 'St.D'])\\n success_rates = pd.DataFrame(index=['1e-1', '1e-2', '1e-3', '1e-4', '1e-5'])\\n peak_ratios = pd.DataFrame(index=['1e-1', '1e-2', '1e-3', '1e-4', '1e-5'])\\n\\n for function in range(1, 21):\\n results = pickle.load(open('results/benchmarking_result_{}.pkl'.format(function), 'rb'))\\n\\n col_name = 'F{}'.format(function)\\n index = function-1\\n\\n convergence_rates_e1.insert(index, col_name, results.ConvergenceRates[0])\\n convergence_rates_e4.insert(index, col_name, results.ConvergenceRates[3])\\n\\n success_rates.insert(index, col_name, results.SuccessRate)\\n peak_ratios.insert(index, col_name, results.PeakRatio)\\n\\n for i in results.SimulationSwarms:\\n\\n x_axis = [k for k, _ in i.items()]\\n y_axis = [v for _, v in i.items()]\\n\\n plt.subplot(4, 5, function) # subplot indexes from 1\\n plt.plot(x_axis, y_axis, 'k-')\\n\\n plt.savefig('results/nmmso_benchmark.png')\\n plt.show()\\n\\n pd.set_option('display.max_columns', None) # make sure all columns are printed below\\n print(convergence_rates_e1)\\n print(convergence_rates_e4)\\n print(success_rates)\\n print(peak_ratios)\\n\\n # Table V from Fieldsend et al.\\n table4 = pd.Series([\\n peak_ratios.stack().median(),\\n peak_ratios.stack().mean(),\\n peak_ratios.stack().std()\\n ])\\n print(table4)\\n\\n # do we want to reproduce Fig. 3\",\n \"def getTestResults():\",\n \"def eval_perf_total(model, X_train, y_train, X_test, y_test):\\n\\n y_hat_train = model.predict(X_train)\\n y_hat_test = model.predict(X_test)\\n \\n train_mae = metrics.mean_absolute_error(y_train, y_hat_train)\\n train_mse = metrics.mean_squared_error(y_train, y_hat_train)\\n train_rmse = np.sqrt(metrics.mean_squared_error(y_train, y_hat_train))\\n train_r = metrics.r2_score(y_train, y_hat_train)\\n\\n print('Evaluating Performance on Training Data:\\\\n')\\n print(f' Train Mean Absolute Error: {train_mae:,.2f}')\\n print(f' Train Mean Squared Error: {train_mse:,.2f}\\\\n')\\n print(f'Train Root Mean Squared Error: {train_rmse:,.2f}')\\n print(f'Train R-Square Value: {round(train_r,2)}')\\n\\n print('\\\\n'+'---'*25+'\\\\n')\\n\\n test_mae = metrics.mean_absolute_error(y_test, y_hat_test)\\n test_mse = metrics.mean_squared_error(y_test, y_hat_test)\\n test_rmse = np.sqrt(metrics.mean_squared_error(y_test, y_hat_test))\\n test_r = metrics.r2_score(y_test, y_hat_test)\\n\\n print('Evaluating Performance on Testing Data:\\\\n')\\n print(f' Test Mean Absolute Error: {test_mae:,.2f}')\\n print(f' Test Mean Squared Error: {test_mse:,.2f}\\\\n')\\n print(f'Test Root Mean Squared Error: {test_rmse:,.2f}')\\n print(f'Test R-Square Value: {round(test_r,2)}')\",\n \"def test_precision_recall_curves_vis_api(experiment_to_use):\\n experiment = experiment_to_use\\n probabilities = experiment.probabilities\\n viz_outputs = (\\\"pdf\\\", \\\"png\\\")\\n positive_label = 1\\n with TemporaryDirectory() as tmpvizdir:\\n for viz_output in viz_outputs:\\n vis_output_pattern_pdf = tmpvizdir + f\\\"/*.{viz_output}\\\"\\n visualize.precision_recall_curves(\\n [probabilities, probabilities],\\n experiment.ground_truth,\\n experiment.ground_truth_metadata,\\n experiment.output_feature_name,\\n positive_label,\\n model_names=[\\\"Model1\\\", \\\"Model2\\\"],\\n output_directory=tmpvizdir,\\n file_format=viz_output,\\n )\\n figure_cnt = glob.glob(vis_output_pattern_pdf)\\n assert 1 == len(figure_cnt)\",\n \"def test_net_on_dataset(args, dataset_name, proposal_file, output_dir, multi_gpu=False, gpu_id=0, use_matlab = False, early_stop=False):\\n\\n \\n # print(\\\"test_net_on_dataset\\\")\\n dataset = JsonDataset(dataset_name)\\n test_timer = Timer()\\n \\n test_timer.tic()\\n \\n all_boxes = test_net(args, dataset_name, proposal_file, output_dir, gpu_id=gpu_id, early_stop=early_stop)\\n test_timer.toc()\\n\\n logger.info('Total inference time: {:.3f}s'.format(test_timer.average_time))\\n\\n roidb = dataset.get_roidb()\\n num_images = len(roidb)\\n num_classes = cfg.MODEL.NUM_CLASSES + 1\\n final_boxes = empty_results(num_classes, num_images)\\n test_corloc = 'train' in dataset_name\\n \\n\\n all_cls_scores = {}\\n\\n for i, entry in enumerate(roidb):\\n\\n if early_stop and i > 10: break\\n\\n boxes = all_boxes[entry['image']]\\n \\n cls_key = entry['image'].replace('.jpg','').split('/')[-1]\\n\\n # print(cls_key)\\n\\n if boxes['scores'] is not None:\\n if test_corloc:\\n # print(\\\"corlooking\\\")\\n _, _, cls_boxes_i = box_results_for_corloc(boxes['scores'], boxes['boxes'])\\n else:\\n _, _, cls_boxes_i = box_results_with_nms_and_limit(boxes['scores'], boxes['boxes'])\\n\\n extend_results(i, final_boxes, cls_boxes_i)\\n else:\\n final_boxes = None\\n \\n results = task_evaluation.evaluate_all(dataset, final_boxes, output_dir, test_corloc, use_matlab = use_matlab)\\n return results\",\n \"def print_results(results, random_counterpart=None, random_concepts=None, num_random_exp=100,\\n min_p_val=0.05):\\n\\n # helper function, returns if this is a random concept\\n def is_random_concept(concept):\\n if random_counterpart:\\n return random_counterpart == concept\\n\\n elif random_concepts:\\n return concept in random_concepts\\n\\n else:\\n return 'random500_' in concept\\n\\n # print class, it will be the same for all\\n print(\\\"Class =\\\", results[0]['target_class'])\\n\\n # prepare data\\n # dict with keys of concepts containing dict with bottlenecks\\n result_summary = {}\\n\\n # random\\n random_i_ups = {}\\n\\n for result in results:\\n if result['cav_concept'] not in result_summary:\\n result_summary[result['cav_concept']] = {}\\n\\n if result['bottleneck'] not in result_summary[result['cav_concept']]:\\n result_summary[result['cav_concept']][result['bottleneck']] = []\\n\\n result_summary[result['cav_concept']][result['bottleneck']].append(result)\\n\\n # store random\\n if is_random_concept(result['cav_concept']):\\n if result['bottleneck'] not in random_i_ups:\\n random_i_ups[result['bottleneck']] = []\\n\\n random_i_ups[result['bottleneck']].append(result['i_up'])\\n\\n # print concepts and classes with indentation\\n for concept in result_summary:\\n\\n # if not random\\n if not is_random_concept(concept):\\n print(\\\" \\\", \\\"Concept =\\\", concept)\\n\\n for bottleneck in result_summary[concept]:\\n i_ups = [item['i_up'] for item in result_summary[concept][bottleneck]]\\n\\n # Calculate statistical significance\\n _, p_val = ttest_ind(random_i_ups[bottleneck], i_ups)\\n\\n print(3 * \\\" \\\", \\\"Bottleneck =\\\", (\\\"%s. TCAV Score = %.2f (+- %.2f), \\\"\\n \\\"random was %.2f (+- %.2f). p-val = %.3f (%s)\\\") % (\\n bottleneck, np.mean(i_ups), np.std(i_ups),\\n np.mean(random_i_ups[bottleneck]),\\n np.std(random_i_ups[bottleneck]), p_val,\\n \\\"not significant\\\" if p_val > min_p_val else \\\"significant\\\"))\",\n \"def analyze_end_performance(self) -> None:\\n self.logger.info(f\\\"###### FINAL RESULTS FOR EACH GAME###### \\\\n\\\")\\n total_results = [0, 0, 0, 0]\\n for game in self.instance_predictions:\\n self.logger.info(f\\\"Simulated game {game.home_team.name} vs {game.away_team.name} \\\"\\n f\\\"{len(self.instance_predictions[game])} times.\\\")\\n confusion_matrix = [0, 0, 0, 0]\\n for predicted_winner in self.instance_predictions[game]:\\n actual_winner = game.home_team if game.home_team.scores.get(\\n GamePeriod.TOTAL) > game.away_team.scores.get(GamePeriod.TOTAL) else game.away_team\\n actual_home_team_won = True if actual_winner == game.home_team else False\\n predicted_home_team_won = True if predicted_winner == game.home_team else False\\n confusion_matrix = recursive_build_confusion_matrix(confusion_matrix, actual_home_team_won,\\n predicted_home_team_won)\\n total_results[0] += confusion_matrix[0]\\n total_results[1] += confusion_matrix[1]\\n total_results[2] += confusion_matrix[2]\\n total_results[3] += confusion_matrix[3]\\n\\n for k, v in analyze_confusion_matrix(confusion_matrix, len(self.instance_predictions[game])).items():\\n self.logger.info(f\\\"{k} : {v}\\\")\\n self.logger.info(\\\"\\\\n###### FINAL RESULTS FOR ALL GAMES###### \\\\n\\\")\\n final_stats = analyze_confusion_matrix(total_results, sum(total_results))\\n for k, v in final_stats.items():\\n self.logger.info(f\\\"{k} : {v}\\\")\\n self.write_stats_to_json(final_stats, self.outfile)\",\n \"def accuracy(num_points, num_dims, p, surv, num_runs=100, num_test=10000):\\n p1 = []\\n p2 = []\\n p3 = []\\n p4 = []\\n for t in range(num_runs):\\n if surv:\\n cloud = generate_survival_curves(num_points + num_test, num_dims)\\n else:\\n cloud = generate_time_series(num_points + num_test, num_dims)\\n opt_cloud = cloud[:num_points]\\n test_cloud = cloud[num_points:]\\n p1.append(proportion(test_cloud, olshen(opt_cloud, p, surv)))\\n p2.append(proportion(test_cloud, two_sided_olshen(opt_cloud, p, surv)))\\n p3.append(proportion(test_cloud, gspie(opt_cloud, p, surv)))\\n p4.append(\\n proportion(test_cloud, pointwise_bonferroni(opt_cloud, p, surv)))\\n plt.boxplot([p1, p2, p3, p4])\\n plt.xticks([1, 2, 3, 4],\\n ['Olshen', 'Two-Sided Olshen', 'GSPIE', 'Pointwise Bonferroni'])\\n plt.ylabel('Observed Coverage')\\n plt.xlabel(f\\\"{p} Prescribed Coverage\\\")\\n plt.show()\",\n \"def test_compare_classifiers_predictions_distribution_vis_api(experiment_to_use):\\n experiment = experiment_to_use\\n predictions = experiment.predictions_num\\n viz_outputs = (\\\"pdf\\\", \\\"png\\\")\\n with TemporaryDirectory() as tmpvizdir:\\n for viz_output in viz_outputs:\\n vis_output_pattern_pdf = tmpvizdir + f\\\"/*.{viz_output}\\\"\\n visualize.compare_classifiers_predictions_distribution(\\n [predictions, predictions],\\n experiment.ground_truth,\\n experiment.ground_truth_metadata,\\n experiment.output_feature_name,\\n labels_limit=0,\\n model_names=[\\\"Model1\\\", \\\"Model2\\\"],\\n output_directory=tmpvizdir,\\n file_format=viz_output,\\n )\\n figure_cnt = glob.glob(vis_output_pattern_pdf)\\n assert 1 == len(figure_cnt)\",\n \"def evaluate_overall(predictions):\\n print(\\\"\\\\nAccuracy for the the models combined (full prediction):\\\\n\\\")\\n\\n perfect_predictions = [0, 0, 0]\\n exact_code_match = [0, 0, 0]\\n exact_code_match_ignore_spaces = [0, 0, 0]\\n correct_syntax = [0, 0, 0]\\n total_predictions = [0, 0, 0]\\n num_failed_predictions = 0\\n \\n for prediction in predictions:\\n # Perfect Prediction\\n if prediction[\\\"correct_data\\\"][\\\"correct_location\\\"] == prediction[\\\"predicted_location\\\"]:\\n # Correct type\\n if prediction[\\\"correct_data\\\"][\\\"correct_type\\\"] == prediction[\\\"predicted_type\\\"]:\\n # Correct token\\n if prediction[\\\"predicted_type\\\"] == \\\"insert\\\" or prediction[\\\"predicted_type\\\"] == \\\"modify\\\":\\n if \\\"correct_token\\\" in prediction[\\\"correct_data\\\"] and prediction[\\\"correct_data\\\"][\\\"correct_token\\\"] == \\\\\\n prediction[\\\"predicted_token\\\"]:\\n perfect_predictions[FixType[prediction[\\\"predicted_type\\\"]].value] = perfect_predictions[FixType[\\n prediction[\\\"predicted_type\\\"]].value] + 1\\n else:\\n perfect_predictions[FixType[prediction[\\\"predicted_type\\\"]].value] = perfect_predictions[FixType[\\n prediction[\\\"predicted_type\\\"]].value] + 1\\n \\n # Exact Code Match\\n if prediction[\\\"predicted_code\\\"] == prediction[\\\"correct_data\\\"][\\\"correct_code\\\"]:\\n exact_code_match[FixType[prediction[\\\"predicted_type\\\"]].value] = exact_code_match[FixType[\\n prediction[\\\"predicted_type\\\"]].value] + 1\\n \\n # Exact Code Match (Ignore Spaces)\\n if prediction[\\\"predicted_code\\\"].replace(\\\" \\\", \\\"\\\") == prediction[\\\"correct_data\\\"][\\\"correct_code\\\"].replace(\\\" \\\", \\\"\\\"):\\n exact_code_match_ignore_spaces[FixType[prediction[\\\"predicted_type\\\"]].value] = exact_code_match_ignore_spaces[FixType[\\n prediction[\\\"predicted_type\\\"]].value] + 1\\n \\n # Correct Syntax Fix\\n try:\\n cst.parse_module(prediction[\\\"predicted_code\\\"])\\n correct_syntax[FixType[prediction[\\\"predicted_type\\\"]].value] = correct_syntax[FixType[\\n prediction[\\\"predicted_type\\\"]].value] + 1\\n except Exception as e:\\n logging.warning(f\\\"{e.__class__.__name__} occurred: {e}\\\")\\n # Happens if parsing fails\\n pass\\n \\n total_predictions[FixType[prediction[\\\"predicted_type\\\"]].value] = total_predictions[FixType[\\n prediction[\\\"predicted_type\\\"]].value] + 1\\n \\n # Perfect Prediction\\n for i in range(3):\\n if total_predictions[i] == 0: # If the type was never predicted\\n accuracy = 0\\n else:\\n accuracy = perfect_predictions[i] / total_predictions[i]\\n print(f\\\"Perfect Prediction for class {FixType(i).name}: {accuracy * 100} %\\\")\\n\\n accuracy = sum(perfect_predictions) / (len(predictions) - num_failed_predictions)\\n print(f\\\"Perfect Prediction accuracy overall is {accuracy * 100} %\\\")\\n \\n # Exact Code Match\\n for i in range(3):\\n if total_predictions[i] == 0: # If the type was never predicted\\n accuracy = 0\\n else:\\n accuracy = exact_code_match[i] / total_predictions[i]\\n print(f\\\"Exact Code Match for class {FixType(i).name}: {accuracy * 100} %\\\")\\n\\n accuracy = sum(exact_code_match) / (len(predictions) - num_failed_predictions)\\n print(f\\\"Exact Code Match accuracy overall is {accuracy * 100} %\\\")\\n \\n # Exact Code Match (Ignore Spaces)\\n for i in range(3):\\n if total_predictions[i] == 0: # If the type was never predicted\\n accuracy = 0\\n else:\\n accuracy = exact_code_match_ignore_spaces[i] / total_predictions[i]\\n print(f\\\"Exact Code Match (Ignore Spaces) for class {FixType(i).name}: {accuracy * 100} %\\\")\\n\\n accuracy = sum(exact_code_match_ignore_spaces) / (len(predictions) - num_failed_predictions)\\n print(f\\\"Exact Code Match (Ignore Spaces) accuracy overall is {accuracy * 100} %\\\")\\n \\n # Correct Syntax Fixes\\n for i in range(3):\\n if total_predictions[i] == 0: # If the type was never predicted\\n accuracy = 0\\n else:\\n accuracy = correct_syntax[i] / total_predictions[i]\\n print(f\\\"Correct Syntax Fixes for class {FixType(i).name}: {accuracy * 100} %\\\")\\n\\n accuracy = sum(correct_syntax) / (len(predictions) - num_failed_predictions)\\n print(f\\\"Correct Syntax Fixes accuracy overall is {accuracy * 100} %\\\")\",\n \"def test_calibration_multiclass_vis_api(experiment_to_use):\\n experiment = experiment_to_use\\n probabilities = experiment.probabilities\\n viz_outputs = (\\\"pdf\\\", \\\"png\\\")\\n with TemporaryDirectory() as tmpvizdir:\\n for viz_output in viz_outputs:\\n vis_output_pattern_pdf = tmpvizdir + f\\\"/*.{viz_output}\\\"\\n visualize.calibration_multiclass(\\n [probabilities, probabilities],\\n experiment.ground_truth,\\n experiment.ground_truth_metadata,\\n experiment.output_feature_name,\\n labels_limit=0,\\n model_names=[\\\"Model1\\\", \\\"Model2\\\"],\\n output_directory=tmpvizdir,\\n file_format=viz_output,\\n )\\n figure_cnt = glob.glob(vis_output_pattern_pdf)\\n assert 2 == len(figure_cnt)\",\n \"def check_expectations(self):\\n self.load_results()\\n\\n for (benchmark, producer), result in self.results.items():\\n if not result.reports:\\n print('No results found for ' + benchmark + ' ' + producer)\\n result.test_passed = False\\n else:\\n for report in result.reports:\\n if check_benchmark_result(report, result.expectation):\\n print('Test passed: ' + result.directory)\\n result.test_passed = True\\n else:\\n print('Test failed: ' + result.directory)\\n result.test_passed = False\",\n \"def evaluation(deploymentID, start_time,\\n end_time, sites, validation, manOrLik):\\n \\n estScoreBound = 5\\n falsePositive_dep = 0\\n falseNegative_dep = 0\\n truePositive_dep = 0\\n trueNegative_dep = 0\\n\\n db_con = MySQLdb.connect(user=\\\"root\\\", db=\\\"qraat\\\")\\n for i in sites:\\n truePositive = 0\\n trueNegative = 0\\n falsePositive = 0\\n falseNegative = 0\\n \\n cur = db_con.cursor()\\n cur.execute(\\\"\\\"\\\"SELECT isPulse, band3, band10, frequency,\\n ec, tnp, edsp, fdsp, edsnr, fdsnr\\n FROM est INNER JOIN est_class%s\\n ON ID = est_class%s.estID\\n INNER JOIN estscore2\\n ON ID = estscore2.estID\\n WHERE timestamp > %s\\n AND timestamp < %s\\n AND deploymentID = %s\\n AND siteID = %s\\n AND setNum = %s;\\n \\\"\\\"\\\"%(manOrLik, manOrLik, start_time, end_time,\\n deploymentID, i, validation))\\n for row in cur.fetchall():\\n estData = {'band3':row[1], 'band10':row[2],\\n 'frequency':row[3], 'ec':row[4], 'tnp':row[5],\\n 'edsp':row[6], 'fdsp':row[7],\\n 'edsnr':row[8], 'fdsnr':row[9]}\\n \\n #Classify data\\n isPulse = modifiedBC(deploymentID, i, start_time,\\n validation, estData, manOrLik)\\n\\n #Determine whether the classification results are correct or not\\n if (row[0] == 1):\\n if (isPulse == 1):\\n truePositive += 1\\n else:\\n falseNegative += 1\\n else:\\n if (isPulse == 1):\\n falsePositive += 1\\n else:\\n trueNegative += 1\\n \\n falsePositive_dep += falsePositive\\n falseNegative_dep += falseNegative\\n truePositive_dep += truePositive\\n trueNegative_dep += trueNegative\\n return [truePositive_dep, trueNegative_dep,\\n falsePositive_dep, falseNegative_dep]\",\n \"def main():\\n\\n # choose number of data-points and sample a pair of vectors: the input\\n # values and the corresponding target values\\n N = 500\\n inputs, targets = sample_data(N, arbitrary_function_2, seed=1)\\n\\n # specify the centres and scale of some rbf basis functions\\n default_centres = np.linspace(0,1,21)\\n default_scale = 0.03\\n default_reg_param = 0.08\\n\\n # get the cross-validation folds\\n num_folds = 4\\n folds = create_cv_folds(N, num_folds)\\n\\n # evaluate then plot the performance of different reg params\\n evaluate_reg_param(inputs, targets, folds, default_centres, default_scale)\\n # evaluate then plot the performance of different scales\\n evaluate_scale(inputs, targets, folds, default_centres, default_reg_param)\\n # evaluate then plot the performance of different numbers of basis\\n # function centres.\\n evaluate_num_centres(\\n inputs, targets, folds, default_scale, default_reg_param)\\n\\n plt.show()\",\n \"def benchmark_algos(algos: List[str], setup_script: str, number: int,\\n *args) -> None:\\n for algo_name in algos:\\n print('\\\\nRAN {} {} times with args {}'.format(algo_name, number,\\n *args))\\n print(timeit('{}({})'.format(algo_name, *args),\\n setup=setup_script, number=number))\",\n \"def run(self):\\n self.evaluate()\\n self.accumulate()\\n self.summarize()\",\n \"def test_variational():\\n # iris\\n #pres = \\\"Test pour le data set Iris (facile, classique)\\\"\\n #test_from_func_variational(pres, 15, 10, 3, True, Iris)\\n\\n # breast cancer\\n pres = \\\"Test pour le data set Breast Cancer (facile, classique)\\\"\\n test_from_func_variational(pres, 15, 10, 3, True, Breast_cancer)\\n\\n # digits\\n # pres = \\\"Test pour le data set Digits (difficile, classique)\\\"\\n # test_from_func(pres, 10, 10, 10, True, Digits, quantum_instance)\\n\\n # wine\\n # pres = \\\"Test pour le data set Wine (moyen, classique)\\\"\\n # test_from_func(pres, 15, 10, 5, True, Wine, quantum_instance)\\n\\n # gaussian\\n pres = \\\"Test pour des données gaussiennes (moyen, classique)\\\"\\n for _ in range(1):\\n print(\\\"\\\\n\\\")\\n print(\\\"New iteration\\\")\\n test_from_func_variational(pres, 25, 10, 2, True, Gaussian)\\n print(\\\"\\\\n\\\")\\n\\n # small adn strings\\n pres = \\\"Test pour des séquences ADN courtes (difficile, classique)\\\"\\n test_from_func_variational(pres, 10, 15, 14, True, Sequence)\\n\\n #Quantum data\\n pres = \\\"Test pour des données générées par ordinateur quantique (facile, quantique)\\\"\\n print(pres)\\n _, samp_train, samp_test, labels = ad_hoc_data(15, 10, 2, 0.3, True)\\n sample_m, sample_p = stock_get(20, 0.3)\\n\\n labels_me = [-1, 1]\\n samp_train_me = {-1: np.array(sample_m[:15]), 1: np.array(sample_p[:15])}\\n samp_test_me = {-1: np.array(sample_m[15:]), 1: np.array(sample_p[15:])}\\n print(samp_train)\\n print(samp_train_me)\\n print(samp_test)\\n print(samp_test_me)\\n\\n my_impl_variational(samp_train, samp_test, labels)\\n print(\\\"Pour autres données quantiques\\\")\\n my_impl_variational(samp_train_me, samp_test_me, labels_me)\",\n \"def test_compare_classifiers_predictions_vis_api(experiment_to_use):\\n experiment = experiment_to_use\\n predictions = experiment.predictions\\n viz_outputs = (\\\"pdf\\\", \\\"png\\\")\\n with TemporaryDirectory() as tmpvizdir:\\n for viz_output in viz_outputs:\\n vis_output_pattern_pdf = tmpvizdir + f\\\"/*.{viz_output}\\\"\\n visualize.compare_classifiers_predictions(\\n [predictions, predictions],\\n experiment.ground_truth,\\n experiment.ground_truth_metadata,\\n experiment.output_feature_name,\\n labels_limit=0,\\n model_names=[\\\"Model1\\\", \\\"Model2\\\"],\\n output_directory=tmpvizdir,\\n file_format=viz_output,\\n )\\n figure_cnt = glob.glob(vis_output_pattern_pdf)\\n assert 1 == len(figure_cnt)\",\n \"def evaluate(clf, dataset, feature_list, features, labels, num_iter, params):\\n\\n features_train, features_test, labels_train, labels_test = \\\\\\n train_test_split(features, labels, test_size=0.3, random_state=42)\\n\\n\\n\\n precision_values = []\\n recall_values = []\\n accuracy_values = []\\n print clf\\n for i in xrange(0, num_iter):\\n #print params\\n clf = GridSearchCV(clf, params)\\n clf.fit(features_train, labels_train)\\n print '*****************************'\\n print clf.best_estimator_\\n print clf.best_params_\\n\\n clf = clf.best_estimator_\\n #test_classifier(clf, dataset, feature_list)\\n pred = clf.predict(features_test)\\n precision_values.append(precision_score(labels_test, pred))\\n recall_values.append(recall_score(labels_test, pred))\\n accuracy_values.append(accuracy_score(labels_test, pred))\\n print 'Recall score: ', mean(recall_values)\\n print 'Precision score: ', mean(precision_values)\\n print 'Accuracy score: ' , mean(accuracy_values)\",\n \"def _compute_experiment_statistics(self):\\n pass\",\n \"def run_mcts(self, runs_per_round):\\n for i in range(runs_per_round):\\n self.select(self.env, 'r')\\n self.env_reset()\\n counts = [self.Nsa[('r', a)] for a in range(self.actions)]\\n # print(\\\"counts \\\", counts)\\n # print(\\\"Q-values\\\", [self.Qsa[('r', a)] for a in range(self.actions)])\\n # print()\\n return np.argmax(counts)\",\n \"def benchmark(options):\\n # Prepare experiments\\n with open(options['<benchmark>']) as f:\\n benchmark_config = json.loads(f.read())\\n generate_agent_configs(benchmark_config)\\n experiments = product(benchmark_config['environments'], benchmark_config['agents'], [options])\\n\\n # Run evaluations\\n with Pool(processes=int(options['--processes'])) as pool:\\n results = pool.starmap(evaluate, experiments)\\n\\n # Clean temporary config files\\n generate_agent_configs(benchmark_config, clean=True)\\n\\n # Write evaluations summary\\n benchmark_filename = os.path.join(Evaluation.OUTPUT_FOLDER, '{}_{}.{}.json'.format(\\n BENCHMARK_FILE, datetime.datetime.now().strftime('%Y%m%d-%H%M%S'), os.getpid()))\\n with open(benchmark_filename, 'w') as f:\\n json.dump(results, f, sort_keys=True, indent=4)\\n gym.logger.info('Benchmark done. Summary written in: {}'.format(benchmark_filename))\",\n \"def plotDifferentSettings():\\n\\t#save location of the results\\n\\tresloc = 'Different_settings_results'\\n\\n\\t#the numrows limits\\n\\tnrows_lims = [1e4, 1e7]\\n\\tnbits = 30\\n\\t\\n\\t#string for the file names of the to be saved files\\n\\tsettingsstr = 'nrows={:.0e}--{:.0e}_nbits={}'.format(nrows_lims[0], nrows_lims[1], nbits)\\n\\t\\n\\t#the relative approximation error for the different counting algorithms\\n\\tll_RAE = []\\n\\tprob_RAE = []\\n\\tcomb_RAE = []\\n\\t#the runtime for the different algorithms\\n\\ttc_runtime = []\\n\\tll_runtime = []\\n\\tprob_runtime = []\\n\\tcomb_runtime = []\\n\\n\\t#the different settings we want to test\\n\\tnumrows = np.linspace(nrows_lims[0], nrows_lims[1], num = 15, dtype = int)\\n\\tnumbits = np.array([nbits])\\n\\t\\n\\tlooplength = len(numrows)\\n\\t\\n\\ttry:\\n\\t\\t(ll_RAE, prob_RAE, comb_RAE, tc_runtime, ll_runtime, prob_runtime, comb_runtime) = np.loadtxt('./{0}/diffset_results_{1}.txt'.format(resloc, settingsstr))\\n\\texcept:\\n\\t\\tfor i in np.arange(len(numrows)):\\n\\t\\t\\toF.progress(i, looplength)\\n\\t\\t\\tfor j in np.arange(len(numbits)):\\n\\t\\t\\t\\tresults = runCounts(numrows[i], numbits[j], doprints = False)\\n\\t\\t\\t\\n\\t\\t\\t\\tll_RAE = np.append(ll_RAE, results[0])\\n\\t\\t\\t\\tprob_RAE = np.append(prob_RAE, results[1])\\n\\t\\t\\t\\tcomb_RAE = np.append(comb_RAE, results[2])\\n\\t\\t\\t\\n\\t\\t\\t\\ttc_runtime = np.append(tc_runtime, results[3])\\n\\t\\t\\t\\tll_runtime = np.append(ll_runtime, results[4])\\n\\t\\t\\t\\tprob_runtime = np.append(prob_runtime, results[5])\\n\\t\\t\\t\\tcomb_runtime = np.append(comb_runtime, results[6])\\n\\t\\t\\t\\n\\t\\tnp.savetxt('./{0}/diffset_results_{1}.txt'.format(resloc, settingsstr), \\n\\t\\t\\tnp.array([ll_RAE, prob_RAE, comb_RAE, tc_runtime, ll_runtime, prob_runtime, comb_runtime]), \\n\\t\\t\\theader = '#ll_RAE, prob_RAE, comb_RAE, tc_runtime, ll_runtime, prob_runtime, comb_runtime')\\n\\t\\n\\tplotTwoValues(numrows, ll_RAE, ll_runtime, 'Number of rows', 'RAE [\\\\%]', 'Runtime [s]', 'RAE and runtime of loglog count for different number of rows. \\\\nNumbits = {}'.format(nbits), 'RAEandRuntime_loglog_{0}.pdf'.format(settingsstr))\\n\\t\\n\\tplotTwoValues(numrows, prob_RAE, prob_runtime, 'Number of rows', 'RAE [\\\\%]', 'Runtime [s]', 'RAE and runtime of probabilisic count for different \\\\nnumber of rows. Numbits = {}'.format(nbits), 'RAEandRuntime_prob_{0}.pdf'.format(settingsstr))\",\n \"def SkoptPaperStats(maxIters, numRuns):\\n\\n comm = MPI.COMM_WORLD\\n rank = comm.Get_rank()\\n\\n assert comm.Get_size() == numRuns, \\\"Please ensure there is one process running per run i.e \\\" + str(numRuns) + \\\" processes.\\\"\\n \\n # Define the problem bounds.\\n skoptBounds = [(10, 1300), (40, 230), (0, 90), (0, 90)]\\n\\n # Use the same seed list as previously.\\n seedList = [572505, 357073, 584216, 604873, 854690, 573165, 298975, 650770, 243921, 191168]\\n\\n # The target for each algorithm. This was determined by using the values in the literature, so there is clearly some deviation either due to the detuning or computation.\\n globalFoM = 1.033\\n\\n if rank == 0:\\n timeList = []\\n iterationList = []\\n\\n # Define which solver will be used.\\n optimiser = skopt.Optimizer(skoptBounds, base_estimator = \\\"GP\\\", n_initial_points = int(np.ceil(maxIters/10)), random_state = seedList[rank])\\n\\n # Start timing.\\n startTime = time.time()\\n timeElapsed = None\\n iterationSuccess = None\\n\\n # Start optimisation.\\n for iteration in range(maxIters):\\n\\n # Make one suggestion.\\n nextParams = optimiser.ask()\\n\\n # Check what FoM this gives. Go negative as this is a minimisation routine.\\n fEval = FitnessSkopt(nextParams)\\n\\n # Update best FoM.\\n if abs(fEval) >= globalFoM:\\n # The algorithm has managed to surpass or equal the paper value.\\n iterationSuccess = iteration\\n timeElapsed = time.time() - startTime\\n \\n if rank == 0:\\n iterationList.append(iterationSuccess)\\n timeList.append(timeElapsed)\\n\\n break\\n \\n # Tell the optimiser about the result.\\n optimiser.tell(nextParams, fEval)\\n\\n # Run complete. Send results to main process. Tags are unique identifiers.\\n if rank != 0:\\n comm.send(timeElapsed, dest = 0, tag = 1)\\n comm.send(iterationSuccess, dest = 0, tag = 2)\\n\\n # Wait for all the processes to end.\\n comm.Barrier()\\n\\n if rank == 0:\\n # Aggregate the data.\\n for process in range(comm.Get_size() - 1):\\n # Get the data.\\n individualTime = None\\n individualTime = comm.recv(individualTime, source = process + 1, tag = 1)\\n\\n individualIter = None\\n individualIter = comm.recv(individualIter, source = process + 1, tag = 2)\\n\\n if individualIter is not None:\\n # Both values must therefore be non-null.\\n iterationList.append(individualIter)\\n timeList.append(individualTime)\\n\\n avgRuntime = np.average(timeList)\\n avgIters = np.average(iterationList)\\n try:\\n\\n fastestTime = np.min(timeList)\\n\\n except ValueError:\\n \\n # List is empty.\\n fastestTime = float('NaN')\\n\\n numSuccess = len(iterationList)\\n successRate = numSuccess/numRuns\\n\\n print(\\\"Bayesian optimisation paper testing complete! Here are the stats:\\\")\\n print(\\\"Number of successful runs: \\\" + str(numSuccess) + \\\" (Success rate of \\\" + str(successRate) + \\\")\\\")\\n print(\\\"Average iterations required for success: \\\" + str(avgIters))\\n print(\\\"Average time required for success: \\\" + str(avgRuntime))\\n print(\\\"Fastest convergence time: \\\" + str(fastestTime))\\n print(\\\"------------------------------------------------------------------------------------------------------------------\\\")\\n \\n return\",\n \"def run_training_and_tests(test_name, dataset, models, n_images = 1000, training_split = 0.7, \\n n_training_images = None, n_test_images = None, \\n n_iterations = 1, dimensions = (50, 50)):\\n\\n aggregate_metrics = {}\\n\\n # Run specified number of iterations\\n for i in range(n_iterations):\\n print(\\\"\\\\nTest iteration\\\", i+1)\\n\\n # Handle if specific training and test set size isn't given\\n if (n_training_images is None):\\n n_training_images = n_images * training_split\\n if (n_test_images is None):\\n n_test_images = n_images * (1-training_split)\\n\\n # Load training and test sets from single dataset\\n train_data, train_labels, test_data, test_labels = image_utils.read_dataset(\\n n_training_images, \\n n_test_images, \\n './datasets/' + dataset, \\n dimensions[0], \\n dimensions[1]\\n )\\n\\n # Train and run tests for each model\\n for model in models:\\n print(\\\"Working with model '\\\" + model.label + \\\"'\\\")\\n\\n # Train model\\n start = time.time()\\n model.train(copy.deepcopy(train_data), train_labels)\\n end = time.time()\\n training_time = round(end - start, 3)\\n\\n # Run predictions on test set\\n start = time.time()\\n predicted = model.run(copy.deepcopy(test_data))\\n end = time.time()\\n test_time = round(end - start, 3)\\n\\n # Calculate metrics and store for aggregate calculations\\n metrics = Metrics(test_labels, predicted, training_time, test_time)\\n if model.label in aggregate_metrics:\\n aggregate_metrics[model.label].append(metrics)\\n else:\\n aggregate_metrics[model.label] = [metrics]\\n\\n # Print results\\n print(\\\"Results\\\\n\\\" + \\\"------\\\")\\n print(str(metrics))\\n\\n # Save model\\n filepath = \\\"./test/\\\" + test_name + \\\"/\\\" + model.label + \\\"/iteration\\\" + str(i+1) + \\\"/\\\"\\n print(\\\"Saving model to '\\\" + filepath + model.label + \\\".joblib'\\\")\\n os.makedirs(os.path.dirname(filepath), exist_ok = True)\\n with open(filepath + model.label + '.joblib', 'wb') as file:\\n dump(model, file)\\n\\n # Save results\\n print(\\\"Saving results to '\\\" + filepath + \\\"results.txt'\\\\n\\\")\\n with open(filepath + \\\"results.txt\\\", 'w') as file:\\n file.write(str(metrics))\\n\\n # Calculate, print and write aggregate metrics\\n print(\\n 'Aggregate Results' + '\\\\n' +\\n '-----------------'\\n )\\n for model in models:\\n aggregate = combine_metrics(aggregate_metrics[model.label])\\n print(model.label)\\n print(aggregate)\\n filepath = \\\"./test/\\\" + test_name + \\\"/\\\" + model.label + \\\"/\\\"\\n print(\\\"Saving results to '\\\" + filepath + \\\"aggregate_results.txt'\\\" + \\\"\\\\n -- -\\\\n\\\")\\n with open(filepath + \\\"aggregate_results.txt\\\", 'w') as file:\\n file.write(str(aggregate))\"\n]","isConversational":false},"negative_scores":{"kind":"list like","value":["0.70623094","0.7007437","0.69541293","0.6884711","0.68788844","0.68726337","0.67315125","0.6698763","0.66550297","0.6629735","0.655879","0.6552913","0.6541094","0.6533274","0.65046316","0.649451","0.6487716","0.64735806","0.6451186","0.6437416","0.64339566","0.6430555","0.6424844","0.64165574","0.64116085","0.64044327","0.6380517","0.6367108","0.63302666","0.63297606","0.6320315","0.631165","0.6309548","0.6306806","0.6304774","0.6296881","0.6294089","0.6268248","0.62646943","0.62541413","0.6244517","0.6240123","0.6229821","0.62199056","0.61869085","0.6179335","0.6175337","0.6159189","0.614777","0.61453754","0.6142987","0.6123919","0.6117597","0.61118615","0.61050224","0.61041033","0.6086952","0.607758","0.6062766","0.60626346","0.60543865","0.6053785","0.6047914","0.6046168","0.60290015","0.6027984","0.60179293","0.60164034","0.6013663","0.6012137","0.6008275","0.60075504","0.60061496","0.60058725","0.6005182","0.60025764","0.6002492","0.6001097","0.60001135","0.599949","0.5997685","0.59974205","0.59945697","0.5986698","0.5976091","0.59693664","0.59659237","0.5963352","0.59601045","0.5957393","0.5954754","0.5954727","0.5953482","0.59530145","0.5951625","0.59452754","0.5942147","0.5941517","0.5938096","0.5929191"],"string":"[\n \"0.70623094\",\n \"0.7007437\",\n \"0.69541293\",\n \"0.6884711\",\n \"0.68788844\",\n \"0.68726337\",\n \"0.67315125\",\n \"0.6698763\",\n \"0.66550297\",\n \"0.6629735\",\n \"0.655879\",\n \"0.6552913\",\n \"0.6541094\",\n \"0.6533274\",\n \"0.65046316\",\n \"0.649451\",\n \"0.6487716\",\n \"0.64735806\",\n \"0.6451186\",\n \"0.6437416\",\n \"0.64339566\",\n \"0.6430555\",\n \"0.6424844\",\n \"0.64165574\",\n \"0.64116085\",\n \"0.64044327\",\n \"0.6380517\",\n \"0.6367108\",\n \"0.63302666\",\n \"0.63297606\",\n \"0.6320315\",\n \"0.631165\",\n \"0.6309548\",\n \"0.6306806\",\n \"0.6304774\",\n \"0.6296881\",\n \"0.6294089\",\n \"0.6268248\",\n \"0.62646943\",\n \"0.62541413\",\n \"0.6244517\",\n \"0.6240123\",\n \"0.6229821\",\n \"0.62199056\",\n \"0.61869085\",\n \"0.6179335\",\n \"0.6175337\",\n \"0.6159189\",\n \"0.614777\",\n \"0.61453754\",\n \"0.6142987\",\n \"0.6123919\",\n \"0.6117597\",\n \"0.61118615\",\n \"0.61050224\",\n \"0.61041033\",\n \"0.6086952\",\n \"0.607758\",\n \"0.6062766\",\n \"0.60626346\",\n \"0.60543865\",\n \"0.6053785\",\n \"0.6047914\",\n \"0.6046168\",\n \"0.60290015\",\n \"0.6027984\",\n \"0.60179293\",\n \"0.60164034\",\n \"0.6013663\",\n \"0.6012137\",\n \"0.6008275\",\n \"0.60075504\",\n \"0.60061496\",\n \"0.60058725\",\n \"0.6005182\",\n \"0.60025764\",\n \"0.6002492\",\n \"0.6001097\",\n \"0.60001135\",\n \"0.599949\",\n \"0.5997685\",\n \"0.59974205\",\n \"0.59945697\",\n \"0.5986698\",\n \"0.5976091\",\n \"0.59693664\",\n \"0.59659237\",\n \"0.5963352\",\n \"0.59601045\",\n \"0.5957393\",\n \"0.5954754\",\n \"0.5954727\",\n \"0.5953482\",\n \"0.59530145\",\n \"0.5951625\",\n \"0.59452754\",\n \"0.5942147\",\n \"0.5941517\",\n \"0.5938096\",\n \"0.5929191\"\n]","isConversational":false},"document_score":{"kind":"string","value":"0.6528349"},"document_rank":{"kind":"string","value":"14"}}},{"rowIdx":93295,"cells":{"query":{"kind":"string","value":"runs an image_matching_module that calculates the precision and recall of multiple comparison algorithms' results"},"document":{"kind":"string","value":"def run_single_advanced_experiment(self, path_to_csv_files, algorithms_weights, thresholds):\n\n splitted_path = os.path.abspath(__file__).split(\"/\")\n if not os.path.isdir(\"/\".join(splitted_path[:-1]) + \"/results\"):\n os.mkdir(\"/\".join(splitted_path[:-1]) + \"/results\")\n\n first_run = True\n for algorithm, weight in algorithms_weights:\n if first_run:\n algorithm_name = algorithm.get_algorithm_name() + weight\n first_run = False\n else:\n algorithm_name = algorithm_name + \"_\" + algorithm.get_algorithm_name() + weight\n\n experiment_dict, list_dict = self.create_experiment_dict(algorithm_name, thresholds)\n\n csv_paths_list = self.reader.read_all_csv_paths_from_path(path_to_csv_files)\n file_number = 1\n for csv_path in csv_paths_list:\n if 'total' in csv_path:\n continue\n df = pd.read_csv(csv_path)\n start_file_time = time.time()\n print(\"started file number:\", file_number)\n for index, row in df.iterrows():\n for algorithm_threshold in thresholds:\n\n algorithm_scores = []\n for algorithm, weight in algorithms_weights:\n algorithm_column_name = algorithm.get_algorithm_name() + \"_actual\"\n initial_algorithm_score = row[algorithm_column_name]\n algorithm_scores.append(initial_algorithm_score * weight)\n\n combined_score = 0.0\n for algorithm_score in algorithm_scores:\n combined_score += algorithm_score\n\n actual_score = row['actual_score']\n if actual_score == 1:\n if combined_score >= algorithm_threshold:\n experiment_dict[str(algorithm_threshold)][algorithm_name][list_dict['TP']] += 1\n else:\n experiment_dict[str(algorithm_threshold)][algorithm_name][list_dict['FN']] += 1\n else:\n if combined_score < algorithm_threshold:\n experiment_dict[str(algorithm_threshold)][algorithm_name][list_dict['TN']] += 1\n else:\n experiment_dict[str(algorithm_threshold)][algorithm_name][list_dict['FP']] += 1\n\n finish_file_time = time.time()\n print(\"finished file number:\", file_number)\n print(\"file number:\", file_number, \"took:\", (finish_file_time - start_file_time) / 60, \"minutes\")\n file_number += 1\n\n # calculate the precision and recall for each threshold and algorithm\n experiment_dict = self.update_precision_and_recall(experiment_dict, list_dict, algorithm_name, thresholds)\n\n # create a dictionary that can be ingested into a dataframe\n experiment_dict_for_csv = self.create_advanced_experiment_dict_for_csv(experiment_dict, list_dict,\n algorithm_name, thresholds)\n\n # create a dataframe from the results\n experiment_dict_df = pd.DataFrame(experiment_dict_for_csv)\n\n # write the results into a csv file\n experiment_csv_file_name = os.path.dirname(\"/\".join(splitted_path)) + \"/results/precision_recall_results_\" \\\n + algorithm_name + \".csv\"\n experiment_dict_df.to_csv(experiment_csv_file_name, index=False, header=True)"},"metadata":{"kind":"string","value":"{\n \"objective\": {\n \"self\": [],\n \"paired\": [],\n \"triplet\": [\n [\n \"query\",\n \"document\",\n \"negatives\"\n ]\n ]\n }\n}"},"negatives":{"kind":"list like","value":["def painting_matching_ml(imgs, db_imgs, method_list, text_masks, author_text, gt_text, metrics, weights, splits=30, max_rank=10):\n descriptor_extractors = [get_descriptor_extractor(method_name) for method_name in method_list]\n tmp_img_format = []\n tmp_mask_format = []\n tmp_text_format = []\n for i in range(len(imgs)):\n for j in range(len(imgs[i])):\n tmp_img_format.append(imgs[i][j])\n tmp_mask_format.append(text_masks[i][j])\n tmp_text_format.append(author_text[i][j])\n\n #db_imgs = [img[0] for img in db_imgs]\n db_img_splits = [i*len(db_imgs)//splits for i in range(splits-1)]\n \n scores = []\n query_descriptors = extract_descriptors(tmp_img_format, descriptor_extractors, method_list, tmp_text_format, tmp_mask_format) \n #np.array([extract_descriptors(img, matching_methods, mask) for img, mask in zip(tmp_img_format, tmp_mask_format)])\n print(\"Starting db extraction + matching\")\n for split in tqdm(range(splits-2)):\n db_descriptors = extract_descriptors(db_imgs[db_img_splits[split]:db_img_splits[split+1]], descriptor_extractors, method_list, gt_text[db_img_splits[split]:db_img_splits[split+1]], None) #np.array([mrhm(db_img) for db_img in db_imgs[db_img_splits[split]:db_img_splits[split+1]]])\n scores.append(compare_descriptors(query_descriptors, db_descriptors, metrics, method_list, weights))\n # compare_descriptors(query_descriptors, db_descriptors, descriptor_comp_methods, descriptor_names, weights)\n db_descriptors = extract_descriptors(db_imgs[db_img_splits[-1]:], descriptor_extractors, method_list, gt_text[db_img_splits[-1]:], None)\n scores.append(compare_descriptors(query_descriptors, db_descriptors, metrics, method_list, weights))\n \n # concatenate all the results\n scores = np.concatenate(scores, 1)\n \n top_k_matches = np.argpartition(scores, list(range(max_rank)))[:, :max_rank]\n return top_k_matches","def patternMatchingTest(matchingAlgorithm,\n observations= defaultCOMPREFdump , \n modelsFolder= defaultWRFdumpsFolder,\n #key1='0828',\n key1='',\n key2='',\n #key3='',\n verbose=True,\n makeSmallImages=True,\n smallDpi=40,\n frameDpi = 600 , # added 2013-09-23\n loadFormerAnalysisResult=False, #added 2013-09-23\n #summary_folder=summary_folder, # needed if loadFormerAnalysisResult is True\n summary_folder=summary_folder+timeString + '/', # 2013-09-27\n matching_algorithm_name=\"\", # needed if loadFormerAnalysisResult is True\n panel_cols = 5,\n panel_rows = 5,\n useCV2 = False, # switch off opencv since it doesn't work yet\n useCV = False,\n averageScoreOrdering=True, # take an average of the ordering to get a total score for the entire model over the time series\n #kwargs = {}, # keyword arguments for the matchinging algorithm, just in case\n **kwargs\n ):\n if key1 != '' or key2 != '':\n summary_folder += '_'.join([key1, key2])\n if not averageScoreOrdering:\n summary_folder += '_nonaver'\n summary_folder += '%dx%d/' % (panel_rows, panel_cols)\n\n t0 = time.time()\n if loadFormerAnalysisResult:\n mv_folder = summary_folder + matching_algorithm_name + '/'\n result = pickle.load(open(mv_folder + 'result.pydump', 'r'))\n\n \"\"\"\n result= {'best matches' : bestMatchAtEachTime, \n 'best overall match' : bestMatchOverall, \n 'all scores' : scores,\n 'matching algorithm' : matching_algorithm_name,\n 'dataset1[obs]' : observations,\n 'dataset2[models]' : modelsFolder,\n 'time spent' : time.time() - t0,\n 'mv folder' : mv_folder, #added 2013-09-23\n 'ordering' : ordering, #added 2013-09-23\n }\n \"\"\"\n bestMatchAtEachTime = result['best matches'] \n bestMatchOverall = result['best overall match']\n scores = result['all scores']\n matching_algorithm_name = result['matching algorithm']\n observations = result['dataset1[obs]']\n modelsFolder = result['dataset2[models]']\n #mv_folder = result['mv folder']\n ordering = result['ordering']\n\n # THE FOLLOWING 30 LINES ARE COPIED FROM \"ELSE...\" BELOW\n ########### ******\n ##\n #\n ds1 = pickle.load(open(observations,'r'))\n ds1.setFloor(0) # added 2013-09-27\n ds1.list = [v for v in ds1.list if (key1 in v.dataTime or key1 in v.name) \\\n and (key2 in v.dataTime or key2 in v.name)]\n ds1.list.sort(key=lambda v:v.dataTime)\n ds1.setVmin(-40) #line added 2013-09-23\n # debug\n #print ds1.list\n print 'loading list with keywords [ %s %s ]' %(key1, key2) \n ds1.load(verbose=verbose)\n # end debug\n modelsList = [v for v in os.listdir(modelsFolder) if '.pydump' in v]\n modelsList.sort(key = lambda v: int(re.findall(r'\\d+',v)[0]))\n print \"modelsList\", modelsList\n scores = {} # scores = {model number: [('dataTime', score), ('dataTime', score),...], ...}\n print 'loading ds2 (model data)'\n wrfs = [] # line added 2013-09-23\n for path in modelsList:\n modelIndex = int(re.findall(r'\\d+', path)[0])\n ds2 = pickle.load(open(modelsFolder+path, 'r'))\n ds2.setFloor(0) # added 2013-09-27\n # debug\n #ds2.list = [v for v in ds2.list if key1 in v.dataTime] # line redundant, 2013-09-23\n ds1.list, ds2.list = ds1.intersect(ds2) #line added 2013-09-23 \n ds2.setVmin(-40) #line added 2013-09-23\n wrfs.append(ds2) # line added 2013-09-23\n # end debug\n\n ####################\n\n\n else:\n # THE FOLLOWING 30 LINES ARE COPIED TO \"IF...\" ABOVE\n ########### ******\n ##\n #\n ds1 = pickle.load(open(observations,'r'))\n ds1.list = [v for v in ds1.list if (key1 in v.dataTime or key1 in v.name) \\\n and (key2 in v.dataTime or key2 in v.name)]\n ds1.list.sort(key=lambda v:v.dataTime)\n ds1.setVmin(-40) #line added 2013-09-23\n # debug\n #print ds1.list\n print 'loading list with keywords [ %s %s ]' %(key1, key2) \n ds1.load(verbose=verbose)\n ds1.setFloor(0) # added 2013-09-27\n # end debug\n modelsList = [v for v in os.listdir(modelsFolder) if '.pydump' in v]\n modelsList.sort(key = lambda v: int(re.findall(r'\\d+',v)[0]))\n print \"modelsList\", modelsList\n scores = {} # scores = {model number: [('dataTime', score), ('dataTime', score),...], ...}\n print 'loading ds2 (model data)'\n wrfs = [] # line added 2013-09-23\n ds1.commonMaskSet=False # tempory attribute added 2013-09-27\n for path in modelsList:\n modelIndex = int(re.findall(r'\\d+', path)[0])\n ds2 = pickle.load(open(modelsFolder+path, 'r'))\n # debug\n #ds2.list = [v for v in ds2.list if key1 in v.dataTime] # line redundant, 2013-09-23\n ds1.list, ds2.list = ds1.intersect(ds2) #line added 2013-09-23 \n ds2.setFloor(0) # added 2013-09-27\n ds2[0].matrix.mask += ds1[0].matrix.mask # added 2013-09-27\n ds2.setCommonMask() # added 2013-09-27\n if not ds1.commonMaskSet: # added 2013-09-27\n ds1[0].matrix.mask +=ds2[0].matrix.mask # added 2013-09-27\n ds1.setCommonMask() # added 2013-09-27\n ds2.setVmin(-40) #line added 2013-09-23\n wrfs.append(ds2) # line added 2013-09-23\n # end debug\n\n #######################################################################################################################\n # key lines\n print 'computing scores between %s and %s ...' %(ds1.name, ds2.name)\n s, matching_algorithm_name = matchingAlgorithm(ds1, ds2, **kwargs) # list of the form [('20120612.0000', 0.9),...] \n scores[modelIndex] = dict(s) # { model index: {'dataTime': score,...}, ...}\n # debug\n if verbose:\n print scores\n print '..................................................'\n #\n ########################################################################################################################\n\n print 'sorting the results...'\n #print 'scores', '\\n', scores #debug\n # bestMatchAtEachTime\n T1 = [v.dataTime for v in ds1]\n T2 = [v.dataTime for v in ds2] #any ds2 will do - they are the same, in this case we pick the last one \n T = set(T1).intersection(T2)\n dataTimeList = sorted(list(T))\n\n #print 'dataTimeList:', dataTimeList #debug\n highestScores = []\n ordering = [] #added 2013-09-23\n\n if averageScoreOrdering: # take the average scores before ordering to make it uniform\n ave_scores = [ (v, np.mean( [scores[v].values() ] ) ) for v in scores.keys()]\n ave_scores = dict(ave_scores)\n print '..taking averages...............................'\n print ave_scores\n for u in scores.keys():\n for t in scores[u].keys():\n scores[u][t] = ave_scores[u]\n print '\\n'.join([str(scores[v]) for v in scores.keys()])\n\n for t in dataTimeList:\n print t\n highestScoreAt_t = max([scores[u][t] for u in scores.keys()]) # u = model index (integer 1-20), t = dataTime (string)\n new_entry = (t, highestScoreAt_t, [u for u in scores.keys() if scores[u][t]==highestScoreAt_t])\n print new_entry\n highestScores.append(new_entry)\n #added 2013-09-23\n ordering.append([0]+sorted(range(1, len(modelsList)+1), key=lambda v: scores[v][t], reverse=True))\n\n\n\n\n # bestMatchOverall\n print scores # debug\n averageScores = [(Model, np.mean(scores[Model].values())) for Model in scores.keys()] # list of [(Model, mean score),...]\n highestScore = max([v[1] for v in averageScores])\n bestMatch = [u[0] for u in averageScores if u[1]==highestScore] \n # debug\n print 'average scores , highest average, best match', averageScores, highestScore, bestMatch\n #xx= raw_input('press enter') \n\n\n bestMatchAtEachTime = highestScores\n bestMatchOverall = (highestScore, bestMatch)\n\n # the following lines do not work well since oftentimes we have multiple pointers (names) to a single object/function\n # and so they are replaced by returning the function name in the function itself\n #allvars = vars()\n #matching_algorithm_name = [v for v in allvars if allvars[v]==matchingAlgorithm][0]\n #mv_folder = summary_folder + matching_algorithm_name + '/'\n mv_folder = summary_folder + matching_algorithm_name + '/' # added 2013-09-27\n\n result= {'best matches' : bestMatchAtEachTime, \n 'best overall match' : bestMatchOverall, \n 'all scores' : scores,\n 'matching algorithm' : matching_algorithm_name,\n 'dataset1[obs]' : observations,\n 'dataset2[models]' : modelsFolder,\n 'time spent' : time.time() - t0,\n 'mv folder' : mv_folder, #added 2013-09-23\n 'ordering' : ordering, #added 2013-09-23\n }\n\n try:\n os.makedirs(mv_folder)\n print 'folder', mv_folder, \"created!\"\n except:\n print 'folder', mv_folder, 'exists!'\n pickle.dump(result, open(mv_folder+'result.pydump','w'))\n\n ############################################################################\n # to make the videos/collected images\n # calling armor.dataStreamTools.makeVideo.makeVideo()\n # 2013-09-23\n\n from . import makeVideo as mv\n \n try:\n os.makedirs(mv_folder)\n print 'folder', mv_folder, \"created!\"\n except:\n print 'folder', mv_folder, 'exists!'\n # make images\n if makeSmallImages:\n print 'making small images'\n for ds in [ds1] + wrfs:\n #ds.imageFolder= mv_folder + ds.name + '/'\n ds.imageFolder= summary_root + ds.name + '/' #added 2013-09-27 to replace the above line - too tired waiting the plots\n ds.saveImages(drawCoast=True, flipud=True, dpi=smallDpi)\n print ds.imageFolder\n else:\n # just set the paths\n for ds in [ds1]+wrfs:\n ds.imageFolder = summary_root + ds.name +'/' # /home/k/ARMOR/data/KONG-REY/summary/WRF01.DBZ/\n for dbzPattern in ds.list:\n dbzPattern.imagePath = ds.imageFolder + ds.name + dbzPattern.dataTime + '.png'\n print dbzPattern.imagePath\n #ds.setImagePaths() # doesn't for for old pickled objects???\n\n\n # make panels\n\n mv.makeVideo(DSS = [ds1]+ wrfs, # [ds0, ds1, ds2, ds3, ds4, ...], a list of armor.pattern.DBZstream objects\n panel_cols = panel_cols, # number of colums in the panel\n panel_rows = panel_rows, # no need to be filled\n #fourcc = cv.CV_FOURCC('F', 'L', 'V', '1'),\n #fps = defaultFps,\n extension= '.avi',\n #fourcc = cv.CV_FOURCC('P', 'I', 'M', '1'),\n outputFileName =\"\",\n outputFolder= mv_folder,\n sandbox = mv_folder ,\n saveFrames = True, # saving the frames as images\n frameDpi = frameDpi,\n useCV2 = useCV2,\n useCV = useCV,\n ordering = ordering, # ordering of the models\n )\n ############################################################################\n # dumping the results\n try:\n os.makedirs(summary_folder+ 'pydumps/')\n except:\n print 'Folder exists! -- ', summary_folder+ 'pydumps/' \n\n pickle.dump(result, open(summary_folder+ 'pydumps/results_' + matching_algorithm_name+ '+' +\\\n key1 + '+' + key2 + str(int(time.time())) + '.pydump', 'w'))\n\n \n\n print \"time spent:\", time.time()-t0\n return result","def compare(model, input1, input2): \n patch_image_time_start = time.time()\n input1_patches = get_patches_non_overlap(input1, 48, 48)\n input2_patches = get_patches_non_overlap(input2, 48, 48)\n patch_image_time_end = time.time()\n compare_image_time_start = time.time()\n pred = model.predict([input1_patches, input2_patches])\n compare_image_time_end = time.time()\n\n result[\"patch_retrieval_time\"].append(patch_image_time_end - patch_image_time_start)\n result[\"image_comparison_time\"].append(compare_image_time_end - compare_image_time_start)\n\n return np.sum(pred)","def ransac(matches, kp1, kp2, s=4, threshold=3, maxIterations=2000, returnMatches=False, inlierRatio=0.05, ransacRatio=0.6):\n\n sizes_kp1 = [kp1[dt[0].queryIdx].size for dt in matches]\n sizes_kp2 = [kp1[dt[0].trainIdx].size for dt in matches]\n tup_matches_kp1 = [kp1[dt[0].queryIdx].pt for dt in matches]\n tup_matches_kp2 = [kp2[dt[0].trainIdx].pt for dt in matches]\n matches_kp1 = np.array([[h for h in kp] + [1] for kp in tup_matches_kp1])\n matches_kp2 = np.array([[h for h in kp] + [1] for kp in tup_matches_kp2])\n\n cnt_matches = len(matches)\n\n max_matches = []\n max_p1, max_p2 = [], []\n max_p1_sizes, max_p2_sizes = [], []\n max_total = 0\n\n for iter in range(maxIterations):\n # Find Homography based on random sample\n data = random.sample(matches, s)\n data_p1 = np.array([matches_kp1[dt[0].queryIdx] for dt in data])\n data_p2 = np.array([matches_kp2[dt[0].trainIdx] for dt in data])\n homography = homomat(data_p1[:, :2], data_p2[:, :2])\n\n # Find P1 projection from the homography matrix\n projected_p2 = np.dot(homography, matches_kp1.transpose())\n projected_p2 = projected_p2[0:3] / projected_p2[2] # make sure w' is 1\n projected_p2 = projected_p2.transpose()\n\n # Initialize Current Matches\n current_matches = []\n current_p1, current_p2 = [], []\n current_p1_sizes, current_p2_sizes = [], []\n current_total = 0\n\n # Check for inliers and outliers for each matches\n for i, (match) in enumerate(matches):\n # normalize the error\n error = np.linalg.norm(matches_kp2[i] - projected_p2[i])\n\n # Check for inliers\n if error < threshold:\n current_matches.append([cv.DMatch(current_total, current_total, match[0].distance)])\n current_p1.append(matches_kp1[i][0:2])\n current_p2.append(matches_kp2[i][0:2])\n current_p1_sizes.append(sizes_kp1[i])\n current_p2_sizes.append(sizes_kp2[i])\n current_total += 1\n\n # If\n if current_total > max_total and current_total >= np.round(inlierRatio*cnt_matches):\n max_matches = current_matches\n max_p1 = current_p1\n max_p2 = current_p2\n max_p1_sizes = current_p1_sizes\n max_p2_sizes = current_p2_sizes\n max_total = current_total\n\n # # we are done in case we have enough inliers\n if current_total > cnt_matches * ransacRatio:\n break\n\n\n # Re-evaluate the Homography based on the best inliers\n max_homography = homomat(np.array(max_p1), np.array(max_p2))\n\n if returnMatches:\n max_kp1 = [cv.KeyPoint(d[0], d[1], max_p1_sizes[i]) for i, d in enumerate(max_p1)]\n max_kp2 = [cv.KeyPoint(d[0], d[1], max_p2_sizes[i]) for i, d in enumerate(max_p2)]\n return max_homography, max_matches, max_kp1, max_kp2\n\n return max_homography","def findMatchesBetweenImages(image_1, image_2):\n # matches - type: list of cv2.DMath\n matches = None\n # image_1_kp - type: list of cv2.KeyPoint items.\n image_1_kp = None\n # image_1_desc - type: numpy.ndarray of numpy.uint8 values.\n image_1_desc = None\n # image_2_kp - type: list of cv2.KeyPoint items.\n image_2_kp = None\n # image_2_desc - type: numpy.ndarray of numpy.uint8 values.\n image_2_desc = None\n # WRITE YOUR CODE HERE.\n\n sift = cv2.ORB_create()\n image_1_kp, image_1_desc = sift.detectAndCompute(image_1, None)\n image_2_kp, image_2_desc = sift.detectAndCompute(image_2, None)\n\n # bf = cv2.BFMatcher(cv2.NORM_HAMMING, crossCheck=True)\n # matches = bf.match(image_1_desc, image_2_desc)\n # matches = sorted(matches, key = lambda x:x.distance)\n # matches = matches[:10]\n\n bf = cv2.BFMatcher()\n matches = bf.knnMatch(image_1_desc, image_2_desc, k=2)\n\n # Apply ratio test\n good = []\n for m, n in matches:\n print(m.distance, n.distance, m.distance < .75*n.distance)\n if m.distance < (0.75 * n.distance):\n good.append([m])\n\n # We coded the return statement for you. You are free to modify it -- just\n # make sure the tests pass.\n print(len(good), good)\n return image_1_kp, image_2_kp, matches\n # END OF FUNCTION.","def compute_precision_recall_bbox(per_image_detections, per_image_gts, num_gt_classes,\n matching_iou_threshold=0.5):\n per_image_eval = per_image_evaluation.PerImageEvaluation(\n num_groundtruth_classes=num_gt_classes,\n matching_iou_threshold=matching_iou_threshold,\n nms_iou_threshold=1.0,\n nms_max_output_boxes=10000\n )\n\n print('Running per-object analysis...')\n\n detection_tp_fp = defaultdict(list) # key is the category; in each list, 1 is tp, 0 is fp\n detection_scores = defaultdict(list)\n num_total_gt = defaultdict(int)\n\n for image_id, dets in tqdm(per_image_detections.items()):\n\n detected_boxes = np.array(dets['boxes'], dtype=np.float32)\n detected_scores = np.array(dets['scores'], dtype=np.float32)\n # labels input to compute_object_detection_metrics() needs to start at 0, not 1\n detected_labels = np.array(dets['labels'], dtype=np.int) - 1 # start at 0\n # num_detections = len(dets['boxes'])\n\n gts = per_image_gts[image_id]\n gt_boxes = np.array(gts['gt_boxes'], dtype=np.float32)\n gt_labels = np.array(gts['gt_labels'], dtype=np.int) - 1 # start at 0\n num_gts = len(gts['gt_boxes'])\n\n groundtruth_is_difficult_list = np.zeros(num_gts, dtype=bool) # place holders - we don't have these\n groundtruth_is_group_of_list = np.zeros(num_gts, dtype=bool)\n\n # to prevent 'Invalid dimensions for box data.' error\n if num_gts == 0:\n # this box will not match any detections\n gt_boxes = np.array([[0, 0, 0, 0]], dtype=np.float32)\n\n scores, tp_fp_labels, is_class_correctly_detected_in_image = (\n per_image_eval.compute_object_detection_metrics(\n detected_boxes=detected_boxes,\n detected_scores=detected_scores,\n detected_class_labels=detected_labels,\n groundtruth_boxes=gt_boxes,\n groundtruth_class_labels=gt_labels,\n groundtruth_is_difficult_list=groundtruth_is_difficult_list,\n groundtruth_is_group_of_list=groundtruth_is_group_of_list\n )\n )\n\n for i, tp_fp_labels_cat in enumerate(tp_fp_labels):\n assert sum(tp_fp_labels_cat) <= sum(gt_labels == i) # true positives < gt of that category\n cat = i + 1 # categories start at 1\n detection_tp_fp[cat].append(tp_fp_labels_cat)\n detection_scores[cat].append(scores[i])\n num_total_gt[cat] += sum(gt_labels == i) # gt_labels start at 0\n\n all_scores = []\n all_tp_fp = []\n\n print('Computing precision recall for each category...')\n per_cat_metrics = {}\n for i in range(num_gt_classes):\n cat = i + 1\n scores_cat = np.concatenate(detection_scores[cat])\n tp_fp_cat = np.concatenate(detection_tp_fp[cat]).astype(np.bool)\n all_scores.append(scores_cat)\n all_tp_fp.append(tp_fp_cat)\n\n precision, recall = metrics.compute_precision_recall(\n scores_cat, tp_fp_cat, num_total_gt[cat]\n )\n average_precision = metrics.compute_average_precision(precision, recall)\n\n per_cat_metrics[cat] = {\n 'category': cat,\n 'precision': precision,\n 'recall': recall,\n 'average_precision': average_precision,\n 'scores': scores_cat,\n 'tp_fp': tp_fp_cat,\n 'num_gt': num_total_gt[cat]\n }\n print('Number of ground truth in category {} is {}'.format(cat, num_total_gt[cat]))\n\n # compute one-class precision/recall/average precision (if every box is just of an object class)\n all_scores = np.concatenate(all_scores)\n all_tp_fp = np.concatenate(all_tp_fp)\n overall_gt_count = sum(num_total_gt.values())\n\n one_class_prec, one_class_recall = metrics.compute_precision_recall(\n all_scores, all_tp_fp, overall_gt_count\n )\n one_class_average_precision = metrics.compute_average_precision(one_class_prec, one_class_recall)\n\n per_cat_metrics['one_class'] = {\n 'category': 'one_class',\n 'precision': one_class_prec,\n 'recall': one_class_recall,\n 'average_precision': one_class_average_precision,\n 'scores': all_scores,\n 'tp_fp': all_tp_fp,\n 'num_gt': overall_gt_count\n }\n\n return per_cat_metrics","def evaluate_matrices(self, precision_matrix, recall_matrix):\n\t\tground_truth_count = precision_matrix.shape[0]\n\t\tdetection_count = precision_matrix.shape[1]\n\t\tground_truth_sets_precision = defaultdict(set) # number of ground truth items that match a particular detection in the precision matrix\n\t\tdetection_sets_precision = defaultdict(set) # number of detection items that match a particular ground truth in the precision matrix\n\t\tground_truth_sets_recall = defaultdict(set) # number of ground truth items that match a particular detection in the recall matrix\n\t\tdetection_sets_recall = defaultdict(set) # number of detection items that match a particular ground truth in the recall matrix\n\n\t\tfor gt_index in range(ground_truth_count):\n\t\t\tfor det_index in range(detection_count):\n\t\t\t\tif precision_matrix[gt_index, det_index] >= self.precision_threshold:\n\t\t\t\t\tground_truth_sets_precision[det_index].add(gt_index)\n\t\t\t\t\tdetection_sets_precision[gt_index].add(det_index)\n\t\t\t\tif recall_matrix[gt_index, det_index] >= self.recall_threshold:\n\t\t\t\t\tground_truth_sets_recall[det_index].add(gt_index)\n\t\t\t\t\tdetection_sets_recall[gt_index].add(det_index)\n\n\t\tmatch_ground_truth = 0. # sum of MatchG\n\t\tmatch_detection = 0. # sum of MatchD\n\n\t\tone_to_one_precision = set()\n\t\tfor gt_index in detection_sets_precision:\n\t\t\tmatching_detections_precision = detection_sets_precision[gt_index]\n\t\t\tif len(matching_detections_precision) == 1:\n\t\t\t\t(detection_precision, ) = matching_detections_precision\n\t\t\t\tif len(ground_truth_sets_precision[detection_precision]) == 1:\n\t\t\t\t\tone_to_one_precision.add((gt_index, detection_precision))\n\t\t\telse:\n\t\t\t\t# one-to-many (one ground truth to many detections)\n\t\t\t\tgt_sum = 0.\n\t\t\t\tfor detection_precision in matching_detections_precision:\n\t\t\t\t\tgt_sum += recall_matrix[gt_index, detection_precision]\n\t\t\t\tif gt_sum >= self.recall_threshold:\n\t\t\t\t\t#print(\"1:N ~ GT {} : DT {}\".format(gt_index,matching_detections_precision))\n\t\t\t\t\tmatch_ground_truth += self.scatter_punishment(matching_detections_precision)\n\t\t\t\t\tmatch_detection += len(matching_detections_precision) * self.scatter_punishment(matching_detections_precision)\n\n\t\tone_to_one_recall = set()\n\t\tfor det_index in ground_truth_sets_recall:\n\t\t\tmatching_ground_truths_recall = ground_truth_sets_recall[det_index]\n\t\t\tif len(matching_ground_truths_recall) == 1:\n\t\t\t\t(ground_truth_recall, ) = matching_ground_truths_recall\n\t\t\t\tif len(detection_sets_recall[ground_truth_recall]) == 1:\n\t\t\t\t\tone_to_one_recall.add((ground_truth_recall, det_index))\n\t\t\telse:\n\t\t\t\t# many-to-one (many ground truths covered by one detection)\n\t\t\t\tdet_sum = 0\n\t\t\t\tfor ground_truth_recall in matching_ground_truths_recall:\n\t\t\t\t\tdet_sum += precision_matrix[ground_truth_recall, det_index]\n\t\t\t\tif det_sum >= self.precision_threshold:\n\t\t\t\t\t#print(\"N:1 ~ DT {} : GT {}\".format(det_index,matching_ground_truths_recall))\n\t\t\t\t\tmatch_detection += self.scatter_punishment(matching_ground_truths_recall)\n\t\t\t\t\tmatch_ground_truth += len(matching_ground_truths_recall) * self.scatter_punishment(matching_ground_truths_recall)\n\n\t\tone_to_one_matches = one_to_one_precision & one_to_one_recall\n\t\tmatch_ground_truth += len(one_to_one_matches)\n\t\tmatch_detection += len(one_to_one_matches)\n\n\t\trecall = match_ground_truth / float(ground_truth_count)\n\t\tprecision = match_detection / float(detection_count)\n\t\treturn (precision, recall, (match_detection, float(detection_count)), (match_ground_truth, float(ground_truth_count)))","def match(image1,image2,threshold,useRansac=False,t_orientation=30,t_scale=0.5):\r\n im1, keypoints1, descriptors1 = ReadKeys(image1)\r\n im2, keypoints2, descriptors2 = ReadKeys(image2)\r\n #\r\n # REPLACE THIS CODE WITH YOUR SOLUTION (ASSIGNMENT 5, QUESTION 3)\r\n #\r\n # Generate five random matches (for testing purposes)\r\n # matched_pairs = []\r\n # num = 5\r\n # for i in range(num):\r\n # matched_pairs.append([keypoints1[i],keypoints2[i]])\r\n # return DisplayMatches(im1, im2, matched_pairs)\r\n\r\n # END OF SECTION OF CODE TO REPLACE\r\n #\r\n\r\n #q3\r\n matched_pairs = []\r\n between_angles = np.arccos(np.dot(descriptors1, np.transpose(descriptors2)))\r\n for i, row in enumerate(between_angles):\r\n \tratio = sorted(row)[0] / sorted(row)[1]\r\n \tif ratio <= threshold:\r\n\t \tmatched_pairs.append([keypoints1[i], keypoints2[np.where(row == sorted(row)[0])[0][0]]])\r\n # print(matched_pairs)\r\n if useRansac is False:\r\n return DisplayMatches(im1, im2, matched_pairs)\r\n\t# \r\n\r\n #q4\r\n repetition = 10\r\n subsets = [[]] * repetition\r\n for i in range(repetition):\r\n r = random.randint(0, len(matched_pairs))\r\n for match in matched_pairs:\r\n ds1, ds2 = matched_pairs[r][1][2]/matched_pairs[r][0][2], match[1][2]/match[0][2]\r\n do1, do2 = (matched_pairs[r][1][3]-matched_pairs[r][0][3]), (match[1][3]-match[0][3])\r\n if abs(ds2 - ds1) <= t_scale * ds1 and abs(do2 - do1) % (2 * math.pi) <= t_orientation:\r\n subsets[i].append(match)\r\n\r\n max_i, max_len = 0, subsets[0]\r\n for i in range(10):\r\n l = len(subsets[i])\r\n if l > max_len:\r\n max_len = l\r\n max_i = i\r\n\r\n im3 = DisplayMatches(im1, im2, subsets[max_i])\r\n return im3","def evaluate_detections(ground_truth, predictions, class_name, overlap_threshold=0.5,\n allow_multiple_matches_per_ignored=True,\n verbose=True):\n\n Detection = namedtuple('Detection', ['image', 'bbox', 'score', 'gt_match'])\n GT = namedtuple('GroundTruth', ['bbox', 'is_matched', 'is_ignored'])\n detections = [Detection(image=img_pred.image_path,\n bbox=np.array(obj_pred[\"bbox\"]),\n score=obj_pred.get(\"score\", 0.0),\n gt_match=-1)\n for img_pred in predictions\n for obj_pred in img_pred\n if obj_pred[\"type\"] == class_name]\n\n scores = np.array([detection.score for detection in detections])\n sorted_ind = np.argsort(-scores)\n detections = [detections[i] for i in sorted_ind]\n\n gts = {}\n for img_gt in ground_truth:\n gts[img_gt.image_path] = GT(\n bbox=np.vstack([np.array(obj_gt[\"bbox\"]) for obj_gt in img_gt]) if img_gt else np.empty(\n (0, 4)),\n is_matched=np.zeros(len(img_gt), dtype=bool),\n is_ignored=np.array([obj_gt.get(\"is_ignored\", False) for obj_gt in img_gt], dtype=bool))\n\n detections_num = len(detections)\n true_pos = np.zeros(detections_num)\n false_pos = np.zeros(detections_num)\n\n for i, detection in tqdm(enumerate(detections), desc=\"Processing detections\",\n disable=not verbose):\n image_path = detection.image\n bboxes_gt = gts[image_path].bbox\n bbox = detection.bbox\n max_overlap = -np.inf\n\n if bboxes_gt is not None and bboxes_gt.shape[0] > 0:\n intersection_xmin = np.maximum(bboxes_gt[:, 0], bbox[0])\n intersection_ymin = np.maximum(bboxes_gt[:, 1], bbox[1])\n intersection_xmax = np.minimum(bboxes_gt[:, 0] + bboxes_gt[:, 2], bbox[0] + bbox[2])\n intersection_ymax = np.minimum(bboxes_gt[:, 1] + bboxes_gt[:, 3], bbox[1] + bbox[3])\n intersection_width = np.maximum(intersection_xmax - intersection_xmin, 0.)\n intersection_height = np.maximum(intersection_ymax - intersection_ymin, 0.)\n intersection = intersection_width * intersection_height\n\n det_area = bbox[2] * bbox[3]\n gt_area = bboxes_gt[:, 2] * bboxes_gt[:, 3]\n union = (det_area + gt_area - intersection)\n ignored_mask = gts[image_path].is_ignored\n if allow_multiple_matches_per_ignored:\n if np.any(ignored_mask):\n union[ignored_mask] = det_area\n\n overlaps = intersection / union\n # Match not ignored ground truths first.\n if np.any(~ignored_mask):\n overlaps_filtered = np.copy(overlaps)\n overlaps_filtered[ignored_mask] = 0.0\n max_overlap = np.max(overlaps_filtered)\n argmax_overlap = np.argmax(overlaps_filtered)\n # If match with non-ignored ground truth is not good enough,\n # try to match with ignored ones.\n if max_overlap < overlap_threshold and np.any(ignored_mask):\n overlaps_filtered = np.copy(overlaps)\n overlaps_filtered[~ignored_mask] = 0.0\n max_overlap = np.max(overlaps_filtered)\n argmax_overlap = np.argmax(overlaps_filtered)\n detections[i] = detection._replace(gt_match=argmax_overlap)\n\n if max_overlap >= overlap_threshold:\n if not gts[image_path].is_ignored[argmax_overlap]:\n if not gts[image_path].is_matched[argmax_overlap]:\n true_pos[i] = 1.\n gts[image_path].is_matched[argmax_overlap] = True\n else:\n false_pos[i] = 1.\n elif not allow_multiple_matches_per_ignored:\n gts[image_path].is_matched[argmax_overlap] = True\n else:\n false_pos[i] = 1.\n\n false_pos = np.cumsum(false_pos)\n true_pos = np.cumsum(true_pos)\n\n debug_visualization = False\n if debug_visualization:\n for image_path, bboxes_gt in gts.items():\n\n print(image_path)\n image = cv2.imread(image_path)\n image_gt = np.copy(image)\n for bbox in bboxes_gt.bbox:\n cv2.rectangle(image_gt, tuple(bbox[:2]), tuple(bbox[2:] + bbox[:2]),\n color=(255, 255, 0), thickness=2)\n cv2.imshow(\"gt\", image_gt)\n for detection in detections:\n if detection.image != image_path:\n continue\n bbox = detection.bbox\n cv2.rectangle(image, tuple(bbox[:2]), tuple(bbox[2:] + bbox[:2]), color=(0, 255, 0),\n thickness=2)\n if detection.gt_match is not None:\n bbox = bboxes_gt.bbox[detection.gt_match]\n cv2.rectangle(image, tuple(bbox[:2]), tuple(bbox[2:] + bbox[:2]),\n color=(0, 0, 255), thickness=1)\n cv2.imshow(\"image\", image)\n cv2.waitKey(0)\n\n # Handle equal-score detections.\n # Get index of the last occurrence of a score.\n ind = len(scores) - np.unique(scores[sorted_ind[::-1]], return_index=True)[1] - 1\n ind = ind[::-1]\n # Though away redundant points.\n false_pos = false_pos[ind]\n true_pos = true_pos[ind]\n\n total_positives_num = np.sum([np.count_nonzero(~gt.is_ignored) for gt in gts.values()])\n recall = true_pos / float(total_positives_num)\n # Avoid divide by zero in case the first detection matches an ignored ground truth.\n precision = true_pos / np.maximum(true_pos + false_pos, np.finfo(np.float64).eps)\n miss_rate = 1.0 - recall\n fppi = false_pos / float(len(gts))\n\n return recall, precision, miss_rate, fppi","def _match_keypoints(self, query_image: Image) -> typing.Tuple[typing.List[int], typing.List[typing.Any]]:\n query_image.get_keypoints_and_descriptors()\n\n # matches = self.matcher.match(query_image.descriptor)\n matches = self.matcher.knnMatch(query_image.descriptor, k=2)\n good = []\n for m_n in matches:\n if len(m_n) == 1:\n good.append(m_n[0])\n continue\n elif len(m_n) != 2:\n continue\n (m, n) = m_n\n if m.distance < 0.7 * n.distance:\n good.append(m)\n\n images_scores = [0] * len(self._candidate_images)\n images_matches = [None] * len(self._candidate_images)\n for image_index, image in enumerate(self._candidate_images):\n matches_scores = []\n matches = []\n for i, match in enumerate(good):\n if match.imgIdx != image_index:\n continue\n matches.append(match)\n matches_scores.append((256 - match.distance) / 256)\n\n match_cnt = len(matches_scores)\n if match_cnt <= 0:\n continue\n\n images_scores[image_index] = (\n 0.5 + ((math.tanh(match_cnt / 3 - 1)) / 2)) * (sum(matches_scores) / match_cnt)\n\n images_matches[image_index] = matches\n\n return images_scores, images_matches","def analyze_similarities():\r\n print('Total number of candidate pairs:', len(pairs))\r\n print(f'\\nNumber of actual item pairs in the train set: {pairs[\"true_match\"].sum()}\\n')\r\n\r\n for feature in ['text_score', 'image_score', 'txt_img_score', 'words_ratio', 'txt_img_words']:\r\n\r\n # Check distribution of True and False predictions for various similarity scores\r\n print('-' * 50)\r\n print(f'\\nDistribution of True/False predictions for {feature}')\r\n for thr in (0.6, 0.65, 0.7, 0.75, 0.8, 0.85, 0.9, 0.95):\r\n print('-' * 50)\r\n print(f'Similarity score over {thr}')\r\n pairs_sample = pairs[pairs[feature] >= thr]\r\n print(f'Number of similar item pairs: {len(pairs_sample)}')\r\n print(pairs_sample['true_match'].value_counts(normalize=True))\r\n\r\n # Check if identical phash can be used to improve the accuracy\r\n same_phash = pairs[pairs['phash_match'] == 1]\r\n different_phash = pairs[pairs['phash_match'] == 0]\r\n\r\n print('\\nFor item pairs with the same phash:')\r\n print(same_phash['true_match'].value_counts(normalize=True))\r\n print('Number of item pairs in this subset:', len(same_phash))\r\n\r\n print('\\nFor item pairs with different phash:')\r\n print(different_phash['true_match'].value_counts(normalize=True))\r\n print('Number of item pairs in this subset:', len(different_phash))\r\n\r\n # Check if numbers in titles can be used to improve the accuracy\r\n same_numbers = pairs[pairs['nums_match'] == 1]\r\n different_numbers = pairs[pairs['nums_match'] == 0]\r\n\r\n print('\\nFor item pairs with the same numbers:')\r\n print(same_numbers['true_match'].value_counts(normalize=True))\r\n print('Number of item pairs in this subset:', len(same_numbers))\r\n\r\n print('\\nFor item pairs with different numbers:')\r\n print(different_numbers['true_match'].value_counts(normalize=True))\r\n print('Number of item pairs in this subset:', len(different_numbers))","def test_compare_classifier_performance_from_pred_vis_api(experiment_to_use):\n experiment = experiment_to_use\n prediction = experiment.predictions\n viz_outputs = (\"pdf\", \"png\")\n with TemporaryDirectory() as tmpvizdir:\n for viz_output in viz_outputs:\n vis_output_pattern_pdf = tmpvizdir + f\"/*.{viz_output}\"\n visualize.compare_classifiers_performance_from_pred(\n [prediction, prediction],\n experiment.ground_truth,\n experiment.ground_truth_metadata,\n experiment.output_feature_name,\n labels_limit=0,\n model_namess=[\"Model1\", \"Model2\"],\n output_directory=tmpvizdir,\n file_format=viz_output,\n )\n figure_cnt = glob.glob(vis_output_pattern_pdf)\n assert 1 == len(figure_cnt)","def run_multiple_advanced_experiments(self, path_to_csv_files, algorithms_weights_list, thresholds):\n starting_time = time.time()\n print(\"started advanced precision recall image_matching_module\")\n for algorithms_weights in algorithms_weights_list:\n start_file_time = time.time()\n\n first_run = True\n for algorithm, weight in algorithms_weights:\n if first_run:\n algorithm_name = algorithm.get_algorithm_name() + weight\n first_run = False\n else:\n algorithm_name = algorithm_name + \"_\" + algorithm.get_algorithm_name() + weight\n\n print(\"started algorithm:\", algorithm_name)\n self.run_single_advanced_experiment(path_to_csv_files, algorithms_weights, thresholds)\n finish_file_time = time.time()\n print(\"finished algorithm:\", algorithm_name)\n print(\"time took:\", (finish_file_time - start_file_time) / 60, \"minutes\")\n finish_time = time.time()\n hours = int((finish_time - starting_time) // 3600)\n minutes = (finish_time - starting_time) / 60\n print(\"finished advanced precision recall image_matching_module\")\n print(\"time image_matching_module took:\", hours, \"hours and\", minutes, \"minutes\")","def run_multiple_simple_experiments(self, path_to_csv_files, algorithms, thresholds):\n starting_time = time.time()\n print(\"started simple precision recall image_matching_module\")\n for algorithm in algorithms:\n start_file_time = time.time()\n print(\"started algorithm:\", algorithm.get_algorithm_name())\n self.run_single_simple_experiment(path_to_csv_files, algorithm, thresholds)\n finish_file_time = time.time()\n print(\"finished algorithm:\", algorithm.get_algorithm_name())\n print(\"time took:\", (finish_file_time - start_file_time) / 60, \"minutes\")\n finish_time = time.time()\n hours = int((finish_time - starting_time) // 3600)\n minutes = (finish_time - starting_time) / 60\n print(\"finished simple precision recall image_matching_module\")\n print(\"time image_matching_module took:\", hours, \"hours and\", minutes, \"minutes\")","def _run():\n matching_terms = {'a', 'b'}\n source_counts = {'a': 10, 'b': 50, 'c': 25}\n target_counts = {'a': 4, 'b': 73, 'c': 15}\n source_chunk = ['a', 'b']\n target_chunk = ['a', 'c', 'b']\n source_distance = score.find_distance(\n matching_terms, source_chunk, source_counts)\n target_distance = score.find_distance(\n matching_terms, target_chunk, target_counts)\n match_score = score.vanilla(\n matching_terms, source_distance, target_distance, source_counts,\n target_counts)\n print('Calculated score:', match_score)","def calculates_results_stats(results_dic): \n # Creates empty dictionary for results_stats_dic\n results_stats_dic = dict()\n \n # Sets all counters to initial values of zero so that they can \n # be incremented while processing through the images in results_dic \n results_stats_dic['n_dogs_img'] = 0\n results_stats_dic['n_match'] = 0\n results_stats_dic['n_correct_dogs'] = 0\n results_stats_dic['n_correct_notdogs'] = 0\n results_stats_dic['n_correct_breed'] = 0\n \n # process through the results dictionary\n for key in results_dic:\n \n # Labels Match Exactly\n if results_dic[key][2] == 1:\n results_stats_dic['n_match'] += 1\n\n # TODO: 5a. REPLACE pass with CODE that counts how many pet images of\n # dogs had their breed correctly classified. This happens \n # when the pet image label indicates the image is-a-dog AND \n # the pet image label and the classifier label match. You \n # will need to write a conditional statement that determines\n # when the dog breed is correctly classified and then \n # increments 'n_correct_breed' by 1. Recall 'n_correct_breed' \n # is a key in the results_stats_dic dictionary with it's value \n # representing the number of correctly classified dog breeds.\n # \n # Pet Image Label is a Dog AND Labels match- counts Correct Breed\n if results_dic[key][3] == 1 and results_dic[key][2] == 1:\n results_stats_dic['n_correct_breed'] += 1\n \n # Pet Image Label is a Dog - counts number of dog images\n if results_dic[key][3] == 1:\n results_stats_dic['n_dogs_img'] += 1\n \n # Classifier classifies image as Dog (& pet image is a dog)\n # counts number of correct dog classifications\n if results_dic[key][4] == 1:\n results_stats_dic['n_correct_dogs'] += 1\n\n # TODO: 5b. REPLACE pass with CODE that counts how many pet images \n # that are NOT dogs were correctly classified. This happens \n # when the pet image label indicates the image is-NOT-a-dog \n # AND the classifier label indicates the images is-NOT-a-dog.\n # You will need to write a conditional statement that \n # determines when the classifier label indicates the image \n # is-NOT-a-dog and then increments 'n_correct_notdogs' by 1. \n # Recall the 'else:' above 'pass' already indicates that the \n # pet image label indicates the image is-NOT-a-dog and \n # 'n_correct_notdogs' is a key in the results_stats_dic dictionary \n # with it's value representing the number of correctly \n # classified NOT-a-dog images.\n # \n # Pet Image Label is NOT a Dog\n else:\n # Classifier classifies image as NOT a Dog(& pet image isn't a dog)\n # counts number of correct NOT dog clasifications.\n if results_dic[key][3] == 0 and results_dic[key][4] == 0:\n results_stats_dic['n_correct_notdogs'] += 1\n\n\n # Calculates run statistics (counts & percentages) below that are calculated\n # using the counters from above.\n\n # calculates number of total images\n results_stats_dic['n_images'] = len(results_dic)\n\n # calculates number of not-a-dog images using - images & dog images counts\n results_stats_dic['n_notdogs_img'] = (results_stats_dic['n_images'] - \n results_stats_dic['n_dogs_img']) \n\n # TODO: 5c. REPLACE zero(0.0) with CODE that calculates the % of correctly\n # matched images. Recall that this can be calculated by the\n # number of correctly matched images ('n_match') divided by the \n # number of images('n_images'). This result will need to be \n # multiplied by 100.0 to provide the percentage.\n # \n # Calculates % correct for matches\n results_stats_dic['pct_match'] = (results_stats_dic['n_match'] / results_stats_dic['n_images']) * 100\n\n # TODO: 5d. REPLACE zero(0.0) with CODE that calculates the % of correctly\n # classified dog images. Recall that this can be calculated by \n # the number of correctly classified dog images('n_correct_dogs')\n # divided by the number of dog images('n_dogs_img'). This result \n # will need to be multiplied by 100.0 to provide the percentage.\n # \n # Calculates % correct dogs\n results_stats_dic['pct_correct_dogs'] = (results_stats_dic['n_correct_dogs'] / results_stats_dic['n_dogs_img']) * 100\n\n # TODO: 5e. REPLACE zero(0.0) with CODE that calculates the % of correctly\n # classified breeds of dogs. Recall that this can be calculated \n # by the number of correctly classified breeds of dog('n_correct_breed') \n # divided by the number of dog images('n_dogs_img'). This result \n # will need to be multiplied by 100.0 to provide the percentage.\n # \n # Calculates % correct breed of dog\n results_stats_dic['pct_correct_breed'] = (results_stats_dic['n_correct_breed'] / results_stats_dic['n_dogs_img']) * 100\n\n # Calculates % correct not-a-dog images\n # Uses conditional statement for when no 'not a dog' images were submitted \n if results_stats_dic['n_notdogs_img'] > 0:\n results_stats_dic['pct_correct_notdogs'] = (results_stats_dic['n_correct_notdogs'] /\n results_stats_dic['n_notdogs_img']) * 100.0\n else:\n results_stats_dic['pct_correct_notdogs'] = 0.0\n\n \n # TODO 5f. REPLACE None with the results_stats_dic dictionary that you \n # created with this function \n return results_stats_dic","def testdata_matcher(fname1='easy1.png', fname2='easy2.png'):\n import utool as ut\n #import vtool as vt\n from vtool import image as gtool\n from vtool import features as feattool\n fpath1 = ut.grab_test_imgpath(fname1)\n fpath2 = ut.grab_test_imgpath(fname2)\n featkw = dict(rotation_invariance=True)\n kpts1, vecs1 = feattool.extract_features(fpath1, **featkw)\n kpts2, vecs2 = feattool.extract_features(fpath2, **featkw)\n #if featkw['rotation_invariance']:\n # print('ori stats 1 ' + ut.get_stats_str(vt.get_oris(kpts2)))\n # print('ori stats 2 ' + ut.get_stats_str(vt.get_oris(kpts1)))\n rchip1 = gtool.imread(fpath1)\n rchip2 = gtool.imread(fpath2)\n #chip1_shape = vt.gtool.open_image_size(fpath1)\n chip2_shape = gtool.open_image_size(fpath2)\n dlen_sqrd2 = chip2_shape[0] ** 2 + chip2_shape[1]\n testtup = (rchip1, rchip2, kpts1, vecs1, kpts2, vecs2, dlen_sqrd2)\n return testtup","def drawMatches(img1, kp1, img2, kp2, matches):\n\n # Create a new output image that concatenates the two images together\n # (a.k.a) a montage\n rows1 = img1.shape[0]\n cols1 = img1.shape[1]\n rows2 = img2.shape[0]\n cols2 = img2.shape[1]\n\n out = np.zeros((max([rows1,rows2]),cols1+cols2,3), dtype='uint8')\n\n # Place the first image to the left\n out[:rows1,:cols1] = np.dstack([img1, img1, img1])\n\n # Place the next image to the right of it\n out[:rows2,cols1:] = np.dstack([img2, img2, img2])\n\n # For each pair of points we have between both images\n # draw circles, then connect a line between them\n for mat in matches:\n\n # Get the matching keypoints for each of the images\n img1_idx = mat.queryIdx\n img2_idx = des_dict[mat.trainIdx][1]\n\n # x - columns\n # y - rows\n (x1,y1) = kp1[img1_idx].pt\n (x2,y2) = kp2[img2_idx].pt\n\n # Draw a small circle at both co-ordinates\n # radius 4\n # colour green\n # thickness = 1\n cv2.circle(out, (int(x1),int(y1)), 4, (0, 255, 0)) \n cv2.circle(out, (int(x2)+cols1,int(y2)), 4, (0, 255, 0))\n\n # Draw a line in between the two points\n # thickness = 1\n # colour green\n cv2.line(out, (int(x1),int(y1)), (int(x2)+cols1,int(y2)), (0, 255, 0), 1)\n\n\n # Show the image\n #cv2.imshow('Matched Features', out)\n #cv2.waitKey(0)\n #cv2.destroyWindow('Matched Features')\n\n # Also return the image if you'd like a copy\n return out","def all_match():\n S1=Spectrum.Spectrum()\n S1.add_peak(50.7,234)\n S1.add_peak(54.6,585)\n S1.add_peak(60.7,773)\n S1.add_peak(65.6,387)\n S1.add_peak(87.7,546)\n S1.add_peak(104.6,598)\n S1.pep_mass=100\n S1.euclidean_scale()\n\n S2=Spectrum.Spectrum()\n S2.add_peak(50.5,234/2)\n S2.add_peak(54.8,585/2)\n S2.add_peak(61.0,773/2)\n S2.add_peak(65.4,387/2)\n S2.add_peak(88.0,546/2)\n S2.add_peak(104.3,598/2)\n S2.pep_mass=100\n S2.euclidean_scale()\n\n score,peaks=similarity.cosine_score_max(S1,S2)\n assert peaks==6, \"Incorrect number of peaks matched with greedy method\"\n assert math.isclose(score,1.0), \"Incorrect score with greedy method\"\n\n score,peaks=similarity.cosine_score_greedy(S1,S2)\n assert peaks==6, \"Incorrect number of peaks matched with maximum weighted method\"\n assert math.isclose(score,1.0), \"Incorrect score with maximum weighted method\"","def ransac(matches, kp1, kp2, sample_points=4, iterations=5, inlier_tolerance=3, inlier_ratio=0.45, check=True, return_max_x=False):\n\n best_inlier_count = 0\n best_h = None\n best_inlier_indices = None\n\n # Get all the corresponing matching pairs for both the images\n pts1 = np.array([kp1[m.queryIdx].pt for m in matches])\n pts2 = np.array([kp2[m.trainIdx].pt for m in matches])\n\n # Re-usable variables for all iterations\n homogeneous_pts1 = np.hstack((pts1, np.ones((pts1.shape[0], 1)))).T\n indices = np.arange(len(pts1))\n num_pts = pts1.shape[0]\n required_inliers = inlier_ratio * num_pts\n\n # For number of iterations\n for _ in range(iterations):\n\n # Sample a small set of points from the point match pairs\n indices_to_sample = np.random.choice(indices, sample_points)\n pts1_sample = pts1[indices_to_sample]\n pts2_sample = pts2[indices_to_sample]\n\n # Get the homography matrix\n h = get_homography_matrix(pts1_sample, pts2_sample)\n\n # Find the new points using the homography matrix\n transformed_points = np.dot(h, homogeneous_pts1).T\n\n # Convert it to world coordinates\n last_col = np.copy(transformed_points[:, -1])\n last_col = last_col[:, np.newaxis]\n transformed_points /= last_col\n transformed_points = transformed_points[:, :-1]\n\n # Find the distance between the actual and the mapped points\n distance = np.linalg.norm(pts2 - transformed_points, axis=1)\n inlier_indices = distance < inlier_tolerance\n inlier_count = inlier_indices.sum()\n\n # Update the best_h if the current h has more inliers\n if inlier_count > best_inlier_count:\n best_h = h\n best_inlier_indices = inlier_indices\n best_inlier_count = inlier_count\n\n # If required inliers is reached break\n if inlier_count > required_inliers:\n break\n\n # Verbose mode - Print the number of inliers\n if check:\n transformed_points = np.dot(best_h, homogeneous_pts1).T\n # Convert it to world coordinates\n last_col = np.copy(transformed_points[:, -1])\n last_col = last_col[:, np.newaxis]\n transformed_points /= last_col\n transformed_points = transformed_points[:, :-1]\n distance = np.linalg.norm(pts2 - transformed_points, axis=1)\n inlier_count = len(distance[distance < inlier_tolerance])\n print('%2.2f of the points are inliers' %\n (inlier_count / num_pts * 100))\n\n # If x coordinates are needed\n if return_max_x:\n max_x_inlier_1 = ceil(pts1[best_inlier_indices].max(axis=0)[0])\n max_x_inlier_2 = ceil(pts2[best_inlier_indices].max(axis=0)[0])\n return best_h, max_x_inlier_1, max_x_inlier_2\n return best_h","def versus(bot1, bot2, num_matches, cpu_count, seed_list):\n W, L, D = 0, 0, 0\n tot_scores = [0, 0]\n avg_scores = [0, 0]\n all_scores = []\n show_match_results = 1\n\n pool = multiprocessing.Pool(cpu_count)\n\n try:\n bot1name = os.path.splitext(os.path.basename(bot1))[0]\n bot2name = os.path.splitext(os.path.basename(bot2))[0]\n start_time = time.time()\n results = [pool.apply_async(run_match, (bot1, bot2, seed_list[i]))\n for i in range(num_matches)]\n \n if show_match_results:\n sys.stdout.write('%20s - %3i: [' % (bot2name, num_matches))\n\n tenths_done = 0\n for r in results:\n scores = r.get(timeout=50)\n if scores[0] > scores[1]:\n W += 1\n elif scores[0] < scores[1]:\n L += 1\n elif scores[0] == scores[1]:\n D += 1\n tot_scores = [tot_scores[x]+scores[x] for x in range(2)]\n avg_scores = [tot_scores[x]//sum([W,L,D]) for x in range(2)]\n all_scores.append(scores[0:2])\n if (sum([W,L,D])*10//num_matches) > tenths_done:\n tenths_done += 1\n if show_match_results:\n sys.stdout.write ('.')\n\n pool.close()\n pool.join()\n \n if show_match_results:\n diff = [s[0] - s[1] for s in all_scores]\n avg, std = meanstd(diff)\n sys.stdout.write (' '*(10-tenths_done))\n print '] - %12s by %8s diff %5.1f std %4.1f in %5.1fs' % (str([W,L,D]), str(avg_scores), avg, std, time.time()-start_time)\n except multiprocessing.TimeoutError:\n if show_match_results:\n sys.stdout.write ('x'*(10-tenths_done))\n print '] - %12s by %8s in %5.1fs' % (str([W,L,D]), str(avg_scores), time.time()-start_time)\n pool.terminate()\n\n return [W,L,D], tot_scores, all_scores","def drawMatches(orbimg1, kp1, orbimg2, kp2, matches):\r\n\r\n # Create a new output image that concatenates the two images together\r\n # (a.k.a) a montage\r\n rows1 = orbimg1.shape[0]\r\n cols1 = orbimg1.shape[1]\r\n rows2 = orbimg2.shape[0]\r\n cols2 = orbimg2.shape[1]\r\n\r\n # Create the output image\r\n # The rows of the output are the largest between the two images\r\n # and the columns are simply the sum of the two together\r\n # The intent is to make this a colour image, so make this 3 channels\r\n out = np.zeros((max([rows1,rows2]),cols1+cols2,3), dtype='uint8')\r\n\r\n # Place the first image to the left\r\n out[:rows1,:cols1] = np.dstack([orbimg1, orbimg1, orbimg1])\r\n\r\n # Place the next image to the right of it\r\n out[:rows2,cols1:] = np.dstack([orbimg2, orbimg2, orbimg2])\r\n\r\n # For each pair of points we have between both images\r\n # draw circles, then connect a line between them\r\n for mat in matches:\r\n\r\n # Get the matching keypoints for each of the images\r\n orbimg1_idx = mat.queryIdx\r\n orbimg2_idx = mat.trainIdx\r\n\r\n # x - columns\r\n # y - rows\r\n (x1,y1) = kp1[orbimg1_idx].pt\r\n (x2,y2) = kp2[orbimg2_idx].pt\r\n\r\n # Draw a small circle at both co-ordinates\r\n # radius 4\r\n # colour blue\r\n # thickness = 1\r\n cv2.circle(out, (int(x1),int(y1)), 4, (255, 0, 0), 1) \r\n cv2.circle(out, (int(x2)+cols1,int(y2)), 4, (255, 0, 0), 1)\r\n\r\n # Draw a line in between the two points\r\n # thickness = 1\r\n # colour blue\r\n cv2.line(out, (int(x1),int(y1)), (int(x2)+cols1,int(y2)), (255,0,0), 1)\r\n\r\n\r\n # Show the image\r\n #cv2.imshow('Matched Features', out)\r\n #cv2.waitKey(0)\r\n #cv2.destroyWindow('Matched Features')\r\n\r\n # Also return the image if you'd like a copy\r\n return out","def calculate_precision_recall_all_images(\n all_prediction_boxes, all_gt_boxes, iou_threshold):\n # Find total true positives, false positives and false negatives\n # over all images\n\n tot_num_tp, tot_num_fp, tot_num_fn = 0, 0, 0\n\n for image in range(len(all_prediction_boxes)):\n prediction_boxes = all_prediction_boxes[image]\n gt_boxes = all_gt_boxes[image]\n\n image_results = calculate_individual_image_result(\n prediction_boxes,\n gt_boxes,\n iou_threshold)\n\n tot_num_tp += image_results[\"true_pos\"]\n tot_num_fp += image_results[\"false_pos\"]\n tot_num_fn += image_results[\"false_neg\"]\n\n # Compute precision, recall\n precision = calculate_precision(tot_num_tp, tot_num_fp, tot_num_fn)\n recall = calculate_recall(tot_num_tp, tot_num_fp, tot_num_fn)\n\n return (precision, recall)","def test_compare_classifier_performance_from_prob_vis_api(experiment_to_use):\n experiment = experiment_to_use\n probability = experiment.probabilities\n viz_outputs = (\"pdf\", \"png\")\n with TemporaryDirectory() as tmpvizdir:\n for viz_output in viz_outputs:\n vis_output_pattern_pdf = tmpvizdir + f\"/*.{viz_output}\"\n visualize.compare_classifiers_performance_from_prob(\n [probability, probability],\n experiment.ground_truth,\n experiment.ground_truth_metadata,\n experiment.output_feature_name,\n top_n_classes=[0],\n labels_limit=0,\n model_namess=[\"Model1\", \"Model2\"],\n output_directory=tmpvizdir,\n file_format=viz_output,\n )\n figure_cnt = glob.glob(vis_output_pattern_pdf)\n assert 1 == len(figure_cnt)","def main():\n\n parser = argparse.ArgumentParser(description='codec_compare')\n parser.add_argument('path', metavar='DIR',\n help='path to images folder')\n args = parser.parse_args()\n classpath = args.path\n classname = classpath.split('/')[1]\n\n images = set(listdir_full_path(classpath))\n if len(images) <= 0:\n print \"\\033[91m[ERROR]\\033[0m\" + \" no source files in ./images.\"\n sys.exit(1)\n\n codeclist_full = set(['aom', 'deepcoder', 'deepcoder-lite', 'fuif', 'fvdo', 'hevc', 'kakadu', 'jpeg',\n 'pik', 'tat', 'xavs', 'xavs-fast', 'xavs-median', 'webp'])\n\n bpp_targets = set([0.06, 0.12, 0.25, 0.50, 0.75, 1.00, 1.50, 2.00])\n for image in images:\n width, height, depth = get_dimensions(image, classname)\n name, imgfmt = os.path.splitext(image)\n imgfmt = os.path.basename(image).split(\".\")[-1]\n derivative_images = []\n if classname[:6] == 'classB':\n derivative_images = create_derivatives(image, classname)\n else:\n derivative_images.append((image, imgfmt))\n\n for derivative_image, pix_fmt in derivative_images:\n json_dir = 'metrics'\n mkdir_p(json_dir)\n json_file = os.path.join(json_dir,\n os.path.splitext(os.path.basename(derivative_image))[0] + \".\" + pix_fmt + \".json\")\n # if os.path.isfile(json_file):\n # print \"\\033[92m[JSON OK]\\033[0m \" + json_file\n # continue\n main_dict = dict()\n derivative_image_metrics = dict()\n for codecname in codeclist_full:\n convertflag = 1\n caseflag = pix_fmt\n if (codecname == 'webp' or codecname == 'tat' or 'deepcoder' in codecname) and depth != '8':\n continue\n if 'xavs' in codecname and depth != '8' and depth != '10':\n continue\n if 'classE' in classname and ('tat' in codecname or 'xavs' in codecname or 'deepcoder' in codecname):\n continue\n if codecname == 'kakadu' and classname[:6] == 'classB':\n convertflag = 0\n caseflag = imgfmt\n bpp_target_metrics = dict()\n for bpp_target in bpp_targets:\n print(codecname)\n if codecname == 'aom' and classname[:6] == 'classB':\n # ('AERIAL2' in image or 'CATS' in image or 'XRAY' in image or 'GOLD' in image or 'TEXTURE1' in image):\n encoded_image_name = os.path.splitext(os.path.basename(derivative_image))[\n 0] + '_' + str(bpp_target) + '_' + imgfmt + '.' + 'av1'\n encoded_image = os.path.join('outputs', codecname, encoded_image_name)\n decoded_image = os.path.join('outputs', codecname, 'decoded', encoded_image_name + '.' + imgfmt)\n original_image = image\n elif codecname == 'kakadu' and classname[:6] == 'classB':\n encoded_image_name = os.path.splitext(os.path.basename(derivative_image))[\n 0] + '_' + str(bpp_target) + '_' + imgfmt + '.' + codecname\n encoded_image = os.path.join('outputs', codecname, encoded_image_name)\n decoded_image = os.path.join('outputs', codecname, 'decoded', encoded_image_name + '.' + imgfmt)\n original_image = image\n elif 'xavs' in codecname and classname[:6] == 'classB':\n encoded_image_name = os.path.splitext(os.path.basename(derivative_image))[\n 0] + '_' + str(bpp_target) + '_' + imgfmt + '.' + codecname\n encoded_image = os.path.join('outputs', codecname, encoded_image_name)\n decoded_image = os.path.join('outputs', codecname, 'decoded', encoded_image_name + '.' + imgfmt)\n original_image = image\n elif codecname == 'fvdo' and classname[:6] == 'classB':\n encoded_image_name = os.path.splitext(os.path.basename(derivative_image))[\n 0] + '_' + str(bpp_target) + '_pgm' + '.' + codecname\n encoded_image = os.path.join('outputs', codecname, encoded_image_name)\n decoded_image = os.path.join('outputs', codecname, 'decoded', encoded_image_name + '.pgm')\n original_image = image\n else:\n if codecname == 'fuif' and 'tif' in imgfmt:\n encoded_image_name = os.path.splitext(os.path.basename(derivative_image))[\n 0] + '.tif_' + str(bpp_target) + '_' + pix_fmt + '.' + codecname\n elif codecname == 'webp' or codecname == 'tat':\n encoded_image_name = os.path.splitext(os.path.basename(derivative_image))[\n 0] + '_' + str(bpp_target) + '_yuv420p.' + codecname\n else:\n encoded_image_name = os.path.splitext(os.path.basename(derivative_image))[\n 0] + '_' + str(bpp_target) + '_' + pix_fmt + '.' + codecname\n encoded_image = os.path.join('outputs', codecname, encoded_image_name)\n decoded_image_path = os.path.join('outputs', codecname, 'decoded')\n decoded_image = ''\n for decodedfile in os.listdir(decoded_image_path):\n encoderoot = '_'.join(os.path.splitext(os.path.basename(encoded_image_name))[0].split('_')[:-1])\n if encoderoot in decodedfile:\n if ('tat' in codecname or 'webp' in codecname) and os.path.splitext(os.path.basename(decodedfile))[1] == '.yuv':\n decoded_image = os.path.join('outputs', codecname, 'decoded', decodedfile)\n print(decoded_image)\n if ('tat' not in codecname or 'webp' not in codecname) and os.path.splitext(os.path.basename(decodedfile))[1] != '.yuv':\n decoded_image = os.path.join('outputs', codecname, 'decoded', decodedfile)\n if 'classE' not in classname and 'classB' not in classname and os.path.isfile(decoded_image):\n decoded_image = convert_decoded(decoded_image, width, height, depth, codecname)\n original_image = convert_decoded(derivative_image, width, height, depth, 'reference')\n else:\n original_image = derivative_image\n\n print('Reference:' + original_image)\n print('Encoded:' + encoded_image)\n print('Decoded:' + decoded_image)\n if (os.path.isfile(original_image) and os.path.isfile(decoded_image) and os.path.isfile(encoded_image)):\n if 'classE' in classname:\n metrics = compute_metrics_HDR(original_image, decoded_image, encoded_image, bpp_target,\n codecname, width, height, pix_fmt, depth)\n\n elif 'classB' in classname:\n metrics = compute_metrics(original_image, decoded_image, encoded_image, bpp_target, codecname,\n width, height, pix_fmt)\n else:\n metrics = compute_metrics_SDR(original_image, decoded_image, encoded_image, bpp_target,\n codecname, width,\n height, imgfmt, depth)\n measured_bpp = (os.path.getsize(encoded_image) * 1.024 * 8) / (float((int(width) * int(height))))\n bpp_target_metrics[measured_bpp] = metrics\n else:\n continue\n \n derivative_image_metrics[codecname] = bpp_target_metrics\n main_dict[derivative_image] = derivative_image_metrics\n\n mkdir_p(json_dir)\n with open(json_file, 'w') as f:\n f.write(json.dumps(main_dict, indent=2))","def run_matching(self):\n paradic = self.cfg['param']['paradic']\n print 'in run_matching() n_bins = ' +str(paradic['n_bins'])\n\n f = open(self.work_dir+'matches.txt','w')\n matching = self.run_proc(['match_cli', 'keys_im0.txt',\n 'keys_im1.txt',\n str(paradic['flag_match']),\n str(paradic['C_match']),\n str(paradic['n_hist']),\n str(paradic['n_ori']),\n str(paradic['n_bins'])],\n stdout=f)\n self.wait_proc(matching, timeout=self.timeout)\n return 1","def perform_match(self, parser_manager):\n log = logging.getLogger(\".\".join([__name__, self.__class__.__name__]))\n log.addFilter(logconfig.ThreadContextFilter())\n extra = self._config_align.all_to_json()\n extra.update(self._config_crystal.all_to_json())\n log = logging.LoggerAdapter(log, extra)\n log.info(\"Matching Started\")\n log.debug(extra)\n\n input_poi = parser_manager.parse_selected_points_from_args()\n beamline_image = parser_manager.get_focused_image()\n parser_manager.save_focused_image(beamline_image)\n focused_image_path = parser_manager.get_focused_image_path()\n formulatrix_image_path = parser_manager.get_formulatrix_image_path()\n job_id = parser_manager.get_job_id()\n\n # Create the images\n image1 = Image.from_file(formulatrix_image_path)\n image2 = beamline_image\n\n # Create results object\n service_result = ServiceResult(job_id, formulatrix_image_path, focused_image_path)\n\n # Perform alignment\n try:\n\n aligned_images, scaled_poi = self._perform_alignment(image1, image2, input_poi)\n service_result.set_image_alignment_results(aligned_images)\n\n # Perform Crystal Matching - only proceed if we have a valid alignment\n if aligned_images.alignment_status_code() == ALIGNED_IMAGE_STATUS_OK:\n match_results = self._perform_matching(aligned_images, scaled_poi, parser_manager)\n\n service_result.append_crystal_matching_results(match_results)\n\n except Exception as e:\n if sys.version_info[0] < 3:\n log.error(\"ERROR: \" + e.message)\n else:\n log.error(\"ERROR: \" + str(e))\n service_result.set_err_state(e)\n\n return service_result","def evaluate(pred_file, ref_file):\n ref_dict, pred_dict, query_dict, id_dict = build_pred_ref_dict(ref_file, pred_file, ref_file)\n total, acc, scores = res_eval_with_type_acc(query_dict, pred_dict, ref_dict, id_dict, save=False)\n em = calculate_exact_match(pred_dict, ref_dict)\n print('Comp Acc: {:.3f}%\\tBleu-4: {:.3f}\\tRouge-L: {:.3f}'.format(acc, scores['Bleu-4'], scores['Rouge-L']))\n print('EM: {:.3f}%'.format(em))\n # calculate_sketch_type_acc(ref_file, pred_file)\n # calculate_exact_match_for_each_q_type(ref_file, pred_file)\n return total, acc, scores, em","def drawMatches(image_1, image_1_keypoints, image_2, image_2_keypoints, matches):\n # Compute number of channels.\n num_channels = 1\n if len(image_1.shape) == 3:\n num_channels = image_1.shape[2]\n # Separation between images.\n margin = 10\n # Create an array that will fit both images (with a margin of 10 to separate\n # the two images)\n joined_image = np.zeros((max(image_1.shape[0], image_2.shape[0]),\n image_1.shape[1] + image_2.shape[1] + margin,\n 3))\n if num_channels == 1:\n for channel_idx in range(3):\n joined_image[:image_1.shape[0],\n :image_1.shape[1],\n channel_idx] = image_1\n joined_image[:image_2.shape[0],\n image_1.shape[1] + margin:,\n channel_idx] = image_2\n else:\n joined_image[:image_1.shape[0], :image_1.shape[1]] = image_1\n joined_image[:image_2.shape[0], image_1.shape[1] + margin:] = image_2\n\n for match in matches:\n image_1_point = (int(image_1_keypoints[match.queryIdx].pt[0]),\n int(image_1_keypoints[match.queryIdx].pt[1]))\n image_2_point = (int(image_2_keypoints[match.trainIdx].pt[0] + \\\n image_1.shape[1] + margin),\n int(image_2_keypoints[match.trainIdx].pt[1]))\n\n cv2.circle(joined_image, image_1_point, 5, (0, 0, 255), thickness = -1)\n cv2.circle(joined_image, image_2_point, 5, (0, 255, 0), thickness = -1)\n cv2.line(joined_image, image_1_point, image_2_point, (255, 0, 0), \\\n thickness = 3)\n return joined_image","def evaluate(self):\n tic = time.time()\n print('Running per image evaluation...')\n p = self.params\n # add backward compatibility if useSegm is specified in params\n if p.useSegm is not None:\n p.iouType = 'segm' if p.useSegm == 1 else 'bbox'\n print(\n \"useSegm (deprecated) is not None.\"\n \"Running {} evaluation\".format(p.iouType)\n )\n print('Evaluate annotation type *{}*'.format(p.iouType))\n p.imgIds = list(np.unique(p.imgIds))\n if p.useCats:\n p.catIds = list(np.unique(p.catIds))\n p.maxDets = sorted(p.maxDets)\n self.params = p\n\n self._prepare()\n # loop through images, area range, max detection number\n catIds = p.catIds if p.useCats else [-1]\n\n self.ious = {\n (imgId, catId): self.computeIoU(imgId, catId)\n for imgId in p.imgIds\n for catId in catIds\n }\n # ignore any categories that is not having any attributes\n self.f1s = {\n (imgId, catId): self.computeF1(imgId, catId)\n for imgId in p.imgIds\n for catId in catIds\n if catId in self.FPParams.catsWithAttributes\n }\n\n # self.gt_attributes_ids = []\n\n # loop through images, area range, max detection number\n self.evalImgs = [\n self.evaluateImg(imgId, catId, areaRng, p.maxDets[-1])\n for catId in catIds\n for areaRng in p.areaRng\n for imgId in p.imgIds\n ]\n # self._paramsEval = copy.deepcopy(self.params) # seems do not need it\n toc = time.time()\n print('DONE (t={:0.2f}s).'.format(toc-tic))","def evaluation(eval_dict, faster_rcnn, test_num=Config.eval_num):\n bboxes, labels, scores = list(), list(), list()\n gt_bboxes, gt_labels, gt_difficult = list(), list(), list()\n for i, [img_dir, img_info] in tqdm(enumerate(eval_dict.items())):\n img, img_info = rescale_image(img_dir, img_info, flip=False)\n img_tensor = create_img_tensor(img)\n box, score, label = faster_rcnn.predict(img_tensor)\n gt_bbox = np.array(img_info['objects'])[:, 1:5].astype(np.float32)\n gt_label = np.array(img_info['objects'])[:, 0]\n difficult = np.array(img_info['difficult'])\n\n # label from text to number\n gt_label = text_to_num(gt_label)\n\n bboxes.append(box)\n labels.append(label)\n scores.append(score)\n gt_bboxes.append(gt_bbox)\n gt_labels.append(gt_label)\n gt_difficult.append(difficult)\n if i == test_num:\n break\n\n AP, mAP = calc_map(bboxes, labels, scores, gt_bboxes, gt_labels, gt_difficult, use_07_metric=True)\n\n return mAP","def evaluate(self, test_data):\n test_results = [(self.decide(self.feedforward(x)), y)\n for (x, y) in test_data]\n\n # Check for and keep track of TP's FP's and FN's\n # Write FP's and FN's to a special directories\n TP = FP = FN = count = 0\n TN = 0\n # what you got = x, what should be = y\n for ((x, y), (image, gt)) in zip(test_results, test_data):\n count += 1\n if x == 1 and y == 1:\n TP += 1\n elif x == 1 and y == 0:\n FP += 1\n if not self.validating: self.save_Image(FP_PATH, count, image)\n elif x == 0 and y == 1:\n FN += 1\n if not self.validating: self.save_Image(FN_PATH, count, image)\n \n if (TP + FP ) == 0:\n return TP, FP, FN, '---------', '---------','---------'\n else:\n false_rate = float(FP) / float(TP + FP)\n\n if (TP + FN ) == 0:\n return TP, FP, FN, '---------', '---------', false_rate\n else:\n detect_rate = float(TP) / float(TP + FN)\n\n if (TP + FP + FN ) == 0:\n return TP, FP, FN, '---------', detect_rate, false_rate\n else:\n quality_rate = float(TP) / float(TP+FP+FN)\n \n return TP, FP, FN, quality_rate, detect_rate, false_rate","def test_compare_classifiers_predictions_vis_api(experiment_to_use):\n experiment = experiment_to_use\n predictions = experiment.predictions\n viz_outputs = (\"pdf\", \"png\")\n with TemporaryDirectory() as tmpvizdir:\n for viz_output in viz_outputs:\n vis_output_pattern_pdf = tmpvizdir + f\"/*.{viz_output}\"\n visualize.compare_classifiers_predictions(\n [predictions, predictions],\n experiment.ground_truth,\n experiment.ground_truth_metadata,\n experiment.output_feature_name,\n labels_limit=0,\n model_names=[\"Model1\", \"Model2\"],\n output_directory=tmpvizdir,\n file_format=viz_output,\n )\n figure_cnt = glob.glob(vis_output_pattern_pdf)\n assert 1 == len(figure_cnt)","def test_compare_single_and_multiprocess_results(sidesweep_image_sequence):\n cc = Cwsim_container_from_ims(ims=sidesweep_image_sequence)\n test_idx = 20\n results_single_proc = cc.self_im_query_all(test_idx, plot_output=False)\n\n test_im = sidesweep_image_sequence[test_idx]\n cc.prepare_memory_bank_outside()\n results_multi_proc = cc.query_image_mp(test_im)\n\n plt.plot(results_single_proc, marker='D', label='single_proc')\n plt.plot(results_multi_proc, marker='D', label='multi_proc')\n plt.legend()\n plt.draw()\n plt.show(block=False)\n plt.pause(3)\n\n assert np.allclose(results_single_proc, results_multi_proc)","def match(desc1,desc2):\n desc1 = array([d/linalg.norm(d) for d in desc1])\n desc2 = array([d/linalg.norm(d) for d in desc2])\n dist_ratio = 0.6\n desc1_size = desc1.shape\n matchscores = zeros((desc1_size[0],1),'int')\n desc2t = desc2.T # precompute matrix transpose\n for i in range(desc1_size[0]):\n dotprods = dot(desc1[i, :], desc2t) # vector of dot products\n dotprods *= 0.9999\n # inverse cosine and sort, return index for features in second image\n indx = argsort(arccos(dotprods))\n # check if nearest neighbor has angle less than dist_ratio times 2nd\n if arccos(dotprods)[indx[0]] < dist_ratio * arccos(dotprods)[indx[1]]:\n matchscores[i] = int(indx[0])\n return matchscores","def compute_matches(self):\n\t\tself.local_database[\"figure_number\"] = [0] * len(self.local_database[\"feature_vectors\"])\n\t\tForensics = wbForensicsHOG(Database=self.local_database)\n\t\tForensics.KDTree_pairs(leaf_size = len(self.local_database)+1)\n\t\tForensics.d_rank(pairs=Forensics.pairs, distances=Forensics.dists, ratios=Forensics.ratios)\n\n\t\tself.local_matches = Forensics.Dist_Rank","def test(label_dict, image_ids, args):\n\n log_header('Starting image retrieval preparations')\n\n queries = []\n\n if args.test_image is not None:\n queries.append(args.test_image)\n else:\n logging.info('Generating random test queries')\n\n number_of_images = len(image_ids)\n\n # Generate random queries to use in the test procedure\n for i in range(args.k):\n queries.append(image_ids[random.randint(0, number_of_images - 1)])\n\n # Calculate score for the retrieved images\n calculate_score(\n label_dict, queries, retrieve_similar_images(queries, args)\n )","def _compute_prediction(self, k, img, kp, des):\n ###taken from example code:\n\n #find matches\n matches = self.matcher.knnMatch(des,self.descs[k],k=2)\n\n # make sure the distance to the closest match is sufficiently better than the second closest\n good_matches = []\n for m,n in matches: \n if (m.distance < 0.7*n.distance and\n kp[m.queryIdx].response > 0.0 and\n self.kps[k][m.trainIdx].response > 0.0):\n good_matches.append((m.queryIdx, m.trainIdx))\n \n if len(good_matches) < 4:\n visual_diff = None\n else:\n #create pt lists\n img_pts = np.zeros((len(good_matches),2))\n tem_pts = np.zeros((len(good_matches),2))\n for idx in range(len(good_matches)):\n match = good_matches[idx]\n img_pts[idx,:] = kp[match[0]].pt\n tem_pts[idx,:] = self.kps[k][match[1]].pt\n \n # Transform input image so that it matches the template image as well as possible\n M, mask = cv2.findHomography(img_pts, tem_pts, cv2.RANSAC, 8.0)\n img_T = cv2.warpPerspective(img, M, self.signs[k].shape[::-1])\n \t\n #calculate error\n visual_diff = compare_images(img_T, self.signs[k])\n return visual_diff\n #not functioning properly (currently does not return highest value for identical image/scene pair)","def my_evaluate(original_image, img_id, annotations, min_score=0.4, max_overlap=0.3, top_k=200):\n\n # Transform\n image = normalize(to_tensor(resize(original_image)))\n\n # Move to default device\n image = image.to(device)\n\n # Forward prop.\n predicted_locs, predicted_scores = model(image.unsqueeze(0))\n\n # Detect objects in SSD output\n det_boxes, det_labels, det_scores = model.detect_objects(predicted_locs, predicted_scores, min_score=min_score,\n max_overlap=max_overlap, top_k=top_k)\n\n # Move detections to the CPU\n det_boxes = det_boxes[0].to('cpu')\n\n # Transform to original image dimensions\n original_dims = torch.FloatTensor(\n [original_image.width, original_image.height, original_image.width, original_image.height]).unsqueeze(0)\n det_boxes = det_boxes * original_dims\n\n # Decode class integer labels\n det_labels = [rev_label_map[l] for l in det_labels[0].to('cpu').tolist()]\n det_scores = det_scores[0].to('cpu').tolist()\n\n if det_labels != ['background']:\n for i in range(det_boxes.size(0)):\n \n # For every detection, see whether it matches a ground truth\n match = False\n for j in range(len(annotations['boxes'])):\n if annotations['labels'][j] == -1: # this is being set when a ground truth already matched\n continue\n match = box_match(det_boxes[i], \n det_labels[i], \n torch.Tensor(annotations['boxes'][j]), \n rev_label_map[annotations['labels'][j]])\n if match:\n annotations['labels'][j] = -1\n break\n \n if match: # true positive if the detection is correct and matched a ground truth\n all_results['true_positive'] += 1\n else: # false positive if the detection did not match a ground truth\n all_results['false_positive'] += 1\n \n # After all detections were checked, let us see whether the detector missed something\n for label in annotations['labels']: \n if label == -1: # This is set after a detection matched this ground truth\n continue\n \n # false negative if we reach this line, since the ground truth object was not found\n all_results['false_negative'] += 1","def test_run_multi_r__(self):\n\n # Test Description\n # ================\n #\n # 1. This test intialises an example *eopy.matchup.matchupIO.MatchUp* object\n #\n # 2. Compare transformed dataset to expected value\n\n ################################################################################################################\n # 1. Initialise Test Data Object\n ################################################################################################################\n\n MatchUpTest = return_MatchUpTest_r__()\n\n ################################################################################################################\n # 2. Define expected values\n ################################################################################################################\n\n # Original dataset values (should be unchanged)\n MatchUpOriginal_expected = return_MatchUpTest_r__()\n\n # Transformed dataset\n values_expected = array([294.0625, 480.3733333, 300.6, 227.3846154, 210.1533333,\n 22.74193548, 22.0625, 21.96875, 22.80645161, 23.5,\n 21.66666667, 21.05882353, 23, 22.40625,\n 38.33333333, 36.63636364, 36.5, 38.42857143,\n 30.1, 32.14893617, 29.37254902, 28.88461538, 28.56603774,\n 33.45238095, 32.81395349, 31.77272727, 32.60465116,\n 40.125, 43.54054054, 38.59090909, 34.08510638,\n 13.72727273, 12, 14.1, 11.79069767, 17.53846154,\n 12.69565217, 31.16666667, 12.26086957, 11.52272727,\n 8.8125, 12, 7.4, 10.13207547])\n unc_expected = [Uncertainty(1, array([1.6, 1.5, 1.5, 1.3, 1.5])),\n Uncertainty(1, array([3.1, 3.2, 3.2, 3.1, 3.0])),\n Uncertainty(1, array([3.3, 3.4, 3.1, 3.2])),\n Uncertainty(1, array([2.1, 2.2, 2.2, 2.1])),\n Uncertainty(1, array([5.0, 4.7, 5.1, 5.2, 5.3])),\n Uncertainty(1, array([4.2, 4.3, 4.4, 4.3])),\n Uncertainty(1, array([4.0, 3.7, 4.4, 4.7])),\n Uncertainty(1, array([2.2, 1.7, 2.0, 4.3, 2.6])),\n Uncertainty(1, array([2.3, 1.2, 2.3, 4.4])),\n Uncertainty(1, array([3.2, 2.7, 3.0, 5.3]))]\n ks_expected = array([4.8, 6.8, 5.2, 5.6, 5.2, 12.10287443, 13.99394856, 12.48108926, 12.85930408])\n unck_expected = [Uncertainty(1, array([0.25, 0.25, 0.25, 0.25, 0.25])),\n Uncertainty(1, array([0.2644, 0.2644, 0.2644, 0.2644]))]\n idx_expected = {\"Nm\": [5, 4],\n \"cNm\": [0, 5, 9],\n \"Im\": [[0, 1], [1, 2]],\n \"sensors\": [-1, 1, 2],\n \"sensor_ms\": [1, 3, 3],\n \"n_sensor\": [0, 1, 1, 2, 1, 1, 2, 1, 1, 2],\n \"n_mu\": [1, 1, 2, 2, 1, 2, 2, 1, 2, 2],\n \"n_cov\": [1, 1, 1, 1, 2, 2, 2, 3, 3, 3],\n \"N_var\": [5, 5, 4, 4, 5, 4, 4, 5, 4, 4],\n \"idx\": [0, 5, 10, 14, 18, 23, 27, 31, 36, 40, 44],\n \"Ia\": [1, 1, 1, 2, 2, 2]}\n a_expected = array([1., 1.3, 0.002, 0.5, 1.1, 0.0005])\n w_matrices_expected = []\n u_matrices_expected = []\n\n ################################################################################################################\n # 3. Run Transform2NormInd.run()\n ################################################################################################################\n\n Transform2NormIndOp = Transform2NormInd()\n MatchUpTransform = Transform2NormIndOp.run(MatchUpTest)\n\n values_test = MatchUpTransform.values\n unc_test = MatchUpTransform.unc\n w_matrices_test = MatchUpTransform.w_matrices\n u_matrices_test = MatchUpTransform.u_matrices\n ks_test = MatchUpTransform.ks\n unck_test = MatchUpTransform.unck\n idx_test = MatchUpTransform.idx\n\n ################################################################################################################\n # 4. Compare retrieve values to expect values\n ################################################################################################################\n\n # Test transformed data object attribute by attribute\n\n # a. values\n for i, (value_expected, value_test) in enumerate(zip(values_expected, values_test)):\n self.assertAlmostEqual(value_expected, value_test, places=5, msg=str(i))\n\n # b. unc\n for block_unc_test, block_unc_expected in zip(unc_test, unc_expected):\n self.assertEqual(block_unc_expected.typeID, block_unc_test.typeID)\n self.assertEqual(block_unc_expected.uR.tolist(), block_unc_test.uR.tolist())\n\n # c. w_matrices\n self.assertEqual(w_matrices_test, w_matrices_expected)\n\n # d. u_matrices\n self.assertEqual(u_matrices_test, u_matrices_expected)\n\n # e. ks\n for k_expected, k_test in zip(ks_expected, ks_test):\n self.assertAlmostEqual(k_test.tolist(), k_expected.tolist(), places=5)\n\n # f. unck\n for block_unck_test, block_unck_expected in zip(unck_test, unck_expected):\n self.assertEqual(block_unck_expected.typeID, block_unck_test.typeID)\n self.assertEqual(block_unck_expected.uR.tolist(), block_unck_test.uR.tolist())\n\n # h. idx\n self.assertEqual(set(idx_expected.keys()), set(idx_test.keys()))\n for key in idx_expected.keys():\n idx_i_test = idx_test[key]\n idx_i_expected = idx_expected[key]\n if isinstance(idx_i_expected, ndarray):\n self.assertEqual(idx_i_test.tolist(), idx_i_expected.tolist())\n else:\n self.assertEqual(idx_i_test, idx_i_expected)\n\n # Test original data object preserved attribute by attribute\n\n # a. values\n for i, (value_original_expected, value_original_test) in enumerate(zip(MatchUpOriginal_expected.values, MatchUpTest.values)):\n self.assertAlmostEqual(value_original_expected, value_original_test, places=5)\n\n # b. unc\n for block_unc_original_expected, block_unc_original_test in zip(MatchUpOriginal_expected.unc, MatchUpTest.unc):\n self.assertEqual(block_unc_original_expected.typeID, block_unc_original_test.typeID)\n self.assertEqual(block_unc_original_expected.uR.tolist(), block_unc_original_test.uR.tolist())\n\n # c. w_matrices\n self.assertEqual(MatchUpOriginal_expected.w_matrices, MatchUpTest.w_matrices)\n\n # d. u_matrices\n self.assertEqual(MatchUpOriginal_expected.u_matrices, MatchUpTest.u_matrices)\n\n # e. ks\n for k_original_expected, k_original_test in zip(MatchUpOriginal_expected.ks, MatchUpTest.ks):\n self.assertAlmostEqual(k_original_test.tolist(), k_original_expected.tolist(), places=5)\n\n # f. unck\n for block_unck_original_expected, block_unck_original_test in zip(MatchUpOriginal_expected.unck, MatchUpTest.unck):\n self.assertEqual(block_unck_original_expected.typeID, block_unck_original_test.typeID)\n self.assertEqual(block_unck_original_expected.uR.tolist(), block_unck_original_test.uR.tolist())\n\n # h. idx\n self.assertEqual(set(MatchUpOriginal_expected.idx), set(MatchUpTest.idx))\n for key in MatchUpOriginal_expected.idx.keys():\n idx_i_original_test = MatchUpTest.idx[key]\n idx_i_original_expected = MatchUpOriginal_expected.idx[key]\n if isinstance(idx_i_original_expected, ndarray):\n self.assertEqual(idx_i_original_test.tolist(), idx_i_original_expected.tolist())\n else:\n self.assertEqual(idx_i_original_test, idx_i_original_expected)","def _calculate_similarity(self):\n self._logger.info(\"Calculating the similarity between images.\")\n\n # Create a helper function to simplify the loops below:\n def get_wrapper(fdx, mdx):\n wrapper = pos_wrappers.image_similarity_wrapper(\n reference_image=self.f['src_gray'](idx=fdx),\n moving_image=self.f['src_gray'](idx=mdx),\n affine_transformation=self.f['part_transf'](mIdx=mdx, fIdx=fdx))\n return copy.copy(wrapper)\n\n commands = [] # Will hold commands for calculating the similarity\n\n # Will hold (moving, fixed) images partial_transforms basically: all\n # partial transformations array\n partial_transforms = []\n\n self._logger.debug(\"Generating similarity measure warppers.\")\n for moving_slice in self.options.slice_range:\n # Get all fixed images to which given moving slice will be aligned:\n tpair = list(flatten(self._get_slice_pair(moving_slice)))\n\n # Append partial transformations for given moving slice to the\n # global partial transformations array\n partial_transforms.append(tpair)\n\n # Generate wrapper for measuring similarity for a given partial\n # transformation.\n for mdx, fdx in tpair:\n commands.append(get_wrapper(fdx, mdx))\n\n # Execute and commands and workflow the similarity measurements.\n stdout, stderr = self.execute(commands)\n simmilarity = map(lambda x: float(x.strip()),\n stdout.strip().split(\"\\n\"))\n simmilarity = dict(zip(flatten(partial_transforms), simmilarity))\n\n self._logger.debug(\"Generating graph edges.\")\n graph_connections = []\n\n # Lambda defines slice skipping is preffered (lower l), or reluctant\n # to slice skipping (higher)\n l = self.options.graphEdgeLambda\n\n for (mdx, fdx), s in simmilarity.iteritems():\n w = (1.0 + s) * abs(mdx - fdx) * (1.0 + l) ** (abs(mdx - fdx))\n graph_connections.append((fdx, mdx, w))\n\n self._logger.info(\"Creating a graph based on image similarities.\")\n # Generate the graph basen on the weight of the edges\n self.G = nx.DiGraph()\n self.G.add_weighted_edges_from(graph_connections)\n\n self._logger.debug(\"Saving the graph to a file.\")\n # Save the edges for some further analysis.\n nx.write_weighted_edgelist(self.G,\n self.f['graph_edges'](sign=self.signature))\n\n # Also, save the individual similarity metrics:\n simm_fh = open(self.f['similarity'](sign=self.signature), 'w')\n for (mdx, fdx), s in sorted(simmilarity.iteritems()):\n simm_fh.write(\"%d %d %f\\n\" % (mdx, fdx, s))\n simm_fh.close()","def compute_matches(gt_boxes, gt_class_ids, gt_masks,\n pred_boxes, pred_class_ids, pred_scores, pred_masks,\n iou_threshold=0.5, score_threshold=0.0):\n # Trim zero padding\n # TODO: cleaner to do zero unpadding upstream\n gt_boxes = trim_zeros(gt_boxes)\n gt_masks = gt_masks[..., :gt_boxes.shape[0]]\n pred_boxes = trim_zeros(pred_boxes)\n pred_scores = pred_scores[:pred_boxes.shape[0]]\n # Sort predictions by score from high to low\n indices = np.argsort(pred_scores)[::-1]\n pred_boxes = pred_boxes[indices]\n pred_class_ids = pred_class_ids[indices]\n pred_scores = pred_scores[indices]\n pred_masks = pred_masks[..., indices]\n\n # Compute IoU overlaps [pred_masks, gt_masks]\n overlaps = compute_overlaps_masks(pred_masks, gt_masks)\n\n # Loop through predictions and find matching ground truth boxes\n match_count = 0\n pred_match = -1 * np.ones([pred_boxes.shape[0]])\n gt_match = -1 * np.ones([gt_boxes.shape[0]])\n for i in range(len(pred_boxes)):\n # Find best matching ground truth box\n # 1. Sort matches by score\n sorted_ixs = np.argsort(overlaps[i])[::-1]\n # 2. Remove low scores\n low_score_idx = np.where(overlaps[i, sorted_ixs] < score_threshold)[0]\n if low_score_idx.size > 0:\n sorted_ixs = sorted_ixs[:low_score_idx[0]]\n # 3. Find the match\n for j in sorted_ixs:\n # If ground truth box is already matched, go to next one\n if gt_match[j] > -1:\n continue\n # If we reach IoU smaller than the threshold, end the loop\n iou = overlaps[i, j]\n if iou < iou_threshold:\n break\n # Do we have a match?\n if pred_class_ids[i] == gt_class_ids[j]:\n match_count += 1\n gt_match[j] = i\n pred_match[i] = j\n break\n\n return gt_match, pred_match, overlaps","def apply_feature_matching(desc1: np.ndarray, desc2: np.ndarray,\n match_calculator: Callable[[list, list], float]) -> list:\n\n # Check descriptors dimensions are 2\n assert desc1.ndim == 2, \"Descriptor 1 shape is not 2\"\n assert desc2.ndim == 2, \"Descriptor 2 shape is not 2\"\n\n # Check that the two features have the same descriptor type\n assert desc1.shape[1] == desc2.shape[1], \"Descriptors shapes are not equal\"\n\n # If there is not key points in any of the images\n if desc1.shape[0] == 0 or desc2.shape[0] == 0:\n return []\n\n # number of key points in each image\n num_key_points1 = desc1.shape[0]\n num_key_points2 = desc2.shape[0]\n\n # List to store the matches scores\n matches = []\n\n # Loop over each key point in image1\n # We need to calculate similarity with each key point in image2\n for kp1 in range(num_key_points1):\n # Initial variables which will be updated in the loop\n distance = -np.inf\n y_index = -1\n\n # Loop over each key point in image2\n for kp2 in range(num_key_points2):\n\n # Match features between the 2 vectors\n value = match_calculator(desc1[kp1], desc2[kp2])\n\n # SSD values examples: (50, 200, 70), we need to minimize SSD (Choose 50)\n # In case of SSD matching: (value is returned as a \"negative\" number) (-50, -200, -70)\n # So we compare it with -np.inf. (The sorting will be reversed later)\n\n # NCC values examples: (0.58, 0.87, 0.42), we need to maximize NCC (Choose 0.87)\n # In case of NCC matching: (value is returned as a \"positive\" number) (-58, -0.87, -0.42)\n # So we compare it with -np.inf. (The sorting will be reversed later)\n\n if value > distance:\n distance = value\n y_index = kp2\n\n # Create a cv2.DMatch object for each match and set attributes:\n # queryIdx: The index of the feature in the first image\n # trainIdx: The index of the feature in the second image\n # distance: The distance between the two features\n cur = cv2.DMatch()\n cur.queryIdx = kp1\n cur.trainIdx = y_index\n cur.distance = distance\n matches.append(cur)\n\n return matches","def find_matching_points(img1, img2, max_pix_movement=50, normalize=True, show=False):\n\n # Initiate ORB detector\n orb = cv2.ORB_create()\n\n # find the keypoints and descriptors with ORB\n kp1, des1 = orb.detectAndCompute(img1, None)\n kp2, des2 = orb.detectAndCompute(img2, None)\n\n # create BFMatcher object\n bf = cv2.BFMatcher(cv2.NORM_HAMMING, crossCheck=True)\n # Match descriptors.\n matches = bf.match(des1,des2)\n # Sort them in the order of their distance.\n matches = sorted(matches, key = lambda x:x.distance)\n # Draw first 10 matches.\n if show:\n img3 = cv2.drawMatches(img1,kp1,img2,kp2,matches[:500], None,flags=2)\n plt.imshow(img3),plt.show()\n # Get the matching keypoints for each of the images\n\n list_kp1 = []\n list_kp2 = []\n for mat in matches:\n img1_idx = mat.queryIdx\n img2_idx = mat.trainIdx\n\n # x - columns\n # y - rows\n list_kp1.append(kp1[img1_idx].pt)\n list_kp2.append(kp2[img2_idx].pt)\n\n n_kp1, n_kp2 = np.float32(list_kp1), np.float32(list_kp2)\n n_kp1 /= np.asarray([img1.shape[1], img1.shape[0]], np.float32)\n n_kp2 /= np.asarray([img2.shape[1], img2.shape[0]], np.float32)\n n_kp1 = n_kp1 * 2. - 1.\n n_kp2 = n_kp2 * 2. - 1.\n\n return np.int32(list_kp1), np.int32(list_kp2), n_kp1, n_kp2","def ratio_test(fx2_match, fx1_match, fx1_norm, match_dist, norm_dist,\n ratio_thresh=.625, fm_dtype=np.int32, fs_dtype=np.float32):\n fx2_to_ratio = np.divide(match_dist, norm_dist).astype(fs_dtype)\n fx2_to_isvalid = np.less(fx2_to_ratio, ratio_thresh)\n fx2_match_RAT = fx2_match.compress(fx2_to_isvalid).astype(fm_dtype)\n fx1_match_RAT = fx1_match.compress(fx2_to_isvalid).astype(fm_dtype)\n fx1_norm_RAT = fx1_norm.compress(fx2_to_isvalid).astype(fm_dtype)\n # Turn the ratio into a score\n fs_RAT = np.subtract(1.0, fx2_to_ratio.compress(fx2_to_isvalid))\n fm_RAT = np.vstack((fx1_match_RAT, fx2_match_RAT)).T\n # return normalizer info as well\n fm_norm_RAT = np.vstack((fx1_norm_RAT, fx2_match_RAT)).T\n ratio_tup = MatchTup3(fm_RAT, fs_RAT, fm_norm_RAT)\n return ratio_tup","def return_MatchUpTest_r__():\n\n ####################################################################################################################\n # 1. Initialise test data\n ####################################################################################################################\n\n values = array([470.5, 720.56, 450.9, 295.6, 315.23,\n 70.5, 70.6, 70.3, 70.7, 70.5,\n 71.5, 71.6, 71.3, 71.7,\n 80.5, 80.6, 80.3, 80.7,\n 150.5, 151.1, 149.8, 150.2, 151.4,\n 140.5, 141.1, 139.8, 140.2,\n 160.5, 161.1, 169.8, 160.2,\n 30.2, 20.4, 28.2, 50.7, 45.6,\n 29.2, 37.4, 28.2, 50.7,\n 28.2, 32.4, 22.2, 53.7, ])\n unc = [Uncertainty(1, array([1.6, 1.5, 1.5, 1.3, 1.5])),\n Uncertainty(1, array([3.1, 3.2, 3.2, 3.1, 3.0])),\n Uncertainty(1, array([3.3, 3.4, 3.1, 3.2])),\n Uncertainty(1, array([2.1, 2.2, 2.2, 2.1])),\n Uncertainty(1, array([5.0, 4.7, 5.1, 5.2, 5.3])),\n Uncertainty(1, array([4.2, 4.3, 4.4, 4.3])),\n Uncertainty(1, array([4.0, 3.7, 4.4, 4.7])),\n Uncertainty(1, array([2.2, 1.7, 2.0, 4.3, 2.6])),\n Uncertainty(1, array([2.3, 1.2, 2.3, 4.4])),\n Uncertainty(1, array([3.2, 2.7, 3.0, 5.3]))]\n ks = array([1.2, 1.7, 1.3, 1.4, 1.3, 3.2, 3.7, 3.3, 3.4])\n unck = [Uncertainty(1, array([0.25, 0.25, 0.25, 0.25, 0.25])),\n Uncertainty(1, array([0.2644, 0.2644, 0.2644, 0.2644]))]\n idx = {\"Nm\": [5, 4],\n \"cNm\": [0, 5, 9],\n \"Im\": [[0, 1], [1, 2]],\n \"sensors\": [-1, 1, 2],\n \"sensor_ms\": [1, 3, 3],\n \"n_sensor\": [0, 1, 1, 2, 1, 1, 2, 1, 1, 2],\n \"n_mu\": [1, 1, 2, 2, 1, 2, 2, 1, 2, 2],\n \"n_cov\": [1, 1, 1, 1, 2, 2, 2, 3, 3, 3],\n \"N_var\": [5, 5, 4, 4, 5, 4, 4, 5, 4, 4],\n \"idx\": [0, 5, 10, 14, 18, 23, 27, 31, 36, 40, 44],\n \"Ia\": [1, 1, 1, 2, 2, 2]}\n a = array([1., 1.3, 0.002, 0.5, 1.1, 0.0005])\n\n ####################################################################################################################\n # 3. Initialise MatchUp object\n ####################################################################################################################\n\n MatchUpTest = MatchUp()\n MatchUpTest.values = values\n MatchUpTest.unc = unc\n MatchUpTest.ks = ks\n MatchUpTest.unck = unck\n MatchUpTest.idx = idx\n MatchUpTest.a = a\n\n return MatchUpTest","def drawMatches(img1, kp1, img2, kp2, matches):\n\n # Create a new output image that concatenates the two images together\n # (a.k.a) a montage\n rows1 = img1.shape[0]\n cols1 = img1.shape[1]\n rows2 = img2.shape[0]\n cols2 = img2.shape[1]\n\n out = np.zeros((max([rows1,rows2]),cols1+cols2,3), dtype='uint8')\n\n # Place the first image to the left\n out[:rows1,:cols1] = np.dstack([img1, img1, img1])\n\n # Place the next image to the right of it\n out[:rows2,cols1:] = np.dstack([img2, img2, img2])\n\n # For each pair of points we have between both images\n # draw circles, then connect a line between them\n for mat in matches:\n\n # Get the matching keypoints for each of the images\n img1_idx = mat.queryIdx\n img2_idx = mat.trainIdx\n\n # x - columns\n # y - rows\n (x1,y1) = kp1[img1_idx].pt\n (x2,y2) = kp2[img2_idx].pt\n\n # Draw a small circle at both co-ordinates\n # radius 4\n # colour blue\n # thickness = 1\n cv2.circle(out, (int(x1),int(y1)), 4, (255, 0, 0), 1) \n cv2.circle(out, (int(x2)+cols1,int(y2)), 4, (255, 0, 0), 1)\n\n # Draw a line in between the two points\n # thickness = 1\n # colour blue\n cv2.line(out, (int(x1),int(y1)), (int(x2)+cols1,int(y2)), (255, 0, 0), 1)\n\n\n # Show the image\n\n # Also return the image if you'd like a copy\n return out","def plot_images(img_A, img_B, match_A, match_B, name=\"p1\", method=\"greedy\"):\n (hA, wA) = img_A.shape[:2]\n (hB, wB) = img_B.shape[:2]\n vis = np.zeros((max(hA, hB), wA + wB, 3), dtype=\"uint8\")\n vis[0:hA, 0:wA] = img_A\n vis[0:hB, wA:] = img_B\n\n # loop over the matches\n for ptA, ptB in zip(match_A, match_B):\n ptB = ptB + np.array([wA, 0])\n cv2.line(vis, tuple(ptA), tuple(ptB), (0, 255, 0), 1)\n\n cv2.imwrite(\"output/sift_\" + method + \"_\" + name + \".png\", vis)\n cv2.imshow(\"Keypoint matching\", vis)\n cv2.waitKey(0)","def ransac(keypoints1, keypoints2, matches, sampling_ratio=0.5, n_iters=500, threshold=20):\n N = matches.shape[0]\n n_samples = int(N * sampling_ratio)\n\n # Please note that coordinates are in the format (y, x)\n matched1 = pad(keypoints1[matches[:,0]])\n matched2 = pad(keypoints2[matches[:,1]])\n matched1_unpad = keypoints1[matches[:,0]]\n matched2_unpad = keypoints2[matches[:,1]]\n\n max_inliers = np.zeros(N)\n n_inliers = 0\n\n # RANSAC iteration start\n ### YOUR CODE HERE\n raise NotImplementedError() # Delete this line\n ### END YOUR CODE\n return H, matches[max_inliers]","def match(desc1,desc2):\n\t\n\tdesc1 = array([d/linalg.norm(d) for d in desc1])\n\tdesc2 = array([d/linalg.norm(d) for d in desc2])\n\t\n\tdist_ratio = 0.6\n\tdesc1_size = desc1.shape\n\t\n\tmatchscores = zeros((desc1_size[0],1))\n\tdesc2t = desc2.T #precompute matrix transpose\n\tfor i in range(desc1_size[0]):\n\t\tdotprods = dot(desc1[i,:],desc2t) #vector of dot products\n\t\tdotprods = 0.9999*dotprods\n\t\t#inverse cosine and sort, return index for features in second image\n\t\tindx = argsort(arccos(dotprods))\n\t\t\n\t\t#check if nearest neighbor has angle less than dist_ratio times 2nd\n#\t\tif arccos(dotprods)[indx[0]] < dist_ratio * arccos(dotprods)[indx[1]]:\n\t\tmatchscores[i] = int(indx[0])\n\t\n\treturn matchscores","def compute_transform(self, image_pair, figure):\t\n\t\timage_pair = [int(image_pair[1]), int(image_pair[3])]\n\t\tindices = np.where(np.all(np.abs(self.Matches[\"image_pairs\"][0] - image_pair) == 0, axis=1)) \n\t\tfeature_pairs = self.Matches[\"image_match_pairs\"][0][indices[0][0]]\n\t\tfeature_pair_scores = self.Matches[\"image_feature_scores\"][0][indices[0][0]] \n\n\t\ti1_loc = self.Database[\"image_locs\"][np.where(np.array(self.Database[\"image_idx\"]) == image_pair[0])[0][0]]\n\t\ti2_loc = self.Database[\"image_locs\"][np.where(np.array(self.Database[\"image_idx\"]) == image_pair[1])[0][0]]\n\n\t\tself.image1 = cv2.cvtColor(extract_image(figure, i1_loc), cv2.COLOR_BGR2GRAY)\n\t\tself.image2 = cv2.cvtColor(extract_image(figure, i2_loc), cv2.COLOR_BGR2GRAY)\n\t\t# self.image1 = extract_image(figure, i1_loc)\n\t\t# self.image2 = extract_image(figure, i2_loc)\n\n\t\tself.match_alignment = {}; self.match_alignment_scores = {}\n\t\tfor pair, score in zip(feature_pairs, feature_pair_scores):\n\t\t\ti1 = self.Database[\"image_idx\"][pair[0]]\n\t\t\ti2 = self.Database[\"image_idx\"][pair[1]]\n\n\t\t\tif i1 == image_pair[0]:\n\t\t\t\ttarget = pair[0] \n\t\t\t\tstart = pair[1]\n\t\t\t\tif target not in self.match_alignment.keys():\n\t\t\t\t\tself.init_match_alignment(target)\n\t\t\telse:\n\t\t\t\ttarget = pair[1]\n\t\t\t\tstart = pair[0]\t\t\t\t\n\t\t\t\tif target not in self.match_alignment.keys():\n\t\t\t\t\tself.init_match_alignment(target)\n\n\t\t\t# get ellipse and convert to three cardinal points\n\t\t\ttarget_ellipse = self.Database[\"orientation\"][target]\n\t\t\tstart_ellipse = self.Database[\"orientation\"][start]\n\t\t\ttarget_points, target_params = utils.ellipse2points(target_ellipse)\n\t\t\tstart_points, start_params = utils.ellipse2points(start_ellipse)\n\n\t\t\t# compute angle between Major axes\n\t\t\trotation_angle = int(np.abs(target_params[0] - start_params[0]))\n\t\t\treflection_truth = target_params[1] * start_params[1]\n\t\t\tif reflection_truth > 0:\n\t\t\t\treflection_truth = \"No\"\n\t\t\telse:\n\t\t\t\treflection_truth = \"Yes\"\n\t\t\tresolution_perc = round(((target_params[2] / start_params[2]) - 1), 1)\n\n\t\t\t# extract features for histogram matching\n\t\t\ttarget_bloc = self.Database[\"blot_locs\"][target]\n\t\t\tstart_bloc = self.Database[\"blot_locs\"][start]\n\t\t\ttarget_feature = extract_image(self.image1, target_bloc, ex=5)\n\t\t\tfinal_feature = extract_image(self.image2, start_bloc, ex=5) \n\n\t\t\t#find affine transform to match\n\t\t\tif self.AFFINE:\n\t\t\t\tT = utils.compute_affine(start_points, target_points)\n\t\t\t\timage2_warped = cv2.warpAffine(self.image2.copy(), T, (self.image1.shape[1], self.image1.shape[0]))\n\t\t\t\tfinal_feature = extract_image(image2_warped, target_bloc, ex=5)\n\t\t\t\n\t\t\t# match hist\n\t\t\tif self.CONTRAST:\n\t\t\t\tfinal_feature = utils.histogram_match(final_feature, target_feature)\n\n\t\t\tself.match_alignment[target].append((final_feature, start, rotation_angle, reflection_truth, \\\n\t\t\t\t\t\t\t\t\t\t\t\t resolution_perc, score))\n\t\t\tself.match_alignment_scores[target].append(score)","def compare_predictions():\n validation_labels = np.array(pd.read_csv(val_true_labels_dir + dataset_version + 'validation_labels.csv', index_col=0))\n validation_labels = np.reshape(validation_labels, (-1))\n\n diff_between_files = []\n also1s = []\n also2s = []\n for filename1 in os.listdir(val_predictions_dir):\n if filename1.endswith(\".csv\"):\n for filename2 in os.listdir(val_predictions_dir):\n if filename2.endswith(\".csv\"):\n if filename1 < filename2:\n wrong1 = 0\n wrong2 = 0\n diff_between = 0\n also1 = 0\n also2 = 0\n diff_corr1 = 0\n diff_corr2 = 0\n f1 = np.array(pd.read_csv(val_predictions_dir + filename1, index_col=0))\n f1 = np.reshape(f1, (-1))\n f2 = np.array(pd.read_csv(val_predictions_dir + filename2, index_col=0))\n f2 = np.reshape(f2, (-1))\n for line in range(f1.shape[0]):\n if f1[line] != validation_labels[line]:\n wrong1 += 1\n if f2[line] != validation_labels[line]:\n wrong2 += 1\n if f1[line] != f2[line]:\n diff_between += 1\n if f1[line] == validation_labels[line]:\n diff_corr1 += 1\n if f2[line] == validation_labels[line]:\n diff_corr2 += 1\n if f1[line] != validation_labels[line]:\n if f2[line] != validation_labels[line]:\n also2 += 1\n if f2[line] != validation_labels[line]:\n if f1[line] != validation_labels[line]:\n also1 += 1\n\n diff_between_files.append(diff_between)\n print(filename1)\n print('Wrongly predicted by 1: ' + str(100 * wrong1 / f1.shape[0]) + '%')\n print(filename2)\n print('Wrongly predicted by 2: ' + str(100 * wrong2 / f1.shape[0]) + '%')\n print()\n print('Differences between files: ' + str(100 * diff_between / f1.shape[0]) + '%')\n print(f'\\t of which correct by 1 {100 * diff_corr1 / diff_between}%, by 2 {100 * diff_corr2 / diff_between}%')\n also1s.append(also1 / wrong2)\n also2s.append(also2 / wrong1)\n print('Wrongly predicted by other among wrong ones: ' + str(100 * also2 / wrong1) + '%, ' + str(\n 100 * also1 / wrong2) + '%\\n\\n\\n')\n\n print('Max, min and avg differences between files:')\n print(str(100 * max(diff_between_files) / validation_labels.shape[0]) + '%')\n print(str(100 * min(diff_between_files) / validation_labels.shape[0]) + '%')\n print(str(100 * np.mean(diff_between_files) / validation_labels.shape[0]) + '%')\n\n print('\\nWrongly predicted by first that were also wrongly predicted by second:')\n print('Max: ' + str(100 * max(also2s)) + '%')\n print('Min: ' + str(100 * min(also2s)) + '%')\n print('Avg: ' + str(100 * np.mean(also2s)) + '%')\n\n print('\\nWrongly predicted by second that were also wrongly predicted by first:')\n print('Max: ' + str(100 * max(also1s)) + '%')\n print('Min: ' + str(100 * min(also1s)) + '%')\n print('Avg: ' + str(100 * np.mean(also1s)) + '%')","def matching_accuracy(pmat_pred, pmat_gt, ns):\n device = pmat_pred.device\n batch_num = pmat_pred.shape[0]\n\n pmat_gt = pmat_gt.to(device)\n\n assert torch.all((pmat_pred == 0) + (pmat_pred == 1)), 'pmat_pred can noly contain 0/1 elements.'\n assert torch.all((pmat_gt == 0) + (pmat_gt == 1)), 'pmat_gt should noly contain 0/1 elements.'\n assert torch.all(torch.sum(pmat_gt, dim=-1) <= 1) and torch.all(torch.sum(pmat_gt, dim=-2) <= 1)\n assert torch.all(torch.sum(pmat_pred, dim=-1) <= 1) and torch.all(torch.sum(pmat_pred, dim=-2) <= 1)\n\n #indices_pred = torch.argmax(pmat_pred, dim=-1)\n #indices_gt = torch.argmax(pmat_gt, dim=-1)\n\n #matched = (indices_gt == indices_pred).type(pmat_pred.dtype)\n match_num_list = []\n gt_num_list = []\n pred_num_list = []\n acc_gt = []\n acc_pred = []\n for b in range(batch_num): #acc_gt,acc_pred\n #match_num += torch.sum(matched[b, :ns[b]])\n #total_num += ns[b].item()\n match_num = torch.sum(pmat_pred[b, :] * pmat_gt[b, :]) + 1e-8 #:ns[b]\n gt_num = torch.sum(pmat_gt[b, :]) + 1e-8\n pred_num = torch.sum(pmat_pred[b,:]) + 1e-8\n match_num_list.append(match_num.cpu().numpy())\n gt_num_list.append(gt_num.cpu().numpy())\n pred_num_list.append(pred_num.cpu().numpy())\n acc_gt.append((match_num/gt_num).cpu().numpy())\n acc_pred.append((match_num/pred_num).cpu().numpy())\n\n return {'acc_gt': np.array(acc_gt),\n 'acc_pred': np.array(acc_pred),\n 'match_num': np.array(match_num_list),\n 'gt_num': np.array(gt_num_list),\n 'pred_num': np.array(pred_num_list)}","def compute_similarity(im1, im2, test_id):\n\n results = dict()\n results['id'] = test_id\n results['test_im'] = im2\n results['mse_value'], results['mse_map'] = compute_mse(im1, im2)\n results['ssim_value'], results['ssim_map'] = compute_ssim(im1, im2, 5)\n results['cw_ssim_value'], results['cw_ssim_map'] = compute_cw_ssim(im1, im2, 30)\n results['gms_value'], results['gms_map'] = compute_gms(im1, im2)\n results['fsim_value'], results['pc_max_map'] = compute_fsim(im1, im2)\n\n return results","def test_detector1pipeline4(_bigdata):\n step = Detector1Pipeline()\n step.save_calibrated_ramp = True\n step.ipc.skip = True\n step.persistence.skip = True\n step.jump.rejection_threshold = 4.0\n step.ramp_fit.save_opt = False\n step.output_file = 'jw84600007001_02101_00001_nrs1_rate.fits'\n step.run(_bigdata+'/pipelines/jw84600007001_02101_00001_nrs1_uncal.fits')\n\n # Compare ramp product\n n_ramp = 'jw84600007001_02101_00001_nrs1_ramp.fits'\n h = fits.open( n_ramp )\n n_ref = _bigdata+'/pipelines/jw84600007001_02101_00001_nrs1_ramp_ref.fits'\n href = fits.open( n_ref )\n newh = fits.HDUList([h['primary'],h['sci'],h['err'],h['groupdq'],h['pixeldq']])\n newhref = fits.HDUList([href['primary'],href['sci'],href['err'],href['groupdq'],h['pixeldq']])\n result = fits.diff.FITSDiff(newh,\n newhref,\n ignore_keywords = ['DATE','CAL_VER','CAL_VCS','CRDS_VER','CRDS_CTX'],\n rtol = 0.00001\n )\n assert result.identical, result.report()\n\n # Compare countrate image product\n n_cr = 'jw84600007001_02101_00001_nrs1_rate.fits'\n h = fits.open( n_cr )\n n_ref = _bigdata+'/pipelines/jw84600007001_02101_00001_nrs1_rate_ref.fits'\n href = fits.open( n_ref )\n newh = fits.HDUList([h['primary'],h['sci'],h['err'],h['dq']])\n newhref = fits.HDUList([href['primary'],href['sci'],href['err'],href['dq']])\n result = fits.diff.FITSDiff(newh,\n newhref,\n ignore_keywords = ['DATE','CAL_VER','CAL_VCS','CRDS_VER','CRDS_CTX'],\n rtol = 0.00001\n )\n assert result.identical, result.report()","def test_compare_classifiers_performance_changing_k_vis_api(experiment_to_use):\n experiment = experiment_to_use\n probabilities = experiment.probabilities\n viz_outputs = (\"pdf\", \"png\")\n with TemporaryDirectory() as tmpvizdir:\n for viz_output in viz_outputs:\n vis_output_pattern_pdf = tmpvizdir + f\"/*.{viz_output}\"\n visualize.compare_classifiers_performance_changing_k(\n [probabilities, probabilities],\n experiment.ground_truth,\n experiment.ground_truth_metadata,\n experiment.output_feature_name,\n top_k=3,\n labels_limit=0,\n model_namess=[\"Model1\", \"Model2\"],\n output_directory=tmpvizdir,\n file_format=viz_output,\n )\n figure_cnt = glob.glob(vis_output_pattern_pdf)\n assert 1 == len(figure_cnt)","def max_disp(left_img, right_img):\n\n # Max initial matches\n MAX_INI_MATCH = np.inf\n\n # ORB detector and brute-force matcher\n detector = cv2.ORB_create(nfeatures=100000, edgeThreshold=5)\n matcher = cv2.BFMatcher(cv2.NORM_HAMMING, crossCheck=True)\n\n # Find the keypoints and descriptors of both images\n keypoint_left, descriptor_left = detector.detectAndCompute(\n image=left_img, mask=None\n )\n keypoint_right, descriptor_right = detector.detectAndCompute(\n image=right_img, mask=None\n )\n\n # Find the best matches\n matches = matcher.match(\n queryDescriptors=descriptor_left, trainDescriptors=descriptor_right\n )\n matches = sorted(matches, key=lambda x: x.distance)\n\n if DEBUG:\n print(f\"Initial matches#\")\n print(f\"match_lr: {len(matches)}\")\n\n # Match refinement with RANSAC\n left_pts = np.float32([keypoint_left[m.queryIdx].pt for m in matches]).reshape(\n -1, 1, 2\n )\n right_pts = np.float32([keypoint_right[m.trainIdx].pt for m in matches]).reshape(\n -1, 1, 2\n )\n homography, mask = cv2.findHomography(left_pts, right_pts, cv2.RANSAC, 5.0)\n matchesMask = mask.ravel().tolist()\n\n ransac_matches = []\n for (m, keep) in zip(matches, matchesMask):\n if keep:\n ransac_matches.append(m)\n\n if DEBUG:\n print(f\"After RANSAC:\")\n print(f\"match: {len(ransac_matches)}\")\n ransac_img = cv2.drawMatches(\n left_img,\n keypoint_left,\n right_img,\n keypoint_right,\n ransac_matches,\n None,\n flags=2,\n )\n cv2.imwrite(\"./debug/RANSAC.png\", ransac_img)\n\n # Get the range of disparity\n disp_min = np.inf\n disp_max = 0\n\n for m in ransac_matches:\n # Keypoints coordinates\n pt_left = np.float32(keypoint_left[m.queryIdx].pt)\n pt_right = np.float32(keypoint_right[m.trainIdx].pt)\n\n # Compute the distance\n dist = int(abs(pt_left - pt_right)[0])\n\n if disp_max < dist:\n disp_max = dist\n\n # Ceiling the number\n disp_max += 1\n\n return disp_max","def illustrate_matching(self):\n work_dir = self.work_dir\n draw_keys_oriented(work_dir+'matching_keys_im0.txt',\n work_dir+'input_0.orig.png',\n work_dir+'matching_keys_im0.png')\n draw_keys_oriented(work_dir+'matching_keys_im1.txt',\n work_dir+'input_1.orig.png',\n work_dir+'matching_keys_im1.png')\n draw_matches(work_dir+'matches.txt',\n work_dir+'input_0.orig.png',\n work_dir+'input_1.orig.png',\n work_dir+'OUTmatches.png')\n return 1","def compute_average_precision_per_class(num_true_cases, gt_boxes,\n\t\t\t\t\t\t\t\t\t\tprediction_file, iou_threshold, use_2007_metric):\n\twith open(prediction_file) as f:\n\t\timage_ids = []\n\t\tboxes = []\n\t\tscores = []\n\t\tfor line in f:\n\t\t\tt = line.rstrip().split(\" \")\n\t\t\timage_ids.append(t[0])\n\t\t\tscores.append(float(t[1]))\n\t\t\tbox = torch.tensor([float(v) for v in t[2:]]).unsqueeze(0)\n\t\t\tbox -= 1.0 # convert to python format where indexes start from 0\n\t\t\tboxes.append(box)\n\t\tscores = np.array(scores)\n\t\tsorted_indexes = np.argsort(-scores)\n\t\tboxes = [boxes[i] for i in sorted_indexes]\n\t\timage_ids = [image_ids[i] for i in sorted_indexes]\n\t\ttrue_positive = np.zeros(len(image_ids))\n\t\tfalse_positive = np.zeros(len(image_ids))\n\t\tmatched = set()\n\t\tfor i, image_id in enumerate(image_ids):\n\t\t\tbox = boxes[i]\n\t\t\tif image_id not in gt_boxes:\n\t\t\t\tfalse_positive[i] = 1\n\t\t\t\tcontinue\n\n\t\t\tgt_box = gt_boxes[image_id]\n\t\t\tious = box_utils.iou_of(box, gt_box)\n\t\t\tmax_iou = torch.max(ious).item()\n\t\t\tmax_arg = torch.argmax(ious).item()\n\t\t\tif max_iou > iou_threshold:\n\t\t\t\tif (image_id, max_arg) not in matched:\n\t\t\t\t\ttrue_positive[i] = 1\n\t\t\t\t\tmatched.add((image_id, max_arg))\n\t\t\t\telse:\n\t\t\t\t\tfalse_positive[i] = 1\n\t\t\telse:\n\t\t\t\tfalse_positive[i] = 1\n\n\ttrue_positive = true_positive.cumsum()\n\tfalse_positive = false_positive.cumsum()\n\tprecision = true_positive / (true_positive + false_positive)\n\trecall = true_positive / num_true_cases\n\tif use_2007_metric:\n\t\treturn measurements.compute_voc2007_average_precision(precision, recall)\n\telse:\n\t\treturn measurements.compute_average_precision(precision, recall)","def test_compare_classifiers_performance_subset_vis_api(experiment_to_use):\n experiment = experiment_to_use\n probabilities = experiment.probabilities\n viz_outputs = (\"pdf\", \"png\")\n with TemporaryDirectory() as tmpvizdir:\n for viz_output in viz_outputs:\n vis_output_pattern_pdf = tmpvizdir + f\"/*.{viz_output}\"\n visualize.compare_classifiers_performance_subset(\n [probabilities, probabilities],\n experiment.ground_truth,\n experiment.ground_truth_metadata,\n experiment.output_feature_name,\n top_n_classes=[6],\n labels_limit=0,\n subset=\"ground_truth\",\n model_namess=[\"Model1\", \"Model2\"],\n output_directory=tmpvizdir,\n file_format=viz_output,\n )\n figure_cnt = glob.glob(vis_output_pattern_pdf)\n assert 1 == len(figure_cnt)","def test_compare_classifiers_predictions_distribution_vis_api(experiment_to_use):\n experiment = experiment_to_use\n predictions = experiment.predictions_num\n viz_outputs = (\"pdf\", \"png\")\n with TemporaryDirectory() as tmpvizdir:\n for viz_output in viz_outputs:\n vis_output_pattern_pdf = tmpvizdir + f\"/*.{viz_output}\"\n visualize.compare_classifiers_predictions_distribution(\n [predictions, predictions],\n experiment.ground_truth,\n experiment.ground_truth_metadata,\n experiment.output_feature_name,\n labels_limit=0,\n model_names=[\"Model1\", \"Model2\"],\n output_directory=tmpvizdir,\n file_format=viz_output,\n )\n figure_cnt = glob.glob(vis_output_pattern_pdf)\n assert 1 == len(figure_cnt)","def ransac_homography(keypoints1, keypoints2, matches, sampling_ratio=0.5, n_iters=500, delta=20):\n N = matches.shape[0]\n n_samples = int(N * sampling_ratio)\n\n matched1_unpad = keypoints1[matches[:,0]]\n matched2_unpad = keypoints2[matches[:,1]]\n\n\n max_inliers = np.zeros(N)\n n_inliers = 0\n\n # RANSAC iteration start\n ### YOUR CODE HERE\n for i in range(n_iters):\n random_sample = np.random.choice(N, n_samples)\n src = matched1_unpad[random_sample] # leave unpadded, is padded in compute homography\n dst = matched2_unpad[random_sample]\n\n # Get H using DLT\n H_matrix = compute_homography(src, dst)\n\n # Matched keypoints are considered inliers if the projected point from one image lies within distance δ to the matched point on the other point.\n\n projection = transform_homography(matched1_unpad, H_matrix) # x point, y point\n distances = np.sqrt(np.sum(np.square(projection-matched2_unpad), axis=1))\n if np.sum(distances < delta) > n_inliers:\n max_inliers = distances < delta\n n_inliers = np.sum(distances < delta)\n \n # Get final H using maximum inliers indices and src dst values\n src = matched1_unpad[max_inliers]\n dst = matched2_unpad[max_inliers]\n H = compute_homography(src, dst)\n \n ### END YOUR CODE\n return H, matches[max_inliers]","def evaluate_rouge_match(ref_data, pred_data, align=\"rouge\"):\n\n scorer = RougeScorer([\"rougeL\"])\n agg = scoring.BootstrapAggregator(confidence_interval=0.95)\n sample_evaluated = 0\n\n for tid in tqdm(pred_data):\n cur_sys = pred_data[tid][\"pred\"]\n if not tid in ref_data: continue\n cur_ref_lst = [x[\"tgt_words\"] for x in ref_data[tid]]\n if len(cur_ref_lst) > 0:\n if align == \"rouge\":\n best_rouge_scrs = __rouge_pairs(cur_sys, cur_ref_lst, scorer)\n agg.add_scores(best_rouge_scrs)\n else:\n best_paired_ref, _ = utils.pair_ref_sys_by_coverage(cur_sys, cur_ref_lst)\n agg.add_scores(scorer.score(best_paired_ref, cur_sys))\n sample_evaluated += 1\n\n print(sample_evaluated)\n result = agg.aggregate()\n for rouge_type in result:\n print(rouge_type)\n print(result[rouge_type].mid)\n\n print(agg.aggregate())\n return","def calculate_metrics(predictions, expected):\n # type: (np.ndarray, np.ndarray) -> (float, float, float)\n clients_count = predictions.shape[0]\n products_count = predictions.shape[1]\n\n true_positive = 0.0\n true_negative = 0.0\n false_positive = 0.0\n false_negative = 0.0\n\n total = float(clients_count * products_count)\n\n for c in range(0, clients_count):\n for p in range(0, products_count):\n if predictions[c, p] == expected[c, p]:\n if predictions[c, p] == 1:\n true_positive += 1\n else:\n true_negative += 1\n else:\n if predictions[c, p] == 1:\n false_positive += 1\n else:\n false_negative += 1\n\n accuracy = float(true_positive + true_negative) / total\n if true_positive + false_positive == 0:\n precision = 0\n else:\n precision = true_positive / float(true_positive + false_positive)\n\n if true_positive + false_negative == 0:\n recall = 0\n else:\n recall = true_positive / float(true_positive + false_negative)\n\n return accuracy, precision, recall","def match2(img1, img2, coordinates1, coordinates2, PATCH_SIZE):\n\n\tpossible_matches = pd.DataFrame(columns=['feature1', 'feature2', 'diff'])\n\n\t# iteration through all the possible pairs of features from img1 and img2\n\tfor (feature1, feature2) in product(coordinates1, coordinates2):\n\t\tpatch1 = make_patch(feature1, PATCH_SIZE, img1)\n\t\tpatch2 = make_patch(feature2, PATCH_SIZE, img2)\n\t\tdiff = count_difference(patch1, patch2)\n\t\t# print(feature1, feature2, diff)\n\t\t# next line is weird, don't ask\n\t\tpossible_matches = possible_matches.append({'feature1': feature1, 'feature2': feature2, 'diff': diff}, ignore_index=True)\n\n\n\t# sorting the possible_matches according to their difference and resetin the index\n\t# inplace=True means that changes will be performed in-place, so they will change the variable and there is no need to reassign\n\tpossible_matches.sort_values(['diff'], inplace=True)\n\tpossible_matches.reset_index(inplace=True, drop=True)\n\t# print(\"SORTED\")\n\t# print(possible_matches)\n\n\t# adding only the best matches\n\tmatches = []\n\twhile not possible_matches.empty:\n\t\tbest_match = possible_matches.loc[0]\n\t\tmatches.append(list(best_match))\n\t\t# creating a list of indicies to be dropped\n\t\ttrash = []\n\n\t\t#finding and deleting rows with features that are already used and reseting the index\n\t\tfor i, (f1, f2, d) in enumerate(possible_matches.values):\n\t\t \n\t\t\tif np.all(f1 == best_match[0]) or np.all(f2 == best_match[1]):\n\t\t\t\ttrash.append(i)\n\t\n\t\tfor i in trash:\n\t\t\tpossible_matches.drop(i, inplace=True)\n\t\tpossible_matches.reset_index(inplace=True, drop=True)\n\t\t# print(possible_matches)\n\n\treturn matches","def detail_matching(self):\n paradic = self.cfg['param']['paradic']\n work_dir = self.work_dir\n \n x = float(self.cfg['param']['x']) # selected pixel in the first image\n y = float(self.cfg['param']['y'])\n \n # sift parameters\n # number of bins in the orientation histogram\n n_bins = int(paradic['n_bins']) \n n_hist = int(paradic['n_hist']) \n # descriptor of n_hist X n_hist weighted histograms with n_ori\n n_ori = int(paradic['n_ori']) \n delta_min = float(paradic['delta_min'])\n sigma_min = float(paradic['sigma_min'])\n sigma_in = float(paradic['sigma_in'])\n lambda_ori = float(paradic['lambda_ori'])\n lambda_descr = float(paradic['lambda_descr'])\n #threshold defining reference orientations\n n_spo = int(paradic['n_spo'])\n \n # Read feature vectors from output files\n if (os.path.getsize(work_dir+'OUTmatches.txt') > 0 ):\n pairdata = find_nearest_keypoint(work_dir+'OUTmatches.txt', y, x)\n \n illustrate_pair(pairdata, n_bins, n_hist, n_ori, work_dir)\n\n \n # Read keys coordinates.\n d = 6+n_bins+n_hist*n_hist*n_ori # size of keydata inside pairdata\n v = n_hist*n_hist*n_ori\n [x1, y1, sigma1, theta1] = [float(x) for x in pairdata[0:4]]\n [o1, s1] = [float(x) for x in pairdata[4+v:4+v+2]]\n [x2a, y2a, sigma2a, theta2a] = [float(x) for x in pairdata[d:d+4]]\n [o2a, s2a] = [float(x) for x in pairdata[d+4+v:d+4+v+2]]\n [x2b, y2b, sigma2b, theta2b] = \\\n [float(x) for x in pairdata[2*d:2*d+4]]\n [o2b, s2b] = [float(x) for x in pairdata[2*d+4+v:2*d+4+v+2]]\n \n draw_one_match(pairdata,\n work_dir+'input_0.png',\n work_dir+'input_1.png',\n d,\n lambda_ori,\n lambda_descr,\n n_hist,\n work_dir+'OUTonepair.png')\n \n \n # Extract thumbnails.\n # keypoint 1 (image 1)\n print ' '.join(['demo_extract_patch', work_dir+'input_0.png',\n str(x1), str(y1), str(sigma1), str(theta1), str(o1), str(s1),\n str(delta_min), str(sigma_min), str(sigma_in), str(n_spo),\n str(lambda_ori), str(lambda_descr), str(n_hist),\n work_dir+\"detail_im1\"])\n proc = self.run_proc(['demo_extract_patch', work_dir+'input_0.png',\n str(x1), str(y1), str(sigma1), str(theta1), str(o1), str(s1),\n str(delta_min), str(sigma_min), str(sigma_in), str(n_spo),\n str(lambda_ori), str(lambda_descr), str(n_hist),\n work_dir+\"detail_im1\"])\n self.wait_proc(proc, timeout=self.timeout)\n \n # keypoint 2a (nearest neighbor in image 2)\n print ' '.join(['demo_extract_patch', work_dir+'input_1.png',\n str(x2a), str(y2a), str(sigma2a), str(theta2a), str(o2a), str(s2a),\n str(delta_min), str(sigma_min), str(sigma_in), str(n_spo),\n str(lambda_ori), str(lambda_descr), str(n_hist),\n work_dir+\"detail_im2a\"])\n proc = self.run_proc(['demo_extract_patch', work_dir+'input_1.png',\n str(x2a), str(y2a), str(sigma2a), str(theta2a), str(o2a), str(s2a),\n str(delta_min), str(sigma_min), str(sigma_in), str(n_spo),\n str(lambda_ori), str(lambda_descr), str(n_hist),\n work_dir+\"detail_im2a\"])\n self.wait_proc(proc, timeout=self.timeout) \n \n # keypoint 2b (second nearest neighbor in image 2)\n proc = self.run_proc(['demo_extract_patch', work_dir+'input_1.png',\n str(x2b), str(y2b), str(sigma2b), str(theta2b), str(o2b), str(s2b),\n str(delta_min), str(sigma_min), str(sigma_in), str(n_spo),\n str(lambda_ori), str(lambda_descr), str(n_hist),\n work_dir+\"detail_im2b\"])\n self.wait_proc(proc, timeout=self.timeout) \n \n \n return 1","def compare_images(self, df):\n score_list = []\n elapsed_list = []\n for index,row in df.iterrows():\n start_time = time.time()\n image1 = imread(row['image1'],as_gray=\"True\")\n image2 = imread(row['image2'],as_gray=\"True\")\n if(image1.shape != image2.shape):\n image2 = resize(image2, (image1.shape)) # if images are of different dimensions, crop the second image based on first\n score = structural_similarity(image1, image2)\n new_score = self.__convert_similarity_score(score)\n elapsed_time = (time.time() - start_time)\n score_list.append(round(new_score,2))\n elapsed_list.append(elapsed_time)\n output_df = self.__create_output_csv(df, score_list, elapsed_list)\n return output_df","def calculate_score(label_dict, queries, results):\n\n total_score = 0.0\n\n log_header('Individual image scores')\n\n # Calculate score for all images\n for image_id in queries:\n if image_id in results.keys():\n # Run the score function\n image_score = score(\n label_dict=label_dict, target=image_id, selection=results[image_id]\n )\n else:\n logging.error('No result generated for %s' % image_id)\n\n image_score = 0.0\n\n total_score += image_score\n\n logging.info('%s: %8.6f' % (image_id, image_score))\n\n log_header('Average score over %d images: %10.8f' % (\n len(queries), total_score / len(queries)\n ))\n\n return total_score","def compare_results(self, benchmark_result, reference_benchmark_result):\n pass","def run_compare(self):\n misses = []\n for row in range(self.data.get_number_of_rows()):\n ref_sentence = self.data.get_row(row)[self.data.get_gold()]\n results = {}\n for team, team_sentence in self.data.get_row_for_teams(self.evaluator.get_teams(row), row).iteritems():\n results[team] = self.get_sentence_score(ref_sentence, team_sentence)\n misses.append(self.evaluator.compare_all(results, row))\n print np.median(misses), np.mean(misses)\n return np.median(misses)","def eval_feature_verify(query_features, gallery_features, query_lst, gallery_lst, metric=\"cosine\"):\n if metric not in [\"euclidean\", \"cosine\"]:\n raise ValueError(\"Invalid metric! \")\n\n num_query = len(query_lst)\n num_gallery = len(gallery_lst)\n\n if metric == \"cosine\":\n gallery_features = normalize(gallery_features, axis=1)\n query_features = normalize(query_features, axis=1)\n\n if metric == \"euclidean\":\n dist = euclidean_distances(gallery_features, query_features).squeeze()\n else:\n dist = np.dot(gallery_features, query_features.T)\n print(dist[:].max())\n print(dist[:].min())\n\n gallery_cam_lst = np.array([x.cam_id for x in gallery_lst], dtype=np.int32)\n gallery_id_lst = np.array([x.class_id for x in gallery_lst], dtype=np.int32)\n query_cam_lst = np.array([x.cam_id for x in query_lst], dtype=np.int32)\n query_id_lst = np.array([x.class_id for x in query_lst], dtype=np.int32)\n\n # same id\n gallery_id_lst_tile = np.tile(gallery_id_lst, (num_query, 1))\n query_id_lst_tile = np.tile(query_id_lst, (num_gallery, 1))\n label_mat = query_id_lst_tile == gallery_id_lst_tile.T # gallery num * query num\n label_mat = label_mat.astype(np.int)\n\n # same camera and id\n gallery_cam_lst_tile = np.tile(gallery_cam_lst, (num_query, 1))\n query_cam_lst_tile = np.tile(query_cam_lst, (num_gallery, 1))\n # same camera, gallery num * query num\n label_cam_mat = query_cam_lst_tile == gallery_cam_lst_tile.T\n # filter by same id\n label_cam_mat = label_cam_mat * label_mat\n # same id but under same camera are labeled as -1,\n label_mat -= 2 * label_cam_mat.astype(np.int)\n\n ind = np.where(label_mat == 0)\n neg_dist = dist[ind]\n ind = np.where(label_mat == 1)\n pos_dist = dist[ind]\n\n new_dist = np.concatenate([pos_dist, neg_dist])\n new_label = np.concatenate([np.ones_like(pos_dist), np.zeros_like(neg_dist)])\n dist_label = zip(new_dist, new_label)\n dist_label_new = np.array(sorted(dist_label, key=lambda d: -d[0]))\n\n # ap = []\n n_good = np.sum(dist_label_new[:, 1])\n n_bad = np.sum(1 - dist_label_new[:, 1])\n print(n_good)\n for thresh in np.arange(0.90, 1, 0.01):\n # print(thresh)\n p = dist_label_new[:, 0] >= thresh\n if np.sum(p) == 0:\n recall, precision, f1 = 0, 0, 0\n else:\n tp = np.logical_and(dist_label_new[:, 1], p)\n recall = np.sum(tp) / n_good\n precision = np.float(np.sum(tp)) / np.sum(p)\n\n f1 = 2*recall*precision/(recall+precision)\n\n print('%.4f\\t%.4f\\t%.4f' % (recall, precision, f1))\n\n # ap.append(compute_ap(dist_label_new, thresh))\n # print(ap)","def matcher(features1, features2):\n #TODO: write a matching function\n #Performing the L2-Norm\n new_features1=[]\n new_features2=[]\n for itr in range(5):\n [rootOfSquare1,rootOfSquare2] = sumOfSquares(features1[itr],features2[itr])\n new_features1.append(np.array(features1[itr])/rootOfSquare1)\n new_features2.append(np.array(features2[itr])/rootOfSquare2)\n indices = []\n for itr in range(len(new_features1)):\n findMinDist=[]\n #findMaxCosineVal=[]\n for itr2 in range(len(new_features2)):\n f1 = new_features1[itr]\n f2 = new_features2[itr2]\n\n #For evaluating the cosine similarity\n # [rootOfSquare1,rootOfSquare2] = sumOfSquares(f1,f2)\n # numerator = np.array(f1)*np.array(f2)\n # numeratorSum = sum(numerator)\n # denominator = rootOfSquare1*rootOfSquare2\n # cosine = np.divide(numeratorSum,denominator)\n # findMaxCosineVal.append(cosine)\n\n #For evaluating the similarity based on euclidean distance\n Dist = np.array(f1) - np.array(f2)\n sum=0\n for i in Dist:\n sum=sum+math.pow(i,2)\n rootOfSum = math.sqrt(sum)\n findMinDist.append(rootOfSum)\n # print \"itr: \", itr, \" Matching scores: \", findMaxCosineVal\n # bestMatch = findMaxCosineVal.index(max(findMaxCosineVal))\n bestMatch = findMinDist.index(min(findMinDist))\n indices.append([itr,bestMatch])\n return indices","def compute(self, result_file_dict):\r\n for part in self.parts:\r\n #=====================Need to change, temporal=========================\r\n if part == 'train':\r\n continue # because the train not have the label\r\n #=======================================================================\r\n gt = self.gt_dict[part]\r\n result_file = result_file_dict[part]\r\n # import ipdb; ipdb.set_trace()\r\n for key, item in result_file.items():\r\n self._result_name = item\r\n # score_records, num_videos = self.load_results(result_file)\r\n score_records, num_videos = self.load_results(item)\r\n logger.info(f'Compute Metric of {item}')\r\n assert num_videos == len(gt), f'the number of saved videos does not match the ground truth, {num_videos} != {len(gt)}'\r\n temp_result = self.eval_method(score_records, gt, str(key))\r\n if temp_result > self.optimal_resulst:\r\n self.optimal_resulst = temp_result\r\n \r\n return self.optimal_resulst","def evaluate(ground_truth, prediction):\n\n def prfs_to_dict(prfs):\n \"\"\"Returns a precision_recall_fscore_support() result as a dict.\"\"\"\n return {\"precision\": prfs[0], \"recall\": prfs[1], \"fscore\": prfs[2]}\n\n results = {}\n items_count = len(ground_truth)\n\n # accuracy\n accuracy = accuracy_score(ground_truth, prediction)\n results[\"accuracy\"] = accuracy\n\n # confusion matrix\n categories = set(ground_truth) | set(prediction)\n confusions = {\n gold: {pred: 0 for pred in categories} for gold in categories\n }\n for g, p in zip(ground_truth, prediction):\n confusions[g][p] += 1\n results[\"confusions\"] = confusions\n\n # class wise precision, recall & f1\n classwise = precision_recall_fscore_support(\n ground_truth, prediction, average=None, warn_for=()\n )\n results[\"true_cat_dist\"] = list(classwise[-1])\n results[\"classwise\"] = {\n str(cl): prfs_to_dict(\n [classwise[0][cl], classwise[1][cl], classwise[2][cl]]\n )\n for cl in categories\n }\n\n # average precision, recall & f1\n results[\"macro_avg\"] = prfs_to_dict(\n precision_recall_fscore_support(\n ground_truth,\n prediction,\n average=\"macro\",\n pos_label=None,\n warn_for=(),\n )\n )\n results[\"micro_avg\"] = prfs_to_dict(\n precision_recall_fscore_support(\n ground_truth,\n prediction,\n average=\"micro\",\n pos_label=None,\n warn_for=(),\n )\n )\n results[\"weigh_avg\"] = prfs_to_dict(\n precision_recall_fscore_support(\n ground_truth,\n prediction,\n average=\"weighted\",\n pos_label=None,\n warn_for=(),\n )\n )\n\n # marginals\n gold_category_distribution = {\n g: sum([confusions[g][p] for p in categories]) for g in categories\n }\n pred_category_distribution = {\n p: sum([confusions[g][p] for g in categories]) for p in categories\n }\n\n # kappa\n expected_agreement_fleiss = sum(\n [\n (\n (gold_category_distribution[c] + pred_category_distribution[c])\n / (2.0 * items_count)\n )\n ** 2\n for c in categories\n ]\n )\n expected_agreement_cohen = sum(\n [\n (float(gold_category_distribution[c]) / items_count)\n * (float(pred_category_distribution[c]) / items_count)\n for c in categories\n ]\n )\n kappa_fleiss = (\n 1.0\n * (accuracy - expected_agreement_fleiss)\n / (1 - expected_agreement_fleiss)\n )\n kappa_cohen = (\n 1.0\n * (accuracy - expected_agreement_cohen)\n / (1 - expected_agreement_cohen)\n )\n results[\"k_fleiss\"] = {\n \"k\": kappa_fleiss,\n \"AE\": expected_agreement_fleiss,\n \"AO\": accuracy,\n }\n results[\"k_cohen\"] = {\n \"k\": kappa_cohen,\n \"AE\": expected_agreement_cohen,\n \"AO\": accuracy,\n }\n\n return results","def test_compare_performance_vis_api(experiment_to_use):\n experiment = experiment_to_use\n # extract test stats only\n test_stats = experiment.test_stats_full\n viz_outputs = (\"pdf\", \"png\")\n with TemporaryDirectory() as tmpvizdir:\n for viz_output in viz_outputs:\n vis_output_pattern_pdf = tmpvizdir + f\"/*.{viz_output}\"\n visualize.compare_performance(\n [test_stats, test_stats],\n output_feature_name=None,\n model_names=[\"Model1\", \"Model2\"],\n output_directory=tmpvizdir,\n file_format=viz_output,\n )\n figure_cnt = glob.glob(vis_output_pattern_pdf)\n assert 1 == len(figure_cnt)","def evaluate_detections(self, ap, phrase_counts, top1acc, total_aug, top1acc_aug, top1acc_oracle, top1acc_aug_oracle):\n # organize mAP by the number of occurrences\n count_thresholds = cfg.TEST.PHRASE_COUNT_THRESHOLDS\n mAP = np.zeros(len(count_thresholds))\n occurrences = np.zeros_like(mAP)\n samples = np.zeros_like(mAP)\n samples_aug = np.zeros_like(mAP)\n acc = np.zeros_like(mAP)\n acc_aug = np.zeros_like(mAP)\n oracle = np.zeros_like(mAP)\n oracle_aug = np.zeros_like(mAP)\n for phrase, phrase_index in self._phrase_to_ind.iteritems():\n n_occurrences = phrase_counts[phrase_index]\n if n_occurrences < 1:\n continue\n\n train_count = 0\n if phrase in self._train_counts:\n train_count = self._train_counts[phrase]\n\n count_index = min(np.where(train_count <= count_thresholds)[0])\n mAP[count_index] += ap[phrase_index]\n occurrences[count_index] += 1\n samples[count_index] += n_occurrences\n acc[count_index] += top1acc[phrase_index]\n acc_aug[count_index] += top1acc_aug[phrase_index]\n samples_aug[count_index] += total_aug[phrase_index]\n oracle[count_index] += top1acc_oracle[phrase_index]\n oracle_aug[count_index] += top1acc_aug_oracle[phrase_index]\n\n mAP = mAP / occurrences\n thresh_string = '\\t'.join([str(thresh) for thresh in count_thresholds])\n print('\\nThresholds: \\t' + thresh_string + '\\tOverall')\n\n ap_string = '\\t'.join(['%.1f' % round(t * 100, 2) for t in mAP])\n print('AP: \\t' + ap_string + '\\t%.1f' % round(np.mean(mAP) * 100, 2))\n\n n_total = np.sum(samples)\n n_aug = np.sum(total_aug)\n loc_acc = np.sum(acc) / (n_total - n_aug)\n group_acc = acc / (samples - samples_aug)\n acc_string = '\\t'.join(['%.1f' % round(t * 100, 2) for t in group_acc])\n print('Loc Acc:\\t' + acc_string + '\\t%.1f' % round(loc_acc * 100, 2))\n \n loc_acc = np.sum(oracle) / (n_total - n_aug)\n group_acc = oracle / (samples - samples_aug)\n acc_string = '\\t'.join(['%.1f' % round(t * 100, 2) for t in group_acc])\n print('Oracle Acc:\\t' + acc_string + '\\t%.1f' % round(loc_acc * 100, 2))\n if cfg.AUGMENTED_POSITIVE_PHRASES:\n loc_acc = (np.sum(acc) + np.sum(acc_aug)) / n_total\n group_acc = (acc+acc_aug) / samples\n acc_string = '\\t'.join(['%.1f' % round(t * 100, 2) for t in group_acc])\n print('Aug Loc Acc:\\t' + acc_string + '\\t%.1f' % round(loc_acc * 100, 2))\n\n loc_acc = (np.sum(oracle) + np.sum(oracle_aug)) / n_total\n group_acc = (oracle+oracle_aug) / samples\n acc_string = '\\t'.join(['%.1f' % round(t * 100, 2) for t in group_acc])\n print('Oracle Aug Acc:\\t' + acc_string + '\\t%.1f' % round(loc_acc * 100, 2))\n\n occ_string = '\\t'.join(['%i' % occ for occ in occurrences])\n print('Per Thresh Cnt:\\t' + occ_string + '\\t%i' % np.sum(occurrences))\n\n sample_string = '\\t'.join(['%i' % item for item in samples])\n print('Instance Cnt: \\t' + sample_string + '\\t%i' % n_total)\n\n if cfg.TOP_K_PER_PHRASE > 1:\n n_correct = np.sum([np.sum(item) for item in gt_labels])\n acc = round((n_correct/n_total)*100, 2)\n print('Portion of phrases with good boxes: %.2f\\n' % acc)\n\n return np.mean(mAP)","def calc_TP_FP(groundtruth, eval_boxes, matches_predicate, confidence_threshold=-9e999):\n TP = 0\n FP = 0\n FN = 0\n for image_key in groundtruth.keys() + eval_boxes.keys():\n gt_bboxes = set(groundtruth.get(image_key, []))\n other_bboxes = set(box\n for box in eval_boxes.get(image_key, [])\n if box.confidence >= confidence_threshold\n or box.confidence is None\n )\n for gt_bbox in gt_bboxes:\n # Do any of their boxes match?\n match = None\n for other_box in other_bboxes:\n if matches_predicate(gt_bbox, other_box):\n match = other_box\n if match:\n other_bboxes.remove(match)\n TP += 1\n else:\n FN += 1\n FP += len(other_bboxes)\n # if len(other_bboxes):\n # print \"False positive on\", image_key\n return {\"TP\": TP, \"FP\": FP, \"FN\": FN, \"TN\": 0}","def match_mfovs_features(matcher_params, sec1_cache, sec2_cache, mfovs1, mfovs2):\n \n thread_local_store = ThreadLocalStorageLRU()\n if 'matcher' in thread_local_store.keys():\n matcher = thread_local_store['matcher']\n else:\n # Initialize the matcher, and store it in the local thread storage\n matcher = FeaturesMatcher(BlobDetector2D.create_matcher, **matcher_params)\n thread_local_store['matcher'] = matcher\n \n# matcher = getattr(threadLocal, 'matcher', None)\n# if matcher is None:\n# # Initialize the matcher, and store it in the local thread storage\n# matcher = FeaturesMatcher(BlobDetector2D.create_matcher, **matcher_params)\n# threadLocal.matcher = matcher\n\n def get_kps_descs(mfovs, sec_cache):\n mfovs = list(mfovs)\n if len(mfovs) == 1:\n mfovs_kps = np.array(sec_cache[\"pre_match_blobs\"][mfovs[0]][0])\n mfovs_descs = np.array(sec_cache[\"pre_match_blobs\"][mfovs[0]][1])\n else:\n mfovs_kps_arrays = []\n mfovs_descs_arrays = []\n for mfov in mfovs:\n kps_descs = sec_cache[\"pre_match_blobs\"][mfov]\n if len(kps_descs[0]) > 0:\n mfovs_kps_arrays.append(kps_descs[0])\n mfovs_descs_arrays.append(kps_descs[1])\n if len(mfovs_kps_arrays) == 0:\n mfovs_kps = np.array([])\n mfovs_descs = np.array([])\n elif len(mfovs_kps_arrays) == 1:\n mfovs_kps = mfovs_kps_arrays[0]\n mfovs_descs = mfovs_descs_arrays[0]\n else:\n mfovs_kps = np.vstack(mfovs_kps_arrays)\n mfovs_descs = np.vstack(mfovs_descs_arrays)\n return np.array(mfovs_kps), np.array(mfovs_descs)\n\n mfovs1_kps, mfovs1_descs = get_kps_descs(mfovs1, sec1_cache)\n mfovs2_kps, mfovs2_descs = get_kps_descs(mfovs2, sec2_cache)\n\n model, filtered_matches = matcher.match_and_filter(mfovs1_kps, mfovs1_descs, mfovs2_kps, mfovs2_descs)\n return mfovs1, model, filtered_matches","def main():\n num_matches = 1000\n matchlist = [\n [ RandomAI, RandomAI ],\n #[ VectorAI, RandomAI ],\n #[ VectorAI, VectorAI ],\n #[ LeftmostAI, RandomAI ],\n #[ LeftmostAI, VectorAI ],\n #[ LeftmostAI, LeftmostAI ],\n\t\t#[ RightmostAI, RandomAI ],\n #[ RightmostAI, VectorAI ],\n #[ RightmostAI, LeftmostAI ],\n #[ RightmostAI, RightmostAI ]\n ]\n\n x=0\n while x < len(matchlist):\n \n print (\"Running \" + str(num_matches) + \" matches: \" + matchlist[x][0].__name__ + \" vs \" + matchlist[x][1].__name__)\n counter = 0 \n \n while counter < num_matches:\n counter = counter + 1\n match = Match(player1_type=matchlist[x][0], player2_type=matchlist[x][1], param_print_game_status=False, param_matchgroup=x)\n match.handle_next_move()\n\n if matchlist[x][0] != matchlist[x][1]:\n print (\"Running \" + str(num_matches) + \" matches: \" + matchlist[x][1].__name__ + \" vs \" + matchlist[x][0].__name__)\n counter = 0\n\n while counter < num_matches:\n counter = counter + 1\n match = Match(player1_type=matchlist[x][1], player2_type=matchlist[x][0], param_print_game_status=False, param_matchgroup=x)\n match.handle_next_move()\n x = x+1\n \n print (\"Matches Finished\")","def evaluate(calculated_results, ground_truth, measure=''):\n score = 0\n truth_pairs = dict.fromkeys(pair[0] + ' ' + pair[1]\n for pair in ground_truth.edges())\n for pair in calculated_results:\n pair_reversed = get_reversed_pair(pair)\n if pair[0] in truth_pairs or pair_reversed in truth_pairs:\n score += 1\n accuracy = (score / len(truth_pairs)) * 100\n print(\"The accuracy is %f %% when using measure: %s\" % (accuracy, measure))","def evaluate_match_quality(matches, thresh=.7):\n marr = np.zeros((len(matches), 2))\n quality = np.zeros((len(matches), 1))\n for i in range(len(matches)):\n marr[i,0] = matches[i][0].distance\n marr[i,1] = matches[i][1].distance\n quality[i] = marr[i, 0] < thresh*marr[i, 1]","def get_metrics(cfg, model, X_anchor, y_anchor, X_gal, y_gal, annoy_index, vec_dim):\n rank10_acc = 0\n rank5_acc = 0\n rank1_acc = 0\n avg_acc = 0\n vote_res = 0\n\n l2 = []\n for anchor in range(0, len(X_anchor)):\n res = get_result(get_image_features(cfg, model, X_anchor[anchor]), annoy_index)\n vote = defaultdict(int)\n # Accuracy\n correct = 0\n for i in res[:10]:\n vote[y_gal[i]] += 1\n\n max_key = max(vote, key=vote.get)\n if max_key == y_anchor[anchor]:\n vote_res += 1\n \n\n for recomm in res[:10]:\n if y_gal[recomm] == y_anchor[anchor]:\n correct += 1 \n\n avg_acc += correct/len(res)\n\n # Mean Average Precision\n l1 = []\n for recomm in res[:10]:\n if y_gal[recomm] == y_anchor[anchor]:\n correct += 1\n l1.append(1)\n else:\n l1.append(0)\n l2.append(l1) \n\n # Rank10 Accuracy\n for each_val in res[:10]:\n if y_gal[each_val] == y_anchor[anchor]:\n rank10_acc += 1\n break\n \n # Rank5 Accuracy\n for each_val in res[:5]:\n if y_gal[each_val] == y_anchor[anchor]:\n rank5_acc += 1\n break\n\n # Rank1 Accuracy\n for each_val in res[:1]:\n if y_gal[each_val] == y_anchor[anchor]:\n rank1_acc += 1\n break\n\n print(\"Avg acc is :: {avg_acc}\".format(avg_acc = avg_acc/len(X_anchor)))\n print(\"Rank 10 acc is :: {rank10_acc}\".format(rank10_acc = rank10_acc/len(X_anchor)))\n print(\"Rank 5 acc is :: {rank5_acc}\".format(rank5_acc = rank5_acc/len(X_anchor)))\n print(\"Rank 1 acc is :: {rank1_acc}\".format(rank1_acc = rank1_acc/len(X_anchor)))\n print(\"Mean Avg Precision is :: {mAP}\".format(mAP=mean_average_precision(l2)))\n print(\"Vote res :: \", vote_res/len(X_anchor))\n\n return rank1_acc/len(X_anchor), mean_average_precision(l2)","def prediction(self, query_image_path_list, evaluation_method, demonstration_subdir, evaluation_subdir):\n semantic_tag_unique_list = list(set(self.node_semantic_tag_list))\n semantic_tag_total_count = np.array([self.node_semantic_tag_list.count(semantic_tag) for semantic_tag in semantic_tag_unique_list])\n\n es_semantic_tag_list = []\n es_target_index_list = []\n es_target_index_rank_list = []\n query_feature_dist = []\n node_index_list = []\n for query_image_path in query_image_path_list:\n query_image = cv2.imread(query_image_path)\n query_image_batch = self.process_batch([query_image])\n query_rolling_features_list = self.sess.run(self.rolling_features_list,\n feed_dict = {self.images_placeholder: query_image_batch})\n\n metric_distance_list = []\n for node_index in range(self.node_number):\n node_metric_feature = self.node_metric_feature_list[node_index]\n rolling_distance_list = []\n for query_rolling_features in query_rolling_features_list:\n rolling_distance = np.sqrt(np.sum(np.square(node_metric_feature - query_rolling_features)))\n rolling_distance_list.append(rolling_distance)\n metric_distance = min(rolling_distance_list)\n metric_distance_list.append(metric_distance)\n metric_distance_list = np.array(metric_distance_list)\n\n query_feature_dist.append(metric_distance_list)\n\n\n node_index_rank = np.argsort(metric_distance_list)\n node_index_list.append(node_index_rank[0])\n es_target_index_rank = [self.node_id_list[node_index] for node_index in node_index_rank]\n es_target_index_rank_list.append(es_target_index_rank)\n es_target_index_list.append(es_target_index_rank[0])\n es_semantic_tag_list.append(set([self.node_semantic_tag_list[node_index_rank[0]]]))\n\n if evaluation_method == 'find_target_position':\n print(\"TARGET INDEX: \"+str(es_target_index))\n print(\"semantic tag: \"+str(self.node_semantic_tag_list[es_target_index]))\n print(\"\\n\")\n return es_semantic_tag_list, es_target_index_list\n\n elif evaluation_method == 'target_position_precision-error':\n return es_target_index_list\n elif evaluation_method == 'target_position_distance-accuracy':\n return es_target_index_list\n elif evaluation_method == 'target_position_recall-N':\n return es_target_index_rank_list\n elif evaluation_method == 'target_position_precision-recall':\n return [node_index_list, query_feature_dist]\n else:\n raise ValueError(\"unknown evaluation method\")","def compare(args):\n p = OptionParser(__doc__)\n opts, args, iopts = p.set_image_options(args, figsize=\"15x5\")\n\n if len(args) != 1:\n sys.exit(not p.print_help())\n\n datafile, = args\n pf = datafile.rsplit(\".\", 1)[0]\n fig, (ax1, ax2, ax3) = plt.subplots(ncols=3, nrows=1,\n figsize=(iopts.w, iopts.h))\n plt.tight_layout(pad=2)\n\n # Huntington risk allele\n infected_thr = 40\n ref_thr = 19\n\n # ax1: Multiple callers at lower range\n df = pd.read_csv(\"Evaluation.csv\")\n truth = df[\"Truth\"]\n\n ax1.plot(truth, df[\"Manta\"], 'bx-')\n ax1.plot(truth, df[\"Isaac\"], 'yo-')\n ax1.plot(truth, df[\"GATK\"], 'md-')\n ax1.plot(truth, df[\"lobSTR\"], 'c+-')\n ax1.plot(truth, truth, 'k--') # to show diagonal\n\n bbox = {'facecolor': 'tomato', 'alpha': .2, 'ec': 'w'}\n pad = 2\n ax1.axhline(infected_thr, color='tomato')\n ax1.text(max(truth) - pad, infected_thr + pad, 'Risk threshold',\n bbox=bbox, ha=\"right\")\n ax1.axhline(ref_thr, color='tomato')\n ax1.text(max(truth) - pad, ref_thr - pad, 'Reference repeat count',\n bbox=bbox, ha=\"right\", va=\"top\")\n\n ax1.set_xlabel(r'Num of CAG repeats inserted ($\\mathit{h}$)')\n ax1.set_ylabel('Num of CAG repeats called')\n ax1.set_title(r'Simulated haploid $\\mathit{h}$')\n ax1.legend(['Manta', 'Isaac', 'GATK', 'lobSTR', 'Truth'], loc='best')\n\n max_insert = 120\n # ax2: lobSTR vs TREDPARSE with haploid model\n lobstr_results = parse_results(\"lobstr_results_homo.txt\")\n tredparse_results = parse_results(\"tredparse_results_homo.txt\")\n truth = range(10, max_insert + 1)\n lx, ly = zip(*lobstr_results)\n tx, ty = zip(*tredparse_results)\n\n ax2.plot(lx, ly, 'c+-')\n ax2.plot(tx, ty, 'gx-')\n ax2.plot(truth, truth, 'k--')\n\n ax2.set_xlabel(r'Num of CAG repeats inserted ($\\mathit{h}$)')\n ax2.set_ylabel('Num of CAG repeats called')\n ax2.set_title(r'Simulated haploid $\\mathit{h}$')\n ax2.legend(['lobSTR', 'TREDPARSE', 'Truth'], loc='best')\n\n pad *= 2\n ax2.axhline(infected_thr, color='tomato')\n ax2.text(max(truth) - pad, infected_thr + pad, 'Risk threshold',\n bbox=bbox, ha=\"right\")\n ax2.set_xlim(10, max_insert)\n\n # ax3: lobSTR vs TREDPARSE with haploid model\n lobstr_results = parse_results(\"lobstr_results_het.txt\", exclude=20)\n tredparse_results = parse_results(\"tredparse_results_het.txt\", exclude=20)\n truth = range(10, max_insert + 1)\n lx, ly = zip(*lobstr_results)\n tx, ty = zip(*tredparse_results)\n\n ax3.plot(lx, ly, 'c+-')\n ax3.plot(tx, ty, 'gx-')\n ax3.plot(truth, truth, 'k--')\n\n ax3.set_xlabel(r'Num of CAG repeats inserted ($\\mathit{h}$)')\n ax3.set_ylabel('Num of CAG repeats called')\n ax3.set_title(r'Simulated diploid $\\mathit{20/h}$')\n ax3.legend(['lobSTR', 'TREDPARSE', 'Truth'], loc='best')\n ax3.axhline(infected_thr, color='tomato')\n ax3.text(max(truth) - pad, infected_thr + pad, 'Risk threshold',\n bbox=bbox, ha=\"right\")\n ax3.set_xlim(10, max_insert)\n\n root = fig.add_axes([0, 0, 1, 1])\n pad = .03\n panel_labels(root, ((pad / 2, 1 - pad, \"A\"), (1 / 3., 1 - pad, \"B\"),\n (2 / 3., 1 - pad, \"C\")))\n normalize_axes(root)\n\n image_name = pf + \".\" + iopts.format\n savefig(image_name, dpi=iopts.dpi, iopts=iopts)","def statistics(img1_blobs, img2_blobs, matches):\n\tstatistics = {}\n\n\tstatistics['#Interest Points in img1'] = len(img1_blobs)\n\tstatistics['#Interest Points in img2'] = len(img2_blobs)\n\tstatistics['Accepted Matches'] = len(matches)\n\tdissimilarity = [match[2] for match in matches]\n\tstatistics['Mean of accepted matches'] = sum(dissimilarity)/len(dissimilarity)\n\tstatistics['SD of accepted matches'] = np.std(dissimilarity)\n\treturn statistics","def test_RawRun_process():\n for style in test_runs:\n test_runs[style].process()\n # now compare all images with ref\n ref = sorted(glob.glob('tests/data/processed_ref/*/*/*'))\n outputs = sorted(glob.glob('tests/data/processed/*/*/*'))\n for ref, out in zip(ref, outputs):\n print ref\n print out\n assert_image_equal(ref, out)","def generate_comparison_metrics(model: sklearn.linear_model,\n x_test: pd.DataFrame,\n y_test: pd.DataFrame,\n y_predictions: pd.DataFrame = None):\n\n if y_predictions is not None:\n if not y_predictions.isnull().values.all():\n y_pred = y_predictions\n elif 'themis_ml' in model.__module__:\n y_pred = model.predict(x_test, x_test[privileged].values)\n else:\n y_pred = model.predict(x_test)\n\n accuracy = sklearn.metrics.accuracy_score(y_true=y_test, y_pred=y_pred)\n precision = sklearn.metrics.precision_score(y_true=y_test, y_pred=y_pred)\n recall = sklearn.metrics.recall_score(y_true=y_test, y_pred=y_pred)\n auc = sklearn.metrics.roc_auc_score(y_true=y_test, y_score=y_pred)\n tn, fp, fn, tp = sklearn.metrics.confusion_matrix(y_true=y_test, y_pred=y_pred).ravel()\n tot_num = tn + fp + fn + tp\n tnp, fpp, fnp, tpp = tn / tot_num, fp / tot_num, fn / tot_num, tp / tot_num\n tpr = true_positive_rate(true_positives=tp, false_negatives=fn) # Also called 'sensitivity' and 'recall'\n fpr = false_positive_rate(false_positives=fp, true_negatives=tn)\n cost = false_positive_cost * fp + false_negative_cost * fn\n output_list = [accuracy, precision, recall, auc, tnp, fpp, fnp, tpp, tpr, fpr, cost]\n return output_list","def painting_matchings_local_desc(query_imgs, db_imgs, text_masks, options):\n tmp_img_format = []\n tmp_mask_format = []\n for i in range(len(query_imgs)):\n for j in range(len(query_imgs[i])):\n tmp_img_format.append(query_imgs[i][j])\n tmp_mask_format.append(text_masks[i][j])\n\n ## Keypoint + local descriptor extraction\n print(\"Obtaining query set descriptors\")\n kp_qs = keypoint_extraction(tmp_img_format, tmp_mask_format, options)\n desc_qs = descriptor_extraction(tmp_img_format, tmp_mask_format, kp_qs, options)\n\n print(\"Obtaining database descriptors\")\n kp_db = keypoint_extraction(db_imgs, None, options)\n desc_db = descriptor_extraction(db_imgs, None, kp_db, options)\n\n ## Match local descriptors \n print(\"Comparing...\") \n num_matches = match_descriptors_qs_db(tmp_img_format, db_imgs, desc_qs, desc_db, kp_qs, kp_db, options.km, options)\n return num_matches","def test_calculate_precision_happy_path(\n mock_engine: DicomImagePiiVerifyEngine,\n gt_labels: List[dict],\n all_pos: List[dict],\n expected_result: float,\n):\n # Act\n test_precision = mock_engine.calculate_precision(gt_labels, all_pos)\n\n # Assert\n assert test_precision == expected_result","def PredictMatch(yourName, tgtName, teamModels, db):\n\n teamTgtPipe, teamYourPipe = QueryTeamData(tgtName, yourName, db)\n featureCoefTgt, probTgt = PredictOp(teamTgtPipe, teamYourPipe, tgtName, teamModels)\n featureCoefYour, probYour = PredictOp(teamYourPipe, teamTgtPipe, yourName, teamModels)\n odds = round(probYour / (probYour + probTgt), 2)\n\n # In featureCoefYour, you want INCREASE those with POSTIVE COEF, DECREASE those with NEGATIVE COEF\n # In featureCoefTgt, you want to do the opposite\n\n # reverse both the sign of the coef, and '_op' in features so as to be the same with featureCoefYour\n featureCoefTgt['coef'] = - featureCoefTgt['coef']\n featureCoefTgt.features = [ii[:-3] if \"_op\" in ii else ii + '_op' for ii in featureCoefTgt.features]\n\n # Combine only the most important 10 features\n # featureBoth = featureCoefTgt[11:].append(featureCoefYour[11:])\n\n # Combine only all the most important features\n featureBoth = featureCoefTgt.append(featureCoefYour)\n\n # get action recommendations\n # Somehow the pandas here uses a deprecated para cols, instaed of the new one subset\n #featureBoth.drop_duplicates(subset = 'features', take_last = True, inplace = True)\n featureBoth.drop_duplicates(cols = 'features', take_last = True, inplace = True)\n actions, featureImprove = GetActions(featureBoth)\n Imp = 0.1\n oddsNew = ImprovedScore(tgtName, yourName, teamModels, featureImprove, teamTgtPipe, teamYourPipe, Imp)\n\n return odds, oddsNew, actions","def ic_compare(args):\n\n config_file = args.setupfn\n conf_base = os.path.basename(config_file).split('.')[0]\n statfile = os.path.join(args.outputdir,\n \"{}_radvel.stat\".format(conf_base))\n\n status = load_status(statfile)\n\n P, post = radvel.utils.initialize_posterior(config_file,\n decorr=args.decorr)\n\n assert status.getboolean('fit', 'run'), \\\n \"Must perform max-liklihood fit before running Information Criteria comparisons\"\n post = radvel.posterior.load(status.get('fit', 'postfile'))\n\n choices = ['nplanets', 'e', 'trend', 'jit', 'gp']\n statsdictlist = []\n paramlist = []\n compareparams = args.type\n\n ipost = copy.deepcopy(post)\n\n if args.simple:\n statsdictlist += radvel.fitting.model_comp(ipost, params=[], verbose=args.verbose)\n else:\n if hasattr(args, 'fixjitter') and args.fixjitter:\n for param in ipost.params:\n if len(param) >= 3 and param[0:3] == 'jit':\n ipost.params[param].vary = False\n\n for compareparam in compareparams:\n assert compareparam in choices, \\\n \"Valid parameter choices for 'ic -t' are combinations of: \"\\\n + \" \".join(choices)\n paramlist.append(compareparam)\n if hasattr(args, 'mixed') and not args.mixed:\n statsdictlist += radvel.fitting.model_comp(ipost, params=[compareparam], verbose=args.verbose)\n if hasattr(args, 'mixed') and not args.mixed:\n new_statsdictlist = []\n for dicti in statsdictlist:\n anymatch = False\n for seendict in new_statsdictlist:\n if collections.Counter(dicti['Free Params'][0]) == \\\n collections.Counter(seendict['Free Params'][0]):\n anymatch = True\n continue\n if not anymatch:\n new_statsdictlist.append(dicti)\n statsdictlist = new_statsdictlist\n\n if not hasattr(args, 'mixed') or (hasattr(args, 'mixed') and args.mixed):\n statsdictlist += radvel.fitting.model_comp(ipost, params=paramlist, verbose=args.verbose)\n\n savestate = {'ic': statsdictlist}\n save_status(statfile, 'ic_compare', savestate)","def run_test(files, full=False, threshold=2):\n columns = ['detector', 'image', 'common', 'repeat']\n data = []\n\n if full:\n det_s = {**DetectorDescriptor.detectors, **DetectorDescriptor.xdetectors}.keys()\n else:\n det_s = DetectorDescriptor.detectors.keys()\n\n for detector in det_s:\n algo = DetectorDescriptor(detector)\n print(\"Running test {}\".format(detector))\n\n for f in files:\n filename = os.path.basename(f).split('.')[0]\n image = cv2.imread(f, 0)\n kps = algo.detect(image)\n\n h = h_for_file(f)\n if h is None: # This will be the case for the base image (img1)\n baseimg = image\n basepts = kps\n\n data.append([detector, filename, len(basepts), len(basepts)])\n continue\n\n hi = linalg.inv(h)\n mask = create_mask(baseimg.shape, hi)\n\n # Only those that are common\n bpts = []\n for pt in basepts:\n if point_in_mask(pt.pt, mask):\n bpts.append(pt)\n bptst = np.vstack([pt.pt for pt in bpts])\n\n rep = 0\n for point in kps:\n tp = transform_point(point.pt, hi)\n if point_in_mask(tp, mask):\n dists = distance.cdist([tp], bptst)\n if np.min(dists) < threshold:\n rep += 1\n\n data.append([detector, filename, len(bpts), rep])\n\n df = pd.DataFrame(data, columns=columns)\n df['repeatability'] = df['repeat'] / df['common']\n return df","def compare_group_images(directory, show_imgs=True, similarity=\"high\", compression=100):\n # list where the found duplicate/similar images are stored\n duplicates = []\n lower_res = []\n\n imgs_matrix = create_imgs_matrix(directory, compression)\n\n # search for similar images\n if similarity == \"low\":\n ref = 13000\n # search for 1:1 duplicate images\n else:\n ref = 21000\n\n main_img = 0\n compared_img = 1\n nrows, ncols = compression, compression\n srow_A = 0\n erow_A = compression\n srow_B = erow_A\n erow_B = srow_B + compression\n\n while erow_B <= imgs_matrix.shape[0]:\n while compared_img < (len(image_files)):\n # select two images from imgs_matrix\n imgA = imgs_matrix[srow_A: erow_A, # rows\n 0: ncols] # columns\n imgB = imgs_matrix[srow_B: erow_B, # rows\n 0: ncols] # columns\n # compare the images\n rotations = 0\n while image_files[main_img] not in duplicates and rotations <= 3:\n if rotations != 0:\n imgB = rotate_img(imgB)\n err = mse(imgA, imgB)\n print ( \"err:\", err)\n if err <= ref:\n if show_imgs == True:\n show_file_info(compared_img, main_img)\n add_to_list(image_files[main_img], duplicates)\n check_img_quality(directory, image_files[main_img], image_files[compared_img], lower_res)\n rotations += 1\n srow_B += compression\n erow_B += compression\n compared_img += 1\n\n srow_A += compression\n erow_A += compression\n srow_B = erow_A\n erow_B = srow_B + compression\n main_img += 1\n compared_img = main_img + 1\n\n msg = \"\\n***\\n DONE: found \" + str(len(duplicates)) + \" duplicate image pairs in \" + str(\n len(image_files)) + \" total images.\\n The following files have lower resolution:\"\n print(msg)\n return set(lower_res)","def compute_classifications2(depc, gid_list, config=None):\n logger.info('[ibs] Process Image Classifications2')\n logger.info('config = {!r}'.format(config))\n # Get controller\n ibs = depc.controller\n depc = ibs.depc_image\n if config['classifier_two_algo'] in ['cnn']:\n config_ = {\n 'draw_annots': False,\n 'thumbsize': (192, 192),\n }\n # depc.delete_property('thumbnails', gid_list, config=config_)\n thumbnail_list = depc.get_property('thumbnails', gid_list, 'img', config=config_)\n result_list = ibs.generate_thumbnail_class2_list(thumbnail_list, **config)\n elif config['classifier_two_algo'] in ['rf']:\n from wbia.algo.detect.rf import classify\n\n config_ = {'algo': 'resnet'}\n vector_list = depc.get_property('features', gid_list, 'vector', config=config_)\n classifier_weight_filepath = config['classifier_weight_filepath']\n result_list = classify(vector_list, weight_filepath=classifier_weight_filepath)\n elif config['classifier_two_algo'] in ['densenet']:\n from wbia.algo.detect import densenet\n\n config_ = {\n 'draw_annots': False,\n 'thumbsize': (densenet.INPUT_SIZE, densenet.INPUT_SIZE),\n }\n thumbpath_list = ibs.depc_image.get(\n 'thumbnails', gid_list, 'img', config=config_, read_extern=False, ensure=True\n )\n config_ = {\n 'classifier_weight_filepath': config['classifier_two_weight_filepath'],\n }\n result_list = densenet.test(\n thumbpath_list,\n ibs=ibs,\n gid_list=gid_list,\n return_dict=True,\n multiclass=True,\n **config_,\n )\n result_list = list(result_list)\n for index in range(len(result_list)):\n best_score, best_key, scores = result_list[index]\n classes = [best_key]\n result_list[index] = (\n scores,\n classes,\n )\n else:\n raise ValueError(\n 'specified classifier_two algo is not supported in config = {!r}'.format(\n config\n )\n )\n\n # yield detections\n for result in result_list:\n yield result","def doMatching(self):\n\n self.alltrans = []\n curdir = os.getcwd()\n os.chdir(self.astromdir)\n\n for cat in self.projcats:\n self.logfile.write(\"Beginning match attempt for %s\" %cat)\n nobj = max(80,self.Nrefobjs)\n base = 'match %s 1 2 3 %s 1 2 3 identity recalc nobj=%d medtf medsigclip=2.7 '\\\n %(cat,self.GSCmatchin,nobj)\n converged=0\n mrad=4.5\n xsh=0.\n ysh=0.\n Retry=0\n while not converged and mrad<10 and Retry<11:\n ferr,mdict=self._runmatch(base,mrad,xsh,ysh)\n if ferr:\n # match not found, make radius bigger and retry\n mrad += 0.5\n Retry += 1\n self.logfile.write(\"Retry: %d ... set mrad: %s\" %(Retry,mrad))\n continue\n\n dx = mdict['dx']\n dy = mdict['dy']\n sigx = mdict['sigx']\n sigy = mdict['sigy']\n\n # if the extra shift is too big compared to the\n # match radius, update input shift and retry\n if max(abs(dx-xsh),abs(dy-ysh)) > min(0.4, mrad/10.):\n xsh=dx\n ysh=dy\n Retry += 1\n continue\n\n # if sigma seems too big, shrink mrad\n if (max(sigx,sigy) > 0.7 and mrad>2.9):\n mrad -= 0.5\n Retry += 1\n continue\n\n # otherwise, it looks like we have a good fit\n # but we want to tune it up with smaller matchrad\n # Don't increment retry in this final case.\n if mrad > 3.9 or (mrad>2.9 and max(sigx,sigy)>0.4):\n mrad = max(mrad-1,2.5)\n continue\n\n self.alltrans.append(mdict)\n converged=1\n self.logfile.write(\"\"\"%s trans:\n mdx,mdy = %.4f %.4f\n edx,edy = %.4f %.4f\n sigx,sigy = %.4f %.4f\n \"\"\"%(cat, mdict['dx'],mdict['dy'],mdict['dx_err'],mdict['dy_err'],\\\n mdict['sigx'],mdict['sigy']))\n if len(self.alltrans)<1:\n errtxt = \"WARNING: No Image successfully matched against extended GSC!\"\n self.logfile.write(errtxt)\n self.errorList.append((self.modName,errtxt))\n return -1\n\n os.chdir(curdir)\n # finally, choose best match transform and calculate\n # the new CRVAL's\n self._pickBestrans()\n\n return 0","def accuracy():\n\n # Creating a list of populations to analyze over\n cells = np.linspace(min_desired_num_cells, max_desired_num_cells, num_data_points)\n list_of_populations = []\n list_of_fpi = []\n for cell_num in cells:\n population = LineageTree.init_from_parameters(pi, T, E2, cell_num)\n\n # Adding populations into a holder for analysing\n list_of_populations.append([population])\n list_of_fpi.append(pi)\n\n return commonAnalyze(list_of_populations, 2, list_of_fpi=list_of_fpi)","def test_classifier(self):\n \n files = 0\n tp = 0\n fp = 0\n tn = 0\n fn = 0\n \n for testFile in os.listdir(self.testing_dir):\n if os.path.splitext(testFile)[1] in self.valid:\n\n files += 1\n fileName = self.testing_dir + \"/\" + testFile\n\n img = cv2.imread(fileName).astype(np.float32)\n self.Helpers.logger.info(\"Loaded test image \" + fileName)\n \n img = cv2.resize(img, (self.Helpers.confs[\"cnn\"][\"data\"][\"dim\"], \n self.Helpers.confs[\"cnn\"][\"data\"][\"dim\"]))\n img = self.reshape(img)\n \n prediction = self.get_predictions(img)\n \n msg = \"\"\n if prediction == 1 and \"_1.\" in testFile:\n tp += 1\n msg = \"Acute Lymphoblastic Leukemia correctly detected (True Positive)\"\n elif prediction == 1 and \"_0.\" in testFile:\n fp += 1\n msg = \"Acute Lymphoblastic Leukemia incorrectly detected (False Positive)\"\n elif prediction == 0 and \"_0.\" in testFile:\n tn += 1\n msg = \"Acute Lymphoblastic Leukemia correctly not detected (True Negative)\"\n elif prediction == 0 and \"_1.\" in testFile:\n fn += 1\n msg = \"Acute Lymphoblastic Leukemia incorrectly not detected (False Negative)\"\n self.Helpers.logger.info(msg)\n \n self.Helpers.logger.info(\"Images Classifier: \" + str(files))\n self.Helpers.logger.info(\"True Positives: \" + str(tp))\n self.Helpers.logger.info(\"False Positives: \" + str(fp))\n self.Helpers.logger.info(\"True Negatives: \" + str(tn))\n self.Helpers.logger.info(\"False Negatives: \" + str(fn))","def test_compare_classifiers_multiclass_multimetric_vis_api(experiment_to_use):\n experiment = experiment_to_use\n # extract test stats only\n test_stats = experiment.test_stats_full\n viz_outputs = (\"pdf\", \"png\")\n with TemporaryDirectory() as tmpvizdir:\n for viz_output in viz_outputs:\n vis_output_pattern_pdf = tmpvizdir + f\"/*.{viz_output}\"\n visualize.compare_classifiers_multiclass_multimetric(\n [test_stats, test_stats],\n experiment.ground_truth_metadata,\n experiment.output_feature_name,\n top_n_classes=[6],\n model_namess=[\"Model1\", \"Model2\"],\n output_directory=tmpvizdir,\n file_format=viz_output,\n )\n figure_cnt = glob.glob(vis_output_pattern_pdf)\n assert 4 == len(figure_cnt)","def compute_map(scores, num_instances):\r\n pr = []\r\n for i, score in enumerate(scores):\r\n score = score[np.argsort(-score[:, 1])] # sort by confidence score\r\n FP = 0\r\n TP = 0\r\n pr_ = []\r\n for prediction in score:\r\n if prediction[0]:\r\n TP += 1\r\n else:\r\n FP += 1\r\n pr_.append([TP/(TP+FP), (TP/num_instances)])\r\n pr.append(pr_)\r\n pr = np.array(pr)\r\n\r\n pinterps = [] # lists of interpolated precisions for every confidence level\r\n ranks = np.linspace(0, 1, 11)\r\n idxs_interpolations = [] # list of indexes of the interpolated precisions, just to plot the recall\r\n for pr_ in pr:\r\n pinterp = []\r\n idxs = []\r\n last_idx = -1\r\n for rank in ranks:\r\n idx = (np.abs(pr_[:, 1] - rank)).argmin() # find the closest recall to the rank\r\n\r\n if rank > pr_[idx, 1]: # this makes sure we are taking the closest recall at the right of the rank\r\n if idx+1 < pr_[:, 0].shape[0]:\r\n idx += 1\r\n interpolated_precision = np.max(pr_[idx:, 0]) # find the max precision within the interval\r\n if idx == last_idx: # just some checks for when the recall doesn't exist\r\n pinterp[-1] = 0\r\n idxs[-1] = 0\r\n pinterp.append(0)\r\n idxs.append(0)\r\n else:\r\n pinterp.append(interpolated_precision)\r\n idxs.append(idx)\r\n last_idx = idx\r\n pinterps.append(pinterp)\r\n idxs_interpolations.append(idxs)\r\n APs = np.array(pinterps).mean(axis=1) # the AP is the average of the interpolated precisions\r\n mAP = APs.mean() # mAP is the mean of all the APs\r\n\r\n return pr, pinterps, idxs_interpolations, mAP, APs","def score_matches(analysis_matches, args):\n\t\n\tscore_types = ('found_score', 'host_score', 'virus_score')\n\t\n\tfor match_type in analysis_matches.values():\n\t\tfor match in match_type:\n\t\t\t# if read crosses an integration\n\t\t\tif match['intID'] is not None:\n\t\t\t\t# if integration was found, and should have been found\n\t\t\t\tif match['found'] is True:\n\t\t\t\t\tmatch['found_score'] = 'tp'\n\t\t\t\t\t\n\t\t\t\t\t# check if integration was found in the correct place in the host\n\t\t\t\t\tif correct_pos(match, args, ref = 'host'):\n\t\t\t\t\t\tmatch['host_score'] = 'tp'\n\t\t\t\t\telse:\n\t\t\t\t\t\tmatch['host_score'] = 'fp'\t\t\t\t\t\t\t\n\t\t\t\t\t\t\n\t\t\t\t\t# check if integration was found in the correct virus\n\t\t\t\t\tif correct_pos(match, args, ref = 'virus'):\n\t\t\t\t\t\tmatch['virus_score'] = 'tp'\n\t\t\t\t\telse:\n\t\t\t\t\t\tmatch['virus_score'] = 'fp'\n\t\t\t\t\t\n\t\t\t\telse:\n\t\t\t\t\tfor score_type in score_types:\n\t\t\t\t\t\tmatch[score_type] = 'fn'\n\t\t\t\n\t\t\t# if read doesn't cross integration\n\t\t\telse:\n\t\t\t\t# doesn't cross but is found - false positive\n\t\t\t\tif match['found'] is True:\n\t\t\t\t\t\tfor score_type in score_types:\n\t\t\t\t\t\t\tmatch[score_type] = 'fp'\n\t\t\t\t# doesn't cross and isn't found - true netgative\n\t\t\t\telse:\n\t\t\t\t\tfor score_type in score_types:\n\t\t\t\t\t\tmatch[score_type] = 'tn'\n\t\t\t\t\t\n\n\t\t\t\t\n\treturn analysis_matches","def test_run(id_, args, kwargs, expected):\n # Remove results for test.\n results_path = expected['results_path']\n shutil.rmtree(results_path, ignore_errors=True)\n # Start code.\n # [future] attempts to run classification with n_jobs>1\n # global cnfg_default\n # sys.modules['cnfg_default'] = get_params(args[0], 'cnfg_default')\n # import cnfg_default\n # #from cnfg_default import custom_score_metric\n objects = pycnfg.run(oid='default', *args, **kwargs)\n tmp = {k: type(v).__name__ for k, v in objects.items()}\n print('OBJECTS:')\n print(tmp)\n # Compare results:\n # * Compare objects (keys and str of values).\n objects_ = expected['objects']\n objects = {k: type(v).__name__ for k, v in objects.items()}\n if objects != objects_:\n print(set(objects.items()) ^ set(objects_.items()))\n assert objects == objects_\n # for k, v in objects.items():\n # assert k in objects_\n # assert type(v).__name__ == objects_[k]\n # * Compare predictions csv(all available).\n pred_path = glob.glob(f\"{results_path}/models/*_pred.csv\")\n pred_path_ = glob.glob(expected['pred_path'])\n assert len(pred_path) == len(pred_path_)\n for act, exp in zip(sorted(pred_path), sorted(pred_path_)):\n file_diff(act, exp)\n assert filecmp.cmp(act, exp)\n # * Compare test logs.\n logs_path = glob.glob(f\"{results_path}/logs*/*_test.log\")[0]\n logs_path_ = expected['logs_path']\n file_diff(logs_path, logs_path_)\n assert filecmp.cmp(logs_path, logs_path_)\n # * Compare runs dataframe, non-universe columns.\n runs_path = f\"{results_path}/runs\"\n runs_path_ = expected['runs_path']\n df = runs_loader(runs_path)\n df_ = runs_loader(runs_path_)\n # First False/True for each element, then check all by columns.\n # col1 True\n # col2 False\n # dtype: bool\n df_diff = df.eq(df_).all()\n # Column names that are not equal.\n columns = sorted(list(df_diff[df_diff==False].dropna().index))\n # columns_eq = sorted(list(df_diff[df_diff==True].dropna().index))\n columns_ = expected['columns_diff']\n print('DIFF:\\n', columns)\n time.sleep(1)\n # assert columns == columns_\n # * Compare model.\n model_path = glob.glob(f\"{results_path}/models/*.model\")\n model_path_ = glob.glob(expected['model_path'])\n assert len(model_path) == len(model_path_)\n for act, exp in zip(sorted(model_path), sorted(model_path_)):\n assert filecmp.cmp(act, exp)\n return"],"string":"[\n \"def painting_matching_ml(imgs, db_imgs, method_list, text_masks, author_text, gt_text, metrics, weights, splits=30, max_rank=10):\\n descriptor_extractors = [get_descriptor_extractor(method_name) for method_name in method_list]\\n tmp_img_format = []\\n tmp_mask_format = []\\n tmp_text_format = []\\n for i in range(len(imgs)):\\n for j in range(len(imgs[i])):\\n tmp_img_format.append(imgs[i][j])\\n tmp_mask_format.append(text_masks[i][j])\\n tmp_text_format.append(author_text[i][j])\\n\\n #db_imgs = [img[0] for img in db_imgs]\\n db_img_splits = [i*len(db_imgs)//splits for i in range(splits-1)]\\n \\n scores = []\\n query_descriptors = extract_descriptors(tmp_img_format, descriptor_extractors, method_list, tmp_text_format, tmp_mask_format) \\n #np.array([extract_descriptors(img, matching_methods, mask) for img, mask in zip(tmp_img_format, tmp_mask_format)])\\n print(\\\"Starting db extraction + matching\\\")\\n for split in tqdm(range(splits-2)):\\n db_descriptors = extract_descriptors(db_imgs[db_img_splits[split]:db_img_splits[split+1]], descriptor_extractors, method_list, gt_text[db_img_splits[split]:db_img_splits[split+1]], None) #np.array([mrhm(db_img) for db_img in db_imgs[db_img_splits[split]:db_img_splits[split+1]]])\\n scores.append(compare_descriptors(query_descriptors, db_descriptors, metrics, method_list, weights))\\n # compare_descriptors(query_descriptors, db_descriptors, descriptor_comp_methods, descriptor_names, weights)\\n db_descriptors = extract_descriptors(db_imgs[db_img_splits[-1]:], descriptor_extractors, method_list, gt_text[db_img_splits[-1]:], None)\\n scores.append(compare_descriptors(query_descriptors, db_descriptors, metrics, method_list, weights))\\n \\n # concatenate all the results\\n scores = np.concatenate(scores, 1)\\n \\n top_k_matches = np.argpartition(scores, list(range(max_rank)))[:, :max_rank]\\n return top_k_matches\",\n \"def patternMatchingTest(matchingAlgorithm,\\n observations= defaultCOMPREFdump , \\n modelsFolder= defaultWRFdumpsFolder,\\n #key1='0828',\\n key1='',\\n key2='',\\n #key3='',\\n verbose=True,\\n makeSmallImages=True,\\n smallDpi=40,\\n frameDpi = 600 , # added 2013-09-23\\n loadFormerAnalysisResult=False, #added 2013-09-23\\n #summary_folder=summary_folder, # needed if loadFormerAnalysisResult is True\\n summary_folder=summary_folder+timeString + '/', # 2013-09-27\\n matching_algorithm_name=\\\"\\\", # needed if loadFormerAnalysisResult is True\\n panel_cols = 5,\\n panel_rows = 5,\\n useCV2 = False, # switch off opencv since it doesn't work yet\\n useCV = False,\\n averageScoreOrdering=True, # take an average of the ordering to get a total score for the entire model over the time series\\n #kwargs = {}, # keyword arguments for the matchinging algorithm, just in case\\n **kwargs\\n ):\\n if key1 != '' or key2 != '':\\n summary_folder += '_'.join([key1, key2])\\n if not averageScoreOrdering:\\n summary_folder += '_nonaver'\\n summary_folder += '%dx%d/' % (panel_rows, panel_cols)\\n\\n t0 = time.time()\\n if loadFormerAnalysisResult:\\n mv_folder = summary_folder + matching_algorithm_name + '/'\\n result = pickle.load(open(mv_folder + 'result.pydump', 'r'))\\n\\n \\\"\\\"\\\"\\n result= {'best matches' : bestMatchAtEachTime, \\n 'best overall match' : bestMatchOverall, \\n 'all scores' : scores,\\n 'matching algorithm' : matching_algorithm_name,\\n 'dataset1[obs]' : observations,\\n 'dataset2[models]' : modelsFolder,\\n 'time spent' : time.time() - t0,\\n 'mv folder' : mv_folder, #added 2013-09-23\\n 'ordering' : ordering, #added 2013-09-23\\n }\\n \\\"\\\"\\\"\\n bestMatchAtEachTime = result['best matches'] \\n bestMatchOverall = result['best overall match']\\n scores = result['all scores']\\n matching_algorithm_name = result['matching algorithm']\\n observations = result['dataset1[obs]']\\n modelsFolder = result['dataset2[models]']\\n #mv_folder = result['mv folder']\\n ordering = result['ordering']\\n\\n # THE FOLLOWING 30 LINES ARE COPIED FROM \\\"ELSE...\\\" BELOW\\n ########### ******\\n ##\\n #\\n ds1 = pickle.load(open(observations,'r'))\\n ds1.setFloor(0) # added 2013-09-27\\n ds1.list = [v for v in ds1.list if (key1 in v.dataTime or key1 in v.name) \\\\\\n and (key2 in v.dataTime or key2 in v.name)]\\n ds1.list.sort(key=lambda v:v.dataTime)\\n ds1.setVmin(-40) #line added 2013-09-23\\n # debug\\n #print ds1.list\\n print 'loading list with keywords [ %s %s ]' %(key1, key2) \\n ds1.load(verbose=verbose)\\n # end debug\\n modelsList = [v for v in os.listdir(modelsFolder) if '.pydump' in v]\\n modelsList.sort(key = lambda v: int(re.findall(r'\\\\d+',v)[0]))\\n print \\\"modelsList\\\", modelsList\\n scores = {} # scores = {model number: [('dataTime', score), ('dataTime', score),...], ...}\\n print 'loading ds2 (model data)'\\n wrfs = [] # line added 2013-09-23\\n for path in modelsList:\\n modelIndex = int(re.findall(r'\\\\d+', path)[0])\\n ds2 = pickle.load(open(modelsFolder+path, 'r'))\\n ds2.setFloor(0) # added 2013-09-27\\n # debug\\n #ds2.list = [v for v in ds2.list if key1 in v.dataTime] # line redundant, 2013-09-23\\n ds1.list, ds2.list = ds1.intersect(ds2) #line added 2013-09-23 \\n ds2.setVmin(-40) #line added 2013-09-23\\n wrfs.append(ds2) # line added 2013-09-23\\n # end debug\\n\\n ####################\\n\\n\\n else:\\n # THE FOLLOWING 30 LINES ARE COPIED TO \\\"IF...\\\" ABOVE\\n ########### ******\\n ##\\n #\\n ds1 = pickle.load(open(observations,'r'))\\n ds1.list = [v for v in ds1.list if (key1 in v.dataTime or key1 in v.name) \\\\\\n and (key2 in v.dataTime or key2 in v.name)]\\n ds1.list.sort(key=lambda v:v.dataTime)\\n ds1.setVmin(-40) #line added 2013-09-23\\n # debug\\n #print ds1.list\\n print 'loading list with keywords [ %s %s ]' %(key1, key2) \\n ds1.load(verbose=verbose)\\n ds1.setFloor(0) # added 2013-09-27\\n # end debug\\n modelsList = [v for v in os.listdir(modelsFolder) if '.pydump' in v]\\n modelsList.sort(key = lambda v: int(re.findall(r'\\\\d+',v)[0]))\\n print \\\"modelsList\\\", modelsList\\n scores = {} # scores = {model number: [('dataTime', score), ('dataTime', score),...], ...}\\n print 'loading ds2 (model data)'\\n wrfs = [] # line added 2013-09-23\\n ds1.commonMaskSet=False # tempory attribute added 2013-09-27\\n for path in modelsList:\\n modelIndex = int(re.findall(r'\\\\d+', path)[0])\\n ds2 = pickle.load(open(modelsFolder+path, 'r'))\\n # debug\\n #ds2.list = [v for v in ds2.list if key1 in v.dataTime] # line redundant, 2013-09-23\\n ds1.list, ds2.list = ds1.intersect(ds2) #line added 2013-09-23 \\n ds2.setFloor(0) # added 2013-09-27\\n ds2[0].matrix.mask += ds1[0].matrix.mask # added 2013-09-27\\n ds2.setCommonMask() # added 2013-09-27\\n if not ds1.commonMaskSet: # added 2013-09-27\\n ds1[0].matrix.mask +=ds2[0].matrix.mask # added 2013-09-27\\n ds1.setCommonMask() # added 2013-09-27\\n ds2.setVmin(-40) #line added 2013-09-23\\n wrfs.append(ds2) # line added 2013-09-23\\n # end debug\\n\\n #######################################################################################################################\\n # key lines\\n print 'computing scores between %s and %s ...' %(ds1.name, ds2.name)\\n s, matching_algorithm_name = matchingAlgorithm(ds1, ds2, **kwargs) # list of the form [('20120612.0000', 0.9),...] \\n scores[modelIndex] = dict(s) # { model index: {'dataTime': score,...}, ...}\\n # debug\\n if verbose:\\n print scores\\n print '..................................................'\\n #\\n ########################################################################################################################\\n\\n print 'sorting the results...'\\n #print 'scores', '\\\\n', scores #debug\\n # bestMatchAtEachTime\\n T1 = [v.dataTime for v in ds1]\\n T2 = [v.dataTime for v in ds2] #any ds2 will do - they are the same, in this case we pick the last one \\n T = set(T1).intersection(T2)\\n dataTimeList = sorted(list(T))\\n\\n #print 'dataTimeList:', dataTimeList #debug\\n highestScores = []\\n ordering = [] #added 2013-09-23\\n\\n if averageScoreOrdering: # take the average scores before ordering to make it uniform\\n ave_scores = [ (v, np.mean( [scores[v].values() ] ) ) for v in scores.keys()]\\n ave_scores = dict(ave_scores)\\n print '..taking averages...............................'\\n print ave_scores\\n for u in scores.keys():\\n for t in scores[u].keys():\\n scores[u][t] = ave_scores[u]\\n print '\\\\n'.join([str(scores[v]) for v in scores.keys()])\\n\\n for t in dataTimeList:\\n print t\\n highestScoreAt_t = max([scores[u][t] for u in scores.keys()]) # u = model index (integer 1-20), t = dataTime (string)\\n new_entry = (t, highestScoreAt_t, [u for u in scores.keys() if scores[u][t]==highestScoreAt_t])\\n print new_entry\\n highestScores.append(new_entry)\\n #added 2013-09-23\\n ordering.append([0]+sorted(range(1, len(modelsList)+1), key=lambda v: scores[v][t], reverse=True))\\n\\n\\n\\n\\n # bestMatchOverall\\n print scores # debug\\n averageScores = [(Model, np.mean(scores[Model].values())) for Model in scores.keys()] # list of [(Model, mean score),...]\\n highestScore = max([v[1] for v in averageScores])\\n bestMatch = [u[0] for u in averageScores if u[1]==highestScore] \\n # debug\\n print 'average scores , highest average, best match', averageScores, highestScore, bestMatch\\n #xx= raw_input('press enter') \\n\\n\\n bestMatchAtEachTime = highestScores\\n bestMatchOverall = (highestScore, bestMatch)\\n\\n # the following lines do not work well since oftentimes we have multiple pointers (names) to a single object/function\\n # and so they are replaced by returning the function name in the function itself\\n #allvars = vars()\\n #matching_algorithm_name = [v for v in allvars if allvars[v]==matchingAlgorithm][0]\\n #mv_folder = summary_folder + matching_algorithm_name + '/'\\n mv_folder = summary_folder + matching_algorithm_name + '/' # added 2013-09-27\\n\\n result= {'best matches' : bestMatchAtEachTime, \\n 'best overall match' : bestMatchOverall, \\n 'all scores' : scores,\\n 'matching algorithm' : matching_algorithm_name,\\n 'dataset1[obs]' : observations,\\n 'dataset2[models]' : modelsFolder,\\n 'time spent' : time.time() - t0,\\n 'mv folder' : mv_folder, #added 2013-09-23\\n 'ordering' : ordering, #added 2013-09-23\\n }\\n\\n try:\\n os.makedirs(mv_folder)\\n print 'folder', mv_folder, \\\"created!\\\"\\n except:\\n print 'folder', mv_folder, 'exists!'\\n pickle.dump(result, open(mv_folder+'result.pydump','w'))\\n\\n ############################################################################\\n # to make the videos/collected images\\n # calling armor.dataStreamTools.makeVideo.makeVideo()\\n # 2013-09-23\\n\\n from . import makeVideo as mv\\n \\n try:\\n os.makedirs(mv_folder)\\n print 'folder', mv_folder, \\\"created!\\\"\\n except:\\n print 'folder', mv_folder, 'exists!'\\n # make images\\n if makeSmallImages:\\n print 'making small images'\\n for ds in [ds1] + wrfs:\\n #ds.imageFolder= mv_folder + ds.name + '/'\\n ds.imageFolder= summary_root + ds.name + '/' #added 2013-09-27 to replace the above line - too tired waiting the plots\\n ds.saveImages(drawCoast=True, flipud=True, dpi=smallDpi)\\n print ds.imageFolder\\n else:\\n # just set the paths\\n for ds in [ds1]+wrfs:\\n ds.imageFolder = summary_root + ds.name +'/' # /home/k/ARMOR/data/KONG-REY/summary/WRF01.DBZ/\\n for dbzPattern in ds.list:\\n dbzPattern.imagePath = ds.imageFolder + ds.name + dbzPattern.dataTime + '.png'\\n print dbzPattern.imagePath\\n #ds.setImagePaths() # doesn't for for old pickled objects???\\n\\n\\n # make panels\\n\\n mv.makeVideo(DSS = [ds1]+ wrfs, # [ds0, ds1, ds2, ds3, ds4, ...], a list of armor.pattern.DBZstream objects\\n panel_cols = panel_cols, # number of colums in the panel\\n panel_rows = panel_rows, # no need to be filled\\n #fourcc = cv.CV_FOURCC('F', 'L', 'V', '1'),\\n #fps = defaultFps,\\n extension= '.avi',\\n #fourcc = cv.CV_FOURCC('P', 'I', 'M', '1'),\\n outputFileName =\\\"\\\",\\n outputFolder= mv_folder,\\n sandbox = mv_folder ,\\n saveFrames = True, # saving the frames as images\\n frameDpi = frameDpi,\\n useCV2 = useCV2,\\n useCV = useCV,\\n ordering = ordering, # ordering of the models\\n )\\n ############################################################################\\n # dumping the results\\n try:\\n os.makedirs(summary_folder+ 'pydumps/')\\n except:\\n print 'Folder exists! -- ', summary_folder+ 'pydumps/' \\n\\n pickle.dump(result, open(summary_folder+ 'pydumps/results_' + matching_algorithm_name+ '+' +\\\\\\n key1 + '+' + key2 + str(int(time.time())) + '.pydump', 'w'))\\n\\n \\n\\n print \\\"time spent:\\\", time.time()-t0\\n return result\",\n \"def compare(model, input1, input2): \\n patch_image_time_start = time.time()\\n input1_patches = get_patches_non_overlap(input1, 48, 48)\\n input2_patches = get_patches_non_overlap(input2, 48, 48)\\n patch_image_time_end = time.time()\\n compare_image_time_start = time.time()\\n pred = model.predict([input1_patches, input2_patches])\\n compare_image_time_end = time.time()\\n\\n result[\\\"patch_retrieval_time\\\"].append(patch_image_time_end - patch_image_time_start)\\n result[\\\"image_comparison_time\\\"].append(compare_image_time_end - compare_image_time_start)\\n\\n return np.sum(pred)\",\n \"def ransac(matches, kp1, kp2, s=4, threshold=3, maxIterations=2000, returnMatches=False, inlierRatio=0.05, ransacRatio=0.6):\\n\\n sizes_kp1 = [kp1[dt[0].queryIdx].size for dt in matches]\\n sizes_kp2 = [kp1[dt[0].trainIdx].size for dt in matches]\\n tup_matches_kp1 = [kp1[dt[0].queryIdx].pt for dt in matches]\\n tup_matches_kp2 = [kp2[dt[0].trainIdx].pt for dt in matches]\\n matches_kp1 = np.array([[h for h in kp] + [1] for kp in tup_matches_kp1])\\n matches_kp2 = np.array([[h for h in kp] + [1] for kp in tup_matches_kp2])\\n\\n cnt_matches = len(matches)\\n\\n max_matches = []\\n max_p1, max_p2 = [], []\\n max_p1_sizes, max_p2_sizes = [], []\\n max_total = 0\\n\\n for iter in range(maxIterations):\\n # Find Homography based on random sample\\n data = random.sample(matches, s)\\n data_p1 = np.array([matches_kp1[dt[0].queryIdx] for dt in data])\\n data_p2 = np.array([matches_kp2[dt[0].trainIdx] for dt in data])\\n homography = homomat(data_p1[:, :2], data_p2[:, :2])\\n\\n # Find P1 projection from the homography matrix\\n projected_p2 = np.dot(homography, matches_kp1.transpose())\\n projected_p2 = projected_p2[0:3] / projected_p2[2] # make sure w' is 1\\n projected_p2 = projected_p2.transpose()\\n\\n # Initialize Current Matches\\n current_matches = []\\n current_p1, current_p2 = [], []\\n current_p1_sizes, current_p2_sizes = [], []\\n current_total = 0\\n\\n # Check for inliers and outliers for each matches\\n for i, (match) in enumerate(matches):\\n # normalize the error\\n error = np.linalg.norm(matches_kp2[i] - projected_p2[i])\\n\\n # Check for inliers\\n if error < threshold:\\n current_matches.append([cv.DMatch(current_total, current_total, match[0].distance)])\\n current_p1.append(matches_kp1[i][0:2])\\n current_p2.append(matches_kp2[i][0:2])\\n current_p1_sizes.append(sizes_kp1[i])\\n current_p2_sizes.append(sizes_kp2[i])\\n current_total += 1\\n\\n # If\\n if current_total > max_total and current_total >= np.round(inlierRatio*cnt_matches):\\n max_matches = current_matches\\n max_p1 = current_p1\\n max_p2 = current_p2\\n max_p1_sizes = current_p1_sizes\\n max_p2_sizes = current_p2_sizes\\n max_total = current_total\\n\\n # # we are done in case we have enough inliers\\n if current_total > cnt_matches * ransacRatio:\\n break\\n\\n\\n # Re-evaluate the Homography based on the best inliers\\n max_homography = homomat(np.array(max_p1), np.array(max_p2))\\n\\n if returnMatches:\\n max_kp1 = [cv.KeyPoint(d[0], d[1], max_p1_sizes[i]) for i, d in enumerate(max_p1)]\\n max_kp2 = [cv.KeyPoint(d[0], d[1], max_p2_sizes[i]) for i, d in enumerate(max_p2)]\\n return max_homography, max_matches, max_kp1, max_kp2\\n\\n return max_homography\",\n \"def findMatchesBetweenImages(image_1, image_2):\\n # matches - type: list of cv2.DMath\\n matches = None\\n # image_1_kp - type: list of cv2.KeyPoint items.\\n image_1_kp = None\\n # image_1_desc - type: numpy.ndarray of numpy.uint8 values.\\n image_1_desc = None\\n # image_2_kp - type: list of cv2.KeyPoint items.\\n image_2_kp = None\\n # image_2_desc - type: numpy.ndarray of numpy.uint8 values.\\n image_2_desc = None\\n # WRITE YOUR CODE HERE.\\n\\n sift = cv2.ORB_create()\\n image_1_kp, image_1_desc = sift.detectAndCompute(image_1, None)\\n image_2_kp, image_2_desc = sift.detectAndCompute(image_2, None)\\n\\n # bf = cv2.BFMatcher(cv2.NORM_HAMMING, crossCheck=True)\\n # matches = bf.match(image_1_desc, image_2_desc)\\n # matches = sorted(matches, key = lambda x:x.distance)\\n # matches = matches[:10]\\n\\n bf = cv2.BFMatcher()\\n matches = bf.knnMatch(image_1_desc, image_2_desc, k=2)\\n\\n # Apply ratio test\\n good = []\\n for m, n in matches:\\n print(m.distance, n.distance, m.distance < .75*n.distance)\\n if m.distance < (0.75 * n.distance):\\n good.append([m])\\n\\n # We coded the return statement for you. You are free to modify it -- just\\n # make sure the tests pass.\\n print(len(good), good)\\n return image_1_kp, image_2_kp, matches\\n # END OF FUNCTION.\",\n \"def compute_precision_recall_bbox(per_image_detections, per_image_gts, num_gt_classes,\\n matching_iou_threshold=0.5):\\n per_image_eval = per_image_evaluation.PerImageEvaluation(\\n num_groundtruth_classes=num_gt_classes,\\n matching_iou_threshold=matching_iou_threshold,\\n nms_iou_threshold=1.0,\\n nms_max_output_boxes=10000\\n )\\n\\n print('Running per-object analysis...')\\n\\n detection_tp_fp = defaultdict(list) # key is the category; in each list, 1 is tp, 0 is fp\\n detection_scores = defaultdict(list)\\n num_total_gt = defaultdict(int)\\n\\n for image_id, dets in tqdm(per_image_detections.items()):\\n\\n detected_boxes = np.array(dets['boxes'], dtype=np.float32)\\n detected_scores = np.array(dets['scores'], dtype=np.float32)\\n # labels input to compute_object_detection_metrics() needs to start at 0, not 1\\n detected_labels = np.array(dets['labels'], dtype=np.int) - 1 # start at 0\\n # num_detections = len(dets['boxes'])\\n\\n gts = per_image_gts[image_id]\\n gt_boxes = np.array(gts['gt_boxes'], dtype=np.float32)\\n gt_labels = np.array(gts['gt_labels'], dtype=np.int) - 1 # start at 0\\n num_gts = len(gts['gt_boxes'])\\n\\n groundtruth_is_difficult_list = np.zeros(num_gts, dtype=bool) # place holders - we don't have these\\n groundtruth_is_group_of_list = np.zeros(num_gts, dtype=bool)\\n\\n # to prevent 'Invalid dimensions for box data.' error\\n if num_gts == 0:\\n # this box will not match any detections\\n gt_boxes = np.array([[0, 0, 0, 0]], dtype=np.float32)\\n\\n scores, tp_fp_labels, is_class_correctly_detected_in_image = (\\n per_image_eval.compute_object_detection_metrics(\\n detected_boxes=detected_boxes,\\n detected_scores=detected_scores,\\n detected_class_labels=detected_labels,\\n groundtruth_boxes=gt_boxes,\\n groundtruth_class_labels=gt_labels,\\n groundtruth_is_difficult_list=groundtruth_is_difficult_list,\\n groundtruth_is_group_of_list=groundtruth_is_group_of_list\\n )\\n )\\n\\n for i, tp_fp_labels_cat in enumerate(tp_fp_labels):\\n assert sum(tp_fp_labels_cat) <= sum(gt_labels == i) # true positives < gt of that category\\n cat = i + 1 # categories start at 1\\n detection_tp_fp[cat].append(tp_fp_labels_cat)\\n detection_scores[cat].append(scores[i])\\n num_total_gt[cat] += sum(gt_labels == i) # gt_labels start at 0\\n\\n all_scores = []\\n all_tp_fp = []\\n\\n print('Computing precision recall for each category...')\\n per_cat_metrics = {}\\n for i in range(num_gt_classes):\\n cat = i + 1\\n scores_cat = np.concatenate(detection_scores[cat])\\n tp_fp_cat = np.concatenate(detection_tp_fp[cat]).astype(np.bool)\\n all_scores.append(scores_cat)\\n all_tp_fp.append(tp_fp_cat)\\n\\n precision, recall = metrics.compute_precision_recall(\\n scores_cat, tp_fp_cat, num_total_gt[cat]\\n )\\n average_precision = metrics.compute_average_precision(precision, recall)\\n\\n per_cat_metrics[cat] = {\\n 'category': cat,\\n 'precision': precision,\\n 'recall': recall,\\n 'average_precision': average_precision,\\n 'scores': scores_cat,\\n 'tp_fp': tp_fp_cat,\\n 'num_gt': num_total_gt[cat]\\n }\\n print('Number of ground truth in category {} is {}'.format(cat, num_total_gt[cat]))\\n\\n # compute one-class precision/recall/average precision (if every box is just of an object class)\\n all_scores = np.concatenate(all_scores)\\n all_tp_fp = np.concatenate(all_tp_fp)\\n overall_gt_count = sum(num_total_gt.values())\\n\\n one_class_prec, one_class_recall = metrics.compute_precision_recall(\\n all_scores, all_tp_fp, overall_gt_count\\n )\\n one_class_average_precision = metrics.compute_average_precision(one_class_prec, one_class_recall)\\n\\n per_cat_metrics['one_class'] = {\\n 'category': 'one_class',\\n 'precision': one_class_prec,\\n 'recall': one_class_recall,\\n 'average_precision': one_class_average_precision,\\n 'scores': all_scores,\\n 'tp_fp': all_tp_fp,\\n 'num_gt': overall_gt_count\\n }\\n\\n return per_cat_metrics\",\n \"def evaluate_matrices(self, precision_matrix, recall_matrix):\\n\\t\\tground_truth_count = precision_matrix.shape[0]\\n\\t\\tdetection_count = precision_matrix.shape[1]\\n\\t\\tground_truth_sets_precision = defaultdict(set) # number of ground truth items that match a particular detection in the precision matrix\\n\\t\\tdetection_sets_precision = defaultdict(set) # number of detection items that match a particular ground truth in the precision matrix\\n\\t\\tground_truth_sets_recall = defaultdict(set) # number of ground truth items that match a particular detection in the recall matrix\\n\\t\\tdetection_sets_recall = defaultdict(set) # number of detection items that match a particular ground truth in the recall matrix\\n\\n\\t\\tfor gt_index in range(ground_truth_count):\\n\\t\\t\\tfor det_index in range(detection_count):\\n\\t\\t\\t\\tif precision_matrix[gt_index, det_index] >= self.precision_threshold:\\n\\t\\t\\t\\t\\tground_truth_sets_precision[det_index].add(gt_index)\\n\\t\\t\\t\\t\\tdetection_sets_precision[gt_index].add(det_index)\\n\\t\\t\\t\\tif recall_matrix[gt_index, det_index] >= self.recall_threshold:\\n\\t\\t\\t\\t\\tground_truth_sets_recall[det_index].add(gt_index)\\n\\t\\t\\t\\t\\tdetection_sets_recall[gt_index].add(det_index)\\n\\n\\t\\tmatch_ground_truth = 0. # sum of MatchG\\n\\t\\tmatch_detection = 0. # sum of MatchD\\n\\n\\t\\tone_to_one_precision = set()\\n\\t\\tfor gt_index in detection_sets_precision:\\n\\t\\t\\tmatching_detections_precision = detection_sets_precision[gt_index]\\n\\t\\t\\tif len(matching_detections_precision) == 1:\\n\\t\\t\\t\\t(detection_precision, ) = matching_detections_precision\\n\\t\\t\\t\\tif len(ground_truth_sets_precision[detection_precision]) == 1:\\n\\t\\t\\t\\t\\tone_to_one_precision.add((gt_index, detection_precision))\\n\\t\\t\\telse:\\n\\t\\t\\t\\t# one-to-many (one ground truth to many detections)\\n\\t\\t\\t\\tgt_sum = 0.\\n\\t\\t\\t\\tfor detection_precision in matching_detections_precision:\\n\\t\\t\\t\\t\\tgt_sum += recall_matrix[gt_index, detection_precision]\\n\\t\\t\\t\\tif gt_sum >= self.recall_threshold:\\n\\t\\t\\t\\t\\t#print(\\\"1:N ~ GT {} : DT {}\\\".format(gt_index,matching_detections_precision))\\n\\t\\t\\t\\t\\tmatch_ground_truth += self.scatter_punishment(matching_detections_precision)\\n\\t\\t\\t\\t\\tmatch_detection += len(matching_detections_precision) * self.scatter_punishment(matching_detections_precision)\\n\\n\\t\\tone_to_one_recall = set()\\n\\t\\tfor det_index in ground_truth_sets_recall:\\n\\t\\t\\tmatching_ground_truths_recall = ground_truth_sets_recall[det_index]\\n\\t\\t\\tif len(matching_ground_truths_recall) == 1:\\n\\t\\t\\t\\t(ground_truth_recall, ) = matching_ground_truths_recall\\n\\t\\t\\t\\tif len(detection_sets_recall[ground_truth_recall]) == 1:\\n\\t\\t\\t\\t\\tone_to_one_recall.add((ground_truth_recall, det_index))\\n\\t\\t\\telse:\\n\\t\\t\\t\\t# many-to-one (many ground truths covered by one detection)\\n\\t\\t\\t\\tdet_sum = 0\\n\\t\\t\\t\\tfor ground_truth_recall in matching_ground_truths_recall:\\n\\t\\t\\t\\t\\tdet_sum += precision_matrix[ground_truth_recall, det_index]\\n\\t\\t\\t\\tif det_sum >= self.precision_threshold:\\n\\t\\t\\t\\t\\t#print(\\\"N:1 ~ DT {} : GT {}\\\".format(det_index,matching_ground_truths_recall))\\n\\t\\t\\t\\t\\tmatch_detection += self.scatter_punishment(matching_ground_truths_recall)\\n\\t\\t\\t\\t\\tmatch_ground_truth += len(matching_ground_truths_recall) * self.scatter_punishment(matching_ground_truths_recall)\\n\\n\\t\\tone_to_one_matches = one_to_one_precision & one_to_one_recall\\n\\t\\tmatch_ground_truth += len(one_to_one_matches)\\n\\t\\tmatch_detection += len(one_to_one_matches)\\n\\n\\t\\trecall = match_ground_truth / float(ground_truth_count)\\n\\t\\tprecision = match_detection / float(detection_count)\\n\\t\\treturn (precision, recall, (match_detection, float(detection_count)), (match_ground_truth, float(ground_truth_count)))\",\n \"def match(image1,image2,threshold,useRansac=False,t_orientation=30,t_scale=0.5):\\r\\n im1, keypoints1, descriptors1 = ReadKeys(image1)\\r\\n im2, keypoints2, descriptors2 = ReadKeys(image2)\\r\\n #\\r\\n # REPLACE THIS CODE WITH YOUR SOLUTION (ASSIGNMENT 5, QUESTION 3)\\r\\n #\\r\\n # Generate five random matches (for testing purposes)\\r\\n # matched_pairs = []\\r\\n # num = 5\\r\\n # for i in range(num):\\r\\n # matched_pairs.append([keypoints1[i],keypoints2[i]])\\r\\n # return DisplayMatches(im1, im2, matched_pairs)\\r\\n\\r\\n # END OF SECTION OF CODE TO REPLACE\\r\\n #\\r\\n\\r\\n #q3\\r\\n matched_pairs = []\\r\\n between_angles = np.arccos(np.dot(descriptors1, np.transpose(descriptors2)))\\r\\n for i, row in enumerate(between_angles):\\r\\n \\tratio = sorted(row)[0] / sorted(row)[1]\\r\\n \\tif ratio <= threshold:\\r\\n\\t \\tmatched_pairs.append([keypoints1[i], keypoints2[np.where(row == sorted(row)[0])[0][0]]])\\r\\n # print(matched_pairs)\\r\\n if useRansac is False:\\r\\n return DisplayMatches(im1, im2, matched_pairs)\\r\\n\\t# \\r\\n\\r\\n #q4\\r\\n repetition = 10\\r\\n subsets = [[]] * repetition\\r\\n for i in range(repetition):\\r\\n r = random.randint(0, len(matched_pairs))\\r\\n for match in matched_pairs:\\r\\n ds1, ds2 = matched_pairs[r][1][2]/matched_pairs[r][0][2], match[1][2]/match[0][2]\\r\\n do1, do2 = (matched_pairs[r][1][3]-matched_pairs[r][0][3]), (match[1][3]-match[0][3])\\r\\n if abs(ds2 - ds1) <= t_scale * ds1 and abs(do2 - do1) % (2 * math.pi) <= t_orientation:\\r\\n subsets[i].append(match)\\r\\n\\r\\n max_i, max_len = 0, subsets[0]\\r\\n for i in range(10):\\r\\n l = len(subsets[i])\\r\\n if l > max_len:\\r\\n max_len = l\\r\\n max_i = i\\r\\n\\r\\n im3 = DisplayMatches(im1, im2, subsets[max_i])\\r\\n return im3\",\n \"def evaluate_detections(ground_truth, predictions, class_name, overlap_threshold=0.5,\\n allow_multiple_matches_per_ignored=True,\\n verbose=True):\\n\\n Detection = namedtuple('Detection', ['image', 'bbox', 'score', 'gt_match'])\\n GT = namedtuple('GroundTruth', ['bbox', 'is_matched', 'is_ignored'])\\n detections = [Detection(image=img_pred.image_path,\\n bbox=np.array(obj_pred[\\\"bbox\\\"]),\\n score=obj_pred.get(\\\"score\\\", 0.0),\\n gt_match=-1)\\n for img_pred in predictions\\n for obj_pred in img_pred\\n if obj_pred[\\\"type\\\"] == class_name]\\n\\n scores = np.array([detection.score for detection in detections])\\n sorted_ind = np.argsort(-scores)\\n detections = [detections[i] for i in sorted_ind]\\n\\n gts = {}\\n for img_gt in ground_truth:\\n gts[img_gt.image_path] = GT(\\n bbox=np.vstack([np.array(obj_gt[\\\"bbox\\\"]) for obj_gt in img_gt]) if img_gt else np.empty(\\n (0, 4)),\\n is_matched=np.zeros(len(img_gt), dtype=bool),\\n is_ignored=np.array([obj_gt.get(\\\"is_ignored\\\", False) for obj_gt in img_gt], dtype=bool))\\n\\n detections_num = len(detections)\\n true_pos = np.zeros(detections_num)\\n false_pos = np.zeros(detections_num)\\n\\n for i, detection in tqdm(enumerate(detections), desc=\\\"Processing detections\\\",\\n disable=not verbose):\\n image_path = detection.image\\n bboxes_gt = gts[image_path].bbox\\n bbox = detection.bbox\\n max_overlap = -np.inf\\n\\n if bboxes_gt is not None and bboxes_gt.shape[0] > 0:\\n intersection_xmin = np.maximum(bboxes_gt[:, 0], bbox[0])\\n intersection_ymin = np.maximum(bboxes_gt[:, 1], bbox[1])\\n intersection_xmax = np.minimum(bboxes_gt[:, 0] + bboxes_gt[:, 2], bbox[0] + bbox[2])\\n intersection_ymax = np.minimum(bboxes_gt[:, 1] + bboxes_gt[:, 3], bbox[1] + bbox[3])\\n intersection_width = np.maximum(intersection_xmax - intersection_xmin, 0.)\\n intersection_height = np.maximum(intersection_ymax - intersection_ymin, 0.)\\n intersection = intersection_width * intersection_height\\n\\n det_area = bbox[2] * bbox[3]\\n gt_area = bboxes_gt[:, 2] * bboxes_gt[:, 3]\\n union = (det_area + gt_area - intersection)\\n ignored_mask = gts[image_path].is_ignored\\n if allow_multiple_matches_per_ignored:\\n if np.any(ignored_mask):\\n union[ignored_mask] = det_area\\n\\n overlaps = intersection / union\\n # Match not ignored ground truths first.\\n if np.any(~ignored_mask):\\n overlaps_filtered = np.copy(overlaps)\\n overlaps_filtered[ignored_mask] = 0.0\\n max_overlap = np.max(overlaps_filtered)\\n argmax_overlap = np.argmax(overlaps_filtered)\\n # If match with non-ignored ground truth is not good enough,\\n # try to match with ignored ones.\\n if max_overlap < overlap_threshold and np.any(ignored_mask):\\n overlaps_filtered = np.copy(overlaps)\\n overlaps_filtered[~ignored_mask] = 0.0\\n max_overlap = np.max(overlaps_filtered)\\n argmax_overlap = np.argmax(overlaps_filtered)\\n detections[i] = detection._replace(gt_match=argmax_overlap)\\n\\n if max_overlap >= overlap_threshold:\\n if not gts[image_path].is_ignored[argmax_overlap]:\\n if not gts[image_path].is_matched[argmax_overlap]:\\n true_pos[i] = 1.\\n gts[image_path].is_matched[argmax_overlap] = True\\n else:\\n false_pos[i] = 1.\\n elif not allow_multiple_matches_per_ignored:\\n gts[image_path].is_matched[argmax_overlap] = True\\n else:\\n false_pos[i] = 1.\\n\\n false_pos = np.cumsum(false_pos)\\n true_pos = np.cumsum(true_pos)\\n\\n debug_visualization = False\\n if debug_visualization:\\n for image_path, bboxes_gt in gts.items():\\n\\n print(image_path)\\n image = cv2.imread(image_path)\\n image_gt = np.copy(image)\\n for bbox in bboxes_gt.bbox:\\n cv2.rectangle(image_gt, tuple(bbox[:2]), tuple(bbox[2:] + bbox[:2]),\\n color=(255, 255, 0), thickness=2)\\n cv2.imshow(\\\"gt\\\", image_gt)\\n for detection in detections:\\n if detection.image != image_path:\\n continue\\n bbox = detection.bbox\\n cv2.rectangle(image, tuple(bbox[:2]), tuple(bbox[2:] + bbox[:2]), color=(0, 255, 0),\\n thickness=2)\\n if detection.gt_match is not None:\\n bbox = bboxes_gt.bbox[detection.gt_match]\\n cv2.rectangle(image, tuple(bbox[:2]), tuple(bbox[2:] + bbox[:2]),\\n color=(0, 0, 255), thickness=1)\\n cv2.imshow(\\\"image\\\", image)\\n cv2.waitKey(0)\\n\\n # Handle equal-score detections.\\n # Get index of the last occurrence of a score.\\n ind = len(scores) - np.unique(scores[sorted_ind[::-1]], return_index=True)[1] - 1\\n ind = ind[::-1]\\n # Though away redundant points.\\n false_pos = false_pos[ind]\\n true_pos = true_pos[ind]\\n\\n total_positives_num = np.sum([np.count_nonzero(~gt.is_ignored) for gt in gts.values()])\\n recall = true_pos / float(total_positives_num)\\n # Avoid divide by zero in case the first detection matches an ignored ground truth.\\n precision = true_pos / np.maximum(true_pos + false_pos, np.finfo(np.float64).eps)\\n miss_rate = 1.0 - recall\\n fppi = false_pos / float(len(gts))\\n\\n return recall, precision, miss_rate, fppi\",\n \"def _match_keypoints(self, query_image: Image) -> typing.Tuple[typing.List[int], typing.List[typing.Any]]:\\n query_image.get_keypoints_and_descriptors()\\n\\n # matches = self.matcher.match(query_image.descriptor)\\n matches = self.matcher.knnMatch(query_image.descriptor, k=2)\\n good = []\\n for m_n in matches:\\n if len(m_n) == 1:\\n good.append(m_n[0])\\n continue\\n elif len(m_n) != 2:\\n continue\\n (m, n) = m_n\\n if m.distance < 0.7 * n.distance:\\n good.append(m)\\n\\n images_scores = [0] * len(self._candidate_images)\\n images_matches = [None] * len(self._candidate_images)\\n for image_index, image in enumerate(self._candidate_images):\\n matches_scores = []\\n matches = []\\n for i, match in enumerate(good):\\n if match.imgIdx != image_index:\\n continue\\n matches.append(match)\\n matches_scores.append((256 - match.distance) / 256)\\n\\n match_cnt = len(matches_scores)\\n if match_cnt <= 0:\\n continue\\n\\n images_scores[image_index] = (\\n 0.5 + ((math.tanh(match_cnt / 3 - 1)) / 2)) * (sum(matches_scores) / match_cnt)\\n\\n images_matches[image_index] = matches\\n\\n return images_scores, images_matches\",\n \"def analyze_similarities():\\r\\n print('Total number of candidate pairs:', len(pairs))\\r\\n print(f'\\\\nNumber of actual item pairs in the train set: {pairs[\\\"true_match\\\"].sum()}\\\\n')\\r\\n\\r\\n for feature in ['text_score', 'image_score', 'txt_img_score', 'words_ratio', 'txt_img_words']:\\r\\n\\r\\n # Check distribution of True and False predictions for various similarity scores\\r\\n print('-' * 50)\\r\\n print(f'\\\\nDistribution of True/False predictions for {feature}')\\r\\n for thr in (0.6, 0.65, 0.7, 0.75, 0.8, 0.85, 0.9, 0.95):\\r\\n print('-' * 50)\\r\\n print(f'Similarity score over {thr}')\\r\\n pairs_sample = pairs[pairs[feature] >= thr]\\r\\n print(f'Number of similar item pairs: {len(pairs_sample)}')\\r\\n print(pairs_sample['true_match'].value_counts(normalize=True))\\r\\n\\r\\n # Check if identical phash can be used to improve the accuracy\\r\\n same_phash = pairs[pairs['phash_match'] == 1]\\r\\n different_phash = pairs[pairs['phash_match'] == 0]\\r\\n\\r\\n print('\\\\nFor item pairs with the same phash:')\\r\\n print(same_phash['true_match'].value_counts(normalize=True))\\r\\n print('Number of item pairs in this subset:', len(same_phash))\\r\\n\\r\\n print('\\\\nFor item pairs with different phash:')\\r\\n print(different_phash['true_match'].value_counts(normalize=True))\\r\\n print('Number of item pairs in this subset:', len(different_phash))\\r\\n\\r\\n # Check if numbers in titles can be used to improve the accuracy\\r\\n same_numbers = pairs[pairs['nums_match'] == 1]\\r\\n different_numbers = pairs[pairs['nums_match'] == 0]\\r\\n\\r\\n print('\\\\nFor item pairs with the same numbers:')\\r\\n print(same_numbers['true_match'].value_counts(normalize=True))\\r\\n print('Number of item pairs in this subset:', len(same_numbers))\\r\\n\\r\\n print('\\\\nFor item pairs with different numbers:')\\r\\n print(different_numbers['true_match'].value_counts(normalize=True))\\r\\n print('Number of item pairs in this subset:', len(different_numbers))\",\n \"def test_compare_classifier_performance_from_pred_vis_api(experiment_to_use):\\n experiment = experiment_to_use\\n prediction = experiment.predictions\\n viz_outputs = (\\\"pdf\\\", \\\"png\\\")\\n with TemporaryDirectory() as tmpvizdir:\\n for viz_output in viz_outputs:\\n vis_output_pattern_pdf = tmpvizdir + f\\\"/*.{viz_output}\\\"\\n visualize.compare_classifiers_performance_from_pred(\\n [prediction, prediction],\\n experiment.ground_truth,\\n experiment.ground_truth_metadata,\\n experiment.output_feature_name,\\n labels_limit=0,\\n model_namess=[\\\"Model1\\\", \\\"Model2\\\"],\\n output_directory=tmpvizdir,\\n file_format=viz_output,\\n )\\n figure_cnt = glob.glob(vis_output_pattern_pdf)\\n assert 1 == len(figure_cnt)\",\n \"def run_multiple_advanced_experiments(self, path_to_csv_files, algorithms_weights_list, thresholds):\\n starting_time = time.time()\\n print(\\\"started advanced precision recall image_matching_module\\\")\\n for algorithms_weights in algorithms_weights_list:\\n start_file_time = time.time()\\n\\n first_run = True\\n for algorithm, weight in algorithms_weights:\\n if first_run:\\n algorithm_name = algorithm.get_algorithm_name() + weight\\n first_run = False\\n else:\\n algorithm_name = algorithm_name + \\\"_\\\" + algorithm.get_algorithm_name() + weight\\n\\n print(\\\"started algorithm:\\\", algorithm_name)\\n self.run_single_advanced_experiment(path_to_csv_files, algorithms_weights, thresholds)\\n finish_file_time = time.time()\\n print(\\\"finished algorithm:\\\", algorithm_name)\\n print(\\\"time took:\\\", (finish_file_time - start_file_time) / 60, \\\"minutes\\\")\\n finish_time = time.time()\\n hours = int((finish_time - starting_time) // 3600)\\n minutes = (finish_time - starting_time) / 60\\n print(\\\"finished advanced precision recall image_matching_module\\\")\\n print(\\\"time image_matching_module took:\\\", hours, \\\"hours and\\\", minutes, \\\"minutes\\\")\",\n \"def run_multiple_simple_experiments(self, path_to_csv_files, algorithms, thresholds):\\n starting_time = time.time()\\n print(\\\"started simple precision recall image_matching_module\\\")\\n for algorithm in algorithms:\\n start_file_time = time.time()\\n print(\\\"started algorithm:\\\", algorithm.get_algorithm_name())\\n self.run_single_simple_experiment(path_to_csv_files, algorithm, thresholds)\\n finish_file_time = time.time()\\n print(\\\"finished algorithm:\\\", algorithm.get_algorithm_name())\\n print(\\\"time took:\\\", (finish_file_time - start_file_time) / 60, \\\"minutes\\\")\\n finish_time = time.time()\\n hours = int((finish_time - starting_time) // 3600)\\n minutes = (finish_time - starting_time) / 60\\n print(\\\"finished simple precision recall image_matching_module\\\")\\n print(\\\"time image_matching_module took:\\\", hours, \\\"hours and\\\", minutes, \\\"minutes\\\")\",\n \"def _run():\\n matching_terms = {'a', 'b'}\\n source_counts = {'a': 10, 'b': 50, 'c': 25}\\n target_counts = {'a': 4, 'b': 73, 'c': 15}\\n source_chunk = ['a', 'b']\\n target_chunk = ['a', 'c', 'b']\\n source_distance = score.find_distance(\\n matching_terms, source_chunk, source_counts)\\n target_distance = score.find_distance(\\n matching_terms, target_chunk, target_counts)\\n match_score = score.vanilla(\\n matching_terms, source_distance, target_distance, source_counts,\\n target_counts)\\n print('Calculated score:', match_score)\",\n \"def calculates_results_stats(results_dic): \\n # Creates empty dictionary for results_stats_dic\\n results_stats_dic = dict()\\n \\n # Sets all counters to initial values of zero so that they can \\n # be incremented while processing through the images in results_dic \\n results_stats_dic['n_dogs_img'] = 0\\n results_stats_dic['n_match'] = 0\\n results_stats_dic['n_correct_dogs'] = 0\\n results_stats_dic['n_correct_notdogs'] = 0\\n results_stats_dic['n_correct_breed'] = 0\\n \\n # process through the results dictionary\\n for key in results_dic:\\n \\n # Labels Match Exactly\\n if results_dic[key][2] == 1:\\n results_stats_dic['n_match'] += 1\\n\\n # TODO: 5a. REPLACE pass with CODE that counts how many pet images of\\n # dogs had their breed correctly classified. This happens \\n # when the pet image label indicates the image is-a-dog AND \\n # the pet image label and the classifier label match. You \\n # will need to write a conditional statement that determines\\n # when the dog breed is correctly classified and then \\n # increments 'n_correct_breed' by 1. Recall 'n_correct_breed' \\n # is a key in the results_stats_dic dictionary with it's value \\n # representing the number of correctly classified dog breeds.\\n # \\n # Pet Image Label is a Dog AND Labels match- counts Correct Breed\\n if results_dic[key][3] == 1 and results_dic[key][2] == 1:\\n results_stats_dic['n_correct_breed'] += 1\\n \\n # Pet Image Label is a Dog - counts number of dog images\\n if results_dic[key][3] == 1:\\n results_stats_dic['n_dogs_img'] += 1\\n \\n # Classifier classifies image as Dog (& pet image is a dog)\\n # counts number of correct dog classifications\\n if results_dic[key][4] == 1:\\n results_stats_dic['n_correct_dogs'] += 1\\n\\n # TODO: 5b. REPLACE pass with CODE that counts how many pet images \\n # that are NOT dogs were correctly classified. This happens \\n # when the pet image label indicates the image is-NOT-a-dog \\n # AND the classifier label indicates the images is-NOT-a-dog.\\n # You will need to write a conditional statement that \\n # determines when the classifier label indicates the image \\n # is-NOT-a-dog and then increments 'n_correct_notdogs' by 1. \\n # Recall the 'else:' above 'pass' already indicates that the \\n # pet image label indicates the image is-NOT-a-dog and \\n # 'n_correct_notdogs' is a key in the results_stats_dic dictionary \\n # with it's value representing the number of correctly \\n # classified NOT-a-dog images.\\n # \\n # Pet Image Label is NOT a Dog\\n else:\\n # Classifier classifies image as NOT a Dog(& pet image isn't a dog)\\n # counts number of correct NOT dog clasifications.\\n if results_dic[key][3] == 0 and results_dic[key][4] == 0:\\n results_stats_dic['n_correct_notdogs'] += 1\\n\\n\\n # Calculates run statistics (counts & percentages) below that are calculated\\n # using the counters from above.\\n\\n # calculates number of total images\\n results_stats_dic['n_images'] = len(results_dic)\\n\\n # calculates number of not-a-dog images using - images & dog images counts\\n results_stats_dic['n_notdogs_img'] = (results_stats_dic['n_images'] - \\n results_stats_dic['n_dogs_img']) \\n\\n # TODO: 5c. REPLACE zero(0.0) with CODE that calculates the % of correctly\\n # matched images. Recall that this can be calculated by the\\n # number of correctly matched images ('n_match') divided by the \\n # number of images('n_images'). This result will need to be \\n # multiplied by 100.0 to provide the percentage.\\n # \\n # Calculates % correct for matches\\n results_stats_dic['pct_match'] = (results_stats_dic['n_match'] / results_stats_dic['n_images']) * 100\\n\\n # TODO: 5d. REPLACE zero(0.0) with CODE that calculates the % of correctly\\n # classified dog images. Recall that this can be calculated by \\n # the number of correctly classified dog images('n_correct_dogs')\\n # divided by the number of dog images('n_dogs_img'). This result \\n # will need to be multiplied by 100.0 to provide the percentage.\\n # \\n # Calculates % correct dogs\\n results_stats_dic['pct_correct_dogs'] = (results_stats_dic['n_correct_dogs'] / results_stats_dic['n_dogs_img']) * 100\\n\\n # TODO: 5e. REPLACE zero(0.0) with CODE that calculates the % of correctly\\n # classified breeds of dogs. Recall that this can be calculated \\n # by the number of correctly classified breeds of dog('n_correct_breed') \\n # divided by the number of dog images('n_dogs_img'). This result \\n # will need to be multiplied by 100.0 to provide the percentage.\\n # \\n # Calculates % correct breed of dog\\n results_stats_dic['pct_correct_breed'] = (results_stats_dic['n_correct_breed'] / results_stats_dic['n_dogs_img']) * 100\\n\\n # Calculates % correct not-a-dog images\\n # Uses conditional statement for when no 'not a dog' images were submitted \\n if results_stats_dic['n_notdogs_img'] > 0:\\n results_stats_dic['pct_correct_notdogs'] = (results_stats_dic['n_correct_notdogs'] /\\n results_stats_dic['n_notdogs_img']) * 100.0\\n else:\\n results_stats_dic['pct_correct_notdogs'] = 0.0\\n\\n \\n # TODO 5f. REPLACE None with the results_stats_dic dictionary that you \\n # created with this function \\n return results_stats_dic\",\n \"def testdata_matcher(fname1='easy1.png', fname2='easy2.png'):\\n import utool as ut\\n #import vtool as vt\\n from vtool import image as gtool\\n from vtool import features as feattool\\n fpath1 = ut.grab_test_imgpath(fname1)\\n fpath2 = ut.grab_test_imgpath(fname2)\\n featkw = dict(rotation_invariance=True)\\n kpts1, vecs1 = feattool.extract_features(fpath1, **featkw)\\n kpts2, vecs2 = feattool.extract_features(fpath2, **featkw)\\n #if featkw['rotation_invariance']:\\n # print('ori stats 1 ' + ut.get_stats_str(vt.get_oris(kpts2)))\\n # print('ori stats 2 ' + ut.get_stats_str(vt.get_oris(kpts1)))\\n rchip1 = gtool.imread(fpath1)\\n rchip2 = gtool.imread(fpath2)\\n #chip1_shape = vt.gtool.open_image_size(fpath1)\\n chip2_shape = gtool.open_image_size(fpath2)\\n dlen_sqrd2 = chip2_shape[0] ** 2 + chip2_shape[1]\\n testtup = (rchip1, rchip2, kpts1, vecs1, kpts2, vecs2, dlen_sqrd2)\\n return testtup\",\n \"def drawMatches(img1, kp1, img2, kp2, matches):\\n\\n # Create a new output image that concatenates the two images together\\n # (a.k.a) a montage\\n rows1 = img1.shape[0]\\n cols1 = img1.shape[1]\\n rows2 = img2.shape[0]\\n cols2 = img2.shape[1]\\n\\n out = np.zeros((max([rows1,rows2]),cols1+cols2,3), dtype='uint8')\\n\\n # Place the first image to the left\\n out[:rows1,:cols1] = np.dstack([img1, img1, img1])\\n\\n # Place the next image to the right of it\\n out[:rows2,cols1:] = np.dstack([img2, img2, img2])\\n\\n # For each pair of points we have between both images\\n # draw circles, then connect a line between them\\n for mat in matches:\\n\\n # Get the matching keypoints for each of the images\\n img1_idx = mat.queryIdx\\n img2_idx = des_dict[mat.trainIdx][1]\\n\\n # x - columns\\n # y - rows\\n (x1,y1) = kp1[img1_idx].pt\\n (x2,y2) = kp2[img2_idx].pt\\n\\n # Draw a small circle at both co-ordinates\\n # radius 4\\n # colour green\\n # thickness = 1\\n cv2.circle(out, (int(x1),int(y1)), 4, (0, 255, 0)) \\n cv2.circle(out, (int(x2)+cols1,int(y2)), 4, (0, 255, 0))\\n\\n # Draw a line in between the two points\\n # thickness = 1\\n # colour green\\n cv2.line(out, (int(x1),int(y1)), (int(x2)+cols1,int(y2)), (0, 255, 0), 1)\\n\\n\\n # Show the image\\n #cv2.imshow('Matched Features', out)\\n #cv2.waitKey(0)\\n #cv2.destroyWindow('Matched Features')\\n\\n # Also return the image if you'd like a copy\\n return out\",\n \"def all_match():\\n S1=Spectrum.Spectrum()\\n S1.add_peak(50.7,234)\\n S1.add_peak(54.6,585)\\n S1.add_peak(60.7,773)\\n S1.add_peak(65.6,387)\\n S1.add_peak(87.7,546)\\n S1.add_peak(104.6,598)\\n S1.pep_mass=100\\n S1.euclidean_scale()\\n\\n S2=Spectrum.Spectrum()\\n S2.add_peak(50.5,234/2)\\n S2.add_peak(54.8,585/2)\\n S2.add_peak(61.0,773/2)\\n S2.add_peak(65.4,387/2)\\n S2.add_peak(88.0,546/2)\\n S2.add_peak(104.3,598/2)\\n S2.pep_mass=100\\n S2.euclidean_scale()\\n\\n score,peaks=similarity.cosine_score_max(S1,S2)\\n assert peaks==6, \\\"Incorrect number of peaks matched with greedy method\\\"\\n assert math.isclose(score,1.0), \\\"Incorrect score with greedy method\\\"\\n\\n score,peaks=similarity.cosine_score_greedy(S1,S2)\\n assert peaks==6, \\\"Incorrect number of peaks matched with maximum weighted method\\\"\\n assert math.isclose(score,1.0), \\\"Incorrect score with maximum weighted method\\\"\",\n \"def ransac(matches, kp1, kp2, sample_points=4, iterations=5, inlier_tolerance=3, inlier_ratio=0.45, check=True, return_max_x=False):\\n\\n best_inlier_count = 0\\n best_h = None\\n best_inlier_indices = None\\n\\n # Get all the corresponing matching pairs for both the images\\n pts1 = np.array([kp1[m.queryIdx].pt for m in matches])\\n pts2 = np.array([kp2[m.trainIdx].pt for m in matches])\\n\\n # Re-usable variables for all iterations\\n homogeneous_pts1 = np.hstack((pts1, np.ones((pts1.shape[0], 1)))).T\\n indices = np.arange(len(pts1))\\n num_pts = pts1.shape[0]\\n required_inliers = inlier_ratio * num_pts\\n\\n # For number of iterations\\n for _ in range(iterations):\\n\\n # Sample a small set of points from the point match pairs\\n indices_to_sample = np.random.choice(indices, sample_points)\\n pts1_sample = pts1[indices_to_sample]\\n pts2_sample = pts2[indices_to_sample]\\n\\n # Get the homography matrix\\n h = get_homography_matrix(pts1_sample, pts2_sample)\\n\\n # Find the new points using the homography matrix\\n transformed_points = np.dot(h, homogeneous_pts1).T\\n\\n # Convert it to world coordinates\\n last_col = np.copy(transformed_points[:, -1])\\n last_col = last_col[:, np.newaxis]\\n transformed_points /= last_col\\n transformed_points = transformed_points[:, :-1]\\n\\n # Find the distance between the actual and the mapped points\\n distance = np.linalg.norm(pts2 - transformed_points, axis=1)\\n inlier_indices = distance < inlier_tolerance\\n inlier_count = inlier_indices.sum()\\n\\n # Update the best_h if the current h has more inliers\\n if inlier_count > best_inlier_count:\\n best_h = h\\n best_inlier_indices = inlier_indices\\n best_inlier_count = inlier_count\\n\\n # If required inliers is reached break\\n if inlier_count > required_inliers:\\n break\\n\\n # Verbose mode - Print the number of inliers\\n if check:\\n transformed_points = np.dot(best_h, homogeneous_pts1).T\\n # Convert it to world coordinates\\n last_col = np.copy(transformed_points[:, -1])\\n last_col = last_col[:, np.newaxis]\\n transformed_points /= last_col\\n transformed_points = transformed_points[:, :-1]\\n distance = np.linalg.norm(pts2 - transformed_points, axis=1)\\n inlier_count = len(distance[distance < inlier_tolerance])\\n print('%2.2f of the points are inliers' %\\n (inlier_count / num_pts * 100))\\n\\n # If x coordinates are needed\\n if return_max_x:\\n max_x_inlier_1 = ceil(pts1[best_inlier_indices].max(axis=0)[0])\\n max_x_inlier_2 = ceil(pts2[best_inlier_indices].max(axis=0)[0])\\n return best_h, max_x_inlier_1, max_x_inlier_2\\n return best_h\",\n \"def versus(bot1, bot2, num_matches, cpu_count, seed_list):\\n W, L, D = 0, 0, 0\\n tot_scores = [0, 0]\\n avg_scores = [0, 0]\\n all_scores = []\\n show_match_results = 1\\n\\n pool = multiprocessing.Pool(cpu_count)\\n\\n try:\\n bot1name = os.path.splitext(os.path.basename(bot1))[0]\\n bot2name = os.path.splitext(os.path.basename(bot2))[0]\\n start_time = time.time()\\n results = [pool.apply_async(run_match, (bot1, bot2, seed_list[i]))\\n for i in range(num_matches)]\\n \\n if show_match_results:\\n sys.stdout.write('%20s - %3i: [' % (bot2name, num_matches))\\n\\n tenths_done = 0\\n for r in results:\\n scores = r.get(timeout=50)\\n if scores[0] > scores[1]:\\n W += 1\\n elif scores[0] < scores[1]:\\n L += 1\\n elif scores[0] == scores[1]:\\n D += 1\\n tot_scores = [tot_scores[x]+scores[x] for x in range(2)]\\n avg_scores = [tot_scores[x]//sum([W,L,D]) for x in range(2)]\\n all_scores.append(scores[0:2])\\n if (sum([W,L,D])*10//num_matches) > tenths_done:\\n tenths_done += 1\\n if show_match_results:\\n sys.stdout.write ('.')\\n\\n pool.close()\\n pool.join()\\n \\n if show_match_results:\\n diff = [s[0] - s[1] for s in all_scores]\\n avg, std = meanstd(diff)\\n sys.stdout.write (' '*(10-tenths_done))\\n print '] - %12s by %8s diff %5.1f std %4.1f in %5.1fs' % (str([W,L,D]), str(avg_scores), avg, std, time.time()-start_time)\\n except multiprocessing.TimeoutError:\\n if show_match_results:\\n sys.stdout.write ('x'*(10-tenths_done))\\n print '] - %12s by %8s in %5.1fs' % (str([W,L,D]), str(avg_scores), time.time()-start_time)\\n pool.terminate()\\n\\n return [W,L,D], tot_scores, all_scores\",\n \"def drawMatches(orbimg1, kp1, orbimg2, kp2, matches):\\r\\n\\r\\n # Create a new output image that concatenates the two images together\\r\\n # (a.k.a) a montage\\r\\n rows1 = orbimg1.shape[0]\\r\\n cols1 = orbimg1.shape[1]\\r\\n rows2 = orbimg2.shape[0]\\r\\n cols2 = orbimg2.shape[1]\\r\\n\\r\\n # Create the output image\\r\\n # The rows of the output are the largest between the two images\\r\\n # and the columns are simply the sum of the two together\\r\\n # The intent is to make this a colour image, so make this 3 channels\\r\\n out = np.zeros((max([rows1,rows2]),cols1+cols2,3), dtype='uint8')\\r\\n\\r\\n # Place the first image to the left\\r\\n out[:rows1,:cols1] = np.dstack([orbimg1, orbimg1, orbimg1])\\r\\n\\r\\n # Place the next image to the right of it\\r\\n out[:rows2,cols1:] = np.dstack([orbimg2, orbimg2, orbimg2])\\r\\n\\r\\n # For each pair of points we have between both images\\r\\n # draw circles, then connect a line between them\\r\\n for mat in matches:\\r\\n\\r\\n # Get the matching keypoints for each of the images\\r\\n orbimg1_idx = mat.queryIdx\\r\\n orbimg2_idx = mat.trainIdx\\r\\n\\r\\n # x - columns\\r\\n # y - rows\\r\\n (x1,y1) = kp1[orbimg1_idx].pt\\r\\n (x2,y2) = kp2[orbimg2_idx].pt\\r\\n\\r\\n # Draw a small circle at both co-ordinates\\r\\n # radius 4\\r\\n # colour blue\\r\\n # thickness = 1\\r\\n cv2.circle(out, (int(x1),int(y1)), 4, (255, 0, 0), 1) \\r\\n cv2.circle(out, (int(x2)+cols1,int(y2)), 4, (255, 0, 0), 1)\\r\\n\\r\\n # Draw a line in between the two points\\r\\n # thickness = 1\\r\\n # colour blue\\r\\n cv2.line(out, (int(x1),int(y1)), (int(x2)+cols1,int(y2)), (255,0,0), 1)\\r\\n\\r\\n\\r\\n # Show the image\\r\\n #cv2.imshow('Matched Features', out)\\r\\n #cv2.waitKey(0)\\r\\n #cv2.destroyWindow('Matched Features')\\r\\n\\r\\n # Also return the image if you'd like a copy\\r\\n return out\",\n \"def calculate_precision_recall_all_images(\\n all_prediction_boxes, all_gt_boxes, iou_threshold):\\n # Find total true positives, false positives and false negatives\\n # over all images\\n\\n tot_num_tp, tot_num_fp, tot_num_fn = 0, 0, 0\\n\\n for image in range(len(all_prediction_boxes)):\\n prediction_boxes = all_prediction_boxes[image]\\n gt_boxes = all_gt_boxes[image]\\n\\n image_results = calculate_individual_image_result(\\n prediction_boxes,\\n gt_boxes,\\n iou_threshold)\\n\\n tot_num_tp += image_results[\\\"true_pos\\\"]\\n tot_num_fp += image_results[\\\"false_pos\\\"]\\n tot_num_fn += image_results[\\\"false_neg\\\"]\\n\\n # Compute precision, recall\\n precision = calculate_precision(tot_num_tp, tot_num_fp, tot_num_fn)\\n recall = calculate_recall(tot_num_tp, tot_num_fp, tot_num_fn)\\n\\n return (precision, recall)\",\n \"def test_compare_classifier_performance_from_prob_vis_api(experiment_to_use):\\n experiment = experiment_to_use\\n probability = experiment.probabilities\\n viz_outputs = (\\\"pdf\\\", \\\"png\\\")\\n with TemporaryDirectory() as tmpvizdir:\\n for viz_output in viz_outputs:\\n vis_output_pattern_pdf = tmpvizdir + f\\\"/*.{viz_output}\\\"\\n visualize.compare_classifiers_performance_from_prob(\\n [probability, probability],\\n experiment.ground_truth,\\n experiment.ground_truth_metadata,\\n experiment.output_feature_name,\\n top_n_classes=[0],\\n labels_limit=0,\\n model_namess=[\\\"Model1\\\", \\\"Model2\\\"],\\n output_directory=tmpvizdir,\\n file_format=viz_output,\\n )\\n figure_cnt = glob.glob(vis_output_pattern_pdf)\\n assert 1 == len(figure_cnt)\",\n \"def main():\\n\\n parser = argparse.ArgumentParser(description='codec_compare')\\n parser.add_argument('path', metavar='DIR',\\n help='path to images folder')\\n args = parser.parse_args()\\n classpath = args.path\\n classname = classpath.split('/')[1]\\n\\n images = set(listdir_full_path(classpath))\\n if len(images) <= 0:\\n print \\\"\\\\033[91m[ERROR]\\\\033[0m\\\" + \\\" no source files in ./images.\\\"\\n sys.exit(1)\\n\\n codeclist_full = set(['aom', 'deepcoder', 'deepcoder-lite', 'fuif', 'fvdo', 'hevc', 'kakadu', 'jpeg',\\n 'pik', 'tat', 'xavs', 'xavs-fast', 'xavs-median', 'webp'])\\n\\n bpp_targets = set([0.06, 0.12, 0.25, 0.50, 0.75, 1.00, 1.50, 2.00])\\n for image in images:\\n width, height, depth = get_dimensions(image, classname)\\n name, imgfmt = os.path.splitext(image)\\n imgfmt = os.path.basename(image).split(\\\".\\\")[-1]\\n derivative_images = []\\n if classname[:6] == 'classB':\\n derivative_images = create_derivatives(image, classname)\\n else:\\n derivative_images.append((image, imgfmt))\\n\\n for derivative_image, pix_fmt in derivative_images:\\n json_dir = 'metrics'\\n mkdir_p(json_dir)\\n json_file = os.path.join(json_dir,\\n os.path.splitext(os.path.basename(derivative_image))[0] + \\\".\\\" + pix_fmt + \\\".json\\\")\\n # if os.path.isfile(json_file):\\n # print \\\"\\\\033[92m[JSON OK]\\\\033[0m \\\" + json_file\\n # continue\\n main_dict = dict()\\n derivative_image_metrics = dict()\\n for codecname in codeclist_full:\\n convertflag = 1\\n caseflag = pix_fmt\\n if (codecname == 'webp' or codecname == 'tat' or 'deepcoder' in codecname) and depth != '8':\\n continue\\n if 'xavs' in codecname and depth != '8' and depth != '10':\\n continue\\n if 'classE' in classname and ('tat' in codecname or 'xavs' in codecname or 'deepcoder' in codecname):\\n continue\\n if codecname == 'kakadu' and classname[:6] == 'classB':\\n convertflag = 0\\n caseflag = imgfmt\\n bpp_target_metrics = dict()\\n for bpp_target in bpp_targets:\\n print(codecname)\\n if codecname == 'aom' and classname[:6] == 'classB':\\n # ('AERIAL2' in image or 'CATS' in image or 'XRAY' in image or 'GOLD' in image or 'TEXTURE1' in image):\\n encoded_image_name = os.path.splitext(os.path.basename(derivative_image))[\\n 0] + '_' + str(bpp_target) + '_' + imgfmt + '.' + 'av1'\\n encoded_image = os.path.join('outputs', codecname, encoded_image_name)\\n decoded_image = os.path.join('outputs', codecname, 'decoded', encoded_image_name + '.' + imgfmt)\\n original_image = image\\n elif codecname == 'kakadu' and classname[:6] == 'classB':\\n encoded_image_name = os.path.splitext(os.path.basename(derivative_image))[\\n 0] + '_' + str(bpp_target) + '_' + imgfmt + '.' + codecname\\n encoded_image = os.path.join('outputs', codecname, encoded_image_name)\\n decoded_image = os.path.join('outputs', codecname, 'decoded', encoded_image_name + '.' + imgfmt)\\n original_image = image\\n elif 'xavs' in codecname and classname[:6] == 'classB':\\n encoded_image_name = os.path.splitext(os.path.basename(derivative_image))[\\n 0] + '_' + str(bpp_target) + '_' + imgfmt + '.' + codecname\\n encoded_image = os.path.join('outputs', codecname, encoded_image_name)\\n decoded_image = os.path.join('outputs', codecname, 'decoded', encoded_image_name + '.' + imgfmt)\\n original_image = image\\n elif codecname == 'fvdo' and classname[:6] == 'classB':\\n encoded_image_name = os.path.splitext(os.path.basename(derivative_image))[\\n 0] + '_' + str(bpp_target) + '_pgm' + '.' + codecname\\n encoded_image = os.path.join('outputs', codecname, encoded_image_name)\\n decoded_image = os.path.join('outputs', codecname, 'decoded', encoded_image_name + '.pgm')\\n original_image = image\\n else:\\n if codecname == 'fuif' and 'tif' in imgfmt:\\n encoded_image_name = os.path.splitext(os.path.basename(derivative_image))[\\n 0] + '.tif_' + str(bpp_target) + '_' + pix_fmt + '.' + codecname\\n elif codecname == 'webp' or codecname == 'tat':\\n encoded_image_name = os.path.splitext(os.path.basename(derivative_image))[\\n 0] + '_' + str(bpp_target) + '_yuv420p.' + codecname\\n else:\\n encoded_image_name = os.path.splitext(os.path.basename(derivative_image))[\\n 0] + '_' + str(bpp_target) + '_' + pix_fmt + '.' + codecname\\n encoded_image = os.path.join('outputs', codecname, encoded_image_name)\\n decoded_image_path = os.path.join('outputs', codecname, 'decoded')\\n decoded_image = ''\\n for decodedfile in os.listdir(decoded_image_path):\\n encoderoot = '_'.join(os.path.splitext(os.path.basename(encoded_image_name))[0].split('_')[:-1])\\n if encoderoot in decodedfile:\\n if ('tat' in codecname or 'webp' in codecname) and os.path.splitext(os.path.basename(decodedfile))[1] == '.yuv':\\n decoded_image = os.path.join('outputs', codecname, 'decoded', decodedfile)\\n print(decoded_image)\\n if ('tat' not in codecname or 'webp' not in codecname) and os.path.splitext(os.path.basename(decodedfile))[1] != '.yuv':\\n decoded_image = os.path.join('outputs', codecname, 'decoded', decodedfile)\\n if 'classE' not in classname and 'classB' not in classname and os.path.isfile(decoded_image):\\n decoded_image = convert_decoded(decoded_image, width, height, depth, codecname)\\n original_image = convert_decoded(derivative_image, width, height, depth, 'reference')\\n else:\\n original_image = derivative_image\\n\\n print('Reference:' + original_image)\\n print('Encoded:' + encoded_image)\\n print('Decoded:' + decoded_image)\\n if (os.path.isfile(original_image) and os.path.isfile(decoded_image) and os.path.isfile(encoded_image)):\\n if 'classE' in classname:\\n metrics = compute_metrics_HDR(original_image, decoded_image, encoded_image, bpp_target,\\n codecname, width, height, pix_fmt, depth)\\n\\n elif 'classB' in classname:\\n metrics = compute_metrics(original_image, decoded_image, encoded_image, bpp_target, codecname,\\n width, height, pix_fmt)\\n else:\\n metrics = compute_metrics_SDR(original_image, decoded_image, encoded_image, bpp_target,\\n codecname, width,\\n height, imgfmt, depth)\\n measured_bpp = (os.path.getsize(encoded_image) * 1.024 * 8) / (float((int(width) * int(height))))\\n bpp_target_metrics[measured_bpp] = metrics\\n else:\\n continue\\n \\n derivative_image_metrics[codecname] = bpp_target_metrics\\n main_dict[derivative_image] = derivative_image_metrics\\n\\n mkdir_p(json_dir)\\n with open(json_file, 'w') as f:\\n f.write(json.dumps(main_dict, indent=2))\",\n \"def run_matching(self):\\n paradic = self.cfg['param']['paradic']\\n print 'in run_matching() n_bins = ' +str(paradic['n_bins'])\\n\\n f = open(self.work_dir+'matches.txt','w')\\n matching = self.run_proc(['match_cli', 'keys_im0.txt',\\n 'keys_im1.txt',\\n str(paradic['flag_match']),\\n str(paradic['C_match']),\\n str(paradic['n_hist']),\\n str(paradic['n_ori']),\\n str(paradic['n_bins'])],\\n stdout=f)\\n self.wait_proc(matching, timeout=self.timeout)\\n return 1\",\n \"def perform_match(self, parser_manager):\\n log = logging.getLogger(\\\".\\\".join([__name__, self.__class__.__name__]))\\n log.addFilter(logconfig.ThreadContextFilter())\\n extra = self._config_align.all_to_json()\\n extra.update(self._config_crystal.all_to_json())\\n log = logging.LoggerAdapter(log, extra)\\n log.info(\\\"Matching Started\\\")\\n log.debug(extra)\\n\\n input_poi = parser_manager.parse_selected_points_from_args()\\n beamline_image = parser_manager.get_focused_image()\\n parser_manager.save_focused_image(beamline_image)\\n focused_image_path = parser_manager.get_focused_image_path()\\n formulatrix_image_path = parser_manager.get_formulatrix_image_path()\\n job_id = parser_manager.get_job_id()\\n\\n # Create the images\\n image1 = Image.from_file(formulatrix_image_path)\\n image2 = beamline_image\\n\\n # Create results object\\n service_result = ServiceResult(job_id, formulatrix_image_path, focused_image_path)\\n\\n # Perform alignment\\n try:\\n\\n aligned_images, scaled_poi = self._perform_alignment(image1, image2, input_poi)\\n service_result.set_image_alignment_results(aligned_images)\\n\\n # Perform Crystal Matching - only proceed if we have a valid alignment\\n if aligned_images.alignment_status_code() == ALIGNED_IMAGE_STATUS_OK:\\n match_results = self._perform_matching(aligned_images, scaled_poi, parser_manager)\\n\\n service_result.append_crystal_matching_results(match_results)\\n\\n except Exception as e:\\n if sys.version_info[0] < 3:\\n log.error(\\\"ERROR: \\\" + e.message)\\n else:\\n log.error(\\\"ERROR: \\\" + str(e))\\n service_result.set_err_state(e)\\n\\n return service_result\",\n \"def evaluate(pred_file, ref_file):\\n ref_dict, pred_dict, query_dict, id_dict = build_pred_ref_dict(ref_file, pred_file, ref_file)\\n total, acc, scores = res_eval_with_type_acc(query_dict, pred_dict, ref_dict, id_dict, save=False)\\n em = calculate_exact_match(pred_dict, ref_dict)\\n print('Comp Acc: {:.3f}%\\\\tBleu-4: {:.3f}\\\\tRouge-L: {:.3f}'.format(acc, scores['Bleu-4'], scores['Rouge-L']))\\n print('EM: {:.3f}%'.format(em))\\n # calculate_sketch_type_acc(ref_file, pred_file)\\n # calculate_exact_match_for_each_q_type(ref_file, pred_file)\\n return total, acc, scores, em\",\n \"def drawMatches(image_1, image_1_keypoints, image_2, image_2_keypoints, matches):\\n # Compute number of channels.\\n num_channels = 1\\n if len(image_1.shape) == 3:\\n num_channels = image_1.shape[2]\\n # Separation between images.\\n margin = 10\\n # Create an array that will fit both images (with a margin of 10 to separate\\n # the two images)\\n joined_image = np.zeros((max(image_1.shape[0], image_2.shape[0]),\\n image_1.shape[1] + image_2.shape[1] + margin,\\n 3))\\n if num_channels == 1:\\n for channel_idx in range(3):\\n joined_image[:image_1.shape[0],\\n :image_1.shape[1],\\n channel_idx] = image_1\\n joined_image[:image_2.shape[0],\\n image_1.shape[1] + margin:,\\n channel_idx] = image_2\\n else:\\n joined_image[:image_1.shape[0], :image_1.shape[1]] = image_1\\n joined_image[:image_2.shape[0], image_1.shape[1] + margin:] = image_2\\n\\n for match in matches:\\n image_1_point = (int(image_1_keypoints[match.queryIdx].pt[0]),\\n int(image_1_keypoints[match.queryIdx].pt[1]))\\n image_2_point = (int(image_2_keypoints[match.trainIdx].pt[0] + \\\\\\n image_1.shape[1] + margin),\\n int(image_2_keypoints[match.trainIdx].pt[1]))\\n\\n cv2.circle(joined_image, image_1_point, 5, (0, 0, 255), thickness = -1)\\n cv2.circle(joined_image, image_2_point, 5, (0, 255, 0), thickness = -1)\\n cv2.line(joined_image, image_1_point, image_2_point, (255, 0, 0), \\\\\\n thickness = 3)\\n return joined_image\",\n \"def evaluate(self):\\n tic = time.time()\\n print('Running per image evaluation...')\\n p = self.params\\n # add backward compatibility if useSegm is specified in params\\n if p.useSegm is not None:\\n p.iouType = 'segm' if p.useSegm == 1 else 'bbox'\\n print(\\n \\\"useSegm (deprecated) is not None.\\\"\\n \\\"Running {} evaluation\\\".format(p.iouType)\\n )\\n print('Evaluate annotation type *{}*'.format(p.iouType))\\n p.imgIds = list(np.unique(p.imgIds))\\n if p.useCats:\\n p.catIds = list(np.unique(p.catIds))\\n p.maxDets = sorted(p.maxDets)\\n self.params = p\\n\\n self._prepare()\\n # loop through images, area range, max detection number\\n catIds = p.catIds if p.useCats else [-1]\\n\\n self.ious = {\\n (imgId, catId): self.computeIoU(imgId, catId)\\n for imgId in p.imgIds\\n for catId in catIds\\n }\\n # ignore any categories that is not having any attributes\\n self.f1s = {\\n (imgId, catId): self.computeF1(imgId, catId)\\n for imgId in p.imgIds\\n for catId in catIds\\n if catId in self.FPParams.catsWithAttributes\\n }\\n\\n # self.gt_attributes_ids = []\\n\\n # loop through images, area range, max detection number\\n self.evalImgs = [\\n self.evaluateImg(imgId, catId, areaRng, p.maxDets[-1])\\n for catId in catIds\\n for areaRng in p.areaRng\\n for imgId in p.imgIds\\n ]\\n # self._paramsEval = copy.deepcopy(self.params) # seems do not need it\\n toc = time.time()\\n print('DONE (t={:0.2f}s).'.format(toc-tic))\",\n \"def evaluation(eval_dict, faster_rcnn, test_num=Config.eval_num):\\n bboxes, labels, scores = list(), list(), list()\\n gt_bboxes, gt_labels, gt_difficult = list(), list(), list()\\n for i, [img_dir, img_info] in tqdm(enumerate(eval_dict.items())):\\n img, img_info = rescale_image(img_dir, img_info, flip=False)\\n img_tensor = create_img_tensor(img)\\n box, score, label = faster_rcnn.predict(img_tensor)\\n gt_bbox = np.array(img_info['objects'])[:, 1:5].astype(np.float32)\\n gt_label = np.array(img_info['objects'])[:, 0]\\n difficult = np.array(img_info['difficult'])\\n\\n # label from text to number\\n gt_label = text_to_num(gt_label)\\n\\n bboxes.append(box)\\n labels.append(label)\\n scores.append(score)\\n gt_bboxes.append(gt_bbox)\\n gt_labels.append(gt_label)\\n gt_difficult.append(difficult)\\n if i == test_num:\\n break\\n\\n AP, mAP = calc_map(bboxes, labels, scores, gt_bboxes, gt_labels, gt_difficult, use_07_metric=True)\\n\\n return mAP\",\n \"def evaluate(self, test_data):\\n test_results = [(self.decide(self.feedforward(x)), y)\\n for (x, y) in test_data]\\n\\n # Check for and keep track of TP's FP's and FN's\\n # Write FP's and FN's to a special directories\\n TP = FP = FN = count = 0\\n TN = 0\\n # what you got = x, what should be = y\\n for ((x, y), (image, gt)) in zip(test_results, test_data):\\n count += 1\\n if x == 1 and y == 1:\\n TP += 1\\n elif x == 1 and y == 0:\\n FP += 1\\n if not self.validating: self.save_Image(FP_PATH, count, image)\\n elif x == 0 and y == 1:\\n FN += 1\\n if not self.validating: self.save_Image(FN_PATH, count, image)\\n \\n if (TP + FP ) == 0:\\n return TP, FP, FN, '---------', '---------','---------'\\n else:\\n false_rate = float(FP) / float(TP + FP)\\n\\n if (TP + FN ) == 0:\\n return TP, FP, FN, '---------', '---------', false_rate\\n else:\\n detect_rate = float(TP) / float(TP + FN)\\n\\n if (TP + FP + FN ) == 0:\\n return TP, FP, FN, '---------', detect_rate, false_rate\\n else:\\n quality_rate = float(TP) / float(TP+FP+FN)\\n \\n return TP, FP, FN, quality_rate, detect_rate, false_rate\",\n \"def test_compare_classifiers_predictions_vis_api(experiment_to_use):\\n experiment = experiment_to_use\\n predictions = experiment.predictions\\n viz_outputs = (\\\"pdf\\\", \\\"png\\\")\\n with TemporaryDirectory() as tmpvizdir:\\n for viz_output in viz_outputs:\\n vis_output_pattern_pdf = tmpvizdir + f\\\"/*.{viz_output}\\\"\\n visualize.compare_classifiers_predictions(\\n [predictions, predictions],\\n experiment.ground_truth,\\n experiment.ground_truth_metadata,\\n experiment.output_feature_name,\\n labels_limit=0,\\n model_names=[\\\"Model1\\\", \\\"Model2\\\"],\\n output_directory=tmpvizdir,\\n file_format=viz_output,\\n )\\n figure_cnt = glob.glob(vis_output_pattern_pdf)\\n assert 1 == len(figure_cnt)\",\n \"def test_compare_single_and_multiprocess_results(sidesweep_image_sequence):\\n cc = Cwsim_container_from_ims(ims=sidesweep_image_sequence)\\n test_idx = 20\\n results_single_proc = cc.self_im_query_all(test_idx, plot_output=False)\\n\\n test_im = sidesweep_image_sequence[test_idx]\\n cc.prepare_memory_bank_outside()\\n results_multi_proc = cc.query_image_mp(test_im)\\n\\n plt.plot(results_single_proc, marker='D', label='single_proc')\\n plt.plot(results_multi_proc, marker='D', label='multi_proc')\\n plt.legend()\\n plt.draw()\\n plt.show(block=False)\\n plt.pause(3)\\n\\n assert np.allclose(results_single_proc, results_multi_proc)\",\n \"def match(desc1,desc2):\\n desc1 = array([d/linalg.norm(d) for d in desc1])\\n desc2 = array([d/linalg.norm(d) for d in desc2])\\n dist_ratio = 0.6\\n desc1_size = desc1.shape\\n matchscores = zeros((desc1_size[0],1),'int')\\n desc2t = desc2.T # precompute matrix transpose\\n for i in range(desc1_size[0]):\\n dotprods = dot(desc1[i, :], desc2t) # vector of dot products\\n dotprods *= 0.9999\\n # inverse cosine and sort, return index for features in second image\\n indx = argsort(arccos(dotprods))\\n # check if nearest neighbor has angle less than dist_ratio times 2nd\\n if arccos(dotprods)[indx[0]] < dist_ratio * arccos(dotprods)[indx[1]]:\\n matchscores[i] = int(indx[0])\\n return matchscores\",\n \"def compute_matches(self):\\n\\t\\tself.local_database[\\\"figure_number\\\"] = [0] * len(self.local_database[\\\"feature_vectors\\\"])\\n\\t\\tForensics = wbForensicsHOG(Database=self.local_database)\\n\\t\\tForensics.KDTree_pairs(leaf_size = len(self.local_database)+1)\\n\\t\\tForensics.d_rank(pairs=Forensics.pairs, distances=Forensics.dists, ratios=Forensics.ratios)\\n\\n\\t\\tself.local_matches = Forensics.Dist_Rank\",\n \"def test(label_dict, image_ids, args):\\n\\n log_header('Starting image retrieval preparations')\\n\\n queries = []\\n\\n if args.test_image is not None:\\n queries.append(args.test_image)\\n else:\\n logging.info('Generating random test queries')\\n\\n number_of_images = len(image_ids)\\n\\n # Generate random queries to use in the test procedure\\n for i in range(args.k):\\n queries.append(image_ids[random.randint(0, number_of_images - 1)])\\n\\n # Calculate score for the retrieved images\\n calculate_score(\\n label_dict, queries, retrieve_similar_images(queries, args)\\n )\",\n \"def _compute_prediction(self, k, img, kp, des):\\n ###taken from example code:\\n\\n #find matches\\n matches = self.matcher.knnMatch(des,self.descs[k],k=2)\\n\\n # make sure the distance to the closest match is sufficiently better than the second closest\\n good_matches = []\\n for m,n in matches: \\n if (m.distance < 0.7*n.distance and\\n kp[m.queryIdx].response > 0.0 and\\n self.kps[k][m.trainIdx].response > 0.0):\\n good_matches.append((m.queryIdx, m.trainIdx))\\n \\n if len(good_matches) < 4:\\n visual_diff = None\\n else:\\n #create pt lists\\n img_pts = np.zeros((len(good_matches),2))\\n tem_pts = np.zeros((len(good_matches),2))\\n for idx in range(len(good_matches)):\\n match = good_matches[idx]\\n img_pts[idx,:] = kp[match[0]].pt\\n tem_pts[idx,:] = self.kps[k][match[1]].pt\\n \\n # Transform input image so that it matches the template image as well as possible\\n M, mask = cv2.findHomography(img_pts, tem_pts, cv2.RANSAC, 8.0)\\n img_T = cv2.warpPerspective(img, M, self.signs[k].shape[::-1])\\n \\t\\n #calculate error\\n visual_diff = compare_images(img_T, self.signs[k])\\n return visual_diff\\n #not functioning properly (currently does not return highest value for identical image/scene pair)\",\n \"def my_evaluate(original_image, img_id, annotations, min_score=0.4, max_overlap=0.3, top_k=200):\\n\\n # Transform\\n image = normalize(to_tensor(resize(original_image)))\\n\\n # Move to default device\\n image = image.to(device)\\n\\n # Forward prop.\\n predicted_locs, predicted_scores = model(image.unsqueeze(0))\\n\\n # Detect objects in SSD output\\n det_boxes, det_labels, det_scores = model.detect_objects(predicted_locs, predicted_scores, min_score=min_score,\\n max_overlap=max_overlap, top_k=top_k)\\n\\n # Move detections to the CPU\\n det_boxes = det_boxes[0].to('cpu')\\n\\n # Transform to original image dimensions\\n original_dims = torch.FloatTensor(\\n [original_image.width, original_image.height, original_image.width, original_image.height]).unsqueeze(0)\\n det_boxes = det_boxes * original_dims\\n\\n # Decode class integer labels\\n det_labels = [rev_label_map[l] for l in det_labels[0].to('cpu').tolist()]\\n det_scores = det_scores[0].to('cpu').tolist()\\n\\n if det_labels != ['background']:\\n for i in range(det_boxes.size(0)):\\n \\n # For every detection, see whether it matches a ground truth\\n match = False\\n for j in range(len(annotations['boxes'])):\\n if annotations['labels'][j] == -1: # this is being set when a ground truth already matched\\n continue\\n match = box_match(det_boxes[i], \\n det_labels[i], \\n torch.Tensor(annotations['boxes'][j]), \\n rev_label_map[annotations['labels'][j]])\\n if match:\\n annotations['labels'][j] = -1\\n break\\n \\n if match: # true positive if the detection is correct and matched a ground truth\\n all_results['true_positive'] += 1\\n else: # false positive if the detection did not match a ground truth\\n all_results['false_positive'] += 1\\n \\n # After all detections were checked, let us see whether the detector missed something\\n for label in annotations['labels']: \\n if label == -1: # This is set after a detection matched this ground truth\\n continue\\n \\n # false negative if we reach this line, since the ground truth object was not found\\n all_results['false_negative'] += 1\",\n \"def test_run_multi_r__(self):\\n\\n # Test Description\\n # ================\\n #\\n # 1. This test intialises an example *eopy.matchup.matchupIO.MatchUp* object\\n #\\n # 2. Compare transformed dataset to expected value\\n\\n ################################################################################################################\\n # 1. Initialise Test Data Object\\n ################################################################################################################\\n\\n MatchUpTest = return_MatchUpTest_r__()\\n\\n ################################################################################################################\\n # 2. Define expected values\\n ################################################################################################################\\n\\n # Original dataset values (should be unchanged)\\n MatchUpOriginal_expected = return_MatchUpTest_r__()\\n\\n # Transformed dataset\\n values_expected = array([294.0625, 480.3733333, 300.6, 227.3846154, 210.1533333,\\n 22.74193548, 22.0625, 21.96875, 22.80645161, 23.5,\\n 21.66666667, 21.05882353, 23, 22.40625,\\n 38.33333333, 36.63636364, 36.5, 38.42857143,\\n 30.1, 32.14893617, 29.37254902, 28.88461538, 28.56603774,\\n 33.45238095, 32.81395349, 31.77272727, 32.60465116,\\n 40.125, 43.54054054, 38.59090909, 34.08510638,\\n 13.72727273, 12, 14.1, 11.79069767, 17.53846154,\\n 12.69565217, 31.16666667, 12.26086957, 11.52272727,\\n 8.8125, 12, 7.4, 10.13207547])\\n unc_expected = [Uncertainty(1, array([1.6, 1.5, 1.5, 1.3, 1.5])),\\n Uncertainty(1, array([3.1, 3.2, 3.2, 3.1, 3.0])),\\n Uncertainty(1, array([3.3, 3.4, 3.1, 3.2])),\\n Uncertainty(1, array([2.1, 2.2, 2.2, 2.1])),\\n Uncertainty(1, array([5.0, 4.7, 5.1, 5.2, 5.3])),\\n Uncertainty(1, array([4.2, 4.3, 4.4, 4.3])),\\n Uncertainty(1, array([4.0, 3.7, 4.4, 4.7])),\\n Uncertainty(1, array([2.2, 1.7, 2.0, 4.3, 2.6])),\\n Uncertainty(1, array([2.3, 1.2, 2.3, 4.4])),\\n Uncertainty(1, array([3.2, 2.7, 3.0, 5.3]))]\\n ks_expected = array([4.8, 6.8, 5.2, 5.6, 5.2, 12.10287443, 13.99394856, 12.48108926, 12.85930408])\\n unck_expected = [Uncertainty(1, array([0.25, 0.25, 0.25, 0.25, 0.25])),\\n Uncertainty(1, array([0.2644, 0.2644, 0.2644, 0.2644]))]\\n idx_expected = {\\\"Nm\\\": [5, 4],\\n \\\"cNm\\\": [0, 5, 9],\\n \\\"Im\\\": [[0, 1], [1, 2]],\\n \\\"sensors\\\": [-1, 1, 2],\\n \\\"sensor_ms\\\": [1, 3, 3],\\n \\\"n_sensor\\\": [0, 1, 1, 2, 1, 1, 2, 1, 1, 2],\\n \\\"n_mu\\\": [1, 1, 2, 2, 1, 2, 2, 1, 2, 2],\\n \\\"n_cov\\\": [1, 1, 1, 1, 2, 2, 2, 3, 3, 3],\\n \\\"N_var\\\": [5, 5, 4, 4, 5, 4, 4, 5, 4, 4],\\n \\\"idx\\\": [0, 5, 10, 14, 18, 23, 27, 31, 36, 40, 44],\\n \\\"Ia\\\": [1, 1, 1, 2, 2, 2]}\\n a_expected = array([1., 1.3, 0.002, 0.5, 1.1, 0.0005])\\n w_matrices_expected = []\\n u_matrices_expected = []\\n\\n ################################################################################################################\\n # 3. Run Transform2NormInd.run()\\n ################################################################################################################\\n\\n Transform2NormIndOp = Transform2NormInd()\\n MatchUpTransform = Transform2NormIndOp.run(MatchUpTest)\\n\\n values_test = MatchUpTransform.values\\n unc_test = MatchUpTransform.unc\\n w_matrices_test = MatchUpTransform.w_matrices\\n u_matrices_test = MatchUpTransform.u_matrices\\n ks_test = MatchUpTransform.ks\\n unck_test = MatchUpTransform.unck\\n idx_test = MatchUpTransform.idx\\n\\n ################################################################################################################\\n # 4. Compare retrieve values to expect values\\n ################################################################################################################\\n\\n # Test transformed data object attribute by attribute\\n\\n # a. values\\n for i, (value_expected, value_test) in enumerate(zip(values_expected, values_test)):\\n self.assertAlmostEqual(value_expected, value_test, places=5, msg=str(i))\\n\\n # b. unc\\n for block_unc_test, block_unc_expected in zip(unc_test, unc_expected):\\n self.assertEqual(block_unc_expected.typeID, block_unc_test.typeID)\\n self.assertEqual(block_unc_expected.uR.tolist(), block_unc_test.uR.tolist())\\n\\n # c. w_matrices\\n self.assertEqual(w_matrices_test, w_matrices_expected)\\n\\n # d. u_matrices\\n self.assertEqual(u_matrices_test, u_matrices_expected)\\n\\n # e. ks\\n for k_expected, k_test in zip(ks_expected, ks_test):\\n self.assertAlmostEqual(k_test.tolist(), k_expected.tolist(), places=5)\\n\\n # f. unck\\n for block_unck_test, block_unck_expected in zip(unck_test, unck_expected):\\n self.assertEqual(block_unck_expected.typeID, block_unck_test.typeID)\\n self.assertEqual(block_unck_expected.uR.tolist(), block_unck_test.uR.tolist())\\n\\n # h. idx\\n self.assertEqual(set(idx_expected.keys()), set(idx_test.keys()))\\n for key in idx_expected.keys():\\n idx_i_test = idx_test[key]\\n idx_i_expected = idx_expected[key]\\n if isinstance(idx_i_expected, ndarray):\\n self.assertEqual(idx_i_test.tolist(), idx_i_expected.tolist())\\n else:\\n self.assertEqual(idx_i_test, idx_i_expected)\\n\\n # Test original data object preserved attribute by attribute\\n\\n # a. values\\n for i, (value_original_expected, value_original_test) in enumerate(zip(MatchUpOriginal_expected.values, MatchUpTest.values)):\\n self.assertAlmostEqual(value_original_expected, value_original_test, places=5)\\n\\n # b. unc\\n for block_unc_original_expected, block_unc_original_test in zip(MatchUpOriginal_expected.unc, MatchUpTest.unc):\\n self.assertEqual(block_unc_original_expected.typeID, block_unc_original_test.typeID)\\n self.assertEqual(block_unc_original_expected.uR.tolist(), block_unc_original_test.uR.tolist())\\n\\n # c. w_matrices\\n self.assertEqual(MatchUpOriginal_expected.w_matrices, MatchUpTest.w_matrices)\\n\\n # d. u_matrices\\n self.assertEqual(MatchUpOriginal_expected.u_matrices, MatchUpTest.u_matrices)\\n\\n # e. ks\\n for k_original_expected, k_original_test in zip(MatchUpOriginal_expected.ks, MatchUpTest.ks):\\n self.assertAlmostEqual(k_original_test.tolist(), k_original_expected.tolist(), places=5)\\n\\n # f. unck\\n for block_unck_original_expected, block_unck_original_test in zip(MatchUpOriginal_expected.unck, MatchUpTest.unck):\\n self.assertEqual(block_unck_original_expected.typeID, block_unck_original_test.typeID)\\n self.assertEqual(block_unck_original_expected.uR.tolist(), block_unck_original_test.uR.tolist())\\n\\n # h. idx\\n self.assertEqual(set(MatchUpOriginal_expected.idx), set(MatchUpTest.idx))\\n for key in MatchUpOriginal_expected.idx.keys():\\n idx_i_original_test = MatchUpTest.idx[key]\\n idx_i_original_expected = MatchUpOriginal_expected.idx[key]\\n if isinstance(idx_i_original_expected, ndarray):\\n self.assertEqual(idx_i_original_test.tolist(), idx_i_original_expected.tolist())\\n else:\\n self.assertEqual(idx_i_original_test, idx_i_original_expected)\",\n \"def _calculate_similarity(self):\\n self._logger.info(\\\"Calculating the similarity between images.\\\")\\n\\n # Create a helper function to simplify the loops below:\\n def get_wrapper(fdx, mdx):\\n wrapper = pos_wrappers.image_similarity_wrapper(\\n reference_image=self.f['src_gray'](idx=fdx),\\n moving_image=self.f['src_gray'](idx=mdx),\\n affine_transformation=self.f['part_transf'](mIdx=mdx, fIdx=fdx))\\n return copy.copy(wrapper)\\n\\n commands = [] # Will hold commands for calculating the similarity\\n\\n # Will hold (moving, fixed) images partial_transforms basically: all\\n # partial transformations array\\n partial_transforms = []\\n\\n self._logger.debug(\\\"Generating similarity measure warppers.\\\")\\n for moving_slice in self.options.slice_range:\\n # Get all fixed images to which given moving slice will be aligned:\\n tpair = list(flatten(self._get_slice_pair(moving_slice)))\\n\\n # Append partial transformations for given moving slice to the\\n # global partial transformations array\\n partial_transforms.append(tpair)\\n\\n # Generate wrapper for measuring similarity for a given partial\\n # transformation.\\n for mdx, fdx in tpair:\\n commands.append(get_wrapper(fdx, mdx))\\n\\n # Execute and commands and workflow the similarity measurements.\\n stdout, stderr = self.execute(commands)\\n simmilarity = map(lambda x: float(x.strip()),\\n stdout.strip().split(\\\"\\\\n\\\"))\\n simmilarity = dict(zip(flatten(partial_transforms), simmilarity))\\n\\n self._logger.debug(\\\"Generating graph edges.\\\")\\n graph_connections = []\\n\\n # Lambda defines slice skipping is preffered (lower l), or reluctant\\n # to slice skipping (higher)\\n l = self.options.graphEdgeLambda\\n\\n for (mdx, fdx), s in simmilarity.iteritems():\\n w = (1.0 + s) * abs(mdx - fdx) * (1.0 + l) ** (abs(mdx - fdx))\\n graph_connections.append((fdx, mdx, w))\\n\\n self._logger.info(\\\"Creating a graph based on image similarities.\\\")\\n # Generate the graph basen on the weight of the edges\\n self.G = nx.DiGraph()\\n self.G.add_weighted_edges_from(graph_connections)\\n\\n self._logger.debug(\\\"Saving the graph to a file.\\\")\\n # Save the edges for some further analysis.\\n nx.write_weighted_edgelist(self.G,\\n self.f['graph_edges'](sign=self.signature))\\n\\n # Also, save the individual similarity metrics:\\n simm_fh = open(self.f['similarity'](sign=self.signature), 'w')\\n for (mdx, fdx), s in sorted(simmilarity.iteritems()):\\n simm_fh.write(\\\"%d %d %f\\\\n\\\" % (mdx, fdx, s))\\n simm_fh.close()\",\n \"def compute_matches(gt_boxes, gt_class_ids, gt_masks,\\n pred_boxes, pred_class_ids, pred_scores, pred_masks,\\n iou_threshold=0.5, score_threshold=0.0):\\n # Trim zero padding\\n # TODO: cleaner to do zero unpadding upstream\\n gt_boxes = trim_zeros(gt_boxes)\\n gt_masks = gt_masks[..., :gt_boxes.shape[0]]\\n pred_boxes = trim_zeros(pred_boxes)\\n pred_scores = pred_scores[:pred_boxes.shape[0]]\\n # Sort predictions by score from high to low\\n indices = np.argsort(pred_scores)[::-1]\\n pred_boxes = pred_boxes[indices]\\n pred_class_ids = pred_class_ids[indices]\\n pred_scores = pred_scores[indices]\\n pred_masks = pred_masks[..., indices]\\n\\n # Compute IoU overlaps [pred_masks, gt_masks]\\n overlaps = compute_overlaps_masks(pred_masks, gt_masks)\\n\\n # Loop through predictions and find matching ground truth boxes\\n match_count = 0\\n pred_match = -1 * np.ones([pred_boxes.shape[0]])\\n gt_match = -1 * np.ones([gt_boxes.shape[0]])\\n for i in range(len(pred_boxes)):\\n # Find best matching ground truth box\\n # 1. Sort matches by score\\n sorted_ixs = np.argsort(overlaps[i])[::-1]\\n # 2. Remove low scores\\n low_score_idx = np.where(overlaps[i, sorted_ixs] < score_threshold)[0]\\n if low_score_idx.size > 0:\\n sorted_ixs = sorted_ixs[:low_score_idx[0]]\\n # 3. Find the match\\n for j in sorted_ixs:\\n # If ground truth box is already matched, go to next one\\n if gt_match[j] > -1:\\n continue\\n # If we reach IoU smaller than the threshold, end the loop\\n iou = overlaps[i, j]\\n if iou < iou_threshold:\\n break\\n # Do we have a match?\\n if pred_class_ids[i] == gt_class_ids[j]:\\n match_count += 1\\n gt_match[j] = i\\n pred_match[i] = j\\n break\\n\\n return gt_match, pred_match, overlaps\",\n \"def apply_feature_matching(desc1: np.ndarray, desc2: np.ndarray,\\n match_calculator: Callable[[list, list], float]) -> list:\\n\\n # Check descriptors dimensions are 2\\n assert desc1.ndim == 2, \\\"Descriptor 1 shape is not 2\\\"\\n assert desc2.ndim == 2, \\\"Descriptor 2 shape is not 2\\\"\\n\\n # Check that the two features have the same descriptor type\\n assert desc1.shape[1] == desc2.shape[1], \\\"Descriptors shapes are not equal\\\"\\n\\n # If there is not key points in any of the images\\n if desc1.shape[0] == 0 or desc2.shape[0] == 0:\\n return []\\n\\n # number of key points in each image\\n num_key_points1 = desc1.shape[0]\\n num_key_points2 = desc2.shape[0]\\n\\n # List to store the matches scores\\n matches = []\\n\\n # Loop over each key point in image1\\n # We need to calculate similarity with each key point in image2\\n for kp1 in range(num_key_points1):\\n # Initial variables which will be updated in the loop\\n distance = -np.inf\\n y_index = -1\\n\\n # Loop over each key point in image2\\n for kp2 in range(num_key_points2):\\n\\n # Match features between the 2 vectors\\n value = match_calculator(desc1[kp1], desc2[kp2])\\n\\n # SSD values examples: (50, 200, 70), we need to minimize SSD (Choose 50)\\n # In case of SSD matching: (value is returned as a \\\"negative\\\" number) (-50, -200, -70)\\n # So we compare it with -np.inf. (The sorting will be reversed later)\\n\\n # NCC values examples: (0.58, 0.87, 0.42), we need to maximize NCC (Choose 0.87)\\n # In case of NCC matching: (value is returned as a \\\"positive\\\" number) (-58, -0.87, -0.42)\\n # So we compare it with -np.inf. (The sorting will be reversed later)\\n\\n if value > distance:\\n distance = value\\n y_index = kp2\\n\\n # Create a cv2.DMatch object for each match and set attributes:\\n # queryIdx: The index of the feature in the first image\\n # trainIdx: The index of the feature in the second image\\n # distance: The distance between the two features\\n cur = cv2.DMatch()\\n cur.queryIdx = kp1\\n cur.trainIdx = y_index\\n cur.distance = distance\\n matches.append(cur)\\n\\n return matches\",\n \"def find_matching_points(img1, img2, max_pix_movement=50, normalize=True, show=False):\\n\\n # Initiate ORB detector\\n orb = cv2.ORB_create()\\n\\n # find the keypoints and descriptors with ORB\\n kp1, des1 = orb.detectAndCompute(img1, None)\\n kp2, des2 = orb.detectAndCompute(img2, None)\\n\\n # create BFMatcher object\\n bf = cv2.BFMatcher(cv2.NORM_HAMMING, crossCheck=True)\\n # Match descriptors.\\n matches = bf.match(des1,des2)\\n # Sort them in the order of their distance.\\n matches = sorted(matches, key = lambda x:x.distance)\\n # Draw first 10 matches.\\n if show:\\n img3 = cv2.drawMatches(img1,kp1,img2,kp2,matches[:500], None,flags=2)\\n plt.imshow(img3),plt.show()\\n # Get the matching keypoints for each of the images\\n\\n list_kp1 = []\\n list_kp2 = []\\n for mat in matches:\\n img1_idx = mat.queryIdx\\n img2_idx = mat.trainIdx\\n\\n # x - columns\\n # y - rows\\n list_kp1.append(kp1[img1_idx].pt)\\n list_kp2.append(kp2[img2_idx].pt)\\n\\n n_kp1, n_kp2 = np.float32(list_kp1), np.float32(list_kp2)\\n n_kp1 /= np.asarray([img1.shape[1], img1.shape[0]], np.float32)\\n n_kp2 /= np.asarray([img2.shape[1], img2.shape[0]], np.float32)\\n n_kp1 = n_kp1 * 2. - 1.\\n n_kp2 = n_kp2 * 2. - 1.\\n\\n return np.int32(list_kp1), np.int32(list_kp2), n_kp1, n_kp2\",\n \"def ratio_test(fx2_match, fx1_match, fx1_norm, match_dist, norm_dist,\\n ratio_thresh=.625, fm_dtype=np.int32, fs_dtype=np.float32):\\n fx2_to_ratio = np.divide(match_dist, norm_dist).astype(fs_dtype)\\n fx2_to_isvalid = np.less(fx2_to_ratio, ratio_thresh)\\n fx2_match_RAT = fx2_match.compress(fx2_to_isvalid).astype(fm_dtype)\\n fx1_match_RAT = fx1_match.compress(fx2_to_isvalid).astype(fm_dtype)\\n fx1_norm_RAT = fx1_norm.compress(fx2_to_isvalid).astype(fm_dtype)\\n # Turn the ratio into a score\\n fs_RAT = np.subtract(1.0, fx2_to_ratio.compress(fx2_to_isvalid))\\n fm_RAT = np.vstack((fx1_match_RAT, fx2_match_RAT)).T\\n # return normalizer info as well\\n fm_norm_RAT = np.vstack((fx1_norm_RAT, fx2_match_RAT)).T\\n ratio_tup = MatchTup3(fm_RAT, fs_RAT, fm_norm_RAT)\\n return ratio_tup\",\n \"def return_MatchUpTest_r__():\\n\\n ####################################################################################################################\\n # 1. Initialise test data\\n ####################################################################################################################\\n\\n values = array([470.5, 720.56, 450.9, 295.6, 315.23,\\n 70.5, 70.6, 70.3, 70.7, 70.5,\\n 71.5, 71.6, 71.3, 71.7,\\n 80.5, 80.6, 80.3, 80.7,\\n 150.5, 151.1, 149.8, 150.2, 151.4,\\n 140.5, 141.1, 139.8, 140.2,\\n 160.5, 161.1, 169.8, 160.2,\\n 30.2, 20.4, 28.2, 50.7, 45.6,\\n 29.2, 37.4, 28.2, 50.7,\\n 28.2, 32.4, 22.2, 53.7, ])\\n unc = [Uncertainty(1, array([1.6, 1.5, 1.5, 1.3, 1.5])),\\n Uncertainty(1, array([3.1, 3.2, 3.2, 3.1, 3.0])),\\n Uncertainty(1, array([3.3, 3.4, 3.1, 3.2])),\\n Uncertainty(1, array([2.1, 2.2, 2.2, 2.1])),\\n Uncertainty(1, array([5.0, 4.7, 5.1, 5.2, 5.3])),\\n Uncertainty(1, array([4.2, 4.3, 4.4, 4.3])),\\n Uncertainty(1, array([4.0, 3.7, 4.4, 4.7])),\\n Uncertainty(1, array([2.2, 1.7, 2.0, 4.3, 2.6])),\\n Uncertainty(1, array([2.3, 1.2, 2.3, 4.4])),\\n Uncertainty(1, array([3.2, 2.7, 3.0, 5.3]))]\\n ks = array([1.2, 1.7, 1.3, 1.4, 1.3, 3.2, 3.7, 3.3, 3.4])\\n unck = [Uncertainty(1, array([0.25, 0.25, 0.25, 0.25, 0.25])),\\n Uncertainty(1, array([0.2644, 0.2644, 0.2644, 0.2644]))]\\n idx = {\\\"Nm\\\": [5, 4],\\n \\\"cNm\\\": [0, 5, 9],\\n \\\"Im\\\": [[0, 1], [1, 2]],\\n \\\"sensors\\\": [-1, 1, 2],\\n \\\"sensor_ms\\\": [1, 3, 3],\\n \\\"n_sensor\\\": [0, 1, 1, 2, 1, 1, 2, 1, 1, 2],\\n \\\"n_mu\\\": [1, 1, 2, 2, 1, 2, 2, 1, 2, 2],\\n \\\"n_cov\\\": [1, 1, 1, 1, 2, 2, 2, 3, 3, 3],\\n \\\"N_var\\\": [5, 5, 4, 4, 5, 4, 4, 5, 4, 4],\\n \\\"idx\\\": [0, 5, 10, 14, 18, 23, 27, 31, 36, 40, 44],\\n \\\"Ia\\\": [1, 1, 1, 2, 2, 2]}\\n a = array([1., 1.3, 0.002, 0.5, 1.1, 0.0005])\\n\\n ####################################################################################################################\\n # 3. Initialise MatchUp object\\n ####################################################################################################################\\n\\n MatchUpTest = MatchUp()\\n MatchUpTest.values = values\\n MatchUpTest.unc = unc\\n MatchUpTest.ks = ks\\n MatchUpTest.unck = unck\\n MatchUpTest.idx = idx\\n MatchUpTest.a = a\\n\\n return MatchUpTest\",\n \"def drawMatches(img1, kp1, img2, kp2, matches):\\n\\n # Create a new output image that concatenates the two images together\\n # (a.k.a) a montage\\n rows1 = img1.shape[0]\\n cols1 = img1.shape[1]\\n rows2 = img2.shape[0]\\n cols2 = img2.shape[1]\\n\\n out = np.zeros((max([rows1,rows2]),cols1+cols2,3), dtype='uint8')\\n\\n # Place the first image to the left\\n out[:rows1,:cols1] = np.dstack([img1, img1, img1])\\n\\n # Place the next image to the right of it\\n out[:rows2,cols1:] = np.dstack([img2, img2, img2])\\n\\n # For each pair of points we have between both images\\n # draw circles, then connect a line between them\\n for mat in matches:\\n\\n # Get the matching keypoints for each of the images\\n img1_idx = mat.queryIdx\\n img2_idx = mat.trainIdx\\n\\n # x - columns\\n # y - rows\\n (x1,y1) = kp1[img1_idx].pt\\n (x2,y2) = kp2[img2_idx].pt\\n\\n # Draw a small circle at both co-ordinates\\n # radius 4\\n # colour blue\\n # thickness = 1\\n cv2.circle(out, (int(x1),int(y1)), 4, (255, 0, 0), 1) \\n cv2.circle(out, (int(x2)+cols1,int(y2)), 4, (255, 0, 0), 1)\\n\\n # Draw a line in between the two points\\n # thickness = 1\\n # colour blue\\n cv2.line(out, (int(x1),int(y1)), (int(x2)+cols1,int(y2)), (255, 0, 0), 1)\\n\\n\\n # Show the image\\n\\n # Also return the image if you'd like a copy\\n return out\",\n \"def plot_images(img_A, img_B, match_A, match_B, name=\\\"p1\\\", method=\\\"greedy\\\"):\\n (hA, wA) = img_A.shape[:2]\\n (hB, wB) = img_B.shape[:2]\\n vis = np.zeros((max(hA, hB), wA + wB, 3), dtype=\\\"uint8\\\")\\n vis[0:hA, 0:wA] = img_A\\n vis[0:hB, wA:] = img_B\\n\\n # loop over the matches\\n for ptA, ptB in zip(match_A, match_B):\\n ptB = ptB + np.array([wA, 0])\\n cv2.line(vis, tuple(ptA), tuple(ptB), (0, 255, 0), 1)\\n\\n cv2.imwrite(\\\"output/sift_\\\" + method + \\\"_\\\" + name + \\\".png\\\", vis)\\n cv2.imshow(\\\"Keypoint matching\\\", vis)\\n cv2.waitKey(0)\",\n \"def ransac(keypoints1, keypoints2, matches, sampling_ratio=0.5, n_iters=500, threshold=20):\\n N = matches.shape[0]\\n n_samples = int(N * sampling_ratio)\\n\\n # Please note that coordinates are in the format (y, x)\\n matched1 = pad(keypoints1[matches[:,0]])\\n matched2 = pad(keypoints2[matches[:,1]])\\n matched1_unpad = keypoints1[matches[:,0]]\\n matched2_unpad = keypoints2[matches[:,1]]\\n\\n max_inliers = np.zeros(N)\\n n_inliers = 0\\n\\n # RANSAC iteration start\\n ### YOUR CODE HERE\\n raise NotImplementedError() # Delete this line\\n ### END YOUR CODE\\n return H, matches[max_inliers]\",\n \"def match(desc1,desc2):\\n\\t\\n\\tdesc1 = array([d/linalg.norm(d) for d in desc1])\\n\\tdesc2 = array([d/linalg.norm(d) for d in desc2])\\n\\t\\n\\tdist_ratio = 0.6\\n\\tdesc1_size = desc1.shape\\n\\t\\n\\tmatchscores = zeros((desc1_size[0],1))\\n\\tdesc2t = desc2.T #precompute matrix transpose\\n\\tfor i in range(desc1_size[0]):\\n\\t\\tdotprods = dot(desc1[i,:],desc2t) #vector of dot products\\n\\t\\tdotprods = 0.9999*dotprods\\n\\t\\t#inverse cosine and sort, return index for features in second image\\n\\t\\tindx = argsort(arccos(dotprods))\\n\\t\\t\\n\\t\\t#check if nearest neighbor has angle less than dist_ratio times 2nd\\n#\\t\\tif arccos(dotprods)[indx[0]] < dist_ratio * arccos(dotprods)[indx[1]]:\\n\\t\\tmatchscores[i] = int(indx[0])\\n\\t\\n\\treturn matchscores\",\n \"def compute_transform(self, image_pair, figure):\\t\\n\\t\\timage_pair = [int(image_pair[1]), int(image_pair[3])]\\n\\t\\tindices = np.where(np.all(np.abs(self.Matches[\\\"image_pairs\\\"][0] - image_pair) == 0, axis=1)) \\n\\t\\tfeature_pairs = self.Matches[\\\"image_match_pairs\\\"][0][indices[0][0]]\\n\\t\\tfeature_pair_scores = self.Matches[\\\"image_feature_scores\\\"][0][indices[0][0]] \\n\\n\\t\\ti1_loc = self.Database[\\\"image_locs\\\"][np.where(np.array(self.Database[\\\"image_idx\\\"]) == image_pair[0])[0][0]]\\n\\t\\ti2_loc = self.Database[\\\"image_locs\\\"][np.where(np.array(self.Database[\\\"image_idx\\\"]) == image_pair[1])[0][0]]\\n\\n\\t\\tself.image1 = cv2.cvtColor(extract_image(figure, i1_loc), cv2.COLOR_BGR2GRAY)\\n\\t\\tself.image2 = cv2.cvtColor(extract_image(figure, i2_loc), cv2.COLOR_BGR2GRAY)\\n\\t\\t# self.image1 = extract_image(figure, i1_loc)\\n\\t\\t# self.image2 = extract_image(figure, i2_loc)\\n\\n\\t\\tself.match_alignment = {}; self.match_alignment_scores = {}\\n\\t\\tfor pair, score in zip(feature_pairs, feature_pair_scores):\\n\\t\\t\\ti1 = self.Database[\\\"image_idx\\\"][pair[0]]\\n\\t\\t\\ti2 = self.Database[\\\"image_idx\\\"][pair[1]]\\n\\n\\t\\t\\tif i1 == image_pair[0]:\\n\\t\\t\\t\\ttarget = pair[0] \\n\\t\\t\\t\\tstart = pair[1]\\n\\t\\t\\t\\tif target not in self.match_alignment.keys():\\n\\t\\t\\t\\t\\tself.init_match_alignment(target)\\n\\t\\t\\telse:\\n\\t\\t\\t\\ttarget = pair[1]\\n\\t\\t\\t\\tstart = pair[0]\\t\\t\\t\\t\\n\\t\\t\\t\\tif target not in self.match_alignment.keys():\\n\\t\\t\\t\\t\\tself.init_match_alignment(target)\\n\\n\\t\\t\\t# get ellipse and convert to three cardinal points\\n\\t\\t\\ttarget_ellipse = self.Database[\\\"orientation\\\"][target]\\n\\t\\t\\tstart_ellipse = self.Database[\\\"orientation\\\"][start]\\n\\t\\t\\ttarget_points, target_params = utils.ellipse2points(target_ellipse)\\n\\t\\t\\tstart_points, start_params = utils.ellipse2points(start_ellipse)\\n\\n\\t\\t\\t# compute angle between Major axes\\n\\t\\t\\trotation_angle = int(np.abs(target_params[0] - start_params[0]))\\n\\t\\t\\treflection_truth = target_params[1] * start_params[1]\\n\\t\\t\\tif reflection_truth > 0:\\n\\t\\t\\t\\treflection_truth = \\\"No\\\"\\n\\t\\t\\telse:\\n\\t\\t\\t\\treflection_truth = \\\"Yes\\\"\\n\\t\\t\\tresolution_perc = round(((target_params[2] / start_params[2]) - 1), 1)\\n\\n\\t\\t\\t# extract features for histogram matching\\n\\t\\t\\ttarget_bloc = self.Database[\\\"blot_locs\\\"][target]\\n\\t\\t\\tstart_bloc = self.Database[\\\"blot_locs\\\"][start]\\n\\t\\t\\ttarget_feature = extract_image(self.image1, target_bloc, ex=5)\\n\\t\\t\\tfinal_feature = extract_image(self.image2, start_bloc, ex=5) \\n\\n\\t\\t\\t#find affine transform to match\\n\\t\\t\\tif self.AFFINE:\\n\\t\\t\\t\\tT = utils.compute_affine(start_points, target_points)\\n\\t\\t\\t\\timage2_warped = cv2.warpAffine(self.image2.copy(), T, (self.image1.shape[1], self.image1.shape[0]))\\n\\t\\t\\t\\tfinal_feature = extract_image(image2_warped, target_bloc, ex=5)\\n\\t\\t\\t\\n\\t\\t\\t# match hist\\n\\t\\t\\tif self.CONTRAST:\\n\\t\\t\\t\\tfinal_feature = utils.histogram_match(final_feature, target_feature)\\n\\n\\t\\t\\tself.match_alignment[target].append((final_feature, start, rotation_angle, reflection_truth, \\\\\\n\\t\\t\\t\\t\\t\\t\\t\\t\\t\\t\\t\\t resolution_perc, score))\\n\\t\\t\\tself.match_alignment_scores[target].append(score)\",\n \"def compare_predictions():\\n validation_labels = np.array(pd.read_csv(val_true_labels_dir + dataset_version + 'validation_labels.csv', index_col=0))\\n validation_labels = np.reshape(validation_labels, (-1))\\n\\n diff_between_files = []\\n also1s = []\\n also2s = []\\n for filename1 in os.listdir(val_predictions_dir):\\n if filename1.endswith(\\\".csv\\\"):\\n for filename2 in os.listdir(val_predictions_dir):\\n if filename2.endswith(\\\".csv\\\"):\\n if filename1 < filename2:\\n wrong1 = 0\\n wrong2 = 0\\n diff_between = 0\\n also1 = 0\\n also2 = 0\\n diff_corr1 = 0\\n diff_corr2 = 0\\n f1 = np.array(pd.read_csv(val_predictions_dir + filename1, index_col=0))\\n f1 = np.reshape(f1, (-1))\\n f2 = np.array(pd.read_csv(val_predictions_dir + filename2, index_col=0))\\n f2 = np.reshape(f2, (-1))\\n for line in range(f1.shape[0]):\\n if f1[line] != validation_labels[line]:\\n wrong1 += 1\\n if f2[line] != validation_labels[line]:\\n wrong2 += 1\\n if f1[line] != f2[line]:\\n diff_between += 1\\n if f1[line] == validation_labels[line]:\\n diff_corr1 += 1\\n if f2[line] == validation_labels[line]:\\n diff_corr2 += 1\\n if f1[line] != validation_labels[line]:\\n if f2[line] != validation_labels[line]:\\n also2 += 1\\n if f2[line] != validation_labels[line]:\\n if f1[line] != validation_labels[line]:\\n also1 += 1\\n\\n diff_between_files.append(diff_between)\\n print(filename1)\\n print('Wrongly predicted by 1: ' + str(100 * wrong1 / f1.shape[0]) + '%')\\n print(filename2)\\n print('Wrongly predicted by 2: ' + str(100 * wrong2 / f1.shape[0]) + '%')\\n print()\\n print('Differences between files: ' + str(100 * diff_between / f1.shape[0]) + '%')\\n print(f'\\\\t of which correct by 1 {100 * diff_corr1 / diff_between}%, by 2 {100 * diff_corr2 / diff_between}%')\\n also1s.append(also1 / wrong2)\\n also2s.append(also2 / wrong1)\\n print('Wrongly predicted by other among wrong ones: ' + str(100 * also2 / wrong1) + '%, ' + str(\\n 100 * also1 / wrong2) + '%\\\\n\\\\n\\\\n')\\n\\n print('Max, min and avg differences between files:')\\n print(str(100 * max(diff_between_files) / validation_labels.shape[0]) + '%')\\n print(str(100 * min(diff_between_files) / validation_labels.shape[0]) + '%')\\n print(str(100 * np.mean(diff_between_files) / validation_labels.shape[0]) + '%')\\n\\n print('\\\\nWrongly predicted by first that were also wrongly predicted by second:')\\n print('Max: ' + str(100 * max(also2s)) + '%')\\n print('Min: ' + str(100 * min(also2s)) + '%')\\n print('Avg: ' + str(100 * np.mean(also2s)) + '%')\\n\\n print('\\\\nWrongly predicted by second that were also wrongly predicted by first:')\\n print('Max: ' + str(100 * max(also1s)) + '%')\\n print('Min: ' + str(100 * min(also1s)) + '%')\\n print('Avg: ' + str(100 * np.mean(also1s)) + '%')\",\n \"def matching_accuracy(pmat_pred, pmat_gt, ns):\\n device = pmat_pred.device\\n batch_num = pmat_pred.shape[0]\\n\\n pmat_gt = pmat_gt.to(device)\\n\\n assert torch.all((pmat_pred == 0) + (pmat_pred == 1)), 'pmat_pred can noly contain 0/1 elements.'\\n assert torch.all((pmat_gt == 0) + (pmat_gt == 1)), 'pmat_gt should noly contain 0/1 elements.'\\n assert torch.all(torch.sum(pmat_gt, dim=-1) <= 1) and torch.all(torch.sum(pmat_gt, dim=-2) <= 1)\\n assert torch.all(torch.sum(pmat_pred, dim=-1) <= 1) and torch.all(torch.sum(pmat_pred, dim=-2) <= 1)\\n\\n #indices_pred = torch.argmax(pmat_pred, dim=-1)\\n #indices_gt = torch.argmax(pmat_gt, dim=-1)\\n\\n #matched = (indices_gt == indices_pred).type(pmat_pred.dtype)\\n match_num_list = []\\n gt_num_list = []\\n pred_num_list = []\\n acc_gt = []\\n acc_pred = []\\n for b in range(batch_num): #acc_gt,acc_pred\\n #match_num += torch.sum(matched[b, :ns[b]])\\n #total_num += ns[b].item()\\n match_num = torch.sum(pmat_pred[b, :] * pmat_gt[b, :]) + 1e-8 #:ns[b]\\n gt_num = torch.sum(pmat_gt[b, :]) + 1e-8\\n pred_num = torch.sum(pmat_pred[b,:]) + 1e-8\\n match_num_list.append(match_num.cpu().numpy())\\n gt_num_list.append(gt_num.cpu().numpy())\\n pred_num_list.append(pred_num.cpu().numpy())\\n acc_gt.append((match_num/gt_num).cpu().numpy())\\n acc_pred.append((match_num/pred_num).cpu().numpy())\\n\\n return {'acc_gt': np.array(acc_gt),\\n 'acc_pred': np.array(acc_pred),\\n 'match_num': np.array(match_num_list),\\n 'gt_num': np.array(gt_num_list),\\n 'pred_num': np.array(pred_num_list)}\",\n \"def compute_similarity(im1, im2, test_id):\\n\\n results = dict()\\n results['id'] = test_id\\n results['test_im'] = im2\\n results['mse_value'], results['mse_map'] = compute_mse(im1, im2)\\n results['ssim_value'], results['ssim_map'] = compute_ssim(im1, im2, 5)\\n results['cw_ssim_value'], results['cw_ssim_map'] = compute_cw_ssim(im1, im2, 30)\\n results['gms_value'], results['gms_map'] = compute_gms(im1, im2)\\n results['fsim_value'], results['pc_max_map'] = compute_fsim(im1, im2)\\n\\n return results\",\n \"def test_detector1pipeline4(_bigdata):\\n step = Detector1Pipeline()\\n step.save_calibrated_ramp = True\\n step.ipc.skip = True\\n step.persistence.skip = True\\n step.jump.rejection_threshold = 4.0\\n step.ramp_fit.save_opt = False\\n step.output_file = 'jw84600007001_02101_00001_nrs1_rate.fits'\\n step.run(_bigdata+'/pipelines/jw84600007001_02101_00001_nrs1_uncal.fits')\\n\\n # Compare ramp product\\n n_ramp = 'jw84600007001_02101_00001_nrs1_ramp.fits'\\n h = fits.open( n_ramp )\\n n_ref = _bigdata+'/pipelines/jw84600007001_02101_00001_nrs1_ramp_ref.fits'\\n href = fits.open( n_ref )\\n newh = fits.HDUList([h['primary'],h['sci'],h['err'],h['groupdq'],h['pixeldq']])\\n newhref = fits.HDUList([href['primary'],href['sci'],href['err'],href['groupdq'],h['pixeldq']])\\n result = fits.diff.FITSDiff(newh,\\n newhref,\\n ignore_keywords = ['DATE','CAL_VER','CAL_VCS','CRDS_VER','CRDS_CTX'],\\n rtol = 0.00001\\n )\\n assert result.identical, result.report()\\n\\n # Compare countrate image product\\n n_cr = 'jw84600007001_02101_00001_nrs1_rate.fits'\\n h = fits.open( n_cr )\\n n_ref = _bigdata+'/pipelines/jw84600007001_02101_00001_nrs1_rate_ref.fits'\\n href = fits.open( n_ref )\\n newh = fits.HDUList([h['primary'],h['sci'],h['err'],h['dq']])\\n newhref = fits.HDUList([href['primary'],href['sci'],href['err'],href['dq']])\\n result = fits.diff.FITSDiff(newh,\\n newhref,\\n ignore_keywords = ['DATE','CAL_VER','CAL_VCS','CRDS_VER','CRDS_CTX'],\\n rtol = 0.00001\\n )\\n assert result.identical, result.report()\",\n \"def test_compare_classifiers_performance_changing_k_vis_api(experiment_to_use):\\n experiment = experiment_to_use\\n probabilities = experiment.probabilities\\n viz_outputs = (\\\"pdf\\\", \\\"png\\\")\\n with TemporaryDirectory() as tmpvizdir:\\n for viz_output in viz_outputs:\\n vis_output_pattern_pdf = tmpvizdir + f\\\"/*.{viz_output}\\\"\\n visualize.compare_classifiers_performance_changing_k(\\n [probabilities, probabilities],\\n experiment.ground_truth,\\n experiment.ground_truth_metadata,\\n experiment.output_feature_name,\\n top_k=3,\\n labels_limit=0,\\n model_namess=[\\\"Model1\\\", \\\"Model2\\\"],\\n output_directory=tmpvizdir,\\n file_format=viz_output,\\n )\\n figure_cnt = glob.glob(vis_output_pattern_pdf)\\n assert 1 == len(figure_cnt)\",\n \"def max_disp(left_img, right_img):\\n\\n # Max initial matches\\n MAX_INI_MATCH = np.inf\\n\\n # ORB detector and brute-force matcher\\n detector = cv2.ORB_create(nfeatures=100000, edgeThreshold=5)\\n matcher = cv2.BFMatcher(cv2.NORM_HAMMING, crossCheck=True)\\n\\n # Find the keypoints and descriptors of both images\\n keypoint_left, descriptor_left = detector.detectAndCompute(\\n image=left_img, mask=None\\n )\\n keypoint_right, descriptor_right = detector.detectAndCompute(\\n image=right_img, mask=None\\n )\\n\\n # Find the best matches\\n matches = matcher.match(\\n queryDescriptors=descriptor_left, trainDescriptors=descriptor_right\\n )\\n matches = sorted(matches, key=lambda x: x.distance)\\n\\n if DEBUG:\\n print(f\\\"Initial matches#\\\")\\n print(f\\\"match_lr: {len(matches)}\\\")\\n\\n # Match refinement with RANSAC\\n left_pts = np.float32([keypoint_left[m.queryIdx].pt for m in matches]).reshape(\\n -1, 1, 2\\n )\\n right_pts = np.float32([keypoint_right[m.trainIdx].pt for m in matches]).reshape(\\n -1, 1, 2\\n )\\n homography, mask = cv2.findHomography(left_pts, right_pts, cv2.RANSAC, 5.0)\\n matchesMask = mask.ravel().tolist()\\n\\n ransac_matches = []\\n for (m, keep) in zip(matches, matchesMask):\\n if keep:\\n ransac_matches.append(m)\\n\\n if DEBUG:\\n print(f\\\"After RANSAC:\\\")\\n print(f\\\"match: {len(ransac_matches)}\\\")\\n ransac_img = cv2.drawMatches(\\n left_img,\\n keypoint_left,\\n right_img,\\n keypoint_right,\\n ransac_matches,\\n None,\\n flags=2,\\n )\\n cv2.imwrite(\\\"./debug/RANSAC.png\\\", ransac_img)\\n\\n # Get the range of disparity\\n disp_min = np.inf\\n disp_max = 0\\n\\n for m in ransac_matches:\\n # Keypoints coordinates\\n pt_left = np.float32(keypoint_left[m.queryIdx].pt)\\n pt_right = np.float32(keypoint_right[m.trainIdx].pt)\\n\\n # Compute the distance\\n dist = int(abs(pt_left - pt_right)[0])\\n\\n if disp_max < dist:\\n disp_max = dist\\n\\n # Ceiling the number\\n disp_max += 1\\n\\n return disp_max\",\n \"def illustrate_matching(self):\\n work_dir = self.work_dir\\n draw_keys_oriented(work_dir+'matching_keys_im0.txt',\\n work_dir+'input_0.orig.png',\\n work_dir+'matching_keys_im0.png')\\n draw_keys_oriented(work_dir+'matching_keys_im1.txt',\\n work_dir+'input_1.orig.png',\\n work_dir+'matching_keys_im1.png')\\n draw_matches(work_dir+'matches.txt',\\n work_dir+'input_0.orig.png',\\n work_dir+'input_1.orig.png',\\n work_dir+'OUTmatches.png')\\n return 1\",\n \"def compute_average_precision_per_class(num_true_cases, gt_boxes,\\n\\t\\t\\t\\t\\t\\t\\t\\t\\t\\tprediction_file, iou_threshold, use_2007_metric):\\n\\twith open(prediction_file) as f:\\n\\t\\timage_ids = []\\n\\t\\tboxes = []\\n\\t\\tscores = []\\n\\t\\tfor line in f:\\n\\t\\t\\tt = line.rstrip().split(\\\" \\\")\\n\\t\\t\\timage_ids.append(t[0])\\n\\t\\t\\tscores.append(float(t[1]))\\n\\t\\t\\tbox = torch.tensor([float(v) for v in t[2:]]).unsqueeze(0)\\n\\t\\t\\tbox -= 1.0 # convert to python format where indexes start from 0\\n\\t\\t\\tboxes.append(box)\\n\\t\\tscores = np.array(scores)\\n\\t\\tsorted_indexes = np.argsort(-scores)\\n\\t\\tboxes = [boxes[i] for i in sorted_indexes]\\n\\t\\timage_ids = [image_ids[i] for i in sorted_indexes]\\n\\t\\ttrue_positive = np.zeros(len(image_ids))\\n\\t\\tfalse_positive = np.zeros(len(image_ids))\\n\\t\\tmatched = set()\\n\\t\\tfor i, image_id in enumerate(image_ids):\\n\\t\\t\\tbox = boxes[i]\\n\\t\\t\\tif image_id not in gt_boxes:\\n\\t\\t\\t\\tfalse_positive[i] = 1\\n\\t\\t\\t\\tcontinue\\n\\n\\t\\t\\tgt_box = gt_boxes[image_id]\\n\\t\\t\\tious = box_utils.iou_of(box, gt_box)\\n\\t\\t\\tmax_iou = torch.max(ious).item()\\n\\t\\t\\tmax_arg = torch.argmax(ious).item()\\n\\t\\t\\tif max_iou > iou_threshold:\\n\\t\\t\\t\\tif (image_id, max_arg) not in matched:\\n\\t\\t\\t\\t\\ttrue_positive[i] = 1\\n\\t\\t\\t\\t\\tmatched.add((image_id, max_arg))\\n\\t\\t\\t\\telse:\\n\\t\\t\\t\\t\\tfalse_positive[i] = 1\\n\\t\\t\\telse:\\n\\t\\t\\t\\tfalse_positive[i] = 1\\n\\n\\ttrue_positive = true_positive.cumsum()\\n\\tfalse_positive = false_positive.cumsum()\\n\\tprecision = true_positive / (true_positive + false_positive)\\n\\trecall = true_positive / num_true_cases\\n\\tif use_2007_metric:\\n\\t\\treturn measurements.compute_voc2007_average_precision(precision, recall)\\n\\telse:\\n\\t\\treturn measurements.compute_average_precision(precision, recall)\",\n \"def test_compare_classifiers_performance_subset_vis_api(experiment_to_use):\\n experiment = experiment_to_use\\n probabilities = experiment.probabilities\\n viz_outputs = (\\\"pdf\\\", \\\"png\\\")\\n with TemporaryDirectory() as tmpvizdir:\\n for viz_output in viz_outputs:\\n vis_output_pattern_pdf = tmpvizdir + f\\\"/*.{viz_output}\\\"\\n visualize.compare_classifiers_performance_subset(\\n [probabilities, probabilities],\\n experiment.ground_truth,\\n experiment.ground_truth_metadata,\\n experiment.output_feature_name,\\n top_n_classes=[6],\\n labels_limit=0,\\n subset=\\\"ground_truth\\\",\\n model_namess=[\\\"Model1\\\", \\\"Model2\\\"],\\n output_directory=tmpvizdir,\\n file_format=viz_output,\\n )\\n figure_cnt = glob.glob(vis_output_pattern_pdf)\\n assert 1 == len(figure_cnt)\",\n \"def test_compare_classifiers_predictions_distribution_vis_api(experiment_to_use):\\n experiment = experiment_to_use\\n predictions = experiment.predictions_num\\n viz_outputs = (\\\"pdf\\\", \\\"png\\\")\\n with TemporaryDirectory() as tmpvizdir:\\n for viz_output in viz_outputs:\\n vis_output_pattern_pdf = tmpvizdir + f\\\"/*.{viz_output}\\\"\\n visualize.compare_classifiers_predictions_distribution(\\n [predictions, predictions],\\n experiment.ground_truth,\\n experiment.ground_truth_metadata,\\n experiment.output_feature_name,\\n labels_limit=0,\\n model_names=[\\\"Model1\\\", \\\"Model2\\\"],\\n output_directory=tmpvizdir,\\n file_format=viz_output,\\n )\\n figure_cnt = glob.glob(vis_output_pattern_pdf)\\n assert 1 == len(figure_cnt)\",\n \"def ransac_homography(keypoints1, keypoints2, matches, sampling_ratio=0.5, n_iters=500, delta=20):\\n N = matches.shape[0]\\n n_samples = int(N * sampling_ratio)\\n\\n matched1_unpad = keypoints1[matches[:,0]]\\n matched2_unpad = keypoints2[matches[:,1]]\\n\\n\\n max_inliers = np.zeros(N)\\n n_inliers = 0\\n\\n # RANSAC iteration start\\n ### YOUR CODE HERE\\n for i in range(n_iters):\\n random_sample = np.random.choice(N, n_samples)\\n src = matched1_unpad[random_sample] # leave unpadded, is padded in compute homography\\n dst = matched2_unpad[random_sample]\\n\\n # Get H using DLT\\n H_matrix = compute_homography(src, dst)\\n\\n # Matched keypoints are considered inliers if the projected point from one image lies within distance δ to the matched point on the other point.\\n\\n projection = transform_homography(matched1_unpad, H_matrix) # x point, y point\\n distances = np.sqrt(np.sum(np.square(projection-matched2_unpad), axis=1))\\n if np.sum(distances < delta) > n_inliers:\\n max_inliers = distances < delta\\n n_inliers = np.sum(distances < delta)\\n \\n # Get final H using maximum inliers indices and src dst values\\n src = matched1_unpad[max_inliers]\\n dst = matched2_unpad[max_inliers]\\n H = compute_homography(src, dst)\\n \\n ### END YOUR CODE\\n return H, matches[max_inliers]\",\n \"def evaluate_rouge_match(ref_data, pred_data, align=\\\"rouge\\\"):\\n\\n scorer = RougeScorer([\\\"rougeL\\\"])\\n agg = scoring.BootstrapAggregator(confidence_interval=0.95)\\n sample_evaluated = 0\\n\\n for tid in tqdm(pred_data):\\n cur_sys = pred_data[tid][\\\"pred\\\"]\\n if not tid in ref_data: continue\\n cur_ref_lst = [x[\\\"tgt_words\\\"] for x in ref_data[tid]]\\n if len(cur_ref_lst) > 0:\\n if align == \\\"rouge\\\":\\n best_rouge_scrs = __rouge_pairs(cur_sys, cur_ref_lst, scorer)\\n agg.add_scores(best_rouge_scrs)\\n else:\\n best_paired_ref, _ = utils.pair_ref_sys_by_coverage(cur_sys, cur_ref_lst)\\n agg.add_scores(scorer.score(best_paired_ref, cur_sys))\\n sample_evaluated += 1\\n\\n print(sample_evaluated)\\n result = agg.aggregate()\\n for rouge_type in result:\\n print(rouge_type)\\n print(result[rouge_type].mid)\\n\\n print(agg.aggregate())\\n return\",\n \"def calculate_metrics(predictions, expected):\\n # type: (np.ndarray, np.ndarray) -> (float, float, float)\\n clients_count = predictions.shape[0]\\n products_count = predictions.shape[1]\\n\\n true_positive = 0.0\\n true_negative = 0.0\\n false_positive = 0.0\\n false_negative = 0.0\\n\\n total = float(clients_count * products_count)\\n\\n for c in range(0, clients_count):\\n for p in range(0, products_count):\\n if predictions[c, p] == expected[c, p]:\\n if predictions[c, p] == 1:\\n true_positive += 1\\n else:\\n true_negative += 1\\n else:\\n if predictions[c, p] == 1:\\n false_positive += 1\\n else:\\n false_negative += 1\\n\\n accuracy = float(true_positive + true_negative) / total\\n if true_positive + false_positive == 0:\\n precision = 0\\n else:\\n precision = true_positive / float(true_positive + false_positive)\\n\\n if true_positive + false_negative == 0:\\n recall = 0\\n else:\\n recall = true_positive / float(true_positive + false_negative)\\n\\n return accuracy, precision, recall\",\n \"def match2(img1, img2, coordinates1, coordinates2, PATCH_SIZE):\\n\\n\\tpossible_matches = pd.DataFrame(columns=['feature1', 'feature2', 'diff'])\\n\\n\\t# iteration through all the possible pairs of features from img1 and img2\\n\\tfor (feature1, feature2) in product(coordinates1, coordinates2):\\n\\t\\tpatch1 = make_patch(feature1, PATCH_SIZE, img1)\\n\\t\\tpatch2 = make_patch(feature2, PATCH_SIZE, img2)\\n\\t\\tdiff = count_difference(patch1, patch2)\\n\\t\\t# print(feature1, feature2, diff)\\n\\t\\t# next line is weird, don't ask\\n\\t\\tpossible_matches = possible_matches.append({'feature1': feature1, 'feature2': feature2, 'diff': diff}, ignore_index=True)\\n\\n\\n\\t# sorting the possible_matches according to their difference and resetin the index\\n\\t# inplace=True means that changes will be performed in-place, so they will change the variable and there is no need to reassign\\n\\tpossible_matches.sort_values(['diff'], inplace=True)\\n\\tpossible_matches.reset_index(inplace=True, drop=True)\\n\\t# print(\\\"SORTED\\\")\\n\\t# print(possible_matches)\\n\\n\\t# adding only the best matches\\n\\tmatches = []\\n\\twhile not possible_matches.empty:\\n\\t\\tbest_match = possible_matches.loc[0]\\n\\t\\tmatches.append(list(best_match))\\n\\t\\t# creating a list of indicies to be dropped\\n\\t\\ttrash = []\\n\\n\\t\\t#finding and deleting rows with features that are already used and reseting the index\\n\\t\\tfor i, (f1, f2, d) in enumerate(possible_matches.values):\\n\\t\\t \\n\\t\\t\\tif np.all(f1 == best_match[0]) or np.all(f2 == best_match[1]):\\n\\t\\t\\t\\ttrash.append(i)\\n\\t\\n\\t\\tfor i in trash:\\n\\t\\t\\tpossible_matches.drop(i, inplace=True)\\n\\t\\tpossible_matches.reset_index(inplace=True, drop=True)\\n\\t\\t# print(possible_matches)\\n\\n\\treturn matches\",\n \"def detail_matching(self):\\n paradic = self.cfg['param']['paradic']\\n work_dir = self.work_dir\\n \\n x = float(self.cfg['param']['x']) # selected pixel in the first image\\n y = float(self.cfg['param']['y'])\\n \\n # sift parameters\\n # number of bins in the orientation histogram\\n n_bins = int(paradic['n_bins']) \\n n_hist = int(paradic['n_hist']) \\n # descriptor of n_hist X n_hist weighted histograms with n_ori\\n n_ori = int(paradic['n_ori']) \\n delta_min = float(paradic['delta_min'])\\n sigma_min = float(paradic['sigma_min'])\\n sigma_in = float(paradic['sigma_in'])\\n lambda_ori = float(paradic['lambda_ori'])\\n lambda_descr = float(paradic['lambda_descr'])\\n #threshold defining reference orientations\\n n_spo = int(paradic['n_spo'])\\n \\n # Read feature vectors from output files\\n if (os.path.getsize(work_dir+'OUTmatches.txt') > 0 ):\\n pairdata = find_nearest_keypoint(work_dir+'OUTmatches.txt', y, x)\\n \\n illustrate_pair(pairdata, n_bins, n_hist, n_ori, work_dir)\\n\\n \\n # Read keys coordinates.\\n d = 6+n_bins+n_hist*n_hist*n_ori # size of keydata inside pairdata\\n v = n_hist*n_hist*n_ori\\n [x1, y1, sigma1, theta1] = [float(x) for x in pairdata[0:4]]\\n [o1, s1] = [float(x) for x in pairdata[4+v:4+v+2]]\\n [x2a, y2a, sigma2a, theta2a] = [float(x) for x in pairdata[d:d+4]]\\n [o2a, s2a] = [float(x) for x in pairdata[d+4+v:d+4+v+2]]\\n [x2b, y2b, sigma2b, theta2b] = \\\\\\n [float(x) for x in pairdata[2*d:2*d+4]]\\n [o2b, s2b] = [float(x) for x in pairdata[2*d+4+v:2*d+4+v+2]]\\n \\n draw_one_match(pairdata,\\n work_dir+'input_0.png',\\n work_dir+'input_1.png',\\n d,\\n lambda_ori,\\n lambda_descr,\\n n_hist,\\n work_dir+'OUTonepair.png')\\n \\n \\n # Extract thumbnails.\\n # keypoint 1 (image 1)\\n print ' '.join(['demo_extract_patch', work_dir+'input_0.png',\\n str(x1), str(y1), str(sigma1), str(theta1), str(o1), str(s1),\\n str(delta_min), str(sigma_min), str(sigma_in), str(n_spo),\\n str(lambda_ori), str(lambda_descr), str(n_hist),\\n work_dir+\\\"detail_im1\\\"])\\n proc = self.run_proc(['demo_extract_patch', work_dir+'input_0.png',\\n str(x1), str(y1), str(sigma1), str(theta1), str(o1), str(s1),\\n str(delta_min), str(sigma_min), str(sigma_in), str(n_spo),\\n str(lambda_ori), str(lambda_descr), str(n_hist),\\n work_dir+\\\"detail_im1\\\"])\\n self.wait_proc(proc, timeout=self.timeout)\\n \\n # keypoint 2a (nearest neighbor in image 2)\\n print ' '.join(['demo_extract_patch', work_dir+'input_1.png',\\n str(x2a), str(y2a), str(sigma2a), str(theta2a), str(o2a), str(s2a),\\n str(delta_min), str(sigma_min), str(sigma_in), str(n_spo),\\n str(lambda_ori), str(lambda_descr), str(n_hist),\\n work_dir+\\\"detail_im2a\\\"])\\n proc = self.run_proc(['demo_extract_patch', work_dir+'input_1.png',\\n str(x2a), str(y2a), str(sigma2a), str(theta2a), str(o2a), str(s2a),\\n str(delta_min), str(sigma_min), str(sigma_in), str(n_spo),\\n str(lambda_ori), str(lambda_descr), str(n_hist),\\n work_dir+\\\"detail_im2a\\\"])\\n self.wait_proc(proc, timeout=self.timeout) \\n \\n # keypoint 2b (second nearest neighbor in image 2)\\n proc = self.run_proc(['demo_extract_patch', work_dir+'input_1.png',\\n str(x2b), str(y2b), str(sigma2b), str(theta2b), str(o2b), str(s2b),\\n str(delta_min), str(sigma_min), str(sigma_in), str(n_spo),\\n str(lambda_ori), str(lambda_descr), str(n_hist),\\n work_dir+\\\"detail_im2b\\\"])\\n self.wait_proc(proc, timeout=self.timeout) \\n \\n \\n return 1\",\n \"def compare_images(self, df):\\n score_list = []\\n elapsed_list = []\\n for index,row in df.iterrows():\\n start_time = time.time()\\n image1 = imread(row['image1'],as_gray=\\\"True\\\")\\n image2 = imread(row['image2'],as_gray=\\\"True\\\")\\n if(image1.shape != image2.shape):\\n image2 = resize(image2, (image1.shape)) # if images are of different dimensions, crop the second image based on first\\n score = structural_similarity(image1, image2)\\n new_score = self.__convert_similarity_score(score)\\n elapsed_time = (time.time() - start_time)\\n score_list.append(round(new_score,2))\\n elapsed_list.append(elapsed_time)\\n output_df = self.__create_output_csv(df, score_list, elapsed_list)\\n return output_df\",\n \"def calculate_score(label_dict, queries, results):\\n\\n total_score = 0.0\\n\\n log_header('Individual image scores')\\n\\n # Calculate score for all images\\n for image_id in queries:\\n if image_id in results.keys():\\n # Run the score function\\n image_score = score(\\n label_dict=label_dict, target=image_id, selection=results[image_id]\\n )\\n else:\\n logging.error('No result generated for %s' % image_id)\\n\\n image_score = 0.0\\n\\n total_score += image_score\\n\\n logging.info('%s: %8.6f' % (image_id, image_score))\\n\\n log_header('Average score over %d images: %10.8f' % (\\n len(queries), total_score / len(queries)\\n ))\\n\\n return total_score\",\n \"def compare_results(self, benchmark_result, reference_benchmark_result):\\n pass\",\n \"def run_compare(self):\\n misses = []\\n for row in range(self.data.get_number_of_rows()):\\n ref_sentence = self.data.get_row(row)[self.data.get_gold()]\\n results = {}\\n for team, team_sentence in self.data.get_row_for_teams(self.evaluator.get_teams(row), row).iteritems():\\n results[team] = self.get_sentence_score(ref_sentence, team_sentence)\\n misses.append(self.evaluator.compare_all(results, row))\\n print np.median(misses), np.mean(misses)\\n return np.median(misses)\",\n \"def eval_feature_verify(query_features, gallery_features, query_lst, gallery_lst, metric=\\\"cosine\\\"):\\n if metric not in [\\\"euclidean\\\", \\\"cosine\\\"]:\\n raise ValueError(\\\"Invalid metric! \\\")\\n\\n num_query = len(query_lst)\\n num_gallery = len(gallery_lst)\\n\\n if metric == \\\"cosine\\\":\\n gallery_features = normalize(gallery_features, axis=1)\\n query_features = normalize(query_features, axis=1)\\n\\n if metric == \\\"euclidean\\\":\\n dist = euclidean_distances(gallery_features, query_features).squeeze()\\n else:\\n dist = np.dot(gallery_features, query_features.T)\\n print(dist[:].max())\\n print(dist[:].min())\\n\\n gallery_cam_lst = np.array([x.cam_id for x in gallery_lst], dtype=np.int32)\\n gallery_id_lst = np.array([x.class_id for x in gallery_lst], dtype=np.int32)\\n query_cam_lst = np.array([x.cam_id for x in query_lst], dtype=np.int32)\\n query_id_lst = np.array([x.class_id for x in query_lst], dtype=np.int32)\\n\\n # same id\\n gallery_id_lst_tile = np.tile(gallery_id_lst, (num_query, 1))\\n query_id_lst_tile = np.tile(query_id_lst, (num_gallery, 1))\\n label_mat = query_id_lst_tile == gallery_id_lst_tile.T # gallery num * query num\\n label_mat = label_mat.astype(np.int)\\n\\n # same camera and id\\n gallery_cam_lst_tile = np.tile(gallery_cam_lst, (num_query, 1))\\n query_cam_lst_tile = np.tile(query_cam_lst, (num_gallery, 1))\\n # same camera, gallery num * query num\\n label_cam_mat = query_cam_lst_tile == gallery_cam_lst_tile.T\\n # filter by same id\\n label_cam_mat = label_cam_mat * label_mat\\n # same id but under same camera are labeled as -1,\\n label_mat -= 2 * label_cam_mat.astype(np.int)\\n\\n ind = np.where(label_mat == 0)\\n neg_dist = dist[ind]\\n ind = np.where(label_mat == 1)\\n pos_dist = dist[ind]\\n\\n new_dist = np.concatenate([pos_dist, neg_dist])\\n new_label = np.concatenate([np.ones_like(pos_dist), np.zeros_like(neg_dist)])\\n dist_label = zip(new_dist, new_label)\\n dist_label_new = np.array(sorted(dist_label, key=lambda d: -d[0]))\\n\\n # ap = []\\n n_good = np.sum(dist_label_new[:, 1])\\n n_bad = np.sum(1 - dist_label_new[:, 1])\\n print(n_good)\\n for thresh in np.arange(0.90, 1, 0.01):\\n # print(thresh)\\n p = dist_label_new[:, 0] >= thresh\\n if np.sum(p) == 0:\\n recall, precision, f1 = 0, 0, 0\\n else:\\n tp = np.logical_and(dist_label_new[:, 1], p)\\n recall = np.sum(tp) / n_good\\n precision = np.float(np.sum(tp)) / np.sum(p)\\n\\n f1 = 2*recall*precision/(recall+precision)\\n\\n print('%.4f\\\\t%.4f\\\\t%.4f' % (recall, precision, f1))\\n\\n # ap.append(compute_ap(dist_label_new, thresh))\\n # print(ap)\",\n \"def matcher(features1, features2):\\n #TODO: write a matching function\\n #Performing the L2-Norm\\n new_features1=[]\\n new_features2=[]\\n for itr in range(5):\\n [rootOfSquare1,rootOfSquare2] = sumOfSquares(features1[itr],features2[itr])\\n new_features1.append(np.array(features1[itr])/rootOfSquare1)\\n new_features2.append(np.array(features2[itr])/rootOfSquare2)\\n indices = []\\n for itr in range(len(new_features1)):\\n findMinDist=[]\\n #findMaxCosineVal=[]\\n for itr2 in range(len(new_features2)):\\n f1 = new_features1[itr]\\n f2 = new_features2[itr2]\\n\\n #For evaluating the cosine similarity\\n # [rootOfSquare1,rootOfSquare2] = sumOfSquares(f1,f2)\\n # numerator = np.array(f1)*np.array(f2)\\n # numeratorSum = sum(numerator)\\n # denominator = rootOfSquare1*rootOfSquare2\\n # cosine = np.divide(numeratorSum,denominator)\\n # findMaxCosineVal.append(cosine)\\n\\n #For evaluating the similarity based on euclidean distance\\n Dist = np.array(f1) - np.array(f2)\\n sum=0\\n for i in Dist:\\n sum=sum+math.pow(i,2)\\n rootOfSum = math.sqrt(sum)\\n findMinDist.append(rootOfSum)\\n # print \\\"itr: \\\", itr, \\\" Matching scores: \\\", findMaxCosineVal\\n # bestMatch = findMaxCosineVal.index(max(findMaxCosineVal))\\n bestMatch = findMinDist.index(min(findMinDist))\\n indices.append([itr,bestMatch])\\n return indices\",\n \"def compute(self, result_file_dict):\\r\\n for part in self.parts:\\r\\n #=====================Need to change, temporal=========================\\r\\n if part == 'train':\\r\\n continue # because the train not have the label\\r\\n #=======================================================================\\r\\n gt = self.gt_dict[part]\\r\\n result_file = result_file_dict[part]\\r\\n # import ipdb; ipdb.set_trace()\\r\\n for key, item in result_file.items():\\r\\n self._result_name = item\\r\\n # score_records, num_videos = self.load_results(result_file)\\r\\n score_records, num_videos = self.load_results(item)\\r\\n logger.info(f'Compute Metric of {item}')\\r\\n assert num_videos == len(gt), f'the number of saved videos does not match the ground truth, {num_videos} != {len(gt)}'\\r\\n temp_result = self.eval_method(score_records, gt, str(key))\\r\\n if temp_result > self.optimal_resulst:\\r\\n self.optimal_resulst = temp_result\\r\\n \\r\\n return self.optimal_resulst\",\n \"def evaluate(ground_truth, prediction):\\n\\n def prfs_to_dict(prfs):\\n \\\"\\\"\\\"Returns a precision_recall_fscore_support() result as a dict.\\\"\\\"\\\"\\n return {\\\"precision\\\": prfs[0], \\\"recall\\\": prfs[1], \\\"fscore\\\": prfs[2]}\\n\\n results = {}\\n items_count = len(ground_truth)\\n\\n # accuracy\\n accuracy = accuracy_score(ground_truth, prediction)\\n results[\\\"accuracy\\\"] = accuracy\\n\\n # confusion matrix\\n categories = set(ground_truth) | set(prediction)\\n confusions = {\\n gold: {pred: 0 for pred in categories} for gold in categories\\n }\\n for g, p in zip(ground_truth, prediction):\\n confusions[g][p] += 1\\n results[\\\"confusions\\\"] = confusions\\n\\n # class wise precision, recall & f1\\n classwise = precision_recall_fscore_support(\\n ground_truth, prediction, average=None, warn_for=()\\n )\\n results[\\\"true_cat_dist\\\"] = list(classwise[-1])\\n results[\\\"classwise\\\"] = {\\n str(cl): prfs_to_dict(\\n [classwise[0][cl], classwise[1][cl], classwise[2][cl]]\\n )\\n for cl in categories\\n }\\n\\n # average precision, recall & f1\\n results[\\\"macro_avg\\\"] = prfs_to_dict(\\n precision_recall_fscore_support(\\n ground_truth,\\n prediction,\\n average=\\\"macro\\\",\\n pos_label=None,\\n warn_for=(),\\n )\\n )\\n results[\\\"micro_avg\\\"] = prfs_to_dict(\\n precision_recall_fscore_support(\\n ground_truth,\\n prediction,\\n average=\\\"micro\\\",\\n pos_label=None,\\n warn_for=(),\\n )\\n )\\n results[\\\"weigh_avg\\\"] = prfs_to_dict(\\n precision_recall_fscore_support(\\n ground_truth,\\n prediction,\\n average=\\\"weighted\\\",\\n pos_label=None,\\n warn_for=(),\\n )\\n )\\n\\n # marginals\\n gold_category_distribution = {\\n g: sum([confusions[g][p] for p in categories]) for g in categories\\n }\\n pred_category_distribution = {\\n p: sum([confusions[g][p] for g in categories]) for p in categories\\n }\\n\\n # kappa\\n expected_agreement_fleiss = sum(\\n [\\n (\\n (gold_category_distribution[c] + pred_category_distribution[c])\\n / (2.0 * items_count)\\n )\\n ** 2\\n for c in categories\\n ]\\n )\\n expected_agreement_cohen = sum(\\n [\\n (float(gold_category_distribution[c]) / items_count)\\n * (float(pred_category_distribution[c]) / items_count)\\n for c in categories\\n ]\\n )\\n kappa_fleiss = (\\n 1.0\\n * (accuracy - expected_agreement_fleiss)\\n / (1 - expected_agreement_fleiss)\\n )\\n kappa_cohen = (\\n 1.0\\n * (accuracy - expected_agreement_cohen)\\n / (1 - expected_agreement_cohen)\\n )\\n results[\\\"k_fleiss\\\"] = {\\n \\\"k\\\": kappa_fleiss,\\n \\\"AE\\\": expected_agreement_fleiss,\\n \\\"AO\\\": accuracy,\\n }\\n results[\\\"k_cohen\\\"] = {\\n \\\"k\\\": kappa_cohen,\\n \\\"AE\\\": expected_agreement_cohen,\\n \\\"AO\\\": accuracy,\\n }\\n\\n return results\",\n \"def test_compare_performance_vis_api(experiment_to_use):\\n experiment = experiment_to_use\\n # extract test stats only\\n test_stats = experiment.test_stats_full\\n viz_outputs = (\\\"pdf\\\", \\\"png\\\")\\n with TemporaryDirectory() as tmpvizdir:\\n for viz_output in viz_outputs:\\n vis_output_pattern_pdf = tmpvizdir + f\\\"/*.{viz_output}\\\"\\n visualize.compare_performance(\\n [test_stats, test_stats],\\n output_feature_name=None,\\n model_names=[\\\"Model1\\\", \\\"Model2\\\"],\\n output_directory=tmpvizdir,\\n file_format=viz_output,\\n )\\n figure_cnt = glob.glob(vis_output_pattern_pdf)\\n assert 1 == len(figure_cnt)\",\n \"def evaluate_detections(self, ap, phrase_counts, top1acc, total_aug, top1acc_aug, top1acc_oracle, top1acc_aug_oracle):\\n # organize mAP by the number of occurrences\\n count_thresholds = cfg.TEST.PHRASE_COUNT_THRESHOLDS\\n mAP = np.zeros(len(count_thresholds))\\n occurrences = np.zeros_like(mAP)\\n samples = np.zeros_like(mAP)\\n samples_aug = np.zeros_like(mAP)\\n acc = np.zeros_like(mAP)\\n acc_aug = np.zeros_like(mAP)\\n oracle = np.zeros_like(mAP)\\n oracle_aug = np.zeros_like(mAP)\\n for phrase, phrase_index in self._phrase_to_ind.iteritems():\\n n_occurrences = phrase_counts[phrase_index]\\n if n_occurrences < 1:\\n continue\\n\\n train_count = 0\\n if phrase in self._train_counts:\\n train_count = self._train_counts[phrase]\\n\\n count_index = min(np.where(train_count <= count_thresholds)[0])\\n mAP[count_index] += ap[phrase_index]\\n occurrences[count_index] += 1\\n samples[count_index] += n_occurrences\\n acc[count_index] += top1acc[phrase_index]\\n acc_aug[count_index] += top1acc_aug[phrase_index]\\n samples_aug[count_index] += total_aug[phrase_index]\\n oracle[count_index] += top1acc_oracle[phrase_index]\\n oracle_aug[count_index] += top1acc_aug_oracle[phrase_index]\\n\\n mAP = mAP / occurrences\\n thresh_string = '\\\\t'.join([str(thresh) for thresh in count_thresholds])\\n print('\\\\nThresholds: \\\\t' + thresh_string + '\\\\tOverall')\\n\\n ap_string = '\\\\t'.join(['%.1f' % round(t * 100, 2) for t in mAP])\\n print('AP: \\\\t' + ap_string + '\\\\t%.1f' % round(np.mean(mAP) * 100, 2))\\n\\n n_total = np.sum(samples)\\n n_aug = np.sum(total_aug)\\n loc_acc = np.sum(acc) / (n_total - n_aug)\\n group_acc = acc / (samples - samples_aug)\\n acc_string = '\\\\t'.join(['%.1f' % round(t * 100, 2) for t in group_acc])\\n print('Loc Acc:\\\\t' + acc_string + '\\\\t%.1f' % round(loc_acc * 100, 2))\\n \\n loc_acc = np.sum(oracle) / (n_total - n_aug)\\n group_acc = oracle / (samples - samples_aug)\\n acc_string = '\\\\t'.join(['%.1f' % round(t * 100, 2) for t in group_acc])\\n print('Oracle Acc:\\\\t' + acc_string + '\\\\t%.1f' % round(loc_acc * 100, 2))\\n if cfg.AUGMENTED_POSITIVE_PHRASES:\\n loc_acc = (np.sum(acc) + np.sum(acc_aug)) / n_total\\n group_acc = (acc+acc_aug) / samples\\n acc_string = '\\\\t'.join(['%.1f' % round(t * 100, 2) for t in group_acc])\\n print('Aug Loc Acc:\\\\t' + acc_string + '\\\\t%.1f' % round(loc_acc * 100, 2))\\n\\n loc_acc = (np.sum(oracle) + np.sum(oracle_aug)) / n_total\\n group_acc = (oracle+oracle_aug) / samples\\n acc_string = '\\\\t'.join(['%.1f' % round(t * 100, 2) for t in group_acc])\\n print('Oracle Aug Acc:\\\\t' + acc_string + '\\\\t%.1f' % round(loc_acc * 100, 2))\\n\\n occ_string = '\\\\t'.join(['%i' % occ for occ in occurrences])\\n print('Per Thresh Cnt:\\\\t' + occ_string + '\\\\t%i' % np.sum(occurrences))\\n\\n sample_string = '\\\\t'.join(['%i' % item for item in samples])\\n print('Instance Cnt: \\\\t' + sample_string + '\\\\t%i' % n_total)\\n\\n if cfg.TOP_K_PER_PHRASE > 1:\\n n_correct = np.sum([np.sum(item) for item in gt_labels])\\n acc = round((n_correct/n_total)*100, 2)\\n print('Portion of phrases with good boxes: %.2f\\\\n' % acc)\\n\\n return np.mean(mAP)\",\n \"def calc_TP_FP(groundtruth, eval_boxes, matches_predicate, confidence_threshold=-9e999):\\n TP = 0\\n FP = 0\\n FN = 0\\n for image_key in groundtruth.keys() + eval_boxes.keys():\\n gt_bboxes = set(groundtruth.get(image_key, []))\\n other_bboxes = set(box\\n for box in eval_boxes.get(image_key, [])\\n if box.confidence >= confidence_threshold\\n or box.confidence is None\\n )\\n for gt_bbox in gt_bboxes:\\n # Do any of their boxes match?\\n match = None\\n for other_box in other_bboxes:\\n if matches_predicate(gt_bbox, other_box):\\n match = other_box\\n if match:\\n other_bboxes.remove(match)\\n TP += 1\\n else:\\n FN += 1\\n FP += len(other_bboxes)\\n # if len(other_bboxes):\\n # print \\\"False positive on\\\", image_key\\n return {\\\"TP\\\": TP, \\\"FP\\\": FP, \\\"FN\\\": FN, \\\"TN\\\": 0}\",\n \"def match_mfovs_features(matcher_params, sec1_cache, sec2_cache, mfovs1, mfovs2):\\n \\n thread_local_store = ThreadLocalStorageLRU()\\n if 'matcher' in thread_local_store.keys():\\n matcher = thread_local_store['matcher']\\n else:\\n # Initialize the matcher, and store it in the local thread storage\\n matcher = FeaturesMatcher(BlobDetector2D.create_matcher, **matcher_params)\\n thread_local_store['matcher'] = matcher\\n \\n# matcher = getattr(threadLocal, 'matcher', None)\\n# if matcher is None:\\n# # Initialize the matcher, and store it in the local thread storage\\n# matcher = FeaturesMatcher(BlobDetector2D.create_matcher, **matcher_params)\\n# threadLocal.matcher = matcher\\n\\n def get_kps_descs(mfovs, sec_cache):\\n mfovs = list(mfovs)\\n if len(mfovs) == 1:\\n mfovs_kps = np.array(sec_cache[\\\"pre_match_blobs\\\"][mfovs[0]][0])\\n mfovs_descs = np.array(sec_cache[\\\"pre_match_blobs\\\"][mfovs[0]][1])\\n else:\\n mfovs_kps_arrays = []\\n mfovs_descs_arrays = []\\n for mfov in mfovs:\\n kps_descs = sec_cache[\\\"pre_match_blobs\\\"][mfov]\\n if len(kps_descs[0]) > 0:\\n mfovs_kps_arrays.append(kps_descs[0])\\n mfovs_descs_arrays.append(kps_descs[1])\\n if len(mfovs_kps_arrays) == 0:\\n mfovs_kps = np.array([])\\n mfovs_descs = np.array([])\\n elif len(mfovs_kps_arrays) == 1:\\n mfovs_kps = mfovs_kps_arrays[0]\\n mfovs_descs = mfovs_descs_arrays[0]\\n else:\\n mfovs_kps = np.vstack(mfovs_kps_arrays)\\n mfovs_descs = np.vstack(mfovs_descs_arrays)\\n return np.array(mfovs_kps), np.array(mfovs_descs)\\n\\n mfovs1_kps, mfovs1_descs = get_kps_descs(mfovs1, sec1_cache)\\n mfovs2_kps, mfovs2_descs = get_kps_descs(mfovs2, sec2_cache)\\n\\n model, filtered_matches = matcher.match_and_filter(mfovs1_kps, mfovs1_descs, mfovs2_kps, mfovs2_descs)\\n return mfovs1, model, filtered_matches\",\n \"def main():\\n num_matches = 1000\\n matchlist = [\\n [ RandomAI, RandomAI ],\\n #[ VectorAI, RandomAI ],\\n #[ VectorAI, VectorAI ],\\n #[ LeftmostAI, RandomAI ],\\n #[ LeftmostAI, VectorAI ],\\n #[ LeftmostAI, LeftmostAI ],\\n\\t\\t#[ RightmostAI, RandomAI ],\\n #[ RightmostAI, VectorAI ],\\n #[ RightmostAI, LeftmostAI ],\\n #[ RightmostAI, RightmostAI ]\\n ]\\n\\n x=0\\n while x < len(matchlist):\\n \\n print (\\\"Running \\\" + str(num_matches) + \\\" matches: \\\" + matchlist[x][0].__name__ + \\\" vs \\\" + matchlist[x][1].__name__)\\n counter = 0 \\n \\n while counter < num_matches:\\n counter = counter + 1\\n match = Match(player1_type=matchlist[x][0], player2_type=matchlist[x][1], param_print_game_status=False, param_matchgroup=x)\\n match.handle_next_move()\\n\\n if matchlist[x][0] != matchlist[x][1]:\\n print (\\\"Running \\\" + str(num_matches) + \\\" matches: \\\" + matchlist[x][1].__name__ + \\\" vs \\\" + matchlist[x][0].__name__)\\n counter = 0\\n\\n while counter < num_matches:\\n counter = counter + 1\\n match = Match(player1_type=matchlist[x][1], player2_type=matchlist[x][0], param_print_game_status=False, param_matchgroup=x)\\n match.handle_next_move()\\n x = x+1\\n \\n print (\\\"Matches Finished\\\")\",\n \"def evaluate(calculated_results, ground_truth, measure=''):\\n score = 0\\n truth_pairs = dict.fromkeys(pair[0] + ' ' + pair[1]\\n for pair in ground_truth.edges())\\n for pair in calculated_results:\\n pair_reversed = get_reversed_pair(pair)\\n if pair[0] in truth_pairs or pair_reversed in truth_pairs:\\n score += 1\\n accuracy = (score / len(truth_pairs)) * 100\\n print(\\\"The accuracy is %f %% when using measure: %s\\\" % (accuracy, measure))\",\n \"def evaluate_match_quality(matches, thresh=.7):\\n marr = np.zeros((len(matches), 2))\\n quality = np.zeros((len(matches), 1))\\n for i in range(len(matches)):\\n marr[i,0] = matches[i][0].distance\\n marr[i,1] = matches[i][1].distance\\n quality[i] = marr[i, 0] < thresh*marr[i, 1]\",\n \"def get_metrics(cfg, model, X_anchor, y_anchor, X_gal, y_gal, annoy_index, vec_dim):\\n rank10_acc = 0\\n rank5_acc = 0\\n rank1_acc = 0\\n avg_acc = 0\\n vote_res = 0\\n\\n l2 = []\\n for anchor in range(0, len(X_anchor)):\\n res = get_result(get_image_features(cfg, model, X_anchor[anchor]), annoy_index)\\n vote = defaultdict(int)\\n # Accuracy\\n correct = 0\\n for i in res[:10]:\\n vote[y_gal[i]] += 1\\n\\n max_key = max(vote, key=vote.get)\\n if max_key == y_anchor[anchor]:\\n vote_res += 1\\n \\n\\n for recomm in res[:10]:\\n if y_gal[recomm] == y_anchor[anchor]:\\n correct += 1 \\n\\n avg_acc += correct/len(res)\\n\\n # Mean Average Precision\\n l1 = []\\n for recomm in res[:10]:\\n if y_gal[recomm] == y_anchor[anchor]:\\n correct += 1\\n l1.append(1)\\n else:\\n l1.append(0)\\n l2.append(l1) \\n\\n # Rank10 Accuracy\\n for each_val in res[:10]:\\n if y_gal[each_val] == y_anchor[anchor]:\\n rank10_acc += 1\\n break\\n \\n # Rank5 Accuracy\\n for each_val in res[:5]:\\n if y_gal[each_val] == y_anchor[anchor]:\\n rank5_acc += 1\\n break\\n\\n # Rank1 Accuracy\\n for each_val in res[:1]:\\n if y_gal[each_val] == y_anchor[anchor]:\\n rank1_acc += 1\\n break\\n\\n print(\\\"Avg acc is :: {avg_acc}\\\".format(avg_acc = avg_acc/len(X_anchor)))\\n print(\\\"Rank 10 acc is :: {rank10_acc}\\\".format(rank10_acc = rank10_acc/len(X_anchor)))\\n print(\\\"Rank 5 acc is :: {rank5_acc}\\\".format(rank5_acc = rank5_acc/len(X_anchor)))\\n print(\\\"Rank 1 acc is :: {rank1_acc}\\\".format(rank1_acc = rank1_acc/len(X_anchor)))\\n print(\\\"Mean Avg Precision is :: {mAP}\\\".format(mAP=mean_average_precision(l2)))\\n print(\\\"Vote res :: \\\", vote_res/len(X_anchor))\\n\\n return rank1_acc/len(X_anchor), mean_average_precision(l2)\",\n \"def prediction(self, query_image_path_list, evaluation_method, demonstration_subdir, evaluation_subdir):\\n semantic_tag_unique_list = list(set(self.node_semantic_tag_list))\\n semantic_tag_total_count = np.array([self.node_semantic_tag_list.count(semantic_tag) for semantic_tag in semantic_tag_unique_list])\\n\\n es_semantic_tag_list = []\\n es_target_index_list = []\\n es_target_index_rank_list = []\\n query_feature_dist = []\\n node_index_list = []\\n for query_image_path in query_image_path_list:\\n query_image = cv2.imread(query_image_path)\\n query_image_batch = self.process_batch([query_image])\\n query_rolling_features_list = self.sess.run(self.rolling_features_list,\\n feed_dict = {self.images_placeholder: query_image_batch})\\n\\n metric_distance_list = []\\n for node_index in range(self.node_number):\\n node_metric_feature = self.node_metric_feature_list[node_index]\\n rolling_distance_list = []\\n for query_rolling_features in query_rolling_features_list:\\n rolling_distance = np.sqrt(np.sum(np.square(node_metric_feature - query_rolling_features)))\\n rolling_distance_list.append(rolling_distance)\\n metric_distance = min(rolling_distance_list)\\n metric_distance_list.append(metric_distance)\\n metric_distance_list = np.array(metric_distance_list)\\n\\n query_feature_dist.append(metric_distance_list)\\n\\n\\n node_index_rank = np.argsort(metric_distance_list)\\n node_index_list.append(node_index_rank[0])\\n es_target_index_rank = [self.node_id_list[node_index] for node_index in node_index_rank]\\n es_target_index_rank_list.append(es_target_index_rank)\\n es_target_index_list.append(es_target_index_rank[0])\\n es_semantic_tag_list.append(set([self.node_semantic_tag_list[node_index_rank[0]]]))\\n\\n if evaluation_method == 'find_target_position':\\n print(\\\"TARGET INDEX: \\\"+str(es_target_index))\\n print(\\\"semantic tag: \\\"+str(self.node_semantic_tag_list[es_target_index]))\\n print(\\\"\\\\n\\\")\\n return es_semantic_tag_list, es_target_index_list\\n\\n elif evaluation_method == 'target_position_precision-error':\\n return es_target_index_list\\n elif evaluation_method == 'target_position_distance-accuracy':\\n return es_target_index_list\\n elif evaluation_method == 'target_position_recall-N':\\n return es_target_index_rank_list\\n elif evaluation_method == 'target_position_precision-recall':\\n return [node_index_list, query_feature_dist]\\n else:\\n raise ValueError(\\\"unknown evaluation method\\\")\",\n \"def compare(args):\\n p = OptionParser(__doc__)\\n opts, args, iopts = p.set_image_options(args, figsize=\\\"15x5\\\")\\n\\n if len(args) != 1:\\n sys.exit(not p.print_help())\\n\\n datafile, = args\\n pf = datafile.rsplit(\\\".\\\", 1)[0]\\n fig, (ax1, ax2, ax3) = plt.subplots(ncols=3, nrows=1,\\n figsize=(iopts.w, iopts.h))\\n plt.tight_layout(pad=2)\\n\\n # Huntington risk allele\\n infected_thr = 40\\n ref_thr = 19\\n\\n # ax1: Multiple callers at lower range\\n df = pd.read_csv(\\\"Evaluation.csv\\\")\\n truth = df[\\\"Truth\\\"]\\n\\n ax1.plot(truth, df[\\\"Manta\\\"], 'bx-')\\n ax1.plot(truth, df[\\\"Isaac\\\"], 'yo-')\\n ax1.plot(truth, df[\\\"GATK\\\"], 'md-')\\n ax1.plot(truth, df[\\\"lobSTR\\\"], 'c+-')\\n ax1.plot(truth, truth, 'k--') # to show diagonal\\n\\n bbox = {'facecolor': 'tomato', 'alpha': .2, 'ec': 'w'}\\n pad = 2\\n ax1.axhline(infected_thr, color='tomato')\\n ax1.text(max(truth) - pad, infected_thr + pad, 'Risk threshold',\\n bbox=bbox, ha=\\\"right\\\")\\n ax1.axhline(ref_thr, color='tomato')\\n ax1.text(max(truth) - pad, ref_thr - pad, 'Reference repeat count',\\n bbox=bbox, ha=\\\"right\\\", va=\\\"top\\\")\\n\\n ax1.set_xlabel(r'Num of CAG repeats inserted ($\\\\mathit{h}$)')\\n ax1.set_ylabel('Num of CAG repeats called')\\n ax1.set_title(r'Simulated haploid $\\\\mathit{h}$')\\n ax1.legend(['Manta', 'Isaac', 'GATK', 'lobSTR', 'Truth'], loc='best')\\n\\n max_insert = 120\\n # ax2: lobSTR vs TREDPARSE with haploid model\\n lobstr_results = parse_results(\\\"lobstr_results_homo.txt\\\")\\n tredparse_results = parse_results(\\\"tredparse_results_homo.txt\\\")\\n truth = range(10, max_insert + 1)\\n lx, ly = zip(*lobstr_results)\\n tx, ty = zip(*tredparse_results)\\n\\n ax2.plot(lx, ly, 'c+-')\\n ax2.plot(tx, ty, 'gx-')\\n ax2.plot(truth, truth, 'k--')\\n\\n ax2.set_xlabel(r'Num of CAG repeats inserted ($\\\\mathit{h}$)')\\n ax2.set_ylabel('Num of CAG repeats called')\\n ax2.set_title(r'Simulated haploid $\\\\mathit{h}$')\\n ax2.legend(['lobSTR', 'TREDPARSE', 'Truth'], loc='best')\\n\\n pad *= 2\\n ax2.axhline(infected_thr, color='tomato')\\n ax2.text(max(truth) - pad, infected_thr + pad, 'Risk threshold',\\n bbox=bbox, ha=\\\"right\\\")\\n ax2.set_xlim(10, max_insert)\\n\\n # ax3: lobSTR vs TREDPARSE with haploid model\\n lobstr_results = parse_results(\\\"lobstr_results_het.txt\\\", exclude=20)\\n tredparse_results = parse_results(\\\"tredparse_results_het.txt\\\", exclude=20)\\n truth = range(10, max_insert + 1)\\n lx, ly = zip(*lobstr_results)\\n tx, ty = zip(*tredparse_results)\\n\\n ax3.plot(lx, ly, 'c+-')\\n ax3.plot(tx, ty, 'gx-')\\n ax3.plot(truth, truth, 'k--')\\n\\n ax3.set_xlabel(r'Num of CAG repeats inserted ($\\\\mathit{h}$)')\\n ax3.set_ylabel('Num of CAG repeats called')\\n ax3.set_title(r'Simulated diploid $\\\\mathit{20/h}$')\\n ax3.legend(['lobSTR', 'TREDPARSE', 'Truth'], loc='best')\\n ax3.axhline(infected_thr, color='tomato')\\n ax3.text(max(truth) - pad, infected_thr + pad, 'Risk threshold',\\n bbox=bbox, ha=\\\"right\\\")\\n ax3.set_xlim(10, max_insert)\\n\\n root = fig.add_axes([0, 0, 1, 1])\\n pad = .03\\n panel_labels(root, ((pad / 2, 1 - pad, \\\"A\\\"), (1 / 3., 1 - pad, \\\"B\\\"),\\n (2 / 3., 1 - pad, \\\"C\\\")))\\n normalize_axes(root)\\n\\n image_name = pf + \\\".\\\" + iopts.format\\n savefig(image_name, dpi=iopts.dpi, iopts=iopts)\",\n \"def statistics(img1_blobs, img2_blobs, matches):\\n\\tstatistics = {}\\n\\n\\tstatistics['#Interest Points in img1'] = len(img1_blobs)\\n\\tstatistics['#Interest Points in img2'] = len(img2_blobs)\\n\\tstatistics['Accepted Matches'] = len(matches)\\n\\tdissimilarity = [match[2] for match in matches]\\n\\tstatistics['Mean of accepted matches'] = sum(dissimilarity)/len(dissimilarity)\\n\\tstatistics['SD of accepted matches'] = np.std(dissimilarity)\\n\\treturn statistics\",\n \"def test_RawRun_process():\\n for style in test_runs:\\n test_runs[style].process()\\n # now compare all images with ref\\n ref = sorted(glob.glob('tests/data/processed_ref/*/*/*'))\\n outputs = sorted(glob.glob('tests/data/processed/*/*/*'))\\n for ref, out in zip(ref, outputs):\\n print ref\\n print out\\n assert_image_equal(ref, out)\",\n \"def generate_comparison_metrics(model: sklearn.linear_model,\\n x_test: pd.DataFrame,\\n y_test: pd.DataFrame,\\n y_predictions: pd.DataFrame = None):\\n\\n if y_predictions is not None:\\n if not y_predictions.isnull().values.all():\\n y_pred = y_predictions\\n elif 'themis_ml' in model.__module__:\\n y_pred = model.predict(x_test, x_test[privileged].values)\\n else:\\n y_pred = model.predict(x_test)\\n\\n accuracy = sklearn.metrics.accuracy_score(y_true=y_test, y_pred=y_pred)\\n precision = sklearn.metrics.precision_score(y_true=y_test, y_pred=y_pred)\\n recall = sklearn.metrics.recall_score(y_true=y_test, y_pred=y_pred)\\n auc = sklearn.metrics.roc_auc_score(y_true=y_test, y_score=y_pred)\\n tn, fp, fn, tp = sklearn.metrics.confusion_matrix(y_true=y_test, y_pred=y_pred).ravel()\\n tot_num = tn + fp + fn + tp\\n tnp, fpp, fnp, tpp = tn / tot_num, fp / tot_num, fn / tot_num, tp / tot_num\\n tpr = true_positive_rate(true_positives=tp, false_negatives=fn) # Also called 'sensitivity' and 'recall'\\n fpr = false_positive_rate(false_positives=fp, true_negatives=tn)\\n cost = false_positive_cost * fp + false_negative_cost * fn\\n output_list = [accuracy, precision, recall, auc, tnp, fpp, fnp, tpp, tpr, fpr, cost]\\n return output_list\",\n \"def painting_matchings_local_desc(query_imgs, db_imgs, text_masks, options):\\n tmp_img_format = []\\n tmp_mask_format = []\\n for i in range(len(query_imgs)):\\n for j in range(len(query_imgs[i])):\\n tmp_img_format.append(query_imgs[i][j])\\n tmp_mask_format.append(text_masks[i][j])\\n\\n ## Keypoint + local descriptor extraction\\n print(\\\"Obtaining query set descriptors\\\")\\n kp_qs = keypoint_extraction(tmp_img_format, tmp_mask_format, options)\\n desc_qs = descriptor_extraction(tmp_img_format, tmp_mask_format, kp_qs, options)\\n\\n print(\\\"Obtaining database descriptors\\\")\\n kp_db = keypoint_extraction(db_imgs, None, options)\\n desc_db = descriptor_extraction(db_imgs, None, kp_db, options)\\n\\n ## Match local descriptors \\n print(\\\"Comparing...\\\") \\n num_matches = match_descriptors_qs_db(tmp_img_format, db_imgs, desc_qs, desc_db, kp_qs, kp_db, options.km, options)\\n return num_matches\",\n \"def test_calculate_precision_happy_path(\\n mock_engine: DicomImagePiiVerifyEngine,\\n gt_labels: List[dict],\\n all_pos: List[dict],\\n expected_result: float,\\n):\\n # Act\\n test_precision = mock_engine.calculate_precision(gt_labels, all_pos)\\n\\n # Assert\\n assert test_precision == expected_result\",\n \"def PredictMatch(yourName, tgtName, teamModels, db):\\n\\n teamTgtPipe, teamYourPipe = QueryTeamData(tgtName, yourName, db)\\n featureCoefTgt, probTgt = PredictOp(teamTgtPipe, teamYourPipe, tgtName, teamModels)\\n featureCoefYour, probYour = PredictOp(teamYourPipe, teamTgtPipe, yourName, teamModels)\\n odds = round(probYour / (probYour + probTgt), 2)\\n\\n # In featureCoefYour, you want INCREASE those with POSTIVE COEF, DECREASE those with NEGATIVE COEF\\n # In featureCoefTgt, you want to do the opposite\\n\\n # reverse both the sign of the coef, and '_op' in features so as to be the same with featureCoefYour\\n featureCoefTgt['coef'] = - featureCoefTgt['coef']\\n featureCoefTgt.features = [ii[:-3] if \\\"_op\\\" in ii else ii + '_op' for ii in featureCoefTgt.features]\\n\\n # Combine only the most important 10 features\\n # featureBoth = featureCoefTgt[11:].append(featureCoefYour[11:])\\n\\n # Combine only all the most important features\\n featureBoth = featureCoefTgt.append(featureCoefYour)\\n\\n # get action recommendations\\n # Somehow the pandas here uses a deprecated para cols, instaed of the new one subset\\n #featureBoth.drop_duplicates(subset = 'features', take_last = True, inplace = True)\\n featureBoth.drop_duplicates(cols = 'features', take_last = True, inplace = True)\\n actions, featureImprove = GetActions(featureBoth)\\n Imp = 0.1\\n oddsNew = ImprovedScore(tgtName, yourName, teamModels, featureImprove, teamTgtPipe, teamYourPipe, Imp)\\n\\n return odds, oddsNew, actions\",\n \"def ic_compare(args):\\n\\n config_file = args.setupfn\\n conf_base = os.path.basename(config_file).split('.')[0]\\n statfile = os.path.join(args.outputdir,\\n \\\"{}_radvel.stat\\\".format(conf_base))\\n\\n status = load_status(statfile)\\n\\n P, post = radvel.utils.initialize_posterior(config_file,\\n decorr=args.decorr)\\n\\n assert status.getboolean('fit', 'run'), \\\\\\n \\\"Must perform max-liklihood fit before running Information Criteria comparisons\\\"\\n post = radvel.posterior.load(status.get('fit', 'postfile'))\\n\\n choices = ['nplanets', 'e', 'trend', 'jit', 'gp']\\n statsdictlist = []\\n paramlist = []\\n compareparams = args.type\\n\\n ipost = copy.deepcopy(post)\\n\\n if args.simple:\\n statsdictlist += radvel.fitting.model_comp(ipost, params=[], verbose=args.verbose)\\n else:\\n if hasattr(args, 'fixjitter') and args.fixjitter:\\n for param in ipost.params:\\n if len(param) >= 3 and param[0:3] == 'jit':\\n ipost.params[param].vary = False\\n\\n for compareparam in compareparams:\\n assert compareparam in choices, \\\\\\n \\\"Valid parameter choices for 'ic -t' are combinations of: \\\"\\\\\\n + \\\" \\\".join(choices)\\n paramlist.append(compareparam)\\n if hasattr(args, 'mixed') and not args.mixed:\\n statsdictlist += radvel.fitting.model_comp(ipost, params=[compareparam], verbose=args.verbose)\\n if hasattr(args, 'mixed') and not args.mixed:\\n new_statsdictlist = []\\n for dicti in statsdictlist:\\n anymatch = False\\n for seendict in new_statsdictlist:\\n if collections.Counter(dicti['Free Params'][0]) == \\\\\\n collections.Counter(seendict['Free Params'][0]):\\n anymatch = True\\n continue\\n if not anymatch:\\n new_statsdictlist.append(dicti)\\n statsdictlist = new_statsdictlist\\n\\n if not hasattr(args, 'mixed') or (hasattr(args, 'mixed') and args.mixed):\\n statsdictlist += radvel.fitting.model_comp(ipost, params=paramlist, verbose=args.verbose)\\n\\n savestate = {'ic': statsdictlist}\\n save_status(statfile, 'ic_compare', savestate)\",\n \"def run_test(files, full=False, threshold=2):\\n columns = ['detector', 'image', 'common', 'repeat']\\n data = []\\n\\n if full:\\n det_s = {**DetectorDescriptor.detectors, **DetectorDescriptor.xdetectors}.keys()\\n else:\\n det_s = DetectorDescriptor.detectors.keys()\\n\\n for detector in det_s:\\n algo = DetectorDescriptor(detector)\\n print(\\\"Running test {}\\\".format(detector))\\n\\n for f in files:\\n filename = os.path.basename(f).split('.')[0]\\n image = cv2.imread(f, 0)\\n kps = algo.detect(image)\\n\\n h = h_for_file(f)\\n if h is None: # This will be the case for the base image (img1)\\n baseimg = image\\n basepts = kps\\n\\n data.append([detector, filename, len(basepts), len(basepts)])\\n continue\\n\\n hi = linalg.inv(h)\\n mask = create_mask(baseimg.shape, hi)\\n\\n # Only those that are common\\n bpts = []\\n for pt in basepts:\\n if point_in_mask(pt.pt, mask):\\n bpts.append(pt)\\n bptst = np.vstack([pt.pt for pt in bpts])\\n\\n rep = 0\\n for point in kps:\\n tp = transform_point(point.pt, hi)\\n if point_in_mask(tp, mask):\\n dists = distance.cdist([tp], bptst)\\n if np.min(dists) < threshold:\\n rep += 1\\n\\n data.append([detector, filename, len(bpts), rep])\\n\\n df = pd.DataFrame(data, columns=columns)\\n df['repeatability'] = df['repeat'] / df['common']\\n return df\",\n \"def compare_group_images(directory, show_imgs=True, similarity=\\\"high\\\", compression=100):\\n # list where the found duplicate/similar images are stored\\n duplicates = []\\n lower_res = []\\n\\n imgs_matrix = create_imgs_matrix(directory, compression)\\n\\n # search for similar images\\n if similarity == \\\"low\\\":\\n ref = 13000\\n # search for 1:1 duplicate images\\n else:\\n ref = 21000\\n\\n main_img = 0\\n compared_img = 1\\n nrows, ncols = compression, compression\\n srow_A = 0\\n erow_A = compression\\n srow_B = erow_A\\n erow_B = srow_B + compression\\n\\n while erow_B <= imgs_matrix.shape[0]:\\n while compared_img < (len(image_files)):\\n # select two images from imgs_matrix\\n imgA = imgs_matrix[srow_A: erow_A, # rows\\n 0: ncols] # columns\\n imgB = imgs_matrix[srow_B: erow_B, # rows\\n 0: ncols] # columns\\n # compare the images\\n rotations = 0\\n while image_files[main_img] not in duplicates and rotations <= 3:\\n if rotations != 0:\\n imgB = rotate_img(imgB)\\n err = mse(imgA, imgB)\\n print ( \\\"err:\\\", err)\\n if err <= ref:\\n if show_imgs == True:\\n show_file_info(compared_img, main_img)\\n add_to_list(image_files[main_img], duplicates)\\n check_img_quality(directory, image_files[main_img], image_files[compared_img], lower_res)\\n rotations += 1\\n srow_B += compression\\n erow_B += compression\\n compared_img += 1\\n\\n srow_A += compression\\n erow_A += compression\\n srow_B = erow_A\\n erow_B = srow_B + compression\\n main_img += 1\\n compared_img = main_img + 1\\n\\n msg = \\\"\\\\n***\\\\n DONE: found \\\" + str(len(duplicates)) + \\\" duplicate image pairs in \\\" + str(\\n len(image_files)) + \\\" total images.\\\\n The following files have lower resolution:\\\"\\n print(msg)\\n return set(lower_res)\",\n \"def compute_classifications2(depc, gid_list, config=None):\\n logger.info('[ibs] Process Image Classifications2')\\n logger.info('config = {!r}'.format(config))\\n # Get controller\\n ibs = depc.controller\\n depc = ibs.depc_image\\n if config['classifier_two_algo'] in ['cnn']:\\n config_ = {\\n 'draw_annots': False,\\n 'thumbsize': (192, 192),\\n }\\n # depc.delete_property('thumbnails', gid_list, config=config_)\\n thumbnail_list = depc.get_property('thumbnails', gid_list, 'img', config=config_)\\n result_list = ibs.generate_thumbnail_class2_list(thumbnail_list, **config)\\n elif config['classifier_two_algo'] in ['rf']:\\n from wbia.algo.detect.rf import classify\\n\\n config_ = {'algo': 'resnet'}\\n vector_list = depc.get_property('features', gid_list, 'vector', config=config_)\\n classifier_weight_filepath = config['classifier_weight_filepath']\\n result_list = classify(vector_list, weight_filepath=classifier_weight_filepath)\\n elif config['classifier_two_algo'] in ['densenet']:\\n from wbia.algo.detect import densenet\\n\\n config_ = {\\n 'draw_annots': False,\\n 'thumbsize': (densenet.INPUT_SIZE, densenet.INPUT_SIZE),\\n }\\n thumbpath_list = ibs.depc_image.get(\\n 'thumbnails', gid_list, 'img', config=config_, read_extern=False, ensure=True\\n )\\n config_ = {\\n 'classifier_weight_filepath': config['classifier_two_weight_filepath'],\\n }\\n result_list = densenet.test(\\n thumbpath_list,\\n ibs=ibs,\\n gid_list=gid_list,\\n return_dict=True,\\n multiclass=True,\\n **config_,\\n )\\n result_list = list(result_list)\\n for index in range(len(result_list)):\\n best_score, best_key, scores = result_list[index]\\n classes = [best_key]\\n result_list[index] = (\\n scores,\\n classes,\\n )\\n else:\\n raise ValueError(\\n 'specified classifier_two algo is not supported in config = {!r}'.format(\\n config\\n )\\n )\\n\\n # yield detections\\n for result in result_list:\\n yield result\",\n \"def doMatching(self):\\n\\n self.alltrans = []\\n curdir = os.getcwd()\\n os.chdir(self.astromdir)\\n\\n for cat in self.projcats:\\n self.logfile.write(\\\"Beginning match attempt for %s\\\" %cat)\\n nobj = max(80,self.Nrefobjs)\\n base = 'match %s 1 2 3 %s 1 2 3 identity recalc nobj=%d medtf medsigclip=2.7 '\\\\\\n %(cat,self.GSCmatchin,nobj)\\n converged=0\\n mrad=4.5\\n xsh=0.\\n ysh=0.\\n Retry=0\\n while not converged and mrad<10 and Retry<11:\\n ferr,mdict=self._runmatch(base,mrad,xsh,ysh)\\n if ferr:\\n # match not found, make radius bigger and retry\\n mrad += 0.5\\n Retry += 1\\n self.logfile.write(\\\"Retry: %d ... set mrad: %s\\\" %(Retry,mrad))\\n continue\\n\\n dx = mdict['dx']\\n dy = mdict['dy']\\n sigx = mdict['sigx']\\n sigy = mdict['sigy']\\n\\n # if the extra shift is too big compared to the\\n # match radius, update input shift and retry\\n if max(abs(dx-xsh),abs(dy-ysh)) > min(0.4, mrad/10.):\\n xsh=dx\\n ysh=dy\\n Retry += 1\\n continue\\n\\n # if sigma seems too big, shrink mrad\\n if (max(sigx,sigy) > 0.7 and mrad>2.9):\\n mrad -= 0.5\\n Retry += 1\\n continue\\n\\n # otherwise, it looks like we have a good fit\\n # but we want to tune it up with smaller matchrad\\n # Don't increment retry in this final case.\\n if mrad > 3.9 or (mrad>2.9 and max(sigx,sigy)>0.4):\\n mrad = max(mrad-1,2.5)\\n continue\\n\\n self.alltrans.append(mdict)\\n converged=1\\n self.logfile.write(\\\"\\\"\\\"%s trans:\\n mdx,mdy = %.4f %.4f\\n edx,edy = %.4f %.4f\\n sigx,sigy = %.4f %.4f\\n \\\"\\\"\\\"%(cat, mdict['dx'],mdict['dy'],mdict['dx_err'],mdict['dy_err'],\\\\\\n mdict['sigx'],mdict['sigy']))\\n if len(self.alltrans)<1:\\n errtxt = \\\"WARNING: No Image successfully matched against extended GSC!\\\"\\n self.logfile.write(errtxt)\\n self.errorList.append((self.modName,errtxt))\\n return -1\\n\\n os.chdir(curdir)\\n # finally, choose best match transform and calculate\\n # the new CRVAL's\\n self._pickBestrans()\\n\\n return 0\",\n \"def accuracy():\\n\\n # Creating a list of populations to analyze over\\n cells = np.linspace(min_desired_num_cells, max_desired_num_cells, num_data_points)\\n list_of_populations = []\\n list_of_fpi = []\\n for cell_num in cells:\\n population = LineageTree.init_from_parameters(pi, T, E2, cell_num)\\n\\n # Adding populations into a holder for analysing\\n list_of_populations.append([population])\\n list_of_fpi.append(pi)\\n\\n return commonAnalyze(list_of_populations, 2, list_of_fpi=list_of_fpi)\",\n \"def test_classifier(self):\\n \\n files = 0\\n tp = 0\\n fp = 0\\n tn = 0\\n fn = 0\\n \\n for testFile in os.listdir(self.testing_dir):\\n if os.path.splitext(testFile)[1] in self.valid:\\n\\n files += 1\\n fileName = self.testing_dir + \\\"/\\\" + testFile\\n\\n img = cv2.imread(fileName).astype(np.float32)\\n self.Helpers.logger.info(\\\"Loaded test image \\\" + fileName)\\n \\n img = cv2.resize(img, (self.Helpers.confs[\\\"cnn\\\"][\\\"data\\\"][\\\"dim\\\"], \\n self.Helpers.confs[\\\"cnn\\\"][\\\"data\\\"][\\\"dim\\\"]))\\n img = self.reshape(img)\\n \\n prediction = self.get_predictions(img)\\n \\n msg = \\\"\\\"\\n if prediction == 1 and \\\"_1.\\\" in testFile:\\n tp += 1\\n msg = \\\"Acute Lymphoblastic Leukemia correctly detected (True Positive)\\\"\\n elif prediction == 1 and \\\"_0.\\\" in testFile:\\n fp += 1\\n msg = \\\"Acute Lymphoblastic Leukemia incorrectly detected (False Positive)\\\"\\n elif prediction == 0 and \\\"_0.\\\" in testFile:\\n tn += 1\\n msg = \\\"Acute Lymphoblastic Leukemia correctly not detected (True Negative)\\\"\\n elif prediction == 0 and \\\"_1.\\\" in testFile:\\n fn += 1\\n msg = \\\"Acute Lymphoblastic Leukemia incorrectly not detected (False Negative)\\\"\\n self.Helpers.logger.info(msg)\\n \\n self.Helpers.logger.info(\\\"Images Classifier: \\\" + str(files))\\n self.Helpers.logger.info(\\\"True Positives: \\\" + str(tp))\\n self.Helpers.logger.info(\\\"False Positives: \\\" + str(fp))\\n self.Helpers.logger.info(\\\"True Negatives: \\\" + str(tn))\\n self.Helpers.logger.info(\\\"False Negatives: \\\" + str(fn))\",\n \"def test_compare_classifiers_multiclass_multimetric_vis_api(experiment_to_use):\\n experiment = experiment_to_use\\n # extract test stats only\\n test_stats = experiment.test_stats_full\\n viz_outputs = (\\\"pdf\\\", \\\"png\\\")\\n with TemporaryDirectory() as tmpvizdir:\\n for viz_output in viz_outputs:\\n vis_output_pattern_pdf = tmpvizdir + f\\\"/*.{viz_output}\\\"\\n visualize.compare_classifiers_multiclass_multimetric(\\n [test_stats, test_stats],\\n experiment.ground_truth_metadata,\\n experiment.output_feature_name,\\n top_n_classes=[6],\\n model_namess=[\\\"Model1\\\", \\\"Model2\\\"],\\n output_directory=tmpvizdir,\\n file_format=viz_output,\\n )\\n figure_cnt = glob.glob(vis_output_pattern_pdf)\\n assert 4 == len(figure_cnt)\",\n \"def compute_map(scores, num_instances):\\r\\n pr = []\\r\\n for i, score in enumerate(scores):\\r\\n score = score[np.argsort(-score[:, 1])] # sort by confidence score\\r\\n FP = 0\\r\\n TP = 0\\r\\n pr_ = []\\r\\n for prediction in score:\\r\\n if prediction[0]:\\r\\n TP += 1\\r\\n else:\\r\\n FP += 1\\r\\n pr_.append([TP/(TP+FP), (TP/num_instances)])\\r\\n pr.append(pr_)\\r\\n pr = np.array(pr)\\r\\n\\r\\n pinterps = [] # lists of interpolated precisions for every confidence level\\r\\n ranks = np.linspace(0, 1, 11)\\r\\n idxs_interpolations = [] # list of indexes of the interpolated precisions, just to plot the recall\\r\\n for pr_ in pr:\\r\\n pinterp = []\\r\\n idxs = []\\r\\n last_idx = -1\\r\\n for rank in ranks:\\r\\n idx = (np.abs(pr_[:, 1] - rank)).argmin() # find the closest recall to the rank\\r\\n\\r\\n if rank > pr_[idx, 1]: # this makes sure we are taking the closest recall at the right of the rank\\r\\n if idx+1 < pr_[:, 0].shape[0]:\\r\\n idx += 1\\r\\n interpolated_precision = np.max(pr_[idx:, 0]) # find the max precision within the interval\\r\\n if idx == last_idx: # just some checks for when the recall doesn't exist\\r\\n pinterp[-1] = 0\\r\\n idxs[-1] = 0\\r\\n pinterp.append(0)\\r\\n idxs.append(0)\\r\\n else:\\r\\n pinterp.append(interpolated_precision)\\r\\n idxs.append(idx)\\r\\n last_idx = idx\\r\\n pinterps.append(pinterp)\\r\\n idxs_interpolations.append(idxs)\\r\\n APs = np.array(pinterps).mean(axis=1) # the AP is the average of the interpolated precisions\\r\\n mAP = APs.mean() # mAP is the mean of all the APs\\r\\n\\r\\n return pr, pinterps, idxs_interpolations, mAP, APs\",\n \"def score_matches(analysis_matches, args):\\n\\t\\n\\tscore_types = ('found_score', 'host_score', 'virus_score')\\n\\t\\n\\tfor match_type in analysis_matches.values():\\n\\t\\tfor match in match_type:\\n\\t\\t\\t# if read crosses an integration\\n\\t\\t\\tif match['intID'] is not None:\\n\\t\\t\\t\\t# if integration was found, and should have been found\\n\\t\\t\\t\\tif match['found'] is True:\\n\\t\\t\\t\\t\\tmatch['found_score'] = 'tp'\\n\\t\\t\\t\\t\\t\\n\\t\\t\\t\\t\\t# check if integration was found in the correct place in the host\\n\\t\\t\\t\\t\\tif correct_pos(match, args, ref = 'host'):\\n\\t\\t\\t\\t\\t\\tmatch['host_score'] = 'tp'\\n\\t\\t\\t\\t\\telse:\\n\\t\\t\\t\\t\\t\\tmatch['host_score'] = 'fp'\\t\\t\\t\\t\\t\\t\\t\\n\\t\\t\\t\\t\\t\\t\\n\\t\\t\\t\\t\\t# check if integration was found in the correct virus\\n\\t\\t\\t\\t\\tif correct_pos(match, args, ref = 'virus'):\\n\\t\\t\\t\\t\\t\\tmatch['virus_score'] = 'tp'\\n\\t\\t\\t\\t\\telse:\\n\\t\\t\\t\\t\\t\\tmatch['virus_score'] = 'fp'\\n\\t\\t\\t\\t\\t\\n\\t\\t\\t\\telse:\\n\\t\\t\\t\\t\\tfor score_type in score_types:\\n\\t\\t\\t\\t\\t\\tmatch[score_type] = 'fn'\\n\\t\\t\\t\\n\\t\\t\\t# if read doesn't cross integration\\n\\t\\t\\telse:\\n\\t\\t\\t\\t# doesn't cross but is found - false positive\\n\\t\\t\\t\\tif match['found'] is True:\\n\\t\\t\\t\\t\\t\\tfor score_type in score_types:\\n\\t\\t\\t\\t\\t\\t\\tmatch[score_type] = 'fp'\\n\\t\\t\\t\\t# doesn't cross and isn't found - true netgative\\n\\t\\t\\t\\telse:\\n\\t\\t\\t\\t\\tfor score_type in score_types:\\n\\t\\t\\t\\t\\t\\tmatch[score_type] = 'tn'\\n\\t\\t\\t\\t\\t\\n\\n\\t\\t\\t\\t\\n\\treturn analysis_matches\",\n \"def test_run(id_, args, kwargs, expected):\\n # Remove results for test.\\n results_path = expected['results_path']\\n shutil.rmtree(results_path, ignore_errors=True)\\n # Start code.\\n # [future] attempts to run classification with n_jobs>1\\n # global cnfg_default\\n # sys.modules['cnfg_default'] = get_params(args[0], 'cnfg_default')\\n # import cnfg_default\\n # #from cnfg_default import custom_score_metric\\n objects = pycnfg.run(oid='default', *args, **kwargs)\\n tmp = {k: type(v).__name__ for k, v in objects.items()}\\n print('OBJECTS:')\\n print(tmp)\\n # Compare results:\\n # * Compare objects (keys and str of values).\\n objects_ = expected['objects']\\n objects = {k: type(v).__name__ for k, v in objects.items()}\\n if objects != objects_:\\n print(set(objects.items()) ^ set(objects_.items()))\\n assert objects == objects_\\n # for k, v in objects.items():\\n # assert k in objects_\\n # assert type(v).__name__ == objects_[k]\\n # * Compare predictions csv(all available).\\n pred_path = glob.glob(f\\\"{results_path}/models/*_pred.csv\\\")\\n pred_path_ = glob.glob(expected['pred_path'])\\n assert len(pred_path) == len(pred_path_)\\n for act, exp in zip(sorted(pred_path), sorted(pred_path_)):\\n file_diff(act, exp)\\n assert filecmp.cmp(act, exp)\\n # * Compare test logs.\\n logs_path = glob.glob(f\\\"{results_path}/logs*/*_test.log\\\")[0]\\n logs_path_ = expected['logs_path']\\n file_diff(logs_path, logs_path_)\\n assert filecmp.cmp(logs_path, logs_path_)\\n # * Compare runs dataframe, non-universe columns.\\n runs_path = f\\\"{results_path}/runs\\\"\\n runs_path_ = expected['runs_path']\\n df = runs_loader(runs_path)\\n df_ = runs_loader(runs_path_)\\n # First False/True for each element, then check all by columns.\\n # col1 True\\n # col2 False\\n # dtype: bool\\n df_diff = df.eq(df_).all()\\n # Column names that are not equal.\\n columns = sorted(list(df_diff[df_diff==False].dropna().index))\\n # columns_eq = sorted(list(df_diff[df_diff==True].dropna().index))\\n columns_ = expected['columns_diff']\\n print('DIFF:\\\\n', columns)\\n time.sleep(1)\\n # assert columns == columns_\\n # * Compare model.\\n model_path = glob.glob(f\\\"{results_path}/models/*.model\\\")\\n model_path_ = glob.glob(expected['model_path'])\\n assert len(model_path) == len(model_path_)\\n for act, exp in zip(sorted(model_path), sorted(model_path_)):\\n assert filecmp.cmp(act, exp)\\n return\"\n]","isConversational":false},"negative_scores":{"kind":"list like","value":["0.70493734","0.6962549","0.6559345","0.6556465","0.6520097","0.645765","0.64450425","0.6405806","0.63550365","0.62989557","0.62902206","0.6247937","0.623757","0.6215632","0.6210621","0.6126112","0.61117905","0.61023283","0.60840875","0.60789233","0.6061147","0.6043648","0.6043097","0.6035426","0.6016283","0.6010682","0.6003366","0.5992554","0.5967746","0.5962685","0.59439194","0.5940636","0.59337485","0.592838","0.5921624","0.59200716","0.59158564","0.59096605","0.59088063","0.5899306","0.5896102","0.5892429","0.58918643","0.5885652","0.58751225","0.58568656","0.58547956","0.58531076","0.585076","0.58433807","0.58418846","0.5834036","0.5825966","0.5822257","0.5801426","0.5795815","0.5790717","0.57893455","0.5787837","0.5785563","0.5780903","0.57775265","0.5775974","0.5773322","0.57649726","0.5763077","0.57615656","0.5753761","0.5753218","0.57473564","0.57454413","0.573491","0.5734811","0.5731087","0.5725846","0.57170415","0.5712467","0.57111156","0.5702868","0.57017","0.5701569","0.5696177","0.56944025","0.5690325","0.5690107","0.5686798","0.56838304","0.5681137","0.5667145","0.56653243","0.5659523","0.565792","0.565059","0.5641997","0.56381136","0.563622","0.5635908","0.5633417","0.56332195","0.5622713","0.5620655"],"string":"[\n \"0.70493734\",\n \"0.6962549\",\n \"0.6559345\",\n \"0.6556465\",\n \"0.6520097\",\n \"0.645765\",\n \"0.64450425\",\n \"0.6405806\",\n \"0.63550365\",\n \"0.62989557\",\n \"0.62902206\",\n \"0.6247937\",\n \"0.623757\",\n \"0.6215632\",\n \"0.6210621\",\n \"0.6126112\",\n \"0.61117905\",\n \"0.61023283\",\n \"0.60840875\",\n \"0.60789233\",\n \"0.6061147\",\n \"0.6043648\",\n \"0.6043097\",\n \"0.6035426\",\n \"0.6016283\",\n \"0.6010682\",\n \"0.6003366\",\n \"0.5992554\",\n \"0.5967746\",\n \"0.5962685\",\n \"0.59439194\",\n \"0.5940636\",\n \"0.59337485\",\n \"0.592838\",\n \"0.5921624\",\n \"0.59200716\",\n \"0.59158564\",\n \"0.59096605\",\n \"0.59088063\",\n \"0.5899306\",\n \"0.5896102\",\n \"0.5892429\",\n \"0.58918643\",\n \"0.5885652\",\n \"0.58751225\",\n \"0.58568656\",\n \"0.58547956\",\n \"0.58531076\",\n \"0.585076\",\n \"0.58433807\",\n \"0.58418846\",\n \"0.5834036\",\n \"0.5825966\",\n \"0.5822257\",\n \"0.5801426\",\n \"0.5795815\",\n \"0.5790717\",\n \"0.57893455\",\n \"0.5787837\",\n \"0.5785563\",\n \"0.5780903\",\n \"0.57775265\",\n \"0.5775974\",\n \"0.5773322\",\n \"0.57649726\",\n \"0.5763077\",\n \"0.57615656\",\n \"0.5753761\",\n \"0.5753218\",\n \"0.57473564\",\n \"0.57454413\",\n \"0.573491\",\n \"0.5734811\",\n \"0.5731087\",\n \"0.5725846\",\n \"0.57170415\",\n \"0.5712467\",\n \"0.57111156\",\n \"0.5702868\",\n \"0.57017\",\n \"0.5701569\",\n \"0.5696177\",\n \"0.56944025\",\n \"0.5690325\",\n \"0.5690107\",\n \"0.5686798\",\n \"0.56838304\",\n \"0.5681137\",\n \"0.5667145\",\n \"0.56653243\",\n \"0.5659523\",\n \"0.565792\",\n \"0.565059\",\n \"0.5641997\",\n \"0.56381136\",\n \"0.563622\",\n \"0.5635908\",\n \"0.5633417\",\n \"0.56332195\",\n \"0.5622713\",\n \"0.5620655\"\n]","isConversational":false},"document_score":{"kind":"string","value":"0.0"},"document_rank":{"kind":"string","value":"-1"}}},{"rowIdx":93296,"cells":{"query":{"kind":"string","value":"create the dictionary holding the image_matching_module's results"},"document":{"kind":"string","value":"def create_experiment_dict(algorithm_name, thresholds):\n experiment_dict = {}\n for threshold in thresholds:\n experiment_dict[str(threshold)] = {}\n experiment_dict[str(threshold)][algorithm_name] = [0, 0, 0, 0, 0, 0, 0, 0, 0, 0]\n list_dict = {'TP': 0, 'FP': 1, 'TN': 2, 'FN': 3, 'TPR': 4, 'TNR': 5, 'PRECISION': 6, 'RECALL': 7,\n 'F1 MEASURE': 8, 'ACCURACY': 9}\n return experiment_dict, list_dict"},"metadata":{"kind":"string","value":"{\n \"objective\": {\n \"self\": [],\n \"paired\": [],\n \"triplet\": [\n [\n \"query\",\n \"document\",\n \"negatives\"\n ]\n ]\n }\n}"},"negatives":{"kind":"list like","value":["def get_detect_result(self):\n\n resultdict = {'class_index' : self.class_index,\n 'obj_name' : self.obj_name,\n 'score' : self.score,\n 'bounding_box' : {\n 'x_min' : self.x_min,\n 'y_min' : self.y_min,\n 'width' : self.width,\n 'height' : self.height}\n }\n return resultdict","def get_results(self):\r\n r ={\r\n 'Gp_s' :np.array(self.Gp_guess_s), #Means of the posteriors\r\n 'Gm_s' :np.array(self.Gm_guess_s), #Means of the posteriors\r\n 'eGp_s':np.array(self.eGp_guess_s) , #Standard deviations of the posteriors\r\n 'eGm_s':np.array(self.eGm_guess_s) , #Standard deviations of the posteriors\r\n 'covGpGm_s':np.array(self.cov_GpGm_s) , #Covariances of the posteriors\r\n 't_tot_pulseSequences_s':np.array(self.t_tot_pulseSequences_s), #Time it takes for each batch of pulse sequences\r\n 't_tot_process_s':np.array(self.t_tot_process_s), #CPU time it takes for processing the Bayes stuff at each iteration\r\n 't_probe_p_s':np.array(self.t_probe_p_s), # Times probed for the diff p\r\n 't_probe_m_s':np.array(self.t_probe_m_s), # Times probed for the diff m\r\n 't_pulseSequences_s':np.array(self.t_pulseSequences_s),\r\n 't_process_s':np.array(self.t_process_s),\r\n 'iteration_s':np.array(self.nb_iteration_s), #Number of iterations performed\r\n 'R_tot_s':np.array(self.R_tot_s) #Total number of readout performed\r\n }\r\n return r","def match_results(self):\n return np.array(list(self._match_result_dict.values()))","def get_matches(self, file_map) -> dict:\r\n get_file_dict = {}\r\n match_dict = {\r\n 'GIF': re.findall(b'(?s)(\\x47\\x49\\x46\\x38\\x39\\x61.{80})', file_map),\r\n 'RTF': re.findall(b'(?s)(.{20}\\x35\\x30\\x34\\x65\\x34\\x37\\x30.{80}|.{20}\\x66\\x66\\x64\\x38\\x66\\x66.{80}|'\r\n b'.{20}\\x66\\x66\\x64\\x38\\x66\\x66\\x65\\x30\\x30\\x30\\x31\\x30.{80})', file_map),\r\n }\r\n if self.jpgsos:\r\n match_dict['JPG_SOS'] = jpg_sos(file_map)\r\n elif self.sof2sos:\r\n match_dict['JPG_SOF2SOS'] = jpg_sof2sos(file_map)\r\n elif self.jump:\r\n match_dict['JPG_JUMP'] = jpg_jump(file_map)\r\n else:\r\n match_dict['JPG'] = re.findall(b'(?s)(\\xff\\xd8\\xff\\xe0\\x00\\x10.{80})', file_map)\r\n match_dict['JPG2'] = re.findall(b'(?s)(\\xff\\xd8\\xff.{80})', file_map)\r\n if self.idat:\r\n match_dict['PNG_IDAT'] = idat(file_map)\r\n else:\r\n match_dict['PNG'] = re.findall(b'(?s)(\\x89\\x50\\x4e\\x47.{82})', file_map)\r\n m = re.match(br'^(?P<magic_beans>\\x49\\x49\\x2a\\x00[^\\x00\\x00]{2}.{80})',file_map,re.S)\r\n if m:\r\n match_dict['TIF'] = [m.group('magic_beans')]\r\n for file_type, regex_match in match_dict.items():\r\n if len(regex_match) > 0:\r\n get_file_dict[file_type] = regex_match\r\n return get_file_dict","def calculates_results_stats(results_dic): \n # Creates empty dictionary for results_stats_dic\n results_stats_dic = dict()\n \n # Sets all counters to initial values of zero so that they can \n # be incremented while processing through the images in results_dic \n results_stats_dic['n_dogs_img'] = 0\n results_stats_dic['n_match'] = 0\n results_stats_dic['n_correct_dogs'] = 0\n results_stats_dic['n_correct_notdogs'] = 0\n results_stats_dic['n_correct_breed'] = 0\n \n # process through the results dictionary\n for key in results_dic:\n \n # Labels Match Exactly\n if results_dic[key][2] == 1:\n results_stats_dic['n_match'] += 1\n\n # TODO: 5a. REPLACE pass with CODE that counts how many pet images of\n # dogs had their breed correctly classified. This happens \n # when the pet image label indicates the image is-a-dog AND \n # the pet image label and the classifier label match. You \n # will need to write a conditional statement that determines\n # when the dog breed is correctly classified and then \n # increments 'n_correct_breed' by 1. Recall 'n_correct_breed' \n # is a key in the results_stats_dic dictionary with it's value \n # representing the number of correctly classified dog breeds.\n # \n # Pet Image Label is a Dog AND Labels match- counts Correct Breed\n if results_dic[key][3] == 1 and results_dic[key][2] == 1:\n results_stats_dic['n_correct_breed'] += 1\n \n # Pet Image Label is a Dog - counts number of dog images\n if results_dic[key][3] == 1:\n results_stats_dic['n_dogs_img'] += 1\n \n # Classifier classifies image as Dog (& pet image is a dog)\n # counts number of correct dog classifications\n if results_dic[key][4] == 1:\n results_stats_dic['n_correct_dogs'] += 1\n\n # TODO: 5b. REPLACE pass with CODE that counts how many pet images \n # that are NOT dogs were correctly classified. This happens \n # when the pet image label indicates the image is-NOT-a-dog \n # AND the classifier label indicates the images is-NOT-a-dog.\n # You will need to write a conditional statement that \n # determines when the classifier label indicates the image \n # is-NOT-a-dog and then increments 'n_correct_notdogs' by 1. \n # Recall the 'else:' above 'pass' already indicates that the \n # pet image label indicates the image is-NOT-a-dog and \n # 'n_correct_notdogs' is a key in the results_stats_dic dictionary \n # with it's value representing the number of correctly \n # classified NOT-a-dog images.\n # \n # Pet Image Label is NOT a Dog\n else:\n # Classifier classifies image as NOT a Dog(& pet image isn't a dog)\n # counts number of correct NOT dog clasifications.\n if results_dic[key][3] == 0 and results_dic[key][4] == 0:\n results_stats_dic['n_correct_notdogs'] += 1\n\n\n # Calculates run statistics (counts & percentages) below that are calculated\n # using the counters from above.\n\n # calculates number of total images\n results_stats_dic['n_images'] = len(results_dic)\n\n # calculates number of not-a-dog images using - images & dog images counts\n results_stats_dic['n_notdogs_img'] = (results_stats_dic['n_images'] - \n results_stats_dic['n_dogs_img']) \n\n # TODO: 5c. REPLACE zero(0.0) with CODE that calculates the % of correctly\n # matched images. Recall that this can be calculated by the\n # number of correctly matched images ('n_match') divided by the \n # number of images('n_images'). This result will need to be \n # multiplied by 100.0 to provide the percentage.\n # \n # Calculates % correct for matches\n results_stats_dic['pct_match'] = (results_stats_dic['n_match'] / results_stats_dic['n_images']) * 100\n\n # TODO: 5d. REPLACE zero(0.0) with CODE that calculates the % of correctly\n # classified dog images. Recall that this can be calculated by \n # the number of correctly classified dog images('n_correct_dogs')\n # divided by the number of dog images('n_dogs_img'). This result \n # will need to be multiplied by 100.0 to provide the percentage.\n # \n # Calculates % correct dogs\n results_stats_dic['pct_correct_dogs'] = (results_stats_dic['n_correct_dogs'] / results_stats_dic['n_dogs_img']) * 100\n\n # TODO: 5e. REPLACE zero(0.0) with CODE that calculates the % of correctly\n # classified breeds of dogs. Recall that this can be calculated \n # by the number of correctly classified breeds of dog('n_correct_breed') \n # divided by the number of dog images('n_dogs_img'). This result \n # will need to be multiplied by 100.0 to provide the percentage.\n # \n # Calculates % correct breed of dog\n results_stats_dic['pct_correct_breed'] = (results_stats_dic['n_correct_breed'] / results_stats_dic['n_dogs_img']) * 100\n\n # Calculates % correct not-a-dog images\n # Uses conditional statement for when no 'not a dog' images were submitted \n if results_stats_dic['n_notdogs_img'] > 0:\n results_stats_dic['pct_correct_notdogs'] = (results_stats_dic['n_correct_notdogs'] /\n results_stats_dic['n_notdogs_img']) * 100.0\n else:\n results_stats_dic['pct_correct_notdogs'] = 0.0\n\n \n # TODO 5f. REPLACE None with the results_stats_dic dictionary that you \n # created with this function \n return results_stats_dic","def search(self, query_image, list_images):\n\t\tresult = {}\n\t\t# Get variables from query image\n\t\tquery_kps, query_descriptor = query_image.key_points_and_descriptors()\n\t\t# Loop in the list of images\n\t\tfor image in list_images:\n\t\t\t# Varuables of the current image\n\t\t\tkps, descriptor = image.key_points_and_descriptors()\n\t\t\t# Get the score of the current image compared with the query image\n\t\t\tscore = self.get_score(query_kps, query_descriptor, kps, descriptor)\n\t\t\tresult[image.name()] = score\n\n\t\t# Sort results by score\n\t\tif len(result) > 0:\n\t\t\tresult = sorted([(v, k) for (k, v) in result.items() if v > 0],\n\t\t\t\treverse = True)\n\t\treturn result","def getOverlapComparisonImagesDict(self) :\n overlap_shift_comparisons = {}\n for o in self.overlaps :\n overlap_shift_comparisons[o.getShiftComparisonDetailTuple()]=o.getShiftComparisonImages()\n return overlap_shift_comparisons","def __call__(self, results):\r\n if isinstance(results['img'], str):\r\n results['filename'] = results['img']\r\n results['ori_filename'] = results['img']\r\n else:\r\n results['filename'] = None\r\n results['ori_filename'] = None\r\n img = mmcv.imread(results['img'])\r\n results['img'] = img\r\n results['img_fields'] = ['img']\r\n results['img_shape'] = img.shape\r\n results['ori_shape'] = img.shape\r\n return results","def get_all_results(pred_root, meta_results):\r\n results_all = {}\r\n for key in tqdm(meta_results, desc='Generating results ..'):\r\n persons = meta_results[key]\r\n\r\n global_seg = cv2.imread(pred_root + 'global_seg/{}.png'.format(key),\r\n cv2.IMREAD_UNCHANGED)\r\n global_tag = cv2.imread(pred_root + 'global_tag/{}.png'.format(key),\r\n cv2.IMREAD_UNCHANGED)\r\n\r\n results = {}\r\n dets, masks = [], []\r\n for p_id, score in persons:\r\n mask = (global_tag == p_id)\r\n if np.sum(mask) == 0:\r\n continue\r\n seg = mask * global_seg\r\n ys, xs = np.where(mask > 0)\r\n x1, y1, x2, y2 = xs.min(), ys.min(), xs.max(), ys.max()\r\n dets.append((x1, y1, x2, y2, score))\r\n masks.append(seg)\r\n\r\n # Reuiqred Field of each result: a list of masks,\r\n # each is a multi-class masks for one person.\r\n # It can also be sparsified to\r\n # [scipy.sparse.csr_matrix(mask) for mask in masks]\r\n # to save memory cost\r\n results['MASKS'] = masks if not Sparse \\\r\n else [scipy.sparse.csr_matrix(m) for m in masks]\r\n # Reuiqred Field of each result,\r\n # a list of detections corresponding to results['MASKS'].\r\n results['DETS'] = dets\r\n\r\n if cache_pkl:\r\n results_cache_add = cache_pkl_path + key + '.pklz'\r\n pickle.dump(results, gzip.open(results_cache_add, 'w'))\r\n results_all[key] = results_cache_add\r\n else:\r\n results_all[key] = results\r\n\r\n if PLOT:\r\n import pylab as plt\r\n plt.figure('seg')\r\n plt.imshow(global_seg)\r\n print('Seg unique:' + str(np.unique(global_seg)))\r\n plt.figure('tag')\r\n plt.imshow(global_tag)\r\n print('Tag unique:' + str(np.unique(global_tag)))\r\n plt.show()\r\n\r\n return results_all","def _match_keypoints(self, query_image: Image) -> typing.Tuple[typing.List[int], typing.List[typing.Any]]:\n query_image.get_keypoints_and_descriptors()\n\n # matches = self.matcher.match(query_image.descriptor)\n matches = self.matcher.knnMatch(query_image.descriptor, k=2)\n good = []\n for m_n in matches:\n if len(m_n) == 1:\n good.append(m_n[0])\n continue\n elif len(m_n) != 2:\n continue\n (m, n) = m_n\n if m.distance < 0.7 * n.distance:\n good.append(m)\n\n images_scores = [0] * len(self._candidate_images)\n images_matches = [None] * len(self._candidate_images)\n for image_index, image in enumerate(self._candidate_images):\n matches_scores = []\n matches = []\n for i, match in enumerate(good):\n if match.imgIdx != image_index:\n continue\n matches.append(match)\n matches_scores.append((256 - match.distance) / 256)\n\n match_cnt = len(matches_scores)\n if match_cnt <= 0:\n continue\n\n images_scores[image_index] = (\n 0.5 + ((math.tanh(match_cnt / 3 - 1)) / 2)) * (sum(matches_scores) / match_cnt)\n\n images_matches[image_index] = matches\n\n return images_scores, images_matches","def _get_images(self):\n raw_outputs = self.interface.get_data(self.target_charge,\n self.charge_deviation,\n n_samples=self.n_samples)\n\n # apply roi to images\n roi_images = []\n for i in range(self.n_samples):\n roi_images += [apply_roi(raw_outputs['raw_images'][i], raw_outputs['ROI'])]\n\n # process and identify blobs in image\n min_size = 100\n outputs = {}\n for ele in self.output_keys:\n outputs[ele] = []\n\n for i in range(len(roi_images)):\n processed_image_data = image_processing.process_and_fit(roi_images[i],\n min_size)\n\n for ele in self.output_keys:\n if ele == 'image_check':\n outputs[ele] += [image_processing.check_image(processed_image_data['binary_image'],\n processed_image_data['smoothed_image'])]\n elif ele == 'processed_images':\n outputs[ele] += [processed_image_data['smoothed_image']]\n else:\n outputs[ele] += [processed_image_data[ele]]\n\n for ele in self.output_keys:\n outputs[ele] = np.array(outputs[ele])\n\n # add in raw data\n outputs.update(raw_outputs)\n\n # if we need to, get averaged results\n if self.average_measurements:\n avg_keys = ['rms_x', 'rms_y', 'CX', 'CY', 'n_blobs', 'FWHMX', 'FWHMY', 'centroid_offset']\n for key in avg_keys:\n outputs[key] = np.nanmean(outputs[key])\n\n return outputs","def statistics(img1_blobs, img2_blobs, matches):\n\tstatistics = {}\n\n\tstatistics['#Interest Points in img1'] = len(img1_blobs)\n\tstatistics['#Interest Points in img2'] = len(img2_blobs)\n\tstatistics['Accepted Matches'] = len(matches)\n\tdissimilarity = [match[2] for match in matches]\n\tstatistics['Mean of accepted matches'] = sum(dissimilarity)/len(dissimilarity)\n\tstatistics['SD of accepted matches'] = np.std(dissimilarity)\n\treturn statistics","def get_stats(filtered_paths : dict) -> dict:\n stats_dict = defaultdict(dict)\n for mag in filtered_paths.keys():\n for spec,paths in filtered_paths[mag].items():\n counter = Counter()\n for path in paths:\n img = cv2.imread(path, -1)\n count = Counter(list(img.ravel()))\n counter += count\n stats_dict[mag][spec] = counter\n print(mag)\n return stats_dict","def get_results(self):\n summary = self.handle.get_summary_data(self.group_name)\n results = {'template': {'status': 'no data'},\n 'complement': {'status': 'no data'},\n '2d': {'status': 'no data'}}\n if 'genome_mapping_template' in summary:\n results['template'] = self._get_results(summary['genome_mapping_template'])\n if 'genome_mapping_complement' in summary:\n results['complement'] = self._get_results(summary['genome_mapping_complement'])\n if 'genome_mapping_2d' in summary:\n results['2d'] = self._get_results(summary['genome_mapping_2d'])\n return results","def init_results(self):\n results = {}\n for i in range(len(self.agents)):\n actor = self.agents[i]\n actor.set_name(\"Actor\") # the filename part of the name is set in actor using filename from init_agents\n results[actor.name] = 0\n return results","def __call__(self, results):\n\n if 'img_fields' in results:\n assert results['img_fields'] == ['img'], \\\n 'Only single img_fields is allowed'\n img = results['img']\n assert img.dtype == np.float32, \\\n 'PhotoMetricDistortion needs the input image of dtype ' \\\n 'np.float32, please set \"to_float32=True\" in ' \\\n '\"LoadImageFromFile\" pipeline'\n # random brightness\n if random.randint(2):\n delta = random.uniform(-self.brightness_delta,\n self.brightness_delta)\n img += delta\n\n # mode == 0 --> do random contrast first\n # mode == 1 --> do random contrast last\n mode = random.randint(2)\n if mode == 1:\n if random.randint(2):\n alpha = random.uniform(self.contrast_lower,\n self.contrast_upper)\n img *= alpha\n\n # convert color from BGR to HSV\n img = general_ocr.bgr2hsv(img)\n\n # random saturation\n if random.randint(2):\n img[..., 1] *= random.uniform(self.saturation_lower,\n self.saturation_upper)\n\n # random hue\n if random.randint(2):\n img[..., 0] += random.uniform(-self.hue_delta, self.hue_delta)\n img[..., 0][img[..., 0] > 360] -= 360\n img[..., 0][img[..., 0] < 0] += 360\n\n # convert color from HSV to BGR\n img = general_ocr.hsv2bgr(img)\n\n # random contrast\n if mode == 0:\n if random.randint(2):\n alpha = random.uniform(self.contrast_lower,\n self.contrast_upper)\n img *= alpha\n\n # randomly swap channels\n if random.randint(2):\n img = img[..., random.permutation(3)]\n\n results['img'] = img\n return results","def get_results(self) -> Dict:\n results = {}\n results['time'] = np.array(self._results['time'])\n results['norm'] = np.array(self._results['norm'])\n results['bond_dimensions'] = np.array(self._results['bond_dimensions'])\n results['dynamics'] = self._results['dynamics']\n results['pt_bond_dimensions'] = self._results['pt_bond_dimensions']\n return results","def write_results(input_dict):\n out_imgs = {}\n for mol in input_dict:\n mols = []\n for new_mols in input_dict[mol]:\n m = Chem.MolFromSmiles(new_mols)\n mols.append(m)\n if len(mols) > 2:\n p = Chem.MolFromSmarts(MCS.FindMCS(mols).smarts)\n AllChem.Compute2DCoords(p)\n for m in mols:\n AllChem.GenerateDepictionMatching2DStructure(m, p)\n out_imgs[mol] = Draw.MolsToGridImage(mols, useSVG=True)\n # Write out the image\n return out_imgs","def findMatchesBetweenImages(image_1, image_2):\n # matches - type: list of cv2.DMath\n matches = None\n # image_1_kp - type: list of cv2.KeyPoint items.\n image_1_kp = None\n # image_1_desc - type: numpy.ndarray of numpy.uint8 values.\n image_1_desc = None\n # image_2_kp - type: list of cv2.KeyPoint items.\n image_2_kp = None\n # image_2_desc - type: numpy.ndarray of numpy.uint8 values.\n image_2_desc = None\n # WRITE YOUR CODE HERE.\n\n sift = cv2.ORB_create()\n image_1_kp, image_1_desc = sift.detectAndCompute(image_1, None)\n image_2_kp, image_2_desc = sift.detectAndCompute(image_2, None)\n\n # bf = cv2.BFMatcher(cv2.NORM_HAMMING, crossCheck=True)\n # matches = bf.match(image_1_desc, image_2_desc)\n # matches = sorted(matches, key = lambda x:x.distance)\n # matches = matches[:10]\n\n bf = cv2.BFMatcher()\n matches = bf.knnMatch(image_1_desc, image_2_desc, k=2)\n\n # Apply ratio test\n good = []\n for m, n in matches:\n print(m.distance, n.distance, m.distance < .75*n.distance)\n if m.distance < (0.75 * n.distance):\n good.append([m])\n\n # We coded the return statement for you. You are free to modify it -- just\n # make sure the tests pass.\n print(len(good), good)\n return image_1_kp, image_2_kp, matches\n # END OF FUNCTION.","def setUp(self):\n img_path = osp.join(osp.dirname(__file__), '../../data/gray.jpg')\n self.results = {\n 'img_path':\n img_path,\n 'img_shape': (300, 400),\n 'instances': [{\n 'bbox': [0, 0, 10, 20],\n 'bbox_label': 1,\n 'mask': [[0, 0, 0, 20, 10, 20, 10, 0]],\n 'ignore_flag': 0\n }, {\n 'bbox': [10, 10, 110, 120],\n 'bbox_label': 2,\n 'mask': [[10, 10, 110, 10, 110, 120, 110, 10]],\n 'ignore_flag': 0\n }, {\n 'bbox': [50, 50, 60, 80],\n 'bbox_label': 2,\n 'mask': [[50, 50, 60, 50, 60, 80, 50, 80]],\n 'ignore_flag': 1\n }]\n }","def results(self):\n \n results = {}\n if self.descriptors is not None:\n for i in range(len(self.items)):\n results[self.items[i]] = {'I': self.ivalue(i),\n 'J': self.jvalue(i)}\n else:\n for i in range(len(self.items)):\n results[self.items[i]] = {'I': self.ivalue(i)}\n return results","def what_image_is_this(self, captured_image, reference_images_dictionary, respawn_filter=False):\n matched_array = []\n captured_image_list = captured_image.tolist()\n total = {}\n\n apply_respawn_filter = hasattr(self, 'respawnFilter') and respawn_filter\n\n for item_name, reference_image in reference_images_dictionary.items():\n total[item_name] = 0\n row = 0\n for reference_row in reference_image: # captured_image must not be larger than any reference image\n pixel = 0\n for reference_pixel in reference_row:\n skip = False\n # print(\"Reference Pixel: \" + str(reference_pixel) +\n # \" Captured Pixel: \" + str(captured_image_list[row][pixel][0]))\n if apply_respawn_filter: # apply filter when respawning\n if self.respawnFilter[\"respawn-filter\"][row][pixel] == 0:\n # filter out this pixel, it could be red or white and will be covering the hero\n skip = True\n if not skip:\n total[item_name] += 1\n if reference_pixel == captured_image_list[row][pixel][0]:\n matched_array.append(item_name)\n pixel = pixel + 1\n row = row + 1\n counter = Counter(matched_array)\n most_common = counter.most_common()\n ratios = {}\n for item in most_common:\n ratios[item[0]] = item[1] / total[item[0]]\n return ratios","def _result_dict_for_single_example(image, key, detections, gt_boxes, gt_labels):\n image_shape = tf.shape(image)\n detection_boxes = detections[\"detection_boxes\"][0]\n detection_scores = detections[\"detection_scores\"][0]\n detection_classes = tf.to_int64(detections[\"detection_classes\"][0])\n num_detections = tf.to_int32(detections[\"num_detections\"][0])\n detection_boxes = tf.slice(detection_boxes, begin=[0, 0], size=[num_detections, -1])\n detection_classes = tf.slice(detection_classes, begin=[0], size=[num_detections])\n detection_scores = tf.slice(detection_scores, begin=[0], size=[num_detections])\n # scale to abs coordinates\n y_scale, x_scale = image_shape[1], image_shape[2]\n detection_boxes = _scale_to_abs_coord(detection_boxes, y_scale, x_scale)\n gt_boxes = _scale_to_abs_coord(gt_boxes, y_scale, x_scale)\n return dict(image=image,\n key=key,\n detection_boxes=detection_boxes,\n detection_scores=detection_scores,\n detection_classes=detection_classes,\n num_detections=num_detections,\n groundtruth_boxes=gt_boxes,\n groundtruth_classes=gt_labels)","def compute_metrics(self, results: list) -> dict:","def generate_scanset_metadata( image_set_dictionary, html_base_path, session_id ):\r\n\tcur_subj_info = {}\r\n\r\n\t\"\"\"need to think through the data structure a bit more.... but can always adjust later \"\"\"\r\n\tcur_subj_info['session_id'] = session_id\r\n\t#cur_subj_info['img_id'] = counter\r\n\tcur_subj_info['subject_id'] = session_id.split('/')[0]\r\n\tglobal counter\r\n\t\r\n\tnii_image_dict = image_set_dictionary['nii_images']\r\n\tpng_image_dict = image_set_dictionary['png_image_set']\r\n\r\n\tscan_metadata = {}\r\n\tfor scan in nii_image_dict:\r\n\t\tprint \"propcessing \", scan\r\n\t\tnii_img = nii_image_dict[scan]['base_image'][0]\r\n\t\tprint nii_img\r\n#\t\tif 'mask' not in scan:\r\n\r\n#\t\tif 'mask' not in scan:\r\n#\t\t\tnii_img = nii_image_dict[scan]['base_image'][0]\r\n#\t\telse:\r\n#\t\t\tcontinue\r\n#\t\tprint \"HI DAVE!\"\r\n\r\n\r\n\t\tif not nii_img:\r\n\t\t\tprint \"did not find base image for\",nii_image_dict\r\n\t\t\tcontinue\r\n\t\tpng_img = html_path_root+ png_image_dict[scan]\r\n\t\tprint nii_img,\"is being passed\"\r\n\r\n\t\t(dim_x, dim_y, dim_z, vox_size_x, vox_size_y, vox_size_z, image_orientation )= igcf.get_nii_image_info(nii_img)\r\n\t\timage_info = Image.open(png_img)\r\n\t\twidth, height = image_info.size\r\n\t\t#print width,height,dim_x,dim_y,dim_z,vox_size_x,vox_size_y,vox_size_z\t\r\n\t\tscan_info = {}\r\n\t\tscan_info['slice_width'] = dim_x\r\n\t\tscan_info['slice_height'] = dim_y\r\n\t\tscan_info['num_slices'] = dim_z\r\n\t\tscan_info['main_image_width'] = width\r\n\t\tscan_info['main_image_height'] = height\r\n\t\tscan_info['nii_image'] = nii_img\r\n\t\tscan_info['base_url'] = png_img.replace(html_path_root,'')\r\n\t\tscan_metadata[scan] = scan_info\r\n\r\n\t\t### There can be one or MORE masks for a given base image... so I will return a list of\r\n\t\t#dictionaries..\r\n\t\tmask_list = nii_image_dict[scan]['masks']\r\n\t\tmask_id = 0\r\n\r\n\t\tmask_info_list = []\r\n\t\tfor mask in mask_list:\r\n\t\t\tcur_mask_info = {}\r\n\t\t\t### I'll call the mask by it's basename\r\n\t\t\tprint mask,\"was passed...\"\r\n\t\t\tmask_base = os.path.basename(mask)\r\n\t\t\tnii_img = nii_image_dict[scan]['masks'][mask_id]\r\n\t\t\tprint nii_image_dict,'mask_id is',mask_id\r\n\t\t\tprint \"nii maeg found should be\",nii_img\r\n\t\t\tif not nii_img:\r\n\t\t\t\tprint \"did not find a valid mask image for \",nii_image_dict\r\n\t\t\t\tcontinue\r\n\r\n\r\n\t\t\tcur_mask_info['name'] = mask_base\r\n\t\t\tcur_mask_info['id'] = mask_id\r\n\t\t\tcur_mask_info['nii_file'] = nii_img\r\n\t\t\t## NEED TO ADD IN THE MASK_URL\r\n#\t\t\tcur_mask_info['mask_url'] = \t\t\t\r\n\t\t\tprint png_image_dict\r\n\t\t\tpng_img = html_path_root+ png_image_dict[scan]\r\n\t\t\tprint nii_img,\"is being passed\"\r\n\t\t\tcur_mask_info['mask_url'] = png_img.replace(html_path_root,'')\r\n\r\n\t\t\tmask_info_list.append( cur_mask_info )\r\n\t\r\n\t\t\r\n\t\t\tmask_id +=1\r\n#\t\tprint cur_mask_info\r\n\t\tcur_subj_info['masks'] = mask_info_list\r\n\t\tscan_metadata[scan]['masks'] = [ mask_info_list]\r\n#\t\tprint mask_info_list\r\n\r\n\tcur_subj_info['image_data'] = scan_metadata\r\n\tcounter += 1\t\r\n\treturn { 'session_name': session_id , 'session_metadata': cur_subj_info }","def images(self) -> dict:\n raise NotImplementedError","def __getitem__(self, index):\n (path_1, path_2, is_same, id0, id1) = self.pairs[index]\n \n try:\n img1, img2 = self._load_img(path_1), self._load_img(path_2)\n return {'img1': img1, 'img2': img2, 'is_same': is_same, 'id0': id0, 'id1': id1}\n except FileNotFoundError:\n return {'is_same': is_same, 'id0': id0, 'id1': id1, 'path1': path_1, 'path2': path_2}","def generate_results(img_name, note_names, note_durations):\n construct_output(indent_level=\"block\",\n message=\"Matching the elements of the image ({}) with the values given by the networks.\"\n .format(img_name))\n\n results = [] # A list that will be returned.\n\n # Path that contains the information about the original image.\n image_path = os.path.abspath(os.path.join(str(Path(__file__).parent.parent.parent), 'resources', 'input_images'))\n image_path = os.path.join(image_path, 'input_images_individual_elements')\n image_path = os.path.join(image_path, img_name[:-4])\n\n # Find all the unrecognized elements and their names.\n unrecognized_elements_path = os.path.join(image_path, \"unrecognized\")\n unrecognized_elements = os.listdir(unrecognized_elements_path)\n\n results_and_positions = []\n for el in unrecognized_elements:\n row_number = el[(el.index(\"_row\") + len(\"_row\")):(el.index(\"_slice\"))]\n el_number = el[(el.index(\"slice\") + len(\"slice\")):]\n el_number = el_number[:el_number.index(\"_\")]\n results_and_positions.append((el, row_number, el_number))\n results_and_positions.sort(key=lambda k: (int(k[1]), int(k[2])))\n if len(note_names) != len(note_durations) != len(unrecognized_elements):\n print(\"generator.py = Not enough elements were recognized!\")\n\n for index, el in enumerate(results_and_positions):\n el = el[0]\n if el.endswith(\".png\"):\n results.append((el[len(\"img02_\"):-len(\".png\")], note_names[index], note_durations[index]))\n\n # Find all the recognized elements and their names.\n recognized_elements_path = os.path.join(image_path, \"recognized\", img_name[:-4] + \".txt\")\n\n with open(recognized_elements_path, 'r') as file:\n recognized_elements = file.readlines()\n if len(recognized_elements) == 0:\n print(\"generator.py = Error in reading image file info!\")\n exit(-1)\n for index, content in enumerate(recognized_elements):\n if content.endswith(\"\\n\"):\n recognized_elements[index] = content[:-1]\n for index, el in enumerate(recognized_elements):\n el_name = el[(el.find(\"template_\") + len(\"template_\")):el.find(\".png\")]\n if el_name.startswith(\"Z_barline\"):\n el_name = \"barline\"\n results.append((el[len(\"img02_\"):-len(\".png\")], el_name, \"-\"))\n\n # Join the individual elements with their descriptions.\n results_and_positions = []\n for result in results:\n x = result[0]\n row_number = x[(x.index(\"row\") + len(\"row\")):x.index(\"_slice\")]\n el_number = x[(x.index(\"slice\") + len(\"slice\")):]\n el_number = el_number[:el_number.index(\"_\")]\n results_and_positions.append((result, row_number, el_number))\n\n # Sort the elements with the order found in the original image.\n results_and_positions.sort(key=lambda k: (int(k[1]), int(k[2])))\n\n results = []\n for x in results_and_positions:\n results.append(x[0])\n\n construct_output(indent_level=\"block\",\n message=\"Results generated.\"\n .format(img_name))\n\n return results","def get_images_and_labels(tampered_path, authentic_path):\n tampered_dir = tampered_path\n authentic_dir = authentic_path\n images = {}\n for im in glob.glob(authentic_dir):\n images[im] = {}\n images[im]['mat'] = cv2.imread(im)\n images[im]['label'] = 0\n for im in glob.glob(tampered_dir):\n images[im] = {}\n images[im]['mat'] = cv2.imread(im)\n images[im]['label'] = 1\n return images","def __call__(self, results):\n for key in results.get('img_fields', ['img']):\n results[key] = general_ocr.imnormalize(results[key], self.mean, self.std,\n self.to_rgb)\n results['img_norm_cfg'] = dict(\n mean=self.mean, std=self.std, to_rgb=self.to_rgb)\n return results","def get_matches(jig_sol, blocks_roi):\n\n match_data = {}\n height, width,= jig_sol.shape\n font = cv2.FONT_HERSHEY_SIMPLEX\n\n #identify the puzzle peice number based on the peice block position in solution image\n identity = {\n (1, 1): 1,\n (2, 1): 2,\n (3, 1): 3,\n (1, 2): 4,\n (2, 2): 5,\n (3, 2): 6,\n (1, 3): 7,\n (2, 3): 8,\n (3, 3): 9,\n }\n\n #iterate through the blocks roi\n for i in blocks_roi.keys():\n blk = blocks_roi[i].copy()\n blk = cv2.cvtColor(blk,cv2.COLOR_BGR2GRAY)\n max_list = []\n\n #for eack blk rotate is by 90 degrees and try template matching\n for k in range(0,360,90):\n #cv2.resize(blk,(int(width/3),int(height/3)), interpolation= cv2.INTER_CUBIC)\n blk_copy = imutils.rotate_bound(blk, -k)\n\n #get the resulting heat map of template matching\n result = cv2.matchTemplate(jig_sol,blk_copy,cv2.TM_CCOEFF_NORMED)\n\n #get the max value and its location in the heat map\n _, max_val, _, max_loc = cv2.minMaxLoc(result)\n\n #append a tuple consisting of max location, value and peice rotation to max_list\n max_list.append((max_loc, max_val, k))#((k+1)*90)%360))\n\n #find the location with maximum value of template matching regardless of peice rotation\n top_left = max(max_list, key=lambda x: x[1])[0]\n\n #get the peice rotation of that template matching\n rot = max(max_list, key=lambda x: x[1])[2]\n\n #calculate the bottom right cordinates of the block\n bottom_right = (top_left[0] + int(width/3), top_left[1] + int(height/3))\n\n #find the center of the block\n centx = 0\n centy = 0\n for (l,m) in [top_left, bottom_right]:\n centx += l\n centy += m\n centx = int(centx/2)\n centy = int(centy/2)\n\n #get the puzzle peice block position in solution image\n piece = (math.ceil(3*centx/width), math.ceil(3*centy/height))\n\n if piece not in identity.keys():\n continue\n\n match_data.update({i: (identity[piece], rot)})\n\n return match_data","def _init_results(self) -> None:\n pt_bond_dimensions = {}\n for site, pt in enumerate(self._process_tensors):\n if pt is not None:\n pt_bond_dimensions[site] = pt.get_bond_dimensions()\n\n self._results = {\n 'time':[],\n 'norm': [],\n 'bond_dimensions': [],\n 'dynamics': {},\n 'pt_bond_dimensions': pt_bond_dimensions,\n }\n for sites in self._dynamics_sites:\n self._results['dynamics'][sites] = Dynamics(name=f\"site{sites}\")","def detail_matching(self):\n paradic = self.cfg['param']['paradic']\n work_dir = self.work_dir\n \n x = float(self.cfg['param']['x']) # selected pixel in the first image\n y = float(self.cfg['param']['y'])\n \n # sift parameters\n # number of bins in the orientation histogram\n n_bins = int(paradic['n_bins']) \n n_hist = int(paradic['n_hist']) \n # descriptor of n_hist X n_hist weighted histograms with n_ori\n n_ori = int(paradic['n_ori']) \n delta_min = float(paradic['delta_min'])\n sigma_min = float(paradic['sigma_min'])\n sigma_in = float(paradic['sigma_in'])\n lambda_ori = float(paradic['lambda_ori'])\n lambda_descr = float(paradic['lambda_descr'])\n #threshold defining reference orientations\n n_spo = int(paradic['n_spo'])\n \n # Read feature vectors from output files\n if (os.path.getsize(work_dir+'OUTmatches.txt') > 0 ):\n pairdata = find_nearest_keypoint(work_dir+'OUTmatches.txt', y, x)\n \n illustrate_pair(pairdata, n_bins, n_hist, n_ori, work_dir)\n\n \n # Read keys coordinates.\n d = 6+n_bins+n_hist*n_hist*n_ori # size of keydata inside pairdata\n v = n_hist*n_hist*n_ori\n [x1, y1, sigma1, theta1] = [float(x) for x in pairdata[0:4]]\n [o1, s1] = [float(x) for x in pairdata[4+v:4+v+2]]\n [x2a, y2a, sigma2a, theta2a] = [float(x) for x in pairdata[d:d+4]]\n [o2a, s2a] = [float(x) for x in pairdata[d+4+v:d+4+v+2]]\n [x2b, y2b, sigma2b, theta2b] = \\\n [float(x) for x in pairdata[2*d:2*d+4]]\n [o2b, s2b] = [float(x) for x in pairdata[2*d+4+v:2*d+4+v+2]]\n \n draw_one_match(pairdata,\n work_dir+'input_0.png',\n work_dir+'input_1.png',\n d,\n lambda_ori,\n lambda_descr,\n n_hist,\n work_dir+'OUTonepair.png')\n \n \n # Extract thumbnails.\n # keypoint 1 (image 1)\n print ' '.join(['demo_extract_patch', work_dir+'input_0.png',\n str(x1), str(y1), str(sigma1), str(theta1), str(o1), str(s1),\n str(delta_min), str(sigma_min), str(sigma_in), str(n_spo),\n str(lambda_ori), str(lambda_descr), str(n_hist),\n work_dir+\"detail_im1\"])\n proc = self.run_proc(['demo_extract_patch', work_dir+'input_0.png',\n str(x1), str(y1), str(sigma1), str(theta1), str(o1), str(s1),\n str(delta_min), str(sigma_min), str(sigma_in), str(n_spo),\n str(lambda_ori), str(lambda_descr), str(n_hist),\n work_dir+\"detail_im1\"])\n self.wait_proc(proc, timeout=self.timeout)\n \n # keypoint 2a (nearest neighbor in image 2)\n print ' '.join(['demo_extract_patch', work_dir+'input_1.png',\n str(x2a), str(y2a), str(sigma2a), str(theta2a), str(o2a), str(s2a),\n str(delta_min), str(sigma_min), str(sigma_in), str(n_spo),\n str(lambda_ori), str(lambda_descr), str(n_hist),\n work_dir+\"detail_im2a\"])\n proc = self.run_proc(['demo_extract_patch', work_dir+'input_1.png',\n str(x2a), str(y2a), str(sigma2a), str(theta2a), str(o2a), str(s2a),\n str(delta_min), str(sigma_min), str(sigma_in), str(n_spo),\n str(lambda_ori), str(lambda_descr), str(n_hist),\n work_dir+\"detail_im2a\"])\n self.wait_proc(proc, timeout=self.timeout) \n \n # keypoint 2b (second nearest neighbor in image 2)\n proc = self.run_proc(['demo_extract_patch', work_dir+'input_1.png',\n str(x2b), str(y2b), str(sigma2b), str(theta2b), str(o2b), str(s2b),\n str(delta_min), str(sigma_min), str(sigma_in), str(n_spo),\n str(lambda_ori), str(lambda_descr), str(n_hist),\n work_dir+\"detail_im2b\"])\n self.wait_proc(proc, timeout=self.timeout) \n \n \n return 1","def extract_results_test(self):\n assert len(self.results.keys()) != 0\n TESTS = [\n {\n \"input\": {\"molecules\": [\"DDSPDLPK\"], \"score_threshold\": 0.95},\n \"output\": {\n \"formula\": \"C(37)H(59)N(9)O(16)\",\n \"file_name\": \"BSA1.mzML\",\n \"scaling_factor\": 100,\n \"spec_id\": 1337,\n },\n }\n ]\n for test_dict in TESTS:\n for key, n, entry in self.results.extract_results(**test_dict[\"input\"]):\n print(key, entry)\n assert key.formula == test_dict[\"output\"][\"formula\"]\n assert key.file_name == test_dict[\"output\"][\"file_name\"]\n assert entry.scaling_factor == test_dict[\"output\"][\"scaling_factor\"]\n assert entry.spec_id == test_dict[\"output\"][\"spec_id\"]\n # print(self.results)\n # print(self.results.lookup)\n assert n == 0","def post_calculations(moving_dataset_image_ids, result=None):\n if result is None:\n result = {}\n\n for _id in moving_dataset_image_ids:\n img = img_data(_id, util.DATA_FOLDER, util.TEMP_FOLDER_PATH)\n img.load_db_transforms()\n\n img_pre = img_data(img.fixed_image, util.DATA_FOLDER, util.TEMP_FOLDER_PATH)\n img_pre.load_db_transforms()\n\n reg_vol = util.transform_volume(img.reg_img_filepath, img_pre.get_transforms())\n vol = util.TEMP_FOLDER_PATH + util.get_basename(basename(reg_vol)) + '_BE.nii.gz'\n\n mult = ants.MultiplyImages()\n mult.inputs.dimension = 3\n mult.inputs.first_input = reg_vol\n mult.inputs.second_input = util.TEMPLATE_MASK\n mult.inputs.output_product_image = vol\n mult.run()\n\n label = \"img\"\n if label in result:\n result[label].append(vol)\n else:\n result[label] = [vol]\n\n for (segmentation, label) in util.find_reg_label_images(_id):\n segmentation = util.transform_volume(segmentation, img_pre.get_transforms(), label_img=True)\n if label in result:\n result[label].append(segmentation)\n else:\n result[label] = [segmentation]\n return result","def parse_image_data(image_urls):\n\n # Initialize images.\n initialized_ClImages = list(map(ClImage, image_urls))\n app = ClarifaiApp(api_key=CLARIFAI_KEY)\n image_info = {}\n\n # Obtain relevant tag information\n try:\n general_response = app.models.get('general-v1.3').predict(initialized_ClImages)\n except ApiError as e:\n error = json.loads(e.response.content)\n pprint('error: {}'.format(error))\n\n else:\n for item in general_response['outputs']:\n video_id = item['input']['data']['image']['url'][23:34]\n thumbnail_tags = set()\n for tag in item['data']['concepts']:\n if tag['value'] > .9:\n tag_string = tag['name'].strip().lower()\n if not Tag.query.filter(Tag.tag == tag_string).first():\n add_tag_data(tag_string)\n thumbnail_tags.add(tag_string)\n image_info[video_id] = {'tags': thumbnail_tags}\n\n\n # Obtain nsfw score\n try:\n nsfw_response = app.models.get('nsfw-v1.0').predict(initialized_ClImages)\n except ApiError as e:\n error = json.loads(e.response.content)\n pprint('error: {}'.format(error))\n\n else:\n for item in nsfw_response['outputs']: #nsfw_r['outputs'] is a list\n video_id = item['input']['data']['image']['url'][23:34]\n nsfw_score = round(item['data']['concepts'][-1]['value'] * 100)\n image_info[video_id]['nsfw_score'] = nsfw_score\n\n\n # Obtain color data\n # try:\n # color_response = app.models.get('color').predict(initialized_ClImages)\n # except ApiError as e:\n # error = json.loads(e.response.content)\n # pprint('error: {}'.format(error[-100:]))\n\n # else:\n # for item in color_response['outputs']:\n # video_id = item['input']['data']['image']['url'][23:34]\n # color_tags = {}\n # for color in item['data']['colors']: # item['data']['colors'] is a list\n # if color['value'] > .2:\n # color_hex = color['w3c']['hex'].rstrip().lower()\n # color_name = color['w3c']['name'].rstrip().lower()\n # if not Color.query.filter(Color.hex_code == color_hex).first():\n # add_color_data(color_hex, color_name)\n # image_info[video_id]['colors'] = color_tags\n\n return image_info","def result_as_dict_matches(query_result):\n matches_keys = ['match_id', 'match_date', 'round_name', 'first_squad_name',\n 'first_squad_score', 'first_squad_points',\n 'second_squad_name', 'second_squad_score',\n 'second_squad_points']\n list_of_dict = []\n for i in range(len(query_result)):\n dictionary = dict(zip(matches_keys, query_result[i]))\n list_of_dict.append(dictionary)\n i += 1\n return list_of_dict","def patternMatchingTest(matchingAlgorithm,\n observations= defaultCOMPREFdump , \n modelsFolder= defaultWRFdumpsFolder,\n #key1='0828',\n key1='',\n key2='',\n #key3='',\n verbose=True,\n makeSmallImages=True,\n smallDpi=40,\n frameDpi = 600 , # added 2013-09-23\n loadFormerAnalysisResult=False, #added 2013-09-23\n #summary_folder=summary_folder, # needed if loadFormerAnalysisResult is True\n summary_folder=summary_folder+timeString + '/', # 2013-09-27\n matching_algorithm_name=\"\", # needed if loadFormerAnalysisResult is True\n panel_cols = 5,\n panel_rows = 5,\n useCV2 = False, # switch off opencv since it doesn't work yet\n useCV = False,\n averageScoreOrdering=True, # take an average of the ordering to get a total score for the entire model over the time series\n #kwargs = {}, # keyword arguments for the matchinging algorithm, just in case\n **kwargs\n ):\n if key1 != '' or key2 != '':\n summary_folder += '_'.join([key1, key2])\n if not averageScoreOrdering:\n summary_folder += '_nonaver'\n summary_folder += '%dx%d/' % (panel_rows, panel_cols)\n\n t0 = time.time()\n if loadFormerAnalysisResult:\n mv_folder = summary_folder + matching_algorithm_name + '/'\n result = pickle.load(open(mv_folder + 'result.pydump', 'r'))\n\n \"\"\"\n result= {'best matches' : bestMatchAtEachTime, \n 'best overall match' : bestMatchOverall, \n 'all scores' : scores,\n 'matching algorithm' : matching_algorithm_name,\n 'dataset1[obs]' : observations,\n 'dataset2[models]' : modelsFolder,\n 'time spent' : time.time() - t0,\n 'mv folder' : mv_folder, #added 2013-09-23\n 'ordering' : ordering, #added 2013-09-23\n }\n \"\"\"\n bestMatchAtEachTime = result['best matches'] \n bestMatchOverall = result['best overall match']\n scores = result['all scores']\n matching_algorithm_name = result['matching algorithm']\n observations = result['dataset1[obs]']\n modelsFolder = result['dataset2[models]']\n #mv_folder = result['mv folder']\n ordering = result['ordering']\n\n # THE FOLLOWING 30 LINES ARE COPIED FROM \"ELSE...\" BELOW\n ########### ******\n ##\n #\n ds1 = pickle.load(open(observations,'r'))\n ds1.setFloor(0) # added 2013-09-27\n ds1.list = [v for v in ds1.list if (key1 in v.dataTime or key1 in v.name) \\\n and (key2 in v.dataTime or key2 in v.name)]\n ds1.list.sort(key=lambda v:v.dataTime)\n ds1.setVmin(-40) #line added 2013-09-23\n # debug\n #print ds1.list\n print 'loading list with keywords [ %s %s ]' %(key1, key2) \n ds1.load(verbose=verbose)\n # end debug\n modelsList = [v for v in os.listdir(modelsFolder) if '.pydump' in v]\n modelsList.sort(key = lambda v: int(re.findall(r'\\d+',v)[0]))\n print \"modelsList\", modelsList\n scores = {} # scores = {model number: [('dataTime', score), ('dataTime', score),...], ...}\n print 'loading ds2 (model data)'\n wrfs = [] # line added 2013-09-23\n for path in modelsList:\n modelIndex = int(re.findall(r'\\d+', path)[0])\n ds2 = pickle.load(open(modelsFolder+path, 'r'))\n ds2.setFloor(0) # added 2013-09-27\n # debug\n #ds2.list = [v for v in ds2.list if key1 in v.dataTime] # line redundant, 2013-09-23\n ds1.list, ds2.list = ds1.intersect(ds2) #line added 2013-09-23 \n ds2.setVmin(-40) #line added 2013-09-23\n wrfs.append(ds2) # line added 2013-09-23\n # end debug\n\n ####################\n\n\n else:\n # THE FOLLOWING 30 LINES ARE COPIED TO \"IF...\" ABOVE\n ########### ******\n ##\n #\n ds1 = pickle.load(open(observations,'r'))\n ds1.list = [v for v in ds1.list if (key1 in v.dataTime or key1 in v.name) \\\n and (key2 in v.dataTime or key2 in v.name)]\n ds1.list.sort(key=lambda v:v.dataTime)\n ds1.setVmin(-40) #line added 2013-09-23\n # debug\n #print ds1.list\n print 'loading list with keywords [ %s %s ]' %(key1, key2) \n ds1.load(verbose=verbose)\n ds1.setFloor(0) # added 2013-09-27\n # end debug\n modelsList = [v for v in os.listdir(modelsFolder) if '.pydump' in v]\n modelsList.sort(key = lambda v: int(re.findall(r'\\d+',v)[0]))\n print \"modelsList\", modelsList\n scores = {} # scores = {model number: [('dataTime', score), ('dataTime', score),...], ...}\n print 'loading ds2 (model data)'\n wrfs = [] # line added 2013-09-23\n ds1.commonMaskSet=False # tempory attribute added 2013-09-27\n for path in modelsList:\n modelIndex = int(re.findall(r'\\d+', path)[0])\n ds2 = pickle.load(open(modelsFolder+path, 'r'))\n # debug\n #ds2.list = [v for v in ds2.list if key1 in v.dataTime] # line redundant, 2013-09-23\n ds1.list, ds2.list = ds1.intersect(ds2) #line added 2013-09-23 \n ds2.setFloor(0) # added 2013-09-27\n ds2[0].matrix.mask += ds1[0].matrix.mask # added 2013-09-27\n ds2.setCommonMask() # added 2013-09-27\n if not ds1.commonMaskSet: # added 2013-09-27\n ds1[0].matrix.mask +=ds2[0].matrix.mask # added 2013-09-27\n ds1.setCommonMask() # added 2013-09-27\n ds2.setVmin(-40) #line added 2013-09-23\n wrfs.append(ds2) # line added 2013-09-23\n # end debug\n\n #######################################################################################################################\n # key lines\n print 'computing scores between %s and %s ...' %(ds1.name, ds2.name)\n s, matching_algorithm_name = matchingAlgorithm(ds1, ds2, **kwargs) # list of the form [('20120612.0000', 0.9),...] \n scores[modelIndex] = dict(s) # { model index: {'dataTime': score,...}, ...}\n # debug\n if verbose:\n print scores\n print '..................................................'\n #\n ########################################################################################################################\n\n print 'sorting the results...'\n #print 'scores', '\\n', scores #debug\n # bestMatchAtEachTime\n T1 = [v.dataTime for v in ds1]\n T2 = [v.dataTime for v in ds2] #any ds2 will do - they are the same, in this case we pick the last one \n T = set(T1).intersection(T2)\n dataTimeList = sorted(list(T))\n\n #print 'dataTimeList:', dataTimeList #debug\n highestScores = []\n ordering = [] #added 2013-09-23\n\n if averageScoreOrdering: # take the average scores before ordering to make it uniform\n ave_scores = [ (v, np.mean( [scores[v].values() ] ) ) for v in scores.keys()]\n ave_scores = dict(ave_scores)\n print '..taking averages...............................'\n print ave_scores\n for u in scores.keys():\n for t in scores[u].keys():\n scores[u][t] = ave_scores[u]\n print '\\n'.join([str(scores[v]) for v in scores.keys()])\n\n for t in dataTimeList:\n print t\n highestScoreAt_t = max([scores[u][t] for u in scores.keys()]) # u = model index (integer 1-20), t = dataTime (string)\n new_entry = (t, highestScoreAt_t, [u for u in scores.keys() if scores[u][t]==highestScoreAt_t])\n print new_entry\n highestScores.append(new_entry)\n #added 2013-09-23\n ordering.append([0]+sorted(range(1, len(modelsList)+1), key=lambda v: scores[v][t], reverse=True))\n\n\n\n\n # bestMatchOverall\n print scores # debug\n averageScores = [(Model, np.mean(scores[Model].values())) for Model in scores.keys()] # list of [(Model, mean score),...]\n highestScore = max([v[1] for v in averageScores])\n bestMatch = [u[0] for u in averageScores if u[1]==highestScore] \n # debug\n print 'average scores , highest average, best match', averageScores, highestScore, bestMatch\n #xx= raw_input('press enter') \n\n\n bestMatchAtEachTime = highestScores\n bestMatchOverall = (highestScore, bestMatch)\n\n # the following lines do not work well since oftentimes we have multiple pointers (names) to a single object/function\n # and so they are replaced by returning the function name in the function itself\n #allvars = vars()\n #matching_algorithm_name = [v for v in allvars if allvars[v]==matchingAlgorithm][0]\n #mv_folder = summary_folder + matching_algorithm_name + '/'\n mv_folder = summary_folder + matching_algorithm_name + '/' # added 2013-09-27\n\n result= {'best matches' : bestMatchAtEachTime, \n 'best overall match' : bestMatchOverall, \n 'all scores' : scores,\n 'matching algorithm' : matching_algorithm_name,\n 'dataset1[obs]' : observations,\n 'dataset2[models]' : modelsFolder,\n 'time spent' : time.time() - t0,\n 'mv folder' : mv_folder, #added 2013-09-23\n 'ordering' : ordering, #added 2013-09-23\n }\n\n try:\n os.makedirs(mv_folder)\n print 'folder', mv_folder, \"created!\"\n except:\n print 'folder', mv_folder, 'exists!'\n pickle.dump(result, open(mv_folder+'result.pydump','w'))\n\n ############################################################################\n # to make the videos/collected images\n # calling armor.dataStreamTools.makeVideo.makeVideo()\n # 2013-09-23\n\n from . import makeVideo as mv\n \n try:\n os.makedirs(mv_folder)\n print 'folder', mv_folder, \"created!\"\n except:\n print 'folder', mv_folder, 'exists!'\n # make images\n if makeSmallImages:\n print 'making small images'\n for ds in [ds1] + wrfs:\n #ds.imageFolder= mv_folder + ds.name + '/'\n ds.imageFolder= summary_root + ds.name + '/' #added 2013-09-27 to replace the above line - too tired waiting the plots\n ds.saveImages(drawCoast=True, flipud=True, dpi=smallDpi)\n print ds.imageFolder\n else:\n # just set the paths\n for ds in [ds1]+wrfs:\n ds.imageFolder = summary_root + ds.name +'/' # /home/k/ARMOR/data/KONG-REY/summary/WRF01.DBZ/\n for dbzPattern in ds.list:\n dbzPattern.imagePath = ds.imageFolder + ds.name + dbzPattern.dataTime + '.png'\n print dbzPattern.imagePath\n #ds.setImagePaths() # doesn't for for old pickled objects???\n\n\n # make panels\n\n mv.makeVideo(DSS = [ds1]+ wrfs, # [ds0, ds1, ds2, ds3, ds4, ...], a list of armor.pattern.DBZstream objects\n panel_cols = panel_cols, # number of colums in the panel\n panel_rows = panel_rows, # no need to be filled\n #fourcc = cv.CV_FOURCC('F', 'L', 'V', '1'),\n #fps = defaultFps,\n extension= '.avi',\n #fourcc = cv.CV_FOURCC('P', 'I', 'M', '1'),\n outputFileName =\"\",\n outputFolder= mv_folder,\n sandbox = mv_folder ,\n saveFrames = True, # saving the frames as images\n frameDpi = frameDpi,\n useCV2 = useCV2,\n useCV = useCV,\n ordering = ordering, # ordering of the models\n )\n ############################################################################\n # dumping the results\n try:\n os.makedirs(summary_folder+ 'pydumps/')\n except:\n print 'Folder exists! -- ', summary_folder+ 'pydumps/' \n\n pickle.dump(result, open(summary_folder+ 'pydumps/results_' + matching_algorithm_name+ '+' +\\\n key1 + '+' + key2 + str(int(time.time())) + '.pydump', 'w'))\n\n \n\n print \"time spent:\", time.time()-t0\n return result","def images_mapped(self):\n try:\n return dict([x for x in enumerate(self.images())])\n except:\n return None","def json(self) -> dict:\n search = {\n 'criteria': self.search_criteria,\n 'matchesExact': {\n 'avgIndexDiff': self.avg_index_diff(TestSearchResult.MatchType.EXACT.value),\n 'firstFailIndex': self.fail_index(TestSearchResult.MatchType.EXACT.value),\n 'missedMatches': self.missed_matches(TestSearchResult.MatchType.EXACT.value),\n 'resultsApi': self.get_results(\n TestSearchResult.MatchType.EXACT.value, TestSearchResult.Source.API.value),\n 'resultsLegacy': self.get_results(\n TestSearchResult.MatchType.EXACT.value, TestSearchResult.Source.LEGACY.value)\n },\n 'matchesSimilar': {\n 'avgIndexDiff': self.avg_index_diff(TestSearchResult.MatchType.SIMILAR.value),\n 'firstFailIndex': self.fail_index(TestSearchResult.MatchType.SIMILAR.value),\n 'missedMatches': self.missed_matches(TestSearchResult.MatchType.SIMILAR.value),\n 'resultsApi': self.get_results(\n TestSearchResult.MatchType.SIMILAR.value, TestSearchResult.Source.API.value),\n 'resultsLegacy': self.get_results(\n TestSearchResult.MatchType.SIMILAR.value, TestSearchResult.Source.LEGACY.value)\n },\n 'runTime': self.run_time,\n }\n\n search['matchesExact']['passed'] = (\n len(search['matchesExact']['missedMatches']) == 0 and\n search['matchesExact']['firstFailIndex'] == -1\n )\n search['matchesSimilar']['passed'] = (\n len(search['matchesSimilar']['missedMatches']) == 0 and\n search['matchesSimilar']['firstFailIndex'] == -1\n )\n\n return search","def match(image1,image2,threshold,useRansac=False,t_orientation=30,t_scale=0.5):\r\n im1, keypoints1, descriptors1 = ReadKeys(image1)\r\n im2, keypoints2, descriptors2 = ReadKeys(image2)\r\n #\r\n # REPLACE THIS CODE WITH YOUR SOLUTION (ASSIGNMENT 5, QUESTION 3)\r\n #\r\n # Generate five random matches (for testing purposes)\r\n # matched_pairs = []\r\n # num = 5\r\n # for i in range(num):\r\n # matched_pairs.append([keypoints1[i],keypoints2[i]])\r\n # return DisplayMatches(im1, im2, matched_pairs)\r\n\r\n # END OF SECTION OF CODE TO REPLACE\r\n #\r\n\r\n #q3\r\n matched_pairs = []\r\n between_angles = np.arccos(np.dot(descriptors1, np.transpose(descriptors2)))\r\n for i, row in enumerate(between_angles):\r\n \tratio = sorted(row)[0] / sorted(row)[1]\r\n \tif ratio <= threshold:\r\n\t \tmatched_pairs.append([keypoints1[i], keypoints2[np.where(row == sorted(row)[0])[0][0]]])\r\n # print(matched_pairs)\r\n if useRansac is False:\r\n return DisplayMatches(im1, im2, matched_pairs)\r\n\t# \r\n\r\n #q4\r\n repetition = 10\r\n subsets = [[]] * repetition\r\n for i in range(repetition):\r\n r = random.randint(0, len(matched_pairs))\r\n for match in matched_pairs:\r\n ds1, ds2 = matched_pairs[r][1][2]/matched_pairs[r][0][2], match[1][2]/match[0][2]\r\n do1, do2 = (matched_pairs[r][1][3]-matched_pairs[r][0][3]), (match[1][3]-match[0][3])\r\n if abs(ds2 - ds1) <= t_scale * ds1 and abs(do2 - do1) % (2 * math.pi) <= t_orientation:\r\n subsets[i].append(match)\r\n\r\n max_i, max_len = 0, subsets[0]\r\n for i in range(10):\r\n l = len(subsets[i])\r\n if l > max_len:\r\n max_len = l\r\n max_i = i\r\n\r\n im3 = DisplayMatches(im1, im2, subsets[max_i])\r\n return im3","def __processFollowerResults(self):\r\n\r\n resultCacheFileName = os.path.join(core.FW_conf['test_result_dir'], os.path.split(core.FW_conf['test_result_dir'])[1] + \".cache\" )\r\n if os.path.isfile(resultCacheFileName):\r\n resultDict = cPickle.load(open(resultCacheFileName, 'rb'))\r\n debug.brf('Loading existing results from cache...')\r\n else:\r\n resultDict = {}\r\n\r\n # read which iteration this was\r\n if not resultDict.has_key(self.tcId):\r\n resultDict[self.tcId] = {}\r\n iteration = len(resultDict[self.tcId]) + 1\r\n resultDict[self.tcId][iteration] = {} # new iteration for the result dictionary\r\n\r\n leaderResult = Dictionary[str, str]()\r\n res = self.result.getResult()\r\n\r\n try:\r\n if str(res) == 'Failed':\r\n errornode = self.result.getError()\r\n if errornode != None and errornode.getAttribute('reason') == 'X-Files generated during testcase execution' or \\\r\n errornode.getAttribute('reason') == 'Phone reset detected':\r\n leaderResult.Add('Result', 'X-files')\r\n elif errornode != None:\r\n leaderResult.Add('Result', str(errornode.getAttribute('reason')))\r\n except:\r\n debug.err('error reading result node')\r\n leaderResult.Add('Result', str(res))\r\n\r\n if not leaderResult.ContainsKey('Result'): # if result was not added\r\n leaderResult.Add('Result', str(res))\r\n\r\n folder = os.path.split(core.FW_conf['test_result_dir'])[1]\r\n\r\n leaderResult.Add('ResultDir', folder)\r\n leaderResult.Add('Duration', str( int( time.time() - self.leader_start_time )) )\r\n #leaderResult.Add('EndTime', strftime(\"%Y-%m-%d %H:%M:%S\", gmtime()))\r\n leaderResult.Add('EndTime', DateTime.Now.ToString(\"yyyy-MM-dd HH:mm:ss\"))\r\n for i in range(1, len(self._capturedStillImages) + 1 ):\r\n debug.out(self._capturedStillImages[i-1])\r\n relativeFileName = os.path.join('videos',self._capturedStillImages[i-1])\r\n leaderResult.Add('Capture%s' % str(i), relativeFileName)\r\n\r\n try:\r\n relativeFileName = os.path.join('videos',os.path.splitext(core.FW_conf['test_result_name'])[0] + '.avi')\r\n videoFile = os.path.join(core.FW_conf['test_result_dir'], relativeFileName)\r\n if os.path.isfile(videoFile): # change video address if video file is found\r\n leaderResult.Add('Video', str(relativeFileName))\r\n #debug.brf('Added leader Video: %s' % str(videoFile))\r\n except Exception as e:\r\n debug.brf('Could not add Leader video to result: %s' % e.message)\r\n\r\n leaderResult.Add('Machine', str(Dns.GetHostName())) # add Leader machine name\r\n\r\n followerResults = core.FW_conf['leader'].GetResults()\r\n followerResults.Add('Leader', leaderResult) # Add leader result to dictionary\r\n\r\n for followerRes in followerResults:\r\n clientId = followerRes.Key # string client id\r\n result = followerRes.Value # dictionary of result, link\r\n\r\n if not resultDict[self.tcId][iteration].has_key(clientId):\r\n resultDict[self.tcId][iteration][clientId] = {}\r\n else:\r\n debug.err('\\n**** %s has already result for iteration %s on test case %s. Will not override.\\n ***' % (clientId, iteration, self.tcId))\r\n continue\r\n\r\n # loop through result\r\n debug.brf('Results for client %s' % clientId)\r\n for res in result:\r\n debug.brf('Adding %s %s' % (str(res.Key), str(res.Value)) )\r\n resultDict[self.tcId][iteration][clientId][str(res.Key)] = str(res.Value)\r\n\r\n if resultDict:\r\n # dump resultDict to cache\r\n try:\r\n cacheFile = open(resultCacheFileName, 'wb')\r\n cPickle.dump(resultDict, cacheFile, 0)\r\n cacheFile.close()\r\n except:\r\n for i in range(10):\r\n debug.err('*' * 40 + ' FAILED TO FLUSH RESULTS TO DISK! ' + '*' * 40)","def getAllForImages(self):\n imageDict = {}\n for id, name in self.getAll().items():\n imageDict[id] = {}\n imageDict[id][\"name\"] = name\n imageDict[id][\"filename\"] = \"The_Steamer_Great_Western_small.jpg\"\n\n return imageDict","def _calculate_distribution(self, images):\n positions = {}\n target_width = target_height = 0\n for block, url in self.image_declarations:\n file_name = normalize_filename(url)\n if file_name in positions:\n positions[file_name]['blocks'].append(block)\n else:\n block_images = images[file_name]\n offset_x = target_width\n offset_y = 0\n\n width, height = block_images[1].size\n positions[file_name] = {\n 'x': offset_x, 'y': offset_y,\n 'width': width, 'height': height,\n 'file_name': file_name, 'images': block_images,\n 'blocks': [block]\n }\n\n target_width += width + self.space_x\n target_height = max(target_height, height)\n\n target_width -= self.space_x\n return {\n 'width': target_width,\n 'height': target_height,\n 'positions': positions.values(),\n }","def compute_similarity(im1, im2, test_id):\n\n results = dict()\n results['id'] = test_id\n results['test_im'] = im2\n results['mse_value'], results['mse_map'] = compute_mse(im1, im2)\n results['ssim_value'], results['ssim_map'] = compute_ssim(im1, im2, 5)\n results['cw_ssim_value'], results['cw_ssim_map'] = compute_cw_ssim(im1, im2, 30)\n results['gms_value'], results['gms_map'] = compute_gms(im1, im2)\n results['fsim_value'], results['pc_max_map'] = compute_fsim(im1, im2)\n\n return results","def _build_results(self):\n results = {}\n cols = []\n for pol in POLLUTANTS:\n for adj in ADJUSTMENTS:\n cols.append(get_rate_column(pol, adjustment=adj, generated=False))\n cols.append(get_column(pol, adjustment=adj))\n cols.append(\"net_consumed_mwh\")\n for ba in self.regions:\n results[ba] = pd.DataFrame(\n index=self.generation.index, columns=cols, dtype=np.float64\n )\n return results","def get_images(self):\n return {'source': utils.image_from_tensor(self.source[0]),\n 'output': utils.image_from_tensor(self.output.data[0]),\n 'target': utils.image_from_tensor(self.target[0])}","def run(self):\n return_dict = {}\n for x in self.file_list:\n return_dict[x] = self.parseFile(x)\n\n return return_dict","def maps(matches):\n if not matches:\n print('no matches')\n return None\n\n maps = {}\n for match in matches:\n map = match['map']\n score = match['score']\n result = match['result']\n # add map to dictionary\n if not map in maps:\n maps[map] = {'wins': 0, 'losses': 0, 'winrate': 0.0}\n map_data = maps[map]\n if result:\n map_data['wins'] += 1\n else:\n map_data['losses'] += 1\n matches_played = map_data['wins'] + map_data['losses']\n\n # compute moving average on winrate\n map_data['winrate'] = map_data['winrate'] + (result - map_data['winrate']) / matches_played\n return maps","def getList(self):\n labelMap = {}\n imageMap = {}\n key = []\n index = 0\n\n for root, dirs, files in os.walk(self.path_data):\n for file in files:\n # If .png or .jpg file found then\n if file.endswith(tuple(config.imageFormat)):\n key.append(index)\n labelMap[index] = preprocessing.getLabel(file)\n imageMap[index] = os.path.join(root, file)\n\n index += 1\n\n else:\n continue\n\n return key, imageMap, labelMap","def get_results(self):\n if not self.sim_stop_flag.value:\n # we are only interested in entries of frames processed > 0\n max_index = np.sum(np.array(self.sim_results_struct.frames[0:50]) > 0)\n\n # convert the cstruct to dictionary\n return dict([(x, getattr(self.sim_results_struct, x)[0:max_index]) for (x,_) in self.sim_results_struct._fields_])\n else:\n return self.results","def get_images_and_labels_nc():\n refs = get_ref_df()\n images = {}\n for _, data in refs.iterrows():\n if data['ProbeFileName'] in images:\n continue\n im = data['ProbeFileName']\n images[im] = 1 if data['IsTarget'] == 'Y' else 0\n return images","def prepare_results(self) -> dict:\n if not hasattr(self, \"results\"):\n raise AttributeError(\n \"Results have not been finalized. Please call \"\n \"finalize_results() before saving output.\"\n )\n\n output = {\n \"armory_version\": armory.__version__,\n \"config\": self.config,\n \"results\": self.results,\n \"timestamp\": int(self.time_stamp),\n }\n return output","def __call__(self, results):\n\n if random.uniform(0, 1) > self.prob:\n return results\n\n if 'img_fields' in results:\n assert results['img_fields'] == ['img'], \\\n 'Only single img_fields is allowed'\n img = results['img']\n\n h, w, c = img.shape\n ratio = random.uniform(self.min_ratio, self.max_ratio)\n # speedup expand when meets large image\n if np.all(self.mean == self.mean[0]):\n expand_img = np.empty((int(h * ratio), int(w * ratio), c),\n img.dtype)\n expand_img.fill(self.mean[0])\n else:\n expand_img = np.full((int(h * ratio), int(w * ratio), c),\n self.mean,\n dtype=img.dtype)\n left = int(random.uniform(0, w * ratio - w))\n top = int(random.uniform(0, h * ratio - h))\n expand_img[top:top + h, left:left + w] = img\n\n results['img'] = expand_img\n # expand bboxes\n for key in results.get('bbox_fields', []):\n results[key] = results[key] + np.tile(\n (left, top), 2).astype(results[key].dtype)\n\n # expand masks\n for key in results.get('mask_fields', []):\n results[key] = results[key].expand(\n int(h * ratio), int(w * ratio), top, left)\n\n # expand segs\n for key in results.get('seg_fields', []):\n gt_seg = results[key]\n expand_gt_seg = np.full((int(h * ratio), int(w * ratio)),\n self.seg_ignore_label,\n dtype=gt_seg.dtype)\n expand_gt_seg[top:top + h, left:left + w] = gt_seg\n results[key] = expand_gt_seg\n return results","def compute(\n self,\n ) -> Dict[str, Tensor]:\n results = torch.stack((self.tp, self.fp, self.tn, self.fn), dim=1)\n\n return {f\"group_{i}\": group / group.sum() for i, group in enumerate(results)}","def result_dict_for_single_example(image, orig_shape,\n key,\n detections,\n groundtruth):\n label_id_offset = 1 # Applying label id offset (b/63711816)\n\n input_data_fields = fields.InputDataFields\n output_dict = {\n input_data_fields.original_image: image,\n input_data_fields.key: key,\n }\n\n detection_fields = fields.DetectionResultFields\n\n detection_boxes = detections[detection_fields.detection_boxes]\n detection_scores = detections[detection_fields.detection_scores]\n\n detection_classes = (\n detections[detection_fields.detection_classes] +\n label_id_offset)\n\n num_detections = int(detections[detection_fields.num_detections])\n detection_boxes = detection_boxes[:num_detections]\n detection_scores = detection_scores[:num_detections]\n detection_classes = detection_classes[:num_detections]\n detection_boxes = toAbsolute(detection_boxes, orig_shape)\n\n output_dict[detection_fields.detection_boxes] = detection_boxes\n output_dict[detection_fields.detection_classes] = detection_classes\n output_dict[detection_fields.detection_scores] = detection_scores\n \n if groundtruth:\n output_dict.update(groundtruth)\n \n return output_dict","def find(image):\n keypoint, description = describe(image)\n # load keypoints, descriptions from mongodb\n\n bf = cv2.BFMatcher(cv2.NORM_HAMMING, crossCheck=True)\n\n best_match_size = float(\"inf\")\n best_match_index = -1\n match_index = 0\n best_matches = 0\n\n for desc in descriptions:\n matches = bf.match(desc,description)\n matches = sorted(matches, key = lambda x:x.distance)\n if len(matches) > 0:\n match_size = sum(x.distance for x in matches[:10])\n\n print \"match size is \", match_size\n if match_size < best_match_size:\n best_match_size = match_size\n best_match_index = match_index\n best_matches = matches\n\n match_index += 1\n\n needle_color = cv2.imread('needle-stripped.png')[:,:,::-1] # needle\n best_match_image = cv2.imread(\"haystack/\"+files[best_match_index])\n print \"best match is \", files[best_match_index]\n\n # Draw first 10 matches.\n outImg = cv2.imread(\"output/outImg.png\")\n match = cv2.drawMatches(needle_color,keypoint,best_match_image[:,:,::-1],keypoints[best_match_index],best_matches[-20:],outImg, flags=6)\n\n plt.imshow(match),plt.show()\n return","def getResults():","def get_pet_labels(image_dir):\n results_dic = dict()\n \n# # Retrieves the file names from the folder specified as 'image_dir' \n filenames_list = listdir(image_dir)\n \n# # Processes the filenames to create the pet image labels\n# # Retrieves the filenames from folder pet_images/\n for i in range (0, len(filenames_list), 1):\n# # Skips file if starts with . (like .DS_Store of Mac OSX) because it \n# # isn't an pet image file\n if filenames_list[i][0] != \".\":\n# # Reads respectively indexed element from filenames_list into temporary string variable 'pet_image' \n pet_image = filenames_list[i]\n# # Sets all characters in 'pet_image' to lower case \n pet_image_lower = pet_image.lower()\n# # Creates list called 'pet_image_word_list' that contains every element in pet_image_lower seperated by '_'\n pet_image_word_list = pet_image_lower.split(\"_\")\n# # Creates temporary variable 'pet_label' to hold pet label name extracted starting as empty string\n pet_image_alpha = \"\"\n# # Iterates through every word in 'pet_image_word_list' and appends word to 'pet_label_alpha' only if word consists \n# # purely of alphabetic characters \n for word in pet_image_word_list:\n if word.isalpha():\n pet_image_alpha += word + \" \"\n# # Removes possible leading or trailing whitespace characters from 'pet_pet_image_alpha' and add stores final label as 'pet_label' \n pet_label = pet_image_alpha.strip()\n\n# # Adds the original filename as 'key' and the created pet_label as 'value' to the 'results_dic' dictionary if 'key' does \n# # not yet exist in 'results_dic', otherwise print Warning message \n if filenames_list[i] not in results_dic:\n results_dic[filenames_list[i]] = [pet_label]\n else:\n print(\"** Warning: Key = \", filenames_list[i], \" already in 'results_dic' with value = \", results_dic[filenames_list[i]])\n \n# # Iterates through the 'results_dic' dictionary and prints its keys and their associated values\n print(\"\\nPrinting: All 'key' - 'value' pairs in dictionary results_dic: \")\n for key in results_dic:\n print(\"Filename = \", key, \" Pet Label = \", results_dic[key])\n \n# # Returns results_dic\n return results_dic","def create_observations_dict(instances, unprocessed_map_data):\n # print(instances)\n results = {}\n for idx in instances:\n results[idx] = {}\n if idx == instances[0]:\n tag_filter_pixel_corners = B.compute_corner_pixels(\n idx, unprocessed_map_data\n ).tolist()\n # prettified_corner_pixels = [tag_filter_pixel_corners[:2,0],tag_filter_pixel_corners[:2,1],tag_filter_pixel_corners[:2,2],tag_filter_pixel_corners[:2,3]]\n # results[idx][\"corner_pixels\"] = [pixel_pair.tolist() for pixel_pair in prettified_corner_pixels]\n results[idx][\"corner_pixels\"] = tag_filter_pixel_corners\n results[idx][\"tag_pose\"] = B.compute_tag_pose(\n idx, unprocessed_map_data\n ).tolist()\n results[idx][\"camera_pose\"] = B.compute_camera_pose(\n idx, unprocessed_map_data\n ).tolist()\n\n return results","def cs4243_histmatch(ori_image, refer_image):\n \n ##your code here ###\n\n # get cdf of ori and ref image\n grey_level = 256\n ori_hist, ori_cum_hist, ori_res_image, ori_uni_hist = cs4243_histequ(ori_image, grey_level)\n ref_hist, ref_cum_hist, ref_res_image, ref_uni_hist = cs4243_histequ(refer_image, grey_level)\n \n # map each ori cdf to ref cdf and get the mapped index as matched grey level\n map_value = []\n for i in range(grey_level):\n ori_cdf = ori_cum_hist[i]\n matched_intensity = np.uint8(np.abs(ref_cum_hist - ori_cdf).argmin())\n map_value.append(matched_intensity)\n ##\n\n # Set the intensity of the pixel in the raw image to its corresponding new intensity \n height, width = ori_image.shape\n res_image = np.zeros(ori_image.shape, dtype='uint8') # Note the type of elements\n for i in range(height):\n for j in range(width):\n res_image[i,j] = map_value[ori_image[i,j]]\n \n res_hist = np.bincount(res_image.flatten(), minlength=256)\n \n return ori_hist, ref_hist, res_image, res_hist","def get_entry_dict(self):\n\n # generating thumbnail URLs is slow, so only generate the ones\n # that will definitely be used.\n ret = {\n 'id': self.id,\n 'vertices': self.vertices,\n 'triangles': self.triangles,\n 'segments': self.segments,\n 'photo': self.photo.get_entry_dict(),\n }\n if self.dominant_rgb0:\n ret['dominant_rgb0'] = self.dominant_rgb0\n #if self.image_pbox:\n #ret['pbox'] = self.pbox\n #ret['image_pbox'] = {\n #'300': self.image_pbox_300.url,\n #'512': self.image_pbox_512.url,\n #'1024': self.image_pbox_1024.url,\n #'orig': self.image_pbox.url,\n #}\n if self.image_bbox:\n ret['image_bbox'] = {\n #'512': self.image_bbox_512.url,\n '1024': self.image_bbox_1024.url,\n #'orig': self.image_bbox.url,\n }\n return ret","def get_results(self):\n d = {}\n# r = {}\n for analyser in self.xml_tree.getroot():\n for child in analyser:\n if child.tag == 'all-records':\n for record in child:\n attributes = record.attrib\n sample = attributes['sampleId']\n assay_id = attributes['assayId']\n genotype = attributes['genotypeId']\n quality = attributes['description'].split('.')[0]\n if re.match(r'rs\\d+', assay_id):\n if sample in d:\n if assay_id in d[sample]:\n for allele in list(genotype):\n if allele not in d[sample][assay_id]['genotype']:\n d[sample][assay_id]['genotype'] += allele\n if quality not in d[sample][assay_id]['quality']:\n d[sample][assay_id]['quality'].append(quality)\n else:\n d[sample][assay_id] = {'genotype': genotype, 'quality': [quality]}\n else:\n d[sample] = {assay_id: {'genotype': genotype, 'quality': [quality]}}\n# if sample in r:\n# if assay_id in r[sample]:\n# for allele in list(genotype):\n# if allele not in r[sample][assay_id]:\n# r[sample][assay_id] += allele\n# else:\n# r[sample][assay_id] = genotype\n# else:\n# r[sample] = {assay_id: genotype}\n# for k, v in r.items():\n# for k1, v1, in v.items():\n# if len(v1) == 1:\n# v[k1] += v1\n# pprint.pprint(r)\n# df = pd.DataFrame.from_dict(r).transpose()\n# print(df)\n# df.to_excel('snpcheck.xlsx')\n return d","def get_image_list(im, dic):\n lst = list(im.getdata())\n tiles = []\n for i in range(len(lst)):\n #print find_similar(lst[i], dic)[random.randrange(10)][1]\n tiles.append(find_similar(lst[i], dic)[random.randrange(10)][1])\n return tiles","def _load_images(self, resolutions=None):\n images = {}\n\n for block, url in self.image_declarations:\n file_name = normalize_filename(url)\n if file_name not in images:\n img_resolutions = {}\n img = Image.open(file_name)\n img_resolutions[1] = img\n width, height = img.size\n\n if resolutions:\n for resolution in resolutions:\n # Get the correct filename for this resolution\n if resolution != 1:\n root, ext = os.path.splitext(file_name)\n res_file_name = '{root}-{resolution}x{ext}'.format(\n root=root, resolution=resolution, ext=ext)\n\n img = Image.open(res_file_name)\n if img.size[0] / resolution != width:\n raise ValueError('Invalid width for {0}'.format(\n res_file_name))\n if img.size[1] / resolution != height:\n raise ValueError('Invalid height for {0}'.format(\n res_file_name))\n img_resolutions[resolution] = img\n\n images[file_name] = img_resolutions\n\n return images","def get_images_to_build(fuzzers, benchmarks):\n images = {}\n templates = _get_image_type_templates()\n for fuzzer in fuzzers:\n for benchmark in benchmarks:\n for name_templ, obj_templ in templates.items():\n name, obj = _instantiate_image_obj(name_templ, obj_templ,\n fuzzer, benchmark)\n images[name] = obj\n return images","def fetch_images(self, images, depth_level):\r\n image_results = {}\r\n initial_area = float(0.0)\r\n total_score = float(0.0)\r\n cnt = float(1.0)\r\n MIN_WIDTH = 50\r\n for image in images[:30]:\r\n src = self.parser.getAttribute(image, attr='src')\r\n src = self.build_image_path(src)\r\n local_image = self.get_local_image(src)\r\n width = local_image.width\r\n height = local_image.height\r\n src = local_image.src\r\n file_extension = local_image.file_extension\r\n\r\n if file_extension != '.gif' or file_extension != 'NA':\r\n if (depth_level >= 1 and local_image.width > 300) or depth_level < 1:\r\n if not self.is_banner_dimensions(width, height):\r\n if width > MIN_WIDTH:\r\n sequence_score = float(1.0 / cnt)\r\n area = float(width * height)\r\n total_score = float(0.0)\r\n\r\n if initial_area == 0:\r\n initial_area = area * float(1.48)\r\n total_score = 1\r\n else:\r\n area_difference = float(area / initial_area)\r\n total_score = sequence_score * area_difference\r\n\r\n image_results.update({local_image: total_score})\r\n cnt += 1\r\n cnt += 1\r\n return image_results","def getPatternDict(self):\n c = \"/cli:python /app:matrix /cmd:getinfo /dev:patternlist\"\n self.sendCMDstring(c)\n time.sleep(self.delay)\n answers = self.readandparseCAM()\n patternlist = {}\n for a in answers:\n if a['dev']=='patternlist':\n for i in range(int(a['count'])):\n nr = a['patternid' +str(i+1)]\n name = a['patternname' +str(i+1)].lower()\n patternlist[name]=nr\n else:\n print \"no patternlist in answers\"\n return patternlist","def init_output_dict(self):\n return {\n \"outputs\": torch.FloatTensor(),\n \"pred_probs\": torch.FloatTensor(),\n \"labels\": torch.LongTensor(),\n }","def _extract_results(self) -> None:\n metric_name = self.metric.name\n for inference_name in ['train', 'test', 'opt']:\n # TODO: Extract information from self.search_results\n data = getattr(self.search_results, f'{inference_name}_metric_dict')[metric_name]\n if all([d is None for d in data]):\n if inference_name not in OPTIONAL_INFERENCE_CHOICES:\n raise ValueError(f\"Expected {metric_name} score for {inference_name} set\"\n f\" to not be None, but got {data}\")\n else:\n continue\n self.data[f'single::{inference_name}::{metric_name}'] = np.array(data)\n\n if self.ensemble_results.empty() or inference_name == 'opt':\n continue\n\n data = getattr(self.ensemble_results, f'{inference_name}_scores')\n if all([d is None for d in data]):\n if inference_name not in OPTIONAL_INFERENCE_CHOICES:\n raise ValueError(f\"Expected {metric_name} score for {inference_name} set\"\n f\" to not be None, but got {data}\")\n else:\n continue\n self.data[f'ensemble::{inference_name}::{metric_name}'] = np.array(data)","def compute_metrics(self, results: list) -> Dict[str, float]:\n logger: MMLogger = MMLogger.get_current_instance()\n\n # pred_coords: [N, K, D]\n pred_coords = np.concatenate(\n [result['pred_coords'] for result in results])\n if pred_coords.ndim == 4 and pred_coords.shape[1] == 1:\n pred_coords = np.squeeze(pred_coords, axis=1)\n # gt_coords: [N, K, D]\n gt_coords = np.stack([result['gt_coords'] for result in results])\n # mask: [N, K]\n mask = np.concatenate([result['mask'] for result in results])\n # action_category_indices: Dict[List[int]]\n action_category_indices = defaultdict(list)\n for idx, result in enumerate(results):\n action_category = result['action'].split('_')[0]\n action_category_indices[action_category].append(idx)\n\n error_name = self.mode.upper()\n\n logger.info(f'Evaluating {self.mode.upper()}...')\n metrics = dict()\n\n metrics[error_name] = keypoint_mpjpe(pred_coords, gt_coords, mask,\n self.ALIGNMENT[self.mode])\n\n for action_category, indices in action_category_indices.items():\n metrics[f'{error_name}_{action_category}'] = keypoint_mpjpe(\n pred_coords[indices], gt_coords[indices], mask[indices])\n\n return metrics","def match_patterns(training_data, num_languages):\n\n ret = {}\n\n for index1, pattern in enumerate(shaman.PatternMatcher.PATTERNS):\n print('Matching pattern %d \"%s\"' % (index1+1, pattern))\n\n matcher = shaman.PatternMatcher(pattern)\n ratios_per_lang = [[] for _ in range(num_languages)]\n\n for index2, (lang_id, code) in enumerate(training_data):\n if len(code) <= 20 or len(code) > 100000:\n continue\n\n ratio = matcher.getratio(code)\n ratios_per_lang[lang_id].append(ratio)\n\n print('Matching patterns %d/%d ' %\n (index2, len(training_data)), end='\\r')\n\n ret[pattern] = [[] for _ in range(num_languages)]\n for lang_id, ratio_list in enumerate(ratios_per_lang):\n ret[pattern][lang_id] = sum(ratio_list) / max(len(ratio_list), 1)\n\n print('Matching patterns completed ')\n return ret","def buildRunDict(self):\n self.showProgressBar()\n ori_images = 0\n if self.img_exist:\n ori_images = len(listDirectory(self.savePathJoin(\"Images\")))\n self.buildRunDictMain(ori_images)\n else:\n self.run_dict[\"Video\"] = {\n \"Run\": True,\n \"Progress\": ori_images,\n \"Text\": \"Preparing video\",\n }\n self.buildParamsDict()\n self.params_dict[\"send_video_frame\"] = True\n\n self.progressLabel.setText(\"Create images from video\")\n\n self.worker = calcRunner.CalculationRunner(self.params_dict) # no parent!\n self.thread = QThread() # no parent!\n\n self.worker.labelUpdate.connect(self.labelUpdate)\n\n self.worker.update.connect(self.progressUpdate)\n self.worker.videoFrame.connect(self.setVidFrame)\n\n self.worker.moveToThread(self.thread)\n self.thread.started.connect(self.worker.startThread)\n self.thread.start()","def calculate_MAP(self):\n testing_images = open('./digitdata/testimages', 'r')\n with testing_images as ti:\n data = list(csv.reader(ti))\n data = [i for i in data if i]\n count = 0\n #loop through all the test images\n for j in range(0,1000):\n classification_dict = {0:0,1:0,2:0,3:0,4:0,5:0,6:0,7:0,8:0,9:0} \n for l in range(0,28):\n coord = count + l\n for w in range(0,28):\n if data[coord][0][w] == \"+\":\n #iterate through each class. z is the class [0-9]\n for z in range(0,10):\n classification_dict[z] += math.log(self.class_probabilities[z][l][w][0]) \n elif data[coord][0][w] == \"#\":\n for z in range(0,10):\n classification_dict[z] += math.log(self.class_probabilities[z][l][w][1])\n elif data[coord][0][w] == \" \":\n for z in range(0,10):\n classification_dict[z] += math.log(self.class_probabilities[z][l][w][2])\n count += 28\n self.solutions.append(max(classification_dict, key=classification_dict.get))","def create_results_dict(self, n_options):\n if n_options == 1:\n self.results_dict[0] = {\"configuration\": [\"user\"]}\n for name in self.results_metadata[\"names\"]:\n self.results_dict[0][name] = []\n else:\n self.results_dict = {}\n for i in range(n_options):\n self.results_dict[i] = {\"configuration\": []}\n for name in self.results_metadata[\"names\"]:\n self.results_dict[i][name] = []","def get_specs(self, results: Dict[str, Dict], params: Dict) -> Dict[str, float] :\n raise NotImplementedError","def predict() -> Any:\n threshold = request.form.get(\"threshold\", type=float)\n source_size = request.form.get(\"source_size\", type=bool)\n images = request.files.getlist(\"images\")\n result = {}\n for image in images:\n input_image = prepare_input(image)\n if input_image is not None:\n output_image = model.predict(input_image, threshold, source_size)\n if output_image is not None:\n result[image.filename] = prepare_output(output_image)\n else:\n result[image.filename] = None\n else:\n result[image.filename] = None\n return result","def extractData(self) -> Dict[str, str]:\n if self.hasScore():\n result = {}\n result[\"date\"] = self.getMatchDate()\n result[\"score\"] = self.getMatchScore()\n for team in self.TEAMS:\n result[\"team_\" + team] = self.getSquadName(team)\n\n return result\n return None","def buildDicts(resComp, comparisons, matrices, atom1, res1, atom2, res2, r, FLAG=False):\r\n if resComp not in matrices:\r\n for pairs in comparisons:\r\n if resComp[0] == pairs[1] and resComp[1] == pairs[0]:\r\n return\r\n matrices[resComp] = [r]\r\n comparisons[resComp] = [(atom1, res1, atom2, res2, r)]\r\n FLAG = True\r\n\r\n elif type(comparisons[resComp][-1]) == list:\r\n if comparisons[resComp][-1][-1][1] != res1 or comparisons[resComp][-1][-1][3] != res2:\r\n # Make new matrix\r\n raise Warning\r\n elif comparisons[resComp][-1][-1][0] == atom1:\r\n comparisons[resComp][-1].append((atom1, res1, atom2, res2, r))\r\n matrices[resComp][-1].append(r)\r\n else:\r\n comparisons[resComp].append([(atom1, res1, atom2, res2, r)])\r\n matrices[resComp].append([r])\r\n\r\n elif type(comparisons[resComp][-1]) == tuple:\r\n if comparisons[resComp][-1][0] == atom1:\r\n comparisons[resComp].append((atom1, res1, atom2, res2, r))\r\n matrices[resComp].append(r)\r\n else:\r\n comparisons[resComp] = [comparisons[resComp], [(atom1, res1, atom2, res2, r)]]\r\n matrices[resComp] = [matrices[resComp], [r]]\r\n\r\n else:\r\n print(\"Error: Dictionary should contain Match objects / Lists of Match objects / List of lists of Match objects\")\r\n\r\n return comparisons, matrices, resComp, FLAG","def _get_output_dictionary(self):\n\n return_dictionary = {}\n\n for output_path in self.provided_outputs:\n return_dictionary[output_path.full_path] = self.get_value(output_path)\n\n return return_dictionary","def testing_on_new_data(directory=parentDir + '/photos-to-categorize'):\n\n known_face_encodings={}\n with open(parentDir + '/data/face_locations.csv', 'r') as csvfile:\n fieldnames=['person', 'fullfilename' , 'top', 'right', 'bottom', 'left']\n reader = csv.DictReader(csvfile,fieldnames=fieldnames)\n for row in reader:\n face_location = [int(row['top']), int(row['right']), int(row['bottom']), int(row['left'])]\n if row['person'] in known_face_encodings:\n known_face_encodings[row['person']] = np.append(known_face_encodings[row['person']],np.array(face_encodings(load_image_file(row['fullfilename']),[face_location])),axis=0)\n else:\n known_face_encodings.update({row['person']:np.array(face_encodings(load_image_file(row['fullfilename']),[face_location]))})\n \n for filename in os.listdir(directory):\n img = load_image_file(directory + '/' + filename)\n \n all_face_locations = face_locations(img)\n for face_location in all_face_locations:\n top,right,bottom,left = face_location\n face_image = img[top:bottom, left:right]\n pil_image = Image.fromarray(face_image)\n \n test_face_encode = face_encodings(img,[face_location])\n result={}\n name='Unknown'\n minval=1\n for person in known_face_encodings:\n r=face_distance(known_face_encodings[person], test_face_encode[0]).mean()\n if r <= 0.6:\n result.update({person:(1-r)*100})\n if r < minval:\n minval=r\n name=person\n \n Image._show(pil_image)\n \n print filename + \": The person is \" + name\n print \"All possiblities: \" + str(result)\n\n #writting in a file\n with open(parentDir + '/data/image_categorizations.csv', 'a') as csvfile:\n fieldnames=['person', 'fullfilename' , 'top', 'right', 'bottom', 'left']\n writer = csv.DictWriter(csvfile, fieldnames=fieldnames)\n if name != 'Unknown': writer.writerow({'person':name, 'fullfilename': directory + '/' + filename, 'top':top, 'right':right, 'bottom':bottom, 'left':left})\n \n\n #raw_input(\"continue?\")","def _add_matches(self):\r\n for record in self.records:\r\n match_dict={key_to_track: record.get(key_to_track)\r\n for key_to_track in self.key_matcher.keys()}\r\n self.key_matcher.add(obj=record,\r\n match_dict=match_dict)","def get_imgid_dict(ann):\n return {item[1][\"file_name\"]: item[0] for item in ann.imgs.items()}","def _results(self):\n results = {}\n results[\"coeff\"] = self._coeff_to_dict()\n results[\"coeff_path\"] = dict(\n zip(\n [f\"{col}\" for col in self.X_train_.columns.tolist()],\n (\n self.model_.coef_path_.reshape(-1, self.model_.coef_path_.shape[-1])\n ).tolist(),\n )\n )\n results[\"cv_standard_error\"] = self.model_.cv_standard_error_.tolist()\n results[\"cv_mean_score\"] = self.model_.cv_mean_score_.tolist()\n results[\"lambda_path\"] = self.model_.lambda_path_.tolist()\n results[\"lambda_best\"] = self.model_.lambda_best_[0]\n results[\"lambda_max\"] = self.model_.lambda_max_\n results[\"n_lambda\"] = self.model_.n_lambda_\n results[\"intercept\"] = self.model_.intercept_\n results[\"intercept_path\"] = self.model_.intercept_path_.tolist()[0]\n results[\"params\"] = self.model_.get_params()\n results[\"module\"] = self.model_.__module__\n\n return results","def calculate_individual_image_result(\n prediction_boxes, gt_boxes, iou_threshold):\n # Find the bounding box matches with the highes IoU threshold\n matched_prediction_boxes, matched_gt_boxes = get_all_box_matches(\n prediction_boxes,\n gt_boxes,\n iou_threshold)\n \n # Compute true positives, false positives, false negatives\n num_tp = len(matched_prediction_boxes)\n num_fp = len(prediction_boxes) - num_tp\n num_fn = np.maximum(len(gt_boxes) - len(matched_gt_boxes), 0, dtype=int)\n\n return {\"true_pos\": num_tp, \"false_pos\": num_fp, \"false_neg\": num_fn}","def __make_processing(self, img_name, abspath_dir_img, id_foot):\n data = {}\n data['data'] = ImageInfo.get_date(abspath_dir_img)\n data['total_part'] = TOTAL_PART\n data['nuvens'] = ImageInfo.get_cloud(abspath_dir_img)\n self.__make_tms(abspath_dir_img)\n data['geom'] = self.__make_footprint(abspath_dir_img, shp_out=id_foot)\n abspath_rgb, img_name_rgb = ImageInfo.get_image_rgb(\n abspath_dir_img, img_name\n )\n data['tms'] = ImageInfo.get_xml_tms(img_name_rgb)\n data['image'] = img_name_rgb\n data['quicklook'] = self.__make_png(abspath_rgb)\n data['path'] = ImageInfo.get_path(img_name)\n return data","def _build_elbv2_mapping_from_resources(resource_to_analyse, result_dict, session):\n for elb_instance, security_group_id, security_group_name in _generate_elb_instances_and_sg(resource_to_analyse, session):\n resource_dict = _check_if_in_list(result_dict, elb_instance[\"LoadBalancerName\"], \"resource_id\")\n if resource_dict is not None:\n resource_dict[\"sg_attached\"].append({\n \"sg_id\": security_group_id,\n \"sg_name\": security_group_name\n })\n else:\n result_dict.append({\n \"resource_id\": elb_instance[\"LoadBalancerName\"],\n \"resource_type\": \"elb\",\n \"sg_attached\": [{\n \"sg_id\": security_group_id,\n \"sg_name\": security_group_name\n }]\n })\n return result_dict","def process():\n config = read_config()\n \n\n img_dir = config['DEFAULT']['images_directory']\n results_dict = {}\n images = list(get_image_files(img_dir))\n for image in tqdm.tqdm(images):\n info = hash_file(image)\n if info == 0:\n continue\n\n hash_value = info['hash']\n\n if hash_value not in results_dict:\n file_name = os.path.basename(info['_id'])\n results_dict[hash_value] = [file_name, 1]\n else:\n results_dict[hash_value][1] += 1\n\n count = list(results_dict.values())\n sorted_count = sorted(count, key=lambda x: x[1], reverse=True)\n \n with ImagesDB(IMG_INFO_DB_FILENAME) as imgDb: \n imgDb.insert_batch(sorted_count)","def iterate_results(results, extract_fn):\n outputs = {}\n for environment, environment_results in results.items():\n if environment not in outputs:\n outputs[environment] = {}\n for experimental_setting, setting_results in environment_results.items():\n outputs[environment][experimental_setting] = []\n for config, seeds_results in setting_results.items():\n for seed, actual_results in seeds_results.items():\n output = extract_fn(actual_results)\n outputs[environment][experimental_setting].append(output)\n outputs[environment][experimental_setting] = np.array(outputs[environment][experimental_setting])\n return outputs","def painting_matching_ml(imgs, db_imgs, method_list, text_masks, author_text, gt_text, metrics, weights, splits=30, max_rank=10):\n descriptor_extractors = [get_descriptor_extractor(method_name) for method_name in method_list]\n tmp_img_format = []\n tmp_mask_format = []\n tmp_text_format = []\n for i in range(len(imgs)):\n for j in range(len(imgs[i])):\n tmp_img_format.append(imgs[i][j])\n tmp_mask_format.append(text_masks[i][j])\n tmp_text_format.append(author_text[i][j])\n\n #db_imgs = [img[0] for img in db_imgs]\n db_img_splits = [i*len(db_imgs)//splits for i in range(splits-1)]\n \n scores = []\n query_descriptors = extract_descriptors(tmp_img_format, descriptor_extractors, method_list, tmp_text_format, tmp_mask_format) \n #np.array([extract_descriptors(img, matching_methods, mask) for img, mask in zip(tmp_img_format, tmp_mask_format)])\n print(\"Starting db extraction + matching\")\n for split in tqdm(range(splits-2)):\n db_descriptors = extract_descriptors(db_imgs[db_img_splits[split]:db_img_splits[split+1]], descriptor_extractors, method_list, gt_text[db_img_splits[split]:db_img_splits[split+1]], None) #np.array([mrhm(db_img) for db_img in db_imgs[db_img_splits[split]:db_img_splits[split+1]]])\n scores.append(compare_descriptors(query_descriptors, db_descriptors, metrics, method_list, weights))\n # compare_descriptors(query_descriptors, db_descriptors, descriptor_comp_methods, descriptor_names, weights)\n db_descriptors = extract_descriptors(db_imgs[db_img_splits[-1]:], descriptor_extractors, method_list, gt_text[db_img_splits[-1]:], None)\n scores.append(compare_descriptors(query_descriptors, db_descriptors, metrics, method_list, weights))\n \n # concatenate all the results\n scores = np.concatenate(scores, 1)\n \n top_k_matches = np.argpartition(scores, list(range(max_rank)))[:, :max_rank]\n return top_k_matches","def copy_to_dict(self):\n search_hit_dict = {}\n search_hit_dict['Offset'] = self.offset\n search_hit_dict['Filename (inode)'] = self.filename\n search_hit_dict['String'] = self.data\n\n return search_hit_dict","def pre_pipeline(self, results):\n results[\"img_prefix\"] = self.img_prefix\n results[\"seg_prefix\"] = self.seg_prefix\n results[\"proposal_file\"] = self.proposal_file\n results[\"bbox_fields\"] = []\n results[\"mask_fields\"] = []\n results[\"seg_fields\"] = []\n results[\"site_fields\"] = []\n results[\"label_fields\"] = []","def get_resource_results():\n resource_results = {}\n resource_results['checks'] = []\n resource_results['green'] = 0\n resource_results['red'] = 0\n resource_results['orange'] = 0\n resource_results['blue'] = 0\n resource_results['failed_accounts'] = 0\n resource_results['total_accounts'] = 0\n resource_results['total_checks'] = 0\n\n # Defaults for when no data is reported, working towards having modules be\n # modular / optional\n resource_results['blue_percent'] = 100\n resource_results['red_percent'] = 0\n resource_results['orange_percent'] = 0\n resource_results['green_percent'] = 0\n resource_results['working_percentage'] = 100\n\n # Check how many accounts failed from each module and add them to the\n # total failed accounts. If the data from a module is considered stale\n # then all of it's accounts will be considered failed.\n milliseconds_since_epoch = time.time() * 1000\n for module in get_all_data('resources_success:*'):\n module_success_json = get_data(module)\n module_success = json.loads(module_success_json)[0]\n resource_results['total_accounts'] += module_success['total_accounts']\n resource_results['total_checks'] += module_success['total_checks']\n milliseconds_since_epoch_module_data_is_valid_until = module_success['valid_until']\n if milliseconds_since_epoch > milliseconds_since_epoch_module_data_is_valid_until:\n resource_results['failed_accounts'] += module_success['total_accounts']\n logger.error('Data for {} is stale, please check the daemon is functioning properly'.format(module))\n else:\n resource_results['failed_accounts'] += module_success['failed_accounts']\n\n # We will count checks in so we can compare it against the number of checks\n # reported by the daemon\n checks_found = 0\n # Get list of keys in the format resources:module#uuid\n for host in get_all_data('resources:*'):\n try:\n # Storing lists with only one value since when I convert\n # dictionaries to json and store them in redis they come back as\n # strings, I am working around this by storing lists,\n # ast.literal_eval also works\n host_data = json.loads(get_data(host))[0]\n resource_results['checks'].append(host_data)\n # get the health status colour of the current check, and then add\n # one to the number of checks with that health status\n resource_results[host_data['health_status']] += 1\n checks_found += 1\n except Exception as e:\n # I would rather log to uwsgi's log but I'll sort this out later\n logger.error('Data for {} is not in a valid format: {}'.format(host, e))\n\n # If we are getting back old checks that are no-longer reporting hence\n # are not in the total_checks variable then they have failed.\n # If we are getting back less checks than we stored then something has\n # gone really wrong or we caught the weekly cron that clears the keys.\n if resource_results['total_checks'] != checks_found:\n logger.info('The number of checks stored in the database doesn\\'t '\\\n 'match the number reported by the daemon, it is likely some '\\\n 'servers are no-longer reporting, run '\\\n 'resources_list_unreporting_servers.py to look into this.')\n\n # The number of checks we are outputing is authoritive over the number\n # we expected to be there, at the moment we are just logging the fact they\n # were different, it would be nice to have a visual display or send an\n # email but there isn't a correct place to do this at the moment\n resource_results['total_checks'] = checks_found\n\n total_results = resource_results['green'] + resource_results['red'] + resource_results['orange'] + resource_results['blue']\n if total_results != 0:\n resource_results['red_percent'] = ( resource_results['red'] / total_results ) * 100\n resource_results['orange_percent'] = ( resource_results['orange'] / total_results ) * 100\n resource_results['blue_percent'] = ( resource_results['blue'] / total_results ) * 100\n # I want the percentage to always be 100 and green seems the most\n # disposable / least affected by any rounding issues\n resource_results['green_percent'] = 100 - ( resource_results['red_percent'] + resource_results['orange_percent'] + resource_results['blue_percent'] )\n\n resource_results['working_percentage'] = 100 - (( resource_results['failed_accounts'] / resource_results['total_accounts'] ) * 100 )\n resource_results['working_accounts'] = resource_results['total_accounts'] - resource_results['failed_accounts']\n logger.debug('working_percentage: {}'.format(resource_results['working_percentage']))\n return resource_results","def get_results(self):\n result = [round(self.mr / self.test_size, 1), round(self.mrr / self.test_size, 3),\n round(self.hits1 / self.test_size, 3), round(self.hits3 / self.test_size, 3),\n round(self.hits5 / self.test_size, 3), round(self.hits10 / self.test_size, 3)]\n return result","def __call__(\n self, \n image: ndarray, \n adversarial_image: ndarray\n ) -> Dict[str, Union[float, int]]:\n ...","def __getitem__(self, idx: int) -> Dict:\n files = self.files[idx]\n build_mask = np.array(Image.open(files[\"build_mask\"]))\n demo_mask = np.array(Image.open(files[\"demolish_mask\"]))\n build_mask = np.clip(build_mask.mean(axis=-1), 0, 1).astype(\"uint8\")\n demo_mask = np.clip(demo_mask.mean(axis=-1), 0, 1).astype(\"uint8\")\n image1 = np.array(Image.open(files[\"image1\"]))\n image2 = np.array(Image.open(files[\"image2\"]))\n image1, image2, build_mask, demo_mask = self.transform([image1, image2, build_mask, demo_mask])\n x = torch.stack([image1, image2], dim=0)\n return dict(x=x, build_mask=build_mask, demolish_mask=demo_mask)","def generate(self):\n return {\n 'json_class': 'Pact::ArrayLike',\n 'contents': from_term(self.matcher),\n 'min': self.minimum}","def process_matches(matches: List[EndpointMatch]) -> List[Result] :\n out=[]\n #sort data\n sort_f = lambda match: match.endpoint_1.name\n sorted_matches = sorted(matches, key = sort_f)\n #group by endpoint 1\n for key, group in itertools.groupby(sorted_matches, key = sort_f ):\n #sort by score\n endpoint_matches = sorted(list(group), key = lambda match: match.score,reverse=True)\n #best match\n best_match = endpoint_matches[0]\n #others\n other_matches = \";\".join([f\"{a.endpoint_2.name}|{a.score:.3g}\" for a in endpoint_matches[1:] ])\n\n out.append(Result(\n best_match.endpoint_1.name,\n best_match.endpoint_2.name,\n best_match.score,\n \";\".join(best_match.endpoint_1.matches),\n \";\".join(best_match.endpoint_2.matches),\n best_match.endpoint_1.regex,\n best_match.endpoint_2.regex,\n other_matches\n ))\n return out","def post(self):\n names = self.request.get('names').split(',')\n results = {}\n for name in names:\n key = 'eol-images-%s' % name\n value = memcache.get(key)\n if value:\n logging.info('memcache value: %s' % value)\n else:\n value = cache.get(key, loads=True)\n if not value:\n name = urllib.quote(name)\n search_url = 'http://eol.org/api/search/%s.json?exact=1' % name\n result = json.loads(urlfetch.fetch(search_url, deadline=60).content)\n page_id = result['results'][0]['id']\n page_url = 'http://eol.org/api/pages/1.0/%s.json' % page_id\n logging.info(page_url)\n result = json.loads(urlfetch.fetch(page_url, deadline=60).content)\n object_id = None\n for x in result['dataObjects']:\n if x['dataType'].endswith('StillImage'):\n object_id = x['identifier']\n if object_id:\n object_url = 'http://eol.org/api/data_objects/1.0/%s.json' % object_id\n value = json.loads(urlfetch.fetch(object_url, deadline=60).content)\n cache.add(key, value, dumps=True)\n memcache.add(key, value) \n results[name] = value \n self.response.headers[\"Content-Type\"] = \"application/json\" \n self.response.out.write(json.dumps(results))","def compute_metrics(self, results: Sequence[Dict]) -> Dict:\n\n preds = []\n gts = []\n for result in results:\n preds.append(result['pred_labels'])\n gts.append(result['gt_labels'])\n preds = torch.cat(preds)\n gts = torch.cat(gts)\n\n assert preds.max() < self.num_classes\n assert gts.max() < self.num_classes\n\n cared_labels = preds.new_tensor(self.cared_labels, dtype=torch.long)\n\n hits = (preds == gts)[None, :]\n preds_per_label = cared_labels[:, None] == preds[None, :]\n gts_per_label = cared_labels[:, None] == gts[None, :]\n\n tp = (hits * preds_per_label).float()\n fp = (~hits * preds_per_label).float()\n fn = (~hits * gts_per_label).float()\n\n result = {}\n if 'macro' in self.mode:\n result['macro_f1'] = self._compute_f1(\n tp.sum(-1), fp.sum(-1), fn.sum(-1))\n if 'micro' in self.mode:\n result['micro_f1'] = self._compute_f1(tp.sum(), fp.sum(), fn.sum())\n\n return result","def checksImages(self):\n metadata=[]\n for image in self.meta['sources']:\n with rasterio.open(image) as src:\n metaData=src.meta\n \n assert metaData['driver'] == 'GTiff', \"Driver is not supported: {0}\".format(metaData['driver'])\n assert metaData['count'] == len(self.meta['bandNames']), \"Nbands incorrect, expected: {0}, {1} provided\".format(metaData['count'],len(self.meta['bandNames']))\n \n metadata.append({'dtype': metaData['dtype'], 'driver': metaData['driver'], 'nodata': metaData['nodata'], 'nBands': metaData['count'],'crs': src.crs.to_string()})\n \n assert len(set([item['dtype'] for item in metadata])) == 1, \"Images list dtypes aren't compatibles. Expected: 1, {1} provided\".format(metaData['count'],len(set([item['dtype'] for item in metadata])))\n assert len(set([item['driver'] for item in metadata])) == 1, \"Images list drivers aren't compatibles. Expected: 1, 1 provided\".format(metaData['count'],len(set([item['driver'] for item in metadata])))\n assert len(set([item['nodata'] for item in metadata])) == 1, \"Images list nodata values aren't compatibles. Expected: 1, {1} provided\".format(metaData['count'],len(set([item['nodata'] for item in metadata])))\n assert len(set([item['nBands'] for item in metadata])) == 1, \"Images list nBands number aren't compatibles. Expected: 1, {1} provided\".format(metaData['count'],len(set([item['nBands'] for item in metadata])))\n assert len(set([item['crs'] for item in metadata])) == 1, \"Images list crs aren't compatibles. Expected: 1, {1} provided\".format(metaData['count'],len(set([item['crs'] for item in metadata]))) \n return metadata[0]"],"string":"[\n \"def get_detect_result(self):\\n\\n resultdict = {'class_index' : self.class_index,\\n 'obj_name' : self.obj_name,\\n 'score' : self.score,\\n 'bounding_box' : {\\n 'x_min' : self.x_min,\\n 'y_min' : self.y_min,\\n 'width' : self.width,\\n 'height' : self.height}\\n }\\n return resultdict\",\n \"def get_results(self):\\r\\n r ={\\r\\n 'Gp_s' :np.array(self.Gp_guess_s), #Means of the posteriors\\r\\n 'Gm_s' :np.array(self.Gm_guess_s), #Means of the posteriors\\r\\n 'eGp_s':np.array(self.eGp_guess_s) , #Standard deviations of the posteriors\\r\\n 'eGm_s':np.array(self.eGm_guess_s) , #Standard deviations of the posteriors\\r\\n 'covGpGm_s':np.array(self.cov_GpGm_s) , #Covariances of the posteriors\\r\\n 't_tot_pulseSequences_s':np.array(self.t_tot_pulseSequences_s), #Time it takes for each batch of pulse sequences\\r\\n 't_tot_process_s':np.array(self.t_tot_process_s), #CPU time it takes for processing the Bayes stuff at each iteration\\r\\n 't_probe_p_s':np.array(self.t_probe_p_s), # Times probed for the diff p\\r\\n 't_probe_m_s':np.array(self.t_probe_m_s), # Times probed for the diff m\\r\\n 't_pulseSequences_s':np.array(self.t_pulseSequences_s),\\r\\n 't_process_s':np.array(self.t_process_s),\\r\\n 'iteration_s':np.array(self.nb_iteration_s), #Number of iterations performed\\r\\n 'R_tot_s':np.array(self.R_tot_s) #Total number of readout performed\\r\\n }\\r\\n return r\",\n \"def match_results(self):\\n return np.array(list(self._match_result_dict.values()))\",\n \"def get_matches(self, file_map) -> dict:\\r\\n get_file_dict = {}\\r\\n match_dict = {\\r\\n 'GIF': re.findall(b'(?s)(\\\\x47\\\\x49\\\\x46\\\\x38\\\\x39\\\\x61.{80})', file_map),\\r\\n 'RTF': re.findall(b'(?s)(.{20}\\\\x35\\\\x30\\\\x34\\\\x65\\\\x34\\\\x37\\\\x30.{80}|.{20}\\\\x66\\\\x66\\\\x64\\\\x38\\\\x66\\\\x66.{80}|'\\r\\n b'.{20}\\\\x66\\\\x66\\\\x64\\\\x38\\\\x66\\\\x66\\\\x65\\\\x30\\\\x30\\\\x30\\\\x31\\\\x30.{80})', file_map),\\r\\n }\\r\\n if self.jpgsos:\\r\\n match_dict['JPG_SOS'] = jpg_sos(file_map)\\r\\n elif self.sof2sos:\\r\\n match_dict['JPG_SOF2SOS'] = jpg_sof2sos(file_map)\\r\\n elif self.jump:\\r\\n match_dict['JPG_JUMP'] = jpg_jump(file_map)\\r\\n else:\\r\\n match_dict['JPG'] = re.findall(b'(?s)(\\\\xff\\\\xd8\\\\xff\\\\xe0\\\\x00\\\\x10.{80})', file_map)\\r\\n match_dict['JPG2'] = re.findall(b'(?s)(\\\\xff\\\\xd8\\\\xff.{80})', file_map)\\r\\n if self.idat:\\r\\n match_dict['PNG_IDAT'] = idat(file_map)\\r\\n else:\\r\\n match_dict['PNG'] = re.findall(b'(?s)(\\\\x89\\\\x50\\\\x4e\\\\x47.{82})', file_map)\\r\\n m = re.match(br'^(?P<magic_beans>\\\\x49\\\\x49\\\\x2a\\\\x00[^\\\\x00\\\\x00]{2}.{80})',file_map,re.S)\\r\\n if m:\\r\\n match_dict['TIF'] = [m.group('magic_beans')]\\r\\n for file_type, regex_match in match_dict.items():\\r\\n if len(regex_match) > 0:\\r\\n get_file_dict[file_type] = regex_match\\r\\n return get_file_dict\",\n \"def calculates_results_stats(results_dic): \\n # Creates empty dictionary for results_stats_dic\\n results_stats_dic = dict()\\n \\n # Sets all counters to initial values of zero so that they can \\n # be incremented while processing through the images in results_dic \\n results_stats_dic['n_dogs_img'] = 0\\n results_stats_dic['n_match'] = 0\\n results_stats_dic['n_correct_dogs'] = 0\\n results_stats_dic['n_correct_notdogs'] = 0\\n results_stats_dic['n_correct_breed'] = 0\\n \\n # process through the results dictionary\\n for key in results_dic:\\n \\n # Labels Match Exactly\\n if results_dic[key][2] == 1:\\n results_stats_dic['n_match'] += 1\\n\\n # TODO: 5a. REPLACE pass with CODE that counts how many pet images of\\n # dogs had their breed correctly classified. This happens \\n # when the pet image label indicates the image is-a-dog AND \\n # the pet image label and the classifier label match. You \\n # will need to write a conditional statement that determines\\n # when the dog breed is correctly classified and then \\n # increments 'n_correct_breed' by 1. Recall 'n_correct_breed' \\n # is a key in the results_stats_dic dictionary with it's value \\n # representing the number of correctly classified dog breeds.\\n # \\n # Pet Image Label is a Dog AND Labels match- counts Correct Breed\\n if results_dic[key][3] == 1 and results_dic[key][2] == 1:\\n results_stats_dic['n_correct_breed'] += 1\\n \\n # Pet Image Label is a Dog - counts number of dog images\\n if results_dic[key][3] == 1:\\n results_stats_dic['n_dogs_img'] += 1\\n \\n # Classifier classifies image as Dog (& pet image is a dog)\\n # counts number of correct dog classifications\\n if results_dic[key][4] == 1:\\n results_stats_dic['n_correct_dogs'] += 1\\n\\n # TODO: 5b. REPLACE pass with CODE that counts how many pet images \\n # that are NOT dogs were correctly classified. This happens \\n # when the pet image label indicates the image is-NOT-a-dog \\n # AND the classifier label indicates the images is-NOT-a-dog.\\n # You will need to write a conditional statement that \\n # determines when the classifier label indicates the image \\n # is-NOT-a-dog and then increments 'n_correct_notdogs' by 1. \\n # Recall the 'else:' above 'pass' already indicates that the \\n # pet image label indicates the image is-NOT-a-dog and \\n # 'n_correct_notdogs' is a key in the results_stats_dic dictionary \\n # with it's value representing the number of correctly \\n # classified NOT-a-dog images.\\n # \\n # Pet Image Label is NOT a Dog\\n else:\\n # Classifier classifies image as NOT a Dog(& pet image isn't a dog)\\n # counts number of correct NOT dog clasifications.\\n if results_dic[key][3] == 0 and results_dic[key][4] == 0:\\n results_stats_dic['n_correct_notdogs'] += 1\\n\\n\\n # Calculates run statistics (counts & percentages) below that are calculated\\n # using the counters from above.\\n\\n # calculates number of total images\\n results_stats_dic['n_images'] = len(results_dic)\\n\\n # calculates number of not-a-dog images using - images & dog images counts\\n results_stats_dic['n_notdogs_img'] = (results_stats_dic['n_images'] - \\n results_stats_dic['n_dogs_img']) \\n\\n # TODO: 5c. REPLACE zero(0.0) with CODE that calculates the % of correctly\\n # matched images. Recall that this can be calculated by the\\n # number of correctly matched images ('n_match') divided by the \\n # number of images('n_images'). This result will need to be \\n # multiplied by 100.0 to provide the percentage.\\n # \\n # Calculates % correct for matches\\n results_stats_dic['pct_match'] = (results_stats_dic['n_match'] / results_stats_dic['n_images']) * 100\\n\\n # TODO: 5d. REPLACE zero(0.0) with CODE that calculates the % of correctly\\n # classified dog images. Recall that this can be calculated by \\n # the number of correctly classified dog images('n_correct_dogs')\\n # divided by the number of dog images('n_dogs_img'). This result \\n # will need to be multiplied by 100.0 to provide the percentage.\\n # \\n # Calculates % correct dogs\\n results_stats_dic['pct_correct_dogs'] = (results_stats_dic['n_correct_dogs'] / results_stats_dic['n_dogs_img']) * 100\\n\\n # TODO: 5e. REPLACE zero(0.0) with CODE that calculates the % of correctly\\n # classified breeds of dogs. Recall that this can be calculated \\n # by the number of correctly classified breeds of dog('n_correct_breed') \\n # divided by the number of dog images('n_dogs_img'). This result \\n # will need to be multiplied by 100.0 to provide the percentage.\\n # \\n # Calculates % correct breed of dog\\n results_stats_dic['pct_correct_breed'] = (results_stats_dic['n_correct_breed'] / results_stats_dic['n_dogs_img']) * 100\\n\\n # Calculates % correct not-a-dog images\\n # Uses conditional statement for when no 'not a dog' images were submitted \\n if results_stats_dic['n_notdogs_img'] > 0:\\n results_stats_dic['pct_correct_notdogs'] = (results_stats_dic['n_correct_notdogs'] /\\n results_stats_dic['n_notdogs_img']) * 100.0\\n else:\\n results_stats_dic['pct_correct_notdogs'] = 0.0\\n\\n \\n # TODO 5f. REPLACE None with the results_stats_dic dictionary that you \\n # created with this function \\n return results_stats_dic\",\n \"def search(self, query_image, list_images):\\n\\t\\tresult = {}\\n\\t\\t# Get variables from query image\\n\\t\\tquery_kps, query_descriptor = query_image.key_points_and_descriptors()\\n\\t\\t# Loop in the list of images\\n\\t\\tfor image in list_images:\\n\\t\\t\\t# Varuables of the current image\\n\\t\\t\\tkps, descriptor = image.key_points_and_descriptors()\\n\\t\\t\\t# Get the score of the current image compared with the query image\\n\\t\\t\\tscore = self.get_score(query_kps, query_descriptor, kps, descriptor)\\n\\t\\t\\tresult[image.name()] = score\\n\\n\\t\\t# Sort results by score\\n\\t\\tif len(result) > 0:\\n\\t\\t\\tresult = sorted([(v, k) for (k, v) in result.items() if v > 0],\\n\\t\\t\\t\\treverse = True)\\n\\t\\treturn result\",\n \"def getOverlapComparisonImagesDict(self) :\\n overlap_shift_comparisons = {}\\n for o in self.overlaps :\\n overlap_shift_comparisons[o.getShiftComparisonDetailTuple()]=o.getShiftComparisonImages()\\n return overlap_shift_comparisons\",\n \"def __call__(self, results):\\r\\n if isinstance(results['img'], str):\\r\\n results['filename'] = results['img']\\r\\n results['ori_filename'] = results['img']\\r\\n else:\\r\\n results['filename'] = None\\r\\n results['ori_filename'] = None\\r\\n img = mmcv.imread(results['img'])\\r\\n results['img'] = img\\r\\n results['img_fields'] = ['img']\\r\\n results['img_shape'] = img.shape\\r\\n results['ori_shape'] = img.shape\\r\\n return results\",\n \"def get_all_results(pred_root, meta_results):\\r\\n results_all = {}\\r\\n for key in tqdm(meta_results, desc='Generating results ..'):\\r\\n persons = meta_results[key]\\r\\n\\r\\n global_seg = cv2.imread(pred_root + 'global_seg/{}.png'.format(key),\\r\\n cv2.IMREAD_UNCHANGED)\\r\\n global_tag = cv2.imread(pred_root + 'global_tag/{}.png'.format(key),\\r\\n cv2.IMREAD_UNCHANGED)\\r\\n\\r\\n results = {}\\r\\n dets, masks = [], []\\r\\n for p_id, score in persons:\\r\\n mask = (global_tag == p_id)\\r\\n if np.sum(mask) == 0:\\r\\n continue\\r\\n seg = mask * global_seg\\r\\n ys, xs = np.where(mask > 0)\\r\\n x1, y1, x2, y2 = xs.min(), ys.min(), xs.max(), ys.max()\\r\\n dets.append((x1, y1, x2, y2, score))\\r\\n masks.append(seg)\\r\\n\\r\\n # Reuiqred Field of each result: a list of masks,\\r\\n # each is a multi-class masks for one person.\\r\\n # It can also be sparsified to\\r\\n # [scipy.sparse.csr_matrix(mask) for mask in masks]\\r\\n # to save memory cost\\r\\n results['MASKS'] = masks if not Sparse \\\\\\r\\n else [scipy.sparse.csr_matrix(m) for m in masks]\\r\\n # Reuiqred Field of each result,\\r\\n # a list of detections corresponding to results['MASKS'].\\r\\n results['DETS'] = dets\\r\\n\\r\\n if cache_pkl:\\r\\n results_cache_add = cache_pkl_path + key + '.pklz'\\r\\n pickle.dump(results, gzip.open(results_cache_add, 'w'))\\r\\n results_all[key] = results_cache_add\\r\\n else:\\r\\n results_all[key] = results\\r\\n\\r\\n if PLOT:\\r\\n import pylab as plt\\r\\n plt.figure('seg')\\r\\n plt.imshow(global_seg)\\r\\n print('Seg unique:' + str(np.unique(global_seg)))\\r\\n plt.figure('tag')\\r\\n plt.imshow(global_tag)\\r\\n print('Tag unique:' + str(np.unique(global_tag)))\\r\\n plt.show()\\r\\n\\r\\n return results_all\",\n \"def _match_keypoints(self, query_image: Image) -> typing.Tuple[typing.List[int], typing.List[typing.Any]]:\\n query_image.get_keypoints_and_descriptors()\\n\\n # matches = self.matcher.match(query_image.descriptor)\\n matches = self.matcher.knnMatch(query_image.descriptor, k=2)\\n good = []\\n for m_n in matches:\\n if len(m_n) == 1:\\n good.append(m_n[0])\\n continue\\n elif len(m_n) != 2:\\n continue\\n (m, n) = m_n\\n if m.distance < 0.7 * n.distance:\\n good.append(m)\\n\\n images_scores = [0] * len(self._candidate_images)\\n images_matches = [None] * len(self._candidate_images)\\n for image_index, image in enumerate(self._candidate_images):\\n matches_scores = []\\n matches = []\\n for i, match in enumerate(good):\\n if match.imgIdx != image_index:\\n continue\\n matches.append(match)\\n matches_scores.append((256 - match.distance) / 256)\\n\\n match_cnt = len(matches_scores)\\n if match_cnt <= 0:\\n continue\\n\\n images_scores[image_index] = (\\n 0.5 + ((math.tanh(match_cnt / 3 - 1)) / 2)) * (sum(matches_scores) / match_cnt)\\n\\n images_matches[image_index] = matches\\n\\n return images_scores, images_matches\",\n \"def _get_images(self):\\n raw_outputs = self.interface.get_data(self.target_charge,\\n self.charge_deviation,\\n n_samples=self.n_samples)\\n\\n # apply roi to images\\n roi_images = []\\n for i in range(self.n_samples):\\n roi_images += [apply_roi(raw_outputs['raw_images'][i], raw_outputs['ROI'])]\\n\\n # process and identify blobs in image\\n min_size = 100\\n outputs = {}\\n for ele in self.output_keys:\\n outputs[ele] = []\\n\\n for i in range(len(roi_images)):\\n processed_image_data = image_processing.process_and_fit(roi_images[i],\\n min_size)\\n\\n for ele in self.output_keys:\\n if ele == 'image_check':\\n outputs[ele] += [image_processing.check_image(processed_image_data['binary_image'],\\n processed_image_data['smoothed_image'])]\\n elif ele == 'processed_images':\\n outputs[ele] += [processed_image_data['smoothed_image']]\\n else:\\n outputs[ele] += [processed_image_data[ele]]\\n\\n for ele in self.output_keys:\\n outputs[ele] = np.array(outputs[ele])\\n\\n # add in raw data\\n outputs.update(raw_outputs)\\n\\n # if we need to, get averaged results\\n if self.average_measurements:\\n avg_keys = ['rms_x', 'rms_y', 'CX', 'CY', 'n_blobs', 'FWHMX', 'FWHMY', 'centroid_offset']\\n for key in avg_keys:\\n outputs[key] = np.nanmean(outputs[key])\\n\\n return outputs\",\n \"def statistics(img1_blobs, img2_blobs, matches):\\n\\tstatistics = {}\\n\\n\\tstatistics['#Interest Points in img1'] = len(img1_blobs)\\n\\tstatistics['#Interest Points in img2'] = len(img2_blobs)\\n\\tstatistics['Accepted Matches'] = len(matches)\\n\\tdissimilarity = [match[2] for match in matches]\\n\\tstatistics['Mean of accepted matches'] = sum(dissimilarity)/len(dissimilarity)\\n\\tstatistics['SD of accepted matches'] = np.std(dissimilarity)\\n\\treturn statistics\",\n \"def get_stats(filtered_paths : dict) -> dict:\\n stats_dict = defaultdict(dict)\\n for mag in filtered_paths.keys():\\n for spec,paths in filtered_paths[mag].items():\\n counter = Counter()\\n for path in paths:\\n img = cv2.imread(path, -1)\\n count = Counter(list(img.ravel()))\\n counter += count\\n stats_dict[mag][spec] = counter\\n print(mag)\\n return stats_dict\",\n \"def get_results(self):\\n summary = self.handle.get_summary_data(self.group_name)\\n results = {'template': {'status': 'no data'},\\n 'complement': {'status': 'no data'},\\n '2d': {'status': 'no data'}}\\n if 'genome_mapping_template' in summary:\\n results['template'] = self._get_results(summary['genome_mapping_template'])\\n if 'genome_mapping_complement' in summary:\\n results['complement'] = self._get_results(summary['genome_mapping_complement'])\\n if 'genome_mapping_2d' in summary:\\n results['2d'] = self._get_results(summary['genome_mapping_2d'])\\n return results\",\n \"def init_results(self):\\n results = {}\\n for i in range(len(self.agents)):\\n actor = self.agents[i]\\n actor.set_name(\\\"Actor\\\") # the filename part of the name is set in actor using filename from init_agents\\n results[actor.name] = 0\\n return results\",\n \"def __call__(self, results):\\n\\n if 'img_fields' in results:\\n assert results['img_fields'] == ['img'], \\\\\\n 'Only single img_fields is allowed'\\n img = results['img']\\n assert img.dtype == np.float32, \\\\\\n 'PhotoMetricDistortion needs the input image of dtype ' \\\\\\n 'np.float32, please set \\\"to_float32=True\\\" in ' \\\\\\n '\\\"LoadImageFromFile\\\" pipeline'\\n # random brightness\\n if random.randint(2):\\n delta = random.uniform(-self.brightness_delta,\\n self.brightness_delta)\\n img += delta\\n\\n # mode == 0 --> do random contrast first\\n # mode == 1 --> do random contrast last\\n mode = random.randint(2)\\n if mode == 1:\\n if random.randint(2):\\n alpha = random.uniform(self.contrast_lower,\\n self.contrast_upper)\\n img *= alpha\\n\\n # convert color from BGR to HSV\\n img = general_ocr.bgr2hsv(img)\\n\\n # random saturation\\n if random.randint(2):\\n img[..., 1] *= random.uniform(self.saturation_lower,\\n self.saturation_upper)\\n\\n # random hue\\n if random.randint(2):\\n img[..., 0] += random.uniform(-self.hue_delta, self.hue_delta)\\n img[..., 0][img[..., 0] > 360] -= 360\\n img[..., 0][img[..., 0] < 0] += 360\\n\\n # convert color from HSV to BGR\\n img = general_ocr.hsv2bgr(img)\\n\\n # random contrast\\n if mode == 0:\\n if random.randint(2):\\n alpha = random.uniform(self.contrast_lower,\\n self.contrast_upper)\\n img *= alpha\\n\\n # randomly swap channels\\n if random.randint(2):\\n img = img[..., random.permutation(3)]\\n\\n results['img'] = img\\n return results\",\n \"def get_results(self) -> Dict:\\n results = {}\\n results['time'] = np.array(self._results['time'])\\n results['norm'] = np.array(self._results['norm'])\\n results['bond_dimensions'] = np.array(self._results['bond_dimensions'])\\n results['dynamics'] = self._results['dynamics']\\n results['pt_bond_dimensions'] = self._results['pt_bond_dimensions']\\n return results\",\n \"def write_results(input_dict):\\n out_imgs = {}\\n for mol in input_dict:\\n mols = []\\n for new_mols in input_dict[mol]:\\n m = Chem.MolFromSmiles(new_mols)\\n mols.append(m)\\n if len(mols) > 2:\\n p = Chem.MolFromSmarts(MCS.FindMCS(mols).smarts)\\n AllChem.Compute2DCoords(p)\\n for m in mols:\\n AllChem.GenerateDepictionMatching2DStructure(m, p)\\n out_imgs[mol] = Draw.MolsToGridImage(mols, useSVG=True)\\n # Write out the image\\n return out_imgs\",\n \"def findMatchesBetweenImages(image_1, image_2):\\n # matches - type: list of cv2.DMath\\n matches = None\\n # image_1_kp - type: list of cv2.KeyPoint items.\\n image_1_kp = None\\n # image_1_desc - type: numpy.ndarray of numpy.uint8 values.\\n image_1_desc = None\\n # image_2_kp - type: list of cv2.KeyPoint items.\\n image_2_kp = None\\n # image_2_desc - type: numpy.ndarray of numpy.uint8 values.\\n image_2_desc = None\\n # WRITE YOUR CODE HERE.\\n\\n sift = cv2.ORB_create()\\n image_1_kp, image_1_desc = sift.detectAndCompute(image_1, None)\\n image_2_kp, image_2_desc = sift.detectAndCompute(image_2, None)\\n\\n # bf = cv2.BFMatcher(cv2.NORM_HAMMING, crossCheck=True)\\n # matches = bf.match(image_1_desc, image_2_desc)\\n # matches = sorted(matches, key = lambda x:x.distance)\\n # matches = matches[:10]\\n\\n bf = cv2.BFMatcher()\\n matches = bf.knnMatch(image_1_desc, image_2_desc, k=2)\\n\\n # Apply ratio test\\n good = []\\n for m, n in matches:\\n print(m.distance, n.distance, m.distance < .75*n.distance)\\n if m.distance < (0.75 * n.distance):\\n good.append([m])\\n\\n # We coded the return statement for you. You are free to modify it -- just\\n # make sure the tests pass.\\n print(len(good), good)\\n return image_1_kp, image_2_kp, matches\\n # END OF FUNCTION.\",\n \"def setUp(self):\\n img_path = osp.join(osp.dirname(__file__), '../../data/gray.jpg')\\n self.results = {\\n 'img_path':\\n img_path,\\n 'img_shape': (300, 400),\\n 'instances': [{\\n 'bbox': [0, 0, 10, 20],\\n 'bbox_label': 1,\\n 'mask': [[0, 0, 0, 20, 10, 20, 10, 0]],\\n 'ignore_flag': 0\\n }, {\\n 'bbox': [10, 10, 110, 120],\\n 'bbox_label': 2,\\n 'mask': [[10, 10, 110, 10, 110, 120, 110, 10]],\\n 'ignore_flag': 0\\n }, {\\n 'bbox': [50, 50, 60, 80],\\n 'bbox_label': 2,\\n 'mask': [[50, 50, 60, 50, 60, 80, 50, 80]],\\n 'ignore_flag': 1\\n }]\\n }\",\n \"def results(self):\\n \\n results = {}\\n if self.descriptors is not None:\\n for i in range(len(self.items)):\\n results[self.items[i]] = {'I': self.ivalue(i),\\n 'J': self.jvalue(i)}\\n else:\\n for i in range(len(self.items)):\\n results[self.items[i]] = {'I': self.ivalue(i)}\\n return results\",\n \"def what_image_is_this(self, captured_image, reference_images_dictionary, respawn_filter=False):\\n matched_array = []\\n captured_image_list = captured_image.tolist()\\n total = {}\\n\\n apply_respawn_filter = hasattr(self, 'respawnFilter') and respawn_filter\\n\\n for item_name, reference_image in reference_images_dictionary.items():\\n total[item_name] = 0\\n row = 0\\n for reference_row in reference_image: # captured_image must not be larger than any reference image\\n pixel = 0\\n for reference_pixel in reference_row:\\n skip = False\\n # print(\\\"Reference Pixel: \\\" + str(reference_pixel) +\\n # \\\" Captured Pixel: \\\" + str(captured_image_list[row][pixel][0]))\\n if apply_respawn_filter: # apply filter when respawning\\n if self.respawnFilter[\\\"respawn-filter\\\"][row][pixel] == 0:\\n # filter out this pixel, it could be red or white and will be covering the hero\\n skip = True\\n if not skip:\\n total[item_name] += 1\\n if reference_pixel == captured_image_list[row][pixel][0]:\\n matched_array.append(item_name)\\n pixel = pixel + 1\\n row = row + 1\\n counter = Counter(matched_array)\\n most_common = counter.most_common()\\n ratios = {}\\n for item in most_common:\\n ratios[item[0]] = item[1] / total[item[0]]\\n return ratios\",\n \"def _result_dict_for_single_example(image, key, detections, gt_boxes, gt_labels):\\n image_shape = tf.shape(image)\\n detection_boxes = detections[\\\"detection_boxes\\\"][0]\\n detection_scores = detections[\\\"detection_scores\\\"][0]\\n detection_classes = tf.to_int64(detections[\\\"detection_classes\\\"][0])\\n num_detections = tf.to_int32(detections[\\\"num_detections\\\"][0])\\n detection_boxes = tf.slice(detection_boxes, begin=[0, 0], size=[num_detections, -1])\\n detection_classes = tf.slice(detection_classes, begin=[0], size=[num_detections])\\n detection_scores = tf.slice(detection_scores, begin=[0], size=[num_detections])\\n # scale to abs coordinates\\n y_scale, x_scale = image_shape[1], image_shape[2]\\n detection_boxes = _scale_to_abs_coord(detection_boxes, y_scale, x_scale)\\n gt_boxes = _scale_to_abs_coord(gt_boxes, y_scale, x_scale)\\n return dict(image=image,\\n key=key,\\n detection_boxes=detection_boxes,\\n detection_scores=detection_scores,\\n detection_classes=detection_classes,\\n num_detections=num_detections,\\n groundtruth_boxes=gt_boxes,\\n groundtruth_classes=gt_labels)\",\n \"def compute_metrics(self, results: list) -> dict:\",\n \"def generate_scanset_metadata( image_set_dictionary, html_base_path, session_id ):\\r\\n\\tcur_subj_info = {}\\r\\n\\r\\n\\t\\\"\\\"\\\"need to think through the data structure a bit more.... but can always adjust later \\\"\\\"\\\"\\r\\n\\tcur_subj_info['session_id'] = session_id\\r\\n\\t#cur_subj_info['img_id'] = counter\\r\\n\\tcur_subj_info['subject_id'] = session_id.split('/')[0]\\r\\n\\tglobal counter\\r\\n\\t\\r\\n\\tnii_image_dict = image_set_dictionary['nii_images']\\r\\n\\tpng_image_dict = image_set_dictionary['png_image_set']\\r\\n\\r\\n\\tscan_metadata = {}\\r\\n\\tfor scan in nii_image_dict:\\r\\n\\t\\tprint \\\"propcessing \\\", scan\\r\\n\\t\\tnii_img = nii_image_dict[scan]['base_image'][0]\\r\\n\\t\\tprint nii_img\\r\\n#\\t\\tif 'mask' not in scan:\\r\\n\\r\\n#\\t\\tif 'mask' not in scan:\\r\\n#\\t\\t\\tnii_img = nii_image_dict[scan]['base_image'][0]\\r\\n#\\t\\telse:\\r\\n#\\t\\t\\tcontinue\\r\\n#\\t\\tprint \\\"HI DAVE!\\\"\\r\\n\\r\\n\\r\\n\\t\\tif not nii_img:\\r\\n\\t\\t\\tprint \\\"did not find base image for\\\",nii_image_dict\\r\\n\\t\\t\\tcontinue\\r\\n\\t\\tpng_img = html_path_root+ png_image_dict[scan]\\r\\n\\t\\tprint nii_img,\\\"is being passed\\\"\\r\\n\\r\\n\\t\\t(dim_x, dim_y, dim_z, vox_size_x, vox_size_y, vox_size_z, image_orientation )= igcf.get_nii_image_info(nii_img)\\r\\n\\t\\timage_info = Image.open(png_img)\\r\\n\\t\\twidth, height = image_info.size\\r\\n\\t\\t#print width,height,dim_x,dim_y,dim_z,vox_size_x,vox_size_y,vox_size_z\\t\\r\\n\\t\\tscan_info = {}\\r\\n\\t\\tscan_info['slice_width'] = dim_x\\r\\n\\t\\tscan_info['slice_height'] = dim_y\\r\\n\\t\\tscan_info['num_slices'] = dim_z\\r\\n\\t\\tscan_info['main_image_width'] = width\\r\\n\\t\\tscan_info['main_image_height'] = height\\r\\n\\t\\tscan_info['nii_image'] = nii_img\\r\\n\\t\\tscan_info['base_url'] = png_img.replace(html_path_root,'')\\r\\n\\t\\tscan_metadata[scan] = scan_info\\r\\n\\r\\n\\t\\t### There can be one or MORE masks for a given base image... so I will return a list of\\r\\n\\t\\t#dictionaries..\\r\\n\\t\\tmask_list = nii_image_dict[scan]['masks']\\r\\n\\t\\tmask_id = 0\\r\\n\\r\\n\\t\\tmask_info_list = []\\r\\n\\t\\tfor mask in mask_list:\\r\\n\\t\\t\\tcur_mask_info = {}\\r\\n\\t\\t\\t### I'll call the mask by it's basename\\r\\n\\t\\t\\tprint mask,\\\"was passed...\\\"\\r\\n\\t\\t\\tmask_base = os.path.basename(mask)\\r\\n\\t\\t\\tnii_img = nii_image_dict[scan]['masks'][mask_id]\\r\\n\\t\\t\\tprint nii_image_dict,'mask_id is',mask_id\\r\\n\\t\\t\\tprint \\\"nii maeg found should be\\\",nii_img\\r\\n\\t\\t\\tif not nii_img:\\r\\n\\t\\t\\t\\tprint \\\"did not find a valid mask image for \\\",nii_image_dict\\r\\n\\t\\t\\t\\tcontinue\\r\\n\\r\\n\\r\\n\\t\\t\\tcur_mask_info['name'] = mask_base\\r\\n\\t\\t\\tcur_mask_info['id'] = mask_id\\r\\n\\t\\t\\tcur_mask_info['nii_file'] = nii_img\\r\\n\\t\\t\\t## NEED TO ADD IN THE MASK_URL\\r\\n#\\t\\t\\tcur_mask_info['mask_url'] = \\t\\t\\t\\r\\n\\t\\t\\tprint png_image_dict\\r\\n\\t\\t\\tpng_img = html_path_root+ png_image_dict[scan]\\r\\n\\t\\t\\tprint nii_img,\\\"is being passed\\\"\\r\\n\\t\\t\\tcur_mask_info['mask_url'] = png_img.replace(html_path_root,'')\\r\\n\\r\\n\\t\\t\\tmask_info_list.append( cur_mask_info )\\r\\n\\t\\r\\n\\t\\t\\r\\n\\t\\t\\tmask_id +=1\\r\\n#\\t\\tprint cur_mask_info\\r\\n\\t\\tcur_subj_info['masks'] = mask_info_list\\r\\n\\t\\tscan_metadata[scan]['masks'] = [ mask_info_list]\\r\\n#\\t\\tprint mask_info_list\\r\\n\\r\\n\\tcur_subj_info['image_data'] = scan_metadata\\r\\n\\tcounter += 1\\t\\r\\n\\treturn { 'session_name': session_id , 'session_metadata': cur_subj_info }\",\n \"def images(self) -> dict:\\n raise NotImplementedError\",\n \"def __getitem__(self, index):\\n (path_1, path_2, is_same, id0, id1) = self.pairs[index]\\n \\n try:\\n img1, img2 = self._load_img(path_1), self._load_img(path_2)\\n return {'img1': img1, 'img2': img2, 'is_same': is_same, 'id0': id0, 'id1': id1}\\n except FileNotFoundError:\\n return {'is_same': is_same, 'id0': id0, 'id1': id1, 'path1': path_1, 'path2': path_2}\",\n \"def generate_results(img_name, note_names, note_durations):\\n construct_output(indent_level=\\\"block\\\",\\n message=\\\"Matching the elements of the image ({}) with the values given by the networks.\\\"\\n .format(img_name))\\n\\n results = [] # A list that will be returned.\\n\\n # Path that contains the information about the original image.\\n image_path = os.path.abspath(os.path.join(str(Path(__file__).parent.parent.parent), 'resources', 'input_images'))\\n image_path = os.path.join(image_path, 'input_images_individual_elements')\\n image_path = os.path.join(image_path, img_name[:-4])\\n\\n # Find all the unrecognized elements and their names.\\n unrecognized_elements_path = os.path.join(image_path, \\\"unrecognized\\\")\\n unrecognized_elements = os.listdir(unrecognized_elements_path)\\n\\n results_and_positions = []\\n for el in unrecognized_elements:\\n row_number = el[(el.index(\\\"_row\\\") + len(\\\"_row\\\")):(el.index(\\\"_slice\\\"))]\\n el_number = el[(el.index(\\\"slice\\\") + len(\\\"slice\\\")):]\\n el_number = el_number[:el_number.index(\\\"_\\\")]\\n results_and_positions.append((el, row_number, el_number))\\n results_and_positions.sort(key=lambda k: (int(k[1]), int(k[2])))\\n if len(note_names) != len(note_durations) != len(unrecognized_elements):\\n print(\\\"generator.py = Not enough elements were recognized!\\\")\\n\\n for index, el in enumerate(results_and_positions):\\n el = el[0]\\n if el.endswith(\\\".png\\\"):\\n results.append((el[len(\\\"img02_\\\"):-len(\\\".png\\\")], note_names[index], note_durations[index]))\\n\\n # Find all the recognized elements and their names.\\n recognized_elements_path = os.path.join(image_path, \\\"recognized\\\", img_name[:-4] + \\\".txt\\\")\\n\\n with open(recognized_elements_path, 'r') as file:\\n recognized_elements = file.readlines()\\n if len(recognized_elements) == 0:\\n print(\\\"generator.py = Error in reading image file info!\\\")\\n exit(-1)\\n for index, content in enumerate(recognized_elements):\\n if content.endswith(\\\"\\\\n\\\"):\\n recognized_elements[index] = content[:-1]\\n for index, el in enumerate(recognized_elements):\\n el_name = el[(el.find(\\\"template_\\\") + len(\\\"template_\\\")):el.find(\\\".png\\\")]\\n if el_name.startswith(\\\"Z_barline\\\"):\\n el_name = \\\"barline\\\"\\n results.append((el[len(\\\"img02_\\\"):-len(\\\".png\\\")], el_name, \\\"-\\\"))\\n\\n # Join the individual elements with their descriptions.\\n results_and_positions = []\\n for result in results:\\n x = result[0]\\n row_number = x[(x.index(\\\"row\\\") + len(\\\"row\\\")):x.index(\\\"_slice\\\")]\\n el_number = x[(x.index(\\\"slice\\\") + len(\\\"slice\\\")):]\\n el_number = el_number[:el_number.index(\\\"_\\\")]\\n results_and_positions.append((result, row_number, el_number))\\n\\n # Sort the elements with the order found in the original image.\\n results_and_positions.sort(key=lambda k: (int(k[1]), int(k[2])))\\n\\n results = []\\n for x in results_and_positions:\\n results.append(x[0])\\n\\n construct_output(indent_level=\\\"block\\\",\\n message=\\\"Results generated.\\\"\\n .format(img_name))\\n\\n return results\",\n \"def get_images_and_labels(tampered_path, authentic_path):\\n tampered_dir = tampered_path\\n authentic_dir = authentic_path\\n images = {}\\n for im in glob.glob(authentic_dir):\\n images[im] = {}\\n images[im]['mat'] = cv2.imread(im)\\n images[im]['label'] = 0\\n for im in glob.glob(tampered_dir):\\n images[im] = {}\\n images[im]['mat'] = cv2.imread(im)\\n images[im]['label'] = 1\\n return images\",\n \"def __call__(self, results):\\n for key in results.get('img_fields', ['img']):\\n results[key] = general_ocr.imnormalize(results[key], self.mean, self.std,\\n self.to_rgb)\\n results['img_norm_cfg'] = dict(\\n mean=self.mean, std=self.std, to_rgb=self.to_rgb)\\n return results\",\n \"def get_matches(jig_sol, blocks_roi):\\n\\n match_data = {}\\n height, width,= jig_sol.shape\\n font = cv2.FONT_HERSHEY_SIMPLEX\\n\\n #identify the puzzle peice number based on the peice block position in solution image\\n identity = {\\n (1, 1): 1,\\n (2, 1): 2,\\n (3, 1): 3,\\n (1, 2): 4,\\n (2, 2): 5,\\n (3, 2): 6,\\n (1, 3): 7,\\n (2, 3): 8,\\n (3, 3): 9,\\n }\\n\\n #iterate through the blocks roi\\n for i in blocks_roi.keys():\\n blk = blocks_roi[i].copy()\\n blk = cv2.cvtColor(blk,cv2.COLOR_BGR2GRAY)\\n max_list = []\\n\\n #for eack blk rotate is by 90 degrees and try template matching\\n for k in range(0,360,90):\\n #cv2.resize(blk,(int(width/3),int(height/3)), interpolation= cv2.INTER_CUBIC)\\n blk_copy = imutils.rotate_bound(blk, -k)\\n\\n #get the resulting heat map of template matching\\n result = cv2.matchTemplate(jig_sol,blk_copy,cv2.TM_CCOEFF_NORMED)\\n\\n #get the max value and its location in the heat map\\n _, max_val, _, max_loc = cv2.minMaxLoc(result)\\n\\n #append a tuple consisting of max location, value and peice rotation to max_list\\n max_list.append((max_loc, max_val, k))#((k+1)*90)%360))\\n\\n #find the location with maximum value of template matching regardless of peice rotation\\n top_left = max(max_list, key=lambda x: x[1])[0]\\n\\n #get the peice rotation of that template matching\\n rot = max(max_list, key=lambda x: x[1])[2]\\n\\n #calculate the bottom right cordinates of the block\\n bottom_right = (top_left[0] + int(width/3), top_left[1] + int(height/3))\\n\\n #find the center of the block\\n centx = 0\\n centy = 0\\n for (l,m) in [top_left, bottom_right]:\\n centx += l\\n centy += m\\n centx = int(centx/2)\\n centy = int(centy/2)\\n\\n #get the puzzle peice block position in solution image\\n piece = (math.ceil(3*centx/width), math.ceil(3*centy/height))\\n\\n if piece not in identity.keys():\\n continue\\n\\n match_data.update({i: (identity[piece], rot)})\\n\\n return match_data\",\n \"def _init_results(self) -> None:\\n pt_bond_dimensions = {}\\n for site, pt in enumerate(self._process_tensors):\\n if pt is not None:\\n pt_bond_dimensions[site] = pt.get_bond_dimensions()\\n\\n self._results = {\\n 'time':[],\\n 'norm': [],\\n 'bond_dimensions': [],\\n 'dynamics': {},\\n 'pt_bond_dimensions': pt_bond_dimensions,\\n }\\n for sites in self._dynamics_sites:\\n self._results['dynamics'][sites] = Dynamics(name=f\\\"site{sites}\\\")\",\n \"def detail_matching(self):\\n paradic = self.cfg['param']['paradic']\\n work_dir = self.work_dir\\n \\n x = float(self.cfg['param']['x']) # selected pixel in the first image\\n y = float(self.cfg['param']['y'])\\n \\n # sift parameters\\n # number of bins in the orientation histogram\\n n_bins = int(paradic['n_bins']) \\n n_hist = int(paradic['n_hist']) \\n # descriptor of n_hist X n_hist weighted histograms with n_ori\\n n_ori = int(paradic['n_ori']) \\n delta_min = float(paradic['delta_min'])\\n sigma_min = float(paradic['sigma_min'])\\n sigma_in = float(paradic['sigma_in'])\\n lambda_ori = float(paradic['lambda_ori'])\\n lambda_descr = float(paradic['lambda_descr'])\\n #threshold defining reference orientations\\n n_spo = int(paradic['n_spo'])\\n \\n # Read feature vectors from output files\\n if (os.path.getsize(work_dir+'OUTmatches.txt') > 0 ):\\n pairdata = find_nearest_keypoint(work_dir+'OUTmatches.txt', y, x)\\n \\n illustrate_pair(pairdata, n_bins, n_hist, n_ori, work_dir)\\n\\n \\n # Read keys coordinates.\\n d = 6+n_bins+n_hist*n_hist*n_ori # size of keydata inside pairdata\\n v = n_hist*n_hist*n_ori\\n [x1, y1, sigma1, theta1] = [float(x) for x in pairdata[0:4]]\\n [o1, s1] = [float(x) for x in pairdata[4+v:4+v+2]]\\n [x2a, y2a, sigma2a, theta2a] = [float(x) for x in pairdata[d:d+4]]\\n [o2a, s2a] = [float(x) for x in pairdata[d+4+v:d+4+v+2]]\\n [x2b, y2b, sigma2b, theta2b] = \\\\\\n [float(x) for x in pairdata[2*d:2*d+4]]\\n [o2b, s2b] = [float(x) for x in pairdata[2*d+4+v:2*d+4+v+2]]\\n \\n draw_one_match(pairdata,\\n work_dir+'input_0.png',\\n work_dir+'input_1.png',\\n d,\\n lambda_ori,\\n lambda_descr,\\n n_hist,\\n work_dir+'OUTonepair.png')\\n \\n \\n # Extract thumbnails.\\n # keypoint 1 (image 1)\\n print ' '.join(['demo_extract_patch', work_dir+'input_0.png',\\n str(x1), str(y1), str(sigma1), str(theta1), str(o1), str(s1),\\n str(delta_min), str(sigma_min), str(sigma_in), str(n_spo),\\n str(lambda_ori), str(lambda_descr), str(n_hist),\\n work_dir+\\\"detail_im1\\\"])\\n proc = self.run_proc(['demo_extract_patch', work_dir+'input_0.png',\\n str(x1), str(y1), str(sigma1), str(theta1), str(o1), str(s1),\\n str(delta_min), str(sigma_min), str(sigma_in), str(n_spo),\\n str(lambda_ori), str(lambda_descr), str(n_hist),\\n work_dir+\\\"detail_im1\\\"])\\n self.wait_proc(proc, timeout=self.timeout)\\n \\n # keypoint 2a (nearest neighbor in image 2)\\n print ' '.join(['demo_extract_patch', work_dir+'input_1.png',\\n str(x2a), str(y2a), str(sigma2a), str(theta2a), str(o2a), str(s2a),\\n str(delta_min), str(sigma_min), str(sigma_in), str(n_spo),\\n str(lambda_ori), str(lambda_descr), str(n_hist),\\n work_dir+\\\"detail_im2a\\\"])\\n proc = self.run_proc(['demo_extract_patch', work_dir+'input_1.png',\\n str(x2a), str(y2a), str(sigma2a), str(theta2a), str(o2a), str(s2a),\\n str(delta_min), str(sigma_min), str(sigma_in), str(n_spo),\\n str(lambda_ori), str(lambda_descr), str(n_hist),\\n work_dir+\\\"detail_im2a\\\"])\\n self.wait_proc(proc, timeout=self.timeout) \\n \\n # keypoint 2b (second nearest neighbor in image 2)\\n proc = self.run_proc(['demo_extract_patch', work_dir+'input_1.png',\\n str(x2b), str(y2b), str(sigma2b), str(theta2b), str(o2b), str(s2b),\\n str(delta_min), str(sigma_min), str(sigma_in), str(n_spo),\\n str(lambda_ori), str(lambda_descr), str(n_hist),\\n work_dir+\\\"detail_im2b\\\"])\\n self.wait_proc(proc, timeout=self.timeout) \\n \\n \\n return 1\",\n \"def extract_results_test(self):\\n assert len(self.results.keys()) != 0\\n TESTS = [\\n {\\n \\\"input\\\": {\\\"molecules\\\": [\\\"DDSPDLPK\\\"], \\\"score_threshold\\\": 0.95},\\n \\\"output\\\": {\\n \\\"formula\\\": \\\"C(37)H(59)N(9)O(16)\\\",\\n \\\"file_name\\\": \\\"BSA1.mzML\\\",\\n \\\"scaling_factor\\\": 100,\\n \\\"spec_id\\\": 1337,\\n },\\n }\\n ]\\n for test_dict in TESTS:\\n for key, n, entry in self.results.extract_results(**test_dict[\\\"input\\\"]):\\n print(key, entry)\\n assert key.formula == test_dict[\\\"output\\\"][\\\"formula\\\"]\\n assert key.file_name == test_dict[\\\"output\\\"][\\\"file_name\\\"]\\n assert entry.scaling_factor == test_dict[\\\"output\\\"][\\\"scaling_factor\\\"]\\n assert entry.spec_id == test_dict[\\\"output\\\"][\\\"spec_id\\\"]\\n # print(self.results)\\n # print(self.results.lookup)\\n assert n == 0\",\n \"def post_calculations(moving_dataset_image_ids, result=None):\\n if result is None:\\n result = {}\\n\\n for _id in moving_dataset_image_ids:\\n img = img_data(_id, util.DATA_FOLDER, util.TEMP_FOLDER_PATH)\\n img.load_db_transforms()\\n\\n img_pre = img_data(img.fixed_image, util.DATA_FOLDER, util.TEMP_FOLDER_PATH)\\n img_pre.load_db_transforms()\\n\\n reg_vol = util.transform_volume(img.reg_img_filepath, img_pre.get_transforms())\\n vol = util.TEMP_FOLDER_PATH + util.get_basename(basename(reg_vol)) + '_BE.nii.gz'\\n\\n mult = ants.MultiplyImages()\\n mult.inputs.dimension = 3\\n mult.inputs.first_input = reg_vol\\n mult.inputs.second_input = util.TEMPLATE_MASK\\n mult.inputs.output_product_image = vol\\n mult.run()\\n\\n label = \\\"img\\\"\\n if label in result:\\n result[label].append(vol)\\n else:\\n result[label] = [vol]\\n\\n for (segmentation, label) in util.find_reg_label_images(_id):\\n segmentation = util.transform_volume(segmentation, img_pre.get_transforms(), label_img=True)\\n if label in result:\\n result[label].append(segmentation)\\n else:\\n result[label] = [segmentation]\\n return result\",\n \"def parse_image_data(image_urls):\\n\\n # Initialize images.\\n initialized_ClImages = list(map(ClImage, image_urls))\\n app = ClarifaiApp(api_key=CLARIFAI_KEY)\\n image_info = {}\\n\\n # Obtain relevant tag information\\n try:\\n general_response = app.models.get('general-v1.3').predict(initialized_ClImages)\\n except ApiError as e:\\n error = json.loads(e.response.content)\\n pprint('error: {}'.format(error))\\n\\n else:\\n for item in general_response['outputs']:\\n video_id = item['input']['data']['image']['url'][23:34]\\n thumbnail_tags = set()\\n for tag in item['data']['concepts']:\\n if tag['value'] > .9:\\n tag_string = tag['name'].strip().lower()\\n if not Tag.query.filter(Tag.tag == tag_string).first():\\n add_tag_data(tag_string)\\n thumbnail_tags.add(tag_string)\\n image_info[video_id] = {'tags': thumbnail_tags}\\n\\n\\n # Obtain nsfw score\\n try:\\n nsfw_response = app.models.get('nsfw-v1.0').predict(initialized_ClImages)\\n except ApiError as e:\\n error = json.loads(e.response.content)\\n pprint('error: {}'.format(error))\\n\\n else:\\n for item in nsfw_response['outputs']: #nsfw_r['outputs'] is a list\\n video_id = item['input']['data']['image']['url'][23:34]\\n nsfw_score = round(item['data']['concepts'][-1]['value'] * 100)\\n image_info[video_id]['nsfw_score'] = nsfw_score\\n\\n\\n # Obtain color data\\n # try:\\n # color_response = app.models.get('color').predict(initialized_ClImages)\\n # except ApiError as e:\\n # error = json.loads(e.response.content)\\n # pprint('error: {}'.format(error[-100:]))\\n\\n # else:\\n # for item in color_response['outputs']:\\n # video_id = item['input']['data']['image']['url'][23:34]\\n # color_tags = {}\\n # for color in item['data']['colors']: # item['data']['colors'] is a list\\n # if color['value'] > .2:\\n # color_hex = color['w3c']['hex'].rstrip().lower()\\n # color_name = color['w3c']['name'].rstrip().lower()\\n # if not Color.query.filter(Color.hex_code == color_hex).first():\\n # add_color_data(color_hex, color_name)\\n # image_info[video_id]['colors'] = color_tags\\n\\n return image_info\",\n \"def result_as_dict_matches(query_result):\\n matches_keys = ['match_id', 'match_date', 'round_name', 'first_squad_name',\\n 'first_squad_score', 'first_squad_points',\\n 'second_squad_name', 'second_squad_score',\\n 'second_squad_points']\\n list_of_dict = []\\n for i in range(len(query_result)):\\n dictionary = dict(zip(matches_keys, query_result[i]))\\n list_of_dict.append(dictionary)\\n i += 1\\n return list_of_dict\",\n \"def patternMatchingTest(matchingAlgorithm,\\n observations= defaultCOMPREFdump , \\n modelsFolder= defaultWRFdumpsFolder,\\n #key1='0828',\\n key1='',\\n key2='',\\n #key3='',\\n verbose=True,\\n makeSmallImages=True,\\n smallDpi=40,\\n frameDpi = 600 , # added 2013-09-23\\n loadFormerAnalysisResult=False, #added 2013-09-23\\n #summary_folder=summary_folder, # needed if loadFormerAnalysisResult is True\\n summary_folder=summary_folder+timeString + '/', # 2013-09-27\\n matching_algorithm_name=\\\"\\\", # needed if loadFormerAnalysisResult is True\\n panel_cols = 5,\\n panel_rows = 5,\\n useCV2 = False, # switch off opencv since it doesn't work yet\\n useCV = False,\\n averageScoreOrdering=True, # take an average of the ordering to get a total score for the entire model over the time series\\n #kwargs = {}, # keyword arguments for the matchinging algorithm, just in case\\n **kwargs\\n ):\\n if key1 != '' or key2 != '':\\n summary_folder += '_'.join([key1, key2])\\n if not averageScoreOrdering:\\n summary_folder += '_nonaver'\\n summary_folder += '%dx%d/' % (panel_rows, panel_cols)\\n\\n t0 = time.time()\\n if loadFormerAnalysisResult:\\n mv_folder = summary_folder + matching_algorithm_name + '/'\\n result = pickle.load(open(mv_folder + 'result.pydump', 'r'))\\n\\n \\\"\\\"\\\"\\n result= {'best matches' : bestMatchAtEachTime, \\n 'best overall match' : bestMatchOverall, \\n 'all scores' : scores,\\n 'matching algorithm' : matching_algorithm_name,\\n 'dataset1[obs]' : observations,\\n 'dataset2[models]' : modelsFolder,\\n 'time spent' : time.time() - t0,\\n 'mv folder' : mv_folder, #added 2013-09-23\\n 'ordering' : ordering, #added 2013-09-23\\n }\\n \\\"\\\"\\\"\\n bestMatchAtEachTime = result['best matches'] \\n bestMatchOverall = result['best overall match']\\n scores = result['all scores']\\n matching_algorithm_name = result['matching algorithm']\\n observations = result['dataset1[obs]']\\n modelsFolder = result['dataset2[models]']\\n #mv_folder = result['mv folder']\\n ordering = result['ordering']\\n\\n # THE FOLLOWING 30 LINES ARE COPIED FROM \\\"ELSE...\\\" BELOW\\n ########### ******\\n ##\\n #\\n ds1 = pickle.load(open(observations,'r'))\\n ds1.setFloor(0) # added 2013-09-27\\n ds1.list = [v for v in ds1.list if (key1 in v.dataTime or key1 in v.name) \\\\\\n and (key2 in v.dataTime or key2 in v.name)]\\n ds1.list.sort(key=lambda v:v.dataTime)\\n ds1.setVmin(-40) #line added 2013-09-23\\n # debug\\n #print ds1.list\\n print 'loading list with keywords [ %s %s ]' %(key1, key2) \\n ds1.load(verbose=verbose)\\n # end debug\\n modelsList = [v for v in os.listdir(modelsFolder) if '.pydump' in v]\\n modelsList.sort(key = lambda v: int(re.findall(r'\\\\d+',v)[0]))\\n print \\\"modelsList\\\", modelsList\\n scores = {} # scores = {model number: [('dataTime', score), ('dataTime', score),...], ...}\\n print 'loading ds2 (model data)'\\n wrfs = [] # line added 2013-09-23\\n for path in modelsList:\\n modelIndex = int(re.findall(r'\\\\d+', path)[0])\\n ds2 = pickle.load(open(modelsFolder+path, 'r'))\\n ds2.setFloor(0) # added 2013-09-27\\n # debug\\n #ds2.list = [v for v in ds2.list if key1 in v.dataTime] # line redundant, 2013-09-23\\n ds1.list, ds2.list = ds1.intersect(ds2) #line added 2013-09-23 \\n ds2.setVmin(-40) #line added 2013-09-23\\n wrfs.append(ds2) # line added 2013-09-23\\n # end debug\\n\\n ####################\\n\\n\\n else:\\n # THE FOLLOWING 30 LINES ARE COPIED TO \\\"IF...\\\" ABOVE\\n ########### ******\\n ##\\n #\\n ds1 = pickle.load(open(observations,'r'))\\n ds1.list = [v for v in ds1.list if (key1 in v.dataTime or key1 in v.name) \\\\\\n and (key2 in v.dataTime or key2 in v.name)]\\n ds1.list.sort(key=lambda v:v.dataTime)\\n ds1.setVmin(-40) #line added 2013-09-23\\n # debug\\n #print ds1.list\\n print 'loading list with keywords [ %s %s ]' %(key1, key2) \\n ds1.load(verbose=verbose)\\n ds1.setFloor(0) # added 2013-09-27\\n # end debug\\n modelsList = [v for v in os.listdir(modelsFolder) if '.pydump' in v]\\n modelsList.sort(key = lambda v: int(re.findall(r'\\\\d+',v)[0]))\\n print \\\"modelsList\\\", modelsList\\n scores = {} # scores = {model number: [('dataTime', score), ('dataTime', score),...], ...}\\n print 'loading ds2 (model data)'\\n wrfs = [] # line added 2013-09-23\\n ds1.commonMaskSet=False # tempory attribute added 2013-09-27\\n for path in modelsList:\\n modelIndex = int(re.findall(r'\\\\d+', path)[0])\\n ds2 = pickle.load(open(modelsFolder+path, 'r'))\\n # debug\\n #ds2.list = [v for v in ds2.list if key1 in v.dataTime] # line redundant, 2013-09-23\\n ds1.list, ds2.list = ds1.intersect(ds2) #line added 2013-09-23 \\n ds2.setFloor(0) # added 2013-09-27\\n ds2[0].matrix.mask += ds1[0].matrix.mask # added 2013-09-27\\n ds2.setCommonMask() # added 2013-09-27\\n if not ds1.commonMaskSet: # added 2013-09-27\\n ds1[0].matrix.mask +=ds2[0].matrix.mask # added 2013-09-27\\n ds1.setCommonMask() # added 2013-09-27\\n ds2.setVmin(-40) #line added 2013-09-23\\n wrfs.append(ds2) # line added 2013-09-23\\n # end debug\\n\\n #######################################################################################################################\\n # key lines\\n print 'computing scores between %s and %s ...' %(ds1.name, ds2.name)\\n s, matching_algorithm_name = matchingAlgorithm(ds1, ds2, **kwargs) # list of the form [('20120612.0000', 0.9),...] \\n scores[modelIndex] = dict(s) # { model index: {'dataTime': score,...}, ...}\\n # debug\\n if verbose:\\n print scores\\n print '..................................................'\\n #\\n ########################################################################################################################\\n\\n print 'sorting the results...'\\n #print 'scores', '\\\\n', scores #debug\\n # bestMatchAtEachTime\\n T1 = [v.dataTime for v in ds1]\\n T2 = [v.dataTime for v in ds2] #any ds2 will do - they are the same, in this case we pick the last one \\n T = set(T1).intersection(T2)\\n dataTimeList = sorted(list(T))\\n\\n #print 'dataTimeList:', dataTimeList #debug\\n highestScores = []\\n ordering = [] #added 2013-09-23\\n\\n if averageScoreOrdering: # take the average scores before ordering to make it uniform\\n ave_scores = [ (v, np.mean( [scores[v].values() ] ) ) for v in scores.keys()]\\n ave_scores = dict(ave_scores)\\n print '..taking averages...............................'\\n print ave_scores\\n for u in scores.keys():\\n for t in scores[u].keys():\\n scores[u][t] = ave_scores[u]\\n print '\\\\n'.join([str(scores[v]) for v in scores.keys()])\\n\\n for t in dataTimeList:\\n print t\\n highestScoreAt_t = max([scores[u][t] for u in scores.keys()]) # u = model index (integer 1-20), t = dataTime (string)\\n new_entry = (t, highestScoreAt_t, [u for u in scores.keys() if scores[u][t]==highestScoreAt_t])\\n print new_entry\\n highestScores.append(new_entry)\\n #added 2013-09-23\\n ordering.append([0]+sorted(range(1, len(modelsList)+1), key=lambda v: scores[v][t], reverse=True))\\n\\n\\n\\n\\n # bestMatchOverall\\n print scores # debug\\n averageScores = [(Model, np.mean(scores[Model].values())) for Model in scores.keys()] # list of [(Model, mean score),...]\\n highestScore = max([v[1] for v in averageScores])\\n bestMatch = [u[0] for u in averageScores if u[1]==highestScore] \\n # debug\\n print 'average scores , highest average, best match', averageScores, highestScore, bestMatch\\n #xx= raw_input('press enter') \\n\\n\\n bestMatchAtEachTime = highestScores\\n bestMatchOverall = (highestScore, bestMatch)\\n\\n # the following lines do not work well since oftentimes we have multiple pointers (names) to a single object/function\\n # and so they are replaced by returning the function name in the function itself\\n #allvars = vars()\\n #matching_algorithm_name = [v for v in allvars if allvars[v]==matchingAlgorithm][0]\\n #mv_folder = summary_folder + matching_algorithm_name + '/'\\n mv_folder = summary_folder + matching_algorithm_name + '/' # added 2013-09-27\\n\\n result= {'best matches' : bestMatchAtEachTime, \\n 'best overall match' : bestMatchOverall, \\n 'all scores' : scores,\\n 'matching algorithm' : matching_algorithm_name,\\n 'dataset1[obs]' : observations,\\n 'dataset2[models]' : modelsFolder,\\n 'time spent' : time.time() - t0,\\n 'mv folder' : mv_folder, #added 2013-09-23\\n 'ordering' : ordering, #added 2013-09-23\\n }\\n\\n try:\\n os.makedirs(mv_folder)\\n print 'folder', mv_folder, \\\"created!\\\"\\n except:\\n print 'folder', mv_folder, 'exists!'\\n pickle.dump(result, open(mv_folder+'result.pydump','w'))\\n\\n ############################################################################\\n # to make the videos/collected images\\n # calling armor.dataStreamTools.makeVideo.makeVideo()\\n # 2013-09-23\\n\\n from . import makeVideo as mv\\n \\n try:\\n os.makedirs(mv_folder)\\n print 'folder', mv_folder, \\\"created!\\\"\\n except:\\n print 'folder', mv_folder, 'exists!'\\n # make images\\n if makeSmallImages:\\n print 'making small images'\\n for ds in [ds1] + wrfs:\\n #ds.imageFolder= mv_folder + ds.name + '/'\\n ds.imageFolder= summary_root + ds.name + '/' #added 2013-09-27 to replace the above line - too tired waiting the plots\\n ds.saveImages(drawCoast=True, flipud=True, dpi=smallDpi)\\n print ds.imageFolder\\n else:\\n # just set the paths\\n for ds in [ds1]+wrfs:\\n ds.imageFolder = summary_root + ds.name +'/' # /home/k/ARMOR/data/KONG-REY/summary/WRF01.DBZ/\\n for dbzPattern in ds.list:\\n dbzPattern.imagePath = ds.imageFolder + ds.name + dbzPattern.dataTime + '.png'\\n print dbzPattern.imagePath\\n #ds.setImagePaths() # doesn't for for old pickled objects???\\n\\n\\n # make panels\\n\\n mv.makeVideo(DSS = [ds1]+ wrfs, # [ds0, ds1, ds2, ds3, ds4, ...], a list of armor.pattern.DBZstream objects\\n panel_cols = panel_cols, # number of colums in the panel\\n panel_rows = panel_rows, # no need to be filled\\n #fourcc = cv.CV_FOURCC('F', 'L', 'V', '1'),\\n #fps = defaultFps,\\n extension= '.avi',\\n #fourcc = cv.CV_FOURCC('P', 'I', 'M', '1'),\\n outputFileName =\\\"\\\",\\n outputFolder= mv_folder,\\n sandbox = mv_folder ,\\n saveFrames = True, # saving the frames as images\\n frameDpi = frameDpi,\\n useCV2 = useCV2,\\n useCV = useCV,\\n ordering = ordering, # ordering of the models\\n )\\n ############################################################################\\n # dumping the results\\n try:\\n os.makedirs(summary_folder+ 'pydumps/')\\n except:\\n print 'Folder exists! -- ', summary_folder+ 'pydumps/' \\n\\n pickle.dump(result, open(summary_folder+ 'pydumps/results_' + matching_algorithm_name+ '+' +\\\\\\n key1 + '+' + key2 + str(int(time.time())) + '.pydump', 'w'))\\n\\n \\n\\n print \\\"time spent:\\\", time.time()-t0\\n return result\",\n \"def images_mapped(self):\\n try:\\n return dict([x for x in enumerate(self.images())])\\n except:\\n return None\",\n \"def json(self) -> dict:\\n search = {\\n 'criteria': self.search_criteria,\\n 'matchesExact': {\\n 'avgIndexDiff': self.avg_index_diff(TestSearchResult.MatchType.EXACT.value),\\n 'firstFailIndex': self.fail_index(TestSearchResult.MatchType.EXACT.value),\\n 'missedMatches': self.missed_matches(TestSearchResult.MatchType.EXACT.value),\\n 'resultsApi': self.get_results(\\n TestSearchResult.MatchType.EXACT.value, TestSearchResult.Source.API.value),\\n 'resultsLegacy': self.get_results(\\n TestSearchResult.MatchType.EXACT.value, TestSearchResult.Source.LEGACY.value)\\n },\\n 'matchesSimilar': {\\n 'avgIndexDiff': self.avg_index_diff(TestSearchResult.MatchType.SIMILAR.value),\\n 'firstFailIndex': self.fail_index(TestSearchResult.MatchType.SIMILAR.value),\\n 'missedMatches': self.missed_matches(TestSearchResult.MatchType.SIMILAR.value),\\n 'resultsApi': self.get_results(\\n TestSearchResult.MatchType.SIMILAR.value, TestSearchResult.Source.API.value),\\n 'resultsLegacy': self.get_results(\\n TestSearchResult.MatchType.SIMILAR.value, TestSearchResult.Source.LEGACY.value)\\n },\\n 'runTime': self.run_time,\\n }\\n\\n search['matchesExact']['passed'] = (\\n len(search['matchesExact']['missedMatches']) == 0 and\\n search['matchesExact']['firstFailIndex'] == -1\\n )\\n search['matchesSimilar']['passed'] = (\\n len(search['matchesSimilar']['missedMatches']) == 0 and\\n search['matchesSimilar']['firstFailIndex'] == -1\\n )\\n\\n return search\",\n \"def match(image1,image2,threshold,useRansac=False,t_orientation=30,t_scale=0.5):\\r\\n im1, keypoints1, descriptors1 = ReadKeys(image1)\\r\\n im2, keypoints2, descriptors2 = ReadKeys(image2)\\r\\n #\\r\\n # REPLACE THIS CODE WITH YOUR SOLUTION (ASSIGNMENT 5, QUESTION 3)\\r\\n #\\r\\n # Generate five random matches (for testing purposes)\\r\\n # matched_pairs = []\\r\\n # num = 5\\r\\n # for i in range(num):\\r\\n # matched_pairs.append([keypoints1[i],keypoints2[i]])\\r\\n # return DisplayMatches(im1, im2, matched_pairs)\\r\\n\\r\\n # END OF SECTION OF CODE TO REPLACE\\r\\n #\\r\\n\\r\\n #q3\\r\\n matched_pairs = []\\r\\n between_angles = np.arccos(np.dot(descriptors1, np.transpose(descriptors2)))\\r\\n for i, row in enumerate(between_angles):\\r\\n \\tratio = sorted(row)[0] / sorted(row)[1]\\r\\n \\tif ratio <= threshold:\\r\\n\\t \\tmatched_pairs.append([keypoints1[i], keypoints2[np.where(row == sorted(row)[0])[0][0]]])\\r\\n # print(matched_pairs)\\r\\n if useRansac is False:\\r\\n return DisplayMatches(im1, im2, matched_pairs)\\r\\n\\t# \\r\\n\\r\\n #q4\\r\\n repetition = 10\\r\\n subsets = [[]] * repetition\\r\\n for i in range(repetition):\\r\\n r = random.randint(0, len(matched_pairs))\\r\\n for match in matched_pairs:\\r\\n ds1, ds2 = matched_pairs[r][1][2]/matched_pairs[r][0][2], match[1][2]/match[0][2]\\r\\n do1, do2 = (matched_pairs[r][1][3]-matched_pairs[r][0][3]), (match[1][3]-match[0][3])\\r\\n if abs(ds2 - ds1) <= t_scale * ds1 and abs(do2 - do1) % (2 * math.pi) <= t_orientation:\\r\\n subsets[i].append(match)\\r\\n\\r\\n max_i, max_len = 0, subsets[0]\\r\\n for i in range(10):\\r\\n l = len(subsets[i])\\r\\n if l > max_len:\\r\\n max_len = l\\r\\n max_i = i\\r\\n\\r\\n im3 = DisplayMatches(im1, im2, subsets[max_i])\\r\\n return im3\",\n \"def __processFollowerResults(self):\\r\\n\\r\\n resultCacheFileName = os.path.join(core.FW_conf['test_result_dir'], os.path.split(core.FW_conf['test_result_dir'])[1] + \\\".cache\\\" )\\r\\n if os.path.isfile(resultCacheFileName):\\r\\n resultDict = cPickle.load(open(resultCacheFileName, 'rb'))\\r\\n debug.brf('Loading existing results from cache...')\\r\\n else:\\r\\n resultDict = {}\\r\\n\\r\\n # read which iteration this was\\r\\n if not resultDict.has_key(self.tcId):\\r\\n resultDict[self.tcId] = {}\\r\\n iteration = len(resultDict[self.tcId]) + 1\\r\\n resultDict[self.tcId][iteration] = {} # new iteration for the result dictionary\\r\\n\\r\\n leaderResult = Dictionary[str, str]()\\r\\n res = self.result.getResult()\\r\\n\\r\\n try:\\r\\n if str(res) == 'Failed':\\r\\n errornode = self.result.getError()\\r\\n if errornode != None and errornode.getAttribute('reason') == 'X-Files generated during testcase execution' or \\\\\\r\\n errornode.getAttribute('reason') == 'Phone reset detected':\\r\\n leaderResult.Add('Result', 'X-files')\\r\\n elif errornode != None:\\r\\n leaderResult.Add('Result', str(errornode.getAttribute('reason')))\\r\\n except:\\r\\n debug.err('error reading result node')\\r\\n leaderResult.Add('Result', str(res))\\r\\n\\r\\n if not leaderResult.ContainsKey('Result'): # if result was not added\\r\\n leaderResult.Add('Result', str(res))\\r\\n\\r\\n folder = os.path.split(core.FW_conf['test_result_dir'])[1]\\r\\n\\r\\n leaderResult.Add('ResultDir', folder)\\r\\n leaderResult.Add('Duration', str( int( time.time() - self.leader_start_time )) )\\r\\n #leaderResult.Add('EndTime', strftime(\\\"%Y-%m-%d %H:%M:%S\\\", gmtime()))\\r\\n leaderResult.Add('EndTime', DateTime.Now.ToString(\\\"yyyy-MM-dd HH:mm:ss\\\"))\\r\\n for i in range(1, len(self._capturedStillImages) + 1 ):\\r\\n debug.out(self._capturedStillImages[i-1])\\r\\n relativeFileName = os.path.join('videos',self._capturedStillImages[i-1])\\r\\n leaderResult.Add('Capture%s' % str(i), relativeFileName)\\r\\n\\r\\n try:\\r\\n relativeFileName = os.path.join('videos',os.path.splitext(core.FW_conf['test_result_name'])[0] + '.avi')\\r\\n videoFile = os.path.join(core.FW_conf['test_result_dir'], relativeFileName)\\r\\n if os.path.isfile(videoFile): # change video address if video file is found\\r\\n leaderResult.Add('Video', str(relativeFileName))\\r\\n #debug.brf('Added leader Video: %s' % str(videoFile))\\r\\n except Exception as e:\\r\\n debug.brf('Could not add Leader video to result: %s' % e.message)\\r\\n\\r\\n leaderResult.Add('Machine', str(Dns.GetHostName())) # add Leader machine name\\r\\n\\r\\n followerResults = core.FW_conf['leader'].GetResults()\\r\\n followerResults.Add('Leader', leaderResult) # Add leader result to dictionary\\r\\n\\r\\n for followerRes in followerResults:\\r\\n clientId = followerRes.Key # string client id\\r\\n result = followerRes.Value # dictionary of result, link\\r\\n\\r\\n if not resultDict[self.tcId][iteration].has_key(clientId):\\r\\n resultDict[self.tcId][iteration][clientId] = {}\\r\\n else:\\r\\n debug.err('\\\\n**** %s has already result for iteration %s on test case %s. Will not override.\\\\n ***' % (clientId, iteration, self.tcId))\\r\\n continue\\r\\n\\r\\n # loop through result\\r\\n debug.brf('Results for client %s' % clientId)\\r\\n for res in result:\\r\\n debug.brf('Adding %s %s' % (str(res.Key), str(res.Value)) )\\r\\n resultDict[self.tcId][iteration][clientId][str(res.Key)] = str(res.Value)\\r\\n\\r\\n if resultDict:\\r\\n # dump resultDict to cache\\r\\n try:\\r\\n cacheFile = open(resultCacheFileName, 'wb')\\r\\n cPickle.dump(resultDict, cacheFile, 0)\\r\\n cacheFile.close()\\r\\n except:\\r\\n for i in range(10):\\r\\n debug.err('*' * 40 + ' FAILED TO FLUSH RESULTS TO DISK! ' + '*' * 40)\",\n \"def getAllForImages(self):\\n imageDict = {}\\n for id, name in self.getAll().items():\\n imageDict[id] = {}\\n imageDict[id][\\\"name\\\"] = name\\n imageDict[id][\\\"filename\\\"] = \\\"The_Steamer_Great_Western_small.jpg\\\"\\n\\n return imageDict\",\n \"def _calculate_distribution(self, images):\\n positions = {}\\n target_width = target_height = 0\\n for block, url in self.image_declarations:\\n file_name = normalize_filename(url)\\n if file_name in positions:\\n positions[file_name]['blocks'].append(block)\\n else:\\n block_images = images[file_name]\\n offset_x = target_width\\n offset_y = 0\\n\\n width, height = block_images[1].size\\n positions[file_name] = {\\n 'x': offset_x, 'y': offset_y,\\n 'width': width, 'height': height,\\n 'file_name': file_name, 'images': block_images,\\n 'blocks': [block]\\n }\\n\\n target_width += width + self.space_x\\n target_height = max(target_height, height)\\n\\n target_width -= self.space_x\\n return {\\n 'width': target_width,\\n 'height': target_height,\\n 'positions': positions.values(),\\n }\",\n \"def compute_similarity(im1, im2, test_id):\\n\\n results = dict()\\n results['id'] = test_id\\n results['test_im'] = im2\\n results['mse_value'], results['mse_map'] = compute_mse(im1, im2)\\n results['ssim_value'], results['ssim_map'] = compute_ssim(im1, im2, 5)\\n results['cw_ssim_value'], results['cw_ssim_map'] = compute_cw_ssim(im1, im2, 30)\\n results['gms_value'], results['gms_map'] = compute_gms(im1, im2)\\n results['fsim_value'], results['pc_max_map'] = compute_fsim(im1, im2)\\n\\n return results\",\n \"def _build_results(self):\\n results = {}\\n cols = []\\n for pol in POLLUTANTS:\\n for adj in ADJUSTMENTS:\\n cols.append(get_rate_column(pol, adjustment=adj, generated=False))\\n cols.append(get_column(pol, adjustment=adj))\\n cols.append(\\\"net_consumed_mwh\\\")\\n for ba in self.regions:\\n results[ba] = pd.DataFrame(\\n index=self.generation.index, columns=cols, dtype=np.float64\\n )\\n return results\",\n \"def get_images(self):\\n return {'source': utils.image_from_tensor(self.source[0]),\\n 'output': utils.image_from_tensor(self.output.data[0]),\\n 'target': utils.image_from_tensor(self.target[0])}\",\n \"def run(self):\\n return_dict = {}\\n for x in self.file_list:\\n return_dict[x] = self.parseFile(x)\\n\\n return return_dict\",\n \"def maps(matches):\\n if not matches:\\n print('no matches')\\n return None\\n\\n maps = {}\\n for match in matches:\\n map = match['map']\\n score = match['score']\\n result = match['result']\\n # add map to dictionary\\n if not map in maps:\\n maps[map] = {'wins': 0, 'losses': 0, 'winrate': 0.0}\\n map_data = maps[map]\\n if result:\\n map_data['wins'] += 1\\n else:\\n map_data['losses'] += 1\\n matches_played = map_data['wins'] + map_data['losses']\\n\\n # compute moving average on winrate\\n map_data['winrate'] = map_data['winrate'] + (result - map_data['winrate']) / matches_played\\n return maps\",\n \"def getList(self):\\n labelMap = {}\\n imageMap = {}\\n key = []\\n index = 0\\n\\n for root, dirs, files in os.walk(self.path_data):\\n for file in files:\\n # If .png or .jpg file found then\\n if file.endswith(tuple(config.imageFormat)):\\n key.append(index)\\n labelMap[index] = preprocessing.getLabel(file)\\n imageMap[index] = os.path.join(root, file)\\n\\n index += 1\\n\\n else:\\n continue\\n\\n return key, imageMap, labelMap\",\n \"def get_results(self):\\n if not self.sim_stop_flag.value:\\n # we are only interested in entries of frames processed > 0\\n max_index = np.sum(np.array(self.sim_results_struct.frames[0:50]) > 0)\\n\\n # convert the cstruct to dictionary\\n return dict([(x, getattr(self.sim_results_struct, x)[0:max_index]) for (x,_) in self.sim_results_struct._fields_])\\n else:\\n return self.results\",\n \"def get_images_and_labels_nc():\\n refs = get_ref_df()\\n images = {}\\n for _, data in refs.iterrows():\\n if data['ProbeFileName'] in images:\\n continue\\n im = data['ProbeFileName']\\n images[im] = 1 if data['IsTarget'] == 'Y' else 0\\n return images\",\n \"def prepare_results(self) -> dict:\\n if not hasattr(self, \\\"results\\\"):\\n raise AttributeError(\\n \\\"Results have not been finalized. Please call \\\"\\n \\\"finalize_results() before saving output.\\\"\\n )\\n\\n output = {\\n \\\"armory_version\\\": armory.__version__,\\n \\\"config\\\": self.config,\\n \\\"results\\\": self.results,\\n \\\"timestamp\\\": int(self.time_stamp),\\n }\\n return output\",\n \"def __call__(self, results):\\n\\n if random.uniform(0, 1) > self.prob:\\n return results\\n\\n if 'img_fields' in results:\\n assert results['img_fields'] == ['img'], \\\\\\n 'Only single img_fields is allowed'\\n img = results['img']\\n\\n h, w, c = img.shape\\n ratio = random.uniform(self.min_ratio, self.max_ratio)\\n # speedup expand when meets large image\\n if np.all(self.mean == self.mean[0]):\\n expand_img = np.empty((int(h * ratio), int(w * ratio), c),\\n img.dtype)\\n expand_img.fill(self.mean[0])\\n else:\\n expand_img = np.full((int(h * ratio), int(w * ratio), c),\\n self.mean,\\n dtype=img.dtype)\\n left = int(random.uniform(0, w * ratio - w))\\n top = int(random.uniform(0, h * ratio - h))\\n expand_img[top:top + h, left:left + w] = img\\n\\n results['img'] = expand_img\\n # expand bboxes\\n for key in results.get('bbox_fields', []):\\n results[key] = results[key] + np.tile(\\n (left, top), 2).astype(results[key].dtype)\\n\\n # expand masks\\n for key in results.get('mask_fields', []):\\n results[key] = results[key].expand(\\n int(h * ratio), int(w * ratio), top, left)\\n\\n # expand segs\\n for key in results.get('seg_fields', []):\\n gt_seg = results[key]\\n expand_gt_seg = np.full((int(h * ratio), int(w * ratio)),\\n self.seg_ignore_label,\\n dtype=gt_seg.dtype)\\n expand_gt_seg[top:top + h, left:left + w] = gt_seg\\n results[key] = expand_gt_seg\\n return results\",\n \"def compute(\\n self,\\n ) -> Dict[str, Tensor]:\\n results = torch.stack((self.tp, self.fp, self.tn, self.fn), dim=1)\\n\\n return {f\\\"group_{i}\\\": group / group.sum() for i, group in enumerate(results)}\",\n \"def result_dict_for_single_example(image, orig_shape,\\n key,\\n detections,\\n groundtruth):\\n label_id_offset = 1 # Applying label id offset (b/63711816)\\n\\n input_data_fields = fields.InputDataFields\\n output_dict = {\\n input_data_fields.original_image: image,\\n input_data_fields.key: key,\\n }\\n\\n detection_fields = fields.DetectionResultFields\\n\\n detection_boxes = detections[detection_fields.detection_boxes]\\n detection_scores = detections[detection_fields.detection_scores]\\n\\n detection_classes = (\\n detections[detection_fields.detection_classes] +\\n label_id_offset)\\n\\n num_detections = int(detections[detection_fields.num_detections])\\n detection_boxes = detection_boxes[:num_detections]\\n detection_scores = detection_scores[:num_detections]\\n detection_classes = detection_classes[:num_detections]\\n detection_boxes = toAbsolute(detection_boxes, orig_shape)\\n\\n output_dict[detection_fields.detection_boxes] = detection_boxes\\n output_dict[detection_fields.detection_classes] = detection_classes\\n output_dict[detection_fields.detection_scores] = detection_scores\\n \\n if groundtruth:\\n output_dict.update(groundtruth)\\n \\n return output_dict\",\n \"def find(image):\\n keypoint, description = describe(image)\\n # load keypoints, descriptions from mongodb\\n\\n bf = cv2.BFMatcher(cv2.NORM_HAMMING, crossCheck=True)\\n\\n best_match_size = float(\\\"inf\\\")\\n best_match_index = -1\\n match_index = 0\\n best_matches = 0\\n\\n for desc in descriptions:\\n matches = bf.match(desc,description)\\n matches = sorted(matches, key = lambda x:x.distance)\\n if len(matches) > 0:\\n match_size = sum(x.distance for x in matches[:10])\\n\\n print \\\"match size is \\\", match_size\\n if match_size < best_match_size:\\n best_match_size = match_size\\n best_match_index = match_index\\n best_matches = matches\\n\\n match_index += 1\\n\\n needle_color = cv2.imread('needle-stripped.png')[:,:,::-1] # needle\\n best_match_image = cv2.imread(\\\"haystack/\\\"+files[best_match_index])\\n print \\\"best match is \\\", files[best_match_index]\\n\\n # Draw first 10 matches.\\n outImg = cv2.imread(\\\"output/outImg.png\\\")\\n match = cv2.drawMatches(needle_color,keypoint,best_match_image[:,:,::-1],keypoints[best_match_index],best_matches[-20:],outImg, flags=6)\\n\\n plt.imshow(match),plt.show()\\n return\",\n \"def getResults():\",\n \"def get_pet_labels(image_dir):\\n results_dic = dict()\\n \\n# # Retrieves the file names from the folder specified as 'image_dir' \\n filenames_list = listdir(image_dir)\\n \\n# # Processes the filenames to create the pet image labels\\n# # Retrieves the filenames from folder pet_images/\\n for i in range (0, len(filenames_list), 1):\\n# # Skips file if starts with . (like .DS_Store of Mac OSX) because it \\n# # isn't an pet image file\\n if filenames_list[i][0] != \\\".\\\":\\n# # Reads respectively indexed element from filenames_list into temporary string variable 'pet_image' \\n pet_image = filenames_list[i]\\n# # Sets all characters in 'pet_image' to lower case \\n pet_image_lower = pet_image.lower()\\n# # Creates list called 'pet_image_word_list' that contains every element in pet_image_lower seperated by '_'\\n pet_image_word_list = pet_image_lower.split(\\\"_\\\")\\n# # Creates temporary variable 'pet_label' to hold pet label name extracted starting as empty string\\n pet_image_alpha = \\\"\\\"\\n# # Iterates through every word in 'pet_image_word_list' and appends word to 'pet_label_alpha' only if word consists \\n# # purely of alphabetic characters \\n for word in pet_image_word_list:\\n if word.isalpha():\\n pet_image_alpha += word + \\\" \\\"\\n# # Removes possible leading or trailing whitespace characters from 'pet_pet_image_alpha' and add stores final label as 'pet_label' \\n pet_label = pet_image_alpha.strip()\\n\\n# # Adds the original filename as 'key' and the created pet_label as 'value' to the 'results_dic' dictionary if 'key' does \\n# # not yet exist in 'results_dic', otherwise print Warning message \\n if filenames_list[i] not in results_dic:\\n results_dic[filenames_list[i]] = [pet_label]\\n else:\\n print(\\\"** Warning: Key = \\\", filenames_list[i], \\\" already in 'results_dic' with value = \\\", results_dic[filenames_list[i]])\\n \\n# # Iterates through the 'results_dic' dictionary and prints its keys and their associated values\\n print(\\\"\\\\nPrinting: All 'key' - 'value' pairs in dictionary results_dic: \\\")\\n for key in results_dic:\\n print(\\\"Filename = \\\", key, \\\" Pet Label = \\\", results_dic[key])\\n \\n# # Returns results_dic\\n return results_dic\",\n \"def create_observations_dict(instances, unprocessed_map_data):\\n # print(instances)\\n results = {}\\n for idx in instances:\\n results[idx] = {}\\n if idx == instances[0]:\\n tag_filter_pixel_corners = B.compute_corner_pixels(\\n idx, unprocessed_map_data\\n ).tolist()\\n # prettified_corner_pixels = [tag_filter_pixel_corners[:2,0],tag_filter_pixel_corners[:2,1],tag_filter_pixel_corners[:2,2],tag_filter_pixel_corners[:2,3]]\\n # results[idx][\\\"corner_pixels\\\"] = [pixel_pair.tolist() for pixel_pair in prettified_corner_pixels]\\n results[idx][\\\"corner_pixels\\\"] = tag_filter_pixel_corners\\n results[idx][\\\"tag_pose\\\"] = B.compute_tag_pose(\\n idx, unprocessed_map_data\\n ).tolist()\\n results[idx][\\\"camera_pose\\\"] = B.compute_camera_pose(\\n idx, unprocessed_map_data\\n ).tolist()\\n\\n return results\",\n \"def cs4243_histmatch(ori_image, refer_image):\\n \\n ##your code here ###\\n\\n # get cdf of ori and ref image\\n grey_level = 256\\n ori_hist, ori_cum_hist, ori_res_image, ori_uni_hist = cs4243_histequ(ori_image, grey_level)\\n ref_hist, ref_cum_hist, ref_res_image, ref_uni_hist = cs4243_histequ(refer_image, grey_level)\\n \\n # map each ori cdf to ref cdf and get the mapped index as matched grey level\\n map_value = []\\n for i in range(grey_level):\\n ori_cdf = ori_cum_hist[i]\\n matched_intensity = np.uint8(np.abs(ref_cum_hist - ori_cdf).argmin())\\n map_value.append(matched_intensity)\\n ##\\n\\n # Set the intensity of the pixel in the raw image to its corresponding new intensity \\n height, width = ori_image.shape\\n res_image = np.zeros(ori_image.shape, dtype='uint8') # Note the type of elements\\n for i in range(height):\\n for j in range(width):\\n res_image[i,j] = map_value[ori_image[i,j]]\\n \\n res_hist = np.bincount(res_image.flatten(), minlength=256)\\n \\n return ori_hist, ref_hist, res_image, res_hist\",\n \"def get_entry_dict(self):\\n\\n # generating thumbnail URLs is slow, so only generate the ones\\n # that will definitely be used.\\n ret = {\\n 'id': self.id,\\n 'vertices': self.vertices,\\n 'triangles': self.triangles,\\n 'segments': self.segments,\\n 'photo': self.photo.get_entry_dict(),\\n }\\n if self.dominant_rgb0:\\n ret['dominant_rgb0'] = self.dominant_rgb0\\n #if self.image_pbox:\\n #ret['pbox'] = self.pbox\\n #ret['image_pbox'] = {\\n #'300': self.image_pbox_300.url,\\n #'512': self.image_pbox_512.url,\\n #'1024': self.image_pbox_1024.url,\\n #'orig': self.image_pbox.url,\\n #}\\n if self.image_bbox:\\n ret['image_bbox'] = {\\n #'512': self.image_bbox_512.url,\\n '1024': self.image_bbox_1024.url,\\n #'orig': self.image_bbox.url,\\n }\\n return ret\",\n \"def get_results(self):\\n d = {}\\n# r = {}\\n for analyser in self.xml_tree.getroot():\\n for child in analyser:\\n if child.tag == 'all-records':\\n for record in child:\\n attributes = record.attrib\\n sample = attributes['sampleId']\\n assay_id = attributes['assayId']\\n genotype = attributes['genotypeId']\\n quality = attributes['description'].split('.')[0]\\n if re.match(r'rs\\\\d+', assay_id):\\n if sample in d:\\n if assay_id in d[sample]:\\n for allele in list(genotype):\\n if allele not in d[sample][assay_id]['genotype']:\\n d[sample][assay_id]['genotype'] += allele\\n if quality not in d[sample][assay_id]['quality']:\\n d[sample][assay_id]['quality'].append(quality)\\n else:\\n d[sample][assay_id] = {'genotype': genotype, 'quality': [quality]}\\n else:\\n d[sample] = {assay_id: {'genotype': genotype, 'quality': [quality]}}\\n# if sample in r:\\n# if assay_id in r[sample]:\\n# for allele in list(genotype):\\n# if allele not in r[sample][assay_id]:\\n# r[sample][assay_id] += allele\\n# else:\\n# r[sample][assay_id] = genotype\\n# else:\\n# r[sample] = {assay_id: genotype}\\n# for k, v in r.items():\\n# for k1, v1, in v.items():\\n# if len(v1) == 1:\\n# v[k1] += v1\\n# pprint.pprint(r)\\n# df = pd.DataFrame.from_dict(r).transpose()\\n# print(df)\\n# df.to_excel('snpcheck.xlsx')\\n return d\",\n \"def get_image_list(im, dic):\\n lst = list(im.getdata())\\n tiles = []\\n for i in range(len(lst)):\\n #print find_similar(lst[i], dic)[random.randrange(10)][1]\\n tiles.append(find_similar(lst[i], dic)[random.randrange(10)][1])\\n return tiles\",\n \"def _load_images(self, resolutions=None):\\n images = {}\\n\\n for block, url in self.image_declarations:\\n file_name = normalize_filename(url)\\n if file_name not in images:\\n img_resolutions = {}\\n img = Image.open(file_name)\\n img_resolutions[1] = img\\n width, height = img.size\\n\\n if resolutions:\\n for resolution in resolutions:\\n # Get the correct filename for this resolution\\n if resolution != 1:\\n root, ext = os.path.splitext(file_name)\\n res_file_name = '{root}-{resolution}x{ext}'.format(\\n root=root, resolution=resolution, ext=ext)\\n\\n img = Image.open(res_file_name)\\n if img.size[0] / resolution != width:\\n raise ValueError('Invalid width for {0}'.format(\\n res_file_name))\\n if img.size[1] / resolution != height:\\n raise ValueError('Invalid height for {0}'.format(\\n res_file_name))\\n img_resolutions[resolution] = img\\n\\n images[file_name] = img_resolutions\\n\\n return images\",\n \"def get_images_to_build(fuzzers, benchmarks):\\n images = {}\\n templates = _get_image_type_templates()\\n for fuzzer in fuzzers:\\n for benchmark in benchmarks:\\n for name_templ, obj_templ in templates.items():\\n name, obj = _instantiate_image_obj(name_templ, obj_templ,\\n fuzzer, benchmark)\\n images[name] = obj\\n return images\",\n \"def fetch_images(self, images, depth_level):\\r\\n image_results = {}\\r\\n initial_area = float(0.0)\\r\\n total_score = float(0.0)\\r\\n cnt = float(1.0)\\r\\n MIN_WIDTH = 50\\r\\n for image in images[:30]:\\r\\n src = self.parser.getAttribute(image, attr='src')\\r\\n src = self.build_image_path(src)\\r\\n local_image = self.get_local_image(src)\\r\\n width = local_image.width\\r\\n height = local_image.height\\r\\n src = local_image.src\\r\\n file_extension = local_image.file_extension\\r\\n\\r\\n if file_extension != '.gif' or file_extension != 'NA':\\r\\n if (depth_level >= 1 and local_image.width > 300) or depth_level < 1:\\r\\n if not self.is_banner_dimensions(width, height):\\r\\n if width > MIN_WIDTH:\\r\\n sequence_score = float(1.0 / cnt)\\r\\n area = float(width * height)\\r\\n total_score = float(0.0)\\r\\n\\r\\n if initial_area == 0:\\r\\n initial_area = area * float(1.48)\\r\\n total_score = 1\\r\\n else:\\r\\n area_difference = float(area / initial_area)\\r\\n total_score = sequence_score * area_difference\\r\\n\\r\\n image_results.update({local_image: total_score})\\r\\n cnt += 1\\r\\n cnt += 1\\r\\n return image_results\",\n \"def getPatternDict(self):\\n c = \\\"/cli:python /app:matrix /cmd:getinfo /dev:patternlist\\\"\\n self.sendCMDstring(c)\\n time.sleep(self.delay)\\n answers = self.readandparseCAM()\\n patternlist = {}\\n for a in answers:\\n if a['dev']=='patternlist':\\n for i in range(int(a['count'])):\\n nr = a['patternid' +str(i+1)]\\n name = a['patternname' +str(i+1)].lower()\\n patternlist[name]=nr\\n else:\\n print \\\"no patternlist in answers\\\"\\n return patternlist\",\n \"def init_output_dict(self):\\n return {\\n \\\"outputs\\\": torch.FloatTensor(),\\n \\\"pred_probs\\\": torch.FloatTensor(),\\n \\\"labels\\\": torch.LongTensor(),\\n }\",\n \"def _extract_results(self) -> None:\\n metric_name = self.metric.name\\n for inference_name in ['train', 'test', 'opt']:\\n # TODO: Extract information from self.search_results\\n data = getattr(self.search_results, f'{inference_name}_metric_dict')[metric_name]\\n if all([d is None for d in data]):\\n if inference_name not in OPTIONAL_INFERENCE_CHOICES:\\n raise ValueError(f\\\"Expected {metric_name} score for {inference_name} set\\\"\\n f\\\" to not be None, but got {data}\\\")\\n else:\\n continue\\n self.data[f'single::{inference_name}::{metric_name}'] = np.array(data)\\n\\n if self.ensemble_results.empty() or inference_name == 'opt':\\n continue\\n\\n data = getattr(self.ensemble_results, f'{inference_name}_scores')\\n if all([d is None for d in data]):\\n if inference_name not in OPTIONAL_INFERENCE_CHOICES:\\n raise ValueError(f\\\"Expected {metric_name} score for {inference_name} set\\\"\\n f\\\" to not be None, but got {data}\\\")\\n else:\\n continue\\n self.data[f'ensemble::{inference_name}::{metric_name}'] = np.array(data)\",\n \"def compute_metrics(self, results: list) -> Dict[str, float]:\\n logger: MMLogger = MMLogger.get_current_instance()\\n\\n # pred_coords: [N, K, D]\\n pred_coords = np.concatenate(\\n [result['pred_coords'] for result in results])\\n if pred_coords.ndim == 4 and pred_coords.shape[1] == 1:\\n pred_coords = np.squeeze(pred_coords, axis=1)\\n # gt_coords: [N, K, D]\\n gt_coords = np.stack([result['gt_coords'] for result in results])\\n # mask: [N, K]\\n mask = np.concatenate([result['mask'] for result in results])\\n # action_category_indices: Dict[List[int]]\\n action_category_indices = defaultdict(list)\\n for idx, result in enumerate(results):\\n action_category = result['action'].split('_')[0]\\n action_category_indices[action_category].append(idx)\\n\\n error_name = self.mode.upper()\\n\\n logger.info(f'Evaluating {self.mode.upper()}...')\\n metrics = dict()\\n\\n metrics[error_name] = keypoint_mpjpe(pred_coords, gt_coords, mask,\\n self.ALIGNMENT[self.mode])\\n\\n for action_category, indices in action_category_indices.items():\\n metrics[f'{error_name}_{action_category}'] = keypoint_mpjpe(\\n pred_coords[indices], gt_coords[indices], mask[indices])\\n\\n return metrics\",\n \"def match_patterns(training_data, num_languages):\\n\\n ret = {}\\n\\n for index1, pattern in enumerate(shaman.PatternMatcher.PATTERNS):\\n print('Matching pattern %d \\\"%s\\\"' % (index1+1, pattern))\\n\\n matcher = shaman.PatternMatcher(pattern)\\n ratios_per_lang = [[] for _ in range(num_languages)]\\n\\n for index2, (lang_id, code) in enumerate(training_data):\\n if len(code) <= 20 or len(code) > 100000:\\n continue\\n\\n ratio = matcher.getratio(code)\\n ratios_per_lang[lang_id].append(ratio)\\n\\n print('Matching patterns %d/%d ' %\\n (index2, len(training_data)), end='\\\\r')\\n\\n ret[pattern] = [[] for _ in range(num_languages)]\\n for lang_id, ratio_list in enumerate(ratios_per_lang):\\n ret[pattern][lang_id] = sum(ratio_list) / max(len(ratio_list), 1)\\n\\n print('Matching patterns completed ')\\n return ret\",\n \"def buildRunDict(self):\\n self.showProgressBar()\\n ori_images = 0\\n if self.img_exist:\\n ori_images = len(listDirectory(self.savePathJoin(\\\"Images\\\")))\\n self.buildRunDictMain(ori_images)\\n else:\\n self.run_dict[\\\"Video\\\"] = {\\n \\\"Run\\\": True,\\n \\\"Progress\\\": ori_images,\\n \\\"Text\\\": \\\"Preparing video\\\",\\n }\\n self.buildParamsDict()\\n self.params_dict[\\\"send_video_frame\\\"] = True\\n\\n self.progressLabel.setText(\\\"Create images from video\\\")\\n\\n self.worker = calcRunner.CalculationRunner(self.params_dict) # no parent!\\n self.thread = QThread() # no parent!\\n\\n self.worker.labelUpdate.connect(self.labelUpdate)\\n\\n self.worker.update.connect(self.progressUpdate)\\n self.worker.videoFrame.connect(self.setVidFrame)\\n\\n self.worker.moveToThread(self.thread)\\n self.thread.started.connect(self.worker.startThread)\\n self.thread.start()\",\n \"def calculate_MAP(self):\\n testing_images = open('./digitdata/testimages', 'r')\\n with testing_images as ti:\\n data = list(csv.reader(ti))\\n data = [i for i in data if i]\\n count = 0\\n #loop through all the test images\\n for j in range(0,1000):\\n classification_dict = {0:0,1:0,2:0,3:0,4:0,5:0,6:0,7:0,8:0,9:0} \\n for l in range(0,28):\\n coord = count + l\\n for w in range(0,28):\\n if data[coord][0][w] == \\\"+\\\":\\n #iterate through each class. z is the class [0-9]\\n for z in range(0,10):\\n classification_dict[z] += math.log(self.class_probabilities[z][l][w][0]) \\n elif data[coord][0][w] == \\\"#\\\":\\n for z in range(0,10):\\n classification_dict[z] += math.log(self.class_probabilities[z][l][w][1])\\n elif data[coord][0][w] == \\\" \\\":\\n for z in range(0,10):\\n classification_dict[z] += math.log(self.class_probabilities[z][l][w][2])\\n count += 28\\n self.solutions.append(max(classification_dict, key=classification_dict.get))\",\n \"def create_results_dict(self, n_options):\\n if n_options == 1:\\n self.results_dict[0] = {\\\"configuration\\\": [\\\"user\\\"]}\\n for name in self.results_metadata[\\\"names\\\"]:\\n self.results_dict[0][name] = []\\n else:\\n self.results_dict = {}\\n for i in range(n_options):\\n self.results_dict[i] = {\\\"configuration\\\": []}\\n for name in self.results_metadata[\\\"names\\\"]:\\n self.results_dict[i][name] = []\",\n \"def get_specs(self, results: Dict[str, Dict], params: Dict) -> Dict[str, float] :\\n raise NotImplementedError\",\n \"def predict() -> Any:\\n threshold = request.form.get(\\\"threshold\\\", type=float)\\n source_size = request.form.get(\\\"source_size\\\", type=bool)\\n images = request.files.getlist(\\\"images\\\")\\n result = {}\\n for image in images:\\n input_image = prepare_input(image)\\n if input_image is not None:\\n output_image = model.predict(input_image, threshold, source_size)\\n if output_image is not None:\\n result[image.filename] = prepare_output(output_image)\\n else:\\n result[image.filename] = None\\n else:\\n result[image.filename] = None\\n return result\",\n \"def extractData(self) -> Dict[str, str]:\\n if self.hasScore():\\n result = {}\\n result[\\\"date\\\"] = self.getMatchDate()\\n result[\\\"score\\\"] = self.getMatchScore()\\n for team in self.TEAMS:\\n result[\\\"team_\\\" + team] = self.getSquadName(team)\\n\\n return result\\n return None\",\n \"def buildDicts(resComp, comparisons, matrices, atom1, res1, atom2, res2, r, FLAG=False):\\r\\n if resComp not in matrices:\\r\\n for pairs in comparisons:\\r\\n if resComp[0] == pairs[1] and resComp[1] == pairs[0]:\\r\\n return\\r\\n matrices[resComp] = [r]\\r\\n comparisons[resComp] = [(atom1, res1, atom2, res2, r)]\\r\\n FLAG = True\\r\\n\\r\\n elif type(comparisons[resComp][-1]) == list:\\r\\n if comparisons[resComp][-1][-1][1] != res1 or comparisons[resComp][-1][-1][3] != res2:\\r\\n # Make new matrix\\r\\n raise Warning\\r\\n elif comparisons[resComp][-1][-1][0] == atom1:\\r\\n comparisons[resComp][-1].append((atom1, res1, atom2, res2, r))\\r\\n matrices[resComp][-1].append(r)\\r\\n else:\\r\\n comparisons[resComp].append([(atom1, res1, atom2, res2, r)])\\r\\n matrices[resComp].append([r])\\r\\n\\r\\n elif type(comparisons[resComp][-1]) == tuple:\\r\\n if comparisons[resComp][-1][0] == atom1:\\r\\n comparisons[resComp].append((atom1, res1, atom2, res2, r))\\r\\n matrices[resComp].append(r)\\r\\n else:\\r\\n comparisons[resComp] = [comparisons[resComp], [(atom1, res1, atom2, res2, r)]]\\r\\n matrices[resComp] = [matrices[resComp], [r]]\\r\\n\\r\\n else:\\r\\n print(\\\"Error: Dictionary should contain Match objects / Lists of Match objects / List of lists of Match objects\\\")\\r\\n\\r\\n return comparisons, matrices, resComp, FLAG\",\n \"def _get_output_dictionary(self):\\n\\n return_dictionary = {}\\n\\n for output_path in self.provided_outputs:\\n return_dictionary[output_path.full_path] = self.get_value(output_path)\\n\\n return return_dictionary\",\n \"def testing_on_new_data(directory=parentDir + '/photos-to-categorize'):\\n\\n known_face_encodings={}\\n with open(parentDir + '/data/face_locations.csv', 'r') as csvfile:\\n fieldnames=['person', 'fullfilename' , 'top', 'right', 'bottom', 'left']\\n reader = csv.DictReader(csvfile,fieldnames=fieldnames)\\n for row in reader:\\n face_location = [int(row['top']), int(row['right']), int(row['bottom']), int(row['left'])]\\n if row['person'] in known_face_encodings:\\n known_face_encodings[row['person']] = np.append(known_face_encodings[row['person']],np.array(face_encodings(load_image_file(row['fullfilename']),[face_location])),axis=0)\\n else:\\n known_face_encodings.update({row['person']:np.array(face_encodings(load_image_file(row['fullfilename']),[face_location]))})\\n \\n for filename in os.listdir(directory):\\n img = load_image_file(directory + '/' + filename)\\n \\n all_face_locations = face_locations(img)\\n for face_location in all_face_locations:\\n top,right,bottom,left = face_location\\n face_image = img[top:bottom, left:right]\\n pil_image = Image.fromarray(face_image)\\n \\n test_face_encode = face_encodings(img,[face_location])\\n result={}\\n name='Unknown'\\n minval=1\\n for person in known_face_encodings:\\n r=face_distance(known_face_encodings[person], test_face_encode[0]).mean()\\n if r <= 0.6:\\n result.update({person:(1-r)*100})\\n if r < minval:\\n minval=r\\n name=person\\n \\n Image._show(pil_image)\\n \\n print filename + \\\": The person is \\\" + name\\n print \\\"All possiblities: \\\" + str(result)\\n\\n #writting in a file\\n with open(parentDir + '/data/image_categorizations.csv', 'a') as csvfile:\\n fieldnames=['person', 'fullfilename' , 'top', 'right', 'bottom', 'left']\\n writer = csv.DictWriter(csvfile, fieldnames=fieldnames)\\n if name != 'Unknown': writer.writerow({'person':name, 'fullfilename': directory + '/' + filename, 'top':top, 'right':right, 'bottom':bottom, 'left':left})\\n \\n\\n #raw_input(\\\"continue?\\\")\",\n \"def _add_matches(self):\\r\\n for record in self.records:\\r\\n match_dict={key_to_track: record.get(key_to_track)\\r\\n for key_to_track in self.key_matcher.keys()}\\r\\n self.key_matcher.add(obj=record,\\r\\n match_dict=match_dict)\",\n \"def get_imgid_dict(ann):\\n return {item[1][\\\"file_name\\\"]: item[0] for item in ann.imgs.items()}\",\n \"def _results(self):\\n results = {}\\n results[\\\"coeff\\\"] = self._coeff_to_dict()\\n results[\\\"coeff_path\\\"] = dict(\\n zip(\\n [f\\\"{col}\\\" for col in self.X_train_.columns.tolist()],\\n (\\n self.model_.coef_path_.reshape(-1, self.model_.coef_path_.shape[-1])\\n ).tolist(),\\n )\\n )\\n results[\\\"cv_standard_error\\\"] = self.model_.cv_standard_error_.tolist()\\n results[\\\"cv_mean_score\\\"] = self.model_.cv_mean_score_.tolist()\\n results[\\\"lambda_path\\\"] = self.model_.lambda_path_.tolist()\\n results[\\\"lambda_best\\\"] = self.model_.lambda_best_[0]\\n results[\\\"lambda_max\\\"] = self.model_.lambda_max_\\n results[\\\"n_lambda\\\"] = self.model_.n_lambda_\\n results[\\\"intercept\\\"] = self.model_.intercept_\\n results[\\\"intercept_path\\\"] = self.model_.intercept_path_.tolist()[0]\\n results[\\\"params\\\"] = self.model_.get_params()\\n results[\\\"module\\\"] = self.model_.__module__\\n\\n return results\",\n \"def calculate_individual_image_result(\\n prediction_boxes, gt_boxes, iou_threshold):\\n # Find the bounding box matches with the highes IoU threshold\\n matched_prediction_boxes, matched_gt_boxes = get_all_box_matches(\\n prediction_boxes,\\n gt_boxes,\\n iou_threshold)\\n \\n # Compute true positives, false positives, false negatives\\n num_tp = len(matched_prediction_boxes)\\n num_fp = len(prediction_boxes) - num_tp\\n num_fn = np.maximum(len(gt_boxes) - len(matched_gt_boxes), 0, dtype=int)\\n\\n return {\\\"true_pos\\\": num_tp, \\\"false_pos\\\": num_fp, \\\"false_neg\\\": num_fn}\",\n \"def __make_processing(self, img_name, abspath_dir_img, id_foot):\\n data = {}\\n data['data'] = ImageInfo.get_date(abspath_dir_img)\\n data['total_part'] = TOTAL_PART\\n data['nuvens'] = ImageInfo.get_cloud(abspath_dir_img)\\n self.__make_tms(abspath_dir_img)\\n data['geom'] = self.__make_footprint(abspath_dir_img, shp_out=id_foot)\\n abspath_rgb, img_name_rgb = ImageInfo.get_image_rgb(\\n abspath_dir_img, img_name\\n )\\n data['tms'] = ImageInfo.get_xml_tms(img_name_rgb)\\n data['image'] = img_name_rgb\\n data['quicklook'] = self.__make_png(abspath_rgb)\\n data['path'] = ImageInfo.get_path(img_name)\\n return data\",\n \"def _build_elbv2_mapping_from_resources(resource_to_analyse, result_dict, session):\\n for elb_instance, security_group_id, security_group_name in _generate_elb_instances_and_sg(resource_to_analyse, session):\\n resource_dict = _check_if_in_list(result_dict, elb_instance[\\\"LoadBalancerName\\\"], \\\"resource_id\\\")\\n if resource_dict is not None:\\n resource_dict[\\\"sg_attached\\\"].append({\\n \\\"sg_id\\\": security_group_id,\\n \\\"sg_name\\\": security_group_name\\n })\\n else:\\n result_dict.append({\\n \\\"resource_id\\\": elb_instance[\\\"LoadBalancerName\\\"],\\n \\\"resource_type\\\": \\\"elb\\\",\\n \\\"sg_attached\\\": [{\\n \\\"sg_id\\\": security_group_id,\\n \\\"sg_name\\\": security_group_name\\n }]\\n })\\n return result_dict\",\n \"def process():\\n config = read_config()\\n \\n\\n img_dir = config['DEFAULT']['images_directory']\\n results_dict = {}\\n images = list(get_image_files(img_dir))\\n for image in tqdm.tqdm(images):\\n info = hash_file(image)\\n if info == 0:\\n continue\\n\\n hash_value = info['hash']\\n\\n if hash_value not in results_dict:\\n file_name = os.path.basename(info['_id'])\\n results_dict[hash_value] = [file_name, 1]\\n else:\\n results_dict[hash_value][1] += 1\\n\\n count = list(results_dict.values())\\n sorted_count = sorted(count, key=lambda x: x[1], reverse=True)\\n \\n with ImagesDB(IMG_INFO_DB_FILENAME) as imgDb: \\n imgDb.insert_batch(sorted_count)\",\n \"def iterate_results(results, extract_fn):\\n outputs = {}\\n for environment, environment_results in results.items():\\n if environment not in outputs:\\n outputs[environment] = {}\\n for experimental_setting, setting_results in environment_results.items():\\n outputs[environment][experimental_setting] = []\\n for config, seeds_results in setting_results.items():\\n for seed, actual_results in seeds_results.items():\\n output = extract_fn(actual_results)\\n outputs[environment][experimental_setting].append(output)\\n outputs[environment][experimental_setting] = np.array(outputs[environment][experimental_setting])\\n return outputs\",\n \"def painting_matching_ml(imgs, db_imgs, method_list, text_masks, author_text, gt_text, metrics, weights, splits=30, max_rank=10):\\n descriptor_extractors = [get_descriptor_extractor(method_name) for method_name in method_list]\\n tmp_img_format = []\\n tmp_mask_format = []\\n tmp_text_format = []\\n for i in range(len(imgs)):\\n for j in range(len(imgs[i])):\\n tmp_img_format.append(imgs[i][j])\\n tmp_mask_format.append(text_masks[i][j])\\n tmp_text_format.append(author_text[i][j])\\n\\n #db_imgs = [img[0] for img in db_imgs]\\n db_img_splits = [i*len(db_imgs)//splits for i in range(splits-1)]\\n \\n scores = []\\n query_descriptors = extract_descriptors(tmp_img_format, descriptor_extractors, method_list, tmp_text_format, tmp_mask_format) \\n #np.array([extract_descriptors(img, matching_methods, mask) for img, mask in zip(tmp_img_format, tmp_mask_format)])\\n print(\\\"Starting db extraction + matching\\\")\\n for split in tqdm(range(splits-2)):\\n db_descriptors = extract_descriptors(db_imgs[db_img_splits[split]:db_img_splits[split+1]], descriptor_extractors, method_list, gt_text[db_img_splits[split]:db_img_splits[split+1]], None) #np.array([mrhm(db_img) for db_img in db_imgs[db_img_splits[split]:db_img_splits[split+1]]])\\n scores.append(compare_descriptors(query_descriptors, db_descriptors, metrics, method_list, weights))\\n # compare_descriptors(query_descriptors, db_descriptors, descriptor_comp_methods, descriptor_names, weights)\\n db_descriptors = extract_descriptors(db_imgs[db_img_splits[-1]:], descriptor_extractors, method_list, gt_text[db_img_splits[-1]:], None)\\n scores.append(compare_descriptors(query_descriptors, db_descriptors, metrics, method_list, weights))\\n \\n # concatenate all the results\\n scores = np.concatenate(scores, 1)\\n \\n top_k_matches = np.argpartition(scores, list(range(max_rank)))[:, :max_rank]\\n return top_k_matches\",\n \"def copy_to_dict(self):\\n search_hit_dict = {}\\n search_hit_dict['Offset'] = self.offset\\n search_hit_dict['Filename (inode)'] = self.filename\\n search_hit_dict['String'] = self.data\\n\\n return search_hit_dict\",\n \"def pre_pipeline(self, results):\\n results[\\\"img_prefix\\\"] = self.img_prefix\\n results[\\\"seg_prefix\\\"] = self.seg_prefix\\n results[\\\"proposal_file\\\"] = self.proposal_file\\n results[\\\"bbox_fields\\\"] = []\\n results[\\\"mask_fields\\\"] = []\\n results[\\\"seg_fields\\\"] = []\\n results[\\\"site_fields\\\"] = []\\n results[\\\"label_fields\\\"] = []\",\n \"def get_resource_results():\\n resource_results = {}\\n resource_results['checks'] = []\\n resource_results['green'] = 0\\n resource_results['red'] = 0\\n resource_results['orange'] = 0\\n resource_results['blue'] = 0\\n resource_results['failed_accounts'] = 0\\n resource_results['total_accounts'] = 0\\n resource_results['total_checks'] = 0\\n\\n # Defaults for when no data is reported, working towards having modules be\\n # modular / optional\\n resource_results['blue_percent'] = 100\\n resource_results['red_percent'] = 0\\n resource_results['orange_percent'] = 0\\n resource_results['green_percent'] = 0\\n resource_results['working_percentage'] = 100\\n\\n # Check how many accounts failed from each module and add them to the\\n # total failed accounts. If the data from a module is considered stale\\n # then all of it's accounts will be considered failed.\\n milliseconds_since_epoch = time.time() * 1000\\n for module in get_all_data('resources_success:*'):\\n module_success_json = get_data(module)\\n module_success = json.loads(module_success_json)[0]\\n resource_results['total_accounts'] += module_success['total_accounts']\\n resource_results['total_checks'] += module_success['total_checks']\\n milliseconds_since_epoch_module_data_is_valid_until = module_success['valid_until']\\n if milliseconds_since_epoch > milliseconds_since_epoch_module_data_is_valid_until:\\n resource_results['failed_accounts'] += module_success['total_accounts']\\n logger.error('Data for {} is stale, please check the daemon is functioning properly'.format(module))\\n else:\\n resource_results['failed_accounts'] += module_success['failed_accounts']\\n\\n # We will count checks in so we can compare it against the number of checks\\n # reported by the daemon\\n checks_found = 0\\n # Get list of keys in the format resources:module#uuid\\n for host in get_all_data('resources:*'):\\n try:\\n # Storing lists with only one value since when I convert\\n # dictionaries to json and store them in redis they come back as\\n # strings, I am working around this by storing lists,\\n # ast.literal_eval also works\\n host_data = json.loads(get_data(host))[0]\\n resource_results['checks'].append(host_data)\\n # get the health status colour of the current check, and then add\\n # one to the number of checks with that health status\\n resource_results[host_data['health_status']] += 1\\n checks_found += 1\\n except Exception as e:\\n # I would rather log to uwsgi's log but I'll sort this out later\\n logger.error('Data for {} is not in a valid format: {}'.format(host, e))\\n\\n # If we are getting back old checks that are no-longer reporting hence\\n # are not in the total_checks variable then they have failed.\\n # If we are getting back less checks than we stored then something has\\n # gone really wrong or we caught the weekly cron that clears the keys.\\n if resource_results['total_checks'] != checks_found:\\n logger.info('The number of checks stored in the database doesn\\\\'t '\\\\\\n 'match the number reported by the daemon, it is likely some '\\\\\\n 'servers are no-longer reporting, run '\\\\\\n 'resources_list_unreporting_servers.py to look into this.')\\n\\n # The number of checks we are outputing is authoritive over the number\\n # we expected to be there, at the moment we are just logging the fact they\\n # were different, it would be nice to have a visual display or send an\\n # email but there isn't a correct place to do this at the moment\\n resource_results['total_checks'] = checks_found\\n\\n total_results = resource_results['green'] + resource_results['red'] + resource_results['orange'] + resource_results['blue']\\n if total_results != 0:\\n resource_results['red_percent'] = ( resource_results['red'] / total_results ) * 100\\n resource_results['orange_percent'] = ( resource_results['orange'] / total_results ) * 100\\n resource_results['blue_percent'] = ( resource_results['blue'] / total_results ) * 100\\n # I want the percentage to always be 100 and green seems the most\\n # disposable / least affected by any rounding issues\\n resource_results['green_percent'] = 100 - ( resource_results['red_percent'] + resource_results['orange_percent'] + resource_results['blue_percent'] )\\n\\n resource_results['working_percentage'] = 100 - (( resource_results['failed_accounts'] / resource_results['total_accounts'] ) * 100 )\\n resource_results['working_accounts'] = resource_results['total_accounts'] - resource_results['failed_accounts']\\n logger.debug('working_percentage: {}'.format(resource_results['working_percentage']))\\n return resource_results\",\n \"def get_results(self):\\n result = [round(self.mr / self.test_size, 1), round(self.mrr / self.test_size, 3),\\n round(self.hits1 / self.test_size, 3), round(self.hits3 / self.test_size, 3),\\n round(self.hits5 / self.test_size, 3), round(self.hits10 / self.test_size, 3)]\\n return result\",\n \"def __call__(\\n self, \\n image: ndarray, \\n adversarial_image: ndarray\\n ) -> Dict[str, Union[float, int]]:\\n ...\",\n \"def __getitem__(self, idx: int) -> Dict:\\n files = self.files[idx]\\n build_mask = np.array(Image.open(files[\\\"build_mask\\\"]))\\n demo_mask = np.array(Image.open(files[\\\"demolish_mask\\\"]))\\n build_mask = np.clip(build_mask.mean(axis=-1), 0, 1).astype(\\\"uint8\\\")\\n demo_mask = np.clip(demo_mask.mean(axis=-1), 0, 1).astype(\\\"uint8\\\")\\n image1 = np.array(Image.open(files[\\\"image1\\\"]))\\n image2 = np.array(Image.open(files[\\\"image2\\\"]))\\n image1, image2, build_mask, demo_mask = self.transform([image1, image2, build_mask, demo_mask])\\n x = torch.stack([image1, image2], dim=0)\\n return dict(x=x, build_mask=build_mask, demolish_mask=demo_mask)\",\n \"def generate(self):\\n return {\\n 'json_class': 'Pact::ArrayLike',\\n 'contents': from_term(self.matcher),\\n 'min': self.minimum}\",\n \"def process_matches(matches: List[EndpointMatch]) -> List[Result] :\\n out=[]\\n #sort data\\n sort_f = lambda match: match.endpoint_1.name\\n sorted_matches = sorted(matches, key = sort_f)\\n #group by endpoint 1\\n for key, group in itertools.groupby(sorted_matches, key = sort_f ):\\n #sort by score\\n endpoint_matches = sorted(list(group), key = lambda match: match.score,reverse=True)\\n #best match\\n best_match = endpoint_matches[0]\\n #others\\n other_matches = \\\";\\\".join([f\\\"{a.endpoint_2.name}|{a.score:.3g}\\\" for a in endpoint_matches[1:] ])\\n\\n out.append(Result(\\n best_match.endpoint_1.name,\\n best_match.endpoint_2.name,\\n best_match.score,\\n \\\";\\\".join(best_match.endpoint_1.matches),\\n \\\";\\\".join(best_match.endpoint_2.matches),\\n best_match.endpoint_1.regex,\\n best_match.endpoint_2.regex,\\n other_matches\\n ))\\n return out\",\n \"def post(self):\\n names = self.request.get('names').split(',')\\n results = {}\\n for name in names:\\n key = 'eol-images-%s' % name\\n value = memcache.get(key)\\n if value:\\n logging.info('memcache value: %s' % value)\\n else:\\n value = cache.get(key, loads=True)\\n if not value:\\n name = urllib.quote(name)\\n search_url = 'http://eol.org/api/search/%s.json?exact=1' % name\\n result = json.loads(urlfetch.fetch(search_url, deadline=60).content)\\n page_id = result['results'][0]['id']\\n page_url = 'http://eol.org/api/pages/1.0/%s.json' % page_id\\n logging.info(page_url)\\n result = json.loads(urlfetch.fetch(page_url, deadline=60).content)\\n object_id = None\\n for x in result['dataObjects']:\\n if x['dataType'].endswith('StillImage'):\\n object_id = x['identifier']\\n if object_id:\\n object_url = 'http://eol.org/api/data_objects/1.0/%s.json' % object_id\\n value = json.loads(urlfetch.fetch(object_url, deadline=60).content)\\n cache.add(key, value, dumps=True)\\n memcache.add(key, value) \\n results[name] = value \\n self.response.headers[\\\"Content-Type\\\"] = \\\"application/json\\\" \\n self.response.out.write(json.dumps(results))\",\n \"def compute_metrics(self, results: Sequence[Dict]) -> Dict:\\n\\n preds = []\\n gts = []\\n for result in results:\\n preds.append(result['pred_labels'])\\n gts.append(result['gt_labels'])\\n preds = torch.cat(preds)\\n gts = torch.cat(gts)\\n\\n assert preds.max() < self.num_classes\\n assert gts.max() < self.num_classes\\n\\n cared_labels = preds.new_tensor(self.cared_labels, dtype=torch.long)\\n\\n hits = (preds == gts)[None, :]\\n preds_per_label = cared_labels[:, None] == preds[None, :]\\n gts_per_label = cared_labels[:, None] == gts[None, :]\\n\\n tp = (hits * preds_per_label).float()\\n fp = (~hits * preds_per_label).float()\\n fn = (~hits * gts_per_label).float()\\n\\n result = {}\\n if 'macro' in self.mode:\\n result['macro_f1'] = self._compute_f1(\\n tp.sum(-1), fp.sum(-1), fn.sum(-1))\\n if 'micro' in self.mode:\\n result['micro_f1'] = self._compute_f1(tp.sum(), fp.sum(), fn.sum())\\n\\n return result\",\n \"def checksImages(self):\\n metadata=[]\\n for image in self.meta['sources']:\\n with rasterio.open(image) as src:\\n metaData=src.meta\\n \\n assert metaData['driver'] == 'GTiff', \\\"Driver is not supported: {0}\\\".format(metaData['driver'])\\n assert metaData['count'] == len(self.meta['bandNames']), \\\"Nbands incorrect, expected: {0}, {1} provided\\\".format(metaData['count'],len(self.meta['bandNames']))\\n \\n metadata.append({'dtype': metaData['dtype'], 'driver': metaData['driver'], 'nodata': metaData['nodata'], 'nBands': metaData['count'],'crs': src.crs.to_string()})\\n \\n assert len(set([item['dtype'] for item in metadata])) == 1, \\\"Images list dtypes aren't compatibles. Expected: 1, {1} provided\\\".format(metaData['count'],len(set([item['dtype'] for item in metadata])))\\n assert len(set([item['driver'] for item in metadata])) == 1, \\\"Images list drivers aren't compatibles. Expected: 1, 1 provided\\\".format(metaData['count'],len(set([item['driver'] for item in metadata])))\\n assert len(set([item['nodata'] for item in metadata])) == 1, \\\"Images list nodata values aren't compatibles. Expected: 1, {1} provided\\\".format(metaData['count'],len(set([item['nodata'] for item in metadata])))\\n assert len(set([item['nBands'] for item in metadata])) == 1, \\\"Images list nBands number aren't compatibles. Expected: 1, {1} provided\\\".format(metaData['count'],len(set([item['nBands'] for item in metadata])))\\n assert len(set([item['crs'] for item in metadata])) == 1, \\\"Images list crs aren't compatibles. Expected: 1, {1} provided\\\".format(metaData['count'],len(set([item['crs'] for item in metadata]))) \\n return metadata[0]\"\n]","isConversational":false},"negative_scores":{"kind":"list like","value":["0.6621895","0.64354163","0.6385068","0.6345569","0.63442147","0.62534124","0.6184932","0.61760575","0.6173091","0.60566914","0.6048575","0.6046165","0.6008114","0.59930813","0.59672993","0.5958381","0.59390056","0.58587366","0.5855858","0.5824432","0.5820238","0.58176214","0.58117014","0.57997525","0.5760925","0.57530844","0.57502156","0.5748866","0.57440513","0.57380676","0.57378685","0.5722579","0.5708604","0.5701054","0.56759876","0.5657722","0.5654448","0.5654025","0.5646867","0.5635593","0.55936265","0.5577351","0.5567221","0.5559705","0.55579376","0.5555611","0.55489624","0.5528617","0.5525503","0.5525323","0.55130625","0.5496515","0.54960114","0.5489149","0.54859614","0.54839224","0.5472803","0.5466476","0.5464672","0.54548645","0.54548293","0.5451661","0.5451248","0.5449377","0.5448407","0.5437631","0.54321694","0.5425948","0.5425262","0.5424067","0.5420165","0.54188406","0.5418336","0.5415256","0.54090303","0.5408035","0.54062897","0.5403069","0.5400889","0.53994715","0.5393601","0.53814304","0.53635406","0.5355368","0.5350279","0.5336311","0.5327998","0.5323576","0.5310025","0.5307986","0.53078073","0.53012747","0.52954996","0.5292379","0.5291169","0.52892077","0.5282658","0.52744615","0.52732986","0.5271746","0.5269578"],"string":"[\n \"0.6621895\",\n \"0.64354163\",\n \"0.6385068\",\n \"0.6345569\",\n \"0.63442147\",\n \"0.62534124\",\n \"0.6184932\",\n \"0.61760575\",\n \"0.6173091\",\n \"0.60566914\",\n \"0.6048575\",\n \"0.6046165\",\n \"0.6008114\",\n \"0.59930813\",\n \"0.59672993\",\n \"0.5958381\",\n \"0.59390056\",\n \"0.58587366\",\n \"0.5855858\",\n \"0.5824432\",\n \"0.5820238\",\n \"0.58176214\",\n \"0.58117014\",\n \"0.57997525\",\n \"0.5760925\",\n \"0.57530844\",\n \"0.57502156\",\n \"0.5748866\",\n \"0.57440513\",\n \"0.57380676\",\n \"0.57378685\",\n \"0.5722579\",\n \"0.5708604\",\n \"0.5701054\",\n \"0.56759876\",\n \"0.5657722\",\n \"0.5654448\",\n \"0.5654025\",\n \"0.5646867\",\n \"0.5635593\",\n \"0.55936265\",\n \"0.5577351\",\n \"0.5567221\",\n \"0.5559705\",\n \"0.55579376\",\n \"0.5555611\",\n \"0.55489624\",\n \"0.5528617\",\n \"0.5525503\",\n \"0.5525323\",\n \"0.55130625\",\n \"0.5496515\",\n \"0.54960114\",\n \"0.5489149\",\n \"0.54859614\",\n \"0.54839224\",\n \"0.5472803\",\n \"0.5466476\",\n \"0.5464672\",\n \"0.54548645\",\n \"0.54548293\",\n \"0.5451661\",\n \"0.5451248\",\n \"0.5449377\",\n \"0.5448407\",\n \"0.5437631\",\n \"0.54321694\",\n \"0.5425948\",\n \"0.5425262\",\n \"0.5424067\",\n \"0.5420165\",\n \"0.54188406\",\n \"0.5418336\",\n \"0.5415256\",\n \"0.54090303\",\n \"0.5408035\",\n \"0.54062897\",\n \"0.5403069\",\n \"0.5400889\",\n \"0.53994715\",\n \"0.5393601\",\n \"0.53814304\",\n \"0.53635406\",\n \"0.5355368\",\n \"0.5350279\",\n \"0.5336311\",\n \"0.5327998\",\n \"0.5323576\",\n \"0.5310025\",\n \"0.5307986\",\n \"0.53078073\",\n \"0.53012747\",\n \"0.52954996\",\n \"0.5292379\",\n \"0.5291169\",\n \"0.52892077\",\n \"0.5282658\",\n \"0.52744615\",\n \"0.52732986\",\n \"0.5271746\",\n \"0.5269578\"\n]","isConversational":false},"document_score":{"kind":"string","value":"0.0"},"document_rank":{"kind":"string","value":"-1"}}},{"rowIdx":93297,"cells":{"query":{"kind":"string","value":"updates the precision and recall data in the image_matching_module's results"},"document":{"kind":"string","value":"def update_precision_and_recall(experiment_dict, list_dict, algorithm_name, thresholds):\n for threshold in thresholds:\n algorithm_column_name = algorithm_name + \"_actual\"\n tp = experiment_dict[str(threshold)][algorithm_name][list_dict['TP']]\n fp = experiment_dict[str(threshold)][algorithm_name][list_dict['FP']]\n fn = experiment_dict[str(threshold)][algorithm_name][list_dict['FN']]\n tn = experiment_dict[str(threshold)][algorithm_name][list_dict['TN']]\n\n accuracy = (tp + tn) / (tp + tn + fp + fn)\n tpr = tp / (tp + fn)\n tnr = tn / (tn + fp)\n precision = tp / (tp + fp)\n recall = tpr\n f1_measure = 2 * ((precision * recall) / (precision + recall))\n\n experiment_dict[str(threshold)][algorithm_name][list_dict['TPR']] = tpr\n experiment_dict[str(threshold)][algorithm_name][list_dict['TNR']] = tnr\n experiment_dict[str(threshold)][algorithm_name][list_dict['PRECISION']] = precision\n experiment_dict[str(threshold)][algorithm_name][list_dict['RECALL']] = recall\n experiment_dict[str(threshold)][algorithm_name][list_dict['F1 MEASURE']] = f1_measure\n experiment_dict[str(threshold)][algorithm_name][list_dict['ACCURACY']] = accuracy\n return experiment_dict"},"metadata":{"kind":"string","value":"{\n \"objective\": {\n \"self\": [],\n \"paired\": [],\n \"triplet\": [\n [\n \"query\",\n \"document\",\n \"negatives\"\n ]\n ]\n }\n}"},"negatives":{"kind":"list like","value":["def evaluate_matrices(self, precision_matrix, recall_matrix):\n\t\tground_truth_count = precision_matrix.shape[0]\n\t\tdetection_count = precision_matrix.shape[1]\n\t\tground_truth_sets_precision = defaultdict(set) # number of ground truth items that match a particular detection in the precision matrix\n\t\tdetection_sets_precision = defaultdict(set) # number of detection items that match a particular ground truth in the precision matrix\n\t\tground_truth_sets_recall = defaultdict(set) # number of ground truth items that match a particular detection in the recall matrix\n\t\tdetection_sets_recall = defaultdict(set) # number of detection items that match a particular ground truth in the recall matrix\n\n\t\tfor gt_index in range(ground_truth_count):\n\t\t\tfor det_index in range(detection_count):\n\t\t\t\tif precision_matrix[gt_index, det_index] >= self.precision_threshold:\n\t\t\t\t\tground_truth_sets_precision[det_index].add(gt_index)\n\t\t\t\t\tdetection_sets_precision[gt_index].add(det_index)\n\t\t\t\tif recall_matrix[gt_index, det_index] >= self.recall_threshold:\n\t\t\t\t\tground_truth_sets_recall[det_index].add(gt_index)\n\t\t\t\t\tdetection_sets_recall[gt_index].add(det_index)\n\n\t\tmatch_ground_truth = 0. # sum of MatchG\n\t\tmatch_detection = 0. # sum of MatchD\n\n\t\tone_to_one_precision = set()\n\t\tfor gt_index in detection_sets_precision:\n\t\t\tmatching_detections_precision = detection_sets_precision[gt_index]\n\t\t\tif len(matching_detections_precision) == 1:\n\t\t\t\t(detection_precision, ) = matching_detections_precision\n\t\t\t\tif len(ground_truth_sets_precision[detection_precision]) == 1:\n\t\t\t\t\tone_to_one_precision.add((gt_index, detection_precision))\n\t\t\telse:\n\t\t\t\t# one-to-many (one ground truth to many detections)\n\t\t\t\tgt_sum = 0.\n\t\t\t\tfor detection_precision in matching_detections_precision:\n\t\t\t\t\tgt_sum += recall_matrix[gt_index, detection_precision]\n\t\t\t\tif gt_sum >= self.recall_threshold:\n\t\t\t\t\t#print(\"1:N ~ GT {} : DT {}\".format(gt_index,matching_detections_precision))\n\t\t\t\t\tmatch_ground_truth += self.scatter_punishment(matching_detections_precision)\n\t\t\t\t\tmatch_detection += len(matching_detections_precision) * self.scatter_punishment(matching_detections_precision)\n\n\t\tone_to_one_recall = set()\n\t\tfor det_index in ground_truth_sets_recall:\n\t\t\tmatching_ground_truths_recall = ground_truth_sets_recall[det_index]\n\t\t\tif len(matching_ground_truths_recall) == 1:\n\t\t\t\t(ground_truth_recall, ) = matching_ground_truths_recall\n\t\t\t\tif len(detection_sets_recall[ground_truth_recall]) == 1:\n\t\t\t\t\tone_to_one_recall.add((ground_truth_recall, det_index))\n\t\t\telse:\n\t\t\t\t# many-to-one (many ground truths covered by one detection)\n\t\t\t\tdet_sum = 0\n\t\t\t\tfor ground_truth_recall in matching_ground_truths_recall:\n\t\t\t\t\tdet_sum += precision_matrix[ground_truth_recall, det_index]\n\t\t\t\tif det_sum >= self.precision_threshold:\n\t\t\t\t\t#print(\"N:1 ~ DT {} : GT {}\".format(det_index,matching_ground_truths_recall))\n\t\t\t\t\tmatch_detection += self.scatter_punishment(matching_ground_truths_recall)\n\t\t\t\t\tmatch_ground_truth += len(matching_ground_truths_recall) * self.scatter_punishment(matching_ground_truths_recall)\n\n\t\tone_to_one_matches = one_to_one_precision & one_to_one_recall\n\t\tmatch_ground_truth += len(one_to_one_matches)\n\t\tmatch_detection += len(one_to_one_matches)\n\n\t\trecall = match_ground_truth / float(ground_truth_count)\n\t\tprecision = match_detection / float(detection_count)\n\t\treturn (precision, recall, (match_detection, float(detection_count)), (match_ground_truth, float(ground_truth_count)))","def _update_result(self, results, clf_numbers):\n # ToDo make results of scoring values dynamic\n names_results = ['Accuracy']\n for number in clf_numbers:\n for name in names_results:\n if name not in self.results:\n self.results[name] = [results[number][name + \"_test_score_\" + str(number)]]\n else:\n self.results[name].append(results[number][name + \"_test_score_\" + str(number)])","def precision(self):\n self.overall_precision = precision_score(\n self.y_true, self.y_pred, average = self.average_type).round(self.digits_count_fp)\n self.classes_precision = precision_score(\n self.y_true, self.y_pred, average = None).round(self.digits_count_fp)","def get_recall_precision(imgs_truth, imgs_pred, is_hit):\n # Numbers of relevant and detect boxes, respectively.\n num_rel = sum([len(img.boxes) for img in imgs_truth])\n num_det = sum([len(img.boxes) for img in imgs_pred])\n\n boxes_for_fname_truth = {img.fname: img.boxes for img in imgs_truth}\n boxes_for_fname_pred = {img.fname: img.boxes for img in imgs_pred}\n\n num_hit = 0\n for fname, boxes in boxes_for_fname_truth.items():\n if fname not in boxes_for_fname_pred:\n continue\n\n num_hit += get_num_hit(boxes, boxes_for_fname_pred[fname], is_hit)\n\n return num_hit / num_rel, num_hit / num_det","def recall(self):\n self.overall_recall = recall_score(\n self.y_true, self.y_pred, average = self.average_type).round(self.digits_count_fp)\n self.classes_recall = recall_score(\n self.y_true, self.y_pred, average = None).round(self.digits_count_fp)","def calculates_results_stats(results_dic): \n # Creates empty dictionary for results_stats_dic\n results_stats_dic = dict()\n \n # Sets all counters to initial values of zero so that they can \n # be incremented while processing through the images in results_dic \n results_stats_dic['n_dogs_img'] = 0\n results_stats_dic['n_match'] = 0\n results_stats_dic['n_correct_dogs'] = 0\n results_stats_dic['n_correct_notdogs'] = 0\n results_stats_dic['n_correct_breed'] = 0\n \n # process through the results dictionary\n for key in results_dic:\n \n # Labels Match Exactly\n if results_dic[key][2] == 1:\n results_stats_dic['n_match'] += 1\n\n # TODO: 5a. REPLACE pass with CODE that counts how many pet images of\n # dogs had their breed correctly classified. This happens \n # when the pet image label indicates the image is-a-dog AND \n # the pet image label and the classifier label match. You \n # will need to write a conditional statement that determines\n # when the dog breed is correctly classified and then \n # increments 'n_correct_breed' by 1. Recall 'n_correct_breed' \n # is a key in the results_stats_dic dictionary with it's value \n # representing the number of correctly classified dog breeds.\n # \n # Pet Image Label is a Dog AND Labels match- counts Correct Breed\n if results_dic[key][3] == 1 and results_dic[key][2] == 1:\n results_stats_dic['n_correct_breed'] += 1\n \n # Pet Image Label is a Dog - counts number of dog images\n if results_dic[key][3] == 1:\n results_stats_dic['n_dogs_img'] += 1\n \n # Classifier classifies image as Dog (& pet image is a dog)\n # counts number of correct dog classifications\n if results_dic[key][4] == 1:\n results_stats_dic['n_correct_dogs'] += 1\n\n # TODO: 5b. REPLACE pass with CODE that counts how many pet images \n # that are NOT dogs were correctly classified. This happens \n # when the pet image label indicates the image is-NOT-a-dog \n # AND the classifier label indicates the images is-NOT-a-dog.\n # You will need to write a conditional statement that \n # determines when the classifier label indicates the image \n # is-NOT-a-dog and then increments 'n_correct_notdogs' by 1. \n # Recall the 'else:' above 'pass' already indicates that the \n # pet image label indicates the image is-NOT-a-dog and \n # 'n_correct_notdogs' is a key in the results_stats_dic dictionary \n # with it's value representing the number of correctly \n # classified NOT-a-dog images.\n # \n # Pet Image Label is NOT a Dog\n else:\n # Classifier classifies image as NOT a Dog(& pet image isn't a dog)\n # counts number of correct NOT dog clasifications.\n if results_dic[key][3] == 0 and results_dic[key][4] == 0:\n results_stats_dic['n_correct_notdogs'] += 1\n\n\n # Calculates run statistics (counts & percentages) below that are calculated\n # using the counters from above.\n\n # calculates number of total images\n results_stats_dic['n_images'] = len(results_dic)\n\n # calculates number of not-a-dog images using - images & dog images counts\n results_stats_dic['n_notdogs_img'] = (results_stats_dic['n_images'] - \n results_stats_dic['n_dogs_img']) \n\n # TODO: 5c. REPLACE zero(0.0) with CODE that calculates the % of correctly\n # matched images. Recall that this can be calculated by the\n # number of correctly matched images ('n_match') divided by the \n # number of images('n_images'). This result will need to be \n # multiplied by 100.0 to provide the percentage.\n # \n # Calculates % correct for matches\n results_stats_dic['pct_match'] = (results_stats_dic['n_match'] / results_stats_dic['n_images']) * 100\n\n # TODO: 5d. REPLACE zero(0.0) with CODE that calculates the % of correctly\n # classified dog images. Recall that this can be calculated by \n # the number of correctly classified dog images('n_correct_dogs')\n # divided by the number of dog images('n_dogs_img'). This result \n # will need to be multiplied by 100.0 to provide the percentage.\n # \n # Calculates % correct dogs\n results_stats_dic['pct_correct_dogs'] = (results_stats_dic['n_correct_dogs'] / results_stats_dic['n_dogs_img']) * 100\n\n # TODO: 5e. REPLACE zero(0.0) with CODE that calculates the % of correctly\n # classified breeds of dogs. Recall that this can be calculated \n # by the number of correctly classified breeds of dog('n_correct_breed') \n # divided by the number of dog images('n_dogs_img'). This result \n # will need to be multiplied by 100.0 to provide the percentage.\n # \n # Calculates % correct breed of dog\n results_stats_dic['pct_correct_breed'] = (results_stats_dic['n_correct_breed'] / results_stats_dic['n_dogs_img']) * 100\n\n # Calculates % correct not-a-dog images\n # Uses conditional statement for when no 'not a dog' images were submitted \n if results_stats_dic['n_notdogs_img'] > 0:\n results_stats_dic['pct_correct_notdogs'] = (results_stats_dic['n_correct_notdogs'] /\n results_stats_dic['n_notdogs_img']) * 100.0\n else:\n results_stats_dic['pct_correct_notdogs'] = 0.0\n\n \n # TODO 5f. REPLACE None with the results_stats_dic dictionary that you \n # created with this function \n return results_stats_dic","def compute_map(scores, num_instances):\r\n pr = []\r\n for i, score in enumerate(scores):\r\n score = score[np.argsort(-score[:, 1])] # sort by confidence score\r\n FP = 0\r\n TP = 0\r\n pr_ = []\r\n for prediction in score:\r\n if prediction[0]:\r\n TP += 1\r\n else:\r\n FP += 1\r\n pr_.append([TP/(TP+FP), (TP/num_instances)])\r\n pr.append(pr_)\r\n pr = np.array(pr)\r\n\r\n pinterps = [] # lists of interpolated precisions for every confidence level\r\n ranks = np.linspace(0, 1, 11)\r\n idxs_interpolations = [] # list of indexes of the interpolated precisions, just to plot the recall\r\n for pr_ in pr:\r\n pinterp = []\r\n idxs = []\r\n last_idx = -1\r\n for rank in ranks:\r\n idx = (np.abs(pr_[:, 1] - rank)).argmin() # find the closest recall to the rank\r\n\r\n if rank > pr_[idx, 1]: # this makes sure we are taking the closest recall at the right of the rank\r\n if idx+1 < pr_[:, 0].shape[0]:\r\n idx += 1\r\n interpolated_precision = np.max(pr_[idx:, 0]) # find the max precision within the interval\r\n if idx == last_idx: # just some checks for when the recall doesn't exist\r\n pinterp[-1] = 0\r\n idxs[-1] = 0\r\n pinterp.append(0)\r\n idxs.append(0)\r\n else:\r\n pinterp.append(interpolated_precision)\r\n idxs.append(idx)\r\n last_idx = idx\r\n pinterps.append(pinterp)\r\n idxs_interpolations.append(idxs)\r\n APs = np.array(pinterps).mean(axis=1) # the AP is the average of the interpolated precisions\r\n mAP = APs.mean() # mAP is the mean of all the APs\r\n\r\n return pr, pinterps, idxs_interpolations, mAP, APs","def recallAndPrecision(self,train = None,test = None,k = 8,nitem = 10):\n train = train or self.traindata\n test = test or self.testdata\n hit = 0\n recall = 0\n precision = 0\n for user in train.keys():\n tu = test.get(user,{})\n rank = self.recommend(user, train = train,k = k,nitem = nitem) \n for item,_ in rank.items():\n if item in tu:\n hit += 1\n recall += len(tu)\n precision += nitem\n return (hit / (recall * 1.0),hit / (precision * 1.0))","def update(\n self, outputs: torch.Tensor, targets: torch.Tensor\n ) -> Tuple[float, float, float]:\n tn, fp, fn, tp, support = super().update(outputs=outputs, targets=targets)\n precision_value, recall_value, f1_value = get_binary_metrics(\n tp=tp, fp=fp, fn=fn, zero_division=self.zero_division\n )\n return precision_value, recall_value, f1_value","def update(self, y_pred: torch.Tensor, y_true: torch.Tensor) -> None:\n self.accuracy.update(y_pred, y_true)\n self.roc_auc.update(y_pred, y_true)\n self.auc_score.update(y_pred, y_true)\n self.precision.update(y_pred, y_true)\n self.recall.update(y_pred, y_true)\n self.f1.update(y_pred, y_true)\n self.hamming_distance.update(y_pred, y_true)\n self.stat_scores.update(y_pred, y_true)","def compute_precision_recall_bbox(per_image_detections, per_image_gts, num_gt_classes,\n matching_iou_threshold=0.5):\n per_image_eval = per_image_evaluation.PerImageEvaluation(\n num_groundtruth_classes=num_gt_classes,\n matching_iou_threshold=matching_iou_threshold,\n nms_iou_threshold=1.0,\n nms_max_output_boxes=10000\n )\n\n print('Running per-object analysis...')\n\n detection_tp_fp = defaultdict(list) # key is the category; in each list, 1 is tp, 0 is fp\n detection_scores = defaultdict(list)\n num_total_gt = defaultdict(int)\n\n for image_id, dets in tqdm(per_image_detections.items()):\n\n detected_boxes = np.array(dets['boxes'], dtype=np.float32)\n detected_scores = np.array(dets['scores'], dtype=np.float32)\n # labels input to compute_object_detection_metrics() needs to start at 0, not 1\n detected_labels = np.array(dets['labels'], dtype=np.int) - 1 # start at 0\n # num_detections = len(dets['boxes'])\n\n gts = per_image_gts[image_id]\n gt_boxes = np.array(gts['gt_boxes'], dtype=np.float32)\n gt_labels = np.array(gts['gt_labels'], dtype=np.int) - 1 # start at 0\n num_gts = len(gts['gt_boxes'])\n\n groundtruth_is_difficult_list = np.zeros(num_gts, dtype=bool) # place holders - we don't have these\n groundtruth_is_group_of_list = np.zeros(num_gts, dtype=bool)\n\n # to prevent 'Invalid dimensions for box data.' error\n if num_gts == 0:\n # this box will not match any detections\n gt_boxes = np.array([[0, 0, 0, 0]], dtype=np.float32)\n\n scores, tp_fp_labels, is_class_correctly_detected_in_image = (\n per_image_eval.compute_object_detection_metrics(\n detected_boxes=detected_boxes,\n detected_scores=detected_scores,\n detected_class_labels=detected_labels,\n groundtruth_boxes=gt_boxes,\n groundtruth_class_labels=gt_labels,\n groundtruth_is_difficult_list=groundtruth_is_difficult_list,\n groundtruth_is_group_of_list=groundtruth_is_group_of_list\n )\n )\n\n for i, tp_fp_labels_cat in enumerate(tp_fp_labels):\n assert sum(tp_fp_labels_cat) <= sum(gt_labels == i) # true positives < gt of that category\n cat = i + 1 # categories start at 1\n detection_tp_fp[cat].append(tp_fp_labels_cat)\n detection_scores[cat].append(scores[i])\n num_total_gt[cat] += sum(gt_labels == i) # gt_labels start at 0\n\n all_scores = []\n all_tp_fp = []\n\n print('Computing precision recall for each category...')\n per_cat_metrics = {}\n for i in range(num_gt_classes):\n cat = i + 1\n scores_cat = np.concatenate(detection_scores[cat])\n tp_fp_cat = np.concatenate(detection_tp_fp[cat]).astype(np.bool)\n all_scores.append(scores_cat)\n all_tp_fp.append(tp_fp_cat)\n\n precision, recall = metrics.compute_precision_recall(\n scores_cat, tp_fp_cat, num_total_gt[cat]\n )\n average_precision = metrics.compute_average_precision(precision, recall)\n\n per_cat_metrics[cat] = {\n 'category': cat,\n 'precision': precision,\n 'recall': recall,\n 'average_precision': average_precision,\n 'scores': scores_cat,\n 'tp_fp': tp_fp_cat,\n 'num_gt': num_total_gt[cat]\n }\n print('Number of ground truth in category {} is {}'.format(cat, num_total_gt[cat]))\n\n # compute one-class precision/recall/average precision (if every box is just of an object class)\n all_scores = np.concatenate(all_scores)\n all_tp_fp = np.concatenate(all_tp_fp)\n overall_gt_count = sum(num_total_gt.values())\n\n one_class_prec, one_class_recall = metrics.compute_precision_recall(\n all_scores, all_tp_fp, overall_gt_count\n )\n one_class_average_precision = metrics.compute_average_precision(one_class_prec, one_class_recall)\n\n per_cat_metrics['one_class'] = {\n 'category': 'one_class',\n 'precision': one_class_prec,\n 'recall': one_class_recall,\n 'average_precision': one_class_average_precision,\n 'scores': all_scores,\n 'tp_fp': all_tp_fp,\n 'num_gt': overall_gt_count\n }\n\n return per_cat_metrics","def update(self, outputs: torch.Tensor, targets: torch.Tensor) -> Tuple[float, float, float]:\n tn, fp, fn, tp, support = super().update(outputs=outputs, targets=targets)\n precision_value, recall_value, f1_value = get_binary_metrics(\n tp=tp, fp=fp, fn=fn, zero_division=self.zero_division\n )\n return precision_value, recall_value, f1_value","def compute_recall_precision_fscore(truth: List[Node], prediction: List[Node]) \\\n -> Tuple[numpy.ndarray, numpy.ndarray, numpy.ndarray]:\n recall = numpy.zeros((len(truth), len(prediction)))\n precision = numpy.zeros((len(truth), len(prediction)))\n fscore = numpy.zeros((len(truth), len(prediction)))\n\n for i, t in enumerate(truth):\n for j, p in enumerate(prediction):\n r, p, f = pixel_metrics(t, p)\n recall[i, j] = r\n precision[i, j] = p\n fscore[i, j] = f\n\n return recall, precision, fscore","def precision(y_test, y_pred):\n\treturn precision_score(y_test, y_pred)","def mask_evaluation(annotation_mask, result_mask, idx):\n\n true_positive = np.sum(np.logical_and(annotation_mask == 255, result_mask == 255)) \n false_positive = np.sum(np.logical_and(result_mask == 255, annotation_mask != result_mask))\n false_negative = np.sum(np.logical_and(annotation_mask == 255, annotation_mask != result_mask))\n\n precision = true_positive / (true_positive + false_positive)\n recall = true_positive / (true_positive + false_negative)\n f1_measure = 2 * ((precision * recall) / (precision + recall))\n\n return recall, precision, f1_measure","def _eval_classifier(self):\n\n y_pred_baseline = self.df_baseline[self.score_column]\n y_pred_sample = self.df_sample[self.score_column]\n\n y_label_baseline = self.df_baseline[self.label_column]\n y_label_sample = self.df_sample[self.label_column]\n\n precision_baseline = precision_score(y_label_baseline, y_pred_baseline)\n recall_baseline = recall_score(y_label_baseline, y_pred_baseline)\n acc_baseline = accuracy_score(y_label_baseline, y_pred_baseline)\n f1_baseline = f1_score(y_label_baseline, y_pred_baseline)\n try:\n auc_baseline = roc_auc_score(y_label_baseline, y_pred_baseline)\n except ValueError:\n auc_baseline = \"NA\"\n\n precision_sample = precision_score(y_label_sample, y_pred_sample)\n recall_sample = recall_score(y_label_sample, y_pred_sample)\n acc_sample = accuracy_score(y_label_sample, y_pred_sample)\n f1_sample = f1_score(y_label_sample, y_pred_sample)\n try:\n auc_sample = roc_auc_score(y_label_sample, y_pred_sample)\n except ValueError:\n auc_sample = \"NA\"\n\n metrics_df = pd.DataFrame(\n {\n \"Accuracy\": [acc_baseline, acc_sample],\n \"Precision\": [precision_baseline, precision_sample],\n \"Recall\": [recall_baseline, recall_sample],\n \"F1\": [f1_baseline, f1_sample],\n \"AUC\": [auc_baseline, auc_sample],\n },\n index=[\"baseline\", \"sample\"],\n )\n\n self.performance_comparison = metrics_df","def get_recall_and_precision(wdb, dets, th_conf, th_iou):\n hit_total, false_total, miss_total = 0, 0, 0\n\n for idx in range(wdb.get_image_count()):\n data = wdb.get_annos_by_image_index(idx)\n img_path = data['image_path']\n fname = os.path.splitext(os.path.basename(img_path))[0]\n\n dets4fname = []\n\n for d in dets:\n if d['filename'] == fname: # we have only ONE corresponding element\n dets4fname = d['data']\n break\n\n dets2compare = [e for e in dets4fname if e['c'] >= th_conf] # detections having confidence larger than threshold only\n\n hit, false, miss = count_hits(data['annos'], dets2compare, th_iou)\n hit_total += hit\n false_total += false\n miss_total += miss\n\n recall = hit_total/float(hit_total+miss_total+0.000001)\n precision = hit_total/float(hit_total+false_total+0.000001)\n\n return recall, precision","def evaluate(ground_truth, prediction):\n\n def prfs_to_dict(prfs):\n \"\"\"Returns a precision_recall_fscore_support() result as a dict.\"\"\"\n return {\"precision\": prfs[0], \"recall\": prfs[1], \"fscore\": prfs[2]}\n\n results = {}\n items_count = len(ground_truth)\n\n # accuracy\n accuracy = accuracy_score(ground_truth, prediction)\n results[\"accuracy\"] = accuracy\n\n # confusion matrix\n categories = set(ground_truth) | set(prediction)\n confusions = {\n gold: {pred: 0 for pred in categories} for gold in categories\n }\n for g, p in zip(ground_truth, prediction):\n confusions[g][p] += 1\n results[\"confusions\"] = confusions\n\n # class wise precision, recall & f1\n classwise = precision_recall_fscore_support(\n ground_truth, prediction, average=None, warn_for=()\n )\n results[\"true_cat_dist\"] = list(classwise[-1])\n results[\"classwise\"] = {\n str(cl): prfs_to_dict(\n [classwise[0][cl], classwise[1][cl], classwise[2][cl]]\n )\n for cl in categories\n }\n\n # average precision, recall & f1\n results[\"macro_avg\"] = prfs_to_dict(\n precision_recall_fscore_support(\n ground_truth,\n prediction,\n average=\"macro\",\n pos_label=None,\n warn_for=(),\n )\n )\n results[\"micro_avg\"] = prfs_to_dict(\n precision_recall_fscore_support(\n ground_truth,\n prediction,\n average=\"micro\",\n pos_label=None,\n warn_for=(),\n )\n )\n results[\"weigh_avg\"] = prfs_to_dict(\n precision_recall_fscore_support(\n ground_truth,\n prediction,\n average=\"weighted\",\n pos_label=None,\n warn_for=(),\n )\n )\n\n # marginals\n gold_category_distribution = {\n g: sum([confusions[g][p] for p in categories]) for g in categories\n }\n pred_category_distribution = {\n p: sum([confusions[g][p] for g in categories]) for p in categories\n }\n\n # kappa\n expected_agreement_fleiss = sum(\n [\n (\n (gold_category_distribution[c] + pred_category_distribution[c])\n / (2.0 * items_count)\n )\n ** 2\n for c in categories\n ]\n )\n expected_agreement_cohen = sum(\n [\n (float(gold_category_distribution[c]) / items_count)\n * (float(pred_category_distribution[c]) / items_count)\n for c in categories\n ]\n )\n kappa_fleiss = (\n 1.0\n * (accuracy - expected_agreement_fleiss)\n / (1 - expected_agreement_fleiss)\n )\n kappa_cohen = (\n 1.0\n * (accuracy - expected_agreement_cohen)\n / (1 - expected_agreement_cohen)\n )\n results[\"k_fleiss\"] = {\n \"k\": kappa_fleiss,\n \"AE\": expected_agreement_fleiss,\n \"AO\": accuracy,\n }\n results[\"k_cohen\"] = {\n \"k\": kappa_cohen,\n \"AE\": expected_agreement_cohen,\n \"AO\": accuracy,\n }\n\n return results","def print_precision_recall(classifier, test_set):\n known_set = collections.defaultdict(set)\n computed_set = collections.defaultdict(set)\n \n for i, (features, label) in enumerate(test_set):\n known_set[label].add(i)\n predicted = classifier.classify(features)\n computed_set[predicted].add(i)\n\n print('pos precision:', precision(known_set['pos'], computed_set['pos']))\n print('pos recall:', recall(known_set['pos'], computed_set['pos']))\n print('neg precision:', precision(known_set['neg'], computed_set['neg']))\n print('neg recall:', recall(known_set['neg'], computed_set['neg']))","def detection_analysis(y_pred, y_true):\n print(\"Precision: \", sm.precision_score(y_pred, y_true))\n print(\"Recall: \", sm.recall_score(y_pred, y_true))\n print(\"Accuracy: \", sm.accuracy_score(y_pred, y_true))\n print(\"\\n\")","def calculate_precision_recall_all_images(\n all_prediction_boxes, all_gt_boxes, iou_threshold):\n # Find total true positives, false positives and false negatives\n # over all images\n\n tot_num_tp, tot_num_fp, tot_num_fn = 0, 0, 0\n\n for image in range(len(all_prediction_boxes)):\n prediction_boxes = all_prediction_boxes[image]\n gt_boxes = all_gt_boxes[image]\n\n image_results = calculate_individual_image_result(\n prediction_boxes,\n gt_boxes,\n iou_threshold)\n\n tot_num_tp += image_results[\"true_pos\"]\n tot_num_fp += image_results[\"false_pos\"]\n tot_num_fn += image_results[\"false_neg\"]\n\n # Compute precision, recall\n precision = calculate_precision(tot_num_tp, tot_num_fp, tot_num_fn)\n recall = calculate_recall(tot_num_tp, tot_num_fp, tot_num_fn)\n\n return (precision, recall)","def precision_recall(conf_matrix):\n tp_count, fp_count, fn_count, tn_count = conf_matrix[0][0], conf_matrix[0][1], conf_matrix[1][0], conf_matrix[1][1]\n\n precision = tp_count / (tp_count + fp_count)\n recall = tp_count / (tp_count + fn_count)\n\n return precision, recall","def painting_matching_ml(imgs, db_imgs, method_list, text_masks, author_text, gt_text, metrics, weights, splits=30, max_rank=10):\n descriptor_extractors = [get_descriptor_extractor(method_name) for method_name in method_list]\n tmp_img_format = []\n tmp_mask_format = []\n tmp_text_format = []\n for i in range(len(imgs)):\n for j in range(len(imgs[i])):\n tmp_img_format.append(imgs[i][j])\n tmp_mask_format.append(text_masks[i][j])\n tmp_text_format.append(author_text[i][j])\n\n #db_imgs = [img[0] for img in db_imgs]\n db_img_splits = [i*len(db_imgs)//splits for i in range(splits-1)]\n \n scores = []\n query_descriptors = extract_descriptors(tmp_img_format, descriptor_extractors, method_list, tmp_text_format, tmp_mask_format) \n #np.array([extract_descriptors(img, matching_methods, mask) for img, mask in zip(tmp_img_format, tmp_mask_format)])\n print(\"Starting db extraction + matching\")\n for split in tqdm(range(splits-2)):\n db_descriptors = extract_descriptors(db_imgs[db_img_splits[split]:db_img_splits[split+1]], descriptor_extractors, method_list, gt_text[db_img_splits[split]:db_img_splits[split+1]], None) #np.array([mrhm(db_img) for db_img in db_imgs[db_img_splits[split]:db_img_splits[split+1]]])\n scores.append(compare_descriptors(query_descriptors, db_descriptors, metrics, method_list, weights))\n # compare_descriptors(query_descriptors, db_descriptors, descriptor_comp_methods, descriptor_names, weights)\n db_descriptors = extract_descriptors(db_imgs[db_img_splits[-1]:], descriptor_extractors, method_list, gt_text[db_img_splits[-1]:], None)\n scores.append(compare_descriptors(query_descriptors, db_descriptors, metrics, method_list, weights))\n \n # concatenate all the results\n scores = np.concatenate(scores, 1)\n \n top_k_matches = np.argpartition(scores, list(range(max_rank)))[:, :max_rank]\n return top_k_matches","def recall(prediction, ground_truth):\n return precision_recall_f1(prediction, ground_truth)[1]","def precision(y_true, y_pred, average, labels):\n y_true, y_pred = check_metric_args(y_true, y_pred, average, labels)\n\n # At this point, you can be sure that y_true and y_pred are one hot encoded.\n result = None\n m = len(y_true)\n n = len(labels)\n\n #call get_confusion_matrix function and put the result in confusion_matrix\n confusion_matrix = get_confusion_matrix(y_true, y_pred, labels)\n\n #compute the result if using micro-averages\n if average == \"micro\":\n numerator = np.trace(confusion_matrix)\n denominator = np.sum(confusion_matrix)\n result = numerator/denominator\n\n #compute the precision independently for each class and then take the average \n elif average == \"macro\":\n diag = np.diag(confusion_matrix)\n row_sums = np.sum(confusion_matrix, axis=1)\n row_sums_adjusted = np.array([1 if val == 0 else val for val in row_sums])\n result = np.mean(diag/row_sums_adjusted)\n\n else:\n diag = np.diag(confusion_matrix)\n row_sums = np.sum(confusion_matrix, axis=1)\n row_sums_adjusted = np.array([1 if val == 0 else val for val in row_sums])\n result = diag/row_sums_adjusted\n\n return result","def add_example(self, precision: float, recall: float, f1: float):\n self._count += 1\n self._total_precision += precision\n self._total_recall += recall\n self._total_f1 += f1","def compare_recall_precisions_from_predictions(true, score_dict, **kwargs):\n pr = OrderedDict()\n for name, score in score_dict.items():\n p, r, threshold = precision_recall_curve(true, score)\n pr[name] = [p, r]\n compare_recall_precision_graph(pr, **kwargs)","def score(label_recall, label_precision, pred_recall, pred_precision):\n f11 = 6*label_recall*label_precision/(5*label_recall+label_precision)\n f12 = 5*pred_recall*pred_precision/(2*pred_recall+3*pred_precision)\n\n return 0.4*f11 + 0.6*f12","def prf_cal(y_pred,y_true,k):\r\n GT=np.sum(y_true[y_true==1.])\r\n instance_num=y_true.shape[0]\r\n prediction_num=instance_num*k\r\n\r\n sort_indices = np.argsort(y_pred)\r\n sort_indices=sort_indices[:,::-1]\r\n static_indices = np.indices(sort_indices.shape)\r\n sorted_annotation= y_true[static_indices[0],sort_indices]\r\n top_k_annotation=sorted_annotation[:,0:k]\r\n TP=np.sum(top_k_annotation[top_k_annotation==1.])\r\n recall=TP/GT\r\n precision=TP/prediction_num\r\n f1=2.*recall*precision/(recall+precision)\r\n return precision, recall, f1","def PredictMatch(yourName, tgtName, teamModels, db):\n\n teamTgtPipe, teamYourPipe = QueryTeamData(tgtName, yourName, db)\n featureCoefTgt, probTgt = PredictOp(teamTgtPipe, teamYourPipe, tgtName, teamModels)\n featureCoefYour, probYour = PredictOp(teamYourPipe, teamTgtPipe, yourName, teamModels)\n odds = round(probYour / (probYour + probTgt), 2)\n\n # In featureCoefYour, you want INCREASE those with POSTIVE COEF, DECREASE those with NEGATIVE COEF\n # In featureCoefTgt, you want to do the opposite\n\n # reverse both the sign of the coef, and '_op' in features so as to be the same with featureCoefYour\n featureCoefTgt['coef'] = - featureCoefTgt['coef']\n featureCoefTgt.features = [ii[:-3] if \"_op\" in ii else ii + '_op' for ii in featureCoefTgt.features]\n\n # Combine only the most important 10 features\n # featureBoth = featureCoefTgt[11:].append(featureCoefYour[11:])\n\n # Combine only all the most important features\n featureBoth = featureCoefTgt.append(featureCoefYour)\n\n # get action recommendations\n # Somehow the pandas here uses a deprecated para cols, instaed of the new one subset\n #featureBoth.drop_duplicates(subset = 'features', take_last = True, inplace = True)\n featureBoth.drop_duplicates(cols = 'features', take_last = True, inplace = True)\n actions, featureImprove = GetActions(featureBoth)\n Imp = 0.1\n oddsNew = ImprovedScore(tgtName, yourName, teamModels, featureImprove, teamTgtPipe, teamYourPipe, Imp)\n\n return odds, oddsNew, actions","def precision_recall_calculator(reactions_removed, reactions_restored):\n \n reactions_removed = set(reactions_removed)\n reactions_restored = set(reactions_restored)\n \n true_positives = reactions_removed.intersection(reactions_restored)\n \n precision = len(true_positives)/len(reactions_restored)\n recall = len(true_positives)/len(reactions_removed)\n \n return precision, recall","def evaluate(self, test_dir='data/dev', target='real'):\n test_data = {c: os.path.join(test_dir, c) for c in self.classes}\n if not target in test_data:\n print('Error: target class does not exist in test data.')\n return\n outcomes = {'TP': 0, 'TN': 0, 'FP': 0, 'FN': 0}\n # >>> YOUR ANSWER HERE\n data = []\n for c in test_data:\n docs = open(test_data[c]).readlines()\n for doc in docs:\n preprocess_doc = doc.strip()\n data.append((c, preprocess_doc))\n for item in data:\n predict_ans = self.predict(item[1])\n if item[0] == 'real':\n if predict_ans == 'real':\n outcomes['TP'] += 1\n else:\n outcomes['FN'] += 1\n else:\n if predict_ans == 'real':\n outcomes['FP'] += 1\n else:\n outcomes['TN'] += 1\n precision = outcomes['TP'] / (outcomes['TP'] + outcomes['FP']) # replace with equation for precision\n recall = outcomes['TP'] / (outcomes['TP'] + outcomes['FN']) # replace with equation for recall\n f1_score = 2 * ((precision * recall) / (precision + recall)) # replace with equation for f1\n # >>> END YOUR ANSWER\n return precision, recall, f1_score","def evaluate_old(goldfile, sysfile):\n gold = {i[\"id\"]: i[\"label\"] for i in read(goldfile)}\n sys = {i[\"id\"]: i[\"prediction\"] for i in read(sysfile)}\n\n labels, preds = [], []\n for idx in gold:\n labels.append(gold[idx])\n preds.append(sys[idx])\n\n precision_macro, recall_macro, f1_macro, _ = precision_recall_fscore_support(labels, preds, labels=[0, 1],\n average=\"macro\")\n print(\"F1-macro score for test data predictions are: %.4f\" % f1_macro)","def __precision_and_recall(self, y_train, y_pred, average='micro', verbose = True):\n # Precision and recall\n pscore = precision_score(y_train, y_pred, average=average)\n rscore = recall_score(y_train, y_pred, average=average)\n\n # harmonic mean of precision\n f1 = f1_score(y_train, y_pred, average='micro')\n\n if verbose:\n message = F\"Precision of model: {pscore} | Recall of model {rscore}\"\n self.__logger.info(message)\n\n message = f\"Harmonic Mean of precision and recall: F1 score: {f1}\"\n self.__logger.info(message)\n return str(f1)","def test_calculate_precision_happy_path(\n mock_engine: DicomImagePiiVerifyEngine,\n gt_labels: List[dict],\n all_pos: List[dict],\n expected_result: float,\n):\n # Act\n test_precision = mock_engine.calculate_precision(gt_labels, all_pos)\n\n # Assert\n assert test_precision == expected_result","def _imequalize(self, results):\n for key in results.get('img_fields', ['image']):\n img = results[key]\n results[key] = mmcv.imequalize(img).astype(img.dtype)","def prf_analysis(y_true: list, y_pred: list) -> None:\n print('Precision: {:,.2f}'.format(precision_score(y_true, y_pred)))\n print('Recall : {:,.2f}'.format(recall_score(y_true, y_pred)))\n print('F1 : {:,.2f}'.format(f1_score(y_true, y_pred)))\n print('Accuracy : {:,.2f}'.format(accuracy_score(y_true, y_pred)))\n return None","def compute(self) -> None:\n \n self.model.eval()\n \n with torch.no_grad():\n for (input, target, _) in self.loader:\n\n # self.model = self.model.train(False) # TEST @lacoupe\n output, _ = self.model(input)\n \n output = (output >= 0.5)\n \n for out, tar in zip(output, target):\n \n tar = bool(tar)\n \n if out and tar:\n self.confusion['true_positive'] += 1\n elif not out and not tar:\n self.confusion['true_negative'] += 1\n elif out and not tar:\n self.confusion['false_positive'] += 1\n elif not out and tar:\n self.confusion['false_negative'] += 1\n \n self.accuracy = (self.confusion['true_positive'] + self.confusion['true_negative']) \\\n / sum(list(self.confusion.values()))\n \n if (self.confusion['true_positive'] + self.confusion['false_positive']) == 0.:\n self.precision = 0.\n else:\n self.precision = self.confusion['true_positive'] \\\n / (self.confusion['true_positive'] + self.confusion['false_positive'])\n \n if (self.confusion['true_positive'] + self.confusion['false_negative']) == 0.:\n self.recall = 0.\n else:\n self.recall = self.confusion['true_positive'] \\\n / (self.confusion['true_positive'] + self.confusion['false_negative'])\n \n if (self.precision + self.recall) == 0.:\n self.f1_score = 0.\n else:\n self.f1_score = 2 * self.precision * self.recall / (self.precision + self.recall)","def calc_accuracy(self, top_matches):\n # Calculate Translation accuracy\n # Accuracy at K\n for k in [1, 5, 10]:\n top_k_matches = top_matches[:, :k]\n # Checks for one match\n one_matching = (top_k_matches == self.tgt_ind[:, None]).sum(1)\n # Checks for multiple translations\n matching = {}\n for i, src_id in enumerate(self.src_ind):\n matching[src_id] = min(matching.get(src_id, 0) + one_matching[i], 1)\n precision_at_k = 100 * np.mean(list(matching.values()))\n print(\"%i source words - Precision at k = %i: %f\" % (len(matching), k, precision_at_k))","def accuracy(self):","def evaluate(clf, dataset, feature_list, features, labels, num_iter, params):\n\n features_train, features_test, labels_train, labels_test = \\\n train_test_split(features, labels, test_size=0.3, random_state=42)\n\n\n\n precision_values = []\n recall_values = []\n accuracy_values = []\n print clf\n for i in xrange(0, num_iter):\n #print params\n clf = GridSearchCV(clf, params)\n clf.fit(features_train, labels_train)\n print '*****************************'\n print clf.best_estimator_\n print clf.best_params_\n\n clf = clf.best_estimator_\n #test_classifier(clf, dataset, feature_list)\n pred = clf.predict(features_test)\n precision_values.append(precision_score(labels_test, pred))\n recall_values.append(recall_score(labels_test, pred))\n accuracy_values.append(accuracy_score(labels_test, pred))\n print 'Recall score: ', mean(recall_values)\n print 'Precision score: ', mean(precision_values)\n print 'Accuracy score: ' , mean(accuracy_values)","def __call__(self, query_result: List, gallery_info: List) -> (float, Dict):\n aps = list()\n\n # For mAP calculation\n pseudo_score = np.arange(0, len(gallery_info))[::-1]\n\n recall_at_k = dict()\n for k in self._hyper_params[\"recall_k\"]:\n recall_at_k[k] = 0\n\n gallery_label = np.array([gallery_info[idx][\"label_idx\"] for idx in range(len(gallery_info))])\n for i in range(len(query_result)):\n ranked_idx = query_result[i][\"ranked_neighbors_idx\"]\n gt = (gallery_label[query_result[i][\"ranked_neighbors_idx\"]] == query_result[i][\"label_idx\"])\n aps.append(average_precision_score(gt, pseudo_score[:len(gt)]))\n\n # deal with 'gallery as query' test\n if gallery_info[ranked_idx[0]][\"path\"] == query_result[i][\"path\"]:\n gt.pop(0)\n self.compute_recall_at_k(gt, recall_at_k)\n\n mAP = np.mean(aps) * 100\n\n for k in recall_at_k:\n recall_at_k[k] = recall_at_k[k] * 100 / len(query_result)\n\n return mAP, recall_at_k","def precision1(y_test, predictions):\n precision = 0.0\n\n for i in range(y_test.shape[0]):\n intersection = 0.0\n hXi = 0.0\n for j in range(y_test.shape[1]):\n hXi = hXi + int(predictions[i,j])\n if int(y_test[i,j]) == 1 and int(predictions[i,j]) == 1:\n intersection += 1\n \n if hXi != 0:\n precision = precision + float(intersection/hXi)\n \n\n precision = float(precision/y_test.shape[0])\n \n return precision","def update(self) -> None:\n self.faithful = self.is_faithful()\n if self.faithful:\n old_class_names = self.class_names\n old_training_image_names = self.training_image_names\n self.class_names = self.find_class_names()\n self.training_image_names = self.find_training_image_names()\n self.extracted_features = list()\n if old_class_names != self.class_names or old_training_image_names != self.training_image_names:\n self.generate_csv_dictionary()\n return","def test_calculate_recall_happy_path(\n mock_engine: DicomImagePiiVerifyEngine,\n gt_labels: dict,\n all_pos: dict,\n expected_result: float,\n):\n # Act\n test_recall = mock_engine.calculate_recall(gt_labels, all_pos)\n\n # Assert\n assert test_recall == expected_result","def compute_average_precision_per_class(num_true_cases, gt_boxes,\n\t\t\t\t\t\t\t\t\t\tprediction_file, iou_threshold, use_2007_metric):\n\twith open(prediction_file) as f:\n\t\timage_ids = []\n\t\tboxes = []\n\t\tscores = []\n\t\tfor line in f:\n\t\t\tt = line.rstrip().split(\" \")\n\t\t\timage_ids.append(t[0])\n\t\t\tscores.append(float(t[1]))\n\t\t\tbox = torch.tensor([float(v) for v in t[2:]]).unsqueeze(0)\n\t\t\tbox -= 1.0 # convert to python format where indexes start from 0\n\t\t\tboxes.append(box)\n\t\tscores = np.array(scores)\n\t\tsorted_indexes = np.argsort(-scores)\n\t\tboxes = [boxes[i] for i in sorted_indexes]\n\t\timage_ids = [image_ids[i] for i in sorted_indexes]\n\t\ttrue_positive = np.zeros(len(image_ids))\n\t\tfalse_positive = np.zeros(len(image_ids))\n\t\tmatched = set()\n\t\tfor i, image_id in enumerate(image_ids):\n\t\t\tbox = boxes[i]\n\t\t\tif image_id not in gt_boxes:\n\t\t\t\tfalse_positive[i] = 1\n\t\t\t\tcontinue\n\n\t\t\tgt_box = gt_boxes[image_id]\n\t\t\tious = box_utils.iou_of(box, gt_box)\n\t\t\tmax_iou = torch.max(ious).item()\n\t\t\tmax_arg = torch.argmax(ious).item()\n\t\t\tif max_iou > iou_threshold:\n\t\t\t\tif (image_id, max_arg) not in matched:\n\t\t\t\t\ttrue_positive[i] = 1\n\t\t\t\t\tmatched.add((image_id, max_arg))\n\t\t\t\telse:\n\t\t\t\t\tfalse_positive[i] = 1\n\t\t\telse:\n\t\t\t\tfalse_positive[i] = 1\n\n\ttrue_positive = true_positive.cumsum()\n\tfalse_positive = false_positive.cumsum()\n\tprecision = true_positive / (true_positive + false_positive)\n\trecall = true_positive / num_true_cases\n\tif use_2007_metric:\n\t\treturn measurements.compute_voc2007_average_precision(precision, recall)\n\telse:\n\t\treturn measurements.compute_average_precision(precision, recall)","def testClassifier(x_train, y_train, x_test, y_test, clf):\n #metrics = []\n start = dt.now()\n clf.fit(x_train, y_train)\n end = dt.now()\n print 'training time: ', (end - start)\n \n # add training time to metrics\n #metrics.append(end-start)\n \n start = dt.now()\n yhat = clf.predict(x_test)\n end = dt.now()\n print 'testing time: ', (end - start)\n \n # add testing time to metrics\n #metrics.append(end-start)\n \n print 'classification report: '\n# print classification_report(y_test, yhat)\n pp(classification_report(y_test, yhat))\n \n print 'f1 score'\n print f1_score(y_test, yhat, average='macro')\n \n print 'accuracy score'\n accuracy = accuracy_score(y_test, yhat)\n print accuracy\n #metrics.append(accuracy)\n #precision = precision_score(y_test, yhat, average=None)\n #recall = recall_score(y_test, yhat, average=None)\n \n # add precision and recall values to metrics\n #for p, r in zip(precision, recall):\n # metrics.append(p)\n # metrics.append(r)\n \n \n #add macro-averaged F1 score to metrics\n #metrics.append(f1_score(y_test, yhat, average='macro'))\n \n print 'confusion matrix:'\n print confusion_matrix(y_test, yhat)\n \n # plot the confusion matrix\n plt.imshow(confusion_matrix(y_test, yhat), interpolation='nearest')\n plt.show()\n \n return accuracy","def do_score_prediction(self):\n \n import iread.myio as mio\n from igui.score_canvas import ScoreCanvas\n exp_name = 'JPS_act_12_exp_4_accv_half_fc_j2'\n exp_name_base = 'ASM_act_12_exp_4'\n exp_base_folder = '/opt/visal/tmp/for_sijin/Data/H36M/H36MExp'\n exp_path = iu.fullfile(exp_base_folder, 'folder_%s' % exp_name, 'batches.meta')\n meta_base_path = iu.fullfile(exp_base_folder, 'folder_%s' % exp_name_base, 'batches.meta')\n meta = mio.unpickle(exp_path)\n meta_base = mio.unpickle(meta_base_path)\n images_path = meta_base['images_path']\n \n pred_pose = meta['feature_list'][0]\n gt_pose = meta['random_feature_list'][0]\n ntotal = gt_pose.shape[-1]\n print 'gt_pose_shape',gt_pose.shape\n print 'pred_pose_shape', pred_pose.shape\n ref_frame = 7600 # This is the index in test range\n ## ref_frame = 2600 # This is the index in test range\n test_range = self.test_data_provider.feature_range\n ref_idx = test_range[ref_frame]\n \n n_to_show = 1000\n \n idx_to_show = np.random.choice(ntotal, n_to_show - 1)\n idx_to_show = [ref_idx] + idx_to_show.tolist() \n idx_to_show = np.asarray(idx_to_show, dtype=np.int).flatten()\n \n ref_pose = pred_pose[...,ref_idx].reshape((-1,1),order='F') \n pose_to_eval =gt_pose[...,idx_to_show]\n output_feature_name = 'fc_2' # <------------------Parameter\n output_layer_idx = self.get_layer_idx(output_feature_name)\n\n # do it once <------------- Maybe it can support multiple batch in the future\n data_dim = self.model_state['layers'][output_layer_idx]['outputs']\n print 'data_dim', data_dim\n \n cur_data = [np.require(np.tile(ref_pose, [1,n_to_show]), \\\n dtype=np.single,requirements='C'), \\\n np.require(pose_to_eval.reshape((-1,n_to_show),order='F'),\\\n dtype=np.single,requirements='C'), \\\n np.require(np.zeros((1,n_to_show),dtype=np.single), \\\n requirements='C'),\n np.require(np.zeros((n_to_show,data_dim),dtype=np.single), \\\n requirements='C')]\n residuals = cur_data[1][...,0].reshape((-1,1),order='F') - cur_data[1]\n dp = self.test_data_provider\n mpjpe = dutils.calc_mpjpe_from_residual(residuals, dp.num_joints)\n\n gt_score = dp.calc_score(mpjpe, dp.mpjpe_factor/dp.max_depth,\\\n dp.mpjpe_offset/dp.max_depth).reshape((1,n_to_show)).flatten()\n self.libmodel.startFeatureWriter(cur_data, output_layer_idx)\n self.finish_batch()\n score = cur_data[-1].T\n print 'dim score', score.shape, 'dim gt_score', gt_score.shape\n score = score.flatten()\n # score = gt_score.flatten()\n def my_sort_f(k):\n if k == 0:\n return 10000000\n else:\n return score[k]\n sorted_idx = sorted(range(n_to_show), key=my_sort_f,reverse=True)\n s_to_show = [idx_to_show[k] for k in sorted_idx]\n sorted_score = np.asarray( [score[k] for k in sorted_idx])\n \n pose_to_plot = self.convert_relskel2rel(cur_data[1])\n sorted_pose = pose_to_plot[...,sorted_idx]\n class ScorePoseCanvas(ScoreCanvas):\n def __init__(self,data_dic):\n import iread.h36m_hmlpe as h36m\n ScoreCanvas.__init__(self,data_dic)\n self.pose_data = data_dic['pose_data']\n self.limbs = h36m.part_idx\n self.tmp = 0\n def show_image(self,ax):\n # ScoreCanvas.show_image(self,ax)\n # return\n import Image\n idx =self.cur_data_idx\n if idx == 0:\n self.tmp = self.tmp + 1\n if self.tmp == 1:\n img = self.load_image(idx)\n ax.imshow(np.asarray(img))\n return\n print 'Current data idx %d ' % self.cur_data_idx\n # params = {'elev':-89, 'azim':-107}\n # params = {'elev':-69, 'azim':-107}\n params = {'elev':-81, 'azim':-91} # frontal view\n fig = plt.figure(100)\n from mpl_toolkits.mplot3d import Axes3D\n import imgproc\n # new_ax = self.fig.add_axes( rng_rel,projection='polar')\n new_ax = fig.add_subplot(111,projection='3d')\n imgproc.turn_off_axis(new_ax)\n cur_pose = self.pose_data[...,idx].reshape((3,-1),order='F')\n dutils.show_3d_skeleton(cur_pose.T,\\\n self.limbs, params)\n xmin,xmax = np.min(cur_pose[0]),np.max(cur_pose[0])\n ymin,ymax = np.min(cur_pose[1]),np.max(cur_pose[1])\n zmin,zmax = np.min(cur_pose[2]),np.max(cur_pose[2])\n def extent(x,y,ratio):\n x = x + (x-y) * ratio\n y = y + (y-x) * ratio\n return -0.4,0.4\n r = 0.1\n new_ax.set_xlim(extent(xmin,xmax,r))\n new_ax.set_ylim(extent(ymin,ymax,r))\n new_ax.set_ylim(extent(zmin,zmax,r))\n tmp_folder = '/public/sijinli2/ibuffer/2014-CVPR2015/tmp/images'\n save_path = iu.fullfile(tmp_folder, 'tmp_image.png')\n plt.savefig(save_path)\n img = Image.open(save_path)\n plt.close(100)\n img_arr = np.asarray(img)\n s = np.int(img_arr.shape[0]/5.0)\n e = np.int(img_arr.shape[0] - s)\n s = 0\n e = img_arr.shape[0]\n img_arr = img_arr[s:e,:,:]\n ax.imshow(np.asarray(img_arr))\n # ax.plot([1,0,0],[0,1,0],[0,0,1])\n\n\n sc = ScorePoseCanvas({'x': np.asarray(range(len(idx_to_show))), 'y':sorted_score,\\\n 'images_path': [images_path[k] for k in s_to_show], \\\n 'pose_data':sorted_pose})\n sc.start()\n print 'max score is ' , sorted_score.max()\n gt_sort_idx = sorted(range(n_to_show), key=lambda k:gt_score[k], reverse=True)\n sorted_gt_score = np.asarray([gt_score[k] for k in gt_sort_idx])\n sorted_score_by_gt = [score[k] for k in gt_sort_idx]\n pl.plot(np.asarray(range(n_to_show)), sorted_gt_score, 'r', label='gt_score')\n pl.plot(np.asarray(range(n_to_show)), sorted_score_by_gt, 'g', label='pred_score')","def compute_and_print_eval_metrics(self):\n s = ('%20s' + '%12s' * 6) % ('Class', 'Images', 'Targets', 'P', 'R', 'mAP@.5', 'mAP@.5:.95')\n precision, recall, f1, mean_precision, mean_recall, map50, map = 0., 0., 0., 0., 0., 0., 0.\n ap = []\n eval_stats = [np.concatenate(x, 0) for x in zip(*self.eval_stats)]\n if len(eval_stats) and eval_stats[0].any():\n precision, recall, ap, f1, ap_class = ap_per_class(*eval_stats)\n precision, recall, ap50, ap = precision[:, 0], recall[:, 0], ap[:, 0], ap.mean(1)\n mean_precision, mean_recall, map50, map = precision.mean(), recall.mean(), ap50.mean(), ap.mean()\n nt = np.bincount(eval_stats[3].astype(np.int64), minlength=len(self.class_names)) # number of targets per class\n else:\n nt = np.zeros(1)\n\n pf = '%20s' + '%12.5g' * 6 # print format\n print(\"\\n EVALUTAION \\n\")\n print(s)\n print(pf % ('all', self.seen, nt.sum(), mean_precision, mean_recall, map50, map))\n if self.cfg.eval.verbose:\n for indx, cls in enumerate(ap_class):\n print(pf % (self.class_names[cls], self.seen, nt[cls], precision[indx], recall[indx], ap50[indx], ap[indx]))","def ratio_test(fx2_match, fx1_match, fx1_norm, match_dist, norm_dist,\n ratio_thresh=.625, fm_dtype=np.int32, fs_dtype=np.float32):\n fx2_to_ratio = np.divide(match_dist, norm_dist).astype(fs_dtype)\n fx2_to_isvalid = np.less(fx2_to_ratio, ratio_thresh)\n fx2_match_RAT = fx2_match.compress(fx2_to_isvalid).astype(fm_dtype)\n fx1_match_RAT = fx1_match.compress(fx2_to_isvalid).astype(fm_dtype)\n fx1_norm_RAT = fx1_norm.compress(fx2_to_isvalid).astype(fm_dtype)\n # Turn the ratio into a score\n fs_RAT = np.subtract(1.0, fx2_to_ratio.compress(fx2_to_isvalid))\n fm_RAT = np.vstack((fx1_match_RAT, fx2_match_RAT)).T\n # return normalizer info as well\n fm_norm_RAT = np.vstack((fx1_norm_RAT, fx2_match_RAT)).T\n ratio_tup = MatchTup3(fm_RAT, fs_RAT, fm_norm_RAT)\n return ratio_tup","def precision(classifier_output, true_labels):\n\n # TODO: finish this.\n conf_matrix = confusion_matrix(classifier_output, true_labels)\n return conf_matrix[0][0]/(conf_matrix[0][0] + conf_matrix[1][0])","def _update(self, results):\n logger = getLogger(\"problog_lfi\")\n fact_body = defaultdict(int)\n fact_par = defaultdict(int)\n\n score = 0.0\n for m, pEvidence, result in results:\n par_marg = dict()\n for fact, value in result.items():\n # use the id and the t variables as index\n index = (fact.args[0], fact.args[1])\n if fact.functor == \"lfi_body\":\n fact_body[index] += value * m\n elif fact.functor == \"lfi_par\":\n if index in par_marg:\n par_marg[index] += value\n for o_index in self._adatomc[index[0]]:\n par_marg[(o_index, *index[1:])] += value\n else:\n par_marg[index] = value\n for o_index in self._adatomc[index[0]]:\n par_marg[(o_index, *index[1:])] = value\n\n for index, value in par_marg.items():\n fact_par[index] += value * m\n try:\n score += math.log(pEvidence)\n except ValueError:\n logger.debug(\"Pr(evidence) == 0.0\")\n\n update_list = fact_body\n\n weight_changed = []\n for weight in self._weights:\n if isinstance(weight, float):\n weight_changed.append(False)\n elif isinstance(weight, dict):\n d = dict()\n for w in weight:\n d[w] = False\n weight_changed.append(d)\n\n score = 0.0\n for index in update_list:\n if float(fact_body[index]) <= 10**-15:\n # if close to zero\n prob = 0.0\n else:\n prob = float(fact_body[index]) / float(fact_par[index])\n try:\n score += math.log(prob)\n except ValueError as ex:\n # prob too close to zero\n pass\n\n logger.debug(\n \"Update probabilistic fact {}: {} / {} = {}\".format(\n index, fact_body[index], fact_par[index], prob\n )\n )\n self._set_weight(index[0], index[1], prob, weight_changed=weight_changed)\n if not index[1]:\n weight_changed[int(index[0])] = True\n elif (\n isinstance(weight_changed[int(index[0])], dict)\n and index[1] in weight_changed[int(index[0])]\n ):\n weight_changed[int(index[0])][index[1]] = True\n else:\n weight_changed[int(index[0])] = {index[1]: True}\n\n if self._enable_normalize:\n self._normalize_weights()\n\n return score","def evaluate(self, test_data):\n test_results = [(self.decide(self.feedforward(x)), y)\n for (x, y) in test_data]\n\n # Check for and keep track of TP's FP's and FN's\n # Write FP's and FN's to a special directories\n TP = FP = FN = count = 0\n TN = 0\n # what you got = x, what should be = y\n for ((x, y), (image, gt)) in zip(test_results, test_data):\n count += 1\n if x == 1 and y == 1:\n TP += 1\n elif x == 1 and y == 0:\n FP += 1\n if not self.validating: self.save_Image(FP_PATH, count, image)\n elif x == 0 and y == 1:\n FN += 1\n if not self.validating: self.save_Image(FN_PATH, count, image)\n \n if (TP + FP ) == 0:\n return TP, FP, FN, '---------', '---------','---------'\n else:\n false_rate = float(FP) / float(TP + FP)\n\n if (TP + FN ) == 0:\n return TP, FP, FN, '---------', '---------', false_rate\n else:\n detect_rate = float(TP) / float(TP + FN)\n\n if (TP + FP + FN ) == 0:\n return TP, FP, FN, '---------', detect_rate, false_rate\n else:\n quality_rate = float(TP) / float(TP+FP+FN)\n \n return TP, FP, FN, quality_rate, detect_rate, false_rate","def calculate_scores(self):\n # Prediction based scores\n #self.report = classification_report(self.y_test, self.y_pred)\n self.accuracy = accuracy_score(self.y_real, self.y_pred)\n self.precision = precision_score(self.y_real, self.y_pred)\n self.recall = recall_score(self.y_real, self.y_pred)\n self.f1 = f1_score(self.y_real, self.y_pred)\n \n # Probability based scores\n self.fpr, self.tpr, _ = roc_curve(self.y_real, self.y_proba)\n self.average_precision = average_precision_score(self.y_real, self.y_proba)\n self.brier_loss = brier_score_loss(self.y_real, self.y_proba)\n self.roc_auc = roc_auc_score(self.y_real, self.y_proba)\n self.prec_cur, self.recall_cur, _ = precision_recall_curve(self.y_real, self.y_proba)","def update_score_other_rounds(self, match):\n self.match_views.update_score(match)","def update_predictions(self, context):\n x, y, o = context.get_predictions()\n self.x_eval += x\n self.y_eval += y\n self.o_eval += o\n self.write_predictions(o)","def evaluate(gold, predictions):\n # gold = {i[\"id\"]:i[\"label\"] for i in read(goldfile)}\n # sys = {i[\"id\"]:i[\"prediction\"] for i in read(sysfile)}\n\n labels, preds = [], []\n for idx in gold:\n labels.append(gold[idx])\n preds.append(predictions[idx])\n\n precision_macro, recall_macro, f1_macro, _ = precision_recall_fscore_support(labels, preds, labels=[0, 1],\n average=\"macro\")\n print(\"F1-macro score for test data predictions are: %.4f\" % f1_macro)\n return preds","def compare(model, input1, input2): \n patch_image_time_start = time.time()\n input1_patches = get_patches_non_overlap(input1, 48, 48)\n input2_patches = get_patches_non_overlap(input2, 48, 48)\n patch_image_time_end = time.time()\n compare_image_time_start = time.time()\n pred = model.predict([input1_patches, input2_patches])\n compare_image_time_end = time.time()\n\n result[\"patch_retrieval_time\"].append(patch_image_time_end - patch_image_time_start)\n result[\"image_comparison_time\"].append(compare_image_time_end - compare_image_time_start)\n\n return np.sum(pred)","def computeMatchInfo( df, inpCol, predCol, true_label=\"label\", method=\"---\", seuil=0. ) :\n df_data = predict( df, inpCol, predCol, seuil=seuil )\n df_ca = computeClassAtt( df_data[true_label], df_data[predCol], method )\n df_ca[\"threshold\"] = seuil\n\n return df_data, df_ca","def evaluate(self):\n tic = time.time()\n print('Running per image evaluation...')\n p = self.params\n # add backward compatibility if useSegm is specified in params\n if p.useSegm is not None:\n p.iouType = 'segm' if p.useSegm == 1 else 'bbox'\n print(\n \"useSegm (deprecated) is not None.\"\n \"Running {} evaluation\".format(p.iouType)\n )\n print('Evaluate annotation type *{}*'.format(p.iouType))\n p.imgIds = list(np.unique(p.imgIds))\n if p.useCats:\n p.catIds = list(np.unique(p.catIds))\n p.maxDets = sorted(p.maxDets)\n self.params = p\n\n self._prepare()\n # loop through images, area range, max detection number\n catIds = p.catIds if p.useCats else [-1]\n\n self.ious = {\n (imgId, catId): self.computeIoU(imgId, catId)\n for imgId in p.imgIds\n for catId in catIds\n }\n # ignore any categories that is not having any attributes\n self.f1s = {\n (imgId, catId): self.computeF1(imgId, catId)\n for imgId in p.imgIds\n for catId in catIds\n if catId in self.FPParams.catsWithAttributes\n }\n\n # self.gt_attributes_ids = []\n\n # loop through images, area range, max detection number\n self.evalImgs = [\n self.evaluateImg(imgId, catId, areaRng, p.maxDets[-1])\n for catId in catIds\n for areaRng in p.areaRng\n for imgId in p.imgIds\n ]\n # self._paramsEval = copy.deepcopy(self.params) # seems do not need it\n toc = time.time()\n print('DONE (t={:0.2f}s).'.format(toc-tic))","def calculate_many_precs_recall_stats(test_class=test_class,\n local_list=local_list,\n Q_stop='',\n no_files_in_label=no_files_in_label,\n no_of_images=no_of_images,\n verbose=verbose):\n if not Q_stop == '':\n Q = Q_stop\n else:\n Q = len(local_list)\n # does all points\n if Q > len(local_list):\n Q = len(local_list)\n if verbose:\n print('The number of points to check (Q) {} is more than the number of points above minx {}'.format(Q, len(local_list)))\n print('Setting Q to len(local_list), you may wish to change your settings')\n N_test = no_files_in_label[test_class] # no of items in class A\n N_not_test = no_of_images - N_test # i.e. set not-A\n # set up counters\n AP = 0 # average precision\n found_recall_precs_p95 = False\n recall_p95 = 0.0\n count_of_test_class = 0\n # loop backwards through the list, abs j is the position in a 1-indexed list\n recall_x = 0 # n.b. this is also sensitivity\n Ave_precs_x = 0\n specificity_x = 0\n count_of_false_positives = 0 # number of not-A on misidentified\n max_informedness = 0\n x_for_max_informedness = 0\n max_f1_stat = 0\n x_for_max_f1 = 0\n recall_for_max_informedness = 0\n specificity_for_max_informedness = 0\n for i in range(Q):\n j = -(i + 1) # 1 indexed\n recall_x_minus_1 = recall_x\n current_class = local_list[j][0]\n #j = j - 1 # really this is here so we can check j\n # break\n if count_of_test_class == N_test:\n # we've found them all\n if verbose:\n print('found all {} of {}, stopping...'.format(N_test, current_class))\n print('{}/{}'.format(count_of_test_class, abs(j)))\n break\n if (current_class == test_class):\n count_of_test_class = count_of_test_class + 1 # n A\n else:\n count_of_false_positives = count_of_false_positives + 1\n precs_x = count_of_test_class / (abs(j)) # N.b. this is the sum, we divide by j on the output\n recall_x = count_of_test_class / N_test\n false_pos_rate = count_of_false_positives / no_of_images\n specificity_x = 1-false_pos_rate\n if (precs_x <= .95) and (found_recall_precs_p95 is False):\n # thingy to grab the recall when precision drops below .95 (if ever)\n recall_p95 = recall_x\n found_recall_precs_p95 = True\n delta_recall_x = recall_x - recall_x_minus_1 # difference in recall between this point nd the next\n weight_precs_x = precs_x * delta_recall_x # weighted precsion at point x (we do average via weighted sum)\n Ave_precs_x = Ave_precs_x + weight_precs_x # average_precision evaluated at point x\n informedness_x = recall_x + specificity_x -1\n if informedness_x > max_informedness:\n max_informedness = informedness_x\n x_for_max_informedness = abs(j)\n recall_for_max_informedness = recall_x\n specificity_for_max_informedness = specificity_x\n if (precs_x > 0 and recall_x > 0):\n f1_x = 2*(precs_x*recall_x) / (precs_x + recall_x)\n else:\n f1_x = 0\n if f1_x > max_f1_stat:\n max_f1_stat = f1_x\n x_for_max_f1 = abs(j)\n if verbose:\n print('Class {}, evaluated at Q={}:, found {}'.format(test_class, Q, count_of_test_class))\n print('Average precision={}; precision={}; recall={}'.format(Ave_precs_x,precs_x,recall_x))\n print('Recall at precision .95 threashold is {}'.format(recall_p95))\n print('specificity = {}'.format(specificity_x))\n print('informedness = {}'.format(informedness_x))\n print('Max informedness was {}, seen at x {}'.format(max_informedness, x_for_max_informedness))\n print('F1 at x={} is {}'.format(abs(j),f1_x))\n print('Max F1 was {}, seen at x = {}'.format(max_f1_stat,x_for_max_f1))\n out = (Ave_precs_x, precs_x, recall_x, recall_p95,\n specificity_x, informedness_x, max_informedness, x_for_max_informedness,\n max_f1_stat, x_for_max_f1,\n recall_for_max_informedness, specificity_for_max_informedness,)\n return out","def statistics_on_test(self, predicted_results, result):\n # Print confusion matrix and mean average precision score\n predicted_results_binary = self.predicted_results_to_binary(predicted_results)\n print(\"\\nConfusion matrix : \")\n print(confusion_matrix(result, predicted_results_binary))\n print(\"\\nAverage precision score : \", average_precision_score(result, predicted_results_binary))","def matching_accuracy(pmat_pred, pmat_gt, ns):\n device = pmat_pred.device\n batch_num = pmat_pred.shape[0]\n\n pmat_gt = pmat_gt.to(device)\n\n assert torch.all((pmat_pred == 0) + (pmat_pred == 1)), 'pmat_pred can noly contain 0/1 elements.'\n assert torch.all((pmat_gt == 0) + (pmat_gt == 1)), 'pmat_gt should noly contain 0/1 elements.'\n assert torch.all(torch.sum(pmat_gt, dim=-1) <= 1) and torch.all(torch.sum(pmat_gt, dim=-2) <= 1)\n assert torch.all(torch.sum(pmat_pred, dim=-1) <= 1) and torch.all(torch.sum(pmat_pred, dim=-2) <= 1)\n\n #indices_pred = torch.argmax(pmat_pred, dim=-1)\n #indices_gt = torch.argmax(pmat_gt, dim=-1)\n\n #matched = (indices_gt == indices_pred).type(pmat_pred.dtype)\n match_num_list = []\n gt_num_list = []\n pred_num_list = []\n acc_gt = []\n acc_pred = []\n for b in range(batch_num): #acc_gt,acc_pred\n #match_num += torch.sum(matched[b, :ns[b]])\n #total_num += ns[b].item()\n match_num = torch.sum(pmat_pred[b, :] * pmat_gt[b, :]) + 1e-8 #:ns[b]\n gt_num = torch.sum(pmat_gt[b, :]) + 1e-8\n pred_num = torch.sum(pmat_pred[b,:]) + 1e-8\n match_num_list.append(match_num.cpu().numpy())\n gt_num_list.append(gt_num.cpu().numpy())\n pred_num_list.append(pred_num.cpu().numpy())\n acc_gt.append((match_num/gt_num).cpu().numpy())\n acc_pred.append((match_num/pred_num).cpu().numpy())\n\n return {'acc_gt': np.array(acc_gt),\n 'acc_pred': np.array(acc_pred),\n 'match_num': np.array(match_num_list),\n 'gt_num': np.array(gt_num_list),\n 'pred_num': np.array(pred_num_list)}","def print_results(precision, recall, f1):\n print \"Precision:\", precision\n print \"Recall:\", recall\n print \"F1 score:\", f1","def scores(self, y, y_pred):\n\n aucroc = 0.\n precision = 0.\n recall = 0.\n f1 = 0.\n aucroc_labs = np.zeros(self.datas[self.train_idx].n_labels)\n precision_labs = np.zeros(self.datas[self.train_idx].n_labels)\n recall_labs = np.zeros(self.datas[self.train_idx].n_labels)\n f1_labs = np.zeros(self.datas[self.train_idx].n_labels)\n label_ratios = np.mean(y, axis=0)\n\n if len(y) > 1:\n y_t = np.transpose(y)\n col_keep = np.ones(len(y_t), dtype=bool)\n for i, col_y in enumerate(y_t):\n if 0 not in col_y or 1 not in col_y:\n col_keep[i] = False\n\n if sum(col_keep) > 0:\n if not col_keep.all():\n y = np.transpose(y_t[col_keep])\n y_pred = np.transpose(np.transpose(y_pred)[col_keep])\n\n f1 = f1_score(y, self._round(y_pred), average=self.metrics_avg)\n s = f1_score(y, self._round(y_pred), average=None)\n f1_labs[col_keep] = s if sum(col_keep) > 1 else s[1]\n aucroc = roc_auc_score(y, y_pred, average=self.metrics_avg)\n aucroc_labs[col_keep] = roc_auc_score(y, y_pred, average=None)\n precision = precision_score(y, self._round(y_pred), average=self.metrics_avg)\n recall = recall_score(y, self._round(y_pred), average=self.metrics_avg)\n if sum(col_keep) > 1:\n precision_labs[col_keep] = precision_score(y, self._round(y_pred), average=None)\n recall_labs[col_keep] = recall_score(y, self._round(y_pred), average=None)\n else:\n precision_labs[col_keep] = precision_score(y, self._round(y_pred))\n recall_labs[col_keep] = recall_score(y, self._round(y_pred))\n elif self.verbose:\n print('*Cannot compute other metrics because no label in Truth has alternatives, only precision*')\n precision = precision_score(y, self._round(y_pred), average=self.metrics_avg)\n precision_labs = precision_score(y, self._round(y_pred), average=None)\n\n elif len(y) == 1:\n if self.verbose:\n print('*Cannot compute other metrics with %d samples, only precision*' % len(y))\n precision = precision_score(y, self._round(y_pred), average=self.metrics_avg)\n precision_labs = precision_score(y, self._round(y_pred), average=None)\n\n result = {\n 'aucroc': aucroc,\n 'precision': precision,\n 'recall': recall,\n 'f1': f1,\n 'aucroc_labs': aucroc_labs,\n 'precision_labs': precision_labs,\n 'recall_labs': recall_labs,\n 'f1_labs': f1_labs,\n 'label_ratios': label_ratios\n }\n\n return result","def update_metrics(self, metrics, predictions, labels):\n return","def compute_precision(data, num_labels = 4): \n \n # Declarating list to store results\n precisions = []\n \n for instance in data:\n \n # Declarating list to store individual results\n instance_precisions = []\n \n for i in np.arange(num_labels):\n \n # Computing and storing precision for each class\n instance_precisions.append(precision_score(instance[:, 2 + i], instance[:, 2 + i + 4]))\n \n # Storing mean results of the instance\n precisions.append(np.mean(instance_precisions))\n \n # Returning mean of all results\n return np.mean(precisions)","def patternMatchingTest(matchingAlgorithm,\n observations= defaultCOMPREFdump , \n modelsFolder= defaultWRFdumpsFolder,\n #key1='0828',\n key1='',\n key2='',\n #key3='',\n verbose=True,\n makeSmallImages=True,\n smallDpi=40,\n frameDpi = 600 , # added 2013-09-23\n loadFormerAnalysisResult=False, #added 2013-09-23\n #summary_folder=summary_folder, # needed if loadFormerAnalysisResult is True\n summary_folder=summary_folder+timeString + '/', # 2013-09-27\n matching_algorithm_name=\"\", # needed if loadFormerAnalysisResult is True\n panel_cols = 5,\n panel_rows = 5,\n useCV2 = False, # switch off opencv since it doesn't work yet\n useCV = False,\n averageScoreOrdering=True, # take an average of the ordering to get a total score for the entire model over the time series\n #kwargs = {}, # keyword arguments for the matchinging algorithm, just in case\n **kwargs\n ):\n if key1 != '' or key2 != '':\n summary_folder += '_'.join([key1, key2])\n if not averageScoreOrdering:\n summary_folder += '_nonaver'\n summary_folder += '%dx%d/' % (panel_rows, panel_cols)\n\n t0 = time.time()\n if loadFormerAnalysisResult:\n mv_folder = summary_folder + matching_algorithm_name + '/'\n result = pickle.load(open(mv_folder + 'result.pydump', 'r'))\n\n \"\"\"\n result= {'best matches' : bestMatchAtEachTime, \n 'best overall match' : bestMatchOverall, \n 'all scores' : scores,\n 'matching algorithm' : matching_algorithm_name,\n 'dataset1[obs]' : observations,\n 'dataset2[models]' : modelsFolder,\n 'time spent' : time.time() - t0,\n 'mv folder' : mv_folder, #added 2013-09-23\n 'ordering' : ordering, #added 2013-09-23\n }\n \"\"\"\n bestMatchAtEachTime = result['best matches'] \n bestMatchOverall = result['best overall match']\n scores = result['all scores']\n matching_algorithm_name = result['matching algorithm']\n observations = result['dataset1[obs]']\n modelsFolder = result['dataset2[models]']\n #mv_folder = result['mv folder']\n ordering = result['ordering']\n\n # THE FOLLOWING 30 LINES ARE COPIED FROM \"ELSE...\" BELOW\n ########### ******\n ##\n #\n ds1 = pickle.load(open(observations,'r'))\n ds1.setFloor(0) # added 2013-09-27\n ds1.list = [v for v in ds1.list if (key1 in v.dataTime or key1 in v.name) \\\n and (key2 in v.dataTime or key2 in v.name)]\n ds1.list.sort(key=lambda v:v.dataTime)\n ds1.setVmin(-40) #line added 2013-09-23\n # debug\n #print ds1.list\n print 'loading list with keywords [ %s %s ]' %(key1, key2) \n ds1.load(verbose=verbose)\n # end debug\n modelsList = [v for v in os.listdir(modelsFolder) if '.pydump' in v]\n modelsList.sort(key = lambda v: int(re.findall(r'\\d+',v)[0]))\n print \"modelsList\", modelsList\n scores = {} # scores = {model number: [('dataTime', score), ('dataTime', score),...], ...}\n print 'loading ds2 (model data)'\n wrfs = [] # line added 2013-09-23\n for path in modelsList:\n modelIndex = int(re.findall(r'\\d+', path)[0])\n ds2 = pickle.load(open(modelsFolder+path, 'r'))\n ds2.setFloor(0) # added 2013-09-27\n # debug\n #ds2.list = [v for v in ds2.list if key1 in v.dataTime] # line redundant, 2013-09-23\n ds1.list, ds2.list = ds1.intersect(ds2) #line added 2013-09-23 \n ds2.setVmin(-40) #line added 2013-09-23\n wrfs.append(ds2) # line added 2013-09-23\n # end debug\n\n ####################\n\n\n else:\n # THE FOLLOWING 30 LINES ARE COPIED TO \"IF...\" ABOVE\n ########### ******\n ##\n #\n ds1 = pickle.load(open(observations,'r'))\n ds1.list = [v for v in ds1.list if (key1 in v.dataTime or key1 in v.name) \\\n and (key2 in v.dataTime or key2 in v.name)]\n ds1.list.sort(key=lambda v:v.dataTime)\n ds1.setVmin(-40) #line added 2013-09-23\n # debug\n #print ds1.list\n print 'loading list with keywords [ %s %s ]' %(key1, key2) \n ds1.load(verbose=verbose)\n ds1.setFloor(0) # added 2013-09-27\n # end debug\n modelsList = [v for v in os.listdir(modelsFolder) if '.pydump' in v]\n modelsList.sort(key = lambda v: int(re.findall(r'\\d+',v)[0]))\n print \"modelsList\", modelsList\n scores = {} # scores = {model number: [('dataTime', score), ('dataTime', score),...], ...}\n print 'loading ds2 (model data)'\n wrfs = [] # line added 2013-09-23\n ds1.commonMaskSet=False # tempory attribute added 2013-09-27\n for path in modelsList:\n modelIndex = int(re.findall(r'\\d+', path)[0])\n ds2 = pickle.load(open(modelsFolder+path, 'r'))\n # debug\n #ds2.list = [v for v in ds2.list if key1 in v.dataTime] # line redundant, 2013-09-23\n ds1.list, ds2.list = ds1.intersect(ds2) #line added 2013-09-23 \n ds2.setFloor(0) # added 2013-09-27\n ds2[0].matrix.mask += ds1[0].matrix.mask # added 2013-09-27\n ds2.setCommonMask() # added 2013-09-27\n if not ds1.commonMaskSet: # added 2013-09-27\n ds1[0].matrix.mask +=ds2[0].matrix.mask # added 2013-09-27\n ds1.setCommonMask() # added 2013-09-27\n ds2.setVmin(-40) #line added 2013-09-23\n wrfs.append(ds2) # line added 2013-09-23\n # end debug\n\n #######################################################################################################################\n # key lines\n print 'computing scores between %s and %s ...' %(ds1.name, ds2.name)\n s, matching_algorithm_name = matchingAlgorithm(ds1, ds2, **kwargs) # list of the form [('20120612.0000', 0.9),...] \n scores[modelIndex] = dict(s) # { model index: {'dataTime': score,...}, ...}\n # debug\n if verbose:\n print scores\n print '..................................................'\n #\n ########################################################################################################################\n\n print 'sorting the results...'\n #print 'scores', '\\n', scores #debug\n # bestMatchAtEachTime\n T1 = [v.dataTime for v in ds1]\n T2 = [v.dataTime for v in ds2] #any ds2 will do - they are the same, in this case we pick the last one \n T = set(T1).intersection(T2)\n dataTimeList = sorted(list(T))\n\n #print 'dataTimeList:', dataTimeList #debug\n highestScores = []\n ordering = [] #added 2013-09-23\n\n if averageScoreOrdering: # take the average scores before ordering to make it uniform\n ave_scores = [ (v, np.mean( [scores[v].values() ] ) ) for v in scores.keys()]\n ave_scores = dict(ave_scores)\n print '..taking averages...............................'\n print ave_scores\n for u in scores.keys():\n for t in scores[u].keys():\n scores[u][t] = ave_scores[u]\n print '\\n'.join([str(scores[v]) for v in scores.keys()])\n\n for t in dataTimeList:\n print t\n highestScoreAt_t = max([scores[u][t] for u in scores.keys()]) # u = model index (integer 1-20), t = dataTime (string)\n new_entry = (t, highestScoreAt_t, [u for u in scores.keys() if scores[u][t]==highestScoreAt_t])\n print new_entry\n highestScores.append(new_entry)\n #added 2013-09-23\n ordering.append([0]+sorted(range(1, len(modelsList)+1), key=lambda v: scores[v][t], reverse=True))\n\n\n\n\n # bestMatchOverall\n print scores # debug\n averageScores = [(Model, np.mean(scores[Model].values())) for Model in scores.keys()] # list of [(Model, mean score),...]\n highestScore = max([v[1] for v in averageScores])\n bestMatch = [u[0] for u in averageScores if u[1]==highestScore] \n # debug\n print 'average scores , highest average, best match', averageScores, highestScore, bestMatch\n #xx= raw_input('press enter') \n\n\n bestMatchAtEachTime = highestScores\n bestMatchOverall = (highestScore, bestMatch)\n\n # the following lines do not work well since oftentimes we have multiple pointers (names) to a single object/function\n # and so they are replaced by returning the function name in the function itself\n #allvars = vars()\n #matching_algorithm_name = [v for v in allvars if allvars[v]==matchingAlgorithm][0]\n #mv_folder = summary_folder + matching_algorithm_name + '/'\n mv_folder = summary_folder + matching_algorithm_name + '/' # added 2013-09-27\n\n result= {'best matches' : bestMatchAtEachTime, \n 'best overall match' : bestMatchOverall, \n 'all scores' : scores,\n 'matching algorithm' : matching_algorithm_name,\n 'dataset1[obs]' : observations,\n 'dataset2[models]' : modelsFolder,\n 'time spent' : time.time() - t0,\n 'mv folder' : mv_folder, #added 2013-09-23\n 'ordering' : ordering, #added 2013-09-23\n }\n\n try:\n os.makedirs(mv_folder)\n print 'folder', mv_folder, \"created!\"\n except:\n print 'folder', mv_folder, 'exists!'\n pickle.dump(result, open(mv_folder+'result.pydump','w'))\n\n ############################################################################\n # to make the videos/collected images\n # calling armor.dataStreamTools.makeVideo.makeVideo()\n # 2013-09-23\n\n from . import makeVideo as mv\n \n try:\n os.makedirs(mv_folder)\n print 'folder', mv_folder, \"created!\"\n except:\n print 'folder', mv_folder, 'exists!'\n # make images\n if makeSmallImages:\n print 'making small images'\n for ds in [ds1] + wrfs:\n #ds.imageFolder= mv_folder + ds.name + '/'\n ds.imageFolder= summary_root + ds.name + '/' #added 2013-09-27 to replace the above line - too tired waiting the plots\n ds.saveImages(drawCoast=True, flipud=True, dpi=smallDpi)\n print ds.imageFolder\n else:\n # just set the paths\n for ds in [ds1]+wrfs:\n ds.imageFolder = summary_root + ds.name +'/' # /home/k/ARMOR/data/KONG-REY/summary/WRF01.DBZ/\n for dbzPattern in ds.list:\n dbzPattern.imagePath = ds.imageFolder + ds.name + dbzPattern.dataTime + '.png'\n print dbzPattern.imagePath\n #ds.setImagePaths() # doesn't for for old pickled objects???\n\n\n # make panels\n\n mv.makeVideo(DSS = [ds1]+ wrfs, # [ds0, ds1, ds2, ds3, ds4, ...], a list of armor.pattern.DBZstream objects\n panel_cols = panel_cols, # number of colums in the panel\n panel_rows = panel_rows, # no need to be filled\n #fourcc = cv.CV_FOURCC('F', 'L', 'V', '1'),\n #fps = defaultFps,\n extension= '.avi',\n #fourcc = cv.CV_FOURCC('P', 'I', 'M', '1'),\n outputFileName =\"\",\n outputFolder= mv_folder,\n sandbox = mv_folder ,\n saveFrames = True, # saving the frames as images\n frameDpi = frameDpi,\n useCV2 = useCV2,\n useCV = useCV,\n ordering = ordering, # ordering of the models\n )\n ############################################################################\n # dumping the results\n try:\n os.makedirs(summary_folder+ 'pydumps/')\n except:\n print 'Folder exists! -- ', summary_folder+ 'pydumps/' \n\n pickle.dump(result, open(summary_folder+ 'pydumps/results_' + matching_algorithm_name+ '+' +\\\n key1 + '+' + key2 + str(int(time.time())) + '.pydump', 'w'))\n\n \n\n print \"time spent:\", time.time()-t0\n return result","def evaluate(test_set, predictions):\n full_report = metrics.classification_report(test_set.labels, predictions,\n labels=range(len(test_set.index2label)),\n target_names=test_set.index2label, digits=3)\n pre, rec, f1, support = metrics.precision_recall_fscore_support(test_set.labels, predictions, average='weighted')\n return pre, rec, f1, support, full_report","def calculate_metrics(predictions, expected):\n # type: (np.ndarray, np.ndarray) -> (float, float, float)\n clients_count = predictions.shape[0]\n products_count = predictions.shape[1]\n\n true_positive = 0.0\n true_negative = 0.0\n false_positive = 0.0\n false_negative = 0.0\n\n total = float(clients_count * products_count)\n\n for c in range(0, clients_count):\n for p in range(0, products_count):\n if predictions[c, p] == expected[c, p]:\n if predictions[c, p] == 1:\n true_positive += 1\n else:\n true_negative += 1\n else:\n if predictions[c, p] == 1:\n false_positive += 1\n else:\n false_negative += 1\n\n accuracy = float(true_positive + true_negative) / total\n if true_positive + false_positive == 0:\n precision = 0\n else:\n precision = true_positive / float(true_positive + false_positive)\n\n if true_positive + false_negative == 0:\n recall = 0\n else:\n recall = true_positive / float(true_positive + false_negative)\n\n return accuracy, precision, recall","def evaluation(eval_dict, faster_rcnn, test_num=Config.eval_num):\n bboxes, labels, scores = list(), list(), list()\n gt_bboxes, gt_labels, gt_difficult = list(), list(), list()\n for i, [img_dir, img_info] in tqdm(enumerate(eval_dict.items())):\n img, img_info = rescale_image(img_dir, img_info, flip=False)\n img_tensor = create_img_tensor(img)\n box, score, label = faster_rcnn.predict(img_tensor)\n gt_bbox = np.array(img_info['objects'])[:, 1:5].astype(np.float32)\n gt_label = np.array(img_info['objects'])[:, 0]\n difficult = np.array(img_info['difficult'])\n\n # label from text to number\n gt_label = text_to_num(gt_label)\n\n bboxes.append(box)\n labels.append(label)\n scores.append(score)\n gt_bboxes.append(gt_bbox)\n gt_labels.append(gt_label)\n gt_difficult.append(difficult)\n if i == test_num:\n break\n\n AP, mAP = calc_map(bboxes, labels, scores, gt_bboxes, gt_labels, gt_difficult, use_07_metric=True)\n\n return mAP","def __call__(self, results):\n for key in results.get('img_fields', ['img']):\n results[key] = general_ocr.imnormalize(results[key], self.mean, self.std,\n self.to_rgb)\n results['img_norm_cfg'] = dict(\n mean=self.mean, std=self.std, to_rgb=self.to_rgb)\n return results","def precision_recall(y_true,\n probas_pred,\n num_positives):\n probas_pred = probas_pred.flatten()\n y_true = y_true.flatten()\n # to handle duplicates scores, we sort (score, NOT(jugement)) for predictions\n # eg,the final order will be (0.5, False), (0.5, False), (0.5, True),\n # (0.4, False), ...\n # This allows to have the worst possible AP.\n # It prevents participants from putting the same score for all predictions\n # to get a good AP.\n order = argsort(list(zip(probas_pred, ~y_true)))\n order = order[::-1] # sort by decreasing score\n probas_pred = probas_pred[order]\n y_true = y_true[order]\n\n ntp = np.cumsum(y_true) # number of true positives <= threshold\n nres = np.arange(len(y_true)) + 1 # number of results\n\n precisions = ntp / nres\n recalls = ntp / num_positives\n return precisions, recalls, probas_pred","def scoring(self):\n pass","def precision(ground_truth, prediction):\n ground_truth = remove_duplicates(ground_truth)\n prediction = remove_duplicates(prediction)\n precision_score = count_a_in_b_unique(prediction, ground_truth) / float(len(prediction))\n assert 0 <= precision_score <= 1\n return precision_score","def update_predictions(data):\n # TODO: Priority 1 - update predictions with inference results\n # TODO: Understand from a research team exactly what the data is going to look like\n trackID = data[0]\n prediction = data[1]\n confidence = data[2]\n to_Insert_Array = [trackID, prediction, confidence]\n OutPut_Data[trackID] = to_Insert_Array","def __train__(self):\n if (self.type_camf == 'CAMF_CI'):\n #users, items, context, ratings\n ci = camf_ci.CI_class(self.__users_array__, self.__items_array__, self.__context_array__, self.__ratings__, self.fold, self.lr, self.factors)\n predictions, losses = ci.fit()\n elif (self.type_camf == 'CAMF_CU'):\n cu = camf_cu.CU_class(self.__users_array__, self.__items_array__, self.__context_array__, self.__ratings__, self.fold, self.lr, self.factors)\n predictions, losses = cu.fit()\n elif (self.type_camf == 'CAMF_C'):\n c = camf_c.C_class(self.__users_array__, self.__items_array__, self.__context_array__, self.__ratings__, self.fold, self.lr, self.factors)\n predictions, losses = c.fit()\n\n dummy_pred = np.zeros((predictions.shape))\n for r, pred_array in enumerate(predictions):\n for c, pred in enumerate(pred_array):\n dummy_pred[r][c] = self.__check_ratings__(pred)\n predictions = dummy_pred\n #save a plot with a loss function\n plots = prs.PlotRSData()\n #print(losses)\n plots.plot_loss_cars(losses, self.type_camf, self.__save_prefix__+\"_loop\"+str(self.loop))\n pd.DataFrame(losses).to_csv(\"./RecSys/out/CAMF/train/\"+self.type_camf+\"/\" + self.__save_prefix__ +\"losses_loop\"+str(self.loop)+\".csv\")\n print('Saving the feature matrix...')\n # set predictions back to the pivot table\n self.__utility_saved_training__(predictions) \n # save results\n self.utility_predictions.to_csv(\"./RecSys/out/CAMF/train/\"+self.type_camf+\"/\" + self.__save_prefix__ + \"_SGD_predictions_loop\"+str(self.loop)+\".csv\")","def __call__(self, results):\n\n for key in results.get('seg_fields', []):\n if self.scale_factor != 1:\n results[key] = general_ocr.imrescale(\n results[key],\n self.scale_factor,\n interpolation='nearest',\n backend=self.backend)\n return results","def update_recall_rate(params, name, recall, k):\n if name in params:\n for key in params[name]:\n params[name][key] += recall[key] / k\n else:\n params[name] = {}\n for key in recall:\n params[name][key] = recall[key] / k","def metric_test(self):\n k = 10\n latent_factor = 10\n n_users = 10\n n_items = 12\n\n interactions, user_features, item_features = util.generate_dummy_data_with_indicator (num_users=n_users, num_items=n_items, interaction_density=.5)\n print (\"interactiosn shape={}\".format( np.shape(interactions) ))\n print (\"user features shape={}\".format( np.shape(user_features.toarray()) ))\n print (\"item features shape={}\".format( np.shape(item_features.toarray()) ))\n\n model = TensorRec(n_components=latent_factor)\n\n model.fit(interactions, user_features, item_features, epochs=19)\n\n ranks = model.predict_rank(user_features=user_features, item_features=item_features)\n\n print (\"Ranks shape={}\".format(np.shape(ranks)))\n\n self.assertTrue(np.shape(interactions) == np.shape(ranks))\n\n tr_recall_result = eval.recall_at_k(predicted_ranks=ranks, test_interactions=interactions, k=k, preserve_rows=False)\n # print (tr_recall_result.mean())\n\n tr_precision_result = eval.precision_at_k(predicted_ranks=ranks, test_interactions=interactions, k=k, preserve_rows=False)\n # print(tr_precision_result.mean())\n\n # we need csr for interactions data\n interactions_ = interactions.tocsr()\n recall_result = metrics.recall_at_k(ranks, interactions_, k=k, preserve_rows=False)\n # print(recall_result.mean())\n\n precision_result = metrics.precision_at_k(ranks, interactions_, k=k, preserve_rows=False)\n # print (precision_result.mean())\n\n self.assertTrue (tr_recall_result.mean() == recall_result.mean())\n self.assertTrue (tr_precision_result.mean() == precision_result.mean())","def __init__(self):\n\n try:\n self.posFreqDict = self.load(\"posFreqDict\")\n self.neuFreqDict = self.load(\"neuFreqDict\")\n self.negFreqDict = self.load(\"negFreqDict\")\n\n except:\n\n lFileList = []\n for fFileObj in os.walk(\"movies_reviews/\"): \n lFileList = fFileObj[2]\n break\n random.shuffle(lFileList)\n \n self.train(lFileList)\n Original_d = {}\n d = {}\n correct_positives =0\n incorrect_positives=0\n correct_negatives =0\n incorrect_negatives = 0\n for file in lFileList[int(float(len(lFileList))*.9):len(lFileList)+1]:\n rating = int(file[7])\n if rating in Original_d:\n Original_d[rating] = Original_d[rating]+1\n else:\n Original_d[rating]=1\n\n fileText = self.loadFile(\"movies_reviews/\" + str(file))\n result = self.classify(fileText)\n if result in d:\n d[result]=d[result]+1\n else:\n d[result]=1\n if rating==5:\n if result ==\"positive\":\n correct_positives+=1\n else:\n incorrect_positives+=1\n else:\n if result == \"negative\":\n correct_negatives+=1\n else:\n incorrect_negatives+=1\n print Original_d\n print d\n print correct_positives #Actually positive, we said positive\n print incorrect_positives #Actually positive, we said negative\n print correct_negatives #Actually negative, we said negative\n print incorrect_negatives #Actually negative, we said positive\n \n A = float(correct_positives)\n B = float(incorrect_positives)\n C = float(incorrect_negatives)\n\n Recall = (A / (A + B)) *100.0\n Precision = (A / (A + C)) *100.0\n FScore = 2 * (Recall*Precision) / (Recall+Precision)\n\n print Recall\n print Precision\n print FScore","def _adjust_brightness_img(self, results, factor=1.0):\n for key in results.get('img_fields', ['image']):\n img = results[key]\n results[key] = mmcv.adjust_brightness(img,\n factor).astype(img.dtype)","def f1_score(model_id, test_set_id, rubric_id):\n result = {'true_positive': 0, 'false_positive': 0, 'true_negative': 0, 'false_negative': 0}\n # right answers\n answers = db.get_rubric_answers(test_set_id, rubric_id)\n # rubrication results\n rubrication_result = db.get_rubrication_result(model_id, test_set_id, rubric_id)\n\n for key in rubrication_result:\n if rubrication_result[key] == answers[key]:\n if rubrication_result[key] == 1:\n result['true_positive'] += 1\n else:\n result['true_negative'] += 1\n else:\n if rubrication_result[key] == 1:\n result['false_positive'] += 1\n else:\n result['false_negative'] += 1\n if (result['true_positive'] + result['false_positive']) > 0:\n result['precision'] = result['true_positive'] / (result['true_positive'] + result['false_positive'])\n else:\n result['precision'] = 0\n if (result['true_positive'] + result['false_negative']) > 0:\n result['recall'] = result['true_positive'] / (result['true_positive'] + result['false_negative'])\n else:\n result['recall'] = 0\n if (result['precision'] + result['recall']) > 0:\n result['f1'] = 2 * result['precision'] * result['recall'] / (result['precision'] + result['recall'])\n else:\n result['f1'] = 0\n return result","def metrics_evaluation(y_true, y_prob, threshold, df_type='train'):\n\n y_pred = (y_prob>=threshold).astype(int)\n \n tn = metrics.confusion_matrix(y_true, y_pred)[0][0]\n fp = metrics.confusion_matrix(y_true, y_pred)[0][1]\n fn = metrics.confusion_matrix(y_true, y_pred)[1][0]\n tp = metrics.confusion_matrix(y_true, y_pred)[1][1]\n\n accuracy_scr = metrics.accuracy_score(y_true, y_pred)\n precision_scr = metrics.precision_score(y_true, y_pred)\n recall_scr = metrics.recall_score(y_true, y_pred)\n f1_scr = metrics.f1_score(y_true, y_pred)\n roc_auc_scr = metrics.roc_auc_score(y_true, y_pred)\n\n result = {'Dataset': df_type, 'No obs': len(y_true), 'Threshold': threshold,\n 'TP':tp, 'FP': fp, 'TN': tn, 'FN':fn , \n 'Accuracy Score':accuracy_scr, 'Precision Score':precision_scr, \n 'Recall Score':recall_scr, 'F1 Score':f1_scr, 'ROC AUC Score':roc_auc_scr}\n\n return result","def get_precision(y_true, y_pred):\n model_entities_filter = (y_pred != 3).astype(\"int\") # of the words our model say has a NER class\n precision_correct_entities = (y_pred[np.where(model_entities_filter)] == y_true[np.where(model_entities_filter)]).astype(\"int\")\n precision = np.sum(precision_correct_entities)/np.sum(model_entities_filter) \n return precision","def test_evaluate(self):\n # test normalized by 'bbox_size'\n jhmdb_pck_metric = JhmdbPCKAccuracy(thr=0.5, norm_item='bbox')\n jhmdb_pck_metric.process(self.data_batch, self.data_samples)\n pck_results = jhmdb_pck_metric.evaluate(self.batch_size)\n target = {\n 'Head PCK': 1.0,\n 'Sho PCK': 1.0,\n 'Elb PCK': 1.0,\n 'Wri PCK': 1.0,\n 'Hip PCK': 1.0,\n 'Knee PCK': 1.0,\n 'Ank PCK': 1.0,\n 'PCK': 1.0,\n }\n self.assertDictEqual(pck_results, target)\n\n # test normalized by 'torso_size'\n jhmdb_tpck_metric = JhmdbPCKAccuracy(thr=0.2, norm_item='torso')\n jhmdb_tpck_metric.process(self.data_batch, self.data_samples)\n tpck_results = jhmdb_tpck_metric.evaluate(self.batch_size)\n target = {\n 'Head tPCK': 1.0,\n 'Sho tPCK': 1.0,\n 'Elb tPCK': 1.0,\n 'Wri tPCK': 1.0,\n 'Hip tPCK': 1.0,\n 'Knee tPCK': 1.0,\n 'Ank tPCK': 1.0,\n 'tPCK': 1.0,\n }\n self.assertDictEqual(tpck_results, target)","def precision(y_true, y_pred):\n true_positives = bk.sum(bk.round(bk.clip(y_true * y_pred, 0, 1)))\n predicted_positives = bk.sum(bk.round(bk.clip(y_pred, 0, 1)))\n precision = true_positives / (predicted_positives + bk.epsilon())\n return precision","def mgcEval(self):\n import numpy as np\n def report_to_df(report):\n\n \"\"\"\n function to convert classification report to dataframe (for visualisation plot)\n \"\"\"\n\n report = re.sub(r\" +\", \" \", report).replace(\"avg / total\", \"avg/total\").replace(\"\\n \", \"\\n\")\n # update this due to sklearn classification report output change\n report = re.sub(r\" +\", \" \", report).replace(\"micro avg\", \"micro_avg\").replace(\"macro avg\", \"macro_avg\").replace(\"weighted avg\", \"weighted_avg\").replace(\"\\n \", \"\\n\")\n report_df = pd.read_csv(StringIO(\"Classes\" + report), sep=' ', index_col=0) \n return(report_df)\n \n #txt report to df\n class_rpttop1 = classification_report(self.y_true, self.y_pred)\n df_report = report_to_df(class_rpttop1)\n\n df_report = df_report.iloc[:self.nb_classes, :].copy()\n df_report.index = df_report.index.astype(int)\n \n\n # classifier prediction metrics\n def classMetrics(averagex):\n precision, recall, fscore, support = score(self.y_true, self.y_pred, average=averagex)\n \n return(\n print(''), \n print('-------------{0:}--------------------'.format(averagex)), \n print('precision: {0:.4f}'.format(precision)),\n print('recall: {0:.4f}'.format(recall)),\n print('fscore: {0:.4f}'.format(fscore)),\n print(''),\n print('kappa score: {0:.4f}'.format(cohen_kappa_score(self.y_true, self.y_pred))),\n print('accuracy score: {0:.4f}'.format(accuracy_score(self.y_true, self.y_pred))))\n \n def predSamp():\n\n correct = np.nonzero(self.y_pred==self.y_true)[0]\n incorrect = np.nonzero(self.y_pred!=self.y_true)[0]\n\n # quick check of the number of correct prediction from validation set\n print(\"\")\n print(\"correct/total = {0: .4f}\".format(len(correct)/(len(correct)+len(incorrect))))\n print(\"total correct sample = {0: .0f}\".format(len(correct)))\n print('------------------------------------------------------------------')\n \n def classReport():\n print('----------------------------- Classfication Report -------------------------------')\n print(classification_report(pd.Series(self.y_true).map(self.dict_label), pd.Series(self.y_pred).map(self.dict_label)))\n \n self.class_rpt = pd.concat([pd.DataFrame(pd.Series(df_report.index.tolist()).map(self.dict_label), columns = ['label']), df_report], axis = 1)\n \n self.classMetricsMac = classMetrics(\"macro\")\n self.classMetricsMic = classMetrics(\"micro\")\n self.predSample = predSamp()\n self.class_rptTop1 = classReport()\n \n return self","def precision_recall_f1(prediction, ground_truth):\n if not isinstance(prediction, list):\n prediction_tokens = prediction.split()\n else:\n prediction_tokens = prediction\n if not isinstance(ground_truth, list):\n ground_truth_tokens = ground_truth.split()\n else:\n ground_truth_tokens = ground_truth\n common = Counter(prediction_tokens) & Counter(ground_truth_tokens)\n num_same = sum(common.values())\n if num_same == 0:\n return 0, 0, 0\n p = 1.0 * num_same / len(prediction_tokens)\n r = 1.0 * num_same / len(ground_truth_tokens)\n f1 = (2 * p * r) / (p + r)\n return p, r, f1","def compute(self) -> Tensor:\n\n if self.samples:\n return self.average_precisions.float() / self.total\n else:\n # pred_image_indices = torch.cat(self.pred_image_indices, dim=0)\n pred_probs = torch.cat(self.pred_probs, dim=0)\n pred_labels = torch.cat(self.pred_labels, dim=0)\n pred_bboxes = torch.cat(self.pred_bboxes, dim=0)\n\n # target_image_indices = torch.cat(self.target_image_indices, dim=0)\n target_labels = torch.cat(self.target_labels, dim=0)\n target_bboxes = torch.cat(self.target_bboxes, dim=0)\n\n # pred_index = torch.nonzero((pred_labels == 1))\n # pred_probs = pred_probs[pred_index]\n # pred_bboxes = pred_bboxes[pred_index]\n # target_index = torch.nonzero((target_labels == 1))\n # target_bboxes = target_bboxes[target_index]\n\n\n # _, index_sorted = torch.sort(pred_probs)\n # pred_bboxes = pred_bboxes[index_sorted].cpu().detach().numpy()\n # target_bboxes = target_bboxes.cpu().detach().numpy()\n pred_probs = pred_probs.cpu().detach().numpy()\n pred_labels = pred_labels.cpu().detach().numpy()\n pred_bboxes = pred_bboxes.cpu().detach().numpy()\n target_labels = target_labels.cpu().detach().numpy()\n target_bboxes = target_bboxes.cpu().detach().numpy()\n\n pred_probs = pred_probs[pred_labels == 1]\n pred_bboxes = pred_bboxes[pred_labels == 1]\n target_bboxes = target_bboxes[target_labels == 1]\n\n preds_sorted_idx = np.argsort(pred_probs)[::-1]\n pred_bboxes = pred_bboxes[preds_sorted_idx]\n\n x, y = calculate_precision_recall(target_bboxes, pred_bboxes)\n\n if len(x) >= 2:\n return auc(x, y)\n else:\n return 0\n\n # return mean_average_precision(\n # pred_image_indices,\n # pred_probs,\n # pred_labels,\n # pred_bboxes,\n # target_image_indices,\n # target_labels,\n # target_bboxes,\n # self.iou_threshold,\n # self.ap_calculation,\n # )","def _adjust_color_img(self, results, factor=1.0):\n for key in results.get('img_fields', ['image']):\n # NOTE defaultly the image should be BGR format\n img = results[key]\n results[key] = mmcv.adjust_color(img, factor).astype(img.dtype)","def _compute_prediction(self, k, img, kp, des):\n ###taken from example code:\n\n #find matches\n matches = self.matcher.knnMatch(des,self.descs[k],k=2)\n\n # make sure the distance to the closest match is sufficiently better than the second closest\n good_matches = []\n for m,n in matches: \n if (m.distance < 0.7*n.distance and\n kp[m.queryIdx].response > 0.0 and\n self.kps[k][m.trainIdx].response > 0.0):\n good_matches.append((m.queryIdx, m.trainIdx))\n \n if len(good_matches) < 4:\n visual_diff = None\n else:\n #create pt lists\n img_pts = np.zeros((len(good_matches),2))\n tem_pts = np.zeros((len(good_matches),2))\n for idx in range(len(good_matches)):\n match = good_matches[idx]\n img_pts[idx,:] = kp[match[0]].pt\n tem_pts[idx,:] = self.kps[k][match[1]].pt\n \n # Transform input image so that it matches the template image as well as possible\n M, mask = cv2.findHomography(img_pts, tem_pts, cv2.RANSAC, 8.0)\n img_T = cv2.warpPerspective(img, M, self.signs[k].shape[::-1])\n \t\n #calculate error\n visual_diff = compare_images(img_T, self.signs[k])\n return visual_diff\n #not functioning properly (currently does not return highest value for identical image/scene pair)","def average_precision_recall(self,frame):\n precs,recls = np.array([]),np.array([])\n\n for num in np.arange(0,10,1): # for each class\n idx = 'Class_'+str(num)+'_precision' # index column name\n precs = np.append(precs,frame[idx].mean()) # add col avg to arr\n for num in np.arange(0,10,1): # for each class\n idx = 'Class_'+str(num)+'_recal' # index column name (misspelled 'recall')\n recls = np.append(recls,frame[idx].mean()) # add col avg to arr\n\n setattr(self,'precision_avg',np.mean(precs)) # average precision across 100 samples x 10 classes\n setattr(self,'recall_avg',np.mean(recls)) # average recall across 100 samples x 10 classes","def update(self, predictions, real_vals):\n for pred, actual in zip(predictions, real_vals):\n self.cm.loc[actual, pred] += 1","def PostProcessing(image, resultList, threshold=0.6):\n\tnum_detections = resultList[0][0].astype(np.int)\n\tscores = resultList[2]\n\tboxes = resultList[3]\n\tbbox_num = 0\n\t\n\t# loop through all the detections and get the confidence and bbox coordinates\n\tfor i in range(num_detections):\n\t\tdet_conf = scores[0, i]\n\t\tdet_ymin = boxes[0, i, 0]\n\t\tdet_xmin = boxes[0, i, 1]\n\t\tdet_ymax = boxes[0, i, 2]\n\t\tdet_xmax = boxes[0, i, 3]\n\n\t\tbbox_width = det_xmax - det_xmin\n\t\tbbox_height = det_ymax - det_ymin\n\t\t# the detection confidence and bbox dimensions must be greater than a minimum value to be a valid detection\n\t\tif threshold <= det_conf and 1 >= det_conf and bbox_width > 0 and bbox_height > 0:\n\t\t\tbbox_num += 1\n\t\t\txmin = int(round(det_xmin * image.shape[1]))\n\t\t\tymin = int(round(det_ymin * image.shape[0]))\n\t\t\txmax = int(round(det_xmax * image.shape[1]))\n\t\t\tymax = int(round(det_ymax * image.shape[0]))\n\t\t\t\n\t\t\tcv2.rectangle(image, (xmin, ymin), (xmax, ymax), (0, 255, 0), 2)\n\t\telse:\n\t\t\tcontinue\n\n\tprint(\"detected bbox num:\", bbox_num)\n\tSRC_PATH = os.path.realpath(__file__).rsplit(\"/\", 1)[0]\n\tOutput_PATH = os.path.join(SRC_PATH, \"../output/output.jpg\")\n\ttry:\n\t\tos.mkdir(os.path.join(SRC_PATH, \"../output/\"))\n\texcept Exception as e:\n\t\tprint(\"Output Path already exists\")\n\tcv2.imwrite(Output_PATH, image)","def compute_recall(data, num_labels = 4): \n \n # Declarating list to store results\n recalls = []\n \n for instance in data:\n \n # Declarating list to store individual results\n instance_recalls = []\n \n for i in np.arange(num_labels):\n \n # Computing and storing accuracy for each class\n instance_recalls.append(recall_score(instance[:, 2 + i], instance[:, 2 + i + 4]))\n \n # Storing mean results of the instance\n recalls.append(np.mean(instance_recalls))\n \n # Returning mean of all results\n return np.mean(recalls)","def compute_prf(self, counts_dict):\n precision = counts_dict[\"correct\"] / counts_dict[\"predicted\"] if counts_dict[\"predicted\"] else 0.0\n recall = counts_dict[\"correct\"] / counts_dict[\"gold\"] if counts_dict[\"gold\"] else 0.0\n fscore = 2 * precision * recall / (precision + recall) if precision + recall else 0.0\n\n return {\"precision\": precision, \"recall\": recall, \"fscore\": fscore}","def test_results_prefit() -> None:\n X_train_val, X_test, y_train_val, y_test = train_test_split(\n X, y, test_size=1/10, random_state=1\n )\n X_train, X_val, y_train, y_val = train_test_split(\n X_train_val, y_train_val, test_size=1/9, random_state=1\n )\n estimator = LinearRegression().fit(X_train, y_train)\n mapie = MapieRegressor(estimator=estimator, cv=\"prefit\")\n mapie.fit(X_val, y_val)\n _, y_pis = mapie.predict(X_test, alpha=0.05)\n width_mean = (y_pis[:, 1, 0] - y_pis[:, 0, 0]).mean()\n coverage = regression_coverage_score(\n y_test, y_pis[:, 0, 0], y_pis[:, 1, 0]\n )\n np.testing.assert_allclose(width_mean, WIDTHS[\"prefit\"], rtol=1e-2)\n np.testing.assert_allclose(coverage, COVERAGES[\"prefit\"], rtol=1e-2)","def score(self, test_data):\n\n\t\tpass","def precision(y_true, y_pred):\n true_positives = K.sum(K.round(K.clip(y_true * y_pred, 0, 1)))\n predicted_positives = K.sum(K.round(K.clip(y_pred, 0, 1)))\n precision = true_positives / (predicted_positives + K.epsilon())\n return precision"],"string":"[\n \"def evaluate_matrices(self, precision_matrix, recall_matrix):\\n\\t\\tground_truth_count = precision_matrix.shape[0]\\n\\t\\tdetection_count = precision_matrix.shape[1]\\n\\t\\tground_truth_sets_precision = defaultdict(set) # number of ground truth items that match a particular detection in the precision matrix\\n\\t\\tdetection_sets_precision = defaultdict(set) # number of detection items that match a particular ground truth in the precision matrix\\n\\t\\tground_truth_sets_recall = defaultdict(set) # number of ground truth items that match a particular detection in the recall matrix\\n\\t\\tdetection_sets_recall = defaultdict(set) # number of detection items that match a particular ground truth in the recall matrix\\n\\n\\t\\tfor gt_index in range(ground_truth_count):\\n\\t\\t\\tfor det_index in range(detection_count):\\n\\t\\t\\t\\tif precision_matrix[gt_index, det_index] >= self.precision_threshold:\\n\\t\\t\\t\\t\\tground_truth_sets_precision[det_index].add(gt_index)\\n\\t\\t\\t\\t\\tdetection_sets_precision[gt_index].add(det_index)\\n\\t\\t\\t\\tif recall_matrix[gt_index, det_index] >= self.recall_threshold:\\n\\t\\t\\t\\t\\tground_truth_sets_recall[det_index].add(gt_index)\\n\\t\\t\\t\\t\\tdetection_sets_recall[gt_index].add(det_index)\\n\\n\\t\\tmatch_ground_truth = 0. # sum of MatchG\\n\\t\\tmatch_detection = 0. # sum of MatchD\\n\\n\\t\\tone_to_one_precision = set()\\n\\t\\tfor gt_index in detection_sets_precision:\\n\\t\\t\\tmatching_detections_precision = detection_sets_precision[gt_index]\\n\\t\\t\\tif len(matching_detections_precision) == 1:\\n\\t\\t\\t\\t(detection_precision, ) = matching_detections_precision\\n\\t\\t\\t\\tif len(ground_truth_sets_precision[detection_precision]) == 1:\\n\\t\\t\\t\\t\\tone_to_one_precision.add((gt_index, detection_precision))\\n\\t\\t\\telse:\\n\\t\\t\\t\\t# one-to-many (one ground truth to many detections)\\n\\t\\t\\t\\tgt_sum = 0.\\n\\t\\t\\t\\tfor detection_precision in matching_detections_precision:\\n\\t\\t\\t\\t\\tgt_sum += recall_matrix[gt_index, detection_precision]\\n\\t\\t\\t\\tif gt_sum >= self.recall_threshold:\\n\\t\\t\\t\\t\\t#print(\\\"1:N ~ GT {} : DT {}\\\".format(gt_index,matching_detections_precision))\\n\\t\\t\\t\\t\\tmatch_ground_truth += self.scatter_punishment(matching_detections_precision)\\n\\t\\t\\t\\t\\tmatch_detection += len(matching_detections_precision) * self.scatter_punishment(matching_detections_precision)\\n\\n\\t\\tone_to_one_recall = set()\\n\\t\\tfor det_index in ground_truth_sets_recall:\\n\\t\\t\\tmatching_ground_truths_recall = ground_truth_sets_recall[det_index]\\n\\t\\t\\tif len(matching_ground_truths_recall) == 1:\\n\\t\\t\\t\\t(ground_truth_recall, ) = matching_ground_truths_recall\\n\\t\\t\\t\\tif len(detection_sets_recall[ground_truth_recall]) == 1:\\n\\t\\t\\t\\t\\tone_to_one_recall.add((ground_truth_recall, det_index))\\n\\t\\t\\telse:\\n\\t\\t\\t\\t# many-to-one (many ground truths covered by one detection)\\n\\t\\t\\t\\tdet_sum = 0\\n\\t\\t\\t\\tfor ground_truth_recall in matching_ground_truths_recall:\\n\\t\\t\\t\\t\\tdet_sum += precision_matrix[ground_truth_recall, det_index]\\n\\t\\t\\t\\tif det_sum >= self.precision_threshold:\\n\\t\\t\\t\\t\\t#print(\\\"N:1 ~ DT {} : GT {}\\\".format(det_index,matching_ground_truths_recall))\\n\\t\\t\\t\\t\\tmatch_detection += self.scatter_punishment(matching_ground_truths_recall)\\n\\t\\t\\t\\t\\tmatch_ground_truth += len(matching_ground_truths_recall) * self.scatter_punishment(matching_ground_truths_recall)\\n\\n\\t\\tone_to_one_matches = one_to_one_precision & one_to_one_recall\\n\\t\\tmatch_ground_truth += len(one_to_one_matches)\\n\\t\\tmatch_detection += len(one_to_one_matches)\\n\\n\\t\\trecall = match_ground_truth / float(ground_truth_count)\\n\\t\\tprecision = match_detection / float(detection_count)\\n\\t\\treturn (precision, recall, (match_detection, float(detection_count)), (match_ground_truth, float(ground_truth_count)))\",\n \"def _update_result(self, results, clf_numbers):\\n # ToDo make results of scoring values dynamic\\n names_results = ['Accuracy']\\n for number in clf_numbers:\\n for name in names_results:\\n if name not in self.results:\\n self.results[name] = [results[number][name + \\\"_test_score_\\\" + str(number)]]\\n else:\\n self.results[name].append(results[number][name + \\\"_test_score_\\\" + str(number)])\",\n \"def precision(self):\\n self.overall_precision = precision_score(\\n self.y_true, self.y_pred, average = self.average_type).round(self.digits_count_fp)\\n self.classes_precision = precision_score(\\n self.y_true, self.y_pred, average = None).round(self.digits_count_fp)\",\n \"def get_recall_precision(imgs_truth, imgs_pred, is_hit):\\n # Numbers of relevant and detect boxes, respectively.\\n num_rel = sum([len(img.boxes) for img in imgs_truth])\\n num_det = sum([len(img.boxes) for img in imgs_pred])\\n\\n boxes_for_fname_truth = {img.fname: img.boxes for img in imgs_truth}\\n boxes_for_fname_pred = {img.fname: img.boxes for img in imgs_pred}\\n\\n num_hit = 0\\n for fname, boxes in boxes_for_fname_truth.items():\\n if fname not in boxes_for_fname_pred:\\n continue\\n\\n num_hit += get_num_hit(boxes, boxes_for_fname_pred[fname], is_hit)\\n\\n return num_hit / num_rel, num_hit / num_det\",\n \"def recall(self):\\n self.overall_recall = recall_score(\\n self.y_true, self.y_pred, average = self.average_type).round(self.digits_count_fp)\\n self.classes_recall = recall_score(\\n self.y_true, self.y_pred, average = None).round(self.digits_count_fp)\",\n \"def calculates_results_stats(results_dic): \\n # Creates empty dictionary for results_stats_dic\\n results_stats_dic = dict()\\n \\n # Sets all counters to initial values of zero so that they can \\n # be incremented while processing through the images in results_dic \\n results_stats_dic['n_dogs_img'] = 0\\n results_stats_dic['n_match'] = 0\\n results_stats_dic['n_correct_dogs'] = 0\\n results_stats_dic['n_correct_notdogs'] = 0\\n results_stats_dic['n_correct_breed'] = 0\\n \\n # process through the results dictionary\\n for key in results_dic:\\n \\n # Labels Match Exactly\\n if results_dic[key][2] == 1:\\n results_stats_dic['n_match'] += 1\\n\\n # TODO: 5a. REPLACE pass with CODE that counts how many pet images of\\n # dogs had their breed correctly classified. This happens \\n # when the pet image label indicates the image is-a-dog AND \\n # the pet image label and the classifier label match. You \\n # will need to write a conditional statement that determines\\n # when the dog breed is correctly classified and then \\n # increments 'n_correct_breed' by 1. Recall 'n_correct_breed' \\n # is a key in the results_stats_dic dictionary with it's value \\n # representing the number of correctly classified dog breeds.\\n # \\n # Pet Image Label is a Dog AND Labels match- counts Correct Breed\\n if results_dic[key][3] == 1 and results_dic[key][2] == 1:\\n results_stats_dic['n_correct_breed'] += 1\\n \\n # Pet Image Label is a Dog - counts number of dog images\\n if results_dic[key][3] == 1:\\n results_stats_dic['n_dogs_img'] += 1\\n \\n # Classifier classifies image as Dog (& pet image is a dog)\\n # counts number of correct dog classifications\\n if results_dic[key][4] == 1:\\n results_stats_dic['n_correct_dogs'] += 1\\n\\n # TODO: 5b. REPLACE pass with CODE that counts how many pet images \\n # that are NOT dogs were correctly classified. This happens \\n # when the pet image label indicates the image is-NOT-a-dog \\n # AND the classifier label indicates the images is-NOT-a-dog.\\n # You will need to write a conditional statement that \\n # determines when the classifier label indicates the image \\n # is-NOT-a-dog and then increments 'n_correct_notdogs' by 1. \\n # Recall the 'else:' above 'pass' already indicates that the \\n # pet image label indicates the image is-NOT-a-dog and \\n # 'n_correct_notdogs' is a key in the results_stats_dic dictionary \\n # with it's value representing the number of correctly \\n # classified NOT-a-dog images.\\n # \\n # Pet Image Label is NOT a Dog\\n else:\\n # Classifier classifies image as NOT a Dog(& pet image isn't a dog)\\n # counts number of correct NOT dog clasifications.\\n if results_dic[key][3] == 0 and results_dic[key][4] == 0:\\n results_stats_dic['n_correct_notdogs'] += 1\\n\\n\\n # Calculates run statistics (counts & percentages) below that are calculated\\n # using the counters from above.\\n\\n # calculates number of total images\\n results_stats_dic['n_images'] = len(results_dic)\\n\\n # calculates number of not-a-dog images using - images & dog images counts\\n results_stats_dic['n_notdogs_img'] = (results_stats_dic['n_images'] - \\n results_stats_dic['n_dogs_img']) \\n\\n # TODO: 5c. REPLACE zero(0.0) with CODE that calculates the % of correctly\\n # matched images. Recall that this can be calculated by the\\n # number of correctly matched images ('n_match') divided by the \\n # number of images('n_images'). This result will need to be \\n # multiplied by 100.0 to provide the percentage.\\n # \\n # Calculates % correct for matches\\n results_stats_dic['pct_match'] = (results_stats_dic['n_match'] / results_stats_dic['n_images']) * 100\\n\\n # TODO: 5d. REPLACE zero(0.0) with CODE that calculates the % of correctly\\n # classified dog images. Recall that this can be calculated by \\n # the number of correctly classified dog images('n_correct_dogs')\\n # divided by the number of dog images('n_dogs_img'). This result \\n # will need to be multiplied by 100.0 to provide the percentage.\\n # \\n # Calculates % correct dogs\\n results_stats_dic['pct_correct_dogs'] = (results_stats_dic['n_correct_dogs'] / results_stats_dic['n_dogs_img']) * 100\\n\\n # TODO: 5e. REPLACE zero(0.0) with CODE that calculates the % of correctly\\n # classified breeds of dogs. Recall that this can be calculated \\n # by the number of correctly classified breeds of dog('n_correct_breed') \\n # divided by the number of dog images('n_dogs_img'). This result \\n # will need to be multiplied by 100.0 to provide the percentage.\\n # \\n # Calculates % correct breed of dog\\n results_stats_dic['pct_correct_breed'] = (results_stats_dic['n_correct_breed'] / results_stats_dic['n_dogs_img']) * 100\\n\\n # Calculates % correct not-a-dog images\\n # Uses conditional statement for when no 'not a dog' images were submitted \\n if results_stats_dic['n_notdogs_img'] > 0:\\n results_stats_dic['pct_correct_notdogs'] = (results_stats_dic['n_correct_notdogs'] /\\n results_stats_dic['n_notdogs_img']) * 100.0\\n else:\\n results_stats_dic['pct_correct_notdogs'] = 0.0\\n\\n \\n # TODO 5f. REPLACE None with the results_stats_dic dictionary that you \\n # created with this function \\n return results_stats_dic\",\n \"def compute_map(scores, num_instances):\\r\\n pr = []\\r\\n for i, score in enumerate(scores):\\r\\n score = score[np.argsort(-score[:, 1])] # sort by confidence score\\r\\n FP = 0\\r\\n TP = 0\\r\\n pr_ = []\\r\\n for prediction in score:\\r\\n if prediction[0]:\\r\\n TP += 1\\r\\n else:\\r\\n FP += 1\\r\\n pr_.append([TP/(TP+FP), (TP/num_instances)])\\r\\n pr.append(pr_)\\r\\n pr = np.array(pr)\\r\\n\\r\\n pinterps = [] # lists of interpolated precisions for every confidence level\\r\\n ranks = np.linspace(0, 1, 11)\\r\\n idxs_interpolations = [] # list of indexes of the interpolated precisions, just to plot the recall\\r\\n for pr_ in pr:\\r\\n pinterp = []\\r\\n idxs = []\\r\\n last_idx = -1\\r\\n for rank in ranks:\\r\\n idx = (np.abs(pr_[:, 1] - rank)).argmin() # find the closest recall to the rank\\r\\n\\r\\n if rank > pr_[idx, 1]: # this makes sure we are taking the closest recall at the right of the rank\\r\\n if idx+1 < pr_[:, 0].shape[0]:\\r\\n idx += 1\\r\\n interpolated_precision = np.max(pr_[idx:, 0]) # find the max precision within the interval\\r\\n if idx == last_idx: # just some checks for when the recall doesn't exist\\r\\n pinterp[-1] = 0\\r\\n idxs[-1] = 0\\r\\n pinterp.append(0)\\r\\n idxs.append(0)\\r\\n else:\\r\\n pinterp.append(interpolated_precision)\\r\\n idxs.append(idx)\\r\\n last_idx = idx\\r\\n pinterps.append(pinterp)\\r\\n idxs_interpolations.append(idxs)\\r\\n APs = np.array(pinterps).mean(axis=1) # the AP is the average of the interpolated precisions\\r\\n mAP = APs.mean() # mAP is the mean of all the APs\\r\\n\\r\\n return pr, pinterps, idxs_interpolations, mAP, APs\",\n \"def recallAndPrecision(self,train = None,test = None,k = 8,nitem = 10):\\n train = train or self.traindata\\n test = test or self.testdata\\n hit = 0\\n recall = 0\\n precision = 0\\n for user in train.keys():\\n tu = test.get(user,{})\\n rank = self.recommend(user, train = train,k = k,nitem = nitem) \\n for item,_ in rank.items():\\n if item in tu:\\n hit += 1\\n recall += len(tu)\\n precision += nitem\\n return (hit / (recall * 1.0),hit / (precision * 1.0))\",\n \"def update(\\n self, outputs: torch.Tensor, targets: torch.Tensor\\n ) -> Tuple[float, float, float]:\\n tn, fp, fn, tp, support = super().update(outputs=outputs, targets=targets)\\n precision_value, recall_value, f1_value = get_binary_metrics(\\n tp=tp, fp=fp, fn=fn, zero_division=self.zero_division\\n )\\n return precision_value, recall_value, f1_value\",\n \"def update(self, y_pred: torch.Tensor, y_true: torch.Tensor) -> None:\\n self.accuracy.update(y_pred, y_true)\\n self.roc_auc.update(y_pred, y_true)\\n self.auc_score.update(y_pred, y_true)\\n self.precision.update(y_pred, y_true)\\n self.recall.update(y_pred, y_true)\\n self.f1.update(y_pred, y_true)\\n self.hamming_distance.update(y_pred, y_true)\\n self.stat_scores.update(y_pred, y_true)\",\n \"def compute_precision_recall_bbox(per_image_detections, per_image_gts, num_gt_classes,\\n matching_iou_threshold=0.5):\\n per_image_eval = per_image_evaluation.PerImageEvaluation(\\n num_groundtruth_classes=num_gt_classes,\\n matching_iou_threshold=matching_iou_threshold,\\n nms_iou_threshold=1.0,\\n nms_max_output_boxes=10000\\n )\\n\\n print('Running per-object analysis...')\\n\\n detection_tp_fp = defaultdict(list) # key is the category; in each list, 1 is tp, 0 is fp\\n detection_scores = defaultdict(list)\\n num_total_gt = defaultdict(int)\\n\\n for image_id, dets in tqdm(per_image_detections.items()):\\n\\n detected_boxes = np.array(dets['boxes'], dtype=np.float32)\\n detected_scores = np.array(dets['scores'], dtype=np.float32)\\n # labels input to compute_object_detection_metrics() needs to start at 0, not 1\\n detected_labels = np.array(dets['labels'], dtype=np.int) - 1 # start at 0\\n # num_detections = len(dets['boxes'])\\n\\n gts = per_image_gts[image_id]\\n gt_boxes = np.array(gts['gt_boxes'], dtype=np.float32)\\n gt_labels = np.array(gts['gt_labels'], dtype=np.int) - 1 # start at 0\\n num_gts = len(gts['gt_boxes'])\\n\\n groundtruth_is_difficult_list = np.zeros(num_gts, dtype=bool) # place holders - we don't have these\\n groundtruth_is_group_of_list = np.zeros(num_gts, dtype=bool)\\n\\n # to prevent 'Invalid dimensions for box data.' error\\n if num_gts == 0:\\n # this box will not match any detections\\n gt_boxes = np.array([[0, 0, 0, 0]], dtype=np.float32)\\n\\n scores, tp_fp_labels, is_class_correctly_detected_in_image = (\\n per_image_eval.compute_object_detection_metrics(\\n detected_boxes=detected_boxes,\\n detected_scores=detected_scores,\\n detected_class_labels=detected_labels,\\n groundtruth_boxes=gt_boxes,\\n groundtruth_class_labels=gt_labels,\\n groundtruth_is_difficult_list=groundtruth_is_difficult_list,\\n groundtruth_is_group_of_list=groundtruth_is_group_of_list\\n )\\n )\\n\\n for i, tp_fp_labels_cat in enumerate(tp_fp_labels):\\n assert sum(tp_fp_labels_cat) <= sum(gt_labels == i) # true positives < gt of that category\\n cat = i + 1 # categories start at 1\\n detection_tp_fp[cat].append(tp_fp_labels_cat)\\n detection_scores[cat].append(scores[i])\\n num_total_gt[cat] += sum(gt_labels == i) # gt_labels start at 0\\n\\n all_scores = []\\n all_tp_fp = []\\n\\n print('Computing precision recall for each category...')\\n per_cat_metrics = {}\\n for i in range(num_gt_classes):\\n cat = i + 1\\n scores_cat = np.concatenate(detection_scores[cat])\\n tp_fp_cat = np.concatenate(detection_tp_fp[cat]).astype(np.bool)\\n all_scores.append(scores_cat)\\n all_tp_fp.append(tp_fp_cat)\\n\\n precision, recall = metrics.compute_precision_recall(\\n scores_cat, tp_fp_cat, num_total_gt[cat]\\n )\\n average_precision = metrics.compute_average_precision(precision, recall)\\n\\n per_cat_metrics[cat] = {\\n 'category': cat,\\n 'precision': precision,\\n 'recall': recall,\\n 'average_precision': average_precision,\\n 'scores': scores_cat,\\n 'tp_fp': tp_fp_cat,\\n 'num_gt': num_total_gt[cat]\\n }\\n print('Number of ground truth in category {} is {}'.format(cat, num_total_gt[cat]))\\n\\n # compute one-class precision/recall/average precision (if every box is just of an object class)\\n all_scores = np.concatenate(all_scores)\\n all_tp_fp = np.concatenate(all_tp_fp)\\n overall_gt_count = sum(num_total_gt.values())\\n\\n one_class_prec, one_class_recall = metrics.compute_precision_recall(\\n all_scores, all_tp_fp, overall_gt_count\\n )\\n one_class_average_precision = metrics.compute_average_precision(one_class_prec, one_class_recall)\\n\\n per_cat_metrics['one_class'] = {\\n 'category': 'one_class',\\n 'precision': one_class_prec,\\n 'recall': one_class_recall,\\n 'average_precision': one_class_average_precision,\\n 'scores': all_scores,\\n 'tp_fp': all_tp_fp,\\n 'num_gt': overall_gt_count\\n }\\n\\n return per_cat_metrics\",\n \"def update(self, outputs: torch.Tensor, targets: torch.Tensor) -> Tuple[float, float, float]:\\n tn, fp, fn, tp, support = super().update(outputs=outputs, targets=targets)\\n precision_value, recall_value, f1_value = get_binary_metrics(\\n tp=tp, fp=fp, fn=fn, zero_division=self.zero_division\\n )\\n return precision_value, recall_value, f1_value\",\n \"def compute_recall_precision_fscore(truth: List[Node], prediction: List[Node]) \\\\\\n -> Tuple[numpy.ndarray, numpy.ndarray, numpy.ndarray]:\\n recall = numpy.zeros((len(truth), len(prediction)))\\n precision = numpy.zeros((len(truth), len(prediction)))\\n fscore = numpy.zeros((len(truth), len(prediction)))\\n\\n for i, t in enumerate(truth):\\n for j, p in enumerate(prediction):\\n r, p, f = pixel_metrics(t, p)\\n recall[i, j] = r\\n precision[i, j] = p\\n fscore[i, j] = f\\n\\n return recall, precision, fscore\",\n \"def precision(y_test, y_pred):\\n\\treturn precision_score(y_test, y_pred)\",\n \"def mask_evaluation(annotation_mask, result_mask, idx):\\n\\n true_positive = np.sum(np.logical_and(annotation_mask == 255, result_mask == 255)) \\n false_positive = np.sum(np.logical_and(result_mask == 255, annotation_mask != result_mask))\\n false_negative = np.sum(np.logical_and(annotation_mask == 255, annotation_mask != result_mask))\\n\\n precision = true_positive / (true_positive + false_positive)\\n recall = true_positive / (true_positive + false_negative)\\n f1_measure = 2 * ((precision * recall) / (precision + recall))\\n\\n return recall, precision, f1_measure\",\n \"def _eval_classifier(self):\\n\\n y_pred_baseline = self.df_baseline[self.score_column]\\n y_pred_sample = self.df_sample[self.score_column]\\n\\n y_label_baseline = self.df_baseline[self.label_column]\\n y_label_sample = self.df_sample[self.label_column]\\n\\n precision_baseline = precision_score(y_label_baseline, y_pred_baseline)\\n recall_baseline = recall_score(y_label_baseline, y_pred_baseline)\\n acc_baseline = accuracy_score(y_label_baseline, y_pred_baseline)\\n f1_baseline = f1_score(y_label_baseline, y_pred_baseline)\\n try:\\n auc_baseline = roc_auc_score(y_label_baseline, y_pred_baseline)\\n except ValueError:\\n auc_baseline = \\\"NA\\\"\\n\\n precision_sample = precision_score(y_label_sample, y_pred_sample)\\n recall_sample = recall_score(y_label_sample, y_pred_sample)\\n acc_sample = accuracy_score(y_label_sample, y_pred_sample)\\n f1_sample = f1_score(y_label_sample, y_pred_sample)\\n try:\\n auc_sample = roc_auc_score(y_label_sample, y_pred_sample)\\n except ValueError:\\n auc_sample = \\\"NA\\\"\\n\\n metrics_df = pd.DataFrame(\\n {\\n \\\"Accuracy\\\": [acc_baseline, acc_sample],\\n \\\"Precision\\\": [precision_baseline, precision_sample],\\n \\\"Recall\\\": [recall_baseline, recall_sample],\\n \\\"F1\\\": [f1_baseline, f1_sample],\\n \\\"AUC\\\": [auc_baseline, auc_sample],\\n },\\n index=[\\\"baseline\\\", \\\"sample\\\"],\\n )\\n\\n self.performance_comparison = metrics_df\",\n \"def get_recall_and_precision(wdb, dets, th_conf, th_iou):\\n hit_total, false_total, miss_total = 0, 0, 0\\n\\n for idx in range(wdb.get_image_count()):\\n data = wdb.get_annos_by_image_index(idx)\\n img_path = data['image_path']\\n fname = os.path.splitext(os.path.basename(img_path))[0]\\n\\n dets4fname = []\\n\\n for d in dets:\\n if d['filename'] == fname: # we have only ONE corresponding element\\n dets4fname = d['data']\\n break\\n\\n dets2compare = [e for e in dets4fname if e['c'] >= th_conf] # detections having confidence larger than threshold only\\n\\n hit, false, miss = count_hits(data['annos'], dets2compare, th_iou)\\n hit_total += hit\\n false_total += false\\n miss_total += miss\\n\\n recall = hit_total/float(hit_total+miss_total+0.000001)\\n precision = hit_total/float(hit_total+false_total+0.000001)\\n\\n return recall, precision\",\n \"def evaluate(ground_truth, prediction):\\n\\n def prfs_to_dict(prfs):\\n \\\"\\\"\\\"Returns a precision_recall_fscore_support() result as a dict.\\\"\\\"\\\"\\n return {\\\"precision\\\": prfs[0], \\\"recall\\\": prfs[1], \\\"fscore\\\": prfs[2]}\\n\\n results = {}\\n items_count = len(ground_truth)\\n\\n # accuracy\\n accuracy = accuracy_score(ground_truth, prediction)\\n results[\\\"accuracy\\\"] = accuracy\\n\\n # confusion matrix\\n categories = set(ground_truth) | set(prediction)\\n confusions = {\\n gold: {pred: 0 for pred in categories} for gold in categories\\n }\\n for g, p in zip(ground_truth, prediction):\\n confusions[g][p] += 1\\n results[\\\"confusions\\\"] = confusions\\n\\n # class wise precision, recall & f1\\n classwise = precision_recall_fscore_support(\\n ground_truth, prediction, average=None, warn_for=()\\n )\\n results[\\\"true_cat_dist\\\"] = list(classwise[-1])\\n results[\\\"classwise\\\"] = {\\n str(cl): prfs_to_dict(\\n [classwise[0][cl], classwise[1][cl], classwise[2][cl]]\\n )\\n for cl in categories\\n }\\n\\n # average precision, recall & f1\\n results[\\\"macro_avg\\\"] = prfs_to_dict(\\n precision_recall_fscore_support(\\n ground_truth,\\n prediction,\\n average=\\\"macro\\\",\\n pos_label=None,\\n warn_for=(),\\n )\\n )\\n results[\\\"micro_avg\\\"] = prfs_to_dict(\\n precision_recall_fscore_support(\\n ground_truth,\\n prediction,\\n average=\\\"micro\\\",\\n pos_label=None,\\n warn_for=(),\\n )\\n )\\n results[\\\"weigh_avg\\\"] = prfs_to_dict(\\n precision_recall_fscore_support(\\n ground_truth,\\n prediction,\\n average=\\\"weighted\\\",\\n pos_label=None,\\n warn_for=(),\\n )\\n )\\n\\n # marginals\\n gold_category_distribution = {\\n g: sum([confusions[g][p] for p in categories]) for g in categories\\n }\\n pred_category_distribution = {\\n p: sum([confusions[g][p] for g in categories]) for p in categories\\n }\\n\\n # kappa\\n expected_agreement_fleiss = sum(\\n [\\n (\\n (gold_category_distribution[c] + pred_category_distribution[c])\\n / (2.0 * items_count)\\n )\\n ** 2\\n for c in categories\\n ]\\n )\\n expected_agreement_cohen = sum(\\n [\\n (float(gold_category_distribution[c]) / items_count)\\n * (float(pred_category_distribution[c]) / items_count)\\n for c in categories\\n ]\\n )\\n kappa_fleiss = (\\n 1.0\\n * (accuracy - expected_agreement_fleiss)\\n / (1 - expected_agreement_fleiss)\\n )\\n kappa_cohen = (\\n 1.0\\n * (accuracy - expected_agreement_cohen)\\n / (1 - expected_agreement_cohen)\\n )\\n results[\\\"k_fleiss\\\"] = {\\n \\\"k\\\": kappa_fleiss,\\n \\\"AE\\\": expected_agreement_fleiss,\\n \\\"AO\\\": accuracy,\\n }\\n results[\\\"k_cohen\\\"] = {\\n \\\"k\\\": kappa_cohen,\\n \\\"AE\\\": expected_agreement_cohen,\\n \\\"AO\\\": accuracy,\\n }\\n\\n return results\",\n \"def print_precision_recall(classifier, test_set):\\n known_set = collections.defaultdict(set)\\n computed_set = collections.defaultdict(set)\\n \\n for i, (features, label) in enumerate(test_set):\\n known_set[label].add(i)\\n predicted = classifier.classify(features)\\n computed_set[predicted].add(i)\\n\\n print('pos precision:', precision(known_set['pos'], computed_set['pos']))\\n print('pos recall:', recall(known_set['pos'], computed_set['pos']))\\n print('neg precision:', precision(known_set['neg'], computed_set['neg']))\\n print('neg recall:', recall(known_set['neg'], computed_set['neg']))\",\n \"def detection_analysis(y_pred, y_true):\\n print(\\\"Precision: \\\", sm.precision_score(y_pred, y_true))\\n print(\\\"Recall: \\\", sm.recall_score(y_pred, y_true))\\n print(\\\"Accuracy: \\\", sm.accuracy_score(y_pred, y_true))\\n print(\\\"\\\\n\\\")\",\n \"def calculate_precision_recall_all_images(\\n all_prediction_boxes, all_gt_boxes, iou_threshold):\\n # Find total true positives, false positives and false negatives\\n # over all images\\n\\n tot_num_tp, tot_num_fp, tot_num_fn = 0, 0, 0\\n\\n for image in range(len(all_prediction_boxes)):\\n prediction_boxes = all_prediction_boxes[image]\\n gt_boxes = all_gt_boxes[image]\\n\\n image_results = calculate_individual_image_result(\\n prediction_boxes,\\n gt_boxes,\\n iou_threshold)\\n\\n tot_num_tp += image_results[\\\"true_pos\\\"]\\n tot_num_fp += image_results[\\\"false_pos\\\"]\\n tot_num_fn += image_results[\\\"false_neg\\\"]\\n\\n # Compute precision, recall\\n precision = calculate_precision(tot_num_tp, tot_num_fp, tot_num_fn)\\n recall = calculate_recall(tot_num_tp, tot_num_fp, tot_num_fn)\\n\\n return (precision, recall)\",\n \"def precision_recall(conf_matrix):\\n tp_count, fp_count, fn_count, tn_count = conf_matrix[0][0], conf_matrix[0][1], conf_matrix[1][0], conf_matrix[1][1]\\n\\n precision = tp_count / (tp_count + fp_count)\\n recall = tp_count / (tp_count + fn_count)\\n\\n return precision, recall\",\n \"def painting_matching_ml(imgs, db_imgs, method_list, text_masks, author_text, gt_text, metrics, weights, splits=30, max_rank=10):\\n descriptor_extractors = [get_descriptor_extractor(method_name) for method_name in method_list]\\n tmp_img_format = []\\n tmp_mask_format = []\\n tmp_text_format = []\\n for i in range(len(imgs)):\\n for j in range(len(imgs[i])):\\n tmp_img_format.append(imgs[i][j])\\n tmp_mask_format.append(text_masks[i][j])\\n tmp_text_format.append(author_text[i][j])\\n\\n #db_imgs = [img[0] for img in db_imgs]\\n db_img_splits = [i*len(db_imgs)//splits for i in range(splits-1)]\\n \\n scores = []\\n query_descriptors = extract_descriptors(tmp_img_format, descriptor_extractors, method_list, tmp_text_format, tmp_mask_format) \\n #np.array([extract_descriptors(img, matching_methods, mask) for img, mask in zip(tmp_img_format, tmp_mask_format)])\\n print(\\\"Starting db extraction + matching\\\")\\n for split in tqdm(range(splits-2)):\\n db_descriptors = extract_descriptors(db_imgs[db_img_splits[split]:db_img_splits[split+1]], descriptor_extractors, method_list, gt_text[db_img_splits[split]:db_img_splits[split+1]], None) #np.array([mrhm(db_img) for db_img in db_imgs[db_img_splits[split]:db_img_splits[split+1]]])\\n scores.append(compare_descriptors(query_descriptors, db_descriptors, metrics, method_list, weights))\\n # compare_descriptors(query_descriptors, db_descriptors, descriptor_comp_methods, descriptor_names, weights)\\n db_descriptors = extract_descriptors(db_imgs[db_img_splits[-1]:], descriptor_extractors, method_list, gt_text[db_img_splits[-1]:], None)\\n scores.append(compare_descriptors(query_descriptors, db_descriptors, metrics, method_list, weights))\\n \\n # concatenate all the results\\n scores = np.concatenate(scores, 1)\\n \\n top_k_matches = np.argpartition(scores, list(range(max_rank)))[:, :max_rank]\\n return top_k_matches\",\n \"def recall(prediction, ground_truth):\\n return precision_recall_f1(prediction, ground_truth)[1]\",\n \"def precision(y_true, y_pred, average, labels):\\n y_true, y_pred = check_metric_args(y_true, y_pred, average, labels)\\n\\n # At this point, you can be sure that y_true and y_pred are one hot encoded.\\n result = None\\n m = len(y_true)\\n n = len(labels)\\n\\n #call get_confusion_matrix function and put the result in confusion_matrix\\n confusion_matrix = get_confusion_matrix(y_true, y_pred, labels)\\n\\n #compute the result if using micro-averages\\n if average == \\\"micro\\\":\\n numerator = np.trace(confusion_matrix)\\n denominator = np.sum(confusion_matrix)\\n result = numerator/denominator\\n\\n #compute the precision independently for each class and then take the average \\n elif average == \\\"macro\\\":\\n diag = np.diag(confusion_matrix)\\n row_sums = np.sum(confusion_matrix, axis=1)\\n row_sums_adjusted = np.array([1 if val == 0 else val for val in row_sums])\\n result = np.mean(diag/row_sums_adjusted)\\n\\n else:\\n diag = np.diag(confusion_matrix)\\n row_sums = np.sum(confusion_matrix, axis=1)\\n row_sums_adjusted = np.array([1 if val == 0 else val for val in row_sums])\\n result = diag/row_sums_adjusted\\n\\n return result\",\n \"def add_example(self, precision: float, recall: float, f1: float):\\n self._count += 1\\n self._total_precision += precision\\n self._total_recall += recall\\n self._total_f1 += f1\",\n \"def compare_recall_precisions_from_predictions(true, score_dict, **kwargs):\\n pr = OrderedDict()\\n for name, score in score_dict.items():\\n p, r, threshold = precision_recall_curve(true, score)\\n pr[name] = [p, r]\\n compare_recall_precision_graph(pr, **kwargs)\",\n \"def score(label_recall, label_precision, pred_recall, pred_precision):\\n f11 = 6*label_recall*label_precision/(5*label_recall+label_precision)\\n f12 = 5*pred_recall*pred_precision/(2*pred_recall+3*pred_precision)\\n\\n return 0.4*f11 + 0.6*f12\",\n \"def prf_cal(y_pred,y_true,k):\\r\\n GT=np.sum(y_true[y_true==1.])\\r\\n instance_num=y_true.shape[0]\\r\\n prediction_num=instance_num*k\\r\\n\\r\\n sort_indices = np.argsort(y_pred)\\r\\n sort_indices=sort_indices[:,::-1]\\r\\n static_indices = np.indices(sort_indices.shape)\\r\\n sorted_annotation= y_true[static_indices[0],sort_indices]\\r\\n top_k_annotation=sorted_annotation[:,0:k]\\r\\n TP=np.sum(top_k_annotation[top_k_annotation==1.])\\r\\n recall=TP/GT\\r\\n precision=TP/prediction_num\\r\\n f1=2.*recall*precision/(recall+precision)\\r\\n return precision, recall, f1\",\n \"def PredictMatch(yourName, tgtName, teamModels, db):\\n\\n teamTgtPipe, teamYourPipe = QueryTeamData(tgtName, yourName, db)\\n featureCoefTgt, probTgt = PredictOp(teamTgtPipe, teamYourPipe, tgtName, teamModels)\\n featureCoefYour, probYour = PredictOp(teamYourPipe, teamTgtPipe, yourName, teamModels)\\n odds = round(probYour / (probYour + probTgt), 2)\\n\\n # In featureCoefYour, you want INCREASE those with POSTIVE COEF, DECREASE those with NEGATIVE COEF\\n # In featureCoefTgt, you want to do the opposite\\n\\n # reverse both the sign of the coef, and '_op' in features so as to be the same with featureCoefYour\\n featureCoefTgt['coef'] = - featureCoefTgt['coef']\\n featureCoefTgt.features = [ii[:-3] if \\\"_op\\\" in ii else ii + '_op' for ii in featureCoefTgt.features]\\n\\n # Combine only the most important 10 features\\n # featureBoth = featureCoefTgt[11:].append(featureCoefYour[11:])\\n\\n # Combine only all the most important features\\n featureBoth = featureCoefTgt.append(featureCoefYour)\\n\\n # get action recommendations\\n # Somehow the pandas here uses a deprecated para cols, instaed of the new one subset\\n #featureBoth.drop_duplicates(subset = 'features', take_last = True, inplace = True)\\n featureBoth.drop_duplicates(cols = 'features', take_last = True, inplace = True)\\n actions, featureImprove = GetActions(featureBoth)\\n Imp = 0.1\\n oddsNew = ImprovedScore(tgtName, yourName, teamModels, featureImprove, teamTgtPipe, teamYourPipe, Imp)\\n\\n return odds, oddsNew, actions\",\n \"def precision_recall_calculator(reactions_removed, reactions_restored):\\n \\n reactions_removed = set(reactions_removed)\\n reactions_restored = set(reactions_restored)\\n \\n true_positives = reactions_removed.intersection(reactions_restored)\\n \\n precision = len(true_positives)/len(reactions_restored)\\n recall = len(true_positives)/len(reactions_removed)\\n \\n return precision, recall\",\n \"def evaluate(self, test_dir='data/dev', target='real'):\\n test_data = {c: os.path.join(test_dir, c) for c in self.classes}\\n if not target in test_data:\\n print('Error: target class does not exist in test data.')\\n return\\n outcomes = {'TP': 0, 'TN': 0, 'FP': 0, 'FN': 0}\\n # >>> YOUR ANSWER HERE\\n data = []\\n for c in test_data:\\n docs = open(test_data[c]).readlines()\\n for doc in docs:\\n preprocess_doc = doc.strip()\\n data.append((c, preprocess_doc))\\n for item in data:\\n predict_ans = self.predict(item[1])\\n if item[0] == 'real':\\n if predict_ans == 'real':\\n outcomes['TP'] += 1\\n else:\\n outcomes['FN'] += 1\\n else:\\n if predict_ans == 'real':\\n outcomes['FP'] += 1\\n else:\\n outcomes['TN'] += 1\\n precision = outcomes['TP'] / (outcomes['TP'] + outcomes['FP']) # replace with equation for precision\\n recall = outcomes['TP'] / (outcomes['TP'] + outcomes['FN']) # replace with equation for recall\\n f1_score = 2 * ((precision * recall) / (precision + recall)) # replace with equation for f1\\n # >>> END YOUR ANSWER\\n return precision, recall, f1_score\",\n \"def evaluate_old(goldfile, sysfile):\\n gold = {i[\\\"id\\\"]: i[\\\"label\\\"] for i in read(goldfile)}\\n sys = {i[\\\"id\\\"]: i[\\\"prediction\\\"] for i in read(sysfile)}\\n\\n labels, preds = [], []\\n for idx in gold:\\n labels.append(gold[idx])\\n preds.append(sys[idx])\\n\\n precision_macro, recall_macro, f1_macro, _ = precision_recall_fscore_support(labels, preds, labels=[0, 1],\\n average=\\\"macro\\\")\\n print(\\\"F1-macro score for test data predictions are: %.4f\\\" % f1_macro)\",\n \"def __precision_and_recall(self, y_train, y_pred, average='micro', verbose = True):\\n # Precision and recall\\n pscore = precision_score(y_train, y_pred, average=average)\\n rscore = recall_score(y_train, y_pred, average=average)\\n\\n # harmonic mean of precision\\n f1 = f1_score(y_train, y_pred, average='micro')\\n\\n if verbose:\\n message = F\\\"Precision of model: {pscore} | Recall of model {rscore}\\\"\\n self.__logger.info(message)\\n\\n message = f\\\"Harmonic Mean of precision and recall: F1 score: {f1}\\\"\\n self.__logger.info(message)\\n return str(f1)\",\n \"def test_calculate_precision_happy_path(\\n mock_engine: DicomImagePiiVerifyEngine,\\n gt_labels: List[dict],\\n all_pos: List[dict],\\n expected_result: float,\\n):\\n # Act\\n test_precision = mock_engine.calculate_precision(gt_labels, all_pos)\\n\\n # Assert\\n assert test_precision == expected_result\",\n \"def _imequalize(self, results):\\n for key in results.get('img_fields', ['image']):\\n img = results[key]\\n results[key] = mmcv.imequalize(img).astype(img.dtype)\",\n \"def prf_analysis(y_true: list, y_pred: list) -> None:\\n print('Precision: {:,.2f}'.format(precision_score(y_true, y_pred)))\\n print('Recall : {:,.2f}'.format(recall_score(y_true, y_pred)))\\n print('F1 : {:,.2f}'.format(f1_score(y_true, y_pred)))\\n print('Accuracy : {:,.2f}'.format(accuracy_score(y_true, y_pred)))\\n return None\",\n \"def compute(self) -> None:\\n \\n self.model.eval()\\n \\n with torch.no_grad():\\n for (input, target, _) in self.loader:\\n\\n # self.model = self.model.train(False) # TEST @lacoupe\\n output, _ = self.model(input)\\n \\n output = (output >= 0.5)\\n \\n for out, tar in zip(output, target):\\n \\n tar = bool(tar)\\n \\n if out and tar:\\n self.confusion['true_positive'] += 1\\n elif not out and not tar:\\n self.confusion['true_negative'] += 1\\n elif out and not tar:\\n self.confusion['false_positive'] += 1\\n elif not out and tar:\\n self.confusion['false_negative'] += 1\\n \\n self.accuracy = (self.confusion['true_positive'] + self.confusion['true_negative']) \\\\\\n / sum(list(self.confusion.values()))\\n \\n if (self.confusion['true_positive'] + self.confusion['false_positive']) == 0.:\\n self.precision = 0.\\n else:\\n self.precision = self.confusion['true_positive'] \\\\\\n / (self.confusion['true_positive'] + self.confusion['false_positive'])\\n \\n if (self.confusion['true_positive'] + self.confusion['false_negative']) == 0.:\\n self.recall = 0.\\n else:\\n self.recall = self.confusion['true_positive'] \\\\\\n / (self.confusion['true_positive'] + self.confusion['false_negative'])\\n \\n if (self.precision + self.recall) == 0.:\\n self.f1_score = 0.\\n else:\\n self.f1_score = 2 * self.precision * self.recall / (self.precision + self.recall)\",\n \"def calc_accuracy(self, top_matches):\\n # Calculate Translation accuracy\\n # Accuracy at K\\n for k in [1, 5, 10]:\\n top_k_matches = top_matches[:, :k]\\n # Checks for one match\\n one_matching = (top_k_matches == self.tgt_ind[:, None]).sum(1)\\n # Checks for multiple translations\\n matching = {}\\n for i, src_id in enumerate(self.src_ind):\\n matching[src_id] = min(matching.get(src_id, 0) + one_matching[i], 1)\\n precision_at_k = 100 * np.mean(list(matching.values()))\\n print(\\\"%i source words - Precision at k = %i: %f\\\" % (len(matching), k, precision_at_k))\",\n \"def accuracy(self):\",\n \"def evaluate(clf, dataset, feature_list, features, labels, num_iter, params):\\n\\n features_train, features_test, labels_train, labels_test = \\\\\\n train_test_split(features, labels, test_size=0.3, random_state=42)\\n\\n\\n\\n precision_values = []\\n recall_values = []\\n accuracy_values = []\\n print clf\\n for i in xrange(0, num_iter):\\n #print params\\n clf = GridSearchCV(clf, params)\\n clf.fit(features_train, labels_train)\\n print '*****************************'\\n print clf.best_estimator_\\n print clf.best_params_\\n\\n clf = clf.best_estimator_\\n #test_classifier(clf, dataset, feature_list)\\n pred = clf.predict(features_test)\\n precision_values.append(precision_score(labels_test, pred))\\n recall_values.append(recall_score(labels_test, pred))\\n accuracy_values.append(accuracy_score(labels_test, pred))\\n print 'Recall score: ', mean(recall_values)\\n print 'Precision score: ', mean(precision_values)\\n print 'Accuracy score: ' , mean(accuracy_values)\",\n \"def __call__(self, query_result: List, gallery_info: List) -> (float, Dict):\\n aps = list()\\n\\n # For mAP calculation\\n pseudo_score = np.arange(0, len(gallery_info))[::-1]\\n\\n recall_at_k = dict()\\n for k in self._hyper_params[\\\"recall_k\\\"]:\\n recall_at_k[k] = 0\\n\\n gallery_label = np.array([gallery_info[idx][\\\"label_idx\\\"] for idx in range(len(gallery_info))])\\n for i in range(len(query_result)):\\n ranked_idx = query_result[i][\\\"ranked_neighbors_idx\\\"]\\n gt = (gallery_label[query_result[i][\\\"ranked_neighbors_idx\\\"]] == query_result[i][\\\"label_idx\\\"])\\n aps.append(average_precision_score(gt, pseudo_score[:len(gt)]))\\n\\n # deal with 'gallery as query' test\\n if gallery_info[ranked_idx[0]][\\\"path\\\"] == query_result[i][\\\"path\\\"]:\\n gt.pop(0)\\n self.compute_recall_at_k(gt, recall_at_k)\\n\\n mAP = np.mean(aps) * 100\\n\\n for k in recall_at_k:\\n recall_at_k[k] = recall_at_k[k] * 100 / len(query_result)\\n\\n return mAP, recall_at_k\",\n \"def precision1(y_test, predictions):\\n precision = 0.0\\n\\n for i in range(y_test.shape[0]):\\n intersection = 0.0\\n hXi = 0.0\\n for j in range(y_test.shape[1]):\\n hXi = hXi + int(predictions[i,j])\\n if int(y_test[i,j]) == 1 and int(predictions[i,j]) == 1:\\n intersection += 1\\n \\n if hXi != 0:\\n precision = precision + float(intersection/hXi)\\n \\n\\n precision = float(precision/y_test.shape[0])\\n \\n return precision\",\n \"def update(self) -> None:\\n self.faithful = self.is_faithful()\\n if self.faithful:\\n old_class_names = self.class_names\\n old_training_image_names = self.training_image_names\\n self.class_names = self.find_class_names()\\n self.training_image_names = self.find_training_image_names()\\n self.extracted_features = list()\\n if old_class_names != self.class_names or old_training_image_names != self.training_image_names:\\n self.generate_csv_dictionary()\\n return\",\n \"def test_calculate_recall_happy_path(\\n mock_engine: DicomImagePiiVerifyEngine,\\n gt_labels: dict,\\n all_pos: dict,\\n expected_result: float,\\n):\\n # Act\\n test_recall = mock_engine.calculate_recall(gt_labels, all_pos)\\n\\n # Assert\\n assert test_recall == expected_result\",\n \"def compute_average_precision_per_class(num_true_cases, gt_boxes,\\n\\t\\t\\t\\t\\t\\t\\t\\t\\t\\tprediction_file, iou_threshold, use_2007_metric):\\n\\twith open(prediction_file) as f:\\n\\t\\timage_ids = []\\n\\t\\tboxes = []\\n\\t\\tscores = []\\n\\t\\tfor line in f:\\n\\t\\t\\tt = line.rstrip().split(\\\" \\\")\\n\\t\\t\\timage_ids.append(t[0])\\n\\t\\t\\tscores.append(float(t[1]))\\n\\t\\t\\tbox = torch.tensor([float(v) for v in t[2:]]).unsqueeze(0)\\n\\t\\t\\tbox -= 1.0 # convert to python format where indexes start from 0\\n\\t\\t\\tboxes.append(box)\\n\\t\\tscores = np.array(scores)\\n\\t\\tsorted_indexes = np.argsort(-scores)\\n\\t\\tboxes = [boxes[i] for i in sorted_indexes]\\n\\t\\timage_ids = [image_ids[i] for i in sorted_indexes]\\n\\t\\ttrue_positive = np.zeros(len(image_ids))\\n\\t\\tfalse_positive = np.zeros(len(image_ids))\\n\\t\\tmatched = set()\\n\\t\\tfor i, image_id in enumerate(image_ids):\\n\\t\\t\\tbox = boxes[i]\\n\\t\\t\\tif image_id not in gt_boxes:\\n\\t\\t\\t\\tfalse_positive[i] = 1\\n\\t\\t\\t\\tcontinue\\n\\n\\t\\t\\tgt_box = gt_boxes[image_id]\\n\\t\\t\\tious = box_utils.iou_of(box, gt_box)\\n\\t\\t\\tmax_iou = torch.max(ious).item()\\n\\t\\t\\tmax_arg = torch.argmax(ious).item()\\n\\t\\t\\tif max_iou > iou_threshold:\\n\\t\\t\\t\\tif (image_id, max_arg) not in matched:\\n\\t\\t\\t\\t\\ttrue_positive[i] = 1\\n\\t\\t\\t\\t\\tmatched.add((image_id, max_arg))\\n\\t\\t\\t\\telse:\\n\\t\\t\\t\\t\\tfalse_positive[i] = 1\\n\\t\\t\\telse:\\n\\t\\t\\t\\tfalse_positive[i] = 1\\n\\n\\ttrue_positive = true_positive.cumsum()\\n\\tfalse_positive = false_positive.cumsum()\\n\\tprecision = true_positive / (true_positive + false_positive)\\n\\trecall = true_positive / num_true_cases\\n\\tif use_2007_metric:\\n\\t\\treturn measurements.compute_voc2007_average_precision(precision, recall)\\n\\telse:\\n\\t\\treturn measurements.compute_average_precision(precision, recall)\",\n \"def testClassifier(x_train, y_train, x_test, y_test, clf):\\n #metrics = []\\n start = dt.now()\\n clf.fit(x_train, y_train)\\n end = dt.now()\\n print 'training time: ', (end - start)\\n \\n # add training time to metrics\\n #metrics.append(end-start)\\n \\n start = dt.now()\\n yhat = clf.predict(x_test)\\n end = dt.now()\\n print 'testing time: ', (end - start)\\n \\n # add testing time to metrics\\n #metrics.append(end-start)\\n \\n print 'classification report: '\\n# print classification_report(y_test, yhat)\\n pp(classification_report(y_test, yhat))\\n \\n print 'f1 score'\\n print f1_score(y_test, yhat, average='macro')\\n \\n print 'accuracy score'\\n accuracy = accuracy_score(y_test, yhat)\\n print accuracy\\n #metrics.append(accuracy)\\n #precision = precision_score(y_test, yhat, average=None)\\n #recall = recall_score(y_test, yhat, average=None)\\n \\n # add precision and recall values to metrics\\n #for p, r in zip(precision, recall):\\n # metrics.append(p)\\n # metrics.append(r)\\n \\n \\n #add macro-averaged F1 score to metrics\\n #metrics.append(f1_score(y_test, yhat, average='macro'))\\n \\n print 'confusion matrix:'\\n print confusion_matrix(y_test, yhat)\\n \\n # plot the confusion matrix\\n plt.imshow(confusion_matrix(y_test, yhat), interpolation='nearest')\\n plt.show()\\n \\n return accuracy\",\n \"def do_score_prediction(self):\\n \\n import iread.myio as mio\\n from igui.score_canvas import ScoreCanvas\\n exp_name = 'JPS_act_12_exp_4_accv_half_fc_j2'\\n exp_name_base = 'ASM_act_12_exp_4'\\n exp_base_folder = '/opt/visal/tmp/for_sijin/Data/H36M/H36MExp'\\n exp_path = iu.fullfile(exp_base_folder, 'folder_%s' % exp_name, 'batches.meta')\\n meta_base_path = iu.fullfile(exp_base_folder, 'folder_%s' % exp_name_base, 'batches.meta')\\n meta = mio.unpickle(exp_path)\\n meta_base = mio.unpickle(meta_base_path)\\n images_path = meta_base['images_path']\\n \\n pred_pose = meta['feature_list'][0]\\n gt_pose = meta['random_feature_list'][0]\\n ntotal = gt_pose.shape[-1]\\n print 'gt_pose_shape',gt_pose.shape\\n print 'pred_pose_shape', pred_pose.shape\\n ref_frame = 7600 # This is the index in test range\\n ## ref_frame = 2600 # This is the index in test range\\n test_range = self.test_data_provider.feature_range\\n ref_idx = test_range[ref_frame]\\n \\n n_to_show = 1000\\n \\n idx_to_show = np.random.choice(ntotal, n_to_show - 1)\\n idx_to_show = [ref_idx] + idx_to_show.tolist() \\n idx_to_show = np.asarray(idx_to_show, dtype=np.int).flatten()\\n \\n ref_pose = pred_pose[...,ref_idx].reshape((-1,1),order='F') \\n pose_to_eval =gt_pose[...,idx_to_show]\\n output_feature_name = 'fc_2' # <------------------Parameter\\n output_layer_idx = self.get_layer_idx(output_feature_name)\\n\\n # do it once <------------- Maybe it can support multiple batch in the future\\n data_dim = self.model_state['layers'][output_layer_idx]['outputs']\\n print 'data_dim', data_dim\\n \\n cur_data = [np.require(np.tile(ref_pose, [1,n_to_show]), \\\\\\n dtype=np.single,requirements='C'), \\\\\\n np.require(pose_to_eval.reshape((-1,n_to_show),order='F'),\\\\\\n dtype=np.single,requirements='C'), \\\\\\n np.require(np.zeros((1,n_to_show),dtype=np.single), \\\\\\n requirements='C'),\\n np.require(np.zeros((n_to_show,data_dim),dtype=np.single), \\\\\\n requirements='C')]\\n residuals = cur_data[1][...,0].reshape((-1,1),order='F') - cur_data[1]\\n dp = self.test_data_provider\\n mpjpe = dutils.calc_mpjpe_from_residual(residuals, dp.num_joints)\\n\\n gt_score = dp.calc_score(mpjpe, dp.mpjpe_factor/dp.max_depth,\\\\\\n dp.mpjpe_offset/dp.max_depth).reshape((1,n_to_show)).flatten()\\n self.libmodel.startFeatureWriter(cur_data, output_layer_idx)\\n self.finish_batch()\\n score = cur_data[-1].T\\n print 'dim score', score.shape, 'dim gt_score', gt_score.shape\\n score = score.flatten()\\n # score = gt_score.flatten()\\n def my_sort_f(k):\\n if k == 0:\\n return 10000000\\n else:\\n return score[k]\\n sorted_idx = sorted(range(n_to_show), key=my_sort_f,reverse=True)\\n s_to_show = [idx_to_show[k] for k in sorted_idx]\\n sorted_score = np.asarray( [score[k] for k in sorted_idx])\\n \\n pose_to_plot = self.convert_relskel2rel(cur_data[1])\\n sorted_pose = pose_to_plot[...,sorted_idx]\\n class ScorePoseCanvas(ScoreCanvas):\\n def __init__(self,data_dic):\\n import iread.h36m_hmlpe as h36m\\n ScoreCanvas.__init__(self,data_dic)\\n self.pose_data = data_dic['pose_data']\\n self.limbs = h36m.part_idx\\n self.tmp = 0\\n def show_image(self,ax):\\n # ScoreCanvas.show_image(self,ax)\\n # return\\n import Image\\n idx =self.cur_data_idx\\n if idx == 0:\\n self.tmp = self.tmp + 1\\n if self.tmp == 1:\\n img = self.load_image(idx)\\n ax.imshow(np.asarray(img))\\n return\\n print 'Current data idx %d ' % self.cur_data_idx\\n # params = {'elev':-89, 'azim':-107}\\n # params = {'elev':-69, 'azim':-107}\\n params = {'elev':-81, 'azim':-91} # frontal view\\n fig = plt.figure(100)\\n from mpl_toolkits.mplot3d import Axes3D\\n import imgproc\\n # new_ax = self.fig.add_axes( rng_rel,projection='polar')\\n new_ax = fig.add_subplot(111,projection='3d')\\n imgproc.turn_off_axis(new_ax)\\n cur_pose = self.pose_data[...,idx].reshape((3,-1),order='F')\\n dutils.show_3d_skeleton(cur_pose.T,\\\\\\n self.limbs, params)\\n xmin,xmax = np.min(cur_pose[0]),np.max(cur_pose[0])\\n ymin,ymax = np.min(cur_pose[1]),np.max(cur_pose[1])\\n zmin,zmax = np.min(cur_pose[2]),np.max(cur_pose[2])\\n def extent(x,y,ratio):\\n x = x + (x-y) * ratio\\n y = y + (y-x) * ratio\\n return -0.4,0.4\\n r = 0.1\\n new_ax.set_xlim(extent(xmin,xmax,r))\\n new_ax.set_ylim(extent(ymin,ymax,r))\\n new_ax.set_ylim(extent(zmin,zmax,r))\\n tmp_folder = '/public/sijinli2/ibuffer/2014-CVPR2015/tmp/images'\\n save_path = iu.fullfile(tmp_folder, 'tmp_image.png')\\n plt.savefig(save_path)\\n img = Image.open(save_path)\\n plt.close(100)\\n img_arr = np.asarray(img)\\n s = np.int(img_arr.shape[0]/5.0)\\n e = np.int(img_arr.shape[0] - s)\\n s = 0\\n e = img_arr.shape[0]\\n img_arr = img_arr[s:e,:,:]\\n ax.imshow(np.asarray(img_arr))\\n # ax.plot([1,0,0],[0,1,0],[0,0,1])\\n\\n\\n sc = ScorePoseCanvas({'x': np.asarray(range(len(idx_to_show))), 'y':sorted_score,\\\\\\n 'images_path': [images_path[k] for k in s_to_show], \\\\\\n 'pose_data':sorted_pose})\\n sc.start()\\n print 'max score is ' , sorted_score.max()\\n gt_sort_idx = sorted(range(n_to_show), key=lambda k:gt_score[k], reverse=True)\\n sorted_gt_score = np.asarray([gt_score[k] for k in gt_sort_idx])\\n sorted_score_by_gt = [score[k] for k in gt_sort_idx]\\n pl.plot(np.asarray(range(n_to_show)), sorted_gt_score, 'r', label='gt_score')\\n pl.plot(np.asarray(range(n_to_show)), sorted_score_by_gt, 'g', label='pred_score')\",\n \"def compute_and_print_eval_metrics(self):\\n s = ('%20s' + '%12s' * 6) % ('Class', 'Images', 'Targets', 'P', 'R', 'mAP@.5', 'mAP@.5:.95')\\n precision, recall, f1, mean_precision, mean_recall, map50, map = 0., 0., 0., 0., 0., 0., 0.\\n ap = []\\n eval_stats = [np.concatenate(x, 0) for x in zip(*self.eval_stats)]\\n if len(eval_stats) and eval_stats[0].any():\\n precision, recall, ap, f1, ap_class = ap_per_class(*eval_stats)\\n precision, recall, ap50, ap = precision[:, 0], recall[:, 0], ap[:, 0], ap.mean(1)\\n mean_precision, mean_recall, map50, map = precision.mean(), recall.mean(), ap50.mean(), ap.mean()\\n nt = np.bincount(eval_stats[3].astype(np.int64), minlength=len(self.class_names)) # number of targets per class\\n else:\\n nt = np.zeros(1)\\n\\n pf = '%20s' + '%12.5g' * 6 # print format\\n print(\\\"\\\\n EVALUTAION \\\\n\\\")\\n print(s)\\n print(pf % ('all', self.seen, nt.sum(), mean_precision, mean_recall, map50, map))\\n if self.cfg.eval.verbose:\\n for indx, cls in enumerate(ap_class):\\n print(pf % (self.class_names[cls], self.seen, nt[cls], precision[indx], recall[indx], ap50[indx], ap[indx]))\",\n \"def ratio_test(fx2_match, fx1_match, fx1_norm, match_dist, norm_dist,\\n ratio_thresh=.625, fm_dtype=np.int32, fs_dtype=np.float32):\\n fx2_to_ratio = np.divide(match_dist, norm_dist).astype(fs_dtype)\\n fx2_to_isvalid = np.less(fx2_to_ratio, ratio_thresh)\\n fx2_match_RAT = fx2_match.compress(fx2_to_isvalid).astype(fm_dtype)\\n fx1_match_RAT = fx1_match.compress(fx2_to_isvalid).astype(fm_dtype)\\n fx1_norm_RAT = fx1_norm.compress(fx2_to_isvalid).astype(fm_dtype)\\n # Turn the ratio into a score\\n fs_RAT = np.subtract(1.0, fx2_to_ratio.compress(fx2_to_isvalid))\\n fm_RAT = np.vstack((fx1_match_RAT, fx2_match_RAT)).T\\n # return normalizer info as well\\n fm_norm_RAT = np.vstack((fx1_norm_RAT, fx2_match_RAT)).T\\n ratio_tup = MatchTup3(fm_RAT, fs_RAT, fm_norm_RAT)\\n return ratio_tup\",\n \"def precision(classifier_output, true_labels):\\n\\n # TODO: finish this.\\n conf_matrix = confusion_matrix(classifier_output, true_labels)\\n return conf_matrix[0][0]/(conf_matrix[0][0] + conf_matrix[1][0])\",\n \"def _update(self, results):\\n logger = getLogger(\\\"problog_lfi\\\")\\n fact_body = defaultdict(int)\\n fact_par = defaultdict(int)\\n\\n score = 0.0\\n for m, pEvidence, result in results:\\n par_marg = dict()\\n for fact, value in result.items():\\n # use the id and the t variables as index\\n index = (fact.args[0], fact.args[1])\\n if fact.functor == \\\"lfi_body\\\":\\n fact_body[index] += value * m\\n elif fact.functor == \\\"lfi_par\\\":\\n if index in par_marg:\\n par_marg[index] += value\\n for o_index in self._adatomc[index[0]]:\\n par_marg[(o_index, *index[1:])] += value\\n else:\\n par_marg[index] = value\\n for o_index in self._adatomc[index[0]]:\\n par_marg[(o_index, *index[1:])] = value\\n\\n for index, value in par_marg.items():\\n fact_par[index] += value * m\\n try:\\n score += math.log(pEvidence)\\n except ValueError:\\n logger.debug(\\\"Pr(evidence) == 0.0\\\")\\n\\n update_list = fact_body\\n\\n weight_changed = []\\n for weight in self._weights:\\n if isinstance(weight, float):\\n weight_changed.append(False)\\n elif isinstance(weight, dict):\\n d = dict()\\n for w in weight:\\n d[w] = False\\n weight_changed.append(d)\\n\\n score = 0.0\\n for index in update_list:\\n if float(fact_body[index]) <= 10**-15:\\n # if close to zero\\n prob = 0.0\\n else:\\n prob = float(fact_body[index]) / float(fact_par[index])\\n try:\\n score += math.log(prob)\\n except ValueError as ex:\\n # prob too close to zero\\n pass\\n\\n logger.debug(\\n \\\"Update probabilistic fact {}: {} / {} = {}\\\".format(\\n index, fact_body[index], fact_par[index], prob\\n )\\n )\\n self._set_weight(index[0], index[1], prob, weight_changed=weight_changed)\\n if not index[1]:\\n weight_changed[int(index[0])] = True\\n elif (\\n isinstance(weight_changed[int(index[0])], dict)\\n and index[1] in weight_changed[int(index[0])]\\n ):\\n weight_changed[int(index[0])][index[1]] = True\\n else:\\n weight_changed[int(index[0])] = {index[1]: True}\\n\\n if self._enable_normalize:\\n self._normalize_weights()\\n\\n return score\",\n \"def evaluate(self, test_data):\\n test_results = [(self.decide(self.feedforward(x)), y)\\n for (x, y) in test_data]\\n\\n # Check for and keep track of TP's FP's and FN's\\n # Write FP's and FN's to a special directories\\n TP = FP = FN = count = 0\\n TN = 0\\n # what you got = x, what should be = y\\n for ((x, y), (image, gt)) in zip(test_results, test_data):\\n count += 1\\n if x == 1 and y == 1:\\n TP += 1\\n elif x == 1 and y == 0:\\n FP += 1\\n if not self.validating: self.save_Image(FP_PATH, count, image)\\n elif x == 0 and y == 1:\\n FN += 1\\n if not self.validating: self.save_Image(FN_PATH, count, image)\\n \\n if (TP + FP ) == 0:\\n return TP, FP, FN, '---------', '---------','---------'\\n else:\\n false_rate = float(FP) / float(TP + FP)\\n\\n if (TP + FN ) == 0:\\n return TP, FP, FN, '---------', '---------', false_rate\\n else:\\n detect_rate = float(TP) / float(TP + FN)\\n\\n if (TP + FP + FN ) == 0:\\n return TP, FP, FN, '---------', detect_rate, false_rate\\n else:\\n quality_rate = float(TP) / float(TP+FP+FN)\\n \\n return TP, FP, FN, quality_rate, detect_rate, false_rate\",\n \"def calculate_scores(self):\\n # Prediction based scores\\n #self.report = classification_report(self.y_test, self.y_pred)\\n self.accuracy = accuracy_score(self.y_real, self.y_pred)\\n self.precision = precision_score(self.y_real, self.y_pred)\\n self.recall = recall_score(self.y_real, self.y_pred)\\n self.f1 = f1_score(self.y_real, self.y_pred)\\n \\n # Probability based scores\\n self.fpr, self.tpr, _ = roc_curve(self.y_real, self.y_proba)\\n self.average_precision = average_precision_score(self.y_real, self.y_proba)\\n self.brier_loss = brier_score_loss(self.y_real, self.y_proba)\\n self.roc_auc = roc_auc_score(self.y_real, self.y_proba)\\n self.prec_cur, self.recall_cur, _ = precision_recall_curve(self.y_real, self.y_proba)\",\n \"def update_score_other_rounds(self, match):\\n self.match_views.update_score(match)\",\n \"def update_predictions(self, context):\\n x, y, o = context.get_predictions()\\n self.x_eval += x\\n self.y_eval += y\\n self.o_eval += o\\n self.write_predictions(o)\",\n \"def evaluate(gold, predictions):\\n # gold = {i[\\\"id\\\"]:i[\\\"label\\\"] for i in read(goldfile)}\\n # sys = {i[\\\"id\\\"]:i[\\\"prediction\\\"] for i in read(sysfile)}\\n\\n labels, preds = [], []\\n for idx in gold:\\n labels.append(gold[idx])\\n preds.append(predictions[idx])\\n\\n precision_macro, recall_macro, f1_macro, _ = precision_recall_fscore_support(labels, preds, labels=[0, 1],\\n average=\\\"macro\\\")\\n print(\\\"F1-macro score for test data predictions are: %.4f\\\" % f1_macro)\\n return preds\",\n \"def compare(model, input1, input2): \\n patch_image_time_start = time.time()\\n input1_patches = get_patches_non_overlap(input1, 48, 48)\\n input2_patches = get_patches_non_overlap(input2, 48, 48)\\n patch_image_time_end = time.time()\\n compare_image_time_start = time.time()\\n pred = model.predict([input1_patches, input2_patches])\\n compare_image_time_end = time.time()\\n\\n result[\\\"patch_retrieval_time\\\"].append(patch_image_time_end - patch_image_time_start)\\n result[\\\"image_comparison_time\\\"].append(compare_image_time_end - compare_image_time_start)\\n\\n return np.sum(pred)\",\n \"def computeMatchInfo( df, inpCol, predCol, true_label=\\\"label\\\", method=\\\"---\\\", seuil=0. ) :\\n df_data = predict( df, inpCol, predCol, seuil=seuil )\\n df_ca = computeClassAtt( df_data[true_label], df_data[predCol], method )\\n df_ca[\\\"threshold\\\"] = seuil\\n\\n return df_data, df_ca\",\n \"def evaluate(self):\\n tic = time.time()\\n print('Running per image evaluation...')\\n p = self.params\\n # add backward compatibility if useSegm is specified in params\\n if p.useSegm is not None:\\n p.iouType = 'segm' if p.useSegm == 1 else 'bbox'\\n print(\\n \\\"useSegm (deprecated) is not None.\\\"\\n \\\"Running {} evaluation\\\".format(p.iouType)\\n )\\n print('Evaluate annotation type *{}*'.format(p.iouType))\\n p.imgIds = list(np.unique(p.imgIds))\\n if p.useCats:\\n p.catIds = list(np.unique(p.catIds))\\n p.maxDets = sorted(p.maxDets)\\n self.params = p\\n\\n self._prepare()\\n # loop through images, area range, max detection number\\n catIds = p.catIds if p.useCats else [-1]\\n\\n self.ious = {\\n (imgId, catId): self.computeIoU(imgId, catId)\\n for imgId in p.imgIds\\n for catId in catIds\\n }\\n # ignore any categories that is not having any attributes\\n self.f1s = {\\n (imgId, catId): self.computeF1(imgId, catId)\\n for imgId in p.imgIds\\n for catId in catIds\\n if catId in self.FPParams.catsWithAttributes\\n }\\n\\n # self.gt_attributes_ids = []\\n\\n # loop through images, area range, max detection number\\n self.evalImgs = [\\n self.evaluateImg(imgId, catId, areaRng, p.maxDets[-1])\\n for catId in catIds\\n for areaRng in p.areaRng\\n for imgId in p.imgIds\\n ]\\n # self._paramsEval = copy.deepcopy(self.params) # seems do not need it\\n toc = time.time()\\n print('DONE (t={:0.2f}s).'.format(toc-tic))\",\n \"def calculate_many_precs_recall_stats(test_class=test_class,\\n local_list=local_list,\\n Q_stop='',\\n no_files_in_label=no_files_in_label,\\n no_of_images=no_of_images,\\n verbose=verbose):\\n if not Q_stop == '':\\n Q = Q_stop\\n else:\\n Q = len(local_list)\\n # does all points\\n if Q > len(local_list):\\n Q = len(local_list)\\n if verbose:\\n print('The number of points to check (Q) {} is more than the number of points above minx {}'.format(Q, len(local_list)))\\n print('Setting Q to len(local_list), you may wish to change your settings')\\n N_test = no_files_in_label[test_class] # no of items in class A\\n N_not_test = no_of_images - N_test # i.e. set not-A\\n # set up counters\\n AP = 0 # average precision\\n found_recall_precs_p95 = False\\n recall_p95 = 0.0\\n count_of_test_class = 0\\n # loop backwards through the list, abs j is the position in a 1-indexed list\\n recall_x = 0 # n.b. this is also sensitivity\\n Ave_precs_x = 0\\n specificity_x = 0\\n count_of_false_positives = 0 # number of not-A on misidentified\\n max_informedness = 0\\n x_for_max_informedness = 0\\n max_f1_stat = 0\\n x_for_max_f1 = 0\\n recall_for_max_informedness = 0\\n specificity_for_max_informedness = 0\\n for i in range(Q):\\n j = -(i + 1) # 1 indexed\\n recall_x_minus_1 = recall_x\\n current_class = local_list[j][0]\\n #j = j - 1 # really this is here so we can check j\\n # break\\n if count_of_test_class == N_test:\\n # we've found them all\\n if verbose:\\n print('found all {} of {}, stopping...'.format(N_test, current_class))\\n print('{}/{}'.format(count_of_test_class, abs(j)))\\n break\\n if (current_class == test_class):\\n count_of_test_class = count_of_test_class + 1 # n A\\n else:\\n count_of_false_positives = count_of_false_positives + 1\\n precs_x = count_of_test_class / (abs(j)) # N.b. this is the sum, we divide by j on the output\\n recall_x = count_of_test_class / N_test\\n false_pos_rate = count_of_false_positives / no_of_images\\n specificity_x = 1-false_pos_rate\\n if (precs_x <= .95) and (found_recall_precs_p95 is False):\\n # thingy to grab the recall when precision drops below .95 (if ever)\\n recall_p95 = recall_x\\n found_recall_precs_p95 = True\\n delta_recall_x = recall_x - recall_x_minus_1 # difference in recall between this point nd the next\\n weight_precs_x = precs_x * delta_recall_x # weighted precsion at point x (we do average via weighted sum)\\n Ave_precs_x = Ave_precs_x + weight_precs_x # average_precision evaluated at point x\\n informedness_x = recall_x + specificity_x -1\\n if informedness_x > max_informedness:\\n max_informedness = informedness_x\\n x_for_max_informedness = abs(j)\\n recall_for_max_informedness = recall_x\\n specificity_for_max_informedness = specificity_x\\n if (precs_x > 0 and recall_x > 0):\\n f1_x = 2*(precs_x*recall_x) / (precs_x + recall_x)\\n else:\\n f1_x = 0\\n if f1_x > max_f1_stat:\\n max_f1_stat = f1_x\\n x_for_max_f1 = abs(j)\\n if verbose:\\n print('Class {}, evaluated at Q={}:, found {}'.format(test_class, Q, count_of_test_class))\\n print('Average precision={}; precision={}; recall={}'.format(Ave_precs_x,precs_x,recall_x))\\n print('Recall at precision .95 threashold is {}'.format(recall_p95))\\n print('specificity = {}'.format(specificity_x))\\n print('informedness = {}'.format(informedness_x))\\n print('Max informedness was {}, seen at x {}'.format(max_informedness, x_for_max_informedness))\\n print('F1 at x={} is {}'.format(abs(j),f1_x))\\n print('Max F1 was {}, seen at x = {}'.format(max_f1_stat,x_for_max_f1))\\n out = (Ave_precs_x, precs_x, recall_x, recall_p95,\\n specificity_x, informedness_x, max_informedness, x_for_max_informedness,\\n max_f1_stat, x_for_max_f1,\\n recall_for_max_informedness, specificity_for_max_informedness,)\\n return out\",\n \"def statistics_on_test(self, predicted_results, result):\\n # Print confusion matrix and mean average precision score\\n predicted_results_binary = self.predicted_results_to_binary(predicted_results)\\n print(\\\"\\\\nConfusion matrix : \\\")\\n print(confusion_matrix(result, predicted_results_binary))\\n print(\\\"\\\\nAverage precision score : \\\", average_precision_score(result, predicted_results_binary))\",\n \"def matching_accuracy(pmat_pred, pmat_gt, ns):\\n device = pmat_pred.device\\n batch_num = pmat_pred.shape[0]\\n\\n pmat_gt = pmat_gt.to(device)\\n\\n assert torch.all((pmat_pred == 0) + (pmat_pred == 1)), 'pmat_pred can noly contain 0/1 elements.'\\n assert torch.all((pmat_gt == 0) + (pmat_gt == 1)), 'pmat_gt should noly contain 0/1 elements.'\\n assert torch.all(torch.sum(pmat_gt, dim=-1) <= 1) and torch.all(torch.sum(pmat_gt, dim=-2) <= 1)\\n assert torch.all(torch.sum(pmat_pred, dim=-1) <= 1) and torch.all(torch.sum(pmat_pred, dim=-2) <= 1)\\n\\n #indices_pred = torch.argmax(pmat_pred, dim=-1)\\n #indices_gt = torch.argmax(pmat_gt, dim=-1)\\n\\n #matched = (indices_gt == indices_pred).type(pmat_pred.dtype)\\n match_num_list = []\\n gt_num_list = []\\n pred_num_list = []\\n acc_gt = []\\n acc_pred = []\\n for b in range(batch_num): #acc_gt,acc_pred\\n #match_num += torch.sum(matched[b, :ns[b]])\\n #total_num += ns[b].item()\\n match_num = torch.sum(pmat_pred[b, :] * pmat_gt[b, :]) + 1e-8 #:ns[b]\\n gt_num = torch.sum(pmat_gt[b, :]) + 1e-8\\n pred_num = torch.sum(pmat_pred[b,:]) + 1e-8\\n match_num_list.append(match_num.cpu().numpy())\\n gt_num_list.append(gt_num.cpu().numpy())\\n pred_num_list.append(pred_num.cpu().numpy())\\n acc_gt.append((match_num/gt_num).cpu().numpy())\\n acc_pred.append((match_num/pred_num).cpu().numpy())\\n\\n return {'acc_gt': np.array(acc_gt),\\n 'acc_pred': np.array(acc_pred),\\n 'match_num': np.array(match_num_list),\\n 'gt_num': np.array(gt_num_list),\\n 'pred_num': np.array(pred_num_list)}\",\n \"def print_results(precision, recall, f1):\\n print \\\"Precision:\\\", precision\\n print \\\"Recall:\\\", recall\\n print \\\"F1 score:\\\", f1\",\n \"def scores(self, y, y_pred):\\n\\n aucroc = 0.\\n precision = 0.\\n recall = 0.\\n f1 = 0.\\n aucroc_labs = np.zeros(self.datas[self.train_idx].n_labels)\\n precision_labs = np.zeros(self.datas[self.train_idx].n_labels)\\n recall_labs = np.zeros(self.datas[self.train_idx].n_labels)\\n f1_labs = np.zeros(self.datas[self.train_idx].n_labels)\\n label_ratios = np.mean(y, axis=0)\\n\\n if len(y) > 1:\\n y_t = np.transpose(y)\\n col_keep = np.ones(len(y_t), dtype=bool)\\n for i, col_y in enumerate(y_t):\\n if 0 not in col_y or 1 not in col_y:\\n col_keep[i] = False\\n\\n if sum(col_keep) > 0:\\n if not col_keep.all():\\n y = np.transpose(y_t[col_keep])\\n y_pred = np.transpose(np.transpose(y_pred)[col_keep])\\n\\n f1 = f1_score(y, self._round(y_pred), average=self.metrics_avg)\\n s = f1_score(y, self._round(y_pred), average=None)\\n f1_labs[col_keep] = s if sum(col_keep) > 1 else s[1]\\n aucroc = roc_auc_score(y, y_pred, average=self.metrics_avg)\\n aucroc_labs[col_keep] = roc_auc_score(y, y_pred, average=None)\\n precision = precision_score(y, self._round(y_pred), average=self.metrics_avg)\\n recall = recall_score(y, self._round(y_pred), average=self.metrics_avg)\\n if sum(col_keep) > 1:\\n precision_labs[col_keep] = precision_score(y, self._round(y_pred), average=None)\\n recall_labs[col_keep] = recall_score(y, self._round(y_pred), average=None)\\n else:\\n precision_labs[col_keep] = precision_score(y, self._round(y_pred))\\n recall_labs[col_keep] = recall_score(y, self._round(y_pred))\\n elif self.verbose:\\n print('*Cannot compute other metrics because no label in Truth has alternatives, only precision*')\\n precision = precision_score(y, self._round(y_pred), average=self.metrics_avg)\\n precision_labs = precision_score(y, self._round(y_pred), average=None)\\n\\n elif len(y) == 1:\\n if self.verbose:\\n print('*Cannot compute other metrics with %d samples, only precision*' % len(y))\\n precision = precision_score(y, self._round(y_pred), average=self.metrics_avg)\\n precision_labs = precision_score(y, self._round(y_pred), average=None)\\n\\n result = {\\n 'aucroc': aucroc,\\n 'precision': precision,\\n 'recall': recall,\\n 'f1': f1,\\n 'aucroc_labs': aucroc_labs,\\n 'precision_labs': precision_labs,\\n 'recall_labs': recall_labs,\\n 'f1_labs': f1_labs,\\n 'label_ratios': label_ratios\\n }\\n\\n return result\",\n \"def update_metrics(self, metrics, predictions, labels):\\n return\",\n \"def compute_precision(data, num_labels = 4): \\n \\n # Declarating list to store results\\n precisions = []\\n \\n for instance in data:\\n \\n # Declarating list to store individual results\\n instance_precisions = []\\n \\n for i in np.arange(num_labels):\\n \\n # Computing and storing precision for each class\\n instance_precisions.append(precision_score(instance[:, 2 + i], instance[:, 2 + i + 4]))\\n \\n # Storing mean results of the instance\\n precisions.append(np.mean(instance_precisions))\\n \\n # Returning mean of all results\\n return np.mean(precisions)\",\n \"def patternMatchingTest(matchingAlgorithm,\\n observations= defaultCOMPREFdump , \\n modelsFolder= defaultWRFdumpsFolder,\\n #key1='0828',\\n key1='',\\n key2='',\\n #key3='',\\n verbose=True,\\n makeSmallImages=True,\\n smallDpi=40,\\n frameDpi = 600 , # added 2013-09-23\\n loadFormerAnalysisResult=False, #added 2013-09-23\\n #summary_folder=summary_folder, # needed if loadFormerAnalysisResult is True\\n summary_folder=summary_folder+timeString + '/', # 2013-09-27\\n matching_algorithm_name=\\\"\\\", # needed if loadFormerAnalysisResult is True\\n panel_cols = 5,\\n panel_rows = 5,\\n useCV2 = False, # switch off opencv since it doesn't work yet\\n useCV = False,\\n averageScoreOrdering=True, # take an average of the ordering to get a total score for the entire model over the time series\\n #kwargs = {}, # keyword arguments for the matchinging algorithm, just in case\\n **kwargs\\n ):\\n if key1 != '' or key2 != '':\\n summary_folder += '_'.join([key1, key2])\\n if not averageScoreOrdering:\\n summary_folder += '_nonaver'\\n summary_folder += '%dx%d/' % (panel_rows, panel_cols)\\n\\n t0 = time.time()\\n if loadFormerAnalysisResult:\\n mv_folder = summary_folder + matching_algorithm_name + '/'\\n result = pickle.load(open(mv_folder + 'result.pydump', 'r'))\\n\\n \\\"\\\"\\\"\\n result= {'best matches' : bestMatchAtEachTime, \\n 'best overall match' : bestMatchOverall, \\n 'all scores' : scores,\\n 'matching algorithm' : matching_algorithm_name,\\n 'dataset1[obs]' : observations,\\n 'dataset2[models]' : modelsFolder,\\n 'time spent' : time.time() - t0,\\n 'mv folder' : mv_folder, #added 2013-09-23\\n 'ordering' : ordering, #added 2013-09-23\\n }\\n \\\"\\\"\\\"\\n bestMatchAtEachTime = result['best matches'] \\n bestMatchOverall = result['best overall match']\\n scores = result['all scores']\\n matching_algorithm_name = result['matching algorithm']\\n observations = result['dataset1[obs]']\\n modelsFolder = result['dataset2[models]']\\n #mv_folder = result['mv folder']\\n ordering = result['ordering']\\n\\n # THE FOLLOWING 30 LINES ARE COPIED FROM \\\"ELSE...\\\" BELOW\\n ########### ******\\n ##\\n #\\n ds1 = pickle.load(open(observations,'r'))\\n ds1.setFloor(0) # added 2013-09-27\\n ds1.list = [v for v in ds1.list if (key1 in v.dataTime or key1 in v.name) \\\\\\n and (key2 in v.dataTime or key2 in v.name)]\\n ds1.list.sort(key=lambda v:v.dataTime)\\n ds1.setVmin(-40) #line added 2013-09-23\\n # debug\\n #print ds1.list\\n print 'loading list with keywords [ %s %s ]' %(key1, key2) \\n ds1.load(verbose=verbose)\\n # end debug\\n modelsList = [v for v in os.listdir(modelsFolder) if '.pydump' in v]\\n modelsList.sort(key = lambda v: int(re.findall(r'\\\\d+',v)[0]))\\n print \\\"modelsList\\\", modelsList\\n scores = {} # scores = {model number: [('dataTime', score), ('dataTime', score),...], ...}\\n print 'loading ds2 (model data)'\\n wrfs = [] # line added 2013-09-23\\n for path in modelsList:\\n modelIndex = int(re.findall(r'\\\\d+', path)[0])\\n ds2 = pickle.load(open(modelsFolder+path, 'r'))\\n ds2.setFloor(0) # added 2013-09-27\\n # debug\\n #ds2.list = [v for v in ds2.list if key1 in v.dataTime] # line redundant, 2013-09-23\\n ds1.list, ds2.list = ds1.intersect(ds2) #line added 2013-09-23 \\n ds2.setVmin(-40) #line added 2013-09-23\\n wrfs.append(ds2) # line added 2013-09-23\\n # end debug\\n\\n ####################\\n\\n\\n else:\\n # THE FOLLOWING 30 LINES ARE COPIED TO \\\"IF...\\\" ABOVE\\n ########### ******\\n ##\\n #\\n ds1 = pickle.load(open(observations,'r'))\\n ds1.list = [v for v in ds1.list if (key1 in v.dataTime or key1 in v.name) \\\\\\n and (key2 in v.dataTime or key2 in v.name)]\\n ds1.list.sort(key=lambda v:v.dataTime)\\n ds1.setVmin(-40) #line added 2013-09-23\\n # debug\\n #print ds1.list\\n print 'loading list with keywords [ %s %s ]' %(key1, key2) \\n ds1.load(verbose=verbose)\\n ds1.setFloor(0) # added 2013-09-27\\n # end debug\\n modelsList = [v for v in os.listdir(modelsFolder) if '.pydump' in v]\\n modelsList.sort(key = lambda v: int(re.findall(r'\\\\d+',v)[0]))\\n print \\\"modelsList\\\", modelsList\\n scores = {} # scores = {model number: [('dataTime', score), ('dataTime', score),...], ...}\\n print 'loading ds2 (model data)'\\n wrfs = [] # line added 2013-09-23\\n ds1.commonMaskSet=False # tempory attribute added 2013-09-27\\n for path in modelsList:\\n modelIndex = int(re.findall(r'\\\\d+', path)[0])\\n ds2 = pickle.load(open(modelsFolder+path, 'r'))\\n # debug\\n #ds2.list = [v for v in ds2.list if key1 in v.dataTime] # line redundant, 2013-09-23\\n ds1.list, ds2.list = ds1.intersect(ds2) #line added 2013-09-23 \\n ds2.setFloor(0) # added 2013-09-27\\n ds2[0].matrix.mask += ds1[0].matrix.mask # added 2013-09-27\\n ds2.setCommonMask() # added 2013-09-27\\n if not ds1.commonMaskSet: # added 2013-09-27\\n ds1[0].matrix.mask +=ds2[0].matrix.mask # added 2013-09-27\\n ds1.setCommonMask() # added 2013-09-27\\n ds2.setVmin(-40) #line added 2013-09-23\\n wrfs.append(ds2) # line added 2013-09-23\\n # end debug\\n\\n #######################################################################################################################\\n # key lines\\n print 'computing scores between %s and %s ...' %(ds1.name, ds2.name)\\n s, matching_algorithm_name = matchingAlgorithm(ds1, ds2, **kwargs) # list of the form [('20120612.0000', 0.9),...] \\n scores[modelIndex] = dict(s) # { model index: {'dataTime': score,...}, ...}\\n # debug\\n if verbose:\\n print scores\\n print '..................................................'\\n #\\n ########################################################################################################################\\n\\n print 'sorting the results...'\\n #print 'scores', '\\\\n', scores #debug\\n # bestMatchAtEachTime\\n T1 = [v.dataTime for v in ds1]\\n T2 = [v.dataTime for v in ds2] #any ds2 will do - they are the same, in this case we pick the last one \\n T = set(T1).intersection(T2)\\n dataTimeList = sorted(list(T))\\n\\n #print 'dataTimeList:', dataTimeList #debug\\n highestScores = []\\n ordering = [] #added 2013-09-23\\n\\n if averageScoreOrdering: # take the average scores before ordering to make it uniform\\n ave_scores = [ (v, np.mean( [scores[v].values() ] ) ) for v in scores.keys()]\\n ave_scores = dict(ave_scores)\\n print '..taking averages...............................'\\n print ave_scores\\n for u in scores.keys():\\n for t in scores[u].keys():\\n scores[u][t] = ave_scores[u]\\n print '\\\\n'.join([str(scores[v]) for v in scores.keys()])\\n\\n for t in dataTimeList:\\n print t\\n highestScoreAt_t = max([scores[u][t] for u in scores.keys()]) # u = model index (integer 1-20), t = dataTime (string)\\n new_entry = (t, highestScoreAt_t, [u for u in scores.keys() if scores[u][t]==highestScoreAt_t])\\n print new_entry\\n highestScores.append(new_entry)\\n #added 2013-09-23\\n ordering.append([0]+sorted(range(1, len(modelsList)+1), key=lambda v: scores[v][t], reverse=True))\\n\\n\\n\\n\\n # bestMatchOverall\\n print scores # debug\\n averageScores = [(Model, np.mean(scores[Model].values())) for Model in scores.keys()] # list of [(Model, mean score),...]\\n highestScore = max([v[1] for v in averageScores])\\n bestMatch = [u[0] for u in averageScores if u[1]==highestScore] \\n # debug\\n print 'average scores , highest average, best match', averageScores, highestScore, bestMatch\\n #xx= raw_input('press enter') \\n\\n\\n bestMatchAtEachTime = highestScores\\n bestMatchOverall = (highestScore, bestMatch)\\n\\n # the following lines do not work well since oftentimes we have multiple pointers (names) to a single object/function\\n # and so they are replaced by returning the function name in the function itself\\n #allvars = vars()\\n #matching_algorithm_name = [v for v in allvars if allvars[v]==matchingAlgorithm][0]\\n #mv_folder = summary_folder + matching_algorithm_name + '/'\\n mv_folder = summary_folder + matching_algorithm_name + '/' # added 2013-09-27\\n\\n result= {'best matches' : bestMatchAtEachTime, \\n 'best overall match' : bestMatchOverall, \\n 'all scores' : scores,\\n 'matching algorithm' : matching_algorithm_name,\\n 'dataset1[obs]' : observations,\\n 'dataset2[models]' : modelsFolder,\\n 'time spent' : time.time() - t0,\\n 'mv folder' : mv_folder, #added 2013-09-23\\n 'ordering' : ordering, #added 2013-09-23\\n }\\n\\n try:\\n os.makedirs(mv_folder)\\n print 'folder', mv_folder, \\\"created!\\\"\\n except:\\n print 'folder', mv_folder, 'exists!'\\n pickle.dump(result, open(mv_folder+'result.pydump','w'))\\n\\n ############################################################################\\n # to make the videos/collected images\\n # calling armor.dataStreamTools.makeVideo.makeVideo()\\n # 2013-09-23\\n\\n from . import makeVideo as mv\\n \\n try:\\n os.makedirs(mv_folder)\\n print 'folder', mv_folder, \\\"created!\\\"\\n except:\\n print 'folder', mv_folder, 'exists!'\\n # make images\\n if makeSmallImages:\\n print 'making small images'\\n for ds in [ds1] + wrfs:\\n #ds.imageFolder= mv_folder + ds.name + '/'\\n ds.imageFolder= summary_root + ds.name + '/' #added 2013-09-27 to replace the above line - too tired waiting the plots\\n ds.saveImages(drawCoast=True, flipud=True, dpi=smallDpi)\\n print ds.imageFolder\\n else:\\n # just set the paths\\n for ds in [ds1]+wrfs:\\n ds.imageFolder = summary_root + ds.name +'/' # /home/k/ARMOR/data/KONG-REY/summary/WRF01.DBZ/\\n for dbzPattern in ds.list:\\n dbzPattern.imagePath = ds.imageFolder + ds.name + dbzPattern.dataTime + '.png'\\n print dbzPattern.imagePath\\n #ds.setImagePaths() # doesn't for for old pickled objects???\\n\\n\\n # make panels\\n\\n mv.makeVideo(DSS = [ds1]+ wrfs, # [ds0, ds1, ds2, ds3, ds4, ...], a list of armor.pattern.DBZstream objects\\n panel_cols = panel_cols, # number of colums in the panel\\n panel_rows = panel_rows, # no need to be filled\\n #fourcc = cv.CV_FOURCC('F', 'L', 'V', '1'),\\n #fps = defaultFps,\\n extension= '.avi',\\n #fourcc = cv.CV_FOURCC('P', 'I', 'M', '1'),\\n outputFileName =\\\"\\\",\\n outputFolder= mv_folder,\\n sandbox = mv_folder ,\\n saveFrames = True, # saving the frames as images\\n frameDpi = frameDpi,\\n useCV2 = useCV2,\\n useCV = useCV,\\n ordering = ordering, # ordering of the models\\n )\\n ############################################################################\\n # dumping the results\\n try:\\n os.makedirs(summary_folder+ 'pydumps/')\\n except:\\n print 'Folder exists! -- ', summary_folder+ 'pydumps/' \\n\\n pickle.dump(result, open(summary_folder+ 'pydumps/results_' + matching_algorithm_name+ '+' +\\\\\\n key1 + '+' + key2 + str(int(time.time())) + '.pydump', 'w'))\\n\\n \\n\\n print \\\"time spent:\\\", time.time()-t0\\n return result\",\n \"def evaluate(test_set, predictions):\\n full_report = metrics.classification_report(test_set.labels, predictions,\\n labels=range(len(test_set.index2label)),\\n target_names=test_set.index2label, digits=3)\\n pre, rec, f1, support = metrics.precision_recall_fscore_support(test_set.labels, predictions, average='weighted')\\n return pre, rec, f1, support, full_report\",\n \"def calculate_metrics(predictions, expected):\\n # type: (np.ndarray, np.ndarray) -> (float, float, float)\\n clients_count = predictions.shape[0]\\n products_count = predictions.shape[1]\\n\\n true_positive = 0.0\\n true_negative = 0.0\\n false_positive = 0.0\\n false_negative = 0.0\\n\\n total = float(clients_count * products_count)\\n\\n for c in range(0, clients_count):\\n for p in range(0, products_count):\\n if predictions[c, p] == expected[c, p]:\\n if predictions[c, p] == 1:\\n true_positive += 1\\n else:\\n true_negative += 1\\n else:\\n if predictions[c, p] == 1:\\n false_positive += 1\\n else:\\n false_negative += 1\\n\\n accuracy = float(true_positive + true_negative) / total\\n if true_positive + false_positive == 0:\\n precision = 0\\n else:\\n precision = true_positive / float(true_positive + false_positive)\\n\\n if true_positive + false_negative == 0:\\n recall = 0\\n else:\\n recall = true_positive / float(true_positive + false_negative)\\n\\n return accuracy, precision, recall\",\n \"def evaluation(eval_dict, faster_rcnn, test_num=Config.eval_num):\\n bboxes, labels, scores = list(), list(), list()\\n gt_bboxes, gt_labels, gt_difficult = list(), list(), list()\\n for i, [img_dir, img_info] in tqdm(enumerate(eval_dict.items())):\\n img, img_info = rescale_image(img_dir, img_info, flip=False)\\n img_tensor = create_img_tensor(img)\\n box, score, label = faster_rcnn.predict(img_tensor)\\n gt_bbox = np.array(img_info['objects'])[:, 1:5].astype(np.float32)\\n gt_label = np.array(img_info['objects'])[:, 0]\\n difficult = np.array(img_info['difficult'])\\n\\n # label from text to number\\n gt_label = text_to_num(gt_label)\\n\\n bboxes.append(box)\\n labels.append(label)\\n scores.append(score)\\n gt_bboxes.append(gt_bbox)\\n gt_labels.append(gt_label)\\n gt_difficult.append(difficult)\\n if i == test_num:\\n break\\n\\n AP, mAP = calc_map(bboxes, labels, scores, gt_bboxes, gt_labels, gt_difficult, use_07_metric=True)\\n\\n return mAP\",\n \"def __call__(self, results):\\n for key in results.get('img_fields', ['img']):\\n results[key] = general_ocr.imnormalize(results[key], self.mean, self.std,\\n self.to_rgb)\\n results['img_norm_cfg'] = dict(\\n mean=self.mean, std=self.std, to_rgb=self.to_rgb)\\n return results\",\n \"def precision_recall(y_true,\\n probas_pred,\\n num_positives):\\n probas_pred = probas_pred.flatten()\\n y_true = y_true.flatten()\\n # to handle duplicates scores, we sort (score, NOT(jugement)) for predictions\\n # eg,the final order will be (0.5, False), (0.5, False), (0.5, True),\\n # (0.4, False), ...\\n # This allows to have the worst possible AP.\\n # It prevents participants from putting the same score for all predictions\\n # to get a good AP.\\n order = argsort(list(zip(probas_pred, ~y_true)))\\n order = order[::-1] # sort by decreasing score\\n probas_pred = probas_pred[order]\\n y_true = y_true[order]\\n\\n ntp = np.cumsum(y_true) # number of true positives <= threshold\\n nres = np.arange(len(y_true)) + 1 # number of results\\n\\n precisions = ntp / nres\\n recalls = ntp / num_positives\\n return precisions, recalls, probas_pred\",\n \"def scoring(self):\\n pass\",\n \"def precision(ground_truth, prediction):\\n ground_truth = remove_duplicates(ground_truth)\\n prediction = remove_duplicates(prediction)\\n precision_score = count_a_in_b_unique(prediction, ground_truth) / float(len(prediction))\\n assert 0 <= precision_score <= 1\\n return precision_score\",\n \"def update_predictions(data):\\n # TODO: Priority 1 - update predictions with inference results\\n # TODO: Understand from a research team exactly what the data is going to look like\\n trackID = data[0]\\n prediction = data[1]\\n confidence = data[2]\\n to_Insert_Array = [trackID, prediction, confidence]\\n OutPut_Data[trackID] = to_Insert_Array\",\n \"def __train__(self):\\n if (self.type_camf == 'CAMF_CI'):\\n #users, items, context, ratings\\n ci = camf_ci.CI_class(self.__users_array__, self.__items_array__, self.__context_array__, self.__ratings__, self.fold, self.lr, self.factors)\\n predictions, losses = ci.fit()\\n elif (self.type_camf == 'CAMF_CU'):\\n cu = camf_cu.CU_class(self.__users_array__, self.__items_array__, self.__context_array__, self.__ratings__, self.fold, self.lr, self.factors)\\n predictions, losses = cu.fit()\\n elif (self.type_camf == 'CAMF_C'):\\n c = camf_c.C_class(self.__users_array__, self.__items_array__, self.__context_array__, self.__ratings__, self.fold, self.lr, self.factors)\\n predictions, losses = c.fit()\\n\\n dummy_pred = np.zeros((predictions.shape))\\n for r, pred_array in enumerate(predictions):\\n for c, pred in enumerate(pred_array):\\n dummy_pred[r][c] = self.__check_ratings__(pred)\\n predictions = dummy_pred\\n #save a plot with a loss function\\n plots = prs.PlotRSData()\\n #print(losses)\\n plots.plot_loss_cars(losses, self.type_camf, self.__save_prefix__+\\\"_loop\\\"+str(self.loop))\\n pd.DataFrame(losses).to_csv(\\\"./RecSys/out/CAMF/train/\\\"+self.type_camf+\\\"/\\\" + self.__save_prefix__ +\\\"losses_loop\\\"+str(self.loop)+\\\".csv\\\")\\n print('Saving the feature matrix...')\\n # set predictions back to the pivot table\\n self.__utility_saved_training__(predictions) \\n # save results\\n self.utility_predictions.to_csv(\\\"./RecSys/out/CAMF/train/\\\"+self.type_camf+\\\"/\\\" + self.__save_prefix__ + \\\"_SGD_predictions_loop\\\"+str(self.loop)+\\\".csv\\\")\",\n \"def __call__(self, results):\\n\\n for key in results.get('seg_fields', []):\\n if self.scale_factor != 1:\\n results[key] = general_ocr.imrescale(\\n results[key],\\n self.scale_factor,\\n interpolation='nearest',\\n backend=self.backend)\\n return results\",\n \"def update_recall_rate(params, name, recall, k):\\n if name in params:\\n for key in params[name]:\\n params[name][key] += recall[key] / k\\n else:\\n params[name] = {}\\n for key in recall:\\n params[name][key] = recall[key] / k\",\n \"def metric_test(self):\\n k = 10\\n latent_factor = 10\\n n_users = 10\\n n_items = 12\\n\\n interactions, user_features, item_features = util.generate_dummy_data_with_indicator (num_users=n_users, num_items=n_items, interaction_density=.5)\\n print (\\\"interactiosn shape={}\\\".format( np.shape(interactions) ))\\n print (\\\"user features shape={}\\\".format( np.shape(user_features.toarray()) ))\\n print (\\\"item features shape={}\\\".format( np.shape(item_features.toarray()) ))\\n\\n model = TensorRec(n_components=latent_factor)\\n\\n model.fit(interactions, user_features, item_features, epochs=19)\\n\\n ranks = model.predict_rank(user_features=user_features, item_features=item_features)\\n\\n print (\\\"Ranks shape={}\\\".format(np.shape(ranks)))\\n\\n self.assertTrue(np.shape(interactions) == np.shape(ranks))\\n\\n tr_recall_result = eval.recall_at_k(predicted_ranks=ranks, test_interactions=interactions, k=k, preserve_rows=False)\\n # print (tr_recall_result.mean())\\n\\n tr_precision_result = eval.precision_at_k(predicted_ranks=ranks, test_interactions=interactions, k=k, preserve_rows=False)\\n # print(tr_precision_result.mean())\\n\\n # we need csr for interactions data\\n interactions_ = interactions.tocsr()\\n recall_result = metrics.recall_at_k(ranks, interactions_, k=k, preserve_rows=False)\\n # print(recall_result.mean())\\n\\n precision_result = metrics.precision_at_k(ranks, interactions_, k=k, preserve_rows=False)\\n # print (precision_result.mean())\\n\\n self.assertTrue (tr_recall_result.mean() == recall_result.mean())\\n self.assertTrue (tr_precision_result.mean() == precision_result.mean())\",\n \"def __init__(self):\\n\\n try:\\n self.posFreqDict = self.load(\\\"posFreqDict\\\")\\n self.neuFreqDict = self.load(\\\"neuFreqDict\\\")\\n self.negFreqDict = self.load(\\\"negFreqDict\\\")\\n\\n except:\\n\\n lFileList = []\\n for fFileObj in os.walk(\\\"movies_reviews/\\\"): \\n lFileList = fFileObj[2]\\n break\\n random.shuffle(lFileList)\\n \\n self.train(lFileList)\\n Original_d = {}\\n d = {}\\n correct_positives =0\\n incorrect_positives=0\\n correct_negatives =0\\n incorrect_negatives = 0\\n for file in lFileList[int(float(len(lFileList))*.9):len(lFileList)+1]:\\n rating = int(file[7])\\n if rating in Original_d:\\n Original_d[rating] = Original_d[rating]+1\\n else:\\n Original_d[rating]=1\\n\\n fileText = self.loadFile(\\\"movies_reviews/\\\" + str(file))\\n result = self.classify(fileText)\\n if result in d:\\n d[result]=d[result]+1\\n else:\\n d[result]=1\\n if rating==5:\\n if result ==\\\"positive\\\":\\n correct_positives+=1\\n else:\\n incorrect_positives+=1\\n else:\\n if result == \\\"negative\\\":\\n correct_negatives+=1\\n else:\\n incorrect_negatives+=1\\n print Original_d\\n print d\\n print correct_positives #Actually positive, we said positive\\n print incorrect_positives #Actually positive, we said negative\\n print correct_negatives #Actually negative, we said negative\\n print incorrect_negatives #Actually negative, we said positive\\n \\n A = float(correct_positives)\\n B = float(incorrect_positives)\\n C = float(incorrect_negatives)\\n\\n Recall = (A / (A + B)) *100.0\\n Precision = (A / (A + C)) *100.0\\n FScore = 2 * (Recall*Precision) / (Recall+Precision)\\n\\n print Recall\\n print Precision\\n print FScore\",\n \"def _adjust_brightness_img(self, results, factor=1.0):\\n for key in results.get('img_fields', ['image']):\\n img = results[key]\\n results[key] = mmcv.adjust_brightness(img,\\n factor).astype(img.dtype)\",\n \"def f1_score(model_id, test_set_id, rubric_id):\\n result = {'true_positive': 0, 'false_positive': 0, 'true_negative': 0, 'false_negative': 0}\\n # right answers\\n answers = db.get_rubric_answers(test_set_id, rubric_id)\\n # rubrication results\\n rubrication_result = db.get_rubrication_result(model_id, test_set_id, rubric_id)\\n\\n for key in rubrication_result:\\n if rubrication_result[key] == answers[key]:\\n if rubrication_result[key] == 1:\\n result['true_positive'] += 1\\n else:\\n result['true_negative'] += 1\\n else:\\n if rubrication_result[key] == 1:\\n result['false_positive'] += 1\\n else:\\n result['false_negative'] += 1\\n if (result['true_positive'] + result['false_positive']) > 0:\\n result['precision'] = result['true_positive'] / (result['true_positive'] + result['false_positive'])\\n else:\\n result['precision'] = 0\\n if (result['true_positive'] + result['false_negative']) > 0:\\n result['recall'] = result['true_positive'] / (result['true_positive'] + result['false_negative'])\\n else:\\n result['recall'] = 0\\n if (result['precision'] + result['recall']) > 0:\\n result['f1'] = 2 * result['precision'] * result['recall'] / (result['precision'] + result['recall'])\\n else:\\n result['f1'] = 0\\n return result\",\n \"def metrics_evaluation(y_true, y_prob, threshold, df_type='train'):\\n\\n y_pred = (y_prob>=threshold).astype(int)\\n \\n tn = metrics.confusion_matrix(y_true, y_pred)[0][0]\\n fp = metrics.confusion_matrix(y_true, y_pred)[0][1]\\n fn = metrics.confusion_matrix(y_true, y_pred)[1][0]\\n tp = metrics.confusion_matrix(y_true, y_pred)[1][1]\\n\\n accuracy_scr = metrics.accuracy_score(y_true, y_pred)\\n precision_scr = metrics.precision_score(y_true, y_pred)\\n recall_scr = metrics.recall_score(y_true, y_pred)\\n f1_scr = metrics.f1_score(y_true, y_pred)\\n roc_auc_scr = metrics.roc_auc_score(y_true, y_pred)\\n\\n result = {'Dataset': df_type, 'No obs': len(y_true), 'Threshold': threshold,\\n 'TP':tp, 'FP': fp, 'TN': tn, 'FN':fn , \\n 'Accuracy Score':accuracy_scr, 'Precision Score':precision_scr, \\n 'Recall Score':recall_scr, 'F1 Score':f1_scr, 'ROC AUC Score':roc_auc_scr}\\n\\n return result\",\n \"def get_precision(y_true, y_pred):\\n model_entities_filter = (y_pred != 3).astype(\\\"int\\\") # of the words our model say has a NER class\\n precision_correct_entities = (y_pred[np.where(model_entities_filter)] == y_true[np.where(model_entities_filter)]).astype(\\\"int\\\")\\n precision = np.sum(precision_correct_entities)/np.sum(model_entities_filter) \\n return precision\",\n \"def test_evaluate(self):\\n # test normalized by 'bbox_size'\\n jhmdb_pck_metric = JhmdbPCKAccuracy(thr=0.5, norm_item='bbox')\\n jhmdb_pck_metric.process(self.data_batch, self.data_samples)\\n pck_results = jhmdb_pck_metric.evaluate(self.batch_size)\\n target = {\\n 'Head PCK': 1.0,\\n 'Sho PCK': 1.0,\\n 'Elb PCK': 1.0,\\n 'Wri PCK': 1.0,\\n 'Hip PCK': 1.0,\\n 'Knee PCK': 1.0,\\n 'Ank PCK': 1.0,\\n 'PCK': 1.0,\\n }\\n self.assertDictEqual(pck_results, target)\\n\\n # test normalized by 'torso_size'\\n jhmdb_tpck_metric = JhmdbPCKAccuracy(thr=0.2, norm_item='torso')\\n jhmdb_tpck_metric.process(self.data_batch, self.data_samples)\\n tpck_results = jhmdb_tpck_metric.evaluate(self.batch_size)\\n target = {\\n 'Head tPCK': 1.0,\\n 'Sho tPCK': 1.0,\\n 'Elb tPCK': 1.0,\\n 'Wri tPCK': 1.0,\\n 'Hip tPCK': 1.0,\\n 'Knee tPCK': 1.0,\\n 'Ank tPCK': 1.0,\\n 'tPCK': 1.0,\\n }\\n self.assertDictEqual(tpck_results, target)\",\n \"def precision(y_true, y_pred):\\n true_positives = bk.sum(bk.round(bk.clip(y_true * y_pred, 0, 1)))\\n predicted_positives = bk.sum(bk.round(bk.clip(y_pred, 0, 1)))\\n precision = true_positives / (predicted_positives + bk.epsilon())\\n return precision\",\n \"def mgcEval(self):\\n import numpy as np\\n def report_to_df(report):\\n\\n \\\"\\\"\\\"\\n function to convert classification report to dataframe (for visualisation plot)\\n \\\"\\\"\\\"\\n\\n report = re.sub(r\\\" +\\\", \\\" \\\", report).replace(\\\"avg / total\\\", \\\"avg/total\\\").replace(\\\"\\\\n \\\", \\\"\\\\n\\\")\\n # update this due to sklearn classification report output change\\n report = re.sub(r\\\" +\\\", \\\" \\\", report).replace(\\\"micro avg\\\", \\\"micro_avg\\\").replace(\\\"macro avg\\\", \\\"macro_avg\\\").replace(\\\"weighted avg\\\", \\\"weighted_avg\\\").replace(\\\"\\\\n \\\", \\\"\\\\n\\\")\\n report_df = pd.read_csv(StringIO(\\\"Classes\\\" + report), sep=' ', index_col=0) \\n return(report_df)\\n \\n #txt report to df\\n class_rpttop1 = classification_report(self.y_true, self.y_pred)\\n df_report = report_to_df(class_rpttop1)\\n\\n df_report = df_report.iloc[:self.nb_classes, :].copy()\\n df_report.index = df_report.index.astype(int)\\n \\n\\n # classifier prediction metrics\\n def classMetrics(averagex):\\n precision, recall, fscore, support = score(self.y_true, self.y_pred, average=averagex)\\n \\n return(\\n print(''), \\n print('-------------{0:}--------------------'.format(averagex)), \\n print('precision: {0:.4f}'.format(precision)),\\n print('recall: {0:.4f}'.format(recall)),\\n print('fscore: {0:.4f}'.format(fscore)),\\n print(''),\\n print('kappa score: {0:.4f}'.format(cohen_kappa_score(self.y_true, self.y_pred))),\\n print('accuracy score: {0:.4f}'.format(accuracy_score(self.y_true, self.y_pred))))\\n \\n def predSamp():\\n\\n correct = np.nonzero(self.y_pred==self.y_true)[0]\\n incorrect = np.nonzero(self.y_pred!=self.y_true)[0]\\n\\n # quick check of the number of correct prediction from validation set\\n print(\\\"\\\")\\n print(\\\"correct/total = {0: .4f}\\\".format(len(correct)/(len(correct)+len(incorrect))))\\n print(\\\"total correct sample = {0: .0f}\\\".format(len(correct)))\\n print('------------------------------------------------------------------')\\n \\n def classReport():\\n print('----------------------------- Classfication Report -------------------------------')\\n print(classification_report(pd.Series(self.y_true).map(self.dict_label), pd.Series(self.y_pred).map(self.dict_label)))\\n \\n self.class_rpt = pd.concat([pd.DataFrame(pd.Series(df_report.index.tolist()).map(self.dict_label), columns = ['label']), df_report], axis = 1)\\n \\n self.classMetricsMac = classMetrics(\\\"macro\\\")\\n self.classMetricsMic = classMetrics(\\\"micro\\\")\\n self.predSample = predSamp()\\n self.class_rptTop1 = classReport()\\n \\n return self\",\n \"def precision_recall_f1(prediction, ground_truth):\\n if not isinstance(prediction, list):\\n prediction_tokens = prediction.split()\\n else:\\n prediction_tokens = prediction\\n if not isinstance(ground_truth, list):\\n ground_truth_tokens = ground_truth.split()\\n else:\\n ground_truth_tokens = ground_truth\\n common = Counter(prediction_tokens) & Counter(ground_truth_tokens)\\n num_same = sum(common.values())\\n if num_same == 0:\\n return 0, 0, 0\\n p = 1.0 * num_same / len(prediction_tokens)\\n r = 1.0 * num_same / len(ground_truth_tokens)\\n f1 = (2 * p * r) / (p + r)\\n return p, r, f1\",\n \"def compute(self) -> Tensor:\\n\\n if self.samples:\\n return self.average_precisions.float() / self.total\\n else:\\n # pred_image_indices = torch.cat(self.pred_image_indices, dim=0)\\n pred_probs = torch.cat(self.pred_probs, dim=0)\\n pred_labels = torch.cat(self.pred_labels, dim=0)\\n pred_bboxes = torch.cat(self.pred_bboxes, dim=0)\\n\\n # target_image_indices = torch.cat(self.target_image_indices, dim=0)\\n target_labels = torch.cat(self.target_labels, dim=0)\\n target_bboxes = torch.cat(self.target_bboxes, dim=0)\\n\\n # pred_index = torch.nonzero((pred_labels == 1))\\n # pred_probs = pred_probs[pred_index]\\n # pred_bboxes = pred_bboxes[pred_index]\\n # target_index = torch.nonzero((target_labels == 1))\\n # target_bboxes = target_bboxes[target_index]\\n\\n\\n # _, index_sorted = torch.sort(pred_probs)\\n # pred_bboxes = pred_bboxes[index_sorted].cpu().detach().numpy()\\n # target_bboxes = target_bboxes.cpu().detach().numpy()\\n pred_probs = pred_probs.cpu().detach().numpy()\\n pred_labels = pred_labels.cpu().detach().numpy()\\n pred_bboxes = pred_bboxes.cpu().detach().numpy()\\n target_labels = target_labels.cpu().detach().numpy()\\n target_bboxes = target_bboxes.cpu().detach().numpy()\\n\\n pred_probs = pred_probs[pred_labels == 1]\\n pred_bboxes = pred_bboxes[pred_labels == 1]\\n target_bboxes = target_bboxes[target_labels == 1]\\n\\n preds_sorted_idx = np.argsort(pred_probs)[::-1]\\n pred_bboxes = pred_bboxes[preds_sorted_idx]\\n\\n x, y = calculate_precision_recall(target_bboxes, pred_bboxes)\\n\\n if len(x) >= 2:\\n return auc(x, y)\\n else:\\n return 0\\n\\n # return mean_average_precision(\\n # pred_image_indices,\\n # pred_probs,\\n # pred_labels,\\n # pred_bboxes,\\n # target_image_indices,\\n # target_labels,\\n # target_bboxes,\\n # self.iou_threshold,\\n # self.ap_calculation,\\n # )\",\n \"def _adjust_color_img(self, results, factor=1.0):\\n for key in results.get('img_fields', ['image']):\\n # NOTE defaultly the image should be BGR format\\n img = results[key]\\n results[key] = mmcv.adjust_color(img, factor).astype(img.dtype)\",\n \"def _compute_prediction(self, k, img, kp, des):\\n ###taken from example code:\\n\\n #find matches\\n matches = self.matcher.knnMatch(des,self.descs[k],k=2)\\n\\n # make sure the distance to the closest match is sufficiently better than the second closest\\n good_matches = []\\n for m,n in matches: \\n if (m.distance < 0.7*n.distance and\\n kp[m.queryIdx].response > 0.0 and\\n self.kps[k][m.trainIdx].response > 0.0):\\n good_matches.append((m.queryIdx, m.trainIdx))\\n \\n if len(good_matches) < 4:\\n visual_diff = None\\n else:\\n #create pt lists\\n img_pts = np.zeros((len(good_matches),2))\\n tem_pts = np.zeros((len(good_matches),2))\\n for idx in range(len(good_matches)):\\n match = good_matches[idx]\\n img_pts[idx,:] = kp[match[0]].pt\\n tem_pts[idx,:] = self.kps[k][match[1]].pt\\n \\n # Transform input image so that it matches the template image as well as possible\\n M, mask = cv2.findHomography(img_pts, tem_pts, cv2.RANSAC, 8.0)\\n img_T = cv2.warpPerspective(img, M, self.signs[k].shape[::-1])\\n \\t\\n #calculate error\\n visual_diff = compare_images(img_T, self.signs[k])\\n return visual_diff\\n #not functioning properly (currently does not return highest value for identical image/scene pair)\",\n \"def average_precision_recall(self,frame):\\n precs,recls = np.array([]),np.array([])\\n\\n for num in np.arange(0,10,1): # for each class\\n idx = 'Class_'+str(num)+'_precision' # index column name\\n precs = np.append(precs,frame[idx].mean()) # add col avg to arr\\n for num in np.arange(0,10,1): # for each class\\n idx = 'Class_'+str(num)+'_recal' # index column name (misspelled 'recall')\\n recls = np.append(recls,frame[idx].mean()) # add col avg to arr\\n\\n setattr(self,'precision_avg',np.mean(precs)) # average precision across 100 samples x 10 classes\\n setattr(self,'recall_avg',np.mean(recls)) # average recall across 100 samples x 10 classes\",\n \"def update(self, predictions, real_vals):\\n for pred, actual in zip(predictions, real_vals):\\n self.cm.loc[actual, pred] += 1\",\n \"def PostProcessing(image, resultList, threshold=0.6):\\n\\tnum_detections = resultList[0][0].astype(np.int)\\n\\tscores = resultList[2]\\n\\tboxes = resultList[3]\\n\\tbbox_num = 0\\n\\t\\n\\t# loop through all the detections and get the confidence and bbox coordinates\\n\\tfor i in range(num_detections):\\n\\t\\tdet_conf = scores[0, i]\\n\\t\\tdet_ymin = boxes[0, i, 0]\\n\\t\\tdet_xmin = boxes[0, i, 1]\\n\\t\\tdet_ymax = boxes[0, i, 2]\\n\\t\\tdet_xmax = boxes[0, i, 3]\\n\\n\\t\\tbbox_width = det_xmax - det_xmin\\n\\t\\tbbox_height = det_ymax - det_ymin\\n\\t\\t# the detection confidence and bbox dimensions must be greater than a minimum value to be a valid detection\\n\\t\\tif threshold <= det_conf and 1 >= det_conf and bbox_width > 0 and bbox_height > 0:\\n\\t\\t\\tbbox_num += 1\\n\\t\\t\\txmin = int(round(det_xmin * image.shape[1]))\\n\\t\\t\\tymin = int(round(det_ymin * image.shape[0]))\\n\\t\\t\\txmax = int(round(det_xmax * image.shape[1]))\\n\\t\\t\\tymax = int(round(det_ymax * image.shape[0]))\\n\\t\\t\\t\\n\\t\\t\\tcv2.rectangle(image, (xmin, ymin), (xmax, ymax), (0, 255, 0), 2)\\n\\t\\telse:\\n\\t\\t\\tcontinue\\n\\n\\tprint(\\\"detected bbox num:\\\", bbox_num)\\n\\tSRC_PATH = os.path.realpath(__file__).rsplit(\\\"/\\\", 1)[0]\\n\\tOutput_PATH = os.path.join(SRC_PATH, \\\"../output/output.jpg\\\")\\n\\ttry:\\n\\t\\tos.mkdir(os.path.join(SRC_PATH, \\\"../output/\\\"))\\n\\texcept Exception as e:\\n\\t\\tprint(\\\"Output Path already exists\\\")\\n\\tcv2.imwrite(Output_PATH, image)\",\n \"def compute_recall(data, num_labels = 4): \\n \\n # Declarating list to store results\\n recalls = []\\n \\n for instance in data:\\n \\n # Declarating list to store individual results\\n instance_recalls = []\\n \\n for i in np.arange(num_labels):\\n \\n # Computing and storing accuracy for each class\\n instance_recalls.append(recall_score(instance[:, 2 + i], instance[:, 2 + i + 4]))\\n \\n # Storing mean results of the instance\\n recalls.append(np.mean(instance_recalls))\\n \\n # Returning mean of all results\\n return np.mean(recalls)\",\n \"def compute_prf(self, counts_dict):\\n precision = counts_dict[\\\"correct\\\"] / counts_dict[\\\"predicted\\\"] if counts_dict[\\\"predicted\\\"] else 0.0\\n recall = counts_dict[\\\"correct\\\"] / counts_dict[\\\"gold\\\"] if counts_dict[\\\"gold\\\"] else 0.0\\n fscore = 2 * precision * recall / (precision + recall) if precision + recall else 0.0\\n\\n return {\\\"precision\\\": precision, \\\"recall\\\": recall, \\\"fscore\\\": fscore}\",\n \"def test_results_prefit() -> None:\\n X_train_val, X_test, y_train_val, y_test = train_test_split(\\n X, y, test_size=1/10, random_state=1\\n )\\n X_train, X_val, y_train, y_val = train_test_split(\\n X_train_val, y_train_val, test_size=1/9, random_state=1\\n )\\n estimator = LinearRegression().fit(X_train, y_train)\\n mapie = MapieRegressor(estimator=estimator, cv=\\\"prefit\\\")\\n mapie.fit(X_val, y_val)\\n _, y_pis = mapie.predict(X_test, alpha=0.05)\\n width_mean = (y_pis[:, 1, 0] - y_pis[:, 0, 0]).mean()\\n coverage = regression_coverage_score(\\n y_test, y_pis[:, 0, 0], y_pis[:, 1, 0]\\n )\\n np.testing.assert_allclose(width_mean, WIDTHS[\\\"prefit\\\"], rtol=1e-2)\\n np.testing.assert_allclose(coverage, COVERAGES[\\\"prefit\\\"], rtol=1e-2)\",\n \"def score(self, test_data):\\n\\n\\t\\tpass\",\n \"def precision(y_true, y_pred):\\n true_positives = K.sum(K.round(K.clip(y_true * y_pred, 0, 1)))\\n predicted_positives = K.sum(K.round(K.clip(y_pred, 0, 1)))\\n precision = true_positives / (predicted_positives + K.epsilon())\\n return precision\"\n]","isConversational":false},"negative_scores":{"kind":"list like","value":["0.625239","0.6167243","0.6078468","0.6062458","0.60042554","0.5945311","0.59133434","0.5907834","0.5897664","0.587507","0.58680505","0.58505803","0.5835626","0.5821899","0.58180785","0.57850605","0.5741462","0.5735853","0.57268965","0.5706414","0.5699125","0.5680613","0.56666785","0.566473","0.5649651","0.56173074","0.5615667","0.56099325","0.5603431","0.5584324","0.55763245","0.55688035","0.5563316","0.55529326","0.5535225","0.5520028","0.5518867","0.55168426","0.5514931","0.55137914","0.55068344","0.55030996","0.5497573","0.54939604","0.5483678","0.54806143","0.5478781","0.54782504","0.5473526","0.54663163","0.5452941","0.54413664","0.5437331","0.5437077","0.5424334","0.5412816","0.5404485","0.5401047","0.5399176","0.53964305","0.5392327","0.5391181","0.5386301","0.53832865","0.53825516","0.53823316","0.5381935","0.53798413","0.5377619","0.53768796","0.5375971","0.5375185","0.53722185","0.5365919","0.5357084","0.53566617","0.53551877","0.53551835","0.5347878","0.5347309","0.5345249","0.5343435","0.533222","0.5330156","0.53270316","0.53151244","0.53084755","0.53052706","0.53022754","0.53017724","0.52956975","0.5289301","0.5286539","0.528572","0.52853996","0.5281019","0.5274307","0.52733576","0.52723455","0.5267981"],"string":"[\n \"0.625239\",\n \"0.6167243\",\n \"0.6078468\",\n \"0.6062458\",\n \"0.60042554\",\n \"0.5945311\",\n \"0.59133434\",\n \"0.5907834\",\n \"0.5897664\",\n \"0.587507\",\n \"0.58680505\",\n \"0.58505803\",\n \"0.5835626\",\n \"0.5821899\",\n \"0.58180785\",\n \"0.57850605\",\n \"0.5741462\",\n \"0.5735853\",\n \"0.57268965\",\n \"0.5706414\",\n \"0.5699125\",\n \"0.5680613\",\n \"0.56666785\",\n \"0.566473\",\n \"0.5649651\",\n \"0.56173074\",\n \"0.5615667\",\n \"0.56099325\",\n \"0.5603431\",\n \"0.5584324\",\n \"0.55763245\",\n \"0.55688035\",\n \"0.5563316\",\n \"0.55529326\",\n \"0.5535225\",\n \"0.5520028\",\n \"0.5518867\",\n \"0.55168426\",\n \"0.5514931\",\n \"0.55137914\",\n \"0.55068344\",\n \"0.55030996\",\n \"0.5497573\",\n \"0.54939604\",\n \"0.5483678\",\n \"0.54806143\",\n \"0.5478781\",\n \"0.54782504\",\n \"0.5473526\",\n \"0.54663163\",\n \"0.5452941\",\n \"0.54413664\",\n \"0.5437331\",\n \"0.5437077\",\n \"0.5424334\",\n \"0.5412816\",\n \"0.5404485\",\n \"0.5401047\",\n \"0.5399176\",\n \"0.53964305\",\n \"0.5392327\",\n \"0.5391181\",\n \"0.5386301\",\n \"0.53832865\",\n \"0.53825516\",\n \"0.53823316\",\n \"0.5381935\",\n \"0.53798413\",\n \"0.5377619\",\n \"0.53768796\",\n \"0.5375971\",\n \"0.5375185\",\n \"0.53722185\",\n \"0.5365919\",\n \"0.5357084\",\n \"0.53566617\",\n \"0.53551877\",\n \"0.53551835\",\n \"0.5347878\",\n \"0.5347309\",\n \"0.5345249\",\n \"0.5343435\",\n \"0.533222\",\n \"0.5330156\",\n \"0.53270316\",\n \"0.53151244\",\n \"0.53084755\",\n \"0.53052706\",\n \"0.53022754\",\n \"0.53017724\",\n \"0.52956975\",\n \"0.5289301\",\n \"0.5286539\",\n \"0.528572\",\n \"0.52853996\",\n \"0.5281019\",\n \"0.5274307\",\n \"0.52733576\",\n \"0.52723455\",\n \"0.5267981\"\n]","isConversational":false},"document_score":{"kind":"string","value":"0.6408821"},"document_rank":{"kind":"string","value":"0"}}},{"rowIdx":93298,"cells":{"query":{"kind":"string","value":"creates a dictionary that can be ingested by pandas DataFrame to write to a csv file"},"document":{"kind":"string","value":"def create_simple_experiment_dict_for_csv(experiment_dict, list_dict, algorithm, thresholds):\n experiment_dict_for_csv = {\n \"attributes\": ['TP', 'FP', 'TN', 'FN', 'TPR (Sensitivity)', 'TNR (Specificity)', 'Precision', 'Recall',\n 'F1 Measure', 'Accuracy']}\n for row_key in list_dict.values():\n for threshold in thresholds:\n algorithm_column_name = algorithm.get_algorithm_name() + \"_actual\"\n column_name = algorithm.get_algorithm_name() + \"_\" + str(threshold)\n row_value = experiment_dict[str(threshold)][algorithm_column_name][row_key]\n if column_name in experiment_dict_for_csv.keys():\n experiment_dict_for_csv[column_name].append(row_value)\n else:\n experiment_dict_for_csv[column_name] = [row_value]\n return experiment_dict_for_csv"},"metadata":{"kind":"string","value":"{\n \"objective\": {\n \"self\": [],\n \"paired\": [],\n \"triplet\": [\n [\n \"query\",\n \"document\",\n \"negatives\"\n ]\n ]\n }\n}"},"negatives":{"kind":"list like","value":["def csv_to_dict(self):\n log = logger.configure(\"default\")\n try: \n df = pd.read_csv(self.__csv_path)\n except IOError as e:\n # file not found\n log.error('Could not import {}. Got error {}'.format(self.__csv_path, e))\n raise \n else:\n cols = list(df.columns)\n metafield_cols = [col for col in cols if 'metafields' in col]\n if metafield_cols == [] or 'Handle' not in cols:\n # relevant columns don't exist\n log.error('{} does not contain `Handle` or `metafields` named columns'.format(self.__csv_path))\n raise\n else:\n new_cols = ['Handle'] + metafield_cols\n df = df[new_cols].set_index('Handle')\n df = df[~df.index.duplicated(keep='first')]\n return df.to_dict('index')","def to_dict_records(df):\r\n return df.to_dict('records')","def DataFrame_to_dict(df):\n return df.to_dict(orient='records')","def df_to_dict(df):\n return df.to_dict('index')","def to_dict(df):\n\t# Extract the names\n\tnames = list(df.columns.values)\n\t# Create sequence\n\tdf_list = []\n\t#Iterate over rows of data frame\n\tfor index, row in df.iterrows():\n\t\t# Create temporary object\n\t\ttemp_dict = {}\n\t\t#Iterate over key names and add values\n\t\tfor name in names:\n\t\t\ttemp_dict[name] = row[name]\n\t\t# Append object to dictionary\n\t\tdf_list.append(temp_dict.copy())\n\t\n\treturn df_list","def write_dataframe(df: pd.DataFrame, filename: str) -> Dict:\n # Ensure that the parent directory for the output file exists.\n dirname = os.path.dirname(filename)\n if dirname:\n os.makedirs(dirname, exist_ok=True)\n # Create a unique list of column names and a mapping from the new uniqye\n # names to the original columns in the given data frame.\n columns = list()\n column_mapping = dict()\n for colidx in range(len(df.columns)):\n colname = 'COL{}'.format(colidx)\n columns.append(colname)\n column_mapping[colname] = df.columns[colidx]\n # Write data frame to temporary CSV file.\n df.to_csv(\n filename,\n header=columns,\n index=False,\n compression=None\n )\n # Return the created column mapping..\n return column_mapping","def create_cris_data_dict(df, filename, outdir):\n json_dict = {}\n for key in df.keys():\n if key != 'participant_id':\n json_dict[key] = {'Units': key.split()[-1]}\n else:\n json_dict[key] = {'Description': 'OpenNeuro ID of the subject.'}\n with open(outdir.joinpath(filename + '.json'), \"w\") as f:\n json.dump(json_dict, f, indent=4)","def __create_csv(self):\n with open(self.__csv_file_name, 'w', newline='', encoding='utf-8') as csv_file:\n writer = csv.DictWriter(csv_file, fieldnames=self.__csv_fields, delimiter=';')\n writer.writeheader()","def dictify(df):\n return {str(k): v for k, v in df.items()}","def WriteDictToCSV(csv_file,csv_columns,dict_data):\n with open('history.csv','wb') as csvfile:\n w = csv.writer(csvfile)\n w.writerows(dict_data.items())\n\n return","def pandas_convert(self):\n data = {}\n\n for names in self.data[0]:\n col_values = []\n\n if names in objects:\n for items in self.data[0][names]:\n col_values = []\n\n col_name = names + \"_\" + items\n\n for i in range(len(self.data)):\n col_values.append(self.data[i][names][items])\n\n data[col_name] = col_values\n else:\n for i in range(len(self.data)):\n col_values.append(self.data[i][names])\n \n data[names] = col_values\n\n self.pandas_df = pd.DataFrame(data=data)\n self.__clean_df()\n\n return self.pandas_df","def exportDict (dict, nameOfFile):\n dictTable = pd.DataFrame(dict)\n dictTable.to_csv(nameOfFile)","def to_dict_index(df):\r\n return df.to_dict('index')","def make_dict(ECOnew,goodObj):\n \n ECO_new = pd.read_csv(ECOnew, delimiter='\\t')\n \n good_Obj = pd.read_csv(goodObj,header=None,names=['ECO_ID'])\n #An array of the ECOIDs obtained from goodObj.txt\n arr1 = good_Obj.ECO_ID.values\n \n #Creates a new dataframe where you grab all the columns that match ECOID\n #in goodObj with ECOID in the formatted ECO catalog\n ECOnew_goodObj = ECO_new.loc[ECO_new.ECOID.isin(arr1), : ]\n \n #Group by ECOID and filters\n ECOnew_groups = ECOnew_goodObj.groupby(['ECOID','filters'])\n #Retrieve keys to the group sorted by ECOID\n ECOnew_keys = sorted(ECOnew_groups.groups.keys())\n \n return ECOnew_groups, ECOnew_keys, ECO_new","def make_csv(data_dict):\n fieldnames = ['name'] + data_dict.itervalues().next().keys()\n with open('data.csv', 'w') as csvfile:\n writer = csv.DictWriter(csvfile, fieldnames=fieldnames)\n writer.writeheader()\n for record in data_dict:\n person = data_dict[record]\n person['name'] = record\n assert set(person.keys()) == set(fieldnames)\n writer.writerow(person)","def make_csv(data_dict):\r\n fieldnames = ['name'] + data_dict.itervalues().next().keys()\r\n with open('data.csv', 'w') as csvfile:\r\n writer = csv.DictWriter(csvfile, fieldnames=fieldnames)\r\n writer.writeheader()\r\n for record in data_dict:\r\n person = data_dict[record]\r\n person['name'] = record\r\n assert set(person.keys()) == set(fieldnames)\r\n writer.writerow(person)","def build_export_row(self, data:dict) -> dict:\n data = {k : v for k, v in data.items() if v is not np.nan}\n for key, value in data.items():\n if key in _column_types:\n if _column_types[key] == list:\n if isinstance(value, (list, set, tuple)):\n data[key] = LIST_DELIMITER.join(value)\n else:\n data[key] = str(value)\n elif _column_types[key] == bool:\n try:\n data[key] = bool(value)\n except:\n data[key] = False\n else:\n data[key] = str(value)\n return data","def convert_to_dict(data_frame: pd.DataFrame) -> List[Dict[str, Any]]:\n type_conversion = {\"date\": str} if \"date\" in data_frame.columns else {}\n return data_frame.replace({np.nan: None}).astype(type_conversion).to_dict(\"records\")","def build_df_dict(df, geneDictCanon):\n\n\toutDict = OrderedDict()\n\n\tfor tr in geneDictCanon:\n\t\toutDict[geneDictCanon[tr][0]] = df.iloc[geneDictCanon[tr][1][0]:geneDictCanon[tr][1][1]]\n\n\treturn outDict","def to_csv(self, filename, **csv_args):\n dataframe = pd.DataFrame.from_dict(self.to_dict())\n return dataframe.to_csv(filename, **csv_args)","async def full_report():\n return DF.to_dict(orient=\"records\")","def write_csv(fd, data):\n # df = pd.DataFrame.from_dict(data)\n df = pd.io.json.json_normalize(data)\n print(df.to_csv(index=False), file=fd)","def tsvRowToDict(row):\n return {col: getattr(row, col) for col in row._columns_}","def make_dict(row):\n return dict((key[0], value) for key, value in zip(colnames, row))","def convert_data(df):\n print(\"Converting history...\")\n return [ dict(row) for i, row in df.iterrows() ]","def load_data(self):\n df = pandas.read_csv(self.path)\n self.data_dict = df.to_dict(orient=\"list\")\n return self.data_dict","def store_csv(self):\n\n with open(self.filepath.with_suffix(\".csv\"), 'w',\n newline='') as csvfile:\n fieldnames = ['counter', 'timestamp', 'acceleration']\n writer = DictWriter(csvfile, fieldnames=fieldnames)\n\n writer.writeheader()\n writer.writerows(self.values)","def df_to_dict(df, key_column, val_column):\n xkey = df[key_column].tolist()\n xval = df[val_column].tolist()\n return dict(zip(xkey,xval))","def creating_dict_from_csv(self) -> dict:\n dictionary = {}\n for row in self.__read_csv():\n if dictionary.get(row[0]):\n dictionary[row[0]].append((row[1], row[2]))\n else:\n dictionary[row[0]] = [(row[1], row[2])]\n\n for key, value in dictionary.items():\n dictionary[key] = sorted(value, key=lambda x: x[1], reverse=True)\n\n return dictionary","def prepare_CSV(self):\n self.drop_columns()\n self.rename_columns()\n self.spilt_columns()\n self.add_vehicle_id_column()\n self.add_source_column()\n self.add_timestamp_columns()\n self.get_colour_columns()\n self.clean_column_formats()\n\n # print(self.data.info())\n # print(self.data.sample(10))\n\n return self.data","def dataframe_to_dict(self, orient=\"dict\", into=dict, index=True): # noqa: PR01\n return self.to_pandas().to_dict(orient, into, index)","def write_to_csv(self, data_points):\n keys = data_points[0].keys()\n with open(self.report_path, 'w') as output_file:\n dict_writer = csv.DictWriter(output_file, keys)\n dict_writer.writeheader()\n dict_writer.writerows(data_points)","def to_df(self) -> Dict[str, Any]:\n return {key: self.render_df(key) for key, value in self.__dict__.items()}","def test_ingest_with_column_names_dict():\n schema = {\"foo\": \"foo1\", \"bar\": \"bar2\"}\n\n data = [{\"foo\": 1, \"bar\": 2}, {\"foo\": 10, \"bar\": 20}]\n\n converted_data = client.ingest_data(data, schema)\n assert converted_data.to_pydict() == {'foo1': [1, 10], 'bar2': [2, 20]}","def dataframe():\n headers = get_headers()\n headers = {'headers': headers}\n headers = pd.DataFrame.from_dict(headers, orient='index')\n headers = headers.replace(r'\\n', ' ', regex=True)\n headers = headers.replace(r'\\r', ' ', regex=True)\n headers = headers.replace(r'\\t', ' ', regex=True)\n headers = headers.replace(r'\\\\t', ' ', regex=True)\n headers = headers.replace(r' ', ' ', regex=True)\n headers = headers.replace(r' ', ' ', regex=True)\n\n paragraphs = get_paragraphs()\n paragraphs = {'paragraphs': paragraphs}\n paragraphs = pd.DataFrame.from_dict(paragraphs, orient='index')\n paragraphs = paragraphs.replace(r'\\n', ' ', regex=True)\n paragraphs = paragraphs.replace(r'\\r', ' ', regex=True)\n paragraphs = paragraphs.replace(r'\\t', ' ', regex=True)\n paragraphs = paragraphs.replace(r'\\\\t', ' ', regex=True)\n paragraphs = paragraphs.replace(r' ', ' ', regex=True)\n paragraphs = paragraphs.replace(r' ', ' ', regex=True)\n\n return headers.to_csv('headers.csv', index=False), paragraphs.to_csv('paragraphs.csv', index=False)","def dictify_df(frame):\n ret = {}\n for row in frame.values:\n cur_level = ret\n for elem in row[:-2]:\n if elem not in cur_level:\n cur_level[elem] = {}\n cur_level = cur_level[elem]\n cur_level[row[-2]] = row[-1]\n return ret","def write_csv(data):\n\n with open('data.csv', 'w') as file:\n writer = csv.DictWriter(file, fieldnames=data[0].keys())\n writer.writeheader()\n for row in data:\n writer.writerow(row)","def output_df(outdict, out_file):\n\tcols = ['#chrom', 'source', 'feature', 'chromStart', 'chromEnd', 'score', 'strand', 'frame', 'transcript_id']\n\tcolOut = ['#chrom', 'source', 'feature', 'start', 'end', 'score', 'strand', 'frame', 'transcript_id']\n\tgtfDF = pd.DataFrame(columns=cols)\n\n\tfor trsp in outdict:\n\t\tgtfDF = gtfDF.append(outdict[trsp], ignore_index=True)\n\t\t\n\tgtfDF.columns = colOut\n\t# print gtfDF.head()\n\tgtfDF.to_csv(out_file, compression='gzip', sep='\\t', index=False)","def csv_dict_writer(path, fieldnames, data):\n with open(path, \"wb\") as out_file:\n writer = csv.DictWriter(out_file, delimiter=',', fieldnames=fieldnames)\n writer.writeheader()\n for row in data:\n writer.writerow(row)","def isolate_df_data(self, dict_df):\n\t\treturn_dict_df = {}\n\t\tfor k,v in dict_df.items():\n\t\t\tnew_df = pd.DataFrame(index=v.index)\n\t\t\tnew_df[k + ' Value'] = v['value']\n\t\t\treturn_dict_df[k] = new_df\n\t\treturn return_dict_df","def gen_dict():\n lines = [line for line in csv.reader(open(__ppath__ + \"/data/occupations.csv\"))] # uses a csv.reader to parse the file, converts the generic iterable to a list\n lines = [(line[0],float(line[1])) for line in lines[1:-2]]# removes the column names and \"Total\" row, re-expresses as a list of tuples to enable dictionary conversion\n lines.append((\"Unemployed\",0.2)) # accounts for missing 0.2% of jobs\n return dict(lines) # converts to dictionary","def df2dict(df):\n attr_list = list(df.columns)\n res_dict = {}\n for attr in attr_list:\n res_dict[attr.lower()] = list(df[attr])\n\n return res_dict","def write_to_csv(path,data_dict):\n\n\n schema = [\"file_name\",\"family\",\"genus\",\"genus_confidence\",\n \"species_1\",\"confidence_1\",\"hall_1\",\n \"species_2\",\"confidence_2\",\"hall_2\",\n \"species_3\",\"confidence_3\",\"hall_3\",\n \"species_4\",\"confidence_4\",\"hall_4\",\"peaks\"]\n\n # if no file exists create a one and inform the user\n if not os.path.exists(path):\n print(\"creating new output file {}\".format(path))\n with open(path, \"w\") as csv_file:\n filewriter = csv.writer(csv_file, delimiter=\",\")\n filewriter.writerow(schema)\n\n row = []\n\n row.append(data_dict[\"file_name\"])\n row.append(data_dict[\"family\"])\n \n row.append(data_dict[\"genus_1\"])\n row.append(data_dict[\"genus_confidence_1\"][:5])\n \n row.append(data_dict[\"species_1\"])\n row.append(data_dict[\"confidence_1\"][:5])\n row.append(data_dict[\"hall_1\"])\n \n row.append(data_dict[\"species_2\"])\n row.append(data_dict[\"confidence_2\"][:5])\n row.append(data_dict[\"hall_2\"])\n\n row.append(data_dict[\"species_3\"])\n row.append(data_dict[\"confidence_3\"][:5])\n row.append(data_dict[\"hall_3\"])\n\n row.append(data_dict[\"species_4\"])\n row.append(data_dict[\"confidence_4\"][:5])\n row.append(data_dict[\"hall_4\"])\n\n row.append(data_dict[\"peaks\"])\n \n with open(path, \"a\") as csv_file:\n filewriter = csv.writer(csv_file, delimiter=\",\")\n filewriter.writerow(row)","def generate_csv(data: dict, cols: list, file_name: str) -> None:\r\n\r\n longest = 0\r\n for i in cols:\r\n if len(data[i]) > longest:\r\n longest = len(data[i])\r\n\r\n len_cols = []\r\n for i in cols:\r\n len_cols.append(str(i) + \" (\" + str(len(data[i])) + \")\")\r\n\r\n for i in cols:\r\n while len(data[i]) < longest:\r\n data[i].append(\"\")\r\n\r\n new_data = {}\r\n count = 0\r\n for i in len_cols:\r\n new_data[i] = data[str(count)]\r\n count += 1\r\n\r\n df = pd.DataFrame(data=new_data, columns=len_cols)\r\n print(df)\r\n\r\n df.to_csv(file_name, encoding='utf-8', index=False)","def save_dataset_csv(self, path):\n cols = list(self.data_dict.keys())\n df = pd.DataFrame(self.data_dict, index=None, columns=cols)\n df.to_csv(path, index=True)","def csv_dict_writer(path, fieldnames, data):\n with open(path, \"wb\") as out_file:\n writer = csv.DictWriter(out_file, delimiter=',', fieldnames=fieldnames)\n writer.writeheader()\n for row in data:\n writer.writerow(row)","def build_dataset_by_df(file_path):\n data = pd.read_csv(file_path, sep=',')\n print(data.head())\n\n user_ids = data['CustomerID'].unique()\n prod_ids = data['StockCode'].unique()\n user_num = len(user_ids)\n product_num = len(prod_ids)\n product_lists_by_user = dict(data.groupby('CustomerID')['StockCode'].apply(list)).values()\n product_to_desc = dict(data.groupby('StockCode')['Description'].apply(set))\n dictionary = {}\n for id in prod_ids:\n dictionary[id] = len(dictionary)\n reversed_dictionary = dict(zip(dictionary.values(), dictionary.keys()))\n return product_lists_by_user, product_to_desc, dictionary, reversed_dictionary, user_num, product_num","def build_data_frame(sql_dict, names):\n\td = defaultdict(list)\n\ti = 0\n\tfor category_data in sql_dict:\n\t\tfor attribute_list in category_data:\n\t\t\tfor attribute in category_data[attribute_list]:\n\t\t\t\td[attribute_list].append(attribute)\n\t\td[\"category\"].extend([names[i]]* len(category_data[attribute_list]) )\n\t\td[\"fct\"].extend([i] * len(category_data[attribute_list]) )\n\t\ti += 1\n\t#Translation into R objects\n\tfor attribute_list in d:\n\t\tif attribute_list == \"category\":\n\t\t\td[attribute_list] = robjects.StrVector(d[attribute_list])\n\t\telif attribute_list in WATCH_FOR_MISSING_VALUES:\n\t\t\ttmp = robjects.FloatVector(range(len(d[attribute_list])))\n\t\t\tfor i in range(len(d[attribute_list])):\n\t\t\t\tif d[attribute_list][i] == \"NA\":\n\t\t\t\t\ttmp.rx[i] = robjects.NA_Real\n\t\t\t\telse:\n\t\t\t\t\ttmp.rx[i] = d[attribute_list][i]\n\t\t\td[attribute_list] = tmp\n\t\telse:\n\t\t\tif attribute_list == \"num_sections\":\n\t\t\t\ttmp = []\n\t\t\t\tfor element in d[attribute_list]:\n\t\t\t\t\tif element == None:\n\t\t\t\t\t\ttmp.append(1)\n\t\t\t\t\telse:\n\t\t\t\t\t\ttmp.append(float(element))\n\t\t\t\td[attribute_list] = robjects.FloatVector(tmp)\n\t\t\t\tcontinue\n\t\t\telif attribute_list == \"speechiness\":\n\t\t\t\ttmp = []\n\t\t\t\tfor element in d[attribute_list]:\n\t\t\t\t\tif element == None:\n\t\t\t\t\t\ttmp.append(1)\n\t\t\t\t\telse:\n\t\t\t\t\t\ttmp.append(float(element))\n\t\t\t\td[attribute_list] = robjects.FloatVector(tmp)\n\t\t\t\tcontinue\n\t\t\t\t \n\t\t\telse:\n\t\t\t\td[attribute_list] = robjects.FloatVector(d[attribute_list])\n\tdata_frame = robjects.DataFrame(d)\n\treturn data_frame","def write_csv(fn, toCSV):\n keys = toCSV[0].keys()\n with open(fn, 'w') as output_file:\n dict_writer = csv.DictWriter(output_file, keys)\n dict_writer.writeheader()\n dict_writer.writerows(toCSV)","def write_dataframe(self, path=os.getcwd(), name=\"Airbnb.csv\") -> None:\n try:\n df = pd.DataFrame(self.__collected_dic)\n except ValueError:\n print(\"ERROR! - Dictionary values are not the same length\")\n return\n\n if not isinstance(name, str):\n raise TypeError\n if \".csv\" != name[-4:]:\n name = f\"{name}.csv\"\n df.to_csv(os.path.join(path, name), index=False)\n print(f\"{name} file was succesfully written in {path}\")","def make_ordered_to_dict(sorted_columns, include_timestamp):\n od = OrderedDict()\n for column in sorted_columns:\n if include_timestamp:\n value = (column.cell.value, column.cell.timestamp)\n else:\n value = column.cell.value\n od[column.columnName] = value\n return od","def make_dict(data_for_dict): \n \n column_name_list = data_for_dict[0]\n db_list = data_for_dict[1:]\n \n column_list1 = []\n column_list2 = []\n column_list3 = []\n column_list4 = []\n column_list5 = []\n column_list6 = []\n column_list7 = []\n column_list8 = []\n column_list9 = []\n column_list10 = []\n column_list11 = []\n hmdb_dict = {}\n for line in db_list:\n my_string1 = '' \n my_string2 = ''\n my_string3 = ''\n my_string4 = ''\n my_string5 = ''\n my_string6 = ''\n my_string7 = ''\n my_string8 = ''\n my_string9 = ''\n my_string10 = ''\n my_string11 = ''\n\n my_string1 = line[0]\n column_list1 += [my_string1]\n my_string2 += line[1]\n column_list2 += [my_string2]\n my_string3 += line[2]\n column_list3 += [my_string3]\n my_string4 += line[3]\n column_list4 += [my_string4]\n my_string5 += line[4]\n column_list5 += [my_string5]\n my_string6 += line[5]\n column_list6 += [my_string6]\n my_string7 += line[6]\n column_list7 += [my_string7]\n my_string8 += line[7]\n column_list8 += [my_string8]\n my_string9 += line[8]\n column_list9 += [my_string9]\n my_string10 += line[9]\n column_list10 += [my_string10]\n my_string11 += line[10]\n column_list11 += [my_string11] \n \n hmdb_dict[column_name_list[0]] = column_list1\n hmdb_dict[column_name_list[1]] = column_list2\n hmdb_dict[column_name_list[2]] = column_list3\n hmdb_dict[column_name_list[3]] = column_list4\n hmdb_dict[column_name_list[4]] = column_list5\n hmdb_dict[column_name_list[5]] = column_list6\n hmdb_dict[column_name_list[6]] = column_list7\n hmdb_dict[column_name_list[7]] = column_list8\n hmdb_dict[column_name_list[8]] = column_list9\n hmdb_dict[column_name_list[9]] = column_list10\n hmdb_dict[column_name_list[10]] = column_list11\n \n return (hmdb_dict)","def chunk_to_dict(chunk):\n csv_cols = chunk.keys()\n return [dict(zip(csv_cols, v)) for v in chunk.values]","def as_dict(self) :\n\n return dict(zip(self.cols, self.data))","def write_csv(dictionary):\n\n with open(constant.METADATA_FILE, 'a', newline='') as file:\n writer = csv.DictWriter(file, fieldnames=dictionary.keys())\n try:\n writer.writerow(dictionary)\n except Exception:\n writeable = {k: str(v).encode(\"utf-8\") for k, v in dictionary.items()}\n writer.writerow(writeable)","def _create_model_out_dictkeys():\r\n model_names = []\r\n result_keys = []\r\n for model_name in model_zoo._ModelZooUrls.CONFIG_PATH_TO_URL_SUFFIX.keys():\r\n try:\r\n print(model_name, \":\")\r\n select_model(model_name)\r\n result = get_features_by_image_path(\"./sample.jpg\")\r\n model_names.append(model_name)\r\n result_keys.append(list(result.keys()))\r\n except RuntimeError as t:\r\n print(t)\r\n\r\n pd.DataFrame(list(zip(model_names, result_keys))).to_csv(\"d2_model_out_dictkeys.csv\")","def create_dataframe(self, dictionary_to_convert, cols):\n\n dataframe_converted = pd.DataFrame.from_dict(dictionary_to_convert, orient='index', columns = cols)\n dataframe_converted = dataframe_converted.reset_index()\n dataframe_converted = dataframe_converted.drop(columns=['index'])\n\n return dataframe_converted","def write_csv(self, _dict, filename):\n with open(filename, 'w') as f:\n f.write('\"'+'\";\"'.join(_dict.keys())+'\"\\n')\n for i in np.arange(len(_dict[list(_dict.keys())[0]])):\n values = []\n for col in _dict.keys():\n try:\n values.append(str(_dict[col][i]))\n except IndexError as e:\n # LSTM don't have first times available because of lacking history\n pass\n f.write(';'.join(values)+'\\n')\n\n logging.info('Wrote {}'.format(filename))\n self._upload_to_bucket(filename, filename)","def _export(self, *args, **kwargs) -> Tuple[pd.DataFrame, dict]:\n return self.data, kwargs","def save_json(df):\n dict = {}\n for row in df.iterrows():\n dict[row[1]['Country']] = {'Region' : row[1]['Region'],\n 'Pop. Density (per sq. mi.)' : row[1]['Pop. Density (per sq. mi.)'],\n 'Infant mortality (per 1000 births)' : row[1]['Infant mortality (per 1000 births)'],\n 'GDP ($ per capita) dollars' : row[1]['GDP ($ per capita) dollars']\n }\n\n with open('data.json', 'w', encoding='utf8') as outfile:\n data = json.dumps(dict, indent=4, sort_keys=False, separators=(',', ': '), ensure_ascii=False)\n outfile.write(data)","def dict_to_csv(filename, dictionary, paramdict=True):\n if paramdict:\n filename += '_params'\n \n with open(filename + '.csv', 'w') as f:\n for key in dictionary.keys():\n if type(dictionary[key]) == tuple:\n f.write(\"%s,%s, %s \\n\"%(key, dictionary[key][0], dictionary[key][1]))\n else:\n f.write(\"%s,%s \\n\"%(key, dictionary[key]))\n print('Saved %s.csv.'%filename)","def as_dict(self):\r\n values = {}\r\n for name, col in self._columns.items():\r\n values[name] = col.to_database(getattr(self, name, None))\r\n return values","def generate_new_row(forecast_date, target, target_end_date, \n location, type, quantile, value):\n new_row = {}\n new_row[\"forecast_date\"] = forecast_date\n new_row[\"target\"] = target\n new_row[\"target_end_date\"] = target_end_date\n new_row[\"location\"] = location\n new_row[\"type\"] = type\n new_row[\"quantile\"] = quantile\n new_row[\"value\"] = value\n return new_row","def create_dataframe():\r\n\r\n df = pd.read_csv('data/data.csv', header=0)\r\n return df","def make_data(rics: list, fields: list, name: str) -> None:\n df = get_data(rics, fields)\n df.index.name = 'ID'\n to_csv(df, name)","def generate_log_df(log_columns, log_values):\n return pd.DataFrame(dict(zip(log_columns, log_values)), index=[0])","def link_datadict_to_dataframe(row, dict):\n d = {}\n for k, v in dict.items():\n d[k] = [row[c] if row[c] == row[c] else None for c in v[\"COLUMN\"]]\n\n if len(d[k]) == 1:\n d[k] = d[k][0]\n elif None in d[k]:\n d[k] = None\n\n return d","def _convert_csv_column_to_dict(csv_data, column):\n results = dict()\n\n for row in csv_data:\n key = row[0]\n data = row[1:][column]\n\n if data:\n if key not in results:\n results[key] = data.strip() if data else \"\"\n else:\n # append multiple choice questions\n results[key] += \"|{0}\".format(data.strip())\n\n return results","def create_dataset(input_file_path, output_file_path):\n col_index_map = {'user_id': 0, 'session_id': 1, 'timestamp': 2, 'step': 3, 'action_type': 4, 'reference': 5,\n 'platform': 6, 'city': 7, 'device': 8,\n 'current_filters': 9, 'impressions': 10, 'prices': 11}\n flat_dict = dict()\n with open(input_file_path, 'r') as csvFile:\n reader = csv.reader(csvFile)\n header = next(reader)\n col_names = [col_name for col_name in col_index_map.keys()]\n col_names.pop(0)\n index = 0\n for row in tqdm(reader):\n if len(flat_dict) > 40000:\n index += 1\n with open(output_file_path + \"_\" + str(index) + \".json\", \"w\") as file:\n json.dump(flat_dict, file)\n print(\" JSON : \", index)\n flat_dict = dict()\n col_values = [row[col_index_map[c_n]] for c_n in col_names]\n dict_for_each_row = dict(zip(col_names, col_values))\n to_list = dict_for_each_row['impressions']\n dict_for_each_row['impressions'] = to_list.split('|')\n to_list = dict_for_each_row['prices']\n dict_for_each_row['prices'] = to_list.split('|')\n user_id = row[col_index_map['user_id']]\n if user_id in flat_dict:\n flat_dict[user_id].append(dict_for_each_row)\n else:\n flat_dict[user_id] = [dict_for_each_row]\n\n print(\"Output is Saved\")","def readData(filename):\r\n data_d = {}\r\n with open(filename) as f:\r\n df = pd.read_csv(f, header=0, dtype='str',sep=';')\r\n df = df.loc[:, ~df.columns.str.contains('^Unnamed')]\r\n df_dict = df.to_dict(orient='index')\r\n for i,val in df_dict.iteritems(): \r\n clean_row = [(k, p.proc(v)) for (k, v) in val.iteritems()]\r\n row_id = val['line_nr']\r\n data_d[row_id] = dict(clean_row)\r\n return data_d\r\n return df","def create_csvfile(options_path, recept_info,df):\n df_csv = df.copy()\n df_csv.index.names = ['Interacting positions']\n\n #Change dynX by full name of receptor, adding the dynid if there is more than a simulation for that receptor\n df_csv.columns = df_csv.columns.map(lambda x: recept_info[x][13])\n\n #Sorting by ballesteros Id's (helixloop column) and clustering order\n df_csv['Interacting positions'] = df_csv.index\n df_csv['helixloop'] = df_csv['Interacting positions'].apply(lambda x: re.sub(r'^(\\d)x',r'\\g<1>0x',x)) \n df_csv = df_csv.sort_values([\"helixloop\"])\n\n #Change jumplines by 'x' to avoid formatting problems\n def new_index(cell):\n cell = cell.replace('\\n\\n', ' ')\n cell = cell.replace('\\n', 'x')\n cell = cell.replace('xx', 'x')\n return cell\n\n df_csv['Interacting positions'] = df_csv['Interacting positions'].apply(lambda x: new_index(x))\n df_csv.index = df_csv['Interacting positions']\n\n #Drop columns\n df_csv.drop(columns = ['helixloop','Interacting positions'], inplace = True)\n\n #Store dataframe as csv\n df_csv.to_csv(path_or_buf = options_path+\"dataframe.csv\", float_format='%.1f')","def write_to_csv(data_in_dictionary, csvfile):\n with open(csvfile, 'wb') as f:\n fieldnames = data_in_dictionary.keys()\n writer = csv.DictWriter(f, fieldnames)\n writer.writeheader()\n writer.writerow(data_in_dictionary)\n logger.info(\"Data written to file: \" + csvfile)","def _as_dict(self):\r\n values = self._dynamic_columns or {}\r\n for name, col in self._columns.items():\r\n values[name] = col.to_database(getattr(self, name, None))\r\n return values","def dictionary_formation():\r\n sales_data = {}\r\n with open('beer_data.csv', \"r\") as data_file:\r\n file_contents = csv.reader(data_file, delimiter=',')\r\n #Value of lines_read used as key value for each dictionary\r\n #in sales_data\r\n lines_read = 1\r\n for line in file_contents:\r\n if lines_read == 1:\r\n lines_read = lines_read + 1\r\n else:\r\n #Stores each column in row as key value in dictionary\r\n sales_data[str(lines_read)] = {\r\n \"invoice_number\": line[0],\r\n \"customer\": line[1],\r\n \"date_required\": line[2],\r\n \"recipe\": line[3],\r\n \"gyle_number\": line[4],\r\n \"quantity_ordered\": int(line[5])\r\n }\r\n lines_read = lines_read + 1\r\n data_file.close()\r\n return sales_data","def exportcsvsumdata(self, log):\r\n csvdata= None\r\n\r\n if (log):\r\n csvdata = ('%s\\t'%(log['CALLSIGN']))\r\n csvdata += ('%s\\t'%(log['OPERATORS']))\r\n csvdata += ('%s\\t'%(log['LOCATION']))\r\n csvdata += ('%d\\t'%(log['COUNT']))\r\n csvdata += ('%s\\t'%(log['NAMES']))\r\n if(log['LASTWORKED']): \r\n csvdata += ('%s/%s UTC'%(log['LASTWORKED'],\r\n log['LWTIME'])) \r\n\r\n return csvdata","def convert_to_dict(df):\n df = df[['ISO3166A2',\n 'ISOen_ro_name',\n 'minlongitude',\n 'maxlongitude',\n 'minlatitude',\n 'maxlatitude',\n 'land_total',\n ]]\n df.ISOen_ro_name = df.ISOen_ro_name.str.strip()\n df.set_index('ISOen_ro_name', inplace=True)\n df.columns = ['country_tag',\n 'min_lon',\n 'max_lon',\n 'min_lat',\n 'max_lat',\n 'area',\n ]\n df.loc[:, \"country_tag\"] = df.loc[:, \"country_tag\"].str.lower()\n return {i[0]: i[1].to_dict() for i in df.iterrows()}","def write_to_csv(self, name_suffix = ''):\n f_path = os.path.join(self.root_dir, 'res' + name_suffix + '.csv')\n field_names = [] # the first field in CSV is 'obj_val'\n\n # put the keys in the cost, prim_var_change, dual_var_change and fea_conditions as field names if any\n for key in self.cost.keys():\n field_names.append(key)\n for key in self.cost_change.keys():\n field_names.append(key)\n for key in self.prim_var_change.keys():\n field_names.append(key)\n for key in self.dual_var_change.keys():\n field_names.append(key)\n for key in self.fea_conditions.keys():\n field_names.append(key)\n\n\tprint f_path\n\n with open(f_path, mode = 'wb') as csv_file: # open the file, if not exist, create it\n writer = csv.DictWriter(csv_file, fieldnames = field_names) # create a writer which maps the dictionaries onto output rows in CSV\n writer.writeheader() # write the field names to the header\n temp_dict = {} # create a temporary dict used to output rows\n row_max = self.get_iter_num() # get the max iters which indicates the number of rows in CSV\n print ('number of rows: ' + str(row_max))\n #print (field_names)\n for row in range(row_max + 1):\n temp_dict.clear() # clear all items\n start_idx = 0\n for i in range(len(self.cost)):\n field = field_names[start_idx + i]\n\t\t if row > len(self.cost[field]) - 1:\n\t\t\ttemp_dict[field] = ''\n\t\t else: temp_dict[field] = self.get_cost_val(field, row)\n\n start_idx = start_idx + len(self.cost) # the start pos of fields in field_names for prim_var_change\n for i in range(len(self.cost_change)): # for each cost_change\n field = field_names[start_idx + i]\n if row == 0: # for row 0 (iter 0), we will set '/' to the change of primal variables\n temp_dict[field] = '/'\n elif row > len(self.cost_change[field]) - 1:\n\t\t\t temp_dict[field] = ''\n\t\t else:\n temp_dict[field] = self.get_cost_change_value(field, row - 1)\n\n\n start_idx = start_idx + len(self.cost_change)\n for i in range(len(self.prim_var_change)): # for each prim_var_change\n field = field_names[start_idx + i]\n if row == 0: # for row 0 (iter 0), we will set '/' to the change of primal variables\n temp_dict[field] = '/'\n\t\t elif row > len(self.prim_var_change[field]) - 1:\n\t\t\ttemp_dict[field] = ''\n else:\n temp_dict[field] = self.get_prim_change_value(field, row - 1)\n\n start_idx = start_idx + len(self.prim_var_change) # go to the start pos of fields in field_names for dual_var_change\n for i in range(len(self.dual_var_change)): # for each dual_var_change\n field = field_names[start_idx + i]\n if row == 0: # for row 0 (iter 0), we will set '/' to the change of dual variables\n temp_dict[field] = '/'\n elif row > len(self.dual_var_change[field]) - 1:\n\t\t\ttemp_dict[field] = '' \n\t\t else:\n temp_dict[field] = self.get_dual_change_value(field, row - 1)\n\n start_idx = start_idx + len(self.dual_var_change) # go the the start pos of fields in field_names for fea_conditions\n for i in range(len(self.fea_conditions)): # for each fea_condition\n field = field_names[start_idx + i]\n\t\t if row > len(self.fea_conditions[field]) - 1:\n\t\t\ttemp_dict[field] = ''\n else: temp_dict[field] = self.get_fea_condition_value(field, row)\n\n writer.writerow(temp_dict)\n\n # we also save the value of primal values if not saved\n if not self.pdv_to_csv:\n self.save_last_prims()","def create_csv(d, f):\n csv_data = list()\n csv_head = [unicode(x) for x in f]\n\n for row in d:\n row_data = list()\n for field in f:\n fields = field.split('.')\n row_data.append(extract_dict(row, fields))\n csv_data.append(row_data)\n\n csv = (csv_head, csv_data)\n return csv","def df_to_json(complete_dataset, output_path, static_columns):\n megajson = {}\n\n static_columns = [\"continent\", \"location\"] + list(static_columns)\n\n complete_dataset = complete_dataset.dropna(axis=\"rows\", subset=[\"iso_code\"])\n\n for _, row in complete_dataset.iterrows():\n\n row_iso = row[\"iso_code\"]\n row_dict_static = row.drop(\"iso_code\")[static_columns].dropna().to_dict()\n row_dict_dynamic = row.drop(\"iso_code\").drop(static_columns).dropna().to_dict()\n\n if row_iso not in megajson:\n megajson[row_iso] = row_dict_static\n megajson[row_iso][\"data\"] = [row_dict_dynamic]\n else:\n megajson[row_iso][\"data\"].append(row_dict_dynamic)\n\n with open(output_path, \"w\") as file:\n file.write(json.dumps(megajson, indent=4))","def _write_csv(self):\n\n # add the label to the header\n if self.input_data.get_value(InputType.TIME_PERIOD) == 'all':\n self.header.append('Date')\n else:\n self.header.append('sample id')\n\n key_list = []\n\n for i, cube in enumerate(self.cube_list):\n if self.input_data.get_value(InputType.TIME_PERIOD) == 'all':\n self._write_sample_with_date(cube, i, key_list)\n else:\n self._write_sample(cube, i, key_list)\n\n output_data_file_path = self._get_full_file_name()\n self._write_data_dict(output_data_file_path, key_list)\n\n return [output_data_file_path]","def generate_csv_file(rows: Collection[dict[str, str]]):\n file = io.StringIO()\n csv_writer = csv.DictWriter(file, (\"label\", \"inventory\", \"collection\"))\n csv_writer.writeheader()\n csv_writer.writerows(rows)\n file.seek(0)\n return io.BytesIO(bytes(file.read(), encoding=\"utf-8\"))","def create_dataframe(dictionary_to_convert, cols):\n\n dataframe_converted = pd.DataFrame.from_dict(dictionary_to_convert, orient='index', columns = cols)\n dataframe_converted = dataframe_converted.reset_index()\n dataframe_converted = dataframe_converted.drop(columns=['index'])\n\n return dataframe_converted","def to_json(self):\n\t\treturn self._dataframe.reset_index().to_json(orient=\"records\")","def csv_to_dict(fp):\n import pandas as pd\n df = pd.read_csv(fp, index_col=0, header=None)\n d = df.to_dict(orient='index')\n d = {k: v.values() for k, v in d.iteritems()}\n return d","def _aggregated_df_to_dict(agg_df: DataFrameLike) -> Dict[str, float]:\n try:\n agg_dicts = agg_df.to_dict(orient='index')\n except TypeError:\n # pd.Series objects do not have to_dict method with orient kwarg\n # so cast to pd.DataFrame and transpose for correct shape\n agg_dicts = agg_df.to_frame().T.to_dict(orient='index')\n\n formatted_agg_dict = {\n f'{key}({idx})': val\n for idx, row in agg_dicts.items()\n for key, val in row.items()\n }\n return formatted_agg_dict","def as_dict(self) -> Dict[str, Any]:\n return {\n column_title: cell.value\n for column_title, cell in self.column_title_to_cell.items()\n }","def create_query_csv(self):\n\n self.query_df.to_csv(self.query_output_file)","def PostgresTableToDict(sqldef, conn):\n df = pd.read_sql(sqldef, con=conn)\n return dict(zip(df.iloc[:, 0], df.iloc[:, 1]))","def pandas_typecast(self) -> dict:\n res = {}\n for feat in self.data_features:\n res[feat.key] = ApiForm.typecast(feat.dtype)\n return res","def read_delsys_csv(filename: str) -> Dict[str, Dict[str, TimeSeries]]:\n # Check the number of rows to skip\n n_rows = 0\n with open(filename, 'r') as fid:\n while True:\n s = fid.readline()\n if s.startswith('X[s]'):\n break\n else:\n n_rows += 1\n\n # Open the CSV\n df = pd.read_csv(filename, skiprows=n_rows)\n\n # Create a TimeSeries for each signal since they all have different time\n # vectors\n n_signals = int(len(df.columns) / 2)\n\n emg = {}\n acc = {}\n gyro = {}\n mag = {}\n\n for i_signal in range(n_signals):\n time = df.iloc[:, i_signal * 2].to_numpy()\n name = df.columns[i_signal * 2 + 1]\n data = df.iloc[:, i_signal * 2 + 1].to_numpy()\n\n if ': Acc' in name:\n short_name = name\n ts = TimeSeries(time=time, data={short_name: data})\n acc[short_name] = ts\n elif ': Mag' in name:\n short_name = name\n ts = TimeSeries(time=time, data={short_name: data})\n mag[short_name] = ts\n elif ': Gyro' in name:\n short_name = name\n ts = TimeSeries(time=time, data={short_name: data})\n gyro[short_name] = ts\n elif ': EMG' in name:\n short_name = name.split(':')[0]\n ts = TimeSeries(time=time, data={short_name: data})\n emg[short_name] = ts\n\n return {'emg': emg, 'acc': acc, 'gyro': gyro, 'mag': mag}","def export_csv(header, data):\n with StringIO() as tmp:\n writer = csv.DictWriter(tmp, fieldnames=header)\n writer.writeheader()\n writer.writerows(data)\n data = tmp.getvalue()\n\n return data","def DictData(self):\n reader = csv.DictReader( open( self.file, \"rU\" ), dialect = \"excel\" )\n return reader","def as_dict(self) -> Dict[str, Any]:\n return {\n column_title: cell.get_value()\n for column_title, cell in self.column_title_to_cell.items()\n }","def write_csv(f, extract, fields=None):\n keys = fields if fields is not None else extract[0].keys()\n dict_writer = csv.DictWriter(f, keys, extrasaction='ignore')\n dict_writer.writeheader()\n dict_writer.writerows(extract)\n return f","def save_csv(connection, query, columns, name):\n\n try:\n df = pd.read_sql(query, connection, columns=columns)\n df.to_csv(name, index=False)\n except Exception as ex:\n print(type(ex))\n print(ex)","def stream_to_db(subreddit: str, \n df_dict: dict, \n db_path: str) -> None:\n file_lkps = {'posts': f\"reddit-{subreddit}-posts.csv\", \n 'comments': f\"reddit-{subreddit}-comments.csv\"}\n \n for _key in df_dict:\n if df_dict.get(_key) is None:\n pass\n full_path = f\"{db_path}/{file_lkps.get(_key)}\"\n df = df_dict.get(_key)\n df.to_csv(full_path, index=False, encoding='utf-8')\n logging.info(f\"Saved {_key} data for subreddit {subreddit} at {datetime.today()}\")\n \n return None","def each_to_csv(data, key, value):\n data.to_csv(\"camelot/clean/nrld_{}_{}.csv\".format(key, value), index=False)\n return data","def make_csv(file_of_data):\n with open(file_of_data, 'w') as f:\n writer = csv.writer(f)\n header = (\"Counter\", \"Date/time\", \"Latitude\", \"Longitude\", \"Temperature\", \"Humidity\")\n writer.writerow(header)","def read_stats_csv(filename):\n\n df_dict = {}\n df = pd.read_csv(filename, header=[0, 1, 2])\n\n # Check if End column data type is datetime - if so use start date as index, otherwise use file number;\n # Use start date as index - Note: df[\"End\"] is interpreted as a dataframe here not a series as in hdf5\n if df[\"End\"].dtypes.all() == pd.Timestamp:\n # Drop redundant columns\n if \"File Number\" in df.columns:\n df = df.drop(\"File Number\", axis=1, level=0)\n df = df.drop(\"End\", axis=1, level=0)\n df = df.set_index(df.columns[0])\n df.index.name = \"Date\"\n\n # Convert timestamps to datetime\n try:\n df.index = pd.to_datetime(df.index, format=\"%Y-%m-%d %H:%M:%S\")\n except:\n try:\n # Timestamp will likely be in local (UK) format if csv file has been subsequently edited and saved\n df.index = pd.to_datetime(df.index, format=\"%d/%m/%Y %H:%M\")\n except:\n raise\n # Use file number as index\n else:\n df = df.drop([\"Start\", \"End\"], axis=1, level=0)\n df = df.set_index(df.columns[0])\n df.index.name = \"File Number\"\n\n df.columns.rename([\"channels\", \"stats\", \"units\"], inplace=True)\n logger = filename.split(\"Statistics_\")[-1].split(\".\")[0]\n df_dict[logger] = df\n\n return df_dict","def create_dataframe(self):\n\n df = pd.DataFrame({'date': [],\n 'RUN': [],\n 'CLONE': [],\n 'GEN': pd.Series(0, index=[], dtype='int'),\n 'frame': pd.Series([], index=[], dtype='int'),\n 'time (ns)': [] }) # set the index\n df.set_index('date')\n print(df)\n\n # Save the DataFrame to disk\n\n ### create a file handle to store the data in (a dict-like) HDF5 format\n store = pd.HDFStore(self.dataframe_path)\n print(store)\n store.put('df', df)\n return store","def create_csv(self, type): \n if os.path.isfile(_path_finder('keydata','{0}_{1}.db'.format(\n self.keyword,type))):\n self.__db_init('{0}'.format(type))\n self.c.execute(\"SELECT MIN(date) FROM tweets\")\n mindate = self.c.fetchone()[0][0:10]\n self.c.execute(\"SELECT MAX(date) FROM tweets\")\n maxdate = self.c.fetchone()[0][0:10]\n start_date = datetime.datetime.strptime(mindate, '%Y-%m-%d')\n end_date = (datetime.datetime.strptime(maxdate, '%Y-%m-%d') + \n datetime.timedelta(days=1))\n \n def __date_range(start, end):\n for n in range((end - start).days):\n yield start + datetime.timedelta(days=n)\n \n def __db_to_list():\n for single_date in __date_range(start_date, end_date):\n d = \"\".join(['%',single_date.strftime(\"%Y-%m-%d\"),'%'])\n self.c.execute('''SELECT count(*) FROM tweets where \n date like('{0}')'''.format(d))\n yield [d[1:11], self.c.fetchone()[0]]\n \n path = _path_finder('keydata','{0}_{1}.csv'.format(\n self.keyword,type))\n if sys.version_info[0] < 3: #Python3 compatibility check\n infile = open(path, 'wb')\n else:\n infile = open(path, 'w', newline='', encoding='utf8')\n with infile as f:\n writer = csv.writer(f)\n writer.writerows(__db_to_list())\n self.conn.commit()\n self.conn.close()\n print('\\nReport has been created:')\n print(os.path.abspath(path))"],"string":"[\n \"def csv_to_dict(self):\\n log = logger.configure(\\\"default\\\")\\n try: \\n df = pd.read_csv(self.__csv_path)\\n except IOError as e:\\n # file not found\\n log.error('Could not import {}. Got error {}'.format(self.__csv_path, e))\\n raise \\n else:\\n cols = list(df.columns)\\n metafield_cols = [col for col in cols if 'metafields' in col]\\n if metafield_cols == [] or 'Handle' not in cols:\\n # relevant columns don't exist\\n log.error('{} does not contain `Handle` or `metafields` named columns'.format(self.__csv_path))\\n raise\\n else:\\n new_cols = ['Handle'] + metafield_cols\\n df = df[new_cols].set_index('Handle')\\n df = df[~df.index.duplicated(keep='first')]\\n return df.to_dict('index')\",\n \"def to_dict_records(df):\\r\\n return df.to_dict('records')\",\n \"def DataFrame_to_dict(df):\\n return df.to_dict(orient='records')\",\n \"def df_to_dict(df):\\n return df.to_dict('index')\",\n \"def to_dict(df):\\n\\t# Extract the names\\n\\tnames = list(df.columns.values)\\n\\t# Create sequence\\n\\tdf_list = []\\n\\t#Iterate over rows of data frame\\n\\tfor index, row in df.iterrows():\\n\\t\\t# Create temporary object\\n\\t\\ttemp_dict = {}\\n\\t\\t#Iterate over key names and add values\\n\\t\\tfor name in names:\\n\\t\\t\\ttemp_dict[name] = row[name]\\n\\t\\t# Append object to dictionary\\n\\t\\tdf_list.append(temp_dict.copy())\\n\\t\\n\\treturn df_list\",\n \"def write_dataframe(df: pd.DataFrame, filename: str) -> Dict:\\n # Ensure that the parent directory for the output file exists.\\n dirname = os.path.dirname(filename)\\n if dirname:\\n os.makedirs(dirname, exist_ok=True)\\n # Create a unique list of column names and a mapping from the new uniqye\\n # names to the original columns in the given data frame.\\n columns = list()\\n column_mapping = dict()\\n for colidx in range(len(df.columns)):\\n colname = 'COL{}'.format(colidx)\\n columns.append(colname)\\n column_mapping[colname] = df.columns[colidx]\\n # Write data frame to temporary CSV file.\\n df.to_csv(\\n filename,\\n header=columns,\\n index=False,\\n compression=None\\n )\\n # Return the created column mapping..\\n return column_mapping\",\n \"def create_cris_data_dict(df, filename, outdir):\\n json_dict = {}\\n for key in df.keys():\\n if key != 'participant_id':\\n json_dict[key] = {'Units': key.split()[-1]}\\n else:\\n json_dict[key] = {'Description': 'OpenNeuro ID of the subject.'}\\n with open(outdir.joinpath(filename + '.json'), \\\"w\\\") as f:\\n json.dump(json_dict, f, indent=4)\",\n \"def __create_csv(self):\\n with open(self.__csv_file_name, 'w', newline='', encoding='utf-8') as csv_file:\\n writer = csv.DictWriter(csv_file, fieldnames=self.__csv_fields, delimiter=';')\\n writer.writeheader()\",\n \"def dictify(df):\\n return {str(k): v for k, v in df.items()}\",\n \"def WriteDictToCSV(csv_file,csv_columns,dict_data):\\n with open('history.csv','wb') as csvfile:\\n w = csv.writer(csvfile)\\n w.writerows(dict_data.items())\\n\\n return\",\n \"def pandas_convert(self):\\n data = {}\\n\\n for names in self.data[0]:\\n col_values = []\\n\\n if names in objects:\\n for items in self.data[0][names]:\\n col_values = []\\n\\n col_name = names + \\\"_\\\" + items\\n\\n for i in range(len(self.data)):\\n col_values.append(self.data[i][names][items])\\n\\n data[col_name] = col_values\\n else:\\n for i in range(len(self.data)):\\n col_values.append(self.data[i][names])\\n \\n data[names] = col_values\\n\\n self.pandas_df = pd.DataFrame(data=data)\\n self.__clean_df()\\n\\n return self.pandas_df\",\n \"def exportDict (dict, nameOfFile):\\n dictTable = pd.DataFrame(dict)\\n dictTable.to_csv(nameOfFile)\",\n \"def to_dict_index(df):\\r\\n return df.to_dict('index')\",\n \"def make_dict(ECOnew,goodObj):\\n \\n ECO_new = pd.read_csv(ECOnew, delimiter='\\\\t')\\n \\n good_Obj = pd.read_csv(goodObj,header=None,names=['ECO_ID'])\\n #An array of the ECOIDs obtained from goodObj.txt\\n arr1 = good_Obj.ECO_ID.values\\n \\n #Creates a new dataframe where you grab all the columns that match ECOID\\n #in goodObj with ECOID in the formatted ECO catalog\\n ECOnew_goodObj = ECO_new.loc[ECO_new.ECOID.isin(arr1), : ]\\n \\n #Group by ECOID and filters\\n ECOnew_groups = ECOnew_goodObj.groupby(['ECOID','filters'])\\n #Retrieve keys to the group sorted by ECOID\\n ECOnew_keys = sorted(ECOnew_groups.groups.keys())\\n \\n return ECOnew_groups, ECOnew_keys, ECO_new\",\n \"def make_csv(data_dict):\\n fieldnames = ['name'] + data_dict.itervalues().next().keys()\\n with open('data.csv', 'w') as csvfile:\\n writer = csv.DictWriter(csvfile, fieldnames=fieldnames)\\n writer.writeheader()\\n for record in data_dict:\\n person = data_dict[record]\\n person['name'] = record\\n assert set(person.keys()) == set(fieldnames)\\n writer.writerow(person)\",\n \"def make_csv(data_dict):\\r\\n fieldnames = ['name'] + data_dict.itervalues().next().keys()\\r\\n with open('data.csv', 'w') as csvfile:\\r\\n writer = csv.DictWriter(csvfile, fieldnames=fieldnames)\\r\\n writer.writeheader()\\r\\n for record in data_dict:\\r\\n person = data_dict[record]\\r\\n person['name'] = record\\r\\n assert set(person.keys()) == set(fieldnames)\\r\\n writer.writerow(person)\",\n \"def build_export_row(self, data:dict) -> dict:\\n data = {k : v for k, v in data.items() if v is not np.nan}\\n for key, value in data.items():\\n if key in _column_types:\\n if _column_types[key] == list:\\n if isinstance(value, (list, set, tuple)):\\n data[key] = LIST_DELIMITER.join(value)\\n else:\\n data[key] = str(value)\\n elif _column_types[key] == bool:\\n try:\\n data[key] = bool(value)\\n except:\\n data[key] = False\\n else:\\n data[key] = str(value)\\n return data\",\n \"def convert_to_dict(data_frame: pd.DataFrame) -> List[Dict[str, Any]]:\\n type_conversion = {\\\"date\\\": str} if \\\"date\\\" in data_frame.columns else {}\\n return data_frame.replace({np.nan: None}).astype(type_conversion).to_dict(\\\"records\\\")\",\n \"def build_df_dict(df, geneDictCanon):\\n\\n\\toutDict = OrderedDict()\\n\\n\\tfor tr in geneDictCanon:\\n\\t\\toutDict[geneDictCanon[tr][0]] = df.iloc[geneDictCanon[tr][1][0]:geneDictCanon[tr][1][1]]\\n\\n\\treturn outDict\",\n \"def to_csv(self, filename, **csv_args):\\n dataframe = pd.DataFrame.from_dict(self.to_dict())\\n return dataframe.to_csv(filename, **csv_args)\",\n \"async def full_report():\\n return DF.to_dict(orient=\\\"records\\\")\",\n \"def write_csv(fd, data):\\n # df = pd.DataFrame.from_dict(data)\\n df = pd.io.json.json_normalize(data)\\n print(df.to_csv(index=False), file=fd)\",\n \"def tsvRowToDict(row):\\n return {col: getattr(row, col) for col in row._columns_}\",\n \"def make_dict(row):\\n return dict((key[0], value) for key, value in zip(colnames, row))\",\n \"def convert_data(df):\\n print(\\\"Converting history...\\\")\\n return [ dict(row) for i, row in df.iterrows() ]\",\n \"def load_data(self):\\n df = pandas.read_csv(self.path)\\n self.data_dict = df.to_dict(orient=\\\"list\\\")\\n return self.data_dict\",\n \"def store_csv(self):\\n\\n with open(self.filepath.with_suffix(\\\".csv\\\"), 'w',\\n newline='') as csvfile:\\n fieldnames = ['counter', 'timestamp', 'acceleration']\\n writer = DictWriter(csvfile, fieldnames=fieldnames)\\n\\n writer.writeheader()\\n writer.writerows(self.values)\",\n \"def df_to_dict(df, key_column, val_column):\\n xkey = df[key_column].tolist()\\n xval = df[val_column].tolist()\\n return dict(zip(xkey,xval))\",\n \"def creating_dict_from_csv(self) -> dict:\\n dictionary = {}\\n for row in self.__read_csv():\\n if dictionary.get(row[0]):\\n dictionary[row[0]].append((row[1], row[2]))\\n else:\\n dictionary[row[0]] = [(row[1], row[2])]\\n\\n for key, value in dictionary.items():\\n dictionary[key] = sorted(value, key=lambda x: x[1], reverse=True)\\n\\n return dictionary\",\n \"def prepare_CSV(self):\\n self.drop_columns()\\n self.rename_columns()\\n self.spilt_columns()\\n self.add_vehicle_id_column()\\n self.add_source_column()\\n self.add_timestamp_columns()\\n self.get_colour_columns()\\n self.clean_column_formats()\\n\\n # print(self.data.info())\\n # print(self.data.sample(10))\\n\\n return self.data\",\n \"def dataframe_to_dict(self, orient=\\\"dict\\\", into=dict, index=True): # noqa: PR01\\n return self.to_pandas().to_dict(orient, into, index)\",\n \"def write_to_csv(self, data_points):\\n keys = data_points[0].keys()\\n with open(self.report_path, 'w') as output_file:\\n dict_writer = csv.DictWriter(output_file, keys)\\n dict_writer.writeheader()\\n dict_writer.writerows(data_points)\",\n \"def to_df(self) -> Dict[str, Any]:\\n return {key: self.render_df(key) for key, value in self.__dict__.items()}\",\n \"def test_ingest_with_column_names_dict():\\n schema = {\\\"foo\\\": \\\"foo1\\\", \\\"bar\\\": \\\"bar2\\\"}\\n\\n data = [{\\\"foo\\\": 1, \\\"bar\\\": 2}, {\\\"foo\\\": 10, \\\"bar\\\": 20}]\\n\\n converted_data = client.ingest_data(data, schema)\\n assert converted_data.to_pydict() == {'foo1': [1, 10], 'bar2': [2, 20]}\",\n \"def dataframe():\\n headers = get_headers()\\n headers = {'headers': headers}\\n headers = pd.DataFrame.from_dict(headers, orient='index')\\n headers = headers.replace(r'\\\\n', ' ', regex=True)\\n headers = headers.replace(r'\\\\r', ' ', regex=True)\\n headers = headers.replace(r'\\\\t', ' ', regex=True)\\n headers = headers.replace(r'\\\\\\\\t', ' ', regex=True)\\n headers = headers.replace(r' ', ' ', regex=True)\\n headers = headers.replace(r' ', ' ', regex=True)\\n\\n paragraphs = get_paragraphs()\\n paragraphs = {'paragraphs': paragraphs}\\n paragraphs = pd.DataFrame.from_dict(paragraphs, orient='index')\\n paragraphs = paragraphs.replace(r'\\\\n', ' ', regex=True)\\n paragraphs = paragraphs.replace(r'\\\\r', ' ', regex=True)\\n paragraphs = paragraphs.replace(r'\\\\t', ' ', regex=True)\\n paragraphs = paragraphs.replace(r'\\\\\\\\t', ' ', regex=True)\\n paragraphs = paragraphs.replace(r' ', ' ', regex=True)\\n paragraphs = paragraphs.replace(r' ', ' ', regex=True)\\n\\n return headers.to_csv('headers.csv', index=False), paragraphs.to_csv('paragraphs.csv', index=False)\",\n \"def dictify_df(frame):\\n ret = {}\\n for row in frame.values:\\n cur_level = ret\\n for elem in row[:-2]:\\n if elem not in cur_level:\\n cur_level[elem] = {}\\n cur_level = cur_level[elem]\\n cur_level[row[-2]] = row[-1]\\n return ret\",\n \"def write_csv(data):\\n\\n with open('data.csv', 'w') as file:\\n writer = csv.DictWriter(file, fieldnames=data[0].keys())\\n writer.writeheader()\\n for row in data:\\n writer.writerow(row)\",\n \"def output_df(outdict, out_file):\\n\\tcols = ['#chrom', 'source', 'feature', 'chromStart', 'chromEnd', 'score', 'strand', 'frame', 'transcript_id']\\n\\tcolOut = ['#chrom', 'source', 'feature', 'start', 'end', 'score', 'strand', 'frame', 'transcript_id']\\n\\tgtfDF = pd.DataFrame(columns=cols)\\n\\n\\tfor trsp in outdict:\\n\\t\\tgtfDF = gtfDF.append(outdict[trsp], ignore_index=True)\\n\\t\\t\\n\\tgtfDF.columns = colOut\\n\\t# print gtfDF.head()\\n\\tgtfDF.to_csv(out_file, compression='gzip', sep='\\\\t', index=False)\",\n \"def csv_dict_writer(path, fieldnames, data):\\n with open(path, \\\"wb\\\") as out_file:\\n writer = csv.DictWriter(out_file, delimiter=',', fieldnames=fieldnames)\\n writer.writeheader()\\n for row in data:\\n writer.writerow(row)\",\n \"def isolate_df_data(self, dict_df):\\n\\t\\treturn_dict_df = {}\\n\\t\\tfor k,v in dict_df.items():\\n\\t\\t\\tnew_df = pd.DataFrame(index=v.index)\\n\\t\\t\\tnew_df[k + ' Value'] = v['value']\\n\\t\\t\\treturn_dict_df[k] = new_df\\n\\t\\treturn return_dict_df\",\n \"def gen_dict():\\n lines = [line for line in csv.reader(open(__ppath__ + \\\"/data/occupations.csv\\\"))] # uses a csv.reader to parse the file, converts the generic iterable to a list\\n lines = [(line[0],float(line[1])) for line in lines[1:-2]]# removes the column names and \\\"Total\\\" row, re-expresses as a list of tuples to enable dictionary conversion\\n lines.append((\\\"Unemployed\\\",0.2)) # accounts for missing 0.2% of jobs\\n return dict(lines) # converts to dictionary\",\n \"def df2dict(df):\\n attr_list = list(df.columns)\\n res_dict = {}\\n for attr in attr_list:\\n res_dict[attr.lower()] = list(df[attr])\\n\\n return res_dict\",\n \"def write_to_csv(path,data_dict):\\n\\n\\n schema = [\\\"file_name\\\",\\\"family\\\",\\\"genus\\\",\\\"genus_confidence\\\",\\n \\\"species_1\\\",\\\"confidence_1\\\",\\\"hall_1\\\",\\n \\\"species_2\\\",\\\"confidence_2\\\",\\\"hall_2\\\",\\n \\\"species_3\\\",\\\"confidence_3\\\",\\\"hall_3\\\",\\n \\\"species_4\\\",\\\"confidence_4\\\",\\\"hall_4\\\",\\\"peaks\\\"]\\n\\n # if no file exists create a one and inform the user\\n if not os.path.exists(path):\\n print(\\\"creating new output file {}\\\".format(path))\\n with open(path, \\\"w\\\") as csv_file:\\n filewriter = csv.writer(csv_file, delimiter=\\\",\\\")\\n filewriter.writerow(schema)\\n\\n row = []\\n\\n row.append(data_dict[\\\"file_name\\\"])\\n row.append(data_dict[\\\"family\\\"])\\n \\n row.append(data_dict[\\\"genus_1\\\"])\\n row.append(data_dict[\\\"genus_confidence_1\\\"][:5])\\n \\n row.append(data_dict[\\\"species_1\\\"])\\n row.append(data_dict[\\\"confidence_1\\\"][:5])\\n row.append(data_dict[\\\"hall_1\\\"])\\n \\n row.append(data_dict[\\\"species_2\\\"])\\n row.append(data_dict[\\\"confidence_2\\\"][:5])\\n row.append(data_dict[\\\"hall_2\\\"])\\n\\n row.append(data_dict[\\\"species_3\\\"])\\n row.append(data_dict[\\\"confidence_3\\\"][:5])\\n row.append(data_dict[\\\"hall_3\\\"])\\n\\n row.append(data_dict[\\\"species_4\\\"])\\n row.append(data_dict[\\\"confidence_4\\\"][:5])\\n row.append(data_dict[\\\"hall_4\\\"])\\n\\n row.append(data_dict[\\\"peaks\\\"])\\n \\n with open(path, \\\"a\\\") as csv_file:\\n filewriter = csv.writer(csv_file, delimiter=\\\",\\\")\\n filewriter.writerow(row)\",\n \"def generate_csv(data: dict, cols: list, file_name: str) -> None:\\r\\n\\r\\n longest = 0\\r\\n for i in cols:\\r\\n if len(data[i]) > longest:\\r\\n longest = len(data[i])\\r\\n\\r\\n len_cols = []\\r\\n for i in cols:\\r\\n len_cols.append(str(i) + \\\" (\\\" + str(len(data[i])) + \\\")\\\")\\r\\n\\r\\n for i in cols:\\r\\n while len(data[i]) < longest:\\r\\n data[i].append(\\\"\\\")\\r\\n\\r\\n new_data = {}\\r\\n count = 0\\r\\n for i in len_cols:\\r\\n new_data[i] = data[str(count)]\\r\\n count += 1\\r\\n\\r\\n df = pd.DataFrame(data=new_data, columns=len_cols)\\r\\n print(df)\\r\\n\\r\\n df.to_csv(file_name, encoding='utf-8', index=False)\",\n \"def save_dataset_csv(self, path):\\n cols = list(self.data_dict.keys())\\n df = pd.DataFrame(self.data_dict, index=None, columns=cols)\\n df.to_csv(path, index=True)\",\n \"def csv_dict_writer(path, fieldnames, data):\\n with open(path, \\\"wb\\\") as out_file:\\n writer = csv.DictWriter(out_file, delimiter=',', fieldnames=fieldnames)\\n writer.writeheader()\\n for row in data:\\n writer.writerow(row)\",\n \"def build_dataset_by_df(file_path):\\n data = pd.read_csv(file_path, sep=',')\\n print(data.head())\\n\\n user_ids = data['CustomerID'].unique()\\n prod_ids = data['StockCode'].unique()\\n user_num = len(user_ids)\\n product_num = len(prod_ids)\\n product_lists_by_user = dict(data.groupby('CustomerID')['StockCode'].apply(list)).values()\\n product_to_desc = dict(data.groupby('StockCode')['Description'].apply(set))\\n dictionary = {}\\n for id in prod_ids:\\n dictionary[id] = len(dictionary)\\n reversed_dictionary = dict(zip(dictionary.values(), dictionary.keys()))\\n return product_lists_by_user, product_to_desc, dictionary, reversed_dictionary, user_num, product_num\",\n \"def build_data_frame(sql_dict, names):\\n\\td = defaultdict(list)\\n\\ti = 0\\n\\tfor category_data in sql_dict:\\n\\t\\tfor attribute_list in category_data:\\n\\t\\t\\tfor attribute in category_data[attribute_list]:\\n\\t\\t\\t\\td[attribute_list].append(attribute)\\n\\t\\td[\\\"category\\\"].extend([names[i]]* len(category_data[attribute_list]) )\\n\\t\\td[\\\"fct\\\"].extend([i] * len(category_data[attribute_list]) )\\n\\t\\ti += 1\\n\\t#Translation into R objects\\n\\tfor attribute_list in d:\\n\\t\\tif attribute_list == \\\"category\\\":\\n\\t\\t\\td[attribute_list] = robjects.StrVector(d[attribute_list])\\n\\t\\telif attribute_list in WATCH_FOR_MISSING_VALUES:\\n\\t\\t\\ttmp = robjects.FloatVector(range(len(d[attribute_list])))\\n\\t\\t\\tfor i in range(len(d[attribute_list])):\\n\\t\\t\\t\\tif d[attribute_list][i] == \\\"NA\\\":\\n\\t\\t\\t\\t\\ttmp.rx[i] = robjects.NA_Real\\n\\t\\t\\t\\telse:\\n\\t\\t\\t\\t\\ttmp.rx[i] = d[attribute_list][i]\\n\\t\\t\\td[attribute_list] = tmp\\n\\t\\telse:\\n\\t\\t\\tif attribute_list == \\\"num_sections\\\":\\n\\t\\t\\t\\ttmp = []\\n\\t\\t\\t\\tfor element in d[attribute_list]:\\n\\t\\t\\t\\t\\tif element == None:\\n\\t\\t\\t\\t\\t\\ttmp.append(1)\\n\\t\\t\\t\\t\\telse:\\n\\t\\t\\t\\t\\t\\ttmp.append(float(element))\\n\\t\\t\\t\\td[attribute_list] = robjects.FloatVector(tmp)\\n\\t\\t\\t\\tcontinue\\n\\t\\t\\telif attribute_list == \\\"speechiness\\\":\\n\\t\\t\\t\\ttmp = []\\n\\t\\t\\t\\tfor element in d[attribute_list]:\\n\\t\\t\\t\\t\\tif element == None:\\n\\t\\t\\t\\t\\t\\ttmp.append(1)\\n\\t\\t\\t\\t\\telse:\\n\\t\\t\\t\\t\\t\\ttmp.append(float(element))\\n\\t\\t\\t\\td[attribute_list] = robjects.FloatVector(tmp)\\n\\t\\t\\t\\tcontinue\\n\\t\\t\\t\\t \\n\\t\\t\\telse:\\n\\t\\t\\t\\td[attribute_list] = robjects.FloatVector(d[attribute_list])\\n\\tdata_frame = robjects.DataFrame(d)\\n\\treturn data_frame\",\n \"def write_csv(fn, toCSV):\\n keys = toCSV[0].keys()\\n with open(fn, 'w') as output_file:\\n dict_writer = csv.DictWriter(output_file, keys)\\n dict_writer.writeheader()\\n dict_writer.writerows(toCSV)\",\n \"def write_dataframe(self, path=os.getcwd(), name=\\\"Airbnb.csv\\\") -> None:\\n try:\\n df = pd.DataFrame(self.__collected_dic)\\n except ValueError:\\n print(\\\"ERROR! - Dictionary values are not the same length\\\")\\n return\\n\\n if not isinstance(name, str):\\n raise TypeError\\n if \\\".csv\\\" != name[-4:]:\\n name = f\\\"{name}.csv\\\"\\n df.to_csv(os.path.join(path, name), index=False)\\n print(f\\\"{name} file was succesfully written in {path}\\\")\",\n \"def make_ordered_to_dict(sorted_columns, include_timestamp):\\n od = OrderedDict()\\n for column in sorted_columns:\\n if include_timestamp:\\n value = (column.cell.value, column.cell.timestamp)\\n else:\\n value = column.cell.value\\n od[column.columnName] = value\\n return od\",\n \"def make_dict(data_for_dict): \\n \\n column_name_list = data_for_dict[0]\\n db_list = data_for_dict[1:]\\n \\n column_list1 = []\\n column_list2 = []\\n column_list3 = []\\n column_list4 = []\\n column_list5 = []\\n column_list6 = []\\n column_list7 = []\\n column_list8 = []\\n column_list9 = []\\n column_list10 = []\\n column_list11 = []\\n hmdb_dict = {}\\n for line in db_list:\\n my_string1 = '' \\n my_string2 = ''\\n my_string3 = ''\\n my_string4 = ''\\n my_string5 = ''\\n my_string6 = ''\\n my_string7 = ''\\n my_string8 = ''\\n my_string9 = ''\\n my_string10 = ''\\n my_string11 = ''\\n\\n my_string1 = line[0]\\n column_list1 += [my_string1]\\n my_string2 += line[1]\\n column_list2 += [my_string2]\\n my_string3 += line[2]\\n column_list3 += [my_string3]\\n my_string4 += line[3]\\n column_list4 += [my_string4]\\n my_string5 += line[4]\\n column_list5 += [my_string5]\\n my_string6 += line[5]\\n column_list6 += [my_string6]\\n my_string7 += line[6]\\n column_list7 += [my_string7]\\n my_string8 += line[7]\\n column_list8 += [my_string8]\\n my_string9 += line[8]\\n column_list9 += [my_string9]\\n my_string10 += line[9]\\n column_list10 += [my_string10]\\n my_string11 += line[10]\\n column_list11 += [my_string11] \\n \\n hmdb_dict[column_name_list[0]] = column_list1\\n hmdb_dict[column_name_list[1]] = column_list2\\n hmdb_dict[column_name_list[2]] = column_list3\\n hmdb_dict[column_name_list[3]] = column_list4\\n hmdb_dict[column_name_list[4]] = column_list5\\n hmdb_dict[column_name_list[5]] = column_list6\\n hmdb_dict[column_name_list[6]] = column_list7\\n hmdb_dict[column_name_list[7]] = column_list8\\n hmdb_dict[column_name_list[8]] = column_list9\\n hmdb_dict[column_name_list[9]] = column_list10\\n hmdb_dict[column_name_list[10]] = column_list11\\n \\n return (hmdb_dict)\",\n \"def chunk_to_dict(chunk):\\n csv_cols = chunk.keys()\\n return [dict(zip(csv_cols, v)) for v in chunk.values]\",\n \"def as_dict(self) :\\n\\n return dict(zip(self.cols, self.data))\",\n \"def write_csv(dictionary):\\n\\n with open(constant.METADATA_FILE, 'a', newline='') as file:\\n writer = csv.DictWriter(file, fieldnames=dictionary.keys())\\n try:\\n writer.writerow(dictionary)\\n except Exception:\\n writeable = {k: str(v).encode(\\\"utf-8\\\") for k, v in dictionary.items()}\\n writer.writerow(writeable)\",\n \"def _create_model_out_dictkeys():\\r\\n model_names = []\\r\\n result_keys = []\\r\\n for model_name in model_zoo._ModelZooUrls.CONFIG_PATH_TO_URL_SUFFIX.keys():\\r\\n try:\\r\\n print(model_name, \\\":\\\")\\r\\n select_model(model_name)\\r\\n result = get_features_by_image_path(\\\"./sample.jpg\\\")\\r\\n model_names.append(model_name)\\r\\n result_keys.append(list(result.keys()))\\r\\n except RuntimeError as t:\\r\\n print(t)\\r\\n\\r\\n pd.DataFrame(list(zip(model_names, result_keys))).to_csv(\\\"d2_model_out_dictkeys.csv\\\")\",\n \"def create_dataframe(self, dictionary_to_convert, cols):\\n\\n dataframe_converted = pd.DataFrame.from_dict(dictionary_to_convert, orient='index', columns = cols)\\n dataframe_converted = dataframe_converted.reset_index()\\n dataframe_converted = dataframe_converted.drop(columns=['index'])\\n\\n return dataframe_converted\",\n \"def write_csv(self, _dict, filename):\\n with open(filename, 'w') as f:\\n f.write('\\\"'+'\\\";\\\"'.join(_dict.keys())+'\\\"\\\\n')\\n for i in np.arange(len(_dict[list(_dict.keys())[0]])):\\n values = []\\n for col in _dict.keys():\\n try:\\n values.append(str(_dict[col][i]))\\n except IndexError as e:\\n # LSTM don't have first times available because of lacking history\\n pass\\n f.write(';'.join(values)+'\\\\n')\\n\\n logging.info('Wrote {}'.format(filename))\\n self._upload_to_bucket(filename, filename)\",\n \"def _export(self, *args, **kwargs) -> Tuple[pd.DataFrame, dict]:\\n return self.data, kwargs\",\n \"def save_json(df):\\n dict = {}\\n for row in df.iterrows():\\n dict[row[1]['Country']] = {'Region' : row[1]['Region'],\\n 'Pop. Density (per sq. mi.)' : row[1]['Pop. Density (per sq. mi.)'],\\n 'Infant mortality (per 1000 births)' : row[1]['Infant mortality (per 1000 births)'],\\n 'GDP ($ per capita) dollars' : row[1]['GDP ($ per capita) dollars']\\n }\\n\\n with open('data.json', 'w', encoding='utf8') as outfile:\\n data = json.dumps(dict, indent=4, sort_keys=False, separators=(',', ': '), ensure_ascii=False)\\n outfile.write(data)\",\n \"def dict_to_csv(filename, dictionary, paramdict=True):\\n if paramdict:\\n filename += '_params'\\n \\n with open(filename + '.csv', 'w') as f:\\n for key in dictionary.keys():\\n if type(dictionary[key]) == tuple:\\n f.write(\\\"%s,%s, %s \\\\n\\\"%(key, dictionary[key][0], dictionary[key][1]))\\n else:\\n f.write(\\\"%s,%s \\\\n\\\"%(key, dictionary[key]))\\n print('Saved %s.csv.'%filename)\",\n \"def as_dict(self):\\r\\n values = {}\\r\\n for name, col in self._columns.items():\\r\\n values[name] = col.to_database(getattr(self, name, None))\\r\\n return values\",\n \"def generate_new_row(forecast_date, target, target_end_date, \\n location, type, quantile, value):\\n new_row = {}\\n new_row[\\\"forecast_date\\\"] = forecast_date\\n new_row[\\\"target\\\"] = target\\n new_row[\\\"target_end_date\\\"] = target_end_date\\n new_row[\\\"location\\\"] = location\\n new_row[\\\"type\\\"] = type\\n new_row[\\\"quantile\\\"] = quantile\\n new_row[\\\"value\\\"] = value\\n return new_row\",\n \"def create_dataframe():\\r\\n\\r\\n df = pd.read_csv('data/data.csv', header=0)\\r\\n return df\",\n \"def make_data(rics: list, fields: list, name: str) -> None:\\n df = get_data(rics, fields)\\n df.index.name = 'ID'\\n to_csv(df, name)\",\n \"def generate_log_df(log_columns, log_values):\\n return pd.DataFrame(dict(zip(log_columns, log_values)), index=[0])\",\n \"def link_datadict_to_dataframe(row, dict):\\n d = {}\\n for k, v in dict.items():\\n d[k] = [row[c] if row[c] == row[c] else None for c in v[\\\"COLUMN\\\"]]\\n\\n if len(d[k]) == 1:\\n d[k] = d[k][0]\\n elif None in d[k]:\\n d[k] = None\\n\\n return d\",\n \"def _convert_csv_column_to_dict(csv_data, column):\\n results = dict()\\n\\n for row in csv_data:\\n key = row[0]\\n data = row[1:][column]\\n\\n if data:\\n if key not in results:\\n results[key] = data.strip() if data else \\\"\\\"\\n else:\\n # append multiple choice questions\\n results[key] += \\\"|{0}\\\".format(data.strip())\\n\\n return results\",\n \"def create_dataset(input_file_path, output_file_path):\\n col_index_map = {'user_id': 0, 'session_id': 1, 'timestamp': 2, 'step': 3, 'action_type': 4, 'reference': 5,\\n 'platform': 6, 'city': 7, 'device': 8,\\n 'current_filters': 9, 'impressions': 10, 'prices': 11}\\n flat_dict = dict()\\n with open(input_file_path, 'r') as csvFile:\\n reader = csv.reader(csvFile)\\n header = next(reader)\\n col_names = [col_name for col_name in col_index_map.keys()]\\n col_names.pop(0)\\n index = 0\\n for row in tqdm(reader):\\n if len(flat_dict) > 40000:\\n index += 1\\n with open(output_file_path + \\\"_\\\" + str(index) + \\\".json\\\", \\\"w\\\") as file:\\n json.dump(flat_dict, file)\\n print(\\\" JSON : \\\", index)\\n flat_dict = dict()\\n col_values = [row[col_index_map[c_n]] for c_n in col_names]\\n dict_for_each_row = dict(zip(col_names, col_values))\\n to_list = dict_for_each_row['impressions']\\n dict_for_each_row['impressions'] = to_list.split('|')\\n to_list = dict_for_each_row['prices']\\n dict_for_each_row['prices'] = to_list.split('|')\\n user_id = row[col_index_map['user_id']]\\n if user_id in flat_dict:\\n flat_dict[user_id].append(dict_for_each_row)\\n else:\\n flat_dict[user_id] = [dict_for_each_row]\\n\\n print(\\\"Output is Saved\\\")\",\n \"def readData(filename):\\r\\n data_d = {}\\r\\n with open(filename) as f:\\r\\n df = pd.read_csv(f, header=0, dtype='str',sep=';')\\r\\n df = df.loc[:, ~df.columns.str.contains('^Unnamed')]\\r\\n df_dict = df.to_dict(orient='index')\\r\\n for i,val in df_dict.iteritems(): \\r\\n clean_row = [(k, p.proc(v)) for (k, v) in val.iteritems()]\\r\\n row_id = val['line_nr']\\r\\n data_d[row_id] = dict(clean_row)\\r\\n return data_d\\r\\n return df\",\n \"def create_csvfile(options_path, recept_info,df):\\n df_csv = df.copy()\\n df_csv.index.names = ['Interacting positions']\\n\\n #Change dynX by full name of receptor, adding the dynid if there is more than a simulation for that receptor\\n df_csv.columns = df_csv.columns.map(lambda x: recept_info[x][13])\\n\\n #Sorting by ballesteros Id's (helixloop column) and clustering order\\n df_csv['Interacting positions'] = df_csv.index\\n df_csv['helixloop'] = df_csv['Interacting positions'].apply(lambda x: re.sub(r'^(\\\\d)x',r'\\\\g<1>0x',x)) \\n df_csv = df_csv.sort_values([\\\"helixloop\\\"])\\n\\n #Change jumplines by 'x' to avoid formatting problems\\n def new_index(cell):\\n cell = cell.replace('\\\\n\\\\n', ' ')\\n cell = cell.replace('\\\\n', 'x')\\n cell = cell.replace('xx', 'x')\\n return cell\\n\\n df_csv['Interacting positions'] = df_csv['Interacting positions'].apply(lambda x: new_index(x))\\n df_csv.index = df_csv['Interacting positions']\\n\\n #Drop columns\\n df_csv.drop(columns = ['helixloop','Interacting positions'], inplace = True)\\n\\n #Store dataframe as csv\\n df_csv.to_csv(path_or_buf = options_path+\\\"dataframe.csv\\\", float_format='%.1f')\",\n \"def write_to_csv(data_in_dictionary, csvfile):\\n with open(csvfile, 'wb') as f:\\n fieldnames = data_in_dictionary.keys()\\n writer = csv.DictWriter(f, fieldnames)\\n writer.writeheader()\\n writer.writerow(data_in_dictionary)\\n logger.info(\\\"Data written to file: \\\" + csvfile)\",\n \"def _as_dict(self):\\r\\n values = self._dynamic_columns or {}\\r\\n for name, col in self._columns.items():\\r\\n values[name] = col.to_database(getattr(self, name, None))\\r\\n return values\",\n \"def dictionary_formation():\\r\\n sales_data = {}\\r\\n with open('beer_data.csv', \\\"r\\\") as data_file:\\r\\n file_contents = csv.reader(data_file, delimiter=',')\\r\\n #Value of lines_read used as key value for each dictionary\\r\\n #in sales_data\\r\\n lines_read = 1\\r\\n for line in file_contents:\\r\\n if lines_read == 1:\\r\\n lines_read = lines_read + 1\\r\\n else:\\r\\n #Stores each column in row as key value in dictionary\\r\\n sales_data[str(lines_read)] = {\\r\\n \\\"invoice_number\\\": line[0],\\r\\n \\\"customer\\\": line[1],\\r\\n \\\"date_required\\\": line[2],\\r\\n \\\"recipe\\\": line[3],\\r\\n \\\"gyle_number\\\": line[4],\\r\\n \\\"quantity_ordered\\\": int(line[5])\\r\\n }\\r\\n lines_read = lines_read + 1\\r\\n data_file.close()\\r\\n return sales_data\",\n \"def exportcsvsumdata(self, log):\\r\\n csvdata= None\\r\\n\\r\\n if (log):\\r\\n csvdata = ('%s\\\\t'%(log['CALLSIGN']))\\r\\n csvdata += ('%s\\\\t'%(log['OPERATORS']))\\r\\n csvdata += ('%s\\\\t'%(log['LOCATION']))\\r\\n csvdata += ('%d\\\\t'%(log['COUNT']))\\r\\n csvdata += ('%s\\\\t'%(log['NAMES']))\\r\\n if(log['LASTWORKED']): \\r\\n csvdata += ('%s/%s UTC'%(log['LASTWORKED'],\\r\\n log['LWTIME'])) \\r\\n\\r\\n return csvdata\",\n \"def convert_to_dict(df):\\n df = df[['ISO3166A2',\\n 'ISOen_ro_name',\\n 'minlongitude',\\n 'maxlongitude',\\n 'minlatitude',\\n 'maxlatitude',\\n 'land_total',\\n ]]\\n df.ISOen_ro_name = df.ISOen_ro_name.str.strip()\\n df.set_index('ISOen_ro_name', inplace=True)\\n df.columns = ['country_tag',\\n 'min_lon',\\n 'max_lon',\\n 'min_lat',\\n 'max_lat',\\n 'area',\\n ]\\n df.loc[:, \\\"country_tag\\\"] = df.loc[:, \\\"country_tag\\\"].str.lower()\\n return {i[0]: i[1].to_dict() for i in df.iterrows()}\",\n \"def write_to_csv(self, name_suffix = ''):\\n f_path = os.path.join(self.root_dir, 'res' + name_suffix + '.csv')\\n field_names = [] # the first field in CSV is 'obj_val'\\n\\n # put the keys in the cost, prim_var_change, dual_var_change and fea_conditions as field names if any\\n for key in self.cost.keys():\\n field_names.append(key)\\n for key in self.cost_change.keys():\\n field_names.append(key)\\n for key in self.prim_var_change.keys():\\n field_names.append(key)\\n for key in self.dual_var_change.keys():\\n field_names.append(key)\\n for key in self.fea_conditions.keys():\\n field_names.append(key)\\n\\n\\tprint f_path\\n\\n with open(f_path, mode = 'wb') as csv_file: # open the file, if not exist, create it\\n writer = csv.DictWriter(csv_file, fieldnames = field_names) # create a writer which maps the dictionaries onto output rows in CSV\\n writer.writeheader() # write the field names to the header\\n temp_dict = {} # create a temporary dict used to output rows\\n row_max = self.get_iter_num() # get the max iters which indicates the number of rows in CSV\\n print ('number of rows: ' + str(row_max))\\n #print (field_names)\\n for row in range(row_max + 1):\\n temp_dict.clear() # clear all items\\n start_idx = 0\\n for i in range(len(self.cost)):\\n field = field_names[start_idx + i]\\n\\t\\t if row > len(self.cost[field]) - 1:\\n\\t\\t\\ttemp_dict[field] = ''\\n\\t\\t else: temp_dict[field] = self.get_cost_val(field, row)\\n\\n start_idx = start_idx + len(self.cost) # the start pos of fields in field_names for prim_var_change\\n for i in range(len(self.cost_change)): # for each cost_change\\n field = field_names[start_idx + i]\\n if row == 0: # for row 0 (iter 0), we will set '/' to the change of primal variables\\n temp_dict[field] = '/'\\n elif row > len(self.cost_change[field]) - 1:\\n\\t\\t\\t temp_dict[field] = ''\\n\\t\\t else:\\n temp_dict[field] = self.get_cost_change_value(field, row - 1)\\n\\n\\n start_idx = start_idx + len(self.cost_change)\\n for i in range(len(self.prim_var_change)): # for each prim_var_change\\n field = field_names[start_idx + i]\\n if row == 0: # for row 0 (iter 0), we will set '/' to the change of primal variables\\n temp_dict[field] = '/'\\n\\t\\t elif row > len(self.prim_var_change[field]) - 1:\\n\\t\\t\\ttemp_dict[field] = ''\\n else:\\n temp_dict[field] = self.get_prim_change_value(field, row - 1)\\n\\n start_idx = start_idx + len(self.prim_var_change) # go to the start pos of fields in field_names for dual_var_change\\n for i in range(len(self.dual_var_change)): # for each dual_var_change\\n field = field_names[start_idx + i]\\n if row == 0: # for row 0 (iter 0), we will set '/' to the change of dual variables\\n temp_dict[field] = '/'\\n elif row > len(self.dual_var_change[field]) - 1:\\n\\t\\t\\ttemp_dict[field] = '' \\n\\t\\t else:\\n temp_dict[field] = self.get_dual_change_value(field, row - 1)\\n\\n start_idx = start_idx + len(self.dual_var_change) # go the the start pos of fields in field_names for fea_conditions\\n for i in range(len(self.fea_conditions)): # for each fea_condition\\n field = field_names[start_idx + i]\\n\\t\\t if row > len(self.fea_conditions[field]) - 1:\\n\\t\\t\\ttemp_dict[field] = ''\\n else: temp_dict[field] = self.get_fea_condition_value(field, row)\\n\\n writer.writerow(temp_dict)\\n\\n # we also save the value of primal values if not saved\\n if not self.pdv_to_csv:\\n self.save_last_prims()\",\n \"def create_csv(d, f):\\n csv_data = list()\\n csv_head = [unicode(x) for x in f]\\n\\n for row in d:\\n row_data = list()\\n for field in f:\\n fields = field.split('.')\\n row_data.append(extract_dict(row, fields))\\n csv_data.append(row_data)\\n\\n csv = (csv_head, csv_data)\\n return csv\",\n \"def df_to_json(complete_dataset, output_path, static_columns):\\n megajson = {}\\n\\n static_columns = [\\\"continent\\\", \\\"location\\\"] + list(static_columns)\\n\\n complete_dataset = complete_dataset.dropna(axis=\\\"rows\\\", subset=[\\\"iso_code\\\"])\\n\\n for _, row in complete_dataset.iterrows():\\n\\n row_iso = row[\\\"iso_code\\\"]\\n row_dict_static = row.drop(\\\"iso_code\\\")[static_columns].dropna().to_dict()\\n row_dict_dynamic = row.drop(\\\"iso_code\\\").drop(static_columns).dropna().to_dict()\\n\\n if row_iso not in megajson:\\n megajson[row_iso] = row_dict_static\\n megajson[row_iso][\\\"data\\\"] = [row_dict_dynamic]\\n else:\\n megajson[row_iso][\\\"data\\\"].append(row_dict_dynamic)\\n\\n with open(output_path, \\\"w\\\") as file:\\n file.write(json.dumps(megajson, indent=4))\",\n \"def _write_csv(self):\\n\\n # add the label to the header\\n if self.input_data.get_value(InputType.TIME_PERIOD) == 'all':\\n self.header.append('Date')\\n else:\\n self.header.append('sample id')\\n\\n key_list = []\\n\\n for i, cube in enumerate(self.cube_list):\\n if self.input_data.get_value(InputType.TIME_PERIOD) == 'all':\\n self._write_sample_with_date(cube, i, key_list)\\n else:\\n self._write_sample(cube, i, key_list)\\n\\n output_data_file_path = self._get_full_file_name()\\n self._write_data_dict(output_data_file_path, key_list)\\n\\n return [output_data_file_path]\",\n \"def generate_csv_file(rows: Collection[dict[str, str]]):\\n file = io.StringIO()\\n csv_writer = csv.DictWriter(file, (\\\"label\\\", \\\"inventory\\\", \\\"collection\\\"))\\n csv_writer.writeheader()\\n csv_writer.writerows(rows)\\n file.seek(0)\\n return io.BytesIO(bytes(file.read(), encoding=\\\"utf-8\\\"))\",\n \"def create_dataframe(dictionary_to_convert, cols):\\n\\n dataframe_converted = pd.DataFrame.from_dict(dictionary_to_convert, orient='index', columns = cols)\\n dataframe_converted = dataframe_converted.reset_index()\\n dataframe_converted = dataframe_converted.drop(columns=['index'])\\n\\n return dataframe_converted\",\n \"def to_json(self):\\n\\t\\treturn self._dataframe.reset_index().to_json(orient=\\\"records\\\")\",\n \"def csv_to_dict(fp):\\n import pandas as pd\\n df = pd.read_csv(fp, index_col=0, header=None)\\n d = df.to_dict(orient='index')\\n d = {k: v.values() for k, v in d.iteritems()}\\n return d\",\n \"def _aggregated_df_to_dict(agg_df: DataFrameLike) -> Dict[str, float]:\\n try:\\n agg_dicts = agg_df.to_dict(orient='index')\\n except TypeError:\\n # pd.Series objects do not have to_dict method with orient kwarg\\n # so cast to pd.DataFrame and transpose for correct shape\\n agg_dicts = agg_df.to_frame().T.to_dict(orient='index')\\n\\n formatted_agg_dict = {\\n f'{key}({idx})': val\\n for idx, row in agg_dicts.items()\\n for key, val in row.items()\\n }\\n return formatted_agg_dict\",\n \"def as_dict(self) -> Dict[str, Any]:\\n return {\\n column_title: cell.value\\n for column_title, cell in self.column_title_to_cell.items()\\n }\",\n \"def create_query_csv(self):\\n\\n self.query_df.to_csv(self.query_output_file)\",\n \"def PostgresTableToDict(sqldef, conn):\\n df = pd.read_sql(sqldef, con=conn)\\n return dict(zip(df.iloc[:, 0], df.iloc[:, 1]))\",\n \"def pandas_typecast(self) -> dict:\\n res = {}\\n for feat in self.data_features:\\n res[feat.key] = ApiForm.typecast(feat.dtype)\\n return res\",\n \"def read_delsys_csv(filename: str) -> Dict[str, Dict[str, TimeSeries]]:\\n # Check the number of rows to skip\\n n_rows = 0\\n with open(filename, 'r') as fid:\\n while True:\\n s = fid.readline()\\n if s.startswith('X[s]'):\\n break\\n else:\\n n_rows += 1\\n\\n # Open the CSV\\n df = pd.read_csv(filename, skiprows=n_rows)\\n\\n # Create a TimeSeries for each signal since they all have different time\\n # vectors\\n n_signals = int(len(df.columns) / 2)\\n\\n emg = {}\\n acc = {}\\n gyro = {}\\n mag = {}\\n\\n for i_signal in range(n_signals):\\n time = df.iloc[:, i_signal * 2].to_numpy()\\n name = df.columns[i_signal * 2 + 1]\\n data = df.iloc[:, i_signal * 2 + 1].to_numpy()\\n\\n if ': Acc' in name:\\n short_name = name\\n ts = TimeSeries(time=time, data={short_name: data})\\n acc[short_name] = ts\\n elif ': Mag' in name:\\n short_name = name\\n ts = TimeSeries(time=time, data={short_name: data})\\n mag[short_name] = ts\\n elif ': Gyro' in name:\\n short_name = name\\n ts = TimeSeries(time=time, data={short_name: data})\\n gyro[short_name] = ts\\n elif ': EMG' in name:\\n short_name = name.split(':')[0]\\n ts = TimeSeries(time=time, data={short_name: data})\\n emg[short_name] = ts\\n\\n return {'emg': emg, 'acc': acc, 'gyro': gyro, 'mag': mag}\",\n \"def export_csv(header, data):\\n with StringIO() as tmp:\\n writer = csv.DictWriter(tmp, fieldnames=header)\\n writer.writeheader()\\n writer.writerows(data)\\n data = tmp.getvalue()\\n\\n return data\",\n \"def DictData(self):\\n reader = csv.DictReader( open( self.file, \\\"rU\\\" ), dialect = \\\"excel\\\" )\\n return reader\",\n \"def as_dict(self) -> Dict[str, Any]:\\n return {\\n column_title: cell.get_value()\\n for column_title, cell in self.column_title_to_cell.items()\\n }\",\n \"def write_csv(f, extract, fields=None):\\n keys = fields if fields is not None else extract[0].keys()\\n dict_writer = csv.DictWriter(f, keys, extrasaction='ignore')\\n dict_writer.writeheader()\\n dict_writer.writerows(extract)\\n return f\",\n \"def save_csv(connection, query, columns, name):\\n\\n try:\\n df = pd.read_sql(query, connection, columns=columns)\\n df.to_csv(name, index=False)\\n except Exception as ex:\\n print(type(ex))\\n print(ex)\",\n \"def stream_to_db(subreddit: str, \\n df_dict: dict, \\n db_path: str) -> None:\\n file_lkps = {'posts': f\\\"reddit-{subreddit}-posts.csv\\\", \\n 'comments': f\\\"reddit-{subreddit}-comments.csv\\\"}\\n \\n for _key in df_dict:\\n if df_dict.get(_key) is None:\\n pass\\n full_path = f\\\"{db_path}/{file_lkps.get(_key)}\\\"\\n df = df_dict.get(_key)\\n df.to_csv(full_path, index=False, encoding='utf-8')\\n logging.info(f\\\"Saved {_key} data for subreddit {subreddit} at {datetime.today()}\\\")\\n \\n return None\",\n \"def each_to_csv(data, key, value):\\n data.to_csv(\\\"camelot/clean/nrld_{}_{}.csv\\\".format(key, value), index=False)\\n return data\",\n \"def make_csv(file_of_data):\\n with open(file_of_data, 'w') as f:\\n writer = csv.writer(f)\\n header = (\\\"Counter\\\", \\\"Date/time\\\", \\\"Latitude\\\", \\\"Longitude\\\", \\\"Temperature\\\", \\\"Humidity\\\")\\n writer.writerow(header)\",\n \"def read_stats_csv(filename):\\n\\n df_dict = {}\\n df = pd.read_csv(filename, header=[0, 1, 2])\\n\\n # Check if End column data type is datetime - if so use start date as index, otherwise use file number;\\n # Use start date as index - Note: df[\\\"End\\\"] is interpreted as a dataframe here not a series as in hdf5\\n if df[\\\"End\\\"].dtypes.all() == pd.Timestamp:\\n # Drop redundant columns\\n if \\\"File Number\\\" in df.columns:\\n df = df.drop(\\\"File Number\\\", axis=1, level=0)\\n df = df.drop(\\\"End\\\", axis=1, level=0)\\n df = df.set_index(df.columns[0])\\n df.index.name = \\\"Date\\\"\\n\\n # Convert timestamps to datetime\\n try:\\n df.index = pd.to_datetime(df.index, format=\\\"%Y-%m-%d %H:%M:%S\\\")\\n except:\\n try:\\n # Timestamp will likely be in local (UK) format if csv file has been subsequently edited and saved\\n df.index = pd.to_datetime(df.index, format=\\\"%d/%m/%Y %H:%M\\\")\\n except:\\n raise\\n # Use file number as index\\n else:\\n df = df.drop([\\\"Start\\\", \\\"End\\\"], axis=1, level=0)\\n df = df.set_index(df.columns[0])\\n df.index.name = \\\"File Number\\\"\\n\\n df.columns.rename([\\\"channels\\\", \\\"stats\\\", \\\"units\\\"], inplace=True)\\n logger = filename.split(\\\"Statistics_\\\")[-1].split(\\\".\\\")[0]\\n df_dict[logger] = df\\n\\n return df_dict\",\n \"def create_dataframe(self):\\n\\n df = pd.DataFrame({'date': [],\\n 'RUN': [],\\n 'CLONE': [],\\n 'GEN': pd.Series(0, index=[], dtype='int'),\\n 'frame': pd.Series([], index=[], dtype='int'),\\n 'time (ns)': [] }) # set the index\\n df.set_index('date')\\n print(df)\\n\\n # Save the DataFrame to disk\\n\\n ### create a file handle to store the data in (a dict-like) HDF5 format\\n store = pd.HDFStore(self.dataframe_path)\\n print(store)\\n store.put('df', df)\\n return store\",\n \"def create_csv(self, type): \\n if os.path.isfile(_path_finder('keydata','{0}_{1}.db'.format(\\n self.keyword,type))):\\n self.__db_init('{0}'.format(type))\\n self.c.execute(\\\"SELECT MIN(date) FROM tweets\\\")\\n mindate = self.c.fetchone()[0][0:10]\\n self.c.execute(\\\"SELECT MAX(date) FROM tweets\\\")\\n maxdate = self.c.fetchone()[0][0:10]\\n start_date = datetime.datetime.strptime(mindate, '%Y-%m-%d')\\n end_date = (datetime.datetime.strptime(maxdate, '%Y-%m-%d') + \\n datetime.timedelta(days=1))\\n \\n def __date_range(start, end):\\n for n in range((end - start).days):\\n yield start + datetime.timedelta(days=n)\\n \\n def __db_to_list():\\n for single_date in __date_range(start_date, end_date):\\n d = \\\"\\\".join(['%',single_date.strftime(\\\"%Y-%m-%d\\\"),'%'])\\n self.c.execute('''SELECT count(*) FROM tweets where \\n date like('{0}')'''.format(d))\\n yield [d[1:11], self.c.fetchone()[0]]\\n \\n path = _path_finder('keydata','{0}_{1}.csv'.format(\\n self.keyword,type))\\n if sys.version_info[0] < 3: #Python3 compatibility check\\n infile = open(path, 'wb')\\n else:\\n infile = open(path, 'w', newline='', encoding='utf8')\\n with infile as f:\\n writer = csv.writer(f)\\n writer.writerows(__db_to_list())\\n self.conn.commit()\\n self.conn.close()\\n print('\\\\nReport has been created:')\\n print(os.path.abspath(path))\"\n]","isConversational":false},"negative_scores":{"kind":"list like","value":["0.68208885","0.6775985","0.65989804","0.6495224","0.6452339","0.642253","0.63207847","0.629015","0.6283443","0.62565273","0.6228677","0.62230223","0.620419","0.6197963","0.6147068","0.61124","0.60970604","0.6051975","0.60255706","0.60189426","0.6013112","0.60058635","0.5984326","0.5951467","0.59006345","0.5899104","0.5880743","0.58761513","0.58639425","0.5861759","0.5861191","0.5855694","0.58507824","0.58458096","0.58213186","0.5820627","0.57988566","0.5796882","0.5786933","0.57715964","0.57690144","0.5767987","0.57617337","0.57576007","0.57444936","0.57180643","0.57113004","0.5708054","0.57066464","0.5706451","0.5706039","0.57057047","0.5683445","0.56831557","0.56777215","0.567376","0.5652035","0.5639124","0.56321126","0.5629922","0.56284565","0.56249344","0.5624708","0.561719","0.56115144","0.56097835","0.5606495","0.55971366","0.5596355","0.55929065","0.55874616","0.5587184","0.5581996","0.55800927","0.5579521","0.55777586","0.55544615","0.5547011","0.554701","0.55431885","0.5537433","0.5534817","0.55021125","0.5499466","0.5496996","0.5495146","0.54905933","0.54863024","0.5472432","0.5463337","0.5458707","0.5455296","0.54552335","0.545504","0.54536897","0.54497355","0.54458135","0.5444228","0.5436945","0.54240847","0.54238445"],"string":"[\n \"0.68208885\",\n \"0.6775985\",\n \"0.65989804\",\n \"0.6495224\",\n \"0.6452339\",\n \"0.642253\",\n \"0.63207847\",\n \"0.629015\",\n \"0.6283443\",\n \"0.62565273\",\n \"0.6228677\",\n \"0.62230223\",\n \"0.620419\",\n \"0.6197963\",\n \"0.6147068\",\n \"0.61124\",\n \"0.60970604\",\n \"0.6051975\",\n \"0.60255706\",\n \"0.60189426\",\n \"0.6013112\",\n \"0.60058635\",\n \"0.5984326\",\n \"0.5951467\",\n \"0.59006345\",\n \"0.5899104\",\n \"0.5880743\",\n \"0.58761513\",\n \"0.58639425\",\n \"0.5861759\",\n \"0.5861191\",\n \"0.5855694\",\n \"0.58507824\",\n \"0.58458096\",\n \"0.58213186\",\n \"0.5820627\",\n \"0.57988566\",\n \"0.5796882\",\n \"0.5786933\",\n \"0.57715964\",\n \"0.57690144\",\n \"0.5767987\",\n \"0.57617337\",\n \"0.57576007\",\n \"0.57444936\",\n \"0.57180643\",\n \"0.57113004\",\n \"0.5708054\",\n \"0.57066464\",\n \"0.5706451\",\n \"0.5706039\",\n \"0.57057047\",\n \"0.5683445\",\n \"0.56831557\",\n \"0.56777215\",\n \"0.567376\",\n \"0.5652035\",\n \"0.5639124\",\n \"0.56321126\",\n \"0.5629922\",\n \"0.56284565\",\n \"0.56249344\",\n \"0.5624708\",\n \"0.561719\",\n \"0.56115144\",\n \"0.56097835\",\n \"0.5606495\",\n \"0.55971366\",\n \"0.5596355\",\n \"0.55929065\",\n \"0.55874616\",\n \"0.5587184\",\n \"0.5581996\",\n \"0.55800927\",\n \"0.5579521\",\n \"0.55777586\",\n \"0.55544615\",\n \"0.5547011\",\n \"0.554701\",\n \"0.55431885\",\n \"0.5537433\",\n \"0.5534817\",\n \"0.55021125\",\n \"0.5499466\",\n \"0.5496996\",\n \"0.5495146\",\n \"0.54905933\",\n \"0.54863024\",\n \"0.5472432\",\n \"0.5463337\",\n \"0.5458707\",\n \"0.5455296\",\n \"0.54552335\",\n \"0.545504\",\n \"0.54536897\",\n \"0.54497355\",\n \"0.54458135\",\n \"0.5444228\",\n \"0.5436945\",\n \"0.54240847\",\n \"0.54238445\"\n]","isConversational":false},"document_score":{"kind":"string","value":"0.0"},"document_rank":{"kind":"string","value":"-1"}}},{"rowIdx":93299,"cells":{"query":{"kind":"string","value":"creates a dictionary that can be ingested by pandas DataFrame to write to a csv file"},"document":{"kind":"string","value":"def create_advanced_experiment_dict_for_csv(experiment_dict, list_dict, algorithm_name, thresholds):\n experiment_dict_for_csv = {\n \"attributes\": ['TP', 'FP', 'TN', 'FN', 'TPR (Sensitivity)', 'TNR (Specificity)', 'Precision', 'Recall',\n 'F1 Measure', 'Accuracy']}\n for row_key in list_dict.values():\n for threshold in thresholds:\n column_name = algorithm_name + \"-\" + str(threshold)\n row_value = experiment_dict[str(threshold)][algorithm_name][row_key]\n if column_name in experiment_dict_for_csv.keys():\n experiment_dict_for_csv[column_name].append(row_value)\n else:\n experiment_dict_for_csv[column_name] = [row_value]\n return experiment_dict_for_csv"},"metadata":{"kind":"string","value":"{\n \"objective\": {\n \"self\": [],\n \"paired\": [],\n \"triplet\": [\n [\n \"query\",\n \"document\",\n \"negatives\"\n ]\n ]\n }\n}"},"negatives":{"kind":"list like","value":["def csv_to_dict(self):\n log = logger.configure(\"default\")\n try: \n df = pd.read_csv(self.__csv_path)\n except IOError as e:\n # file not found\n log.error('Could not import {}. Got error {}'.format(self.__csv_path, e))\n raise \n else:\n cols = list(df.columns)\n metafield_cols = [col for col in cols if 'metafields' in col]\n if metafield_cols == [] or 'Handle' not in cols:\n # relevant columns don't exist\n log.error('{} does not contain `Handle` or `metafields` named columns'.format(self.__csv_path))\n raise\n else:\n new_cols = ['Handle'] + metafield_cols\n df = df[new_cols].set_index('Handle')\n df = df[~df.index.duplicated(keep='first')]\n return df.to_dict('index')","def to_dict_records(df):\r\n return df.to_dict('records')","def DataFrame_to_dict(df):\n return df.to_dict(orient='records')","def df_to_dict(df):\n return df.to_dict('index')","def to_dict(df):\n\t# Extract the names\n\tnames = list(df.columns.values)\n\t# Create sequence\n\tdf_list = []\n\t#Iterate over rows of data frame\n\tfor index, row in df.iterrows():\n\t\t# Create temporary object\n\t\ttemp_dict = {}\n\t\t#Iterate over key names and add values\n\t\tfor name in names:\n\t\t\ttemp_dict[name] = row[name]\n\t\t# Append object to dictionary\n\t\tdf_list.append(temp_dict.copy())\n\t\n\treturn df_list","def write_dataframe(df: pd.DataFrame, filename: str) -> Dict:\n # Ensure that the parent directory for the output file exists.\n dirname = os.path.dirname(filename)\n if dirname:\n os.makedirs(dirname, exist_ok=True)\n # Create a unique list of column names and a mapping from the new uniqye\n # names to the original columns in the given data frame.\n columns = list()\n column_mapping = dict()\n for colidx in range(len(df.columns)):\n colname = 'COL{}'.format(colidx)\n columns.append(colname)\n column_mapping[colname] = df.columns[colidx]\n # Write data frame to temporary CSV file.\n df.to_csv(\n filename,\n header=columns,\n index=False,\n compression=None\n )\n # Return the created column mapping..\n return column_mapping","def create_cris_data_dict(df, filename, outdir):\n json_dict = {}\n for key in df.keys():\n if key != 'participant_id':\n json_dict[key] = {'Units': key.split()[-1]}\n else:\n json_dict[key] = {'Description': 'OpenNeuro ID of the subject.'}\n with open(outdir.joinpath(filename + '.json'), \"w\") as f:\n json.dump(json_dict, f, indent=4)","def __create_csv(self):\n with open(self.__csv_file_name, 'w', newline='', encoding='utf-8') as csv_file:\n writer = csv.DictWriter(csv_file, fieldnames=self.__csv_fields, delimiter=';')\n writer.writeheader()","def dictify(df):\n return {str(k): v for k, v in df.items()}","def WriteDictToCSV(csv_file,csv_columns,dict_data):\n with open('history.csv','wb') as csvfile:\n w = csv.writer(csvfile)\n w.writerows(dict_data.items())\n\n return","def pandas_convert(self):\n data = {}\n\n for names in self.data[0]:\n col_values = []\n\n if names in objects:\n for items in self.data[0][names]:\n col_values = []\n\n col_name = names + \"_\" + items\n\n for i in range(len(self.data)):\n col_values.append(self.data[i][names][items])\n\n data[col_name] = col_values\n else:\n for i in range(len(self.data)):\n col_values.append(self.data[i][names])\n \n data[names] = col_values\n\n self.pandas_df = pd.DataFrame(data=data)\n self.__clean_df()\n\n return self.pandas_df","def exportDict (dict, nameOfFile):\n dictTable = pd.DataFrame(dict)\n dictTable.to_csv(nameOfFile)","def to_dict_index(df):\r\n return df.to_dict('index')","def make_dict(ECOnew,goodObj):\n \n ECO_new = pd.read_csv(ECOnew, delimiter='\\t')\n \n good_Obj = pd.read_csv(goodObj,header=None,names=['ECO_ID'])\n #An array of the ECOIDs obtained from goodObj.txt\n arr1 = good_Obj.ECO_ID.values\n \n #Creates a new dataframe where you grab all the columns that match ECOID\n #in goodObj with ECOID in the formatted ECO catalog\n ECOnew_goodObj = ECO_new.loc[ECO_new.ECOID.isin(arr1), : ]\n \n #Group by ECOID and filters\n ECOnew_groups = ECOnew_goodObj.groupby(['ECOID','filters'])\n #Retrieve keys to the group sorted by ECOID\n ECOnew_keys = sorted(ECOnew_groups.groups.keys())\n \n return ECOnew_groups, ECOnew_keys, ECO_new","def make_csv(data_dict):\n fieldnames = ['name'] + data_dict.itervalues().next().keys()\n with open('data.csv', 'w') as csvfile:\n writer = csv.DictWriter(csvfile, fieldnames=fieldnames)\n writer.writeheader()\n for record in data_dict:\n person = data_dict[record]\n person['name'] = record\n assert set(person.keys()) == set(fieldnames)\n writer.writerow(person)","def make_csv(data_dict):\r\n fieldnames = ['name'] + data_dict.itervalues().next().keys()\r\n with open('data.csv', 'w') as csvfile:\r\n writer = csv.DictWriter(csvfile, fieldnames=fieldnames)\r\n writer.writeheader()\r\n for record in data_dict:\r\n person = data_dict[record]\r\n person['name'] = record\r\n assert set(person.keys()) == set(fieldnames)\r\n writer.writerow(person)","def build_export_row(self, data:dict) -> dict:\n data = {k : v for k, v in data.items() if v is not np.nan}\n for key, value in data.items():\n if key in _column_types:\n if _column_types[key] == list:\n if isinstance(value, (list, set, tuple)):\n data[key] = LIST_DELIMITER.join(value)\n else:\n data[key] = str(value)\n elif _column_types[key] == bool:\n try:\n data[key] = bool(value)\n except:\n data[key] = False\n else:\n data[key] = str(value)\n return data","def convert_to_dict(data_frame: pd.DataFrame) -> List[Dict[str, Any]]:\n type_conversion = {\"date\": str} if \"date\" in data_frame.columns else {}\n return data_frame.replace({np.nan: None}).astype(type_conversion).to_dict(\"records\")","def build_df_dict(df, geneDictCanon):\n\n\toutDict = OrderedDict()\n\n\tfor tr in geneDictCanon:\n\t\toutDict[geneDictCanon[tr][0]] = df.iloc[geneDictCanon[tr][1][0]:geneDictCanon[tr][1][1]]\n\n\treturn outDict","def to_csv(self, filename, **csv_args):\n dataframe = pd.DataFrame.from_dict(self.to_dict())\n return dataframe.to_csv(filename, **csv_args)","async def full_report():\n return DF.to_dict(orient=\"records\")","def write_csv(fd, data):\n # df = pd.DataFrame.from_dict(data)\n df = pd.io.json.json_normalize(data)\n print(df.to_csv(index=False), file=fd)","def tsvRowToDict(row):\n return {col: getattr(row, col) for col in row._columns_}","def make_dict(row):\n return dict((key[0], value) for key, value in zip(colnames, row))","def convert_data(df):\n print(\"Converting history...\")\n return [ dict(row) for i, row in df.iterrows() ]","def load_data(self):\n df = pandas.read_csv(self.path)\n self.data_dict = df.to_dict(orient=\"list\")\n return self.data_dict","def store_csv(self):\n\n with open(self.filepath.with_suffix(\".csv\"), 'w',\n newline='') as csvfile:\n fieldnames = ['counter', 'timestamp', 'acceleration']\n writer = DictWriter(csvfile, fieldnames=fieldnames)\n\n writer.writeheader()\n writer.writerows(self.values)","def df_to_dict(df, key_column, val_column):\n xkey = df[key_column].tolist()\n xval = df[val_column].tolist()\n return dict(zip(xkey,xval))","def creating_dict_from_csv(self) -> dict:\n dictionary = {}\n for row in self.__read_csv():\n if dictionary.get(row[0]):\n dictionary[row[0]].append((row[1], row[2]))\n else:\n dictionary[row[0]] = [(row[1], row[2])]\n\n for key, value in dictionary.items():\n dictionary[key] = sorted(value, key=lambda x: x[1], reverse=True)\n\n return dictionary","def prepare_CSV(self):\n self.drop_columns()\n self.rename_columns()\n self.spilt_columns()\n self.add_vehicle_id_column()\n self.add_source_column()\n self.add_timestamp_columns()\n self.get_colour_columns()\n self.clean_column_formats()\n\n # print(self.data.info())\n # print(self.data.sample(10))\n\n return self.data","def dataframe_to_dict(self, orient=\"dict\", into=dict, index=True): # noqa: PR01\n return self.to_pandas().to_dict(orient, into, index)","def write_to_csv(self, data_points):\n keys = data_points[0].keys()\n with open(self.report_path, 'w') as output_file:\n dict_writer = csv.DictWriter(output_file, keys)\n dict_writer.writeheader()\n dict_writer.writerows(data_points)","def to_df(self) -> Dict[str, Any]:\n return {key: self.render_df(key) for key, value in self.__dict__.items()}","def test_ingest_with_column_names_dict():\n schema = {\"foo\": \"foo1\", \"bar\": \"bar2\"}\n\n data = [{\"foo\": 1, \"bar\": 2}, {\"foo\": 10, \"bar\": 20}]\n\n converted_data = client.ingest_data(data, schema)\n assert converted_data.to_pydict() == {'foo1': [1, 10], 'bar2': [2, 20]}","def dictify_df(frame):\n ret = {}\n for row in frame.values:\n cur_level = ret\n for elem in row[:-2]:\n if elem not in cur_level:\n cur_level[elem] = {}\n cur_level = cur_level[elem]\n cur_level[row[-2]] = row[-1]\n return ret","def dataframe():\n headers = get_headers()\n headers = {'headers': headers}\n headers = pd.DataFrame.from_dict(headers, orient='index')\n headers = headers.replace(r'\\n', ' ', regex=True)\n headers = headers.replace(r'\\r', ' ', regex=True)\n headers = headers.replace(r'\\t', ' ', regex=True)\n headers = headers.replace(r'\\\\t', ' ', regex=True)\n headers = headers.replace(r' ', ' ', regex=True)\n headers = headers.replace(r' ', ' ', regex=True)\n\n paragraphs = get_paragraphs()\n paragraphs = {'paragraphs': paragraphs}\n paragraphs = pd.DataFrame.from_dict(paragraphs, orient='index')\n paragraphs = paragraphs.replace(r'\\n', ' ', regex=True)\n paragraphs = paragraphs.replace(r'\\r', ' ', regex=True)\n paragraphs = paragraphs.replace(r'\\t', ' ', regex=True)\n paragraphs = paragraphs.replace(r'\\\\t', ' ', regex=True)\n paragraphs = paragraphs.replace(r' ', ' ', regex=True)\n paragraphs = paragraphs.replace(r' ', ' ', regex=True)\n\n return headers.to_csv('headers.csv', index=False), paragraphs.to_csv('paragraphs.csv', index=False)","def write_csv(data):\n\n with open('data.csv', 'w') as file:\n writer = csv.DictWriter(file, fieldnames=data[0].keys())\n writer.writeheader()\n for row in data:\n writer.writerow(row)","def output_df(outdict, out_file):\n\tcols = ['#chrom', 'source', 'feature', 'chromStart', 'chromEnd', 'score', 'strand', 'frame', 'transcript_id']\n\tcolOut = ['#chrom', 'source', 'feature', 'start', 'end', 'score', 'strand', 'frame', 'transcript_id']\n\tgtfDF = pd.DataFrame(columns=cols)\n\n\tfor trsp in outdict:\n\t\tgtfDF = gtfDF.append(outdict[trsp], ignore_index=True)\n\t\t\n\tgtfDF.columns = colOut\n\t# print gtfDF.head()\n\tgtfDF.to_csv(out_file, compression='gzip', sep='\\t', index=False)","def csv_dict_writer(path, fieldnames, data):\n with open(path, \"wb\") as out_file:\n writer = csv.DictWriter(out_file, delimiter=',', fieldnames=fieldnames)\n writer.writeheader()\n for row in data:\n writer.writerow(row)","def isolate_df_data(self, dict_df):\n\t\treturn_dict_df = {}\n\t\tfor k,v in dict_df.items():\n\t\t\tnew_df = pd.DataFrame(index=v.index)\n\t\t\tnew_df[k + ' Value'] = v['value']\n\t\t\treturn_dict_df[k] = new_df\n\t\treturn return_dict_df","def gen_dict():\n lines = [line for line in csv.reader(open(__ppath__ + \"/data/occupations.csv\"))] # uses a csv.reader to parse the file, converts the generic iterable to a list\n lines = [(line[0],float(line[1])) for line in lines[1:-2]]# removes the column names and \"Total\" row, re-expresses as a list of tuples to enable dictionary conversion\n lines.append((\"Unemployed\",0.2)) # accounts for missing 0.2% of jobs\n return dict(lines) # converts to dictionary","def df2dict(df):\n attr_list = list(df.columns)\n res_dict = {}\n for attr in attr_list:\n res_dict[attr.lower()] = list(df[attr])\n\n return res_dict","def write_to_csv(path,data_dict):\n\n\n schema = [\"file_name\",\"family\",\"genus\",\"genus_confidence\",\n \"species_1\",\"confidence_1\",\"hall_1\",\n \"species_2\",\"confidence_2\",\"hall_2\",\n \"species_3\",\"confidence_3\",\"hall_3\",\n \"species_4\",\"confidence_4\",\"hall_4\",\"peaks\"]\n\n # if no file exists create a one and inform the user\n if not os.path.exists(path):\n print(\"creating new output file {}\".format(path))\n with open(path, \"w\") as csv_file:\n filewriter = csv.writer(csv_file, delimiter=\",\")\n filewriter.writerow(schema)\n\n row = []\n\n row.append(data_dict[\"file_name\"])\n row.append(data_dict[\"family\"])\n \n row.append(data_dict[\"genus_1\"])\n row.append(data_dict[\"genus_confidence_1\"][:5])\n \n row.append(data_dict[\"species_1\"])\n row.append(data_dict[\"confidence_1\"][:5])\n row.append(data_dict[\"hall_1\"])\n \n row.append(data_dict[\"species_2\"])\n row.append(data_dict[\"confidence_2\"][:5])\n row.append(data_dict[\"hall_2\"])\n\n row.append(data_dict[\"species_3\"])\n row.append(data_dict[\"confidence_3\"][:5])\n row.append(data_dict[\"hall_3\"])\n\n row.append(data_dict[\"species_4\"])\n row.append(data_dict[\"confidence_4\"][:5])\n row.append(data_dict[\"hall_4\"])\n\n row.append(data_dict[\"peaks\"])\n \n with open(path, \"a\") as csv_file:\n filewriter = csv.writer(csv_file, delimiter=\",\")\n filewriter.writerow(row)","def generate_csv(data: dict, cols: list, file_name: str) -> None:\r\n\r\n longest = 0\r\n for i in cols:\r\n if len(data[i]) > longest:\r\n longest = len(data[i])\r\n\r\n len_cols = []\r\n for i in cols:\r\n len_cols.append(str(i) + \" (\" + str(len(data[i])) + \")\")\r\n\r\n for i in cols:\r\n while len(data[i]) < longest:\r\n data[i].append(\"\")\r\n\r\n new_data = {}\r\n count = 0\r\n for i in len_cols:\r\n new_data[i] = data[str(count)]\r\n count += 1\r\n\r\n df = pd.DataFrame(data=new_data, columns=len_cols)\r\n print(df)\r\n\r\n df.to_csv(file_name, encoding='utf-8', index=False)","def save_dataset_csv(self, path):\n cols = list(self.data_dict.keys())\n df = pd.DataFrame(self.data_dict, index=None, columns=cols)\n df.to_csv(path, index=True)","def csv_dict_writer(path, fieldnames, data):\n with open(path, \"wb\") as out_file:\n writer = csv.DictWriter(out_file, delimiter=',', fieldnames=fieldnames)\n writer.writeheader()\n for row in data:\n writer.writerow(row)","def build_dataset_by_df(file_path):\n data = pd.read_csv(file_path, sep=',')\n print(data.head())\n\n user_ids = data['CustomerID'].unique()\n prod_ids = data['StockCode'].unique()\n user_num = len(user_ids)\n product_num = len(prod_ids)\n product_lists_by_user = dict(data.groupby('CustomerID')['StockCode'].apply(list)).values()\n product_to_desc = dict(data.groupby('StockCode')['Description'].apply(set))\n dictionary = {}\n for id in prod_ids:\n dictionary[id] = len(dictionary)\n reversed_dictionary = dict(zip(dictionary.values(), dictionary.keys()))\n return product_lists_by_user, product_to_desc, dictionary, reversed_dictionary, user_num, product_num","def build_data_frame(sql_dict, names):\n\td = defaultdict(list)\n\ti = 0\n\tfor category_data in sql_dict:\n\t\tfor attribute_list in category_data:\n\t\t\tfor attribute in category_data[attribute_list]:\n\t\t\t\td[attribute_list].append(attribute)\n\t\td[\"category\"].extend([names[i]]* len(category_data[attribute_list]) )\n\t\td[\"fct\"].extend([i] * len(category_data[attribute_list]) )\n\t\ti += 1\n\t#Translation into R objects\n\tfor attribute_list in d:\n\t\tif attribute_list == \"category\":\n\t\t\td[attribute_list] = robjects.StrVector(d[attribute_list])\n\t\telif attribute_list in WATCH_FOR_MISSING_VALUES:\n\t\t\ttmp = robjects.FloatVector(range(len(d[attribute_list])))\n\t\t\tfor i in range(len(d[attribute_list])):\n\t\t\t\tif d[attribute_list][i] == \"NA\":\n\t\t\t\t\ttmp.rx[i] = robjects.NA_Real\n\t\t\t\telse:\n\t\t\t\t\ttmp.rx[i] = d[attribute_list][i]\n\t\t\td[attribute_list] = tmp\n\t\telse:\n\t\t\tif attribute_list == \"num_sections\":\n\t\t\t\ttmp = []\n\t\t\t\tfor element in d[attribute_list]:\n\t\t\t\t\tif element == None:\n\t\t\t\t\t\ttmp.append(1)\n\t\t\t\t\telse:\n\t\t\t\t\t\ttmp.append(float(element))\n\t\t\t\td[attribute_list] = robjects.FloatVector(tmp)\n\t\t\t\tcontinue\n\t\t\telif attribute_list == \"speechiness\":\n\t\t\t\ttmp = []\n\t\t\t\tfor element in d[attribute_list]:\n\t\t\t\t\tif element == None:\n\t\t\t\t\t\ttmp.append(1)\n\t\t\t\t\telse:\n\t\t\t\t\t\ttmp.append(float(element))\n\t\t\t\td[attribute_list] = robjects.FloatVector(tmp)\n\t\t\t\tcontinue\n\t\t\t\t \n\t\t\telse:\n\t\t\t\td[attribute_list] = robjects.FloatVector(d[attribute_list])\n\tdata_frame = robjects.DataFrame(d)\n\treturn data_frame","def make_dict(data_for_dict): \n \n column_name_list = data_for_dict[0]\n db_list = data_for_dict[1:]\n \n column_list1 = []\n column_list2 = []\n column_list3 = []\n column_list4 = []\n column_list5 = []\n column_list6 = []\n column_list7 = []\n column_list8 = []\n column_list9 = []\n column_list10 = []\n column_list11 = []\n hmdb_dict = {}\n for line in db_list:\n my_string1 = '' \n my_string2 = ''\n my_string3 = ''\n my_string4 = ''\n my_string5 = ''\n my_string6 = ''\n my_string7 = ''\n my_string8 = ''\n my_string9 = ''\n my_string10 = ''\n my_string11 = ''\n\n my_string1 = line[0]\n column_list1 += [my_string1]\n my_string2 += line[1]\n column_list2 += [my_string2]\n my_string3 += line[2]\n column_list3 += [my_string3]\n my_string4 += line[3]\n column_list4 += [my_string4]\n my_string5 += line[4]\n column_list5 += [my_string5]\n my_string6 += line[5]\n column_list6 += [my_string6]\n my_string7 += line[6]\n column_list7 += [my_string7]\n my_string8 += line[7]\n column_list8 += [my_string8]\n my_string9 += line[8]\n column_list9 += [my_string9]\n my_string10 += line[9]\n column_list10 += [my_string10]\n my_string11 += line[10]\n column_list11 += [my_string11] \n \n hmdb_dict[column_name_list[0]] = column_list1\n hmdb_dict[column_name_list[1]] = column_list2\n hmdb_dict[column_name_list[2]] = column_list3\n hmdb_dict[column_name_list[3]] = column_list4\n hmdb_dict[column_name_list[4]] = column_list5\n hmdb_dict[column_name_list[5]] = column_list6\n hmdb_dict[column_name_list[6]] = column_list7\n hmdb_dict[column_name_list[7]] = column_list8\n hmdb_dict[column_name_list[8]] = column_list9\n hmdb_dict[column_name_list[9]] = column_list10\n hmdb_dict[column_name_list[10]] = column_list11\n \n return (hmdb_dict)","def make_ordered_to_dict(sorted_columns, include_timestamp):\n od = OrderedDict()\n for column in sorted_columns:\n if include_timestamp:\n value = (column.cell.value, column.cell.timestamp)\n else:\n value = column.cell.value\n od[column.columnName] = value\n return od","def write_dataframe(self, path=os.getcwd(), name=\"Airbnb.csv\") -> None:\n try:\n df = pd.DataFrame(self.__collected_dic)\n except ValueError:\n print(\"ERROR! - Dictionary values are not the same length\")\n return\n\n if not isinstance(name, str):\n raise TypeError\n if \".csv\" != name[-4:]:\n name = f\"{name}.csv\"\n df.to_csv(os.path.join(path, name), index=False)\n print(f\"{name} file was succesfully written in {path}\")","def write_csv(fn, toCSV):\n keys = toCSV[0].keys()\n with open(fn, 'w') as output_file:\n dict_writer = csv.DictWriter(output_file, keys)\n dict_writer.writeheader()\n dict_writer.writerows(toCSV)","def as_dict(self) :\n\n return dict(zip(self.cols, self.data))","def chunk_to_dict(chunk):\n csv_cols = chunk.keys()\n return [dict(zip(csv_cols, v)) for v in chunk.values]","def write_csv(dictionary):\n\n with open(constant.METADATA_FILE, 'a', newline='') as file:\n writer = csv.DictWriter(file, fieldnames=dictionary.keys())\n try:\n writer.writerow(dictionary)\n except Exception:\n writeable = {k: str(v).encode(\"utf-8\") for k, v in dictionary.items()}\n writer.writerow(writeable)","def _create_model_out_dictkeys():\r\n model_names = []\r\n result_keys = []\r\n for model_name in model_zoo._ModelZooUrls.CONFIG_PATH_TO_URL_SUFFIX.keys():\r\n try:\r\n print(model_name, \":\")\r\n select_model(model_name)\r\n result = get_features_by_image_path(\"./sample.jpg\")\r\n model_names.append(model_name)\r\n result_keys.append(list(result.keys()))\r\n except RuntimeError as t:\r\n print(t)\r\n\r\n pd.DataFrame(list(zip(model_names, result_keys))).to_csv(\"d2_model_out_dictkeys.csv\")","def create_dataframe(self, dictionary_to_convert, cols):\n\n dataframe_converted = pd.DataFrame.from_dict(dictionary_to_convert, orient='index', columns = cols)\n dataframe_converted = dataframe_converted.reset_index()\n dataframe_converted = dataframe_converted.drop(columns=['index'])\n\n return dataframe_converted","def write_csv(self, _dict, filename):\n with open(filename, 'w') as f:\n f.write('\"'+'\";\"'.join(_dict.keys())+'\"\\n')\n for i in np.arange(len(_dict[list(_dict.keys())[0]])):\n values = []\n for col in _dict.keys():\n try:\n values.append(str(_dict[col][i]))\n except IndexError as e:\n # LSTM don't have first times available because of lacking history\n pass\n f.write(';'.join(values)+'\\n')\n\n logging.info('Wrote {}'.format(filename))\n self._upload_to_bucket(filename, filename)","def _export(self, *args, **kwargs) -> Tuple[pd.DataFrame, dict]:\n return self.data, kwargs","def save_json(df):\n dict = {}\n for row in df.iterrows():\n dict[row[1]['Country']] = {'Region' : row[1]['Region'],\n 'Pop. Density (per sq. mi.)' : row[1]['Pop. Density (per sq. mi.)'],\n 'Infant mortality (per 1000 births)' : row[1]['Infant mortality (per 1000 births)'],\n 'GDP ($ per capita) dollars' : row[1]['GDP ($ per capita) dollars']\n }\n\n with open('data.json', 'w', encoding='utf8') as outfile:\n data = json.dumps(dict, indent=4, sort_keys=False, separators=(',', ': '), ensure_ascii=False)\n outfile.write(data)","def dict_to_csv(filename, dictionary, paramdict=True):\n if paramdict:\n filename += '_params'\n \n with open(filename + '.csv', 'w') as f:\n for key in dictionary.keys():\n if type(dictionary[key]) == tuple:\n f.write(\"%s,%s, %s \\n\"%(key, dictionary[key][0], dictionary[key][1]))\n else:\n f.write(\"%s,%s \\n\"%(key, dictionary[key]))\n print('Saved %s.csv.'%filename)","def generate_new_row(forecast_date, target, target_end_date, \n location, type, quantile, value):\n new_row = {}\n new_row[\"forecast_date\"] = forecast_date\n new_row[\"target\"] = target\n new_row[\"target_end_date\"] = target_end_date\n new_row[\"location\"] = location\n new_row[\"type\"] = type\n new_row[\"quantile\"] = quantile\n new_row[\"value\"] = value\n return new_row","def as_dict(self):\r\n values = {}\r\n for name, col in self._columns.items():\r\n values[name] = col.to_database(getattr(self, name, None))\r\n return values","def create_dataframe():\r\n\r\n df = pd.read_csv('data/data.csv', header=0)\r\n return df","def make_data(rics: list, fields: list, name: str) -> None:\n df = get_data(rics, fields)\n df.index.name = 'ID'\n to_csv(df, name)","def generate_log_df(log_columns, log_values):\n return pd.DataFrame(dict(zip(log_columns, log_values)), index=[0])","def link_datadict_to_dataframe(row, dict):\n d = {}\n for k, v in dict.items():\n d[k] = [row[c] if row[c] == row[c] else None for c in v[\"COLUMN\"]]\n\n if len(d[k]) == 1:\n d[k] = d[k][0]\n elif None in d[k]:\n d[k] = None\n\n return d","def _convert_csv_column_to_dict(csv_data, column):\n results = dict()\n\n for row in csv_data:\n key = row[0]\n data = row[1:][column]\n\n if data:\n if key not in results:\n results[key] = data.strip() if data else \"\"\n else:\n # append multiple choice questions\n results[key] += \"|{0}\".format(data.strip())\n\n return results","def create_dataset(input_file_path, output_file_path):\n col_index_map = {'user_id': 0, 'session_id': 1, 'timestamp': 2, 'step': 3, 'action_type': 4, 'reference': 5,\n 'platform': 6, 'city': 7, 'device': 8,\n 'current_filters': 9, 'impressions': 10, 'prices': 11}\n flat_dict = dict()\n with open(input_file_path, 'r') as csvFile:\n reader = csv.reader(csvFile)\n header = next(reader)\n col_names = [col_name for col_name in col_index_map.keys()]\n col_names.pop(0)\n index = 0\n for row in tqdm(reader):\n if len(flat_dict) > 40000:\n index += 1\n with open(output_file_path + \"_\" + str(index) + \".json\", \"w\") as file:\n json.dump(flat_dict, file)\n print(\" JSON : \", index)\n flat_dict = dict()\n col_values = [row[col_index_map[c_n]] for c_n in col_names]\n dict_for_each_row = dict(zip(col_names, col_values))\n to_list = dict_for_each_row['impressions']\n dict_for_each_row['impressions'] = to_list.split('|')\n to_list = dict_for_each_row['prices']\n dict_for_each_row['prices'] = to_list.split('|')\n user_id = row[col_index_map['user_id']]\n if user_id in flat_dict:\n flat_dict[user_id].append(dict_for_each_row)\n else:\n flat_dict[user_id] = [dict_for_each_row]\n\n print(\"Output is Saved\")","def readData(filename):\r\n data_d = {}\r\n with open(filename) as f:\r\n df = pd.read_csv(f, header=0, dtype='str',sep=';')\r\n df = df.loc[:, ~df.columns.str.contains('^Unnamed')]\r\n df_dict = df.to_dict(orient='index')\r\n for i,val in df_dict.iteritems(): \r\n clean_row = [(k, p.proc(v)) for (k, v) in val.iteritems()]\r\n row_id = val['line_nr']\r\n data_d[row_id] = dict(clean_row)\r\n return data_d\r\n return df","def create_csvfile(options_path, recept_info,df):\n df_csv = df.copy()\n df_csv.index.names = ['Interacting positions']\n\n #Change dynX by full name of receptor, adding the dynid if there is more than a simulation for that receptor\n df_csv.columns = df_csv.columns.map(lambda x: recept_info[x][13])\n\n #Sorting by ballesteros Id's (helixloop column) and clustering order\n df_csv['Interacting positions'] = df_csv.index\n df_csv['helixloop'] = df_csv['Interacting positions'].apply(lambda x: re.sub(r'^(\\d)x',r'\\g<1>0x',x)) \n df_csv = df_csv.sort_values([\"helixloop\"])\n\n #Change jumplines by 'x' to avoid formatting problems\n def new_index(cell):\n cell = cell.replace('\\n\\n', ' ')\n cell = cell.replace('\\n', 'x')\n cell = cell.replace('xx', 'x')\n return cell\n\n df_csv['Interacting positions'] = df_csv['Interacting positions'].apply(lambda x: new_index(x))\n df_csv.index = df_csv['Interacting positions']\n\n #Drop columns\n df_csv.drop(columns = ['helixloop','Interacting positions'], inplace = True)\n\n #Store dataframe as csv\n df_csv.to_csv(path_or_buf = options_path+\"dataframe.csv\", float_format='%.1f')","def write_to_csv(data_in_dictionary, csvfile):\n with open(csvfile, 'wb') as f:\n fieldnames = data_in_dictionary.keys()\n writer = csv.DictWriter(f, fieldnames)\n writer.writeheader()\n writer.writerow(data_in_dictionary)\n logger.info(\"Data written to file: \" + csvfile)","def _as_dict(self):\r\n values = self._dynamic_columns or {}\r\n for name, col in self._columns.items():\r\n values[name] = col.to_database(getattr(self, name, None))\r\n return values","def dictionary_formation():\r\n sales_data = {}\r\n with open('beer_data.csv', \"r\") as data_file:\r\n file_contents = csv.reader(data_file, delimiter=',')\r\n #Value of lines_read used as key value for each dictionary\r\n #in sales_data\r\n lines_read = 1\r\n for line in file_contents:\r\n if lines_read == 1:\r\n lines_read = lines_read + 1\r\n else:\r\n #Stores each column in row as key value in dictionary\r\n sales_data[str(lines_read)] = {\r\n \"invoice_number\": line[0],\r\n \"customer\": line[1],\r\n \"date_required\": line[2],\r\n \"recipe\": line[3],\r\n \"gyle_number\": line[4],\r\n \"quantity_ordered\": int(line[5])\r\n }\r\n lines_read = lines_read + 1\r\n data_file.close()\r\n return sales_data","def convert_to_dict(df):\n df = df[['ISO3166A2',\n 'ISOen_ro_name',\n 'minlongitude',\n 'maxlongitude',\n 'minlatitude',\n 'maxlatitude',\n 'land_total',\n ]]\n df.ISOen_ro_name = df.ISOen_ro_name.str.strip()\n df.set_index('ISOen_ro_name', inplace=True)\n df.columns = ['country_tag',\n 'min_lon',\n 'max_lon',\n 'min_lat',\n 'max_lat',\n 'area',\n ]\n df.loc[:, \"country_tag\"] = df.loc[:, \"country_tag\"].str.lower()\n return {i[0]: i[1].to_dict() for i in df.iterrows()}","def exportcsvsumdata(self, log):\r\n csvdata= None\r\n\r\n if (log):\r\n csvdata = ('%s\\t'%(log['CALLSIGN']))\r\n csvdata += ('%s\\t'%(log['OPERATORS']))\r\n csvdata += ('%s\\t'%(log['LOCATION']))\r\n csvdata += ('%d\\t'%(log['COUNT']))\r\n csvdata += ('%s\\t'%(log['NAMES']))\r\n if(log['LASTWORKED']): \r\n csvdata += ('%s/%s UTC'%(log['LASTWORKED'],\r\n log['LWTIME'])) \r\n\r\n return csvdata","def write_to_csv(self, name_suffix = ''):\n f_path = os.path.join(self.root_dir, 'res' + name_suffix + '.csv')\n field_names = [] # the first field in CSV is 'obj_val'\n\n # put the keys in the cost, prim_var_change, dual_var_change and fea_conditions as field names if any\n for key in self.cost.keys():\n field_names.append(key)\n for key in self.cost_change.keys():\n field_names.append(key)\n for key in self.prim_var_change.keys():\n field_names.append(key)\n for key in self.dual_var_change.keys():\n field_names.append(key)\n for key in self.fea_conditions.keys():\n field_names.append(key)\n\n\tprint f_path\n\n with open(f_path, mode = 'wb') as csv_file: # open the file, if not exist, create it\n writer = csv.DictWriter(csv_file, fieldnames = field_names) # create a writer which maps the dictionaries onto output rows in CSV\n writer.writeheader() # write the field names to the header\n temp_dict = {} # create a temporary dict used to output rows\n row_max = self.get_iter_num() # get the max iters which indicates the number of rows in CSV\n print ('number of rows: ' + str(row_max))\n #print (field_names)\n for row in range(row_max + 1):\n temp_dict.clear() # clear all items\n start_idx = 0\n for i in range(len(self.cost)):\n field = field_names[start_idx + i]\n\t\t if row > len(self.cost[field]) - 1:\n\t\t\ttemp_dict[field] = ''\n\t\t else: temp_dict[field] = self.get_cost_val(field, row)\n\n start_idx = start_idx + len(self.cost) # the start pos of fields in field_names for prim_var_change\n for i in range(len(self.cost_change)): # for each cost_change\n field = field_names[start_idx + i]\n if row == 0: # for row 0 (iter 0), we will set '/' to the change of primal variables\n temp_dict[field] = '/'\n elif row > len(self.cost_change[field]) - 1:\n\t\t\t temp_dict[field] = ''\n\t\t else:\n temp_dict[field] = self.get_cost_change_value(field, row - 1)\n\n\n start_idx = start_idx + len(self.cost_change)\n for i in range(len(self.prim_var_change)): # for each prim_var_change\n field = field_names[start_idx + i]\n if row == 0: # for row 0 (iter 0), we will set '/' to the change of primal variables\n temp_dict[field] = '/'\n\t\t elif row > len(self.prim_var_change[field]) - 1:\n\t\t\ttemp_dict[field] = ''\n else:\n temp_dict[field] = self.get_prim_change_value(field, row - 1)\n\n start_idx = start_idx + len(self.prim_var_change) # go to the start pos of fields in field_names for dual_var_change\n for i in range(len(self.dual_var_change)): # for each dual_var_change\n field = field_names[start_idx + i]\n if row == 0: # for row 0 (iter 0), we will set '/' to the change of dual variables\n temp_dict[field] = '/'\n elif row > len(self.dual_var_change[field]) - 1:\n\t\t\ttemp_dict[field] = '' \n\t\t else:\n temp_dict[field] = self.get_dual_change_value(field, row - 1)\n\n start_idx = start_idx + len(self.dual_var_change) # go the the start pos of fields in field_names for fea_conditions\n for i in range(len(self.fea_conditions)): # for each fea_condition\n field = field_names[start_idx + i]\n\t\t if row > len(self.fea_conditions[field]) - 1:\n\t\t\ttemp_dict[field] = ''\n else: temp_dict[field] = self.get_fea_condition_value(field, row)\n\n writer.writerow(temp_dict)\n\n # we also save the value of primal values if not saved\n if not self.pdv_to_csv:\n self.save_last_prims()","def df_to_json(complete_dataset, output_path, static_columns):\n megajson = {}\n\n static_columns = [\"continent\", \"location\"] + list(static_columns)\n\n complete_dataset = complete_dataset.dropna(axis=\"rows\", subset=[\"iso_code\"])\n\n for _, row in complete_dataset.iterrows():\n\n row_iso = row[\"iso_code\"]\n row_dict_static = row.drop(\"iso_code\")[static_columns].dropna().to_dict()\n row_dict_dynamic = row.drop(\"iso_code\").drop(static_columns).dropna().to_dict()\n\n if row_iso not in megajson:\n megajson[row_iso] = row_dict_static\n megajson[row_iso][\"data\"] = [row_dict_dynamic]\n else:\n megajson[row_iso][\"data\"].append(row_dict_dynamic)\n\n with open(output_path, \"w\") as file:\n file.write(json.dumps(megajson, indent=4))","def create_csv(d, f):\n csv_data = list()\n csv_head = [unicode(x) for x in f]\n\n for row in d:\n row_data = list()\n for field in f:\n fields = field.split('.')\n row_data.append(extract_dict(row, fields))\n csv_data.append(row_data)\n\n csv = (csv_head, csv_data)\n return csv","def _write_csv(self):\n\n # add the label to the header\n if self.input_data.get_value(InputType.TIME_PERIOD) == 'all':\n self.header.append('Date')\n else:\n self.header.append('sample id')\n\n key_list = []\n\n for i, cube in enumerate(self.cube_list):\n if self.input_data.get_value(InputType.TIME_PERIOD) == 'all':\n self._write_sample_with_date(cube, i, key_list)\n else:\n self._write_sample(cube, i, key_list)\n\n output_data_file_path = self._get_full_file_name()\n self._write_data_dict(output_data_file_path, key_list)\n\n return [output_data_file_path]","def generate_csv_file(rows: Collection[dict[str, str]]):\n file = io.StringIO()\n csv_writer = csv.DictWriter(file, (\"label\", \"inventory\", \"collection\"))\n csv_writer.writeheader()\n csv_writer.writerows(rows)\n file.seek(0)\n return io.BytesIO(bytes(file.read(), encoding=\"utf-8\"))","def create_dataframe(dictionary_to_convert, cols):\n\n dataframe_converted = pd.DataFrame.from_dict(dictionary_to_convert, orient='index', columns = cols)\n dataframe_converted = dataframe_converted.reset_index()\n dataframe_converted = dataframe_converted.drop(columns=['index'])\n\n return dataframe_converted","def to_json(self):\n\t\treturn self._dataframe.reset_index().to_json(orient=\"records\")","def csv_to_dict(fp):\n import pandas as pd\n df = pd.read_csv(fp, index_col=0, header=None)\n d = df.to_dict(orient='index')\n d = {k: v.values() for k, v in d.iteritems()}\n return d","def _aggregated_df_to_dict(agg_df: DataFrameLike) -> Dict[str, float]:\n try:\n agg_dicts = agg_df.to_dict(orient='index')\n except TypeError:\n # pd.Series objects do not have to_dict method with orient kwarg\n # so cast to pd.DataFrame and transpose for correct shape\n agg_dicts = agg_df.to_frame().T.to_dict(orient='index')\n\n formatted_agg_dict = {\n f'{key}({idx})': val\n for idx, row in agg_dicts.items()\n for key, val in row.items()\n }\n return formatted_agg_dict","def as_dict(self) -> Dict[str, Any]:\n return {\n column_title: cell.value\n for column_title, cell in self.column_title_to_cell.items()\n }","def create_query_csv(self):\n\n self.query_df.to_csv(self.query_output_file)","def PostgresTableToDict(sqldef, conn):\n df = pd.read_sql(sqldef, con=conn)\n return dict(zip(df.iloc[:, 0], df.iloc[:, 1]))","def pandas_typecast(self) -> dict:\n res = {}\n for feat in self.data_features:\n res[feat.key] = ApiForm.typecast(feat.dtype)\n return res","def read_delsys_csv(filename: str) -> Dict[str, Dict[str, TimeSeries]]:\n # Check the number of rows to skip\n n_rows = 0\n with open(filename, 'r') as fid:\n while True:\n s = fid.readline()\n if s.startswith('X[s]'):\n break\n else:\n n_rows += 1\n\n # Open the CSV\n df = pd.read_csv(filename, skiprows=n_rows)\n\n # Create a TimeSeries for each signal since they all have different time\n # vectors\n n_signals = int(len(df.columns) / 2)\n\n emg = {}\n acc = {}\n gyro = {}\n mag = {}\n\n for i_signal in range(n_signals):\n time = df.iloc[:, i_signal * 2].to_numpy()\n name = df.columns[i_signal * 2 + 1]\n data = df.iloc[:, i_signal * 2 + 1].to_numpy()\n\n if ': Acc' in name:\n short_name = name\n ts = TimeSeries(time=time, data={short_name: data})\n acc[short_name] = ts\n elif ': Mag' in name:\n short_name = name\n ts = TimeSeries(time=time, data={short_name: data})\n mag[short_name] = ts\n elif ': Gyro' in name:\n short_name = name\n ts = TimeSeries(time=time, data={short_name: data})\n gyro[short_name] = ts\n elif ': EMG' in name:\n short_name = name.split(':')[0]\n ts = TimeSeries(time=time, data={short_name: data})\n emg[short_name] = ts\n\n return {'emg': emg, 'acc': acc, 'gyro': gyro, 'mag': mag}","def as_dict(self) -> Dict[str, Any]:\n return {\n column_title: cell.get_value()\n for column_title, cell in self.column_title_to_cell.items()\n }","def DictData(self):\n reader = csv.DictReader( open( self.file, \"rU\" ), dialect = \"excel\" )\n return reader","def export_csv(header, data):\n with StringIO() as tmp:\n writer = csv.DictWriter(tmp, fieldnames=header)\n writer.writeheader()\n writer.writerows(data)\n data = tmp.getvalue()\n\n return data","def write_csv(f, extract, fields=None):\n keys = fields if fields is not None else extract[0].keys()\n dict_writer = csv.DictWriter(f, keys, extrasaction='ignore')\n dict_writer.writeheader()\n dict_writer.writerows(extract)\n return f","def save_csv(connection, query, columns, name):\n\n try:\n df = pd.read_sql(query, connection, columns=columns)\n df.to_csv(name, index=False)\n except Exception as ex:\n print(type(ex))\n print(ex)","def stream_to_db(subreddit: str, \n df_dict: dict, \n db_path: str) -> None:\n file_lkps = {'posts': f\"reddit-{subreddit}-posts.csv\", \n 'comments': f\"reddit-{subreddit}-comments.csv\"}\n \n for _key in df_dict:\n if df_dict.get(_key) is None:\n pass\n full_path = f\"{db_path}/{file_lkps.get(_key)}\"\n df = df_dict.get(_key)\n df.to_csv(full_path, index=False, encoding='utf-8')\n logging.info(f\"Saved {_key} data for subreddit {subreddit} at {datetime.today()}\")\n \n return None","def each_to_csv(data, key, value):\n data.to_csv(\"camelot/clean/nrld_{}_{}.csv\".format(key, value), index=False)\n return data","def make_csv(file_of_data):\n with open(file_of_data, 'w') as f:\n writer = csv.writer(f)\n header = (\"Counter\", \"Date/time\", \"Latitude\", \"Longitude\", \"Temperature\", \"Humidity\")\n writer.writerow(header)","def read_stats_csv(filename):\n\n df_dict = {}\n df = pd.read_csv(filename, header=[0, 1, 2])\n\n # Check if End column data type is datetime - if so use start date as index, otherwise use file number;\n # Use start date as index - Note: df[\"End\"] is interpreted as a dataframe here not a series as in hdf5\n if df[\"End\"].dtypes.all() == pd.Timestamp:\n # Drop redundant columns\n if \"File Number\" in df.columns:\n df = df.drop(\"File Number\", axis=1, level=0)\n df = df.drop(\"End\", axis=1, level=0)\n df = df.set_index(df.columns[0])\n df.index.name = \"Date\"\n\n # Convert timestamps to datetime\n try:\n df.index = pd.to_datetime(df.index, format=\"%Y-%m-%d %H:%M:%S\")\n except:\n try:\n # Timestamp will likely be in local (UK) format if csv file has been subsequently edited and saved\n df.index = pd.to_datetime(df.index, format=\"%d/%m/%Y %H:%M\")\n except:\n raise\n # Use file number as index\n else:\n df = df.drop([\"Start\", \"End\"], axis=1, level=0)\n df = df.set_index(df.columns[0])\n df.index.name = \"File Number\"\n\n df.columns.rename([\"channels\", \"stats\", \"units\"], inplace=True)\n logger = filename.split(\"Statistics_\")[-1].split(\".\")[0]\n df_dict[logger] = df\n\n return df_dict","def create_dataframe(self):\n\n df = pd.DataFrame({'date': [],\n 'RUN': [],\n 'CLONE': [],\n 'GEN': pd.Series(0, index=[], dtype='int'),\n 'frame': pd.Series([], index=[], dtype='int'),\n 'time (ns)': [] }) # set the index\n df.set_index('date')\n print(df)\n\n # Save the DataFrame to disk\n\n ### create a file handle to store the data in (a dict-like) HDF5 format\n store = pd.HDFStore(self.dataframe_path)\n print(store)\n store.put('df', df)\n return store","def create_csv(self, type): \n if os.path.isfile(_path_finder('keydata','{0}_{1}.db'.format(\n self.keyword,type))):\n self.__db_init('{0}'.format(type))\n self.c.execute(\"SELECT MIN(date) FROM tweets\")\n mindate = self.c.fetchone()[0][0:10]\n self.c.execute(\"SELECT MAX(date) FROM tweets\")\n maxdate = self.c.fetchone()[0][0:10]\n start_date = datetime.datetime.strptime(mindate, '%Y-%m-%d')\n end_date = (datetime.datetime.strptime(maxdate, '%Y-%m-%d') + \n datetime.timedelta(days=1))\n \n def __date_range(start, end):\n for n in range((end - start).days):\n yield start + datetime.timedelta(days=n)\n \n def __db_to_list():\n for single_date in __date_range(start_date, end_date):\n d = \"\".join(['%',single_date.strftime(\"%Y-%m-%d\"),'%'])\n self.c.execute('''SELECT count(*) FROM tweets where \n date like('{0}')'''.format(d))\n yield [d[1:11], self.c.fetchone()[0]]\n \n path = _path_finder('keydata','{0}_{1}.csv'.format(\n self.keyword,type))\n if sys.version_info[0] < 3: #Python3 compatibility check\n infile = open(path, 'wb')\n else:\n infile = open(path, 'w', newline='', encoding='utf8')\n with infile as f:\n writer = csv.writer(f)\n writer.writerows(__db_to_list())\n self.conn.commit()\n self.conn.close()\n print('\\nReport has been created:')\n print(os.path.abspath(path))"],"string":"[\n \"def csv_to_dict(self):\\n log = logger.configure(\\\"default\\\")\\n try: \\n df = pd.read_csv(self.__csv_path)\\n except IOError as e:\\n # file not found\\n log.error('Could not import {}. Got error {}'.format(self.__csv_path, e))\\n raise \\n else:\\n cols = list(df.columns)\\n metafield_cols = [col for col in cols if 'metafields' in col]\\n if metafield_cols == [] or 'Handle' not in cols:\\n # relevant columns don't exist\\n log.error('{} does not contain `Handle` or `metafields` named columns'.format(self.__csv_path))\\n raise\\n else:\\n new_cols = ['Handle'] + metafield_cols\\n df = df[new_cols].set_index('Handle')\\n df = df[~df.index.duplicated(keep='first')]\\n return df.to_dict('index')\",\n \"def to_dict_records(df):\\r\\n return df.to_dict('records')\",\n \"def DataFrame_to_dict(df):\\n return df.to_dict(orient='records')\",\n \"def df_to_dict(df):\\n return df.to_dict('index')\",\n \"def to_dict(df):\\n\\t# Extract the names\\n\\tnames = list(df.columns.values)\\n\\t# Create sequence\\n\\tdf_list = []\\n\\t#Iterate over rows of data frame\\n\\tfor index, row in df.iterrows():\\n\\t\\t# Create temporary object\\n\\t\\ttemp_dict = {}\\n\\t\\t#Iterate over key names and add values\\n\\t\\tfor name in names:\\n\\t\\t\\ttemp_dict[name] = row[name]\\n\\t\\t# Append object to dictionary\\n\\t\\tdf_list.append(temp_dict.copy())\\n\\t\\n\\treturn df_list\",\n \"def write_dataframe(df: pd.DataFrame, filename: str) -> Dict:\\n # Ensure that the parent directory for the output file exists.\\n dirname = os.path.dirname(filename)\\n if dirname:\\n os.makedirs(dirname, exist_ok=True)\\n # Create a unique list of column names and a mapping from the new uniqye\\n # names to the original columns in the given data frame.\\n columns = list()\\n column_mapping = dict()\\n for colidx in range(len(df.columns)):\\n colname = 'COL{}'.format(colidx)\\n columns.append(colname)\\n column_mapping[colname] = df.columns[colidx]\\n # Write data frame to temporary CSV file.\\n df.to_csv(\\n filename,\\n header=columns,\\n index=False,\\n compression=None\\n )\\n # Return the created column mapping..\\n return column_mapping\",\n \"def create_cris_data_dict(df, filename, outdir):\\n json_dict = {}\\n for key in df.keys():\\n if key != 'participant_id':\\n json_dict[key] = {'Units': key.split()[-1]}\\n else:\\n json_dict[key] = {'Description': 'OpenNeuro ID of the subject.'}\\n with open(outdir.joinpath(filename + '.json'), \\\"w\\\") as f:\\n json.dump(json_dict, f, indent=4)\",\n \"def __create_csv(self):\\n with open(self.__csv_file_name, 'w', newline='', encoding='utf-8') as csv_file:\\n writer = csv.DictWriter(csv_file, fieldnames=self.__csv_fields, delimiter=';')\\n writer.writeheader()\",\n \"def dictify(df):\\n return {str(k): v for k, v in df.items()}\",\n \"def WriteDictToCSV(csv_file,csv_columns,dict_data):\\n with open('history.csv','wb') as csvfile:\\n w = csv.writer(csvfile)\\n w.writerows(dict_data.items())\\n\\n return\",\n \"def pandas_convert(self):\\n data = {}\\n\\n for names in self.data[0]:\\n col_values = []\\n\\n if names in objects:\\n for items in self.data[0][names]:\\n col_values = []\\n\\n col_name = names + \\\"_\\\" + items\\n\\n for i in range(len(self.data)):\\n col_values.append(self.data[i][names][items])\\n\\n data[col_name] = col_values\\n else:\\n for i in range(len(self.data)):\\n col_values.append(self.data[i][names])\\n \\n data[names] = col_values\\n\\n self.pandas_df = pd.DataFrame(data=data)\\n self.__clean_df()\\n\\n return self.pandas_df\",\n \"def exportDict (dict, nameOfFile):\\n dictTable = pd.DataFrame(dict)\\n dictTable.to_csv(nameOfFile)\",\n \"def to_dict_index(df):\\r\\n return df.to_dict('index')\",\n \"def make_dict(ECOnew,goodObj):\\n \\n ECO_new = pd.read_csv(ECOnew, delimiter='\\\\t')\\n \\n good_Obj = pd.read_csv(goodObj,header=None,names=['ECO_ID'])\\n #An array of the ECOIDs obtained from goodObj.txt\\n arr1 = good_Obj.ECO_ID.values\\n \\n #Creates a new dataframe where you grab all the columns that match ECOID\\n #in goodObj with ECOID in the formatted ECO catalog\\n ECOnew_goodObj = ECO_new.loc[ECO_new.ECOID.isin(arr1), : ]\\n \\n #Group by ECOID and filters\\n ECOnew_groups = ECOnew_goodObj.groupby(['ECOID','filters'])\\n #Retrieve keys to the group sorted by ECOID\\n ECOnew_keys = sorted(ECOnew_groups.groups.keys())\\n \\n return ECOnew_groups, ECOnew_keys, ECO_new\",\n \"def make_csv(data_dict):\\n fieldnames = ['name'] + data_dict.itervalues().next().keys()\\n with open('data.csv', 'w') as csvfile:\\n writer = csv.DictWriter(csvfile, fieldnames=fieldnames)\\n writer.writeheader()\\n for record in data_dict:\\n person = data_dict[record]\\n person['name'] = record\\n assert set(person.keys()) == set(fieldnames)\\n writer.writerow(person)\",\n \"def make_csv(data_dict):\\r\\n fieldnames = ['name'] + data_dict.itervalues().next().keys()\\r\\n with open('data.csv', 'w') as csvfile:\\r\\n writer = csv.DictWriter(csvfile, fieldnames=fieldnames)\\r\\n writer.writeheader()\\r\\n for record in data_dict:\\r\\n person = data_dict[record]\\r\\n person['name'] = record\\r\\n assert set(person.keys()) == set(fieldnames)\\r\\n writer.writerow(person)\",\n \"def build_export_row(self, data:dict) -> dict:\\n data = {k : v for k, v in data.items() if v is not np.nan}\\n for key, value in data.items():\\n if key in _column_types:\\n if _column_types[key] == list:\\n if isinstance(value, (list, set, tuple)):\\n data[key] = LIST_DELIMITER.join(value)\\n else:\\n data[key] = str(value)\\n elif _column_types[key] == bool:\\n try:\\n data[key] = bool(value)\\n except:\\n data[key] = False\\n else:\\n data[key] = str(value)\\n return data\",\n \"def convert_to_dict(data_frame: pd.DataFrame) -> List[Dict[str, Any]]:\\n type_conversion = {\\\"date\\\": str} if \\\"date\\\" in data_frame.columns else {}\\n return data_frame.replace({np.nan: None}).astype(type_conversion).to_dict(\\\"records\\\")\",\n \"def build_df_dict(df, geneDictCanon):\\n\\n\\toutDict = OrderedDict()\\n\\n\\tfor tr in geneDictCanon:\\n\\t\\toutDict[geneDictCanon[tr][0]] = df.iloc[geneDictCanon[tr][1][0]:geneDictCanon[tr][1][1]]\\n\\n\\treturn outDict\",\n \"def to_csv(self, filename, **csv_args):\\n dataframe = pd.DataFrame.from_dict(self.to_dict())\\n return dataframe.to_csv(filename, **csv_args)\",\n \"async def full_report():\\n return DF.to_dict(orient=\\\"records\\\")\",\n \"def write_csv(fd, data):\\n # df = pd.DataFrame.from_dict(data)\\n df = pd.io.json.json_normalize(data)\\n print(df.to_csv(index=False), file=fd)\",\n \"def tsvRowToDict(row):\\n return {col: getattr(row, col) for col in row._columns_}\",\n \"def make_dict(row):\\n return dict((key[0], value) for key, value in zip(colnames, row))\",\n \"def convert_data(df):\\n print(\\\"Converting history...\\\")\\n return [ dict(row) for i, row in df.iterrows() ]\",\n \"def load_data(self):\\n df = pandas.read_csv(self.path)\\n self.data_dict = df.to_dict(orient=\\\"list\\\")\\n return self.data_dict\",\n \"def store_csv(self):\\n\\n with open(self.filepath.with_suffix(\\\".csv\\\"), 'w',\\n newline='') as csvfile:\\n fieldnames = ['counter', 'timestamp', 'acceleration']\\n writer = DictWriter(csvfile, fieldnames=fieldnames)\\n\\n writer.writeheader()\\n writer.writerows(self.values)\",\n \"def df_to_dict(df, key_column, val_column):\\n xkey = df[key_column].tolist()\\n xval = df[val_column].tolist()\\n return dict(zip(xkey,xval))\",\n \"def creating_dict_from_csv(self) -> dict:\\n dictionary = {}\\n for row in self.__read_csv():\\n if dictionary.get(row[0]):\\n dictionary[row[0]].append((row[1], row[2]))\\n else:\\n dictionary[row[0]] = [(row[1], row[2])]\\n\\n for key, value in dictionary.items():\\n dictionary[key] = sorted(value, key=lambda x: x[1], reverse=True)\\n\\n return dictionary\",\n \"def prepare_CSV(self):\\n self.drop_columns()\\n self.rename_columns()\\n self.spilt_columns()\\n self.add_vehicle_id_column()\\n self.add_source_column()\\n self.add_timestamp_columns()\\n self.get_colour_columns()\\n self.clean_column_formats()\\n\\n # print(self.data.info())\\n # print(self.data.sample(10))\\n\\n return self.data\",\n \"def dataframe_to_dict(self, orient=\\\"dict\\\", into=dict, index=True): # noqa: PR01\\n return self.to_pandas().to_dict(orient, into, index)\",\n \"def write_to_csv(self, data_points):\\n keys = data_points[0].keys()\\n with open(self.report_path, 'w') as output_file:\\n dict_writer = csv.DictWriter(output_file, keys)\\n dict_writer.writeheader()\\n dict_writer.writerows(data_points)\",\n \"def to_df(self) -> Dict[str, Any]:\\n return {key: self.render_df(key) for key, value in self.__dict__.items()}\",\n \"def test_ingest_with_column_names_dict():\\n schema = {\\\"foo\\\": \\\"foo1\\\", \\\"bar\\\": \\\"bar2\\\"}\\n\\n data = [{\\\"foo\\\": 1, \\\"bar\\\": 2}, {\\\"foo\\\": 10, \\\"bar\\\": 20}]\\n\\n converted_data = client.ingest_data(data, schema)\\n assert converted_data.to_pydict() == {'foo1': [1, 10], 'bar2': [2, 20]}\",\n \"def dictify_df(frame):\\n ret = {}\\n for row in frame.values:\\n cur_level = ret\\n for elem in row[:-2]:\\n if elem not in cur_level:\\n cur_level[elem] = {}\\n cur_level = cur_level[elem]\\n cur_level[row[-2]] = row[-1]\\n return ret\",\n \"def dataframe():\\n headers = get_headers()\\n headers = {'headers': headers}\\n headers = pd.DataFrame.from_dict(headers, orient='index')\\n headers = headers.replace(r'\\\\n', ' ', regex=True)\\n headers = headers.replace(r'\\\\r', ' ', regex=True)\\n headers = headers.replace(r'\\\\t', ' ', regex=True)\\n headers = headers.replace(r'\\\\\\\\t', ' ', regex=True)\\n headers = headers.replace(r' ', ' ', regex=True)\\n headers = headers.replace(r' ', ' ', regex=True)\\n\\n paragraphs = get_paragraphs()\\n paragraphs = {'paragraphs': paragraphs}\\n paragraphs = pd.DataFrame.from_dict(paragraphs, orient='index')\\n paragraphs = paragraphs.replace(r'\\\\n', ' ', regex=True)\\n paragraphs = paragraphs.replace(r'\\\\r', ' ', regex=True)\\n paragraphs = paragraphs.replace(r'\\\\t', ' ', regex=True)\\n paragraphs = paragraphs.replace(r'\\\\\\\\t', ' ', regex=True)\\n paragraphs = paragraphs.replace(r' ', ' ', regex=True)\\n paragraphs = paragraphs.replace(r' ', ' ', regex=True)\\n\\n return headers.to_csv('headers.csv', index=False), paragraphs.to_csv('paragraphs.csv', index=False)\",\n \"def write_csv(data):\\n\\n with open('data.csv', 'w') as file:\\n writer = csv.DictWriter(file, fieldnames=data[0].keys())\\n writer.writeheader()\\n for row in data:\\n writer.writerow(row)\",\n \"def output_df(outdict, out_file):\\n\\tcols = ['#chrom', 'source', 'feature', 'chromStart', 'chromEnd', 'score', 'strand', 'frame', 'transcript_id']\\n\\tcolOut = ['#chrom', 'source', 'feature', 'start', 'end', 'score', 'strand', 'frame', 'transcript_id']\\n\\tgtfDF = pd.DataFrame(columns=cols)\\n\\n\\tfor trsp in outdict:\\n\\t\\tgtfDF = gtfDF.append(outdict[trsp], ignore_index=True)\\n\\t\\t\\n\\tgtfDF.columns = colOut\\n\\t# print gtfDF.head()\\n\\tgtfDF.to_csv(out_file, compression='gzip', sep='\\\\t', index=False)\",\n \"def csv_dict_writer(path, fieldnames, data):\\n with open(path, \\\"wb\\\") as out_file:\\n writer = csv.DictWriter(out_file, delimiter=',', fieldnames=fieldnames)\\n writer.writeheader()\\n for row in data:\\n writer.writerow(row)\",\n \"def isolate_df_data(self, dict_df):\\n\\t\\treturn_dict_df = {}\\n\\t\\tfor k,v in dict_df.items():\\n\\t\\t\\tnew_df = pd.DataFrame(index=v.index)\\n\\t\\t\\tnew_df[k + ' Value'] = v['value']\\n\\t\\t\\treturn_dict_df[k] = new_df\\n\\t\\treturn return_dict_df\",\n \"def gen_dict():\\n lines = [line for line in csv.reader(open(__ppath__ + \\\"/data/occupations.csv\\\"))] # uses a csv.reader to parse the file, converts the generic iterable to a list\\n lines = [(line[0],float(line[1])) for line in lines[1:-2]]# removes the column names and \\\"Total\\\" row, re-expresses as a list of tuples to enable dictionary conversion\\n lines.append((\\\"Unemployed\\\",0.2)) # accounts for missing 0.2% of jobs\\n return dict(lines) # converts to dictionary\",\n \"def df2dict(df):\\n attr_list = list(df.columns)\\n res_dict = {}\\n for attr in attr_list:\\n res_dict[attr.lower()] = list(df[attr])\\n\\n return res_dict\",\n \"def write_to_csv(path,data_dict):\\n\\n\\n schema = [\\\"file_name\\\",\\\"family\\\",\\\"genus\\\",\\\"genus_confidence\\\",\\n \\\"species_1\\\",\\\"confidence_1\\\",\\\"hall_1\\\",\\n \\\"species_2\\\",\\\"confidence_2\\\",\\\"hall_2\\\",\\n \\\"species_3\\\",\\\"confidence_3\\\",\\\"hall_3\\\",\\n \\\"species_4\\\",\\\"confidence_4\\\",\\\"hall_4\\\",\\\"peaks\\\"]\\n\\n # if no file exists create a one and inform the user\\n if not os.path.exists(path):\\n print(\\\"creating new output file {}\\\".format(path))\\n with open(path, \\\"w\\\") as csv_file:\\n filewriter = csv.writer(csv_file, delimiter=\\\",\\\")\\n filewriter.writerow(schema)\\n\\n row = []\\n\\n row.append(data_dict[\\\"file_name\\\"])\\n row.append(data_dict[\\\"family\\\"])\\n \\n row.append(data_dict[\\\"genus_1\\\"])\\n row.append(data_dict[\\\"genus_confidence_1\\\"][:5])\\n \\n row.append(data_dict[\\\"species_1\\\"])\\n row.append(data_dict[\\\"confidence_1\\\"][:5])\\n row.append(data_dict[\\\"hall_1\\\"])\\n \\n row.append(data_dict[\\\"species_2\\\"])\\n row.append(data_dict[\\\"confidence_2\\\"][:5])\\n row.append(data_dict[\\\"hall_2\\\"])\\n\\n row.append(data_dict[\\\"species_3\\\"])\\n row.append(data_dict[\\\"confidence_3\\\"][:5])\\n row.append(data_dict[\\\"hall_3\\\"])\\n\\n row.append(data_dict[\\\"species_4\\\"])\\n row.append(data_dict[\\\"confidence_4\\\"][:5])\\n row.append(data_dict[\\\"hall_4\\\"])\\n\\n row.append(data_dict[\\\"peaks\\\"])\\n \\n with open(path, \\\"a\\\") as csv_file:\\n filewriter = csv.writer(csv_file, delimiter=\\\",\\\")\\n filewriter.writerow(row)\",\n \"def generate_csv(data: dict, cols: list, file_name: str) -> None:\\r\\n\\r\\n longest = 0\\r\\n for i in cols:\\r\\n if len(data[i]) > longest:\\r\\n longest = len(data[i])\\r\\n\\r\\n len_cols = []\\r\\n for i in cols:\\r\\n len_cols.append(str(i) + \\\" (\\\" + str(len(data[i])) + \\\")\\\")\\r\\n\\r\\n for i in cols:\\r\\n while len(data[i]) < longest:\\r\\n data[i].append(\\\"\\\")\\r\\n\\r\\n new_data = {}\\r\\n count = 0\\r\\n for i in len_cols:\\r\\n new_data[i] = data[str(count)]\\r\\n count += 1\\r\\n\\r\\n df = pd.DataFrame(data=new_data, columns=len_cols)\\r\\n print(df)\\r\\n\\r\\n df.to_csv(file_name, encoding='utf-8', index=False)\",\n \"def save_dataset_csv(self, path):\\n cols = list(self.data_dict.keys())\\n df = pd.DataFrame(self.data_dict, index=None, columns=cols)\\n df.to_csv(path, index=True)\",\n \"def csv_dict_writer(path, fieldnames, data):\\n with open(path, \\\"wb\\\") as out_file:\\n writer = csv.DictWriter(out_file, delimiter=',', fieldnames=fieldnames)\\n writer.writeheader()\\n for row in data:\\n writer.writerow(row)\",\n \"def build_dataset_by_df(file_path):\\n data = pd.read_csv(file_path, sep=',')\\n print(data.head())\\n\\n user_ids = data['CustomerID'].unique()\\n prod_ids = data['StockCode'].unique()\\n user_num = len(user_ids)\\n product_num = len(prod_ids)\\n product_lists_by_user = dict(data.groupby('CustomerID')['StockCode'].apply(list)).values()\\n product_to_desc = dict(data.groupby('StockCode')['Description'].apply(set))\\n dictionary = {}\\n for id in prod_ids:\\n dictionary[id] = len(dictionary)\\n reversed_dictionary = dict(zip(dictionary.values(), dictionary.keys()))\\n return product_lists_by_user, product_to_desc, dictionary, reversed_dictionary, user_num, product_num\",\n \"def build_data_frame(sql_dict, names):\\n\\td = defaultdict(list)\\n\\ti = 0\\n\\tfor category_data in sql_dict:\\n\\t\\tfor attribute_list in category_data:\\n\\t\\t\\tfor attribute in category_data[attribute_list]:\\n\\t\\t\\t\\td[attribute_list].append(attribute)\\n\\t\\td[\\\"category\\\"].extend([names[i]]* len(category_data[attribute_list]) )\\n\\t\\td[\\\"fct\\\"].extend([i] * len(category_data[attribute_list]) )\\n\\t\\ti += 1\\n\\t#Translation into R objects\\n\\tfor attribute_list in d:\\n\\t\\tif attribute_list == \\\"category\\\":\\n\\t\\t\\td[attribute_list] = robjects.StrVector(d[attribute_list])\\n\\t\\telif attribute_list in WATCH_FOR_MISSING_VALUES:\\n\\t\\t\\ttmp = robjects.FloatVector(range(len(d[attribute_list])))\\n\\t\\t\\tfor i in range(len(d[attribute_list])):\\n\\t\\t\\t\\tif d[attribute_list][i] == \\\"NA\\\":\\n\\t\\t\\t\\t\\ttmp.rx[i] = robjects.NA_Real\\n\\t\\t\\t\\telse:\\n\\t\\t\\t\\t\\ttmp.rx[i] = d[attribute_list][i]\\n\\t\\t\\td[attribute_list] = tmp\\n\\t\\telse:\\n\\t\\t\\tif attribute_list == \\\"num_sections\\\":\\n\\t\\t\\t\\ttmp = []\\n\\t\\t\\t\\tfor element in d[attribute_list]:\\n\\t\\t\\t\\t\\tif element == None:\\n\\t\\t\\t\\t\\t\\ttmp.append(1)\\n\\t\\t\\t\\t\\telse:\\n\\t\\t\\t\\t\\t\\ttmp.append(float(element))\\n\\t\\t\\t\\td[attribute_list] = robjects.FloatVector(tmp)\\n\\t\\t\\t\\tcontinue\\n\\t\\t\\telif attribute_list == \\\"speechiness\\\":\\n\\t\\t\\t\\ttmp = []\\n\\t\\t\\t\\tfor element in d[attribute_list]:\\n\\t\\t\\t\\t\\tif element == None:\\n\\t\\t\\t\\t\\t\\ttmp.append(1)\\n\\t\\t\\t\\t\\telse:\\n\\t\\t\\t\\t\\t\\ttmp.append(float(element))\\n\\t\\t\\t\\td[attribute_list] = robjects.FloatVector(tmp)\\n\\t\\t\\t\\tcontinue\\n\\t\\t\\t\\t \\n\\t\\t\\telse:\\n\\t\\t\\t\\td[attribute_list] = robjects.FloatVector(d[attribute_list])\\n\\tdata_frame = robjects.DataFrame(d)\\n\\treturn data_frame\",\n \"def make_dict(data_for_dict): \\n \\n column_name_list = data_for_dict[0]\\n db_list = data_for_dict[1:]\\n \\n column_list1 = []\\n column_list2 = []\\n column_list3 = []\\n column_list4 = []\\n column_list5 = []\\n column_list6 = []\\n column_list7 = []\\n column_list8 = []\\n column_list9 = []\\n column_list10 = []\\n column_list11 = []\\n hmdb_dict = {}\\n for line in db_list:\\n my_string1 = '' \\n my_string2 = ''\\n my_string3 = ''\\n my_string4 = ''\\n my_string5 = ''\\n my_string6 = ''\\n my_string7 = ''\\n my_string8 = ''\\n my_string9 = ''\\n my_string10 = ''\\n my_string11 = ''\\n\\n my_string1 = line[0]\\n column_list1 += [my_string1]\\n my_string2 += line[1]\\n column_list2 += [my_string2]\\n my_string3 += line[2]\\n column_list3 += [my_string3]\\n my_string4 += line[3]\\n column_list4 += [my_string4]\\n my_string5 += line[4]\\n column_list5 += [my_string5]\\n my_string6 += line[5]\\n column_list6 += [my_string6]\\n my_string7 += line[6]\\n column_list7 += [my_string7]\\n my_string8 += line[7]\\n column_list8 += [my_string8]\\n my_string9 += line[8]\\n column_list9 += [my_string9]\\n my_string10 += line[9]\\n column_list10 += [my_string10]\\n my_string11 += line[10]\\n column_list11 += [my_string11] \\n \\n hmdb_dict[column_name_list[0]] = column_list1\\n hmdb_dict[column_name_list[1]] = column_list2\\n hmdb_dict[column_name_list[2]] = column_list3\\n hmdb_dict[column_name_list[3]] = column_list4\\n hmdb_dict[column_name_list[4]] = column_list5\\n hmdb_dict[column_name_list[5]] = column_list6\\n hmdb_dict[column_name_list[6]] = column_list7\\n hmdb_dict[column_name_list[7]] = column_list8\\n hmdb_dict[column_name_list[8]] = column_list9\\n hmdb_dict[column_name_list[9]] = column_list10\\n hmdb_dict[column_name_list[10]] = column_list11\\n \\n return (hmdb_dict)\",\n \"def make_ordered_to_dict(sorted_columns, include_timestamp):\\n od = OrderedDict()\\n for column in sorted_columns:\\n if include_timestamp:\\n value = (column.cell.value, column.cell.timestamp)\\n else:\\n value = column.cell.value\\n od[column.columnName] = value\\n return od\",\n \"def write_dataframe(self, path=os.getcwd(), name=\\\"Airbnb.csv\\\") -> None:\\n try:\\n df = pd.DataFrame(self.__collected_dic)\\n except ValueError:\\n print(\\\"ERROR! - Dictionary values are not the same length\\\")\\n return\\n\\n if not isinstance(name, str):\\n raise TypeError\\n if \\\".csv\\\" != name[-4:]:\\n name = f\\\"{name}.csv\\\"\\n df.to_csv(os.path.join(path, name), index=False)\\n print(f\\\"{name} file was succesfully written in {path}\\\")\",\n \"def write_csv(fn, toCSV):\\n keys = toCSV[0].keys()\\n with open(fn, 'w') as output_file:\\n dict_writer = csv.DictWriter(output_file, keys)\\n dict_writer.writeheader()\\n dict_writer.writerows(toCSV)\",\n \"def as_dict(self) :\\n\\n return dict(zip(self.cols, self.data))\",\n \"def chunk_to_dict(chunk):\\n csv_cols = chunk.keys()\\n return [dict(zip(csv_cols, v)) for v in chunk.values]\",\n \"def write_csv(dictionary):\\n\\n with open(constant.METADATA_FILE, 'a', newline='') as file:\\n writer = csv.DictWriter(file, fieldnames=dictionary.keys())\\n try:\\n writer.writerow(dictionary)\\n except Exception:\\n writeable = {k: str(v).encode(\\\"utf-8\\\") for k, v in dictionary.items()}\\n writer.writerow(writeable)\",\n \"def _create_model_out_dictkeys():\\r\\n model_names = []\\r\\n result_keys = []\\r\\n for model_name in model_zoo._ModelZooUrls.CONFIG_PATH_TO_URL_SUFFIX.keys():\\r\\n try:\\r\\n print(model_name, \\\":\\\")\\r\\n select_model(model_name)\\r\\n result = get_features_by_image_path(\\\"./sample.jpg\\\")\\r\\n model_names.append(model_name)\\r\\n result_keys.append(list(result.keys()))\\r\\n except RuntimeError as t:\\r\\n print(t)\\r\\n\\r\\n pd.DataFrame(list(zip(model_names, result_keys))).to_csv(\\\"d2_model_out_dictkeys.csv\\\")\",\n \"def create_dataframe(self, dictionary_to_convert, cols):\\n\\n dataframe_converted = pd.DataFrame.from_dict(dictionary_to_convert, orient='index', columns = cols)\\n dataframe_converted = dataframe_converted.reset_index()\\n dataframe_converted = dataframe_converted.drop(columns=['index'])\\n\\n return dataframe_converted\",\n \"def write_csv(self, _dict, filename):\\n with open(filename, 'w') as f:\\n f.write('\\\"'+'\\\";\\\"'.join(_dict.keys())+'\\\"\\\\n')\\n for i in np.arange(len(_dict[list(_dict.keys())[0]])):\\n values = []\\n for col in _dict.keys():\\n try:\\n values.append(str(_dict[col][i]))\\n except IndexError as e:\\n # LSTM don't have first times available because of lacking history\\n pass\\n f.write(';'.join(values)+'\\\\n')\\n\\n logging.info('Wrote {}'.format(filename))\\n self._upload_to_bucket(filename, filename)\",\n \"def _export(self, *args, **kwargs) -> Tuple[pd.DataFrame, dict]:\\n return self.data, kwargs\",\n \"def save_json(df):\\n dict = {}\\n for row in df.iterrows():\\n dict[row[1]['Country']] = {'Region' : row[1]['Region'],\\n 'Pop. Density (per sq. mi.)' : row[1]['Pop. Density (per sq. mi.)'],\\n 'Infant mortality (per 1000 births)' : row[1]['Infant mortality (per 1000 births)'],\\n 'GDP ($ per capita) dollars' : row[1]['GDP ($ per capita) dollars']\\n }\\n\\n with open('data.json', 'w', encoding='utf8') as outfile:\\n data = json.dumps(dict, indent=4, sort_keys=False, separators=(',', ': '), ensure_ascii=False)\\n outfile.write(data)\",\n \"def dict_to_csv(filename, dictionary, paramdict=True):\\n if paramdict:\\n filename += '_params'\\n \\n with open(filename + '.csv', 'w') as f:\\n for key in dictionary.keys():\\n if type(dictionary[key]) == tuple:\\n f.write(\\\"%s,%s, %s \\\\n\\\"%(key, dictionary[key][0], dictionary[key][1]))\\n else:\\n f.write(\\\"%s,%s \\\\n\\\"%(key, dictionary[key]))\\n print('Saved %s.csv.'%filename)\",\n \"def generate_new_row(forecast_date, target, target_end_date, \\n location, type, quantile, value):\\n new_row = {}\\n new_row[\\\"forecast_date\\\"] = forecast_date\\n new_row[\\\"target\\\"] = target\\n new_row[\\\"target_end_date\\\"] = target_end_date\\n new_row[\\\"location\\\"] = location\\n new_row[\\\"type\\\"] = type\\n new_row[\\\"quantile\\\"] = quantile\\n new_row[\\\"value\\\"] = value\\n return new_row\",\n \"def as_dict(self):\\r\\n values = {}\\r\\n for name, col in self._columns.items():\\r\\n values[name] = col.to_database(getattr(self, name, None))\\r\\n return values\",\n \"def create_dataframe():\\r\\n\\r\\n df = pd.read_csv('data/data.csv', header=0)\\r\\n return df\",\n \"def make_data(rics: list, fields: list, name: str) -> None:\\n df = get_data(rics, fields)\\n df.index.name = 'ID'\\n to_csv(df, name)\",\n \"def generate_log_df(log_columns, log_values):\\n return pd.DataFrame(dict(zip(log_columns, log_values)), index=[0])\",\n \"def link_datadict_to_dataframe(row, dict):\\n d = {}\\n for k, v in dict.items():\\n d[k] = [row[c] if row[c] == row[c] else None for c in v[\\\"COLUMN\\\"]]\\n\\n if len(d[k]) == 1:\\n d[k] = d[k][0]\\n elif None in d[k]:\\n d[k] = None\\n\\n return d\",\n \"def _convert_csv_column_to_dict(csv_data, column):\\n results = dict()\\n\\n for row in csv_data:\\n key = row[0]\\n data = row[1:][column]\\n\\n if data:\\n if key not in results:\\n results[key] = data.strip() if data else \\\"\\\"\\n else:\\n # append multiple choice questions\\n results[key] += \\\"|{0}\\\".format(data.strip())\\n\\n return results\",\n \"def create_dataset(input_file_path, output_file_path):\\n col_index_map = {'user_id': 0, 'session_id': 1, 'timestamp': 2, 'step': 3, 'action_type': 4, 'reference': 5,\\n 'platform': 6, 'city': 7, 'device': 8,\\n 'current_filters': 9, 'impressions': 10, 'prices': 11}\\n flat_dict = dict()\\n with open(input_file_path, 'r') as csvFile:\\n reader = csv.reader(csvFile)\\n header = next(reader)\\n col_names = [col_name for col_name in col_index_map.keys()]\\n col_names.pop(0)\\n index = 0\\n for row in tqdm(reader):\\n if len(flat_dict) > 40000:\\n index += 1\\n with open(output_file_path + \\\"_\\\" + str(index) + \\\".json\\\", \\\"w\\\") as file:\\n json.dump(flat_dict, file)\\n print(\\\" JSON : \\\", index)\\n flat_dict = dict()\\n col_values = [row[col_index_map[c_n]] for c_n in col_names]\\n dict_for_each_row = dict(zip(col_names, col_values))\\n to_list = dict_for_each_row['impressions']\\n dict_for_each_row['impressions'] = to_list.split('|')\\n to_list = dict_for_each_row['prices']\\n dict_for_each_row['prices'] = to_list.split('|')\\n user_id = row[col_index_map['user_id']]\\n if user_id in flat_dict:\\n flat_dict[user_id].append(dict_for_each_row)\\n else:\\n flat_dict[user_id] = [dict_for_each_row]\\n\\n print(\\\"Output is Saved\\\")\",\n \"def readData(filename):\\r\\n data_d = {}\\r\\n with open(filename) as f:\\r\\n df = pd.read_csv(f, header=0, dtype='str',sep=';')\\r\\n df = df.loc[:, ~df.columns.str.contains('^Unnamed')]\\r\\n df_dict = df.to_dict(orient='index')\\r\\n for i,val in df_dict.iteritems(): \\r\\n clean_row = [(k, p.proc(v)) for (k, v) in val.iteritems()]\\r\\n row_id = val['line_nr']\\r\\n data_d[row_id] = dict(clean_row)\\r\\n return data_d\\r\\n return df\",\n \"def create_csvfile(options_path, recept_info,df):\\n df_csv = df.copy()\\n df_csv.index.names = ['Interacting positions']\\n\\n #Change dynX by full name of receptor, adding the dynid if there is more than a simulation for that receptor\\n df_csv.columns = df_csv.columns.map(lambda x: recept_info[x][13])\\n\\n #Sorting by ballesteros Id's (helixloop column) and clustering order\\n df_csv['Interacting positions'] = df_csv.index\\n df_csv['helixloop'] = df_csv['Interacting positions'].apply(lambda x: re.sub(r'^(\\\\d)x',r'\\\\g<1>0x',x)) \\n df_csv = df_csv.sort_values([\\\"helixloop\\\"])\\n\\n #Change jumplines by 'x' to avoid formatting problems\\n def new_index(cell):\\n cell = cell.replace('\\\\n\\\\n', ' ')\\n cell = cell.replace('\\\\n', 'x')\\n cell = cell.replace('xx', 'x')\\n return cell\\n\\n df_csv['Interacting positions'] = df_csv['Interacting positions'].apply(lambda x: new_index(x))\\n df_csv.index = df_csv['Interacting positions']\\n\\n #Drop columns\\n df_csv.drop(columns = ['helixloop','Interacting positions'], inplace = True)\\n\\n #Store dataframe as csv\\n df_csv.to_csv(path_or_buf = options_path+\\\"dataframe.csv\\\", float_format='%.1f')\",\n \"def write_to_csv(data_in_dictionary, csvfile):\\n with open(csvfile, 'wb') as f:\\n fieldnames = data_in_dictionary.keys()\\n writer = csv.DictWriter(f, fieldnames)\\n writer.writeheader()\\n writer.writerow(data_in_dictionary)\\n logger.info(\\\"Data written to file: \\\" + csvfile)\",\n \"def _as_dict(self):\\r\\n values = self._dynamic_columns or {}\\r\\n for name, col in self._columns.items():\\r\\n values[name] = col.to_database(getattr(self, name, None))\\r\\n return values\",\n \"def dictionary_formation():\\r\\n sales_data = {}\\r\\n with open('beer_data.csv', \\\"r\\\") as data_file:\\r\\n file_contents = csv.reader(data_file, delimiter=',')\\r\\n #Value of lines_read used as key value for each dictionary\\r\\n #in sales_data\\r\\n lines_read = 1\\r\\n for line in file_contents:\\r\\n if lines_read == 1:\\r\\n lines_read = lines_read + 1\\r\\n else:\\r\\n #Stores each column in row as key value in dictionary\\r\\n sales_data[str(lines_read)] = {\\r\\n \\\"invoice_number\\\": line[0],\\r\\n \\\"customer\\\": line[1],\\r\\n \\\"date_required\\\": line[2],\\r\\n \\\"recipe\\\": line[3],\\r\\n \\\"gyle_number\\\": line[4],\\r\\n \\\"quantity_ordered\\\": int(line[5])\\r\\n }\\r\\n lines_read = lines_read + 1\\r\\n data_file.close()\\r\\n return sales_data\",\n \"def convert_to_dict(df):\\n df = df[['ISO3166A2',\\n 'ISOen_ro_name',\\n 'minlongitude',\\n 'maxlongitude',\\n 'minlatitude',\\n 'maxlatitude',\\n 'land_total',\\n ]]\\n df.ISOen_ro_name = df.ISOen_ro_name.str.strip()\\n df.set_index('ISOen_ro_name', inplace=True)\\n df.columns = ['country_tag',\\n 'min_lon',\\n 'max_lon',\\n 'min_lat',\\n 'max_lat',\\n 'area',\\n ]\\n df.loc[:, \\\"country_tag\\\"] = df.loc[:, \\\"country_tag\\\"].str.lower()\\n return {i[0]: i[1].to_dict() for i in df.iterrows()}\",\n \"def exportcsvsumdata(self, log):\\r\\n csvdata= None\\r\\n\\r\\n if (log):\\r\\n csvdata = ('%s\\\\t'%(log['CALLSIGN']))\\r\\n csvdata += ('%s\\\\t'%(log['OPERATORS']))\\r\\n csvdata += ('%s\\\\t'%(log['LOCATION']))\\r\\n csvdata += ('%d\\\\t'%(log['COUNT']))\\r\\n csvdata += ('%s\\\\t'%(log['NAMES']))\\r\\n if(log['LASTWORKED']): \\r\\n csvdata += ('%s/%s UTC'%(log['LASTWORKED'],\\r\\n log['LWTIME'])) \\r\\n\\r\\n return csvdata\",\n \"def write_to_csv(self, name_suffix = ''):\\n f_path = os.path.join(self.root_dir, 'res' + name_suffix + '.csv')\\n field_names = [] # the first field in CSV is 'obj_val'\\n\\n # put the keys in the cost, prim_var_change, dual_var_change and fea_conditions as field names if any\\n for key in self.cost.keys():\\n field_names.append(key)\\n for key in self.cost_change.keys():\\n field_names.append(key)\\n for key in self.prim_var_change.keys():\\n field_names.append(key)\\n for key in self.dual_var_change.keys():\\n field_names.append(key)\\n for key in self.fea_conditions.keys():\\n field_names.append(key)\\n\\n\\tprint f_path\\n\\n with open(f_path, mode = 'wb') as csv_file: # open the file, if not exist, create it\\n writer = csv.DictWriter(csv_file, fieldnames = field_names) # create a writer which maps the dictionaries onto output rows in CSV\\n writer.writeheader() # write the field names to the header\\n temp_dict = {} # create a temporary dict used to output rows\\n row_max = self.get_iter_num() # get the max iters which indicates the number of rows in CSV\\n print ('number of rows: ' + str(row_max))\\n #print (field_names)\\n for row in range(row_max + 1):\\n temp_dict.clear() # clear all items\\n start_idx = 0\\n for i in range(len(self.cost)):\\n field = field_names[start_idx + i]\\n\\t\\t if row > len(self.cost[field]) - 1:\\n\\t\\t\\ttemp_dict[field] = ''\\n\\t\\t else: temp_dict[field] = self.get_cost_val(field, row)\\n\\n start_idx = start_idx + len(self.cost) # the start pos of fields in field_names for prim_var_change\\n for i in range(len(self.cost_change)): # for each cost_change\\n field = field_names[start_idx + i]\\n if row == 0: # for row 0 (iter 0), we will set '/' to the change of primal variables\\n temp_dict[field] = '/'\\n elif row > len(self.cost_change[field]) - 1:\\n\\t\\t\\t temp_dict[field] = ''\\n\\t\\t else:\\n temp_dict[field] = self.get_cost_change_value(field, row - 1)\\n\\n\\n start_idx = start_idx + len(self.cost_change)\\n for i in range(len(self.prim_var_change)): # for each prim_var_change\\n field = field_names[start_idx + i]\\n if row == 0: # for row 0 (iter 0), we will set '/' to the change of primal variables\\n temp_dict[field] = '/'\\n\\t\\t elif row > len(self.prim_var_change[field]) - 1:\\n\\t\\t\\ttemp_dict[field] = ''\\n else:\\n temp_dict[field] = self.get_prim_change_value(field, row - 1)\\n\\n start_idx = start_idx + len(self.prim_var_change) # go to the start pos of fields in field_names for dual_var_change\\n for i in range(len(self.dual_var_change)): # for each dual_var_change\\n field = field_names[start_idx + i]\\n if row == 0: # for row 0 (iter 0), we will set '/' to the change of dual variables\\n temp_dict[field] = '/'\\n elif row > len(self.dual_var_change[field]) - 1:\\n\\t\\t\\ttemp_dict[field] = '' \\n\\t\\t else:\\n temp_dict[field] = self.get_dual_change_value(field, row - 1)\\n\\n start_idx = start_idx + len(self.dual_var_change) # go the the start pos of fields in field_names for fea_conditions\\n for i in range(len(self.fea_conditions)): # for each fea_condition\\n field = field_names[start_idx + i]\\n\\t\\t if row > len(self.fea_conditions[field]) - 1:\\n\\t\\t\\ttemp_dict[field] = ''\\n else: temp_dict[field] = self.get_fea_condition_value(field, row)\\n\\n writer.writerow(temp_dict)\\n\\n # we also save the value of primal values if not saved\\n if not self.pdv_to_csv:\\n self.save_last_prims()\",\n \"def df_to_json(complete_dataset, output_path, static_columns):\\n megajson = {}\\n\\n static_columns = [\\\"continent\\\", \\\"location\\\"] + list(static_columns)\\n\\n complete_dataset = complete_dataset.dropna(axis=\\\"rows\\\", subset=[\\\"iso_code\\\"])\\n\\n for _, row in complete_dataset.iterrows():\\n\\n row_iso = row[\\\"iso_code\\\"]\\n row_dict_static = row.drop(\\\"iso_code\\\")[static_columns].dropna().to_dict()\\n row_dict_dynamic = row.drop(\\\"iso_code\\\").drop(static_columns).dropna().to_dict()\\n\\n if row_iso not in megajson:\\n megajson[row_iso] = row_dict_static\\n megajson[row_iso][\\\"data\\\"] = [row_dict_dynamic]\\n else:\\n megajson[row_iso][\\\"data\\\"].append(row_dict_dynamic)\\n\\n with open(output_path, \\\"w\\\") as file:\\n file.write(json.dumps(megajson, indent=4))\",\n \"def create_csv(d, f):\\n csv_data = list()\\n csv_head = [unicode(x) for x in f]\\n\\n for row in d:\\n row_data = list()\\n for field in f:\\n fields = field.split('.')\\n row_data.append(extract_dict(row, fields))\\n csv_data.append(row_data)\\n\\n csv = (csv_head, csv_data)\\n return csv\",\n \"def _write_csv(self):\\n\\n # add the label to the header\\n if self.input_data.get_value(InputType.TIME_PERIOD) == 'all':\\n self.header.append('Date')\\n else:\\n self.header.append('sample id')\\n\\n key_list = []\\n\\n for i, cube in enumerate(self.cube_list):\\n if self.input_data.get_value(InputType.TIME_PERIOD) == 'all':\\n self._write_sample_with_date(cube, i, key_list)\\n else:\\n self._write_sample(cube, i, key_list)\\n\\n output_data_file_path = self._get_full_file_name()\\n self._write_data_dict(output_data_file_path, key_list)\\n\\n return [output_data_file_path]\",\n \"def generate_csv_file(rows: Collection[dict[str, str]]):\\n file = io.StringIO()\\n csv_writer = csv.DictWriter(file, (\\\"label\\\", \\\"inventory\\\", \\\"collection\\\"))\\n csv_writer.writeheader()\\n csv_writer.writerows(rows)\\n file.seek(0)\\n return io.BytesIO(bytes(file.read(), encoding=\\\"utf-8\\\"))\",\n \"def create_dataframe(dictionary_to_convert, cols):\\n\\n dataframe_converted = pd.DataFrame.from_dict(dictionary_to_convert, orient='index', columns = cols)\\n dataframe_converted = dataframe_converted.reset_index()\\n dataframe_converted = dataframe_converted.drop(columns=['index'])\\n\\n return dataframe_converted\",\n \"def to_json(self):\\n\\t\\treturn self._dataframe.reset_index().to_json(orient=\\\"records\\\")\",\n \"def csv_to_dict(fp):\\n import pandas as pd\\n df = pd.read_csv(fp, index_col=0, header=None)\\n d = df.to_dict(orient='index')\\n d = {k: v.values() for k, v in d.iteritems()}\\n return d\",\n \"def _aggregated_df_to_dict(agg_df: DataFrameLike) -> Dict[str, float]:\\n try:\\n agg_dicts = agg_df.to_dict(orient='index')\\n except TypeError:\\n # pd.Series objects do not have to_dict method with orient kwarg\\n # so cast to pd.DataFrame and transpose for correct shape\\n agg_dicts = agg_df.to_frame().T.to_dict(orient='index')\\n\\n formatted_agg_dict = {\\n f'{key}({idx})': val\\n for idx, row in agg_dicts.items()\\n for key, val in row.items()\\n }\\n return formatted_agg_dict\",\n \"def as_dict(self) -> Dict[str, Any]:\\n return {\\n column_title: cell.value\\n for column_title, cell in self.column_title_to_cell.items()\\n }\",\n \"def create_query_csv(self):\\n\\n self.query_df.to_csv(self.query_output_file)\",\n \"def PostgresTableToDict(sqldef, conn):\\n df = pd.read_sql(sqldef, con=conn)\\n return dict(zip(df.iloc[:, 0], df.iloc[:, 1]))\",\n \"def pandas_typecast(self) -> dict:\\n res = {}\\n for feat in self.data_features:\\n res[feat.key] = ApiForm.typecast(feat.dtype)\\n return res\",\n \"def read_delsys_csv(filename: str) -> Dict[str, Dict[str, TimeSeries]]:\\n # Check the number of rows to skip\\n n_rows = 0\\n with open(filename, 'r') as fid:\\n while True:\\n s = fid.readline()\\n if s.startswith('X[s]'):\\n break\\n else:\\n n_rows += 1\\n\\n # Open the CSV\\n df = pd.read_csv(filename, skiprows=n_rows)\\n\\n # Create a TimeSeries for each signal since they all have different time\\n # vectors\\n n_signals = int(len(df.columns) / 2)\\n\\n emg = {}\\n acc = {}\\n gyro = {}\\n mag = {}\\n\\n for i_signal in range(n_signals):\\n time = df.iloc[:, i_signal * 2].to_numpy()\\n name = df.columns[i_signal * 2 + 1]\\n data = df.iloc[:, i_signal * 2 + 1].to_numpy()\\n\\n if ': Acc' in name:\\n short_name = name\\n ts = TimeSeries(time=time, data={short_name: data})\\n acc[short_name] = ts\\n elif ': Mag' in name:\\n short_name = name\\n ts = TimeSeries(time=time, data={short_name: data})\\n mag[short_name] = ts\\n elif ': Gyro' in name:\\n short_name = name\\n ts = TimeSeries(time=time, data={short_name: data})\\n gyro[short_name] = ts\\n elif ': EMG' in name:\\n short_name = name.split(':')[0]\\n ts = TimeSeries(time=time, data={short_name: data})\\n emg[short_name] = ts\\n\\n return {'emg': emg, 'acc': acc, 'gyro': gyro, 'mag': mag}\",\n \"def as_dict(self) -> Dict[str, Any]:\\n return {\\n column_title: cell.get_value()\\n for column_title, cell in self.column_title_to_cell.items()\\n }\",\n \"def DictData(self):\\n reader = csv.DictReader( open( self.file, \\\"rU\\\" ), dialect = \\\"excel\\\" )\\n return reader\",\n \"def export_csv(header, data):\\n with StringIO() as tmp:\\n writer = csv.DictWriter(tmp, fieldnames=header)\\n writer.writeheader()\\n writer.writerows(data)\\n data = tmp.getvalue()\\n\\n return data\",\n \"def write_csv(f, extract, fields=None):\\n keys = fields if fields is not None else extract[0].keys()\\n dict_writer = csv.DictWriter(f, keys, extrasaction='ignore')\\n dict_writer.writeheader()\\n dict_writer.writerows(extract)\\n return f\",\n \"def save_csv(connection, query, columns, name):\\n\\n try:\\n df = pd.read_sql(query, connection, columns=columns)\\n df.to_csv(name, index=False)\\n except Exception as ex:\\n print(type(ex))\\n print(ex)\",\n \"def stream_to_db(subreddit: str, \\n df_dict: dict, \\n db_path: str) -> None:\\n file_lkps = {'posts': f\\\"reddit-{subreddit}-posts.csv\\\", \\n 'comments': f\\\"reddit-{subreddit}-comments.csv\\\"}\\n \\n for _key in df_dict:\\n if df_dict.get(_key) is None:\\n pass\\n full_path = f\\\"{db_path}/{file_lkps.get(_key)}\\\"\\n df = df_dict.get(_key)\\n df.to_csv(full_path, index=False, encoding='utf-8')\\n logging.info(f\\\"Saved {_key} data for subreddit {subreddit} at {datetime.today()}\\\")\\n \\n return None\",\n \"def each_to_csv(data, key, value):\\n data.to_csv(\\\"camelot/clean/nrld_{}_{}.csv\\\".format(key, value), index=False)\\n return data\",\n \"def make_csv(file_of_data):\\n with open(file_of_data, 'w') as f:\\n writer = csv.writer(f)\\n header = (\\\"Counter\\\", \\\"Date/time\\\", \\\"Latitude\\\", \\\"Longitude\\\", \\\"Temperature\\\", \\\"Humidity\\\")\\n writer.writerow(header)\",\n \"def read_stats_csv(filename):\\n\\n df_dict = {}\\n df = pd.read_csv(filename, header=[0, 1, 2])\\n\\n # Check if End column data type is datetime - if so use start date as index, otherwise use file number;\\n # Use start date as index - Note: df[\\\"End\\\"] is interpreted as a dataframe here not a series as in hdf5\\n if df[\\\"End\\\"].dtypes.all() == pd.Timestamp:\\n # Drop redundant columns\\n if \\\"File Number\\\" in df.columns:\\n df = df.drop(\\\"File Number\\\", axis=1, level=0)\\n df = df.drop(\\\"End\\\", axis=1, level=0)\\n df = df.set_index(df.columns[0])\\n df.index.name = \\\"Date\\\"\\n\\n # Convert timestamps to datetime\\n try:\\n df.index = pd.to_datetime(df.index, format=\\\"%Y-%m-%d %H:%M:%S\\\")\\n except:\\n try:\\n # Timestamp will likely be in local (UK) format if csv file has been subsequently edited and saved\\n df.index = pd.to_datetime(df.index, format=\\\"%d/%m/%Y %H:%M\\\")\\n except:\\n raise\\n # Use file number as index\\n else:\\n df = df.drop([\\\"Start\\\", \\\"End\\\"], axis=1, level=0)\\n df = df.set_index(df.columns[0])\\n df.index.name = \\\"File Number\\\"\\n\\n df.columns.rename([\\\"channels\\\", \\\"stats\\\", \\\"units\\\"], inplace=True)\\n logger = filename.split(\\\"Statistics_\\\")[-1].split(\\\".\\\")[0]\\n df_dict[logger] = df\\n\\n return df_dict\",\n \"def create_dataframe(self):\\n\\n df = pd.DataFrame({'date': [],\\n 'RUN': [],\\n 'CLONE': [],\\n 'GEN': pd.Series(0, index=[], dtype='int'),\\n 'frame': pd.Series([], index=[], dtype='int'),\\n 'time (ns)': [] }) # set the index\\n df.set_index('date')\\n print(df)\\n\\n # Save the DataFrame to disk\\n\\n ### create a file handle to store the data in (a dict-like) HDF5 format\\n store = pd.HDFStore(self.dataframe_path)\\n print(store)\\n store.put('df', df)\\n return store\",\n \"def create_csv(self, type): \\n if os.path.isfile(_path_finder('keydata','{0}_{1}.db'.format(\\n self.keyword,type))):\\n self.__db_init('{0}'.format(type))\\n self.c.execute(\\\"SELECT MIN(date) FROM tweets\\\")\\n mindate = self.c.fetchone()[0][0:10]\\n self.c.execute(\\\"SELECT MAX(date) FROM tweets\\\")\\n maxdate = self.c.fetchone()[0][0:10]\\n start_date = datetime.datetime.strptime(mindate, '%Y-%m-%d')\\n end_date = (datetime.datetime.strptime(maxdate, '%Y-%m-%d') + \\n datetime.timedelta(days=1))\\n \\n def __date_range(start, end):\\n for n in range((end - start).days):\\n yield start + datetime.timedelta(days=n)\\n \\n def __db_to_list():\\n for single_date in __date_range(start_date, end_date):\\n d = \\\"\\\".join(['%',single_date.strftime(\\\"%Y-%m-%d\\\"),'%'])\\n self.c.execute('''SELECT count(*) FROM tweets where \\n date like('{0}')'''.format(d))\\n yield [d[1:11], self.c.fetchone()[0]]\\n \\n path = _path_finder('keydata','{0}_{1}.csv'.format(\\n self.keyword,type))\\n if sys.version_info[0] < 3: #Python3 compatibility check\\n infile = open(path, 'wb')\\n else:\\n infile = open(path, 'w', newline='', encoding='utf8')\\n with infile as f:\\n writer = csv.writer(f)\\n writer.writerows(__db_to_list())\\n self.conn.commit()\\n self.conn.close()\\n print('\\\\nReport has been created:')\\n print(os.path.abspath(path))\"\n]","isConversational":false},"negative_scores":{"kind":"list like","value":["0.68224066","0.67767906","0.65991676","0.6495553","0.64526737","0.6419535","0.6320348","0.62898755","0.6284467","0.6255064","0.62294525","0.62211573","0.6205003","0.6200207","0.61460024","0.61113054","0.6097703","0.60526484","0.60267997","0.6017852","0.6014081","0.600382","0.59845316","0.5953394","0.59025145","0.5901035","0.5879255","0.58771026","0.586691","0.586303","0.58621913","0.58540696","0.585137","0.5846501","0.5821844","0.5820106","0.5796997","0.5795294","0.5784772","0.57721674","0.57703835","0.5768244","0.5759943","0.5756479","0.5742448","0.5715866","0.57127535","0.570887","0.57077116","0.5706913","0.5705207","0.5704369","0.56854403","0.5685327","0.5676551","0.56737196","0.5653622","0.5637295","0.5631909","0.5629384","0.5626578","0.56265086","0.56261635","0.56182086","0.561101","0.5610304","0.560665","0.5597874","0.55974984","0.5593769","0.5586","0.55851823","0.5583944","0.55823964","0.5579883","0.5579153","0.555253","0.55476177","0.5546526","0.55408955","0.5536764","0.5536331","0.5502579","0.55008876","0.54978484","0.5496857","0.5489109","0.54875535","0.5473396","0.5464363","0.5457154","0.5457118","0.5456807","0.5453595","0.5451142","0.54476786","0.5443968","0.54428047","0.5437501","0.5424835","0.5423739"],"string":"[\n \"0.68224066\",\n \"0.67767906\",\n \"0.65991676\",\n \"0.6495553\",\n \"0.64526737\",\n \"0.6419535\",\n \"0.6320348\",\n \"0.62898755\",\n \"0.6284467\",\n \"0.6255064\",\n \"0.62294525\",\n \"0.62211573\",\n \"0.6205003\",\n \"0.6200207\",\n \"0.61460024\",\n \"0.61113054\",\n \"0.6097703\",\n \"0.60526484\",\n \"0.60267997\",\n \"0.6017852\",\n \"0.6014081\",\n \"0.600382\",\n \"0.59845316\",\n \"0.5953394\",\n \"0.59025145\",\n \"0.5901035\",\n \"0.5879255\",\n \"0.58771026\",\n \"0.586691\",\n \"0.586303\",\n \"0.58621913\",\n \"0.58540696\",\n \"0.585137\",\n \"0.5846501\",\n \"0.5821844\",\n \"0.5820106\",\n \"0.5796997\",\n \"0.5795294\",\n \"0.5784772\",\n \"0.57721674\",\n \"0.57703835\",\n \"0.5768244\",\n \"0.5759943\",\n \"0.5756479\",\n \"0.5742448\",\n \"0.5715866\",\n \"0.57127535\",\n \"0.570887\",\n \"0.57077116\",\n \"0.5706913\",\n \"0.5705207\",\n \"0.5704369\",\n \"0.56854403\",\n \"0.5685327\",\n \"0.5676551\",\n \"0.56737196\",\n \"0.5653622\",\n \"0.5637295\",\n \"0.5631909\",\n \"0.5629384\",\n \"0.5626578\",\n \"0.56265086\",\n \"0.56261635\",\n \"0.56182086\",\n \"0.561101\",\n \"0.5610304\",\n \"0.560665\",\n \"0.5597874\",\n \"0.55974984\",\n \"0.5593769\",\n \"0.5586\",\n \"0.55851823\",\n \"0.5583944\",\n \"0.55823964\",\n \"0.5579883\",\n \"0.5579153\",\n \"0.555253\",\n \"0.55476177\",\n \"0.5546526\",\n \"0.55408955\",\n \"0.5536764\",\n \"0.5536331\",\n \"0.5502579\",\n \"0.55008876\",\n \"0.54978484\",\n \"0.5496857\",\n \"0.5489109\",\n \"0.54875535\",\n \"0.5473396\",\n \"0.5464363\",\n \"0.5457154\",\n \"0.5457118\",\n \"0.5456807\",\n \"0.5453595\",\n \"0.5451142\",\n \"0.54476786\",\n \"0.5443968\",\n \"0.54428047\",\n \"0.5437501\",\n \"0.5424835\",\n \"0.5423739\"\n]","isConversational":false},"document_score":{"kind":"string","value":"0.0"},"document_rank":{"kind":"string","value":"-1"}}}],"truncated":false,"partial":true},"paginationData":{"pageIndex":932,"numItemsPerPage":100,"numTotalItems":95772,"offset":93200,"length":100}},"jwt":"eyJhbGciOiJFZERTQSIsImtpZCI6IjVHZDBvd0g5MTM2eDZjc1FvbE1zcktNWUZoRVFoUm5PVVVybEpjOUhaUEEifQ.eyJyZWFkIjp0cnVlLCJwZXJtaXNzaW9ucyI6eyJyZXBvLmNvbnRlbnQucmVhZCI6dHJ1ZX0sImlhdCI6MTc4NTU0NzU4MCwianRpIjoiMGQ3MDhmYTYtMzdjNy00YzlmLTlmOTktMTMzYzA3OTQwOGVhIiwic3ViIjoiL2RhdGFzZXRzL25vbWljLWFpL2Nvcm5zdGFjay1weXRob24tdjEiLCJleHAiOjE3ODU1NTExODAsImlzcyI6Imh0dHBzOi8vaHVnZ2luZ2ZhY2UuY28ifQ.LLr8-2QVLtq8HMDINZZYYymGzt2G5lgOU4OFUE7PCoc30AF8lsEYAFtSzwVR69DJD2XMjNjMNmez4PvFVVGtAw","displayUrls":true,"splitSizeSummaries":[{"config":"default","split":"train","numRows":95772,"numBytesParquet":1928375891}]},"discussionsStats":{"closed":1,"open":1,"total":2},"fullWidth":true,"hasGatedAccess":true,"hasFullAccess":true,"isEmbedded":false,"savedQueries":{"community":[],"user":[]}}">
query stringlengths 9 3.4k | document stringlengths 9 87.4k | metadata dict | negatives listlengths 4 101 | negative_scores listlengths 4 101 | document_score stringlengths 3 10 | document_rank stringclasses 102
values |
|---|---|---|---|---|---|---|
Create new lists were previously doubled points in dist list removed and averaged in elev_list | def average_double_pts(elev_list, dist_list, minimum_point_distance=0.5):
import numpy as np
dist_array = np.array(dist_list)
diffs = dist_array[1:] - dist_array[:-1]
bad_pts = diffs < minimum_point_distance
bad_indices = [i for i in range(len(bad_pts)) if bad_pts[i]]
new_elev_array = np.arra... | {
"objective": {
"self": [],
"paired": [],
"triplet": [
[
"query",
"document",
"negatives"
]
]
}
} | [
"def v_dist_space(trial):\n return [np.mean(space) for space in trial]",
"def _euclidian_distances(stop_list):\n e_dists2 = [transitfeed.approximate_distance_between_stops(stop, tail) for\n (stop, tail) in zip(stop_list, stop_list[1:])]\n\n return e_dists2",
"def _transform_d... | [
"0.59104544",
"0.5468914",
"0.5397144",
"0.53817797",
"0.5364048",
"0.535594",
"0.53433156",
"0.5334994",
"0.53302234",
"0.53162",
"0.52459675",
"0.52387244",
"0.52223456",
"0.519468",
"0.5191414",
"0.51864374",
"0.5181687",
"0.51815665",
"0.51673615",
"0.5149777",
"0.5147351... | 0.72335047 | 0 |
Return an array of item comparisons, avoiding selfcomparisons. | def pairwise_equalities(things):
return np.array([things[i] == things[j] for i in range(len(things))
for j in range(i + 1, len(things))]) | {
"objective": {
"self": [],
"paired": [],
"triplet": [
[
"query",
"document",
"negatives"
]
]
}
} | [
"def item_comparer(self):\n return self.item_comparer_value",
"def OnCompareItems(self, item1, item2):\r\n\r\n return cmp(self.GetItemText(item1), self.GetItemText(item2))",
"def by_equality(cls, *expected_items):\n return cls(*(to_matcher(EqualsMatcher, i) for i in expected_items))",
"de... | [
"0.58421564",
"0.56897765",
"0.5672012",
"0.5588486",
"0.5552559",
"0.5403639",
"0.5401066",
"0.53403103",
"0.5271652",
"0.5211439",
"0.51990104",
"0.5190706",
"0.5167556",
"0.51661444",
"0.51609856",
"0.5154473",
"0.51290977",
"0.5125599",
"0.5107754",
"0.5106961",
"0.510658... | 0.48618746 | 46 |
Because of slight numerical instabilities, the mostlikelytree method does _not_ in fact always return the same tree when there are multiple trees with the same or very nearly the same likelihood. The best we can ask for is that the method returns _a_ best tree, defined as one with the biggest likelihood, or nearly the ... | def test_that_most_probable_always_returns_a_tree_of_the_same_logprob():
grammar = Grammar(rulebooks)
actual_tree = grammar.sample_tree()
sentence = actual_tree.terminals
inner = grammar.compute_inside_probabilities(sentence)
best_tree = grammar.compute_most_likely_tree(sentence, inner, root=0)
... | {
"objective": {
"self": [],
"paired": [],
"triplet": [
[
"query",
"document",
"negatives"
]
]
}
} | [
"def find_maximum_value(self):\n if self.root: \n self.max_val = self.root.value\n else:\n return 'No tree found'\n def inner(root):\n if root.left:\n inner(root.left)\n\n if root.right:\n inner(root.right)\n\n if self.max_val < root.value:\n self.max_val = roo... | [
"0.67465556",
"0.6164014",
"0.61626226",
"0.60900533",
"0.6089753",
"0.60731304",
"0.6045642",
"0.60375273",
"0.6031647",
"0.6021071",
"0.5996487",
"0.59402007",
"0.59293586",
"0.58979386",
"0.58588094",
"0.58377635",
"0.58170396",
"0.58130044",
"0.58093256",
"0.5809125",
"0.... | 0.7200898 | 0 |
Load in arguments for applying coefficients for Ensemble Model Output Statistics (EMOS), otherwise known as Nonhomogeneous Gaussian Regression (NGR). The coefficients are applied to the forecast that is supplied, so as to calibrate the forecast. The calibrated forecast is written to a netCDF file. If no coefficients ar... | def main(argv=None):
parser = ArgParser(
description='Apply coefficients for Ensemble Model Output '
'Statistics (EMOS), otherwise known as Non-homogeneous '
'Gaussian Regression (NGR). The supported input formats '
'are realizations, probabilities... | {
"objective": {
"self": [],
"paired": [],
"triplet": [
[
"query",
"document",
"negatives"
]
]
}
} | [
"def process(current_forecast, coeffs, num_realizations=None,\n random_ordering=False, random_seed=None,\n ecc_bounds_warning=False, predictor_of_mean='mean'):\n if coeffs is None:\n msg = (\"There are no coefficients provided for calibration. The \"\n \"uncalibrated fo... | [
"0.5519549",
"0.5417745",
"0.53862226",
"0.52634716",
"0.50273323",
"0.5010265",
"0.49229556",
"0.49013907",
"0.4877792",
"0.48454607",
"0.4783449",
"0.47544655",
"0.47529",
"0.47140917",
"0.47133023",
"0.47060087",
"0.46985283",
"0.4687954",
"0.46867755",
"0.4664076",
"0.465... | 0.7194427 | 0 |
Applying coefficients for Ensemble Model Output Statistics. Load in arguments for applying coefficients for Ensemble Model Output Statistics (EMOS), otherwise known as Nonhomogeneous Gaussian Regression (NGR). The coefficients are applied to the forecast that is supplied, so as to calibrate the forecast. The calibrated... | def process(current_forecast, coeffs, num_realizations=None,
random_ordering=False, random_seed=None,
ecc_bounds_warning=False, predictor_of_mean='mean'):
if coeffs is None:
msg = ("There are no coefficients provided for calibration. The "
"uncalibrated forecast will b... | {
"objective": {
"self": [],
"paired": [],
"triplet": [
[
"query",
"document",
"negatives"
]
]
}
} | [
"def main(argv=None):\n parser = ArgParser(\n description='Apply coefficients for Ensemble Model Output '\n 'Statistics (EMOS), otherwise known as Non-homogeneous '\n 'Gaussian Regression (NGR). The supported input formats '\n 'are realizations, pro... | [
"0.6839221",
"0.5753612",
"0.5297529",
"0.50588846",
"0.486943",
"0.48606175",
"0.48570108",
"0.4810559",
"0.4788578",
"0.47579268",
"0.4745183",
"0.47238272",
"0.47165662",
"0.4668261",
"0.46644482",
"0.4629341",
"0.46219748",
"0.46173713",
"0.46099064",
"0.46055603",
"0.460... | 0.6488239 | 1 |
Iterate over batches of a given size when the input is simply the raw image itself | def iterate_batches(inputs, targets, batch_size, shuffle=True):
# Make sure that the input and target arrays are also of the same size
assert len(inputs) == len(targets)
indices = np.arange(len(inputs))
if shuffle:
np.random.shuffle(indices)
for start_index in rang... | {
"objective": {
"self": [],
"paired": [],
"triplet": [
[
"query",
"document",
"negatives"
]
]
}
} | [
"def generator(array, batch_size):\n start = 0 # pointer to where we are in iteration\n while True:\n stop = start + batch_size\n diff = stop - array.shape[0]\n if diff <= 0:\n batch = array[start:stop]\n start += batch_size\n else:\n batch = np.concatenate((array[start:], array[:diff]... | [
"0.7110718",
"0.7075221",
"0.70680904",
"0.7002261",
"0.6975876",
"0.69697165",
"0.69678855",
"0.69515705",
"0.6930055",
"0.6887131",
"0.68486667",
"0.68431544",
"0.67821205",
"0.6775559",
"0.6767301",
"0.6744574",
"0.67342305",
"0.6732619",
"0.67257315",
"0.6717631",
"0.6696... | 0.6239594 | 64 |
Retrieve a list of all panelist scores from nonBest Of and nonRepeat shows | def retrieve_all_scores(database_connection: mysql.connector.connect
) -> List[int]:
cursor = database_connection.cursor()
query = ("SELECT pm.panelistscore FROM ww_showpnlmap pm "
"JOIN ww_shows s ON s.showid = pm.showid "
"WHERE s.bestof = 0 AND s.repeatshowid ... | {
"objective": {
"self": [],
"paired": [],
"triplet": [
[
"query",
"document",
"negatives"
]
]
}
} | [
"def retrieve_grouped_scores(database_connection: mysql.connector.connect\n ) -> Dict:\n cursor = database_connection.cursor()\n query = (\"SELECT pm.panelistscore, COUNT(pm.panelistscore) \"\n \"FROM ww_showpnlmap pm \"\n \"JOIN ww_shows s ON s.showid = pm.sh... | [
"0.63882565",
"0.6345482",
"0.60516936",
"0.60274804",
"0.5971298",
"0.59415215",
"0.5897459",
"0.58885854",
"0.58742666",
"0.56829435",
"0.567487",
"0.56604165",
"0.56549484",
"0.5630021",
"0.5613433",
"0.56064445",
"0.5598407",
"0.5592181",
"0.55896866",
"0.55654347",
"0.55... | 0.6766444 | 0 |
Retrieve a list of grouped panelist scores from nonBest Of and nonRepeat shows | def retrieve_grouped_scores(database_connection: mysql.connector.connect
) -> Dict:
cursor = database_connection.cursor()
query = ("SELECT pm.panelistscore, COUNT(pm.panelistscore) "
"FROM ww_showpnlmap pm "
"JOIN ww_shows s ON s.showid = pm.showid "
... | {
"objective": {
"self": [],
"paired": [],
"triplet": [
[
"query",
"document",
"negatives"
]
]
}
} | [
"def retrieve_all_scores(database_connection: mysql.connector.connect\n ) -> List[int]:\n cursor = database_connection.cursor()\n query = (\"SELECT pm.panelistscore FROM ww_showpnlmap pm \"\n \"JOIN ww_shows s ON s.showid = pm.showid \"\n \"WHERE s.bestof = 0 AND ... | [
"0.61236584",
"0.60025114",
"0.5887449",
"0.58714366",
"0.58647263",
"0.57514274",
"0.57352406",
"0.5672734",
"0.5664091",
"0.5638059",
"0.5633964",
"0.56201035",
"0.5512606",
"0.55024105",
"0.5480762",
"0.5474629",
"0.5465591",
"0.54459333",
"0.5442375",
"0.53806484",
"0.537... | 0.6751662 | 0 |
Calculate stats for all of the panelist scores | def calculate_stats(scores: List[int]):
stats = OrderedDict()
stats["count"] = len(scores)
stats["minimum"] = int(numpy.amin(scores))
stats["maximum"] = int(numpy.amax(scores))
stats["mean"] = round(numpy.mean(scores), 4)
stats["median"] = int(numpy.median(scores))
stats["standard_deviation"... | {
"objective": {
"self": [],
"paired": [],
"triplet": [
[
"query",
"document",
"negatives"
]
]
}
} | [
"def disp_score():",
"def update_scores(self):\r\n totals = [0, 0, 0, 0]\r\n for player in range(0, 4):\r\n for round_num in range(0, 17):\r\n try:\r\n bid = int(self.spin_bids[player][round_num].get())\r\n tricks = int(self.spin_tricks... | [
"0.69189066",
"0.6732025",
"0.6481974",
"0.64566064",
"0.6448772",
"0.6393949",
"0.6384861",
"0.63422036",
"0.63363665",
"0.6307522",
"0.62796324",
"0.6164866",
"0.61432964",
"0.6126191",
"0.6116369",
"0.61115795",
"0.60974866",
"0.60919774",
"0.6089504",
"0.6074493",
"0.6064... | 0.6238826 | 11 |
Print out the score stats | def print_stats(stats: Dict):
print()
print(" Panelist Scores")
print(" Count: {}".format(stats["count"]))
print(" Minimum: {}".format(stats["minimum"]))
print(" Maximum: {}".format(stats["maximum"]))
print(" Median: {}".format(stats["median"]))
print(" Mean: ... | {
"objective": {
"self": [],
"paired": [],
"triplet": [
[
"query",
"document",
"negatives"
]
]
}
} | [
"def print_scores(self):\n ### FILL IN ###",
"def print_scores(self):\n print(\"scores: \", self.get_scores())",
"def disp_score():",
"def display_scores(scores):\n\n\t#print(\"\\t\\t\\tScores: \", list(scores))\n\tprint(\"\\t\\t\\tMean: \", scores.mean())\n\tprint(\"\\t\\t\\tStandard Deviation:... | [
"0.84830403",
"0.8449821",
"0.78642124",
"0.77374864",
"0.75699306",
"0.75461483",
"0.7507333",
"0.74400777",
"0.7432176",
"0.7403559",
"0.727528",
"0.71789753",
"0.71708167",
"0.70969886",
"0.70647675",
"0.7057524",
"0.7038564",
"0.70344925",
"0.7024657",
"0.7006421",
"0.697... | 0.7133117 | 13 |
Print out the scrore spread | def print_score_spread(score_spread: Dict):
print(" Score Spread\n")
print(" Score Count")
for score in score_spread:
print(" {:>7}{:>12}".format(score["score"], score["count"]))
print()
return | {
"objective": {
"self": [],
"paired": [],
"triplet": [
[
"query",
"document",
"negatives"
]
]
}
} | [
"def print(self):\r\n self.print_avec_separateur()",
"def print(self):\n self.print_avec_separateur(\" \")",
"def print_scores(self):\n ### FILL IN ###",
"def print_output(self):\n print(\"Reference score: \" + str(self.PotTax_reference.sum().TFI))\n print(\"Intervention sco... | [
"0.6561322",
"0.643337",
"0.6330498",
"0.6281885",
"0.62810534",
"0.62737805",
"0.62556595",
"0.6224368",
"0.6221346",
"0.6209357",
"0.6202504",
"0.6195883",
"0.6191251",
"0.6188209",
"0.6159958",
"0.6145935",
"0.6139892",
"0.60920966",
"0.6085532",
"0.60845447",
"0.6076933",... | 0.71421385 | 0 |
Load configuration file from config.json | def load_config(app_environment) -> Dict:
with open('config.json', 'r') as config_file:
config_dict = json.load(config_file)
if app_environment.startswith("develop"):
if "development" in config_dict:
config = config_dict["development"]
else:
raise Exception("Miss... | {
"objective": {
"self": [],
"paired": [],
"triplet": [
[
"query",
"document",
"negatives"
]
]
}
} | [
"def load_config(self):\r\n with open('config.json', 'r') as f:\r\n self.config = json.load(f)",
"def load_config():\n global config\n\n with open(\"config.json\") as f:\n json_config = f.read()\n f.close()\n config = json.loads(json_config)",
"def load_config():\n here =... | [
"0.8881662",
"0.8681146",
"0.86279327",
"0.8495189",
"0.8393106",
"0.8161093",
"0.8107652",
"0.80916744",
"0.8027772",
"0.79656696",
"0.7927089",
"0.791812",
"0.7890831",
"0.7879447",
"0.78793275",
"0.78693116",
"0.78693116",
"0.78527486",
"0.7845725",
"0.7770614",
"0.7750379... | 0.69013244 | 97 |
Pull in scoring data and generate stats based on the data | def main():
app_environment = os.getenv("APP_ENV", "local").strip().lower()
config = load_config(app_environment)
database_connection = mysql.connector.connect(**config["database"])
all_scores = retrieve_all_scores(database_connection)
stats = calculate_stats(all_scores)
print_stats(stats)
... | {
"objective": {
"self": [],
"paired": [],
"triplet": [
[
"query",
"document",
"negatives"
]
]
}
} | [
"def get_score(self):\n files_flare = self.generate_flare_set()\n files_non_flare = self.generate_non_flare_set()\n timeseries = []\n y = []\n scores = {}\n column_mapping = self.__get_column_mapping()\n for col in tqdm(range(1, 25)):\n for file in tqdm(fi... | [
"0.6874266",
"0.6695014",
"0.6617427",
"0.65822417",
"0.63687444",
"0.6361665",
"0.6314557",
"0.6211419",
"0.6166223",
"0.6105088",
"0.6104523",
"0.60632145",
"0.6057583",
"0.60560167",
"0.60233086",
"0.5951729",
"0.59515303",
"0.5930597",
"0.59092224",
"0.59012306",
"0.58973... | 0.5663475 | 53 |
constructeur de la classe | def __init__(self, t, a, v):
self.time = t
self.attribute = a
self.value = v | {
"objective": {
"self": [],
"paired": [],
"triplet": [
[
"query",
"document",
"negatives"
]
]
}
} | [
"def __init__(self) -> None:\n # TODO: Provide the complete constructor for this object",
"def __init__(self):\n raise NotImplementedError",
"def __init__(self):\n raise NotImplementedError",
"def __init__(self):\n raise NotImplementedError",
"def __init__(self):\n raise N... | [
"0.8200863",
"0.80114734",
"0.80114734",
"0.80114734",
"0.80114734",
"0.80000806",
"0.79996276",
"0.7898036",
"0.7823019",
"0.7815178",
"0.7815178",
"0.7815178",
"0.7815178",
"0.7815178",
"0.7815178",
"0.7815178",
"0.7815178",
"0.7815178",
"0.7815178",
"0.7815178",
"0.7815178... | 0.0 | -1 |
constructeur de la classe | def __init__(self, n, t, s):
self.name = n
self.target = t
self.start = s
self.keyframes = [] | {
"objective": {
"self": [],
"paired": [],
"triplet": [
[
"query",
"document",
"negatives"
]
]
}
} | [
"def __init__(self) -> None:\n # TODO: Provide the complete constructor for this object",
"def __init__(self):\n raise NotImplementedError",
"def __init__(self):\n raise NotImplementedError",
"def __init__(self):\n raise NotImplementedError",
"def __init__(self):\n raise N... | [
"0.8200863",
"0.80114734",
"0.80114734",
"0.80114734",
"0.80114734",
"0.80000806",
"0.79996276",
"0.7898036",
"0.7823019",
"0.7815178",
"0.7815178",
"0.7815178",
"0.7815178",
"0.7815178",
"0.7815178",
"0.7815178",
"0.7815178",
"0.7815178",
"0.7815178",
"0.7815178",
"0.7815178... | 0.0 | -1 |
constructeur de la classe | def __init__(self, n):
self.name = n
self.timelines = {} | {
"objective": {
"self": [],
"paired": [],
"triplet": [
[
"query",
"document",
"negatives"
]
]
}
} | [
"def __init__(self) -> None:\n # TODO: Provide the complete constructor for this object",
"def __init__(self):\n raise NotImplementedError",
"def __init__(self):\n raise NotImplementedError",
"def __init__(self):\n raise NotImplementedError",
"def __init__(self):\n raise N... | [
"0.8200863",
"0.80114734",
"0.80114734",
"0.80114734",
"0.80114734",
"0.80000806",
"0.79996276",
"0.7898036",
"0.7823019",
"0.7815178",
"0.7815178",
"0.7815178",
"0.7815178",
"0.7815178",
"0.7815178",
"0.7815178",
"0.7815178",
"0.7815178",
"0.7815178",
"0.7815178",
"0.7815178... | 0.0 | -1 |
fonction qui ajoute une ligne de temps dans l'animation | def add_timeline(self, t):
self.timelines.update({t.name : t}) | {
"objective": {
"self": [],
"paired": [],
"triplet": [
[
"query",
"document",
"negatives"
]
]
}
} | [
"def on_timer(self, event):\n \n o = Unicorn()\n data = o.get_data(rt)\n k = len(data[0])\n y[:, :-k] = y[:, k:]\n y[:, -k:] = remap((data), -40, 40, -1, 1 ) \n t2 = _thread.start_new_thread(printT, ())\n #y2 = np.array([lfilter(b, a, y[i]) for i in range(17)]... | [
"0.6608005",
"0.6595933",
"0.6549832",
"0.6526188",
"0.6441607",
"0.6407639",
"0.64039797",
"0.63954604",
"0.6334338",
"0.6334338",
"0.63001573",
"0.62417245",
"0.6231882",
"0.62167317",
"0.61714906",
"0.611921",
"0.61162287",
"0.60838497",
"0.60744506",
"0.60744506",
"0.6051... | 0.0 | -1 |
Return version key to use given a specific version. For example, if a dictionary only has instructions for version 'generic', will return 'generic' given any version string. Raises `ValueError` if no key could be found for requested version. | def get_version_key(self, version):
if self._generic_only:
return GENERIC_VERSION
else:
self.check_version_exists(version)
return version | {
"objective": {
"self": [],
"paired": [],
"triplet": [
[
"query",
"document",
"negatives"
]
]
}
} | [
"def GetKey(self, version_number):\n return self.dict[str(version_number)]",
"def key_version(self) -> Optional[str]:\n return pulumi.get(self, \"key_version\")",
"def get_version(self, key, version=None):\n if version is None:\n if key not in self._data:\n raise KeyEr... | [
"0.7516901",
"0.6918039",
"0.6749801",
"0.6605578",
"0.6401275",
"0.6225745",
"0.620228",
"0.619842",
"0.6191113",
"0.6155921",
"0.6147764",
"0.60847867",
"0.6057561",
"0.60508114",
"0.5988967",
"0.59874403",
"0.59791094",
"0.59724796",
"0.5945546",
"0.59303916",
"0.58354455"... | 0.8208527 | 0 |
Return dictionary with rendered keys and values. | def render_env(self):
return {
jinja2.Template(k).render({self.name: self}):
jinja2.Template(v).render({self.name: self})
for k, v in self.env.items()
} if self.env else self.env | {
"objective": {
"self": [],
"paired": [],
"triplet": [
[
"query",
"document",
"negatives"
]
]
}
} | [
"def make_dict(self):\n return self.generate_widgets()",
"def render_dict(dict):\n\t\treturn str.encode(str(dict))",
"def _template_data(self):\n return {\"form\": self.form.render()}",
"def render_dictionary(self): \n asset_json = {\n 'name': self.name,\n 'prod... | [
"0.6552589",
"0.6363904",
"0.62682635",
"0.62654954",
"0.6263933",
"0.6260918",
"0.62299466",
"0.60744417",
"0.6046848",
"0.5990396",
"0.59854126",
"0.5968035",
"0.59228987",
"0.59225416",
"0.5914001",
"0.5899469",
"0.5892299",
"0.5860215",
"0.58501613",
"0.5825743",
"0.58257... | 0.6336243 | 2 |
Creates a keras regularizer. | def get_regularizer(type, l1, l2):
if type == 'l1':
reg = keras.regularizers.l1(l1)
elif type == 'l2':
reg = keras.regularizers.l2(l2)
elif type == 'l1_l2':
reg = keras.regularizers.l1_l2(l1, l2)
else:
reg = None
return reg | {
"objective": {
"self": [],
"paired": [],
"triplet": [
[
"query",
"document",
"negatives"
]
]
}
} | [
"def update_regularizer(self, regularizer = regularizers.l1(0.1)):\n # for layer in self.layers:\n # layer.kernel_regularizer = regularizer\n self.list_cnn[-1].kernel_regularizer = regularizer",
"def add_regularization(self, regularizer):\n for layer in self.prenet_layers:\n ... | [
"0.6614855",
"0.6163753",
"0.6110547",
"0.60351515",
"0.5921308",
"0.5716683",
"0.56689423",
"0.5552135",
"0.54837906",
"0.5478638",
"0.546309",
"0.5458307",
"0.54565173",
"0.5401835",
"0.53959936",
"0.5376531",
"0.5373115",
"0.5337177",
"0.5316122",
"0.53155696",
"0.52800167... | 0.6125757 | 2 |
Input layer class that defines the input shape for the network. An input layer is required for the first layer in all models. | def __init__(self, shape):
self.shape = literal_eval(shape)
self.keras_layer = keras.layers.Input(shape=self.shape)
self.type = 'Input'
self.name = ':'.join([self.type, str(self.shape)]) | {
"objective": {
"self": [],
"paired": [],
"triplet": [
[
"query",
"document",
"negatives"
]
]
}
} | [
"def __init__(self, shape, input_var=None):\n\n self.output = layers.InputLayer(shape, input_var=input_var)",
"def build(self, input_layer, trainable=True):\n\n with tf.variable_scope(self.name):\n # Determine the size of the input when flattened\n input_layer_shape = input_layer.get_s... | [
"0.73620754",
"0.68776786",
"0.68701315",
"0.6866467",
"0.6684106",
"0.66439646",
"0.66422045",
"0.65750545",
"0.6553184",
"0.6525942",
"0.63709074",
"0.63709074",
"0.63709074",
"0.6291092",
"0.62881696",
"0.6286421",
"0.6284356",
"0.6267243",
"0.62483495",
"0.62224865",
"0.6... | 0.700692 | 1 |
layer_exist = [conv_act, pool_max_act, pool_ave_act, pool_max_no_act, pool_ave_no_act] | def layers(arr):
layer_exist = [0, 0, 0, 0, 0]
arr = arr.split()
for i, a in enumerate(arr):
if arr[i].lower().startswith("convolution") or arr[i].lower().startswith("innerproduct"):
if i != len(arr) and arr[i+1].lower() == "relu":
layer_exist[0] = 1
if arr[i].lower() == "avepooling":
if (i != (len... | {
"objective": {
"self": [],
"paired": [],
"triplet": [
[
"query",
"document",
"negatives"
]
]
}
} | [
"def get_features(inp_layer, pad=0):\n # Note: tweaked to use average pooling instead of maxpooling\n net = OrderedDict()\n net['conv1_1'] = ConvLayer(inp_layer, 64, 3, pad=pad, flip_filters=False, nonlinearity=rectify)\n net['conv1_2'] = ConvLayer(net['conv1_1'], 64, 3, pad=pad, flip_filters=False, non... | [
"0.6113794",
"0.60285205",
"0.5972099",
"0.59599495",
"0.58484864",
"0.5841038",
"0.5795003",
"0.5780913",
"0.57801276",
"0.5778729",
"0.5769082",
"0.57289445",
"0.5715246",
"0.5696411",
"0.56913024",
"0.5674453",
"0.5619215",
"0.5609114",
"0.558215",
"0.5575851",
"0.5573412"... | 0.6529583 | 0 |
Interfaces with individually compatible ports that are ordered in different ways should still be deemed compatible. | def test_is_compatible_sel_order(self):
# Selectors with multiple levels:
i = Interface('/foo[0:2],/bar[0:2]')
i['/foo[0:2]', 'interface', 'io'] = [0, 'in']
i['/bar[0:2]', 'interface', 'io'] = [0, 'out']
j = Interface('/bar[0:2],/foo[0:2]')
j['/bar[0:2]', 'interface', 'i... | {
"objective": {
"self": [],
"paired": [],
"triplet": [
[
"query",
"document",
"negatives"
]
]
}
} | [
"def target_interfaces(self):",
"def target_interfaces(self):",
"def temporally_align_data_interfaces(self):\n pass",
"def connect_interfaces(self, iface1: str, ip1_name: str,\n iface2: str, ip2_name: str):\n if (ip1_name not in self._ips.keys() or\n ip2_... | [
"0.6219634",
"0.6219634",
"0.5754795",
"0.57205325",
"0.56868404",
"0.56806505",
"0.5673285",
"0.5562755",
"0.55203426",
"0.55015105",
"0.55015105",
"0.5495435",
"0.54481024",
"0.54418063",
"0.5391868",
"0.53848755",
"0.5348518",
"0.5347069",
"0.53385854",
"0.53363526",
"0.53... | 0.51881975 | 30 |
It should be possible to define compatible interfaces containing both input and output ports. | def test_is_compatible_both_dirs(self):
i = Interface('/foo[0:4]')
i['/foo[0:2]', 'interface', 'io'] = [0, 'out']
i['/foo[2:4]', 'interface', 'io'] = [0, 'in']
j = Interface('/foo[0:4]')
j['/foo[0:2]', 'interface', 'io'] = [1, 'in']
j['/foo[2:4]', 'interface', 'io'] = [1... | {
"objective": {
"self": [],
"paired": [],
"triplet": [
[
"query",
"document",
"negatives"
]
]
}
} | [
"def target_interfaces(self):",
"def target_interfaces(self):",
"def interfaces(self):",
"def interfaces(self):",
"def __init__(__self__, *,\n from_port: pulumi.Input[int],\n to_port: pulumi.Input[int]):\n pulumi.set(__self__, \"from_port\", from_port)\n pulumi.... | [
"0.67224735",
"0.67224735",
"0.61528635",
"0.61528635",
"0.5994254",
"0.5912999",
"0.58316606",
"0.58027184",
"0.5625491",
"0.56192183",
"0.5570189",
"0.55607367",
"0.5545737",
"0.5538456",
"0.54931146",
"0.54783744",
"0.5430864",
"0.5430864",
"0.5412712",
"0.54087794",
"0.53... | 0.0 | -1 |
It should be possible to define compatible interfaces containing both input and output ports with specified types. | def test_is_compatible_both_dirs_types(self):
i = Interface('/foo[0:4]')
i['/foo[0:2]'] = [0, 'out', 'gpot']
i['/foo[2:4]'] = [0, 'in', 'spike']
j = Interface('/foo[0:4]')
j['/foo[0:2]'] = [1, 'in', 'gpot']
j['/foo[2:4]'] = [1, 'out', 'spike']
assert i.is_compati... | {
"objective": {
"self": [],
"paired": [],
"triplet": [
[
"query",
"document",
"negatives"
]
]
}
} | [
"def target_interfaces(self):",
"def target_interfaces(self):",
"def set_io_types(self, other):\n other.set_input_type(self.input_type).set_output_type(self.output_type)\n self.set_input_type(other.input_type).set_output_type(other.output_type)\n return self",
"def convert_interface(self, access_modi... | [
"0.62524253",
"0.62524253",
"0.59509814",
"0.5787149",
"0.56537616",
"0.56537616",
"0.55518574",
"0.55134493",
"0.5413066",
"0.54043597",
"0.5389857",
"0.53701156",
"0.5331732",
"0.52904046",
"0.52746",
"0.52578366",
"0.52531266",
"0.52442455",
"0.5240985",
"0.52286166",
"0.5... | 0.46521124 | 87 |
Interfaces can be compatible even if some of their ports do not have a set input or output status. | def test_is_compatible_with_nulls(self):
i = Interface('/foo[0:3]')
i['/foo[0:2]', 'interface', 'io'] = [0, 'out']
i['/foo[2]', 'interface'] = 0
j = Interface('/foo[0:3]')
j['/foo[0:2]', 'interface', 'io'] = [1, 'in']
j['/foo[2]', 'interface'] = 1
assert i.is_com... | {
"objective": {
"self": [],
"paired": [],
"triplet": [
[
"query",
"document",
"negatives"
]
]
}
} | [
"def target_interfaces(self):",
"def target_interfaces(self):",
"def connect_interfaces(self, iface1: str, ip1_name: str,\n iface2: str, ip2_name: str):\n if (ip1_name not in self._ips.keys() or\n ip2_name not in self._ips.keys()):\n raise ValueError(\n... | [
"0.6464873",
"0.6464873",
"0.61978257",
"0.59971684",
"0.59839207",
"0.57873195",
"0.57873195",
"0.5763124",
"0.5751498",
"0.5732146",
"0.56955546",
"0.5691122",
"0.56707364",
"0.5632924",
"0.5616398",
"0.5613331",
"0.56106734",
"0.55786246",
"0.55636865",
"0.55598676",
"0.55... | 0.54418343 | 22 |
Interfaces can be compatible even if some of their ports do not have a set type. | def test_is_compatible_with_nulls_types(self):
i = Interface('/foo[0:3]')
i['/foo[0:2]'] = [0, 'out', 'gpot']
i['/foo[2]', 'interface'] = 0
j = Interface('/foo[0:3]')
j['/foo[0:2]'] = [1, 'in', 'gpot']
j['/foo[2]', 'interface'] = 1
assert i.is_compatible(0, j, 1) | {
"objective": {
"self": [],
"paired": [],
"triplet": [
[
"query",
"document",
"negatives"
]
]
}
} | [
"def target_interfaces(self):",
"def target_interfaces(self):",
"def connect_interfaces(self, iface1: str, ip1_name: str,\n iface2: str, ip2_name: str):\n if (ip1_name not in self._ips.keys() or\n ip2_name not in self._ips.keys()):\n raise ValueError(\n... | [
"0.6660175",
"0.6660175",
"0.62674147",
"0.6096234",
"0.6096234",
"0.59134775",
"0.5881471",
"0.5868384",
"0.5858607",
"0.5845406",
"0.5821904",
"0.5656214",
"0.5649054",
"0.5648483",
"0.56465447",
"0.5585523",
"0.557584",
"0.55142313",
"0.5511947",
"0.5511947",
"0.5503032",
... | 0.6222775 | 3 |
Interfaces that both share a subset of compatible ports can be deemed compatible by setting the `allow_subsets` option of the compatibility test. | def test_is_compatible_subsets(self):
i = Interface('/foo[0:6]')
i['/foo[0:3]'] = [0, 'out', 'gpot']
i['/foo[3:6]'] = [0, 'out', 'spike']
j = Interface('/foo[0:6]')
j['/foo[0:2]'] = [1, 'in', 'gpot']
j['/foo[3:5]'] = [1, 'in', 'spike']
k = Interface('/foo[0:6]')
... | {
"objective": {
"self": [],
"paired": [],
"triplet": [
[
"query",
"document",
"negatives"
]
]
}
} | [
"def test_is_compatible_subsets_with_null_types(self):\n\n i = Interface('/foo[0:6]')\n i['/foo[0:3]'] = [0, 'out']\n i['/foo[3:6]'] = [0, 'out']\n j = Interface('/foo[0:6]')\n j['/foo[0:2]'] = [1, 'in']\n j['/foo[3:5]'] = [1, 'in']\n k = Interface('/foo[0:6]')\n ... | [
"0.6700613",
"0.5888596",
"0.5888596",
"0.5716942",
"0.5570211",
"0.5458658",
"0.5365004",
"0.5320249",
"0.52785486",
"0.51995903",
"0.5177569",
"0.515498",
"0.51448685",
"0.5117207",
"0.49495307",
"0.49465245",
"0.49463874",
"0.49270806",
"0.4923223",
"0.491944",
"0.4907876"... | 0.78658915 | 0 |
Interfaces that both share a subset of compatible ports can be deemed compatible by setting the `allow_subsets` option of the compatibility test even when the types are null. | def test_is_compatible_subsets_with_null_types(self):
i = Interface('/foo[0:6]')
i['/foo[0:3]'] = [0, 'out']
i['/foo[3:6]'] = [0, 'out']
j = Interface('/foo[0:6]')
j['/foo[0:2]'] = [1, 'in']
j['/foo[3:5]'] = [1, 'in']
k = Interface('/foo[0:6]')
assert i.i... | {
"objective": {
"self": [],
"paired": [],
"triplet": [
[
"query",
"document",
"negatives"
]
]
}
} | [
"def test_is_compatible_subsets(self):\n\n i = Interface('/foo[0:6]')\n i['/foo[0:3]'] = [0, 'out', 'gpot']\n i['/foo[3:6]'] = [0, 'out', 'spike']\n j = Interface('/foo[0:6]')\n j['/foo[0:2]'] = [1, 'in', 'gpot']\n j['/foo[3:5]'] = [1, 'in', 'spike']\n k = Interface(... | [
"0.76551175",
"0.6187195",
"0.5959266",
"0.5550304",
"0.5550304",
"0.52693313",
"0.5247899",
"0.5175491",
"0.5172923",
"0.51695246",
"0.51173824",
"0.5093611",
"0.5085437",
"0.50665164",
"0.504104",
"0.5036018",
"0.50319415",
"0.5031793",
"0.49958917",
"0.49873957",
"0.497339... | 0.7800067 | 0 |
Return a strategy that holds if and only if pending >= x or player reaches goal. | def hold_at(x):
def strategy(state):
# your code here
p, me, you, pending = state
return "hold" if (pending >= x or me + pending >= goal) else "roll"
strategy.__name__ = 'hold_at(%d)' % x
return strategy | {
"objective": {
"self": [],
"paired": [],
"triplet": [
[
"query",
"document",
"negatives"
]
]
}
} | [
"def hold_at(x):\n\n def strategy(state):\n (p, me, you, pending) = state\n return 'hold' if (pending >= x or me + pending >= goal) else 'roll'\n\n strategy.__name__ = 'hold_at(%d)' % x\n return strategy",
"def Pwin(state):\n # Assumes opponent also plays with optimal strategy\n p, me... | [
"0.7355454",
"0.60242695",
"0.6023155",
"0.5951468",
"0.5937637",
"0.572351",
"0.5663513",
"0.56209296",
"0.56154317",
"0.56149787",
"0.5597981",
"0.55769897",
"0.556702",
"0.55660975",
"0.5485379",
"0.54448724",
"0.5435684",
"0.54253024",
"0.5404603",
"0.5403368",
"0.5395781... | 0.7068357 | 1 |
Play a game of pig between two players, represented by their strategies. Each time through the main loop we ask the current player for one decision, which must be 'hold' or 'roll', and we update the state accordingly. When one player's score exceeds the goal, return that player. | def play_pig(A, B):
# your code here
strategies = [A,B]
state = (0,0,0,0)
while True:
p, me, you, pending = state
if me >= goal:
return strategies[p]
elif you >= goal:
return strategies[other_turn[p]]
# elif strategies[p](state) == 'hold':
... | {
"objective": {
"self": [],
"paired": [],
"triplet": [
[
"query",
"document",
"negatives"
]
]
}
} | [
"def play_pig(A, B):\n # your code here\n strategies = [A, B]\n state = (0, 0, 0, 0)\n while True:\n (p, me, you, pending) = state\n if me >= goal:\n return strategies[p]\n elif you >= goal:\n return strategies[other[p]]\n elif strategies[p](state) == 'h... | [
"0.7907119",
"0.7647033",
"0.7639917",
"0.723368",
"0.7189156",
"0.7187271",
"0.7064494",
"0.7062731",
"0.7054494",
"0.68677187",
"0.674492",
"0.670514",
"0.6567981",
"0.65314806",
"0.6519332",
"0.6519332",
"0.6519332",
"0.65100443",
"0.65067357",
"0.6433185",
"0.632696",
"... | 0.7669755 | 1 |
Decorator that caches the return value for each call to f(args). Then when called again with same args, we can just look it up. | def memo(f):
cache = {}
def _f(*args):
try:
return cache[args]
except KeyError:
cache[args] = result = f(*args)
return result
except TypeError:
# some element of args can't be a dict key
return f(args)
return _f | {
"objective": {
"self": [],
"paired": [],
"triplet": [
[
"query",
"document",
"negatives"
]
]
}
} | [
"def memoize(f):\n cache = {}\n @functools.wraps(f)\n def g(*args):\n ret = cache.get(args, cache)\n if ret is cache:\n ret = cache[args] = f(*args)\n return ret\n return g",
"def __call__(self, *args):\n if args not in self.memo:\n self.memo[args] = self.f(*arg... | [
"0.81131274",
"0.80451596",
"0.8038009",
"0.8008783",
"0.79322904",
"0.7912004",
"0.7912004",
"0.7912004",
"0.7751491",
"0.7736098",
"0.7705755",
"0.76081574",
"0.7593774",
"0.7568509",
"0.7536149",
"0.7459705",
"0.7363271",
"0.7348799",
"0.7338997",
"0.73304653",
"0.73113525... | 0.7772014 | 9 |
The utility of a state, here just the probability that an optimal player whose turn it to move can win from the current state | def Pwin(state):
# Assumes opponent also plays with optimal strategy
p, me, you, pending = state
if me + pending >= goal:
return 1
elif you >= goal:
return 0
else:
return max(Q_pig(state, action, Pwin) for action in pig_actions(state)) | {
"objective": {
"self": [],
"paired": [],
"triplet": [
[
"query",
"document",
"negatives"
]
]
}
} | [
"def rough_outcome(self) -> float:\n # HUYNH YOU PRICK WHY THE FUCK DO YOU MAKE US WRITE THIS SHIT EVEN IT'S NOT USED ANYWHERE\n # pick move based on this may not be optimal but better than random\n # return 1 if win immediately\n # return -1 if all states reachable will result the other... | [
"0.77746516",
"0.7501392",
"0.73808473",
"0.7227007",
"0.69487715",
"0.68937486",
"0.68130076",
"0.6812178",
"0.68047494",
"0.6799885",
"0.6795713",
"0.6783112",
"0.67188185",
"0.665067",
"0.65943986",
"0.6592056",
"0.65892106",
"0.6571575",
"0.65390325",
"0.6532277",
"0.6526... | 0.75051713 | 1 |
A strategy that maximizes the expected difference between my final score and my opponent's. | def max_diffs(state):
# your code here
return best_action(state, pig_actions, Q_pig, win_diff) | {
"objective": {
"self": [],
"paired": [],
"triplet": [
[
"query",
"document",
"negatives"
]
]
}
} | [
"def final_strategy(score, opponent_score):\r\n free, k = opponent_score // 10 + 1, 1\r\n winning = score > opponent_score\r\n ratio = (goal - score) / (goal - opponent_score)\r\n mean = make_mean_strategy(3)(score, opponent_score)\r\n if goal - score <= free: return 0\r\n while k < 4:\r\n ... | [
"0.7319641",
"0.7241692",
"0.72058785",
"0.70882905",
"0.69610775",
"0.6886896",
"0.688641",
"0.6814296",
"0.6810675",
"0.67275155",
"0.67186815",
"0.6703836",
"0.6694766",
"0.6693455",
"0.6671631",
"0.666438",
"0.6662398",
"0.66298085",
"0.6617561",
"0.6569686",
"0.6525511",... | 0.6654747 | 17 |
Parse number from string with automatic type detection. Parses input string and converts to a number using automatic type detection. If the number contains any fractional parts, then the number will be converted to float, otherwise the number will be converted to an int. | def parse_number_auto_dtype(x: str) -> Union[int, float]:
value: Union[int, float] = float(x)
if value.is_integer():
value = int(value)
return value | {
"objective": {
"self": [],
"paired": [],
"triplet": [
[
"query",
"document",
"negatives"
]
]
}
} | [
"def _parseNumber(self, str):\r\n\t\tif (str.count(\".\") == 0):\r\n\t\t\treturn int(str)\r\n\t\tif (str.count(\".\") == 1):\r\n\t\t\treturn float(str)\r\n\t\treturn str",
"def str2num(s):\r\n try:\r\n return int(s)\r\n except ValueError:\r\n return float(s)",
"def str_to_num(s):\n\n meth... | [
"0.806247",
"0.7663109",
"0.76058894",
"0.7603773",
"0.7543853",
"0.75322175",
"0.74632233",
"0.7321434",
"0.72316444",
"0.7026305",
"0.7019086",
"0.7009801",
"0.6992993",
"0.69265276",
"0.690676",
"0.6881941",
"0.68817985",
"0.68067026",
"0.67965007",
"0.6696791",
"0.6664161... | 0.73269945 | 7 |
Helper function for `_GenericAlias.__repr__`. | def _to_str(obj: object) -> str:
if obj is Ellipsis:
return '...'
elif isinstance(obj, type) and not isinstance(obj, _GENERIC_ALIAS_TYPE):
if obj.__module__ == 'builtins':
return obj.__qualname__
else:
return f'{obj.__module__}.{obj.__qualname__}'
else:
... | {
"objective": {
"self": [],
"paired": [],
"triplet": [
[
"query",
"document",
"negatives"
]
]
}
} | [
"def __repr__(self):\n args = []\n if self.name != \"alpha\":\n args.append(repr(self.name))\n if self.propertiesstr:\n args.append(repr(self.propertiesstr))\n elif self.propertiesstr:\n args.append(\"attr=%r\" % self.propertiesstr)\n retur... | [
"0.74621296",
"0.73563325",
"0.7346365",
"0.72829545",
"0.7205059",
"0.7160061",
"0.715092",
"0.71405625",
"0.71347225",
"0.71313053",
"0.7101834",
"0.7092644",
"0.7092644",
"0.7092644",
"0.7092644",
"0.7092644",
"0.7081821",
"0.70372033",
"0.69983757",
"0.69691986",
"0.69690... | 0.0 | -1 |
Search for all typevars and typevarcontaining objects in `args`. Helper function for `_GenericAlias.__init__`. | def _parse_parameters(args: Iterable[Any]) -> Generator[TypeVar, None, None]:
for i in args:
if hasattr(i, "__parameters__"):
yield from i.__parameters__
elif isinstance(i, TypeVar):
yield i | {
"objective": {
"self": [],
"paired": [],
"triplet": [
[
"query",
"document",
"negatives"
]
]
}
} | [
"def findType(*args, deep: bool=True, exact: bool=True, forward: bool=True, type: AnyStr=\"\",\n **kwargs)->List[AnyStr]:\n pass",
"def find(self, *args):\n return _ida_hexrays.lvars_t_find(self, *args)",
"def used_variables(*terms):\n\n t = terms[0] if len(terms) == 1 else terms\n\n ... | [
"0.6083603",
"0.5493589",
"0.53747714",
"0.53291667",
"0.5164382",
"0.5154936",
"0.50176644",
"0.5015225",
"0.49874654",
"0.49854574",
"0.49834588",
"0.4983403",
"0.49549422",
"0.49442652",
"0.4919461",
"0.48979554",
"0.48860407",
"0.47833136",
"0.4778925",
"0.47677916",
"0.4... | 0.591873 | 1 |
Recursively replace all typevars with those from `parameters`. Helper function for `_GenericAlias.__getitem__`. | def _reconstruct_alias(alias: _T, parameters: Iterator[TypeVar]) -> _T:
args = []
for i in alias.__args__:
if isinstance(i, TypeVar):
value: Any = next(parameters)
elif isinstance(i, _GenericAlias):
value = _reconstruct_alias(i, parameters)
elif hasattr(i, "__para... | {
"objective": {
"self": [],
"paired": [],
"triplet": [
[
"query",
"document",
"negatives"
]
]
}
} | [
"def replace_vars(params, contents):\n if isinstance(contents, str):\n contents = [contents]\n replace_contents = []\n\n if params != None and contents != None:\n for content in contents:\n replace_content = content\n for match in regexp_replace_var.findall(content):\n ... | [
"0.5673922",
"0.5523126",
"0.5507102",
"0.54108083",
"0.54027843",
"0.5336006",
"0.5278679",
"0.5244699",
"0.5235492",
"0.5232679",
"0.5227746",
"0.5176148",
"0.5109751",
"0.5095588",
"0.5087228",
"0.505932",
"0.5014933",
"0.5012798",
"0.49798912",
"0.49677464",
"0.49254423",... | 0.6224574 | 0 |
Type variables in the ``GenericAlias``. | def __parameters__(self) -> tuple[TypeVar, ...]:
return super().__getattribute__("_parameters") | {
"objective": {
"self": [],
"paired": [],
"triplet": [
[
"query",
"document",
"negatives"
]
]
}
} | [
"def _variable_types(self):\n return self._variable_single_types + self._variable_array_types",
"def injectTypes (g):\n\tself=__module__\n\ts=g.symbols\n\tg.token('TYPE_VAR', '_|[A-Z][A-Z0-9]*')\n\tg.rule('TypeParameter', s.LSB, listOf(g.agroup(s.TYPE_VAR, s.FQNAME), s.COMMA, g), s.RSB)\n\tg.rule('TypeRefe... | [
"0.64707834",
"0.60676146",
"0.5990629",
"0.59615344",
"0.588528",
"0.5847107",
"0.5746555",
"0.56999505",
"0.55221957",
"0.54898584",
"0.5446973",
"0.5353904",
"0.53393316",
"0.53158456",
"0.5303082",
"0.525656",
"0.52356243",
"0.5229733",
"0.5203652",
"0.52030146",
"0.51546... | 0.52475464 | 16 |
Check if an `obj` is an instance. | def __instancecheck__(self, obj: object) -> NoReturn:
raise TypeError("isinstance() argument 2 cannot be a "
"parameterized generic") | {
"objective": {
"self": [],
"paired": [],
"triplet": [
[
"query",
"document",
"negatives"
]
]
}
} | [
"def assert_is_instance(self, obj, cls, msg=\"\"):\r\n assert isinstance(obj, cls)",
"def is_dataclass_instance(obj: Any) -> bool:\n return dataclasses.is_dataclass(obj) and not isinstance(obj, type)",
"def assertIsInstance(self, obj, cls, msg=None):\r\n if not _is_instance(obj, cls):\r\n ... | [
"0.77252495",
"0.76500654",
"0.7596194",
"0.7542007",
"0.7537573",
"0.74013054",
"0.7116655",
"0.71137726",
"0.70612186",
"0.7052341",
"0.6999939",
"0.69481283",
"0.6900092",
"0.6855783",
"0.6831825",
"0.67850405",
"0.67850405",
"0.6783268",
"0.67796606",
"0.67789185",
"0.676... | 0.7006438 | 10 |
Check if a `cls` is a subclass. | def __subclasscheck__(self, cls: type) -> NoReturn:
raise TypeError("issubclass() argument 2 cannot be a "
"parameterized generic") | {
"objective": {
"self": [],
"paired": [],
"triplet": [
[
"query",
"document",
"negatives"
]
]
}
} | [
"def is_proper_subclass(cls, maybe_proper_superclass):\n return cls is not maybe_proper_superclass and issubclass(cls, maybe_proper_superclass)",
"def is_subclass(parent_class, child_class_name):\n for child_class in parent_class.__subclasses__():\n if child_class.__name__ == child_class_name:\n ... | [
"0.7876223",
"0.7705129",
"0.7567947",
"0.75527894",
"0.7437532",
"0.7431865",
"0.7421365",
"0.7377788",
"0.7368424",
"0.7331841",
"0.7313958",
"0.7293447",
"0.72841907",
"0.7261879",
"0.7256828",
"0.7247368",
"0.7236065",
"0.72272855",
"0.7211588",
"0.7210452",
"0.71509",
... | 0.70036453 | 31 |
Return ``self == value``. | def __eq__(self, value: object) -> bool:
if not isinstance(value, _GENERIC_ALIAS_TYPE):
return NotImplemented
return (
self.__origin__ == value.__origin__ and
self.__args__ == value.__args__ and
self.__unpacked__ == getattr(
value, "__unpac... | {
"objective": {
"self": [],
"paired": [],
"triplet": [
[
"query",
"document",
"negatives"
]
]
}
} | [
"def __eq__(self, other):\n return self.value == other.value",
"def __eq__(self, other):\n return self.value == other or self.value == other.value",
"def __eq__(self, other):\n if other is None:\n return False\n if self.value == other.value:\n return True\n ... | [
"0.81785744",
"0.81545824",
"0.7910936",
"0.77516854",
"0.7656241",
"0.7656241",
"0.76422757",
"0.7580233",
"0.7384331",
"0.7323974",
"0.72412944",
"0.72412944",
"0.7133235",
"0.70649093",
"0.7051817",
"0.70497334",
"0.70427805",
"0.70339525",
"0.70094466",
"0.6975119",
"0.69... | 0.6931704 | 23 |
Generates a population of bitarray strings (individuals) to solve the NQueen problem, here N = 'board_size'. 'pop_size' parameter decides the number of bitarray strings (individuals) generated by this function. Creates a list of bitarrray strings (individuals) representing the population for solving the NQueen problem.... | def genPopulation(self):
self.population_list = []
for i in xrange(0, self.pop_size):
individual = bitarray(self.indv_size)
# Loop for randomizing the 'individual' string.
for j in xrange(0, self.board_size):
vert_pos = random.randint(0, self.boa... | {
"objective": {
"self": [],
"paired": [],
"triplet": [
[
"query",
"document",
"negatives"
]
]
}
} | [
"def create_population(board_size, population_size):\n return [Nonogram(board_size) for x in range(0, population_size)]",
"def generate_population(population_size, nn_architecture):\n population = []\n for _ in range(population_size):\n population.append(nn.create_nn_from_arch(nn_architecture))\n\... | [
"0.72544706",
"0.67998624",
"0.6587251",
"0.65856385",
"0.64451057",
"0.6369007",
"0.62210387",
"0.6044195",
"0.6014805",
"0.58983636",
"0.58905756",
"0.58675313",
"0.58664834",
"0.5811403",
"0.5697839",
"0.5692206",
"0.56726235",
"0.5658943",
"0.5640834",
"0.5509586",
"0.548... | 0.727817 | 0 |
Populates the fitness list with fitness function values of coresponding entries in the population list. | def computeFitnessList(self,fitnessFunction):
self.fitness_list = []
cmlsum = 0
for individual in self.population_list:
cmlsum = cmlsum + fitnessFunction(individual, self.board_size, self.pos_bits_size)
self.fitness_list.append(cmlsum) | {
"objective": {
"self": [],
"paired": [],
"triplet": [
[
"query",
"document",
"negatives"
]
]
}
} | [
"def init_population(self):\n for idx in xrange(0, self.population_size):\n individual = self.individual_factory.create()\n self.population.append(individual)\n\n self.population_fitness = numpy.asarray(map(lambda individual: individual.get_fitness(), self.population))\n\n ... | [
"0.7050818",
"0.7024004",
"0.6900333",
"0.68683046",
"0.67409855",
"0.6698061",
"0.6585672",
"0.6527284",
"0.65243554",
"0.6502384",
"0.6475374",
"0.6410031",
"0.6401292",
"0.6385652",
"0.6321939",
"0.6284058",
"0.62710774",
"0.6265765",
"0.6243417",
"0.62331927",
"0.6226014"... | 0.7070026 | 0 |
Binary Search for finding appropriate index for key in fitness list. | def findInFitnessList(self, key):
low = 0
high = len(self.fitness_list)
mid = 0
while(low <= high):
mid = (low + high)/2
if key > self.fitness_list[mid]:
low = mid + 1
elif key < self.fitness_list[mid]:
high = m... | {
"objective": {
"self": [],
"paired": [],
"triplet": [
[
"query",
"document",
"negatives"
]
]
}
} | [
"def binary_search(self, num_lst, key):\r\n # Running time: O(log n) with O(n logn) overhead\r\n # get sorted list\r\n num_lst = sorted(num_lst)\r\n \r\n low, high, idx = 0, len(num_lst), -1\r\n \r\n while low < high:\r\n mid = int(math.floor((low+high) / ... | [
"0.71013904",
"0.70412946",
"0.69745183",
"0.68652153",
"0.6833065",
"0.68021667",
"0.68017167",
"0.6780906",
"0.6739357",
"0.6714616",
"0.6690004",
"0.65812993",
"0.6545252",
"0.6515856",
"0.65082943",
"0.6490799",
"0.6479905",
"0.64732397",
"0.64555985",
"0.6413893",
"0.636... | 0.78341144 | 0 |
Converts a number in int format to a bitarray of length 'size', in binary representation. Returns the bitarray. | def toBitArray(number, size):
temp_bitnum = bitarray(64)
count = 0
number = number & 0xFFFFFFFFFFFFFFFF # enforces the number to be 64 bit.
while count < size:
temp_bitnum[63 - count] = (number % 2)
# print "digit ", count, " : ", (number % 2)
number = number >> 1
... | {
"objective": {
"self": [],
"paired": [],
"triplet": [
[
"query",
"document",
"negatives"
]
]
}
} | [
"def binary(num, size=32):\n ret = np.zeros(size, dtype=np.int)\n n = np.array([int(x) for x in list(bin(num)[2:])])\n ret[ret.size - n.size:] = n\n return ret",
"def binary(num, size=32):\n ret = np.zeros(size, dtype=np.int)\n n = np.array([int(x) for x in list(bin(num)[2:])])\n ret[ret.size... | [
"0.7790668",
"0.7790668",
"0.7245732",
"0.6670738",
"0.64003235",
"0.6200373",
"0.6070387",
"0.60702974",
"0.60172224",
"0.5953754",
"0.5915167",
"0.591251",
"0.58911425",
"0.5887502",
"0.5887502",
"0.5879545",
"0.586908",
"0.5858048",
"0.58292425",
"0.58278644",
"0.5824009",... | 0.7832017 | 0 |
Converts a bitarray in binary format to a number in int format, to its coresponding decimal representation. Returns the number. | def fromBitArray(bitnum):
number = 0 & 0xFFFFFFFFFFFFFFFF # enforces the number to be 64 bit.
idx = len(bitnum) - 1
number = number + bitnum[idx]
idx = idx - 1
count = 1
while idx != -1:
number = number + (bitnum[idx] << count)
count = count + 1
idx = idx - 1
... | {
"objective": {
"self": [],
"paired": [],
"triplet": [
[
"query",
"document",
"negatives"
]
]
}
} | [
"def convert_bit_array_to_decimal(bit_array):\n decimal_int = 0\n ii = 0\n for i in np.arange(-(len(bit_array)), 0)*-1-1:\n if bit_array[ii]>0:\n decimal_int += pow(2.,i)\n ii+=1\n return int(decimal_int)",
"def bitarray2dec(in_bitarray):\n\n number = 0\n\n for i in rang... | [
"0.7937204",
"0.7795629",
"0.7240436",
"0.6926016",
"0.6884594",
"0.6877382",
"0.6808722",
"0.6694597",
"0.6669835",
"0.6577601",
"0.64981467",
"0.64833474",
"0.6417316",
"0.6362759",
"0.634778",
"0.6308411",
"0.6292181",
"0.62766623",
"0.62646604",
"0.62482375",
"0.6226497",... | 0.6684726 | 8 |
Calculates the finess value of an individual and returns it. Has a computational complexity of O(1) per piece. | def fitnessFunction(individual, board_size, pos_bits_size):
right_diag = [0] * (2 * board_size - 1)
left_diag = [0] * (2 * board_size - 1)
vertical = [0] * board_size
conflicts = 0
idx = 0
while idx < board_size:
# print "idx: ",idx,individual[idx * pos_bits_size : idx * pos_bits... | {
"objective": {
"self": [],
"paired": [],
"triplet": [
[
"query",
"document",
"negatives"
]
]
}
} | [
"def frobenius(self, value):\n return sum(self.hadamard(value))",
"def relative_abundance_fodder(self, animal):\n species = animal.__class__.__name__\n return self.relevant_food(animal) / (\n (len(self.fauna_list[species]) + 1) * animal.parameters['F'])",
"def get_fare(self):... | [
"0.6012158",
"0.576322",
"0.5754833",
"0.56664544",
"0.5625633",
"0.5573023",
"0.555937",
"0.5543563",
"0.5542946",
"0.55332303",
"0.5504458",
"0.54831654",
"0.5437056",
"0.5428136",
"0.5416342",
"0.54088235",
"0.5377262",
"0.5372197",
"0.5363929",
"0.5355039",
"0.53508043",
... | 0.0 | -1 |
Initialize a symmetry operation object from a string representation of symmetry operation. | def __init__(self, operation_string):
self.operation_string = operation_string.lower()
self.operation_function = None
self.__set_operation_function() | {
"objective": {
"self": [],
"paired": [],
"triplet": [
[
"query",
"document",
"negatives"
]
]
}
} | [
"def __init__(self, op, symbolicExpression1, symbolicExpression2):\n\n SymbolicExpression.__init__(self)\n \n self.op = op\n self.symbolicExpression1 = symbolicExpression1\n self.symbolicExpression2 = symbolicExpression2",
"def __init__(self, operand_string):\n\... | [
"0.65855575",
"0.6394667",
"0.60201573",
"0.58792925",
"0.5829858",
"0.5817785",
"0.58118224",
"0.5685034",
"0.5679989",
"0.5663519",
"0.5663512",
"0.5659587",
"0.5653366",
"0.5648754",
"0.5644039",
"0.56371164",
"0.56318647",
"0.56213385",
"0.5609781",
"0.5595345",
"0.555548... | 0.66930246 | 0 |
Convert the string form of the symmetry operation into the form of a mathematical function that can be directly applied to a vector to transform a point to a symmetry related point. | def __set_operation_function(self):
if self.operation_function is not None:
return self.operation_function
else:
self.operation_function = symm_eval | {
"objective": {
"self": [],
"paired": [],
"triplet": [
[
"query",
"document",
"negatives"
]
]
}
} | [
"def process_fn(fn_string, symbols):\n fn_string = fn_string.replace('^', '**')\n fn = lambdify([sympy.symbols(symbols)], fn_string, 'numpy')\n return fn",
"def matrix_sym_op(x):\n return (x + tf.transpose(x))/2",
"def convertStoichiometryMath(self):\n return _libsbml.Model_convertStoichiomet... | [
"0.64556384",
"0.5763819",
"0.5679932",
"0.5665814",
"0.5638721",
"0.55064327",
"0.5426737",
"0.54216206",
"0.5392853",
"0.53914815",
"0.53572667",
"0.53297466",
"0.5317339",
"0.5299851",
"0.5280908",
"0.527228",
"0.52579755",
"0.52576876",
"0.52001554",
"0.513768",
"0.513490... | 0.47996485 | 70 |
Applying this symmetry operation on a 3D coordinate to transform it to a symmetryrelated position in the crystal. | def transform_scaled_position(self, data):
return self.operation_function(prepare_operation(self.operation_string), data) | {
"objective": {
"self": [],
"paired": [],
"triplet": [
[
"query",
"document",
"negatives"
]
]
}
} | [
"def sat_3d_position(sat_2d_position):\n return np.dot(transformation_parameter, xi_eta(sat_2d_position))",
"def convert_coordinate_system_3d(x, y, z):\n\n return x, -z, y",
"def LEIsotropic3D(self):\n const = self.ymod / ((1+self.Nu) * (1-(2*self.Nu)))\n a = const * self.Nu\n b = con... | [
"0.6256417",
"0.62044483",
"0.5919591",
"0.5910444",
"0.5888323",
"0.588735",
"0.5876213",
"0.5783937",
"0.5778394",
"0.57703584",
"0.5756848",
"0.57547265",
"0.57113785",
"0.56860083",
"0.5667107",
"0.5617889",
"0.55534124",
"0.5551718",
"0.55438286",
"0.5534934",
"0.552158"... | 0.0 | -1 |
Given the LATT directive in a res file, return the corresponding centered lattice type. | def construct(cls, latt):
latt = abs(latt)
if latt == 1:
return cls.primitive()
elif latt == 2:
return cls.body_centered()
elif latt == 3:
# default setting from reading in a res file is Rhombohedral
return cls.rhombohedral()
elif l... | {
"objective": {
"self": [],
"paired": [],
"triplet": [
[
"query",
"document",
"negatives"
]
]
}
} | [
"def specified_lattice_constant(self):\n return load_node(self.get_attribute('specified_lattice_constant'))",
"def print_abs_type(self):\n return 'DLA'",
"def get_ligand_resname_from_topology( topfile ):\n\n file = open( topfile, 'r')\n text = file.readlines()\n file.close()\n indices ... | [
"0.5302461",
"0.51673067",
"0.51398134",
"0.4979288",
"0.48934972",
"0.48730606",
"0.48607278",
"0.48193753",
"0.47571704",
"0.473777",
"0.4727024",
"0.46984535",
"0.46911123",
"0.46570626",
"0.46286234",
"0.46027964",
"0.46002376",
"0.45856792",
"0.45732707",
"0.4515798",
"0... | 0.6210394 | 0 |
returns a list containing the al points of the bounding box, starting from bottom left and going anticlockwise | def boundingBox(points):
min_x, min_y = float('inf'), float('inf')
max_x, max_y = float('-inf'), float('-inf')
for x, _, y in points:
min_x = min(min_x, x)
min_y = min(min_y, y)
max_x = max(max_x, x)
max_y = max(max_y, y)
return (min_x, min_y), (max_x, min_y), (max_x, ma... | {
"objective": {
"self": [],
"paired": [],
"triplet": [
[
"query",
"document",
"negatives"
]
]
}
} | [
"def boundingBox(self):\n y_max = np.max(self.points[:,0])\n x_max = np.max(self.points[:,1])\n y_min = np.min(self.points[:,0])\n x_min = np.min(self.points[:,1])\n \n return ((x_max, y_max), (x_min, y_min))",
"def bbox(self):\n return [self._x0, self._y0, self._x... | [
"0.75965375",
"0.7497875",
"0.74912584",
"0.7396015",
"0.73149467",
"0.7234918",
"0.7175359",
"0.7174177",
"0.71489733",
"0.7111562",
"0.71065605",
"0.70789665",
"0.7042901",
"0.704142",
"0.7023269",
"0.7008294",
"0.69650996",
"0.69617414",
"0.69278795",
"0.6900412",
"0.68911... | 0.7039391 | 14 |
Basic plotting feature. Can be used standalone, in which case you can ignore the 'newfig' and 'plotfig' parameters. Otherwise, you may suppress these by passing False and using the returned axes object. | def basicplot(data, plotter='plot', newfig=True, showplot=True, **kwargs):
if plotter not in ('plot', 'step'):
raise ValueError
if newfig:
plt.figure(figsize=kwargs.get('figsize', (8, 6)))
# ............................................................................ Plot per type
if isinstance(... | {
"objective": {
"self": [],
"paired": [],
"triplet": [
[
"query",
"document",
"negatives"
]
]
}
} | [
"def init_axes(self):\n plt.switch_backend(\"cairo\")\n fig = plt.figure(figsize=(15,10))\n ax = fig.add_axes([0.05, 0.15, 0.9, 0.80,])\n return (fig, ax)",
"def _newax(ax=None):\n from matplotlib import pyplot as plt\n if ax is not None:\n return ax\n fig = plt.figure(... | [
"0.6635028",
"0.6569405",
"0.64893675",
"0.6439919",
"0.6360829",
"0.6326854",
"0.6290503",
"0.627174",
"0.62165153",
"0.6209581",
"0.6181125",
"0.61717874",
"0.6159054",
"0.6155145",
"0.6095297",
"0.60936546",
"0.60824245",
"0.60354114",
"0.6033826",
"0.6010859",
"0.59907717... | 0.6620118 | 1 |
Template for polishing a matplotlib plot. | def basic_plot_polishing(ax, **kwargs):
# Title
ax.set_title(kwargs.get('title'), **kwargs.get('title_kw', {}))
# ............................................... X- and Y-axes
# Axes Labels
ax.set_xlabel(kwargs.get('xlabel'), **kwargs.get('xlabel_kw', {}))
ax.set_ylabel(kwargs.get('ylabel'), **kwarg... | {
"objective": {
"self": [],
"paired": [],
"triplet": [
[
"query",
"document",
"negatives"
]
]
}
} | [
"def builtin_plot(self, **kwargs):\n self.gp.plot(**kwargs)\n return",
"def plot():\n pass",
"def evaluate(self, plot):",
"def plot(self, ax=..., *, name=..., **kwargs):\n ...",
"def plot_figure(param1, param2):\n return 0",
"def PygletPlot(*args, **kwargs):\n\n from sympy.plotti... | [
"0.65788615",
"0.64977914",
"0.63758653",
"0.6249835",
"0.6161063",
"0.61453396",
"0.6137437",
"0.61225075",
"0.61015445",
"0.6099689",
"0.60799134",
"0.6052406",
"0.5981926",
"0.5980428",
"0.5944208",
"0.59270716",
"0.5923354",
"0.59175104",
"0.5909176",
"0.59045833",
"0.589... | 0.60760945 | 11 |
delegates to base interface run method, then attempts to generate reports | def _run_interface(self, runtime):
try:
runtime = super(
ReportCapableInterface, self)._run_interface(runtime)
except NotImplementedError:
pass # the interface is derived from BaseInterface
# leave early if there's nothing to do
if not self.input... | {
"objective": {
"self": [],
"paired": [],
"triplet": [
[
"query",
"document",
"negatives"
]
]
}
} | [
"def _generate_report(self):\n raise NotImplementedError",
"def __execute_reporter(self):\n if not self.__args.report:\n return\n reporter.HTMLReporter().generate_report_from_file(\n self.__lst_json_files)",
"def run(self):\n report_file = self.get_report_file_n... | [
"0.7437408",
"0.73895234",
"0.7325222",
"0.7179931",
"0.6891885",
"0.6887828",
"0.6871864",
"0.6871864",
"0.67655283",
"0.67655283",
"0.67655283",
"0.67655283",
"0.67655283",
"0.67655283",
"0.67655283",
"0.67655283",
"0.67655283",
"0.6731759",
"0.67242306",
"0.6709329",
"0.66... | 0.7569578 | 0 |
A placeholder to run stuff after the normal execution of the interface (i.e. assign proper inputs to reporting functions) | def _post_run_hook(self, runtime):
pass | {
"objective": {
"self": [],
"paired": [],
"triplet": [
[
"query",
"document",
"negatives"
]
]
}
} | [
"def postRun(self):\n pass",
"def after_test(self, func, *args, **kwargs):\n pass",
"def after_step():\n raise NotImplementedError",
"def post_execute(self):",
"def post_exec(self):\n raise NotImplementedError(\"Must implement in frontend subclass.\")",
"def _initialise_run(self) -> No... | [
"0.64498556",
"0.63633126",
"0.633806",
"0.6311236",
"0.62261844",
"0.622094",
"0.62078357",
"0.6205322",
"0.61407",
"0.6081844",
"0.6081844",
"0.6050914",
"0.60314965",
"0.6028222",
"0.6022563",
"0.59925395",
"0.5990467",
"0.5945822",
"0.5933975",
"0.5908431",
"0.5903639",
... | 0.5792072 | 25 |
Saves an svg object. | def _generate_report(self):
raise NotImplementedError | {
"objective": {
"self": [],
"paired": [],
"triplet": [
[
"query",
"document",
"negatives"
]
]
}
} | [
"def save_svg(string, file_name):\n file_handle = file(file_name, \"w\")\n file_handle.write(string)\n file_handle.close()",
"def save_canvas_svg(self, filename):\n canvasvg.saveall(filename, self.canvas)",
"def export_as_svg(self):\n from ExportCommand import ExportCommand\n\n opt... | [
"0.7298411",
"0.72474957",
"0.7201746",
"0.69764864",
"0.69075066",
"0.68920994",
"0.6787797",
"0.65999645",
"0.6587811",
"0.65750474",
"0.6531958",
"0.65172744",
"0.64560294",
"0.64560294",
"0.6405152",
"0.6390808",
"0.6390808",
"0.6390808",
"0.6378808",
"0.63608027",
"0.634... | 0.0 | -1 |
Saves an svg snippet | def _generate_error_report(self, errno=None):
# as of now we think this will be the same for every interface
NIWORKFLOWS_LOG.warn('Report was not generated')
errorstr = '<div><span class="error">Failed to generate report!</span>.\n'
if errno:
errorstr += (' <span class="erro... | {
"objective": {
"self": [],
"paired": [],
"triplet": [
[
"query",
"document",
"negatives"
]
]
}
} | [
"def save_svg(string, file_name):\n file_handle = file(file_name, \"w\")\n file_handle.write(string)\n file_handle.close()",
"def save_plot(p, file_name, path='../static/images/'):\n p.output_backend = \"svg\"\n export_svgs(p, filename=path + file_name + '.svg')",
"def save(filename, canvas):\n ... | [
"0.7796538",
"0.72046244",
"0.7060236",
"0.70434374",
"0.694006",
"0.6827847",
"0.6754541",
"0.65522254",
"0.65489715",
"0.65106666",
"0.6446291",
"0.6428489",
"0.6419784",
"0.6399665",
"0.63871783",
"0.6345695",
"0.63159513",
"0.6313837",
"0.6306657",
"0.6256271",
"0.6255564... | 0.0 | -1 |
Generates the visual report | def _generate_report(self):
from niworkflows.viz.utils import plot_registration
NIWORKFLOWS_LOG.info('Generating visual report')
fixed_image_nii = load_img(self._fixed_image)
moving_image_nii = load_img(self._moving_image)
contour_nii = load_img(self._contour) if self._contour i... | {
"objective": {
"self": [],
"paired": [],
"triplet": [
[
"query",
"document",
"negatives"
]
]
}
} | [
"def _generate_report(self):\n from niworkflows.viz.utils import plot_registration\n NIWORKFLOWS_LOG.info('Generating visual report')\n\n anat = load_img(self._anat_file)\n contour_nii = load_img(self._contour) if self._contour is not None else None\n\n if self._mask_file:\n ... | [
"0.7264522",
"0.72419435",
"0.72344077",
"0.7205322",
"0.7167486",
"0.70792633",
"0.69737315",
"0.68265104",
"0.67954844",
"0.6770184",
"0.6706242",
"0.6649476",
"0.6630725",
"0.66305715",
"0.66226053",
"0.66017485",
"0.6577003",
"0.6519854",
"0.6497918",
"0.64795333",
"0.645... | 0.70916015 | 5 |
Generates the visual report | def _generate_report(self):
from niworkflows.viz.utils import plot_registration
NIWORKFLOWS_LOG.info('Generating visual report')
anat = load_img(self._anat_file)
contour_nii = load_img(self._contour) if self._contour is not None else None
if self._mask_file:
anat = ... | {
"objective": {
"self": [],
"paired": [],
"triplet": [
[
"query",
"document",
"negatives"
]
]
}
} | [
"def generate_report(self) -> Report:\n # equity_curve = self._generate_equity_curve()\n # summary_stats = self._generate_summary_stats(equity_curve)\n # return Report(equity_curve, summary_stats)\n pass",
"def _generate_report(self):\n raise NotImplementedError",
"def generat... | [
"0.72419435",
"0.72344077",
"0.7205322",
"0.7167486",
"0.70916015",
"0.70792633",
"0.69737315",
"0.68265104",
"0.67954844",
"0.6770184",
"0.6706242",
"0.6649476",
"0.6630725",
"0.66305715",
"0.66226053",
"0.66017485",
"0.6577003",
"0.6519854",
"0.6497918",
"0.64795333",
"0.64... | 0.7264522 | 0 |
Initialize a new Calibration | def __init__(self, poses=None, selected_point=None):
super(Calibration, self).__init__("calibration")
self.marker_array = MarkerArray()
if selected_point is None:
selected_point = 0
if poses is not None:
self.poses = poses
else:
self.load()
... | {
"objective": {
"self": [],
"paired": [],
"triplet": [
[
"query",
"document",
"negatives"
]
]
}
} | [
"def __init__(self):\n self._read_calibration_data()\n self.configure_sensor(\n TemperatureOversamplings.x08,\n PressureOversamplings.x16,\n HumidityOversamplings.x08,\n IIRFilterCoefficients.FC_003,\n 250,\n 250)",
"def initialise_ca... | [
"0.7170853",
"0.7100124",
"0.7099024",
"0.7032544",
"0.7022527",
"0.6957133",
"0.6374656",
"0.63612366",
"0.6234849",
"0.6194527",
"0.6192426",
"0.61858934",
"0.6176256",
"0.61386317",
"0.6112563",
"0.61066556",
"0.60895455",
"0.6060279",
"0.6034259",
"0.6027615",
"0.59995145... | 0.73918635 | 0 |
Calibration changed event, publish the markers and the calibration data to ROS | def calibration_changed(self):
for marker in self.marker_array.markers:
marker.action = Marker.DELETE
for idx, pose in enumerate(self.poses):
marker = Marker()
marker.id = int(str(int(round(time.time() * 1000)) + idx)[6:])
marker.header.frame_id = Calibr... | {
"objective": {
"self": [],
"paired": [],
"triplet": [
[
"query",
"document",
"negatives"
]
]
}
} | [
"def _doCalibration(self):\n self._cmdCalibration(2)",
"def rDataChanged(self):\n\n self._queues.uResolutionTab.refreshData()\n self._layerManager.updateReviewLayer()",
"def fDataChanged(self):\n\n self._layerManager.getAimsFeatures()",
"def XPLMDataChanged_f(inRefcon):",
"def ge... | [
"0.59780914",
"0.5974194",
"0.57603115",
"0.5716506",
"0.5700313",
"0.56652844",
"0.551917",
"0.550133",
"0.548774",
"0.546406",
"0.54577005",
"0.5453123",
"0.5446209",
"0.54401296",
"0.5424774",
"0.5401204",
"0.53942555",
"0.5393695",
"0.5350199",
"0.5338598",
"0.5327624",
... | 0.7535387 | 0 |
Add new model point to the calibration points | def add(self, p):
self.poses.append(CalibrationPoint(p))
self.selected_point = len(self.poses) - 1
self.calibration_changed() | {
"objective": {
"self": [],
"paired": [],
"triplet": [
[
"query",
"document",
"negatives"
]
]
}
} | [
"def edit(self, p):\n self.poses[self.selected_point].model = p\n self.calibration_changed()",
"def add_point(self, xx, fx):\n dim = len(xx)\n self._realloc(dim)\n self.x[self.nump, :] = xx\n self.fx[self.nump, :] = fx\n self.nump += 1\n # Update the leave-o... | [
"0.69154406",
"0.6766578",
"0.6623721",
"0.6548858",
"0.6522341",
"0.6482697",
"0.6431424",
"0.63752323",
"0.63269496",
"0.6270424",
"0.6258767",
"0.62286866",
"0.6175748",
"0.6123516",
"0.61176217",
"0.61059564",
"0.6047522",
"0.5996935",
"0.59870833",
"0.5979229",
"0.593303... | 0.7142621 | 0 |
Edit the seleted model point | def edit(self, p):
self.poses[self.selected_point].model = p
self.calibration_changed() | {
"objective": {
"self": [],
"paired": [],
"triplet": [
[
"query",
"document",
"negatives"
]
]
}
} | [
"def _point_edited(self, _var_name, _list_index, _operation):\r\n #update the point only if:\r\n #- the user edited the var (not just a point selection)\r\n #- the input string can be parsed to ints\r\n if (not self.inhibit_callbacks and\r\n is_float(self.editable_x.get())... | [
"0.7051569",
"0.67868155",
"0.6757438",
"0.66552246",
"0.65705884",
"0.6387175",
"0.6364775",
"0.6216918",
"0.62072617",
"0.59917426",
"0.5985266",
"0.59831333",
"0.59759885",
"0.5964893",
"0.58861",
"0.5845197",
"0.58253455",
"0.5799752",
"0.57940185",
"0.5790111",
"0.579011... | 0.7967514 | 0 |
Record the real world position of the selected model point | def record(self):
if self.selected_point is None:
RosProxy().notify("No calibration point selected", STATE.ERROR)
return
if len(self.poses) == 0:
RosProxy().notify("No calibration point added", STATE.ERROR)
return
if RobotControllerHandler.curre... | {
"objective": {
"self": [],
"paired": [],
"triplet": [
[
"query",
"document",
"negatives"
]
]
}
} | [
"def _world_point(self, point_3d):\n return self.obj.matrix_world @ point_3d",
"def coord(self):\r\n return self.model.coord",
"def edit(self, p):\n self.poses[self.selected_point].model = p\n self.calibration_changed()",
"def record_pose(self):\n drone_pos = client.getMulti... | [
"0.661618",
"0.6589321",
"0.6392966",
"0.6373264",
"0.6324207",
"0.6305534",
"0.6298948",
"0.6229923",
"0.6171466",
"0.61672115",
"0.60906225",
"0.60764337",
"0.6038905",
"0.6004793",
"0.6004699",
"0.6002335",
"0.5980551",
"0.5974565",
"0.59389585",
"0.593828",
"0.5920619",
... | 0.6123218 | 10 |
Remove selected calibration point | def remove(self):
if self.selected_point is None:
RosProxy().notify("No calibration point selected", STATE.ERROR)
return
if len(self.poses) == 0:
RosProxy().notify("No calibration point added", STATE.ERROR)
return
self.poses.remove(self.poses[se... | {
"objective": {
"self": [],
"paired": [],
"triplet": [
[
"query",
"document",
"negatives"
]
]
}
} | [
"def delete_selection(self):\n if self.selected_point_index is not None:\n del self.current_shape[self.selected_point_index]\n self.selected_point_index = None\n self.changed()",
"def rm_calibration(self):\n\n self.bin_edges_kev = None",
"def removePick(self):\n ... | [
"0.7046191",
"0.68148166",
"0.6728017",
"0.6607091",
"0.6529173",
"0.63954777",
"0.6347519",
"0.60552454",
"0.6048657",
"0.6048657",
"0.60466725",
"0.6028333",
"0.601033",
"0.60037434",
"0.60037434",
"0.5995959",
"0.5992583",
"0.59890836",
"0.5910162",
"0.5888631",
"0.5885889... | 0.8378994 | 0 |
Load a given object | def load_object(self, obj):
self.poses = obj.poses
self.selected_point = obj.selected_point
self.calibration_changed() | {
"objective": {
"self": [],
"paired": [],
"triplet": [
[
"query",
"document",
"negatives"
]
]
}
} | [
"def load_object(path):\r\n with open(path,\"rb\") as f:\r\n object = pickle.load(f) \r\n return object",
"def load_object(path):\n try:\n dot = path.rindex('.')\n except ValueError:\n raise ValueError(\"Error loading object '%s': not a full path\" % path)\n\n modul... | [
"0.7695731",
"0.74446046",
"0.7410183",
"0.73371464",
"0.73294044",
"0.72880054",
"0.7247311",
"0.7246651",
"0.7245202",
"0.7236421",
"0.7236421",
"0.7191883",
"0.71651185",
"0.71360666",
"0.71178216",
"0.7112144",
"0.69774926",
"0.6947777",
"0.68680066",
"0.68327016",
"0.682... | 0.6548898 | 31 |
Load a default calibration as an example | def load_default(self):
poses = []
cp = CalibrationPoint(Pose(), Pose())
cp.measured.position = Point(0.0, 0.0, 0.12)
cp.measured.orientation.x = -0.025
cp.measured.orientation.y = 1.
cp.measured.orientation.z = 0.011
cp.measured.orientation.w = 0.002
cp... | {
"objective": {
"self": [],
"paired": [],
"triplet": [
[
"query",
"document",
"negatives"
]
]
}
} | [
"def __init__ (self) :\n self.loadCSPAD2x2CalibParsDefault()",
"def load_cal(self):\n global DEBUG, dtParameterDesc\n if DEBUG:\n print(self.__class__.__name__+'.load_cal(): calibration is loaded')\n if 'refatt' in self.parameters and 'refoutpower' in self.parameters:\n ... | [
"0.64449143",
"0.62831944",
"0.59671456",
"0.5753438",
"0.5722229",
"0.570451",
"0.5687627",
"0.5635133",
"0.558993",
"0.55866563",
"0.55388427",
"0.55273575",
"0.5505142",
"0.5503746",
"0.5497588",
"0.5470468",
"0.5460154",
"0.5457385",
"0.5434277",
"0.5433322",
"0.54281676"... | 0.6843389 | 0 |
Goto to the selected position | def goto(self, speed=1):
self.safe_goto(speed, 0) | {
"objective": {
"self": [],
"paired": [],
"triplet": [
[
"query",
"document",
"negatives"
]
]
}
} | [
"def __goto(self):\n from QScintilla.GotoDialog import GotoDialog\n \n aw = self.activeWindow()\n lines = aw.lines()\n curLine = aw.getCursorPosition()[0] + 1\n dlg = GotoDialog(lines, curLine, self.ui, None, True)\n if dlg.exec_() == QDialog.Accepted:\n a... | [
"0.75331914",
"0.744304",
"0.73540735",
"0.71401715",
"0.7106989",
"0.70884335",
"0.70847714",
"0.69586277",
"0.6954708",
"0.686003",
"0.6822451",
"0.6774556",
"0.6698407",
"0.66824645",
"0.6607257",
"0.64744914",
"0.6432751",
"0.6329947",
"0.6266078",
"0.62590426",
"0.624844... | 0.5589648 | 85 |
Safe goto to the selected position (the Z is increased with 2cm) | def safe_goto(self, speed=1.0, distance=SAFETY_DISTANCE):
moves = []
point = copy.deepcopy(
self.poses[self.selected_point].measured.position)
orientation = self.poses[self.selected_point].measured.orientation
point.z += distance
move = Move()
move.pose.posit... | {
"objective": {
"self": [],
"paired": [],
"triplet": [
[
"query",
"document",
"negatives"
]
]
}
} | [
"def move_to_position2(self):",
"def move_to_position1(self):",
"def goto(x, y):\n turtleTmp.setposition(x, y)",
"def goto_position(editor, pos):\n cursor = editor.textCursor()\n editor.moveCursor(cursor.End)\n cursor.setPosition(pos)\n editor.setTextCursor(cursor)",
"def jump(self):\n ... | [
"0.7087748",
"0.7047049",
"0.6761017",
"0.6383471",
"0.6282584",
"0.6282203",
"0.62701017",
"0.62499607",
"0.6248314",
"0.6211238",
"0.6206618",
"0.6184875",
"0.61233246",
"0.6099291",
"0.6097524",
"0.6097524",
"0.6070837",
"0.6049628",
"0.60357076",
"0.603397",
"0.6002929",
... | 0.572487 | 48 |
Set the selected point | def set_selected_point(self, i):
if i < len(self.poses):
self.selected_point = min(len(self.poses), max(0, i))
self.calibration_changed() | {
"objective": {
"self": [],
"paired": [],
"triplet": [
[
"query",
"document",
"negatives"
]
]
}
} | [
"def _on_point_selected(self, _event):\r\n selected_iid = self._tree.selection()\r\n self._index_of_sel_point = self._tree.index(selected_iid)\r\n self._edit_zone.set_editable_point(self._tree.item(selected_iid)[\"values\"][0])\r\n self._notify(\"focus\", {})",
"def setSetpoint(self, p... | [
"0.7825963",
"0.780328",
"0.7588371",
"0.7408778",
"0.7331143",
"0.7323788",
"0.72938347",
"0.72737926",
"0.71781445",
"0.71363586",
"0.71363586",
"0.70924747",
"0.70924747",
"0.70924747",
"0.7033677",
"0.7014697",
"0.6998659",
"0.6984284",
"0.6957518",
"0.6957341",
"0.693528... | 0.77336913 | 2 |
Calibration setter service handler | def set_calibration(self, req):
cmd = req.command
data = json.loads(req.json)
if cmd == "add":
if "x" in data and "y" in data and "z" in data:
x, y, z = float(data["x"]), float(data["y"]), float(data["z"])
self.add(Point(x, y, z))
else:
... | {
"objective": {
"self": [],
"paired": [],
"triplet": [
[
"query",
"document",
"negatives"
]
]
}
} | [
"def _doCalibration(self):\n self._cmdCalibration(2)",
"def calibration(self, cal: int, /) -> None:",
"def __init__(self, poses=None, selected_point=None):\n super(Calibration, self).__init__(\"calibration\")\n\n self.marker_array = MarkerArray()\n if selected_point is None:\n ... | [
"0.6910026",
"0.6533155",
"0.64192873",
"0.6146642",
"0.60480934",
"0.6035951",
"0.6031567",
"0.6006333",
"0.5834597",
"0.58197874",
"0.57702",
"0.5759993",
"0.5720985",
"0.57105434",
"0.5708835",
"0.5705385",
"0.56633836",
"0.55804294",
"0.5518535",
"0.549152",
"0.54897946",... | 0.60191 | 7 |
Same as cloudvolume.paths.extract, but raise an additional cloudvolume.exceptions.UnsupportedProtocolError if either dataset or layer is not set. | def strict_extract(cloudpath, windows=None, disable_toabs=False):
path = extract(cloudpath, windows, disable_toabs)
if path.dataset == '' or path.layer == '':
raise UnsupportedProtocolError(cloudpath_error(cloudpath))
return path | {
"objective": {
"self": [],
"paired": [],
"triplet": [
[
"query",
"document",
"negatives"
]
]
}
} | [
"def extract(cloudpath, windows=None, disable_toabs=False):\n if len(cloudpath) == 0:\n return ExtractedPath('','','','','','','')\n\n windows_file_re = re.compile(r'((?:\\w:\\\\)[\\d\\w_\\.\\-]+)')\n bucket_re = re.compile(r'^(/?[~\\d\\w_\\.\\-]+(?::\\d+)?)(?:\\b|$)') # posix /what/a/great/path\n \n error ... | [
"0.57432127",
"0.5345072",
"0.5141064",
"0.48350242",
"0.471617",
"0.4686658",
"0.468103",
"0.46336773",
"0.462907",
"0.46199074",
"0.46039414",
"0.45926678",
"0.4592578",
"0.45727795",
"0.45651555",
"0.4543723",
"0.45103985",
"0.4503522",
"0.4497334",
"0.44893438",
"0.448435... | 0.7009644 | 0 |
Given a valid cloudpath of the form | def extract(cloudpath, windows=None, disable_toabs=False):
if len(cloudpath) == 0:
return ExtractedPath('','','','','','','')
windows_file_re = re.compile(r'((?:\w:\\)[\d\w_\.\-]+)')
bucket_re = re.compile(r'^(/?[~\d\w_\.\-]+(?::\d+)?)(?:\b|$)') # posix /what/a/great/path
error = UnsupportedProtocolErro... | {
"objective": {
"self": [],
"paired": [],
"triplet": [
[
"query",
"document",
"negatives"
]
]
}
} | [
"def path_validate(path):\n # functionality to be added later\n return path",
"def validate_short_path(short_path):",
"def parse_managed_path(path):\n fields = path.split(':', 1)\n return fields[0], fields[1]",
"def clean_path(self, path):\n if('.flaccuesplit.' in path):\n ... | [
"0.5782255",
"0.5485256",
"0.5111033",
"0.5109678",
"0.5098054",
"0.4996147",
"0.49742427",
"0.4955958",
"0.49552155",
"0.49309495",
"0.49085343",
"0.49030942",
"0.48504016",
"0.48399198",
"0.48375192",
"0.48148483",
"0.47820082",
"0.47807506",
"0.47638345",
"0.47635552",
"0.... | 0.5554715 | 1 |
This function checks to see if user guess the correct letter | def check_guess(letter, letters_pending, letters_found):
# correct guess
if letter in letters_pending:
letter_index = letters_pending[letter].pop(0) # return first letter's index
letters_found.add(letter_index)
# check if list is empty
if len(letters_pending[letter]) ==... | {
"objective": {
"self": [],
"paired": [],
"triplet": [
[
"query",
"document",
"negatives"
]
]
}
} | [
"def checkLetter():\n\tguess = False\n\twhile guess != True:\n\t\tguess = str(raw_input(\"Guess a letter: \"))\n\t\tif guess.isalpha() and len(guess) == 1 :\n\t\t\treturn guess\n\t\telif not guess.isalpha() or len(guess) > 1:\n\t\t\tprint \"The input may be one letter only!\"\n\t\telse:\n\t\t\tprint \"Error in chec... | [
"0.8326243",
"0.797419",
"0.7970673",
"0.78612447",
"0.78342074",
"0.7764369",
"0.77154046",
"0.7664183",
"0.7633212",
"0.7632182",
"0.74513006",
"0.73829585",
"0.7381481",
"0.7340003",
"0.73062885",
"0.72928417",
"0.72888887",
"0.7277303",
"0.7275198",
"0.7268439",
"0.722009... | 0.6817726 | 38 |
This function displays the letters that user found and underscores for letters remain to be found | def display_word_status(word, indexes_found):
status = []
for i in range(len(word)):
if i in indexes_found:
status.append(word[i])
else:
status.append('_')
print('\n[', " ".join(status), '] \n') | {
"objective": {
"self": [],
"paired": [],
"triplet": [
[
"query",
"document",
"negatives"
]
]
}
} | [
"def show_hidden_word(secret_word, old_letters_guessed):\n print(\"\\n\")\n for letter in secret_word:\n if letter in old_letters_guessed:\n print(letter, end=' ')\n else:\n print('_', end=' ')\n return ' '",
"def display_letters(word, guesses):\n pass",
"def show... | [
"0.70420015",
"0.6987924",
"0.69101435",
"0.68977433",
"0.67463523",
"0.674226",
"0.6739028",
"0.66745555",
"0.6641939",
"0.6510712",
"0.6438312",
"0.6392078",
"0.6339939",
"0.6339912",
"0.6318629",
"0.62372696",
"0.6155217",
"0.61542034",
"0.614742",
"0.6090914",
"0.6011447"... | 0.5992686 | 22 |
SCons entry point for this tool. | def generate(env):
# NOTE: SCons requires the use of this name, which fails gpylint.
# Set target platform bits
env.SetBits('debug')
env['TARGET_DEBUG'] = True
env.Append(
CPPDEFINES=['_DEBUG'] + env.get('CPPDEFINES_DEBUG', []),
CCFLAGS=env.get('CCFLAGS_DEBUG', []),
LINKFLAGS=env.get('LIN... | {
"objective": {
"self": [],
"paired": [],
"triplet": [
[
"query",
"document",
"negatives"
]
]
}
} | [
"def generate(env):\n\n # FIXME: this is already too late\n #if env.get('quiet', False):\n # quietCommandLines(env)\n\n # shortcuts\n debug = env['debug']\n machine = env['machine']\n platform = env['platform']\n x86 = env['machine'] == 'x86'\n gcc = env['platform'] in ('linux', 'freeb... | [
"0.6464546",
"0.6412459",
"0.6104805",
"0.6104805",
"0.60741913",
"0.6071515",
"0.5993396",
"0.59785044",
"0.59724134",
"0.59724134",
"0.59724104",
"0.59639555",
"0.59407467",
"0.5926329",
"0.5874594",
"0.58656955",
"0.5856603",
"0.58456635",
"0.5837887",
"0.5832198",
"0.5824... | 0.0 | -1 |
Extract tags and mentions from a string. Returns a tuple with a set of tags and mentioned users, if they exist. | def extract_tags_and_mentions(s: str) -> typing.Tuple[set, set]:
if not s:
return set(), set()
# Build a set after stripping the # or @ sign before every item found with
# the regular expression.
tags = set(t for t in find_hashtags_re.findall(s))
mentions = set(m[1:] for m in find_mentions_... | {
"objective": {
"self": [],
"paired": [],
"triplet": [
[
"query",
"document",
"negatives"
]
]
}
} | [
"def get_tagged_user(line, unique_users):\n tagged_user = None\n\n for user in unique_users:\n \n tagged_user = re.search(f\"@{user}\\s*\", line)\n \n if tagged_user != None:\n tagged_user = tagged_user.group(0).replace(\"@\", \"\").strip()\n line = line.replace(f\"@{... | [
"0.5984887",
"0.5866291",
"0.5816416",
"0.57498",
"0.57427794",
"0.56972706",
"0.5674774",
"0.5634183",
"0.5608683",
"0.5608683",
"0.5608683",
"0.5608683",
"0.5608683",
"0.56065565",
"0.56022114",
"0.55869055",
"0.55849683",
"0.5581752",
"0.5581752",
"0.5576589",
"0.5564691",... | 0.8448166 | 0 |
On post(type) creation, set hash_id. | def on_created_post(sender, instance, created, **kwargs):
if not created:
return
if not issubclass(sender, dillo.models.posts.Post):
return
instance.hash_id = instance.id
instance.save()
log.debug('Set user %s as follower of own post %i' % (instance.user, instance.id))
follow(ins... | {
"objective": {
"self": [],
"paired": [],
"triplet": [
[
"query",
"document",
"negatives"
]
]
}
} | [
"def handler(event, _context):\n model = PostModel()\n post_id = model.create(**json.loads(event['body']))\n return dump_result({'post_id': post_id}, status_code=201)",
"def save(self, *args, **kwargs):\n if not self.pkhash:\n self.pkhash = compute_hash(self.script)\n super(DataO... | [
"0.6146022",
"0.60286593",
"0.5837189",
"0.5832979",
"0.5738559",
"0.57207257",
"0.56405866",
"0.5623517",
"0.5464062",
"0.54599774",
"0.5413562",
"0.5413562",
"0.53702575",
"0.5368633",
"0.53681165",
"0.5332242",
"0.5332242",
"0.5330535",
"0.5305785",
"0.5294041",
"0.529331"... | 0.57402194 | 4 |
Assign Tags and Mentions to a Post(type) by parsing the description. | def on_saved_post(sender, instance, created, **kwargs):
# If the post was just created, stop here since we have no tags yet
if created:
return
if not issubclass(sender, dillo.models.posts.Post):
return
# Extract tags and mentions from text and assign them to the Post
tags, mentions =... | {
"objective": {
"self": [],
"paired": [],
"triplet": [
[
"query",
"document",
"negatives"
]
]
}
} | [
"def add_meta(self, post, *args, **kwargs):\n\t\tsuper(Command, self).add_meta(post, *args, **kwargs)\n\t\tpost.gen_description = False\n\t\tpost.description = description_from_content(post)\n\t\tpost.save()",
"def post(self, post_id=None):\n\n if post_id:\n abort(400)\n else:\n ... | [
"0.612241",
"0.59482175",
"0.5902697",
"0.5901005",
"0.5845779",
"0.58072627",
"0.5710131",
"0.56681836",
"0.5645754",
"0.5642677",
"0.5604725",
"0.5604725",
"0.55815965",
"0.55642414",
"0.54975563",
"0.54782796",
"0.5466466",
"0.54545844",
"0.54401195",
"0.54401195",
"0.5419... | 0.6026173 | 1 |
Create activity for published Post. | def on_post_published(sender, instance: dillo.models.posts.Post, **kwargs):
# If activity was already created (the post was already published),
# skip this signal to prevent the activity from appearing twice in the feeds.
if models_actstream.Action.objects.filter(
actor_object_id=instance.user.id, a... | {
"objective": {
"self": [],
"paired": [],
"triplet": [
[
"query",
"document",
"negatives"
]
]
}
} | [
"def post_to_activity(self, post):\n id = None\n if post.get('id'):\n # strip USERID_ prefix if it's there\n post['id'] = post['id'].split('_', 1)[-1]\n id = post['id']\n\n obj = self.post_to_object(post)\n activity = {\n 'verb': VERBS.get(post.get('type', obj.get('objectType')), 'po... | [
"0.74787325",
"0.6533647",
"0.64678",
"0.6233577",
"0.62174255",
"0.61825395",
"0.6088378",
"0.60738593",
"0.60658085",
"0.6014791",
"0.6014791",
"0.6007368",
"0.5983524",
"0.5974287",
"0.5948988",
"0.5934463",
"0.58406353",
"0.58193",
"0.58059466",
"0.5795128",
"0.5768188",
... | 0.6908637 | 1 |
Assign tags to a comment by parsing the content. | def on_created_comment(sender, instance: dillo.models.comments.Comment, created, **kwargs):
if not created:
return
# Extract tags and mentions from text and assign them to the Comment
tags, mentions = extract_tags_and_mentions(instance.content)
instance.tags.set(*tags)
for mentioned_user in ... | {
"objective": {
"self": [],
"paired": [],
"triplet": [
[
"query",
"document",
"negatives"
]
]
}
} | [
"def comment(self, content):\n pass",
"def __init__(self, node, declare):\n self.comment = getOptionalTag(node, \"comment\", [])\n if self.comment != []:\n self.comment = self.comment.replace(\"\\n\",\" \")\n while self.comment.find(\" \") != -1:\n self.c... | [
"0.6160088",
"0.6029694",
"0.58862054",
"0.5713247",
"0.5622352",
"0.561385",
"0.5606108",
"0.557518",
"0.5571938",
"0.5537934",
"0.54423904",
"0.54145694",
"0.5373428",
"0.5349663",
"0.53335774",
"0.5330848",
"0.5328987",
"0.5298146",
"0.52952015",
"0.52707106",
"0.5269429",... | 0.5155063 | 31 |
Actions to perform once a Like is created. | def on_created_like(sender, instance: dillo.models.mixins.Likes, created, **kwargs):
if (
isinstance(instance.content_object, dillo.models.comments.Comment)
and instance.content_object.parent_comment
):
target = instance.content_object.parent_comment
else:
target = None
... | {
"objective": {
"self": [],
"paired": [],
"triplet": [
[
"query",
"document",
"negatives"
]
]
}
} | [
"def like(self):\n self.like_count = self.like_count + 1 if self.like_count else 1",
"def like():\n if CURR_USER_KEY in session:\n cafe_id = int(request.json[\"cafe_id\"])\n print(type(cafe_id))\n user_id = g.user.id\n print(type(user_id))\n like = Like(\n c... | [
"0.6593319",
"0.6169452",
"0.6148844",
"0.6112476",
"0.6108443",
"0.6105551",
"0.60661376",
"0.60419214",
"0.6040893",
"0.59646094",
"0.59646094",
"0.59571815",
"0.5945115",
"0.5939536",
"0.5915928",
"0.5865272",
"0.58126223",
"0.5807308",
"0.5788529",
"0.57406265",
"0.572260... | 0.74319637 | 0 |
Decrease user profile likes of 1. The function is available as standalone for easier unit testing. | def profile_likes_count_decrease(target_user: User):
try:
with transaction.atomic():
dillo.models.profiles.Profile.objects.filter(user=target_user).update(
likes_count=F('likes_count') - 1
)
except IntegrityError:
log.warning('Integrity error when incremen... | {
"objective": {
"self": [],
"paired": [],
"triplet": [
[
"query",
"document",
"negatives"
]
]
}
} | [
"def on_deleted_like(sender, instance: dillo.models.mixins.Likes, **kwargs):\n if not instance.content_object:\n return\n target_user = instance.content_object.user\n profile_likes_count_decrease(target_user)\n log.debug('Decreased like count for user %s' % target_user)",
"def unlike(self, data... | [
"0.7092867",
"0.69812095",
"0.6517912",
"0.61881864",
"0.61731464",
"0.6097247",
"0.6096408",
"0.6036876",
"0.5903274",
"0.58781475",
"0.58778524",
"0.5823332",
"0.5778853",
"0.5737949",
"0.5679264",
"0.563741",
"0.55705523",
"0.5514937",
"0.54584485",
"0.5454046",
"0.5450492... | 0.8061975 | 0 |
Decrease likes_count for profile when Entity or Comment is unliked. | def on_deleted_like(sender, instance: dillo.models.mixins.Likes, **kwargs):
if not instance.content_object:
return
target_user = instance.content_object.user
profile_likes_count_decrease(target_user)
log.debug('Decreased like count for user %s' % target_user) | {
"objective": {
"self": [],
"paired": [],
"triplet": [
[
"query",
"document",
"negatives"
]
]
}
} | [
"def profile_likes_count_decrease(target_user: User):\n try:\n with transaction.atomic():\n dillo.models.profiles.Profile.objects.filter(user=target_user).update(\n likes_count=F('likes_count') - 1\n )\n except IntegrityError:\n log.warning('Integrity error w... | [
"0.74382937",
"0.68135536",
"0.62009215",
"0.61690074",
"0.6078528",
"0.5979719",
"0.5800343",
"0.5772288",
"0.5705149",
"0.56960106",
"0.5675965",
"0.56723356",
"0.5662375",
"0.56615674",
"0.5642525",
"0.56078047",
"0.56078047",
"0.56078047",
"0.5589605",
"0.5575719",
"0.557... | 0.7310249 | 1 |
Fetch social account name and avatar, and add them to the Profile. | def on_social_account_added(sender, instance: SocialAccount, created, **kwargs):
if not created:
return
name = instance.extra_data.get('name')
if name:
log.debug('Updating name via socialaccount for user %i' % instance.user.id)
instance.user.profile.name = name
instance.user.... | {
"objective": {
"self": [],
"paired": [],
"triplet": [
[
"query",
"document",
"negatives"
]
]
}
} | [
"def get_profile(self,soup,Name,Profile_url):\n h3=soup.find(\"h3\")\n \n if h3 !=None:\n if h3.find(\"a\")!=None: \n name=h3.a.get_text()\n if h3.a.has_attr('href') :\n h3_a_tag=h3.a['href']\n h3_a_tag=h3_... | [
"0.62179977",
"0.61199266",
"0.61171144",
"0.6097713",
"0.60976326",
"0.60648817",
"0.605172",
"0.60389084",
"0.6031145",
"0.60300606",
"0.60167927",
"0.6015737",
"0.6003023",
"0.5930042",
"0.5929104",
"0.5922604",
"0.59004813",
"0.59002423",
"0.58768433",
"0.58611965",
"0.58... | 0.76166373 | 0 |
On action created, fan out notifications | def on_created_action(sender, instance: models_actstream.Action, created, **kwargs):
if not created:
return
# If content_type is Post or Profile
if (
instance.action_object_content_type
== ContentType.objects.get_for_model(dillo.models.posts.Post)
and instance.verb == 'poste... | {
"objective": {
"self": [],
"paired": [],
"triplet": [
[
"query",
"document",
"negatives"
]
]
}
} | [
"def action_defined(sender, instance, created, raw, using, **kwargs):\n if created:\n raw_hook_event.send(\n sender=None,\n event_name=\"action_defined\",\n instance=instance,\n payload=ActionSerializer(instance).data,\n user=instance.team,\n )... | [
"0.63869154",
"0.6225311",
"0.6225311",
"0.6225311",
"0.6225311",
"0.6144163",
"0.5890172",
"0.58513165",
"0.58451253",
"0.5807999",
"0.57855886",
"0.57809436",
"0.5763071",
"0.5763071",
"0.57493734",
"0.5724776",
"0.572035",
"0.5714243",
"0.5679732",
"0.5625895",
"0.55907756... | 0.7528839 | 0 |
User stops following something. | def on_deleted_follow(sender, instance: models_actstream.Follow, **kwargs):
content_type = ContentType.objects.get_for_id(instance.content_type_id)
log.debug("Unfollowing %s %s" % (content_type.name, instance.object_id))
dillo.tasks.feeds.repopulate_timeline_content(
instance.content_type_id, instan... | {
"objective": {
"self": [],
"paired": [],
"triplet": [
[
"query",
"document",
"negatives"
]
]
}
} | [
"def stop(self):\r\n self.terminating = True",
"def stop(self) -> None:",
"def stop(self) -> None:",
"def stop(self):\r\n pass",
"def stop(self):",
"def stop(self):",
"def stop(self):\n return",
"def stop() -> None:",
"def stop(self):\n pass",
"def stop(self):\n ... | [
"0.758307",
"0.7484069",
"0.7484069",
"0.7418194",
"0.73704475",
"0.73704475",
"0.7342334",
"0.733363",
"0.73308474",
"0.73308474",
"0.73308474",
"0.73308474",
"0.73308474",
"0.73308474",
"0.73308474",
"0.73308474",
"0.73308474",
"0.73308474",
"0.73308474",
"0.7324925",
"0.73... | 0.0 | -1 |
If confirmed, subscribe to newsletter. | def on_email_confirmed(request, email_address: EmailAddress, **kwargs):
dillo.tasks.profile.update_mailing_list_subscription(email_address.email, True) | {
"objective": {
"self": [],
"paired": [],
"triplet": [
[
"query",
"document",
"negatives"
]
]
}
} | [
"def subscribe(self, request):\n email = self.cleaned_data.get('email')\n\n email_name, domain_part = email.rsplit('@', 1)\n domain_name = '@' + domain_part\n email_domain, created = Domain.objects.get_or_create(name=domain_name)\n\n subscriber, created = Subscriber.objects.get_or... | [
"0.7161424",
"0.6835099",
"0.6736429",
"0.65938777",
"0.6326111",
"0.6227474",
"0.6216683",
"0.6181522",
"0.6166237",
"0.61440307",
"0.6120497",
"0.61191845",
"0.60010856",
"0.5983212",
"0.595965",
"0.5947182",
"0.5929845",
"0.5913996",
"0.5880866",
"0.5878438",
"0.58783746",... | 0.66522044 | 3 |
Unsubscribe previous email, wait for new one to be confirmed. | def on_email_email_changed(
request, user, from_email_address: EmailAddress, to_email_address, **kwargs
):
if from_email_address:
dillo.tasks.profile.update_mailing_list_subscription(from_email_address.email, False) | {
"objective": {
"self": [],
"paired": [],
"triplet": [
[
"query",
"document",
"negatives"
]
]
}
} | [
"def unsubscribe_all_emails_from_cloudwatch(self):\n for subscription in self._get_cloudwatch_subscriptions():\n subscription_arn = subscription['SubscriptionArn']\n if not subscription_arn == 'PendingConfirmation':\n self.conn.unsubscribe(subscription_arn)",
"def unsub... | [
"0.6626395",
"0.64564025",
"0.6453663",
"0.6379453",
"0.6170811",
"0.61062586",
"0.61062586",
"0.61062586",
"0.61062586",
"0.61062586",
"0.60732317",
"0.6026578",
"0.58910805",
"0.5749465",
"0.57386756",
"0.5738046",
"0.5716363",
"0.5714279",
"0.57047355",
"0.5697949",
"0.565... | 0.0 | -1 |
Delete the user from the newsletter. | def user_pre_delete(sender, instance: User, **kwargs):
if not hasattr(settings, 'ANYMAIL'):
log.info("Mailgun not configured, skipping mailing list subscription update")
return
if not hasattr(settings, 'MAILING_LIST_NEWSLETTER_EMAIL'):
log.debug("Newsletter not configured, skipping mai... | {
"objective": {
"self": [],
"paired": [],
"triplet": [
[
"query",
"document",
"negatives"
]
]
}
} | [
"def delete_user():",
"def delete_user():\n #TODO user delete\n pass",
"def delete_user(self, user):\n self.delete(user)",
"def delete_user(id):\n pass",
"def delete_user(self):\n User.user_list.remove(self)",
"def delete_user(self):\n User.user_list.remove(self)",
"def del... | [
"0.7976897",
"0.7772087",
"0.770037",
"0.7479432",
"0.746371",
"0.746371",
"0.746371",
"0.7442444",
"0.7436067",
"0.7431538",
"0.73823726",
"0.7339518",
"0.7288484",
"0.7285742",
"0.7276345",
"0.725602",
"0.72481054",
"0.72303545",
"0.72023326",
"0.7187994",
"0.71587986",
"... | 0.71425474 | 22 |
runs multiple experiments that calculates the precision and recall of the comparison algorithms' results | def run_multiple_simple_experiments(self, path_to_csv_files, algorithms, thresholds):
starting_time = time.time()
print("started simple precision recall image_matching_module")
for algorithm in algorithms:
start_file_time = time.time()
print("started algorithm:", algorith... | {
"objective": {
"self": [],
"paired": [],
"triplet": [
[
"query",
"document",
"negatives"
]
]
}
} | [
"def run_experiments():\n if True: # Change to False when done finding max_scoring_num_rolls\n six_sided_max = max_scoring_num_rolls(six_sided)\n print('Max scoring num rolls for six-sided dice:', six_sided_max)\n rerolled_max = max_scoring_num_rolls(reroll(six_sided))\n print('Max s... | [
"0.7036995",
"0.6971232",
"0.69174814",
"0.68236893",
"0.6810289",
"0.67737514",
"0.6682389",
"0.66659284",
"0.6623596",
"0.6608259",
"0.65498793",
"0.6528051",
"0.65186554",
"0.64956766",
"0.6495176",
"0.6481607",
"0.6480974",
"0.6465037",
"0.6449332",
"0.64473736",
"0.64420... | 0.66045517 | 10 |
runs an image_matching_module that calculates the precision and recall of the comparison algorithm's results | def run_single_simple_experiment(self, path_to_csv_files, algorithm, thresholds):
splitted_path = os.path.abspath(__file__).split("/")
if not os.path.isdir("/".join(splitted_path[:-1]) + "/results"):
os.mkdir("/".join(splitted_path[:-1]) + "/results")
experiment_dict, list_dict = s... | {
"objective": {
"self": [],
"paired": [],
"triplet": [
[
"query",
"document",
"negatives"
]
]
}
} | [
"def patternMatchingTest(matchingAlgorithm,\n observations= defaultCOMPREFdump , \n modelsFolder= defaultWRFdumpsFolder,\n #key1='0828',\n key1='',\n key2='',\n #key3='',\n verbose=True,\n makeSmallImages=True,\n smal... | [
"0.686315",
"0.66412973",
"0.6606622",
"0.6439272",
"0.64282435",
"0.636913",
"0.63653",
"0.6287904",
"0.6282749",
"0.61645573",
"0.61143124",
"0.6112964",
"0.6099393",
"0.60614634",
"0.6056027",
"0.60436153",
"0.60412055",
"0.60172075",
"0.6009123",
"0.6001078",
"0.59953034"... | 0.0 | -1 |
runs multiple experiments that calculates the precision and recall of multiple comparison algorithms' results | def run_multiple_advanced_experiments(self, path_to_csv_files, algorithms_weights_list, thresholds):
starting_time = time.time()
print("started advanced precision recall image_matching_module")
for algorithms_weights in algorithms_weights_list:
start_file_time = time.time()
... | {
"objective": {
"self": [],
"paired": [],
"triplet": [
[
"query",
"document",
"negatives"
]
]
}
} | [
"def run_experiments():\n if True: # Change to False when done finding max_scoring_num_rolls\n six_sided_max = max_scoring_num_rolls(six_sided)\n print('Max scoring num rolls for six-sided dice:', six_sided_max)\n rerolled_max = max_scoring_num_rolls(reroll(six_sided))\n print('Max s... | [
"0.70623094",
"0.7007437",
"0.69541293",
"0.6884711",
"0.68788844",
"0.68726337",
"0.67315125",
"0.6698763",
"0.66550297",
"0.6629735",
"0.655879",
"0.6552913",
"0.6541094",
"0.6533274",
"0.65046316",
"0.649451",
"0.6487716",
"0.64735806",
"0.6451186",
"0.6437416",
"0.6433956... | 0.6528349 | 14 |
runs an image_matching_module that calculates the precision and recall of multiple comparison algorithms' results | def run_single_advanced_experiment(self, path_to_csv_files, algorithms_weights, thresholds):
splitted_path = os.path.abspath(__file__).split("/")
if not os.path.isdir("/".join(splitted_path[:-1]) + "/results"):
os.mkdir("/".join(splitted_path[:-1]) + "/results")
first_run = True
... | {
"objective": {
"self": [],
"paired": [],
"triplet": [
[
"query",
"document",
"negatives"
]
]
}
} | [
"def painting_matching_ml(imgs, db_imgs, method_list, text_masks, author_text, gt_text, metrics, weights, splits=30, max_rank=10):\n descriptor_extractors = [get_descriptor_extractor(method_name) for method_name in method_list]\n tmp_img_format = []\n tmp_mask_format = []\n tmp_text_format = []\n for... | [
"0.70493734",
"0.6962549",
"0.6559345",
"0.6556465",
"0.6520097",
"0.645765",
"0.64450425",
"0.6405806",
"0.63550365",
"0.62989557",
"0.62902206",
"0.6247937",
"0.623757",
"0.6215632",
"0.6210621",
"0.6126112",
"0.61117905",
"0.61023283",
"0.60840875",
"0.60789233",
"0.606114... | 0.0 | -1 |
create the dictionary holding the image_matching_module's results | def create_experiment_dict(algorithm_name, thresholds):
experiment_dict = {}
for threshold in thresholds:
experiment_dict[str(threshold)] = {}
experiment_dict[str(threshold)][algorithm_name] = [0, 0, 0, 0, 0, 0, 0, 0, 0, 0]
list_dict = {'TP': 0, 'FP': 1, 'TN': 2, 'FN': 3,... | {
"objective": {
"self": [],
"paired": [],
"triplet": [
[
"query",
"document",
"negatives"
]
]
}
} | [
"def get_detect_result(self):\n\n resultdict = {'class_index' : self.class_index,\n 'obj_name' : self.obj_name,\n 'score' : self.score,\n 'bounding_box' : {\n 'x_min' : self.x_min,\n 'y_min' : sel... | [
"0.6621895",
"0.64354163",
"0.6385068",
"0.6345569",
"0.63442147",
"0.62534124",
"0.6184932",
"0.61760575",
"0.6173091",
"0.60566914",
"0.6048575",
"0.6046165",
"0.6008114",
"0.59930813",
"0.59672993",
"0.5958381",
"0.59390056",
"0.58587366",
"0.5855858",
"0.5824432",
"0.5820... | 0.0 | -1 |
updates the precision and recall data in the image_matching_module's results | def update_precision_and_recall(experiment_dict, list_dict, algorithm_name, thresholds):
for threshold in thresholds:
algorithm_column_name = algorithm_name + "_actual"
tp = experiment_dict[str(threshold)][algorithm_name][list_dict['TP']]
fp = experiment_dict[str(threshold)][... | {
"objective": {
"self": [],
"paired": [],
"triplet": [
[
"query",
"document",
"negatives"
]
]
}
} | [
"def evaluate_matrices(self, precision_matrix, recall_matrix):\n\t\tground_truth_count = precision_matrix.shape[0]\n\t\tdetection_count = precision_matrix.shape[1]\n\t\tground_truth_sets_precision = defaultdict(set) # number of ground truth items that match a particular detection in the precision matrix\n\t\tdetect... | [
"0.625239",
"0.6167243",
"0.6078468",
"0.6062458",
"0.60042554",
"0.5945311",
"0.59133434",
"0.5907834",
"0.5897664",
"0.587507",
"0.58680505",
"0.58505803",
"0.5835626",
"0.5821899",
"0.58180785",
"0.57850605",
"0.5741462",
"0.5735853",
"0.57268965",
"0.5706414",
"0.5699125"... | 0.6408821 | 0 |
creates a dictionary that can be ingested by pandas DataFrame to write to a csv file | def create_simple_experiment_dict_for_csv(experiment_dict, list_dict, algorithm, thresholds):
experiment_dict_for_csv = {
"attributes": ['TP', 'FP', 'TN', 'FN', 'TPR (Sensitivity)', 'TNR (Specificity)', 'Precision', 'Recall',
'F1 Measure', 'Accuracy']}
for row_key ... | {
"objective": {
"self": [],
"paired": [],
"triplet": [
[
"query",
"document",
"negatives"
]
]
}
} | [
"def csv_to_dict(self):\n log = logger.configure(\"default\")\n try: \n df = pd.read_csv(self.__csv_path)\n except IOError as e:\n # file not found\n log.error('Could not import {}. Got error {}'.format(self.__csv_path, e))\n raise \n else:\n ... | [
"0.68208885",
"0.6775985",
"0.65989804",
"0.6495224",
"0.6452339",
"0.642253",
"0.63207847",
"0.629015",
"0.6283443",
"0.62565273",
"0.6228677",
"0.62230223",
"0.620419",
"0.6197963",
"0.6147068",
"0.61124",
"0.60970604",
"0.6051975",
"0.60255706",
"0.60189426",
"0.6013112",
... | 0.0 | -1 |
creates a dictionary that can be ingested by pandas DataFrame to write to a csv file | def create_advanced_experiment_dict_for_csv(experiment_dict, list_dict, algorithm_name, thresholds):
experiment_dict_for_csv = {
"attributes": ['TP', 'FP', 'TN', 'FN', 'TPR (Sensitivity)', 'TNR (Specificity)', 'Precision', 'Recall',
'F1 Measure', 'Accuracy']}
for r... | {
"objective": {
"self": [],
"paired": [],
"triplet": [
[
"query",
"document",
"negatives"
]
]
}
} | [
"def csv_to_dict(self):\n log = logger.configure(\"default\")\n try: \n df = pd.read_csv(self.__csv_path)\n except IOError as e:\n # file not found\n log.error('Could not import {}. Got error {}'.format(self.__csv_path, e))\n raise \n else:\n ... | [
"0.68224066",
"0.67767906",
"0.65991676",
"0.6495553",
"0.64526737",
"0.6419535",
"0.6320348",
"0.62898755",
"0.6284467",
"0.6255064",
"0.62294525",
"0.62211573",
"0.6205003",
"0.6200207",
"0.61460024",
"0.61113054",
"0.6097703",
"0.60526484",
"0.60267997",
"0.6017852",
"0.60... | 0.0 | -1 |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.