repo stringlengths 1 99 | file stringlengths 13 215 | code stringlengths 12 59.2M | file_length int64 12 59.2M | avg_line_length float64 3.82 1.48M | max_line_length int64 12 2.51M | extension_type stringclasses 1
value |
|---|---|---|---|---|---|---|
BrnoLM | BrnoLM-master/scripts/eval/eval-independent.py | #!/usr/bin/env python
import argparse
import logging
import math
import torch
from brnolm.runtime.runtime_utils import init_seeds
from brnolm.runtime.evaluation import IndependentLinesEvaluator
def main():
logging.basicConfig(level=logging.INFO, format='[%(levelname)s::%(name)s] %(message)s')
parser = argpa... | 2,106 | 33.540984 | 166 | py |
BrnoLM | BrnoLM-master/scripts/eval/eval-noivecs-domain-adaptation.py | #!/usr/bin/env python
import argparse
import math
import torch
from brnolm.data_pipeline.split_corpus_dataset import DomainAdaptationSplitFFMultiTarget
from brnolm.data_pipeline.multistream import BatchBuilder
from brnolm.runtime.runtime_utils import CudaStream, filelist_to_objects, init_seeds
from brnolm.runtime.ru... | 2,447 | 36.090909 | 96 | py |
BrnoLM | BrnoLM-master/scripts/eval/eval-chime-v2.py | #!/usr/bin/env python
import argparse
import logging
import math
import torch
from safe_gpu.safe_gpu import GPUOwner
from brnolm.runtime.runtime_utils import init_seeds
from brnolm.runtime.evaluation import SubstitutionalEnblockEvaluator_v2
from brnolm.data_pipeline.aug_paper_pipeline import Corruptor
if __name__ =... | 2,866 | 41.161765 | 166 | py |
BrnoLM | BrnoLM-master/scripts/eval/eval-ivecs-domain-adaptation.py | #!/usr/bin/env python
import argparse
import math
import torch
from brnolm.smm_itf import ivec_appenders
from brnolm.data_pipeline.split_corpus_dataset import DomainAdaptationSplitFFMultiTarget
from brnolm.data_pipeline.multistream import BatchBuilder
from brnolm.smm_itf import smm_ivec_extractor
from brnolm.runtime... | 2,993 | 37.883117 | 96 | py |
BrnoLM | BrnoLM-master/scripts/eval/eval-multifile.py | #!/usr/bin/env python
import argparse
import math
import torch
from brnolm.data_pipeline.multistream import BatchBuilder
from brnolm.data_pipeline.reading import tokens_from_file
from brnolm.data_pipeline.temporal_splitting import TemporalSplits
from brnolm.runtime.runtime_utils import CudaStream, init_seeds, fileli... | 2,075 | 36.071429 | 84 | py |
BrnoLM | BrnoLM-master/scripts/eval/eval-chime.py | #!/usr/bin/env python
import argparse
import logging
import math
import torch
from brnolm.runtime.runtime_utils import init_seeds
from brnolm.runtime.evaluation import SubstitutionalEnblockEvaluator
from brnolm.data_pipeline.augmentation import Corruptor
if __name__ == '__main__':
logging.basicConfig(level=logg... | 2,589 | 40.774194 | 166 | py |
BrnoLM | BrnoLM-master/scripts/eval/eval.py | #!/usr/bin/env python
import argparse
import logging
import math
import torch
from safe_gpu.safe_gpu import GPUOwner
from brnolm.runtime.runtime_utils import init_seeds
from brnolm.runtime.evaluation import EnblockEvaluator
def main(args):
print(args)
init_seeds(args.seed, args.cuda)
print("loading mo... | 1,975 | 31.933333 | 166 | py |
BrnoLM | BrnoLM-master/scripts/eval/eval-ivecs-oracle.py | #!/usr/bin/env python
import argparse
import math
import torch
from brnolm.data_pipeline.split_corpus_dataset import TokenizedSplitFFBase
from brnolm.smm_itf import ivec_appenders
from brnolm.smm_itf import smm_ivec_extractor
from brnolm.data_pipeline.multistream import BatchBuilder
from brnolm.data_pipeline.temporal... | 2,812 | 36.013158 | 90 | py |
BrnoLM | BrnoLM-master/scripts/oov-clustering/collect-embeddings.py | #!/usr/bin/env python
import argparse
import sys
import torch
from brnolm.oov_clustering.embeddings_io import str_from_embedding
from brnolm.oov_clustering.embeddings_computation import tensor_from_words
def embs_from_words(words, lm):
words = ["</s>"] + words
th_data = tensor_from_words(words, lm.vocab)[:... | 1,868 | 30.15 | 93 | py |
BrnoLM | BrnoLM-master/scripts/oov-clustering/predict-embeddings.py | #!/usr/bin/env python
import argparse
import sys
import torch
from brnolm.oov_clustering.embeddings_io import str_from_embedding
from brnolm.oov_clustering.embeddings_computation import tensor_from_words
def relevant_prefix(transcript, word_of_interest):
first_oov_oi_loc = transcript.index(word_of_interest)
... | 2,076 | 28.253521 | 93 | py |
BrnoLM | BrnoLM-master/scripts/train/train-independent.py | #!/usr/bin/env python
import argparse
import math
import random
import torch
import time
from brnolm.data_pipeline.reading import get_independent_lines
from brnolm.data_pipeline.threaded import OndemandDataProvider
from brnolm.data_pipeline.multistream import Batcher
from brnolm.runtime.runtime_utils import init_see... | 5,205 | 38.439394 | 198 | py |
BrnoLM | BrnoLM-master/scripts/train/train-chime-robust-v6.py | #!/usr/bin/env python
import argparse
import logging
import math
import torch
from brnolm.data_pipeline.reading import tokens_from_fn
from brnolm.data_pipeline.threaded import OndemandDataProvider
from brnolm.data_pipeline.aug_paper_pipeline import InputTargetCorruptor, form_input_targets, LazyBatcher, TemplSplitterC... | 7,071 | 41.095238 | 174 | py |
BrnoLM | BrnoLM-master/scripts/train/train-multifile.py | #!/usr/bin/env python
import argparse
import random
import torch
from brnolm.data_pipeline.multistream import BatchBuilder
from brnolm.data_pipeline.reading import tokens_from_file
from brnolm.data_pipeline.temporal_splitting import TemporalSplits
from brnolm.runtime.runtime_utils import CudaStream, init_seeds, fi... | 4,945 | 39.876033 | 112 | py |
BrnoLM | BrnoLM-master/scripts/train/logger.py | # Code referenced from https://gist.github.com/gyglim/1f8dfb1b5c82627ae3efcfbbadb9f514
from torch.utils.tensorboard import SummaryWriter
class Logger(object):
def __init__(self, log_dir, update_freq):
"""Create a summary writer logging to log_dir."""
self.writer = SummaryWriter(log_dir)
se... | 1,124 | 31.142857 | 86 | py |
BrnoLM | BrnoLM-master/scripts/train/train-ivecs-oracle.py | #!/usr/bin/env python
import argparse
import random
import torch
from brnolm.data_pipeline.multistream import BatchBuilder
from brnolm.data_pipeline.temporal_splitting import TemporalSplits
from brnolm.data_pipeline.split_corpus_dataset import TokenizedSplitFFBase
from brnolm.smm_itf import ivec_appenders
from brnol... | 5,252 | 37.911111 | 112 | py |
BrnoLM | BrnoLM-master/scripts/train/train-ivecs-partial.py | #!/usr/bin/env python
import argparse
import random
import torch
from brnolm.data_pipeline.multistream import BatchBuilder
from brnolm.smm_itf import ivec_appenders
from brnolm.smm_itf import smm_ivec_extractor
from brnolm.runtime.runtime_utils import CudaStream, init_seeds, filelist_to_tokenized_splits, BatchFilte... | 5,080 | 38.695313 | 121 | py |
BrnoLM | BrnoLM-master/scripts/train/train-flat.py | #!/usr/bin/env python
import argparse
import logging
import math
import sys
import torch
from safe_gpu.safe_gpu import GPUOwner
from brnolm.data_pipeline.pipeline_factories import plain_factory, yaml_factory
from brnolm.runtime.runtime_utils import init_seeds, epoch_summary
from brnolm.runtime.runtime_multifile imp... | 6,422 | 38.164634 | 114 | py |
BrnoLM | BrnoLM-master/scripts/train/train-no-epoch.py | #!/usr/bin/env python
import argparse
import logging
import math
import sys
import torch
from safe_gpu.safe_gpu import GPUOwner
from brnolm import zoo
from brnolm.data_pipeline.reading import tokenizer_factory
from brnolm.data_pipeline.pipeline_factories import plain_factory_noepoch, yaml_factory_noepoch
from brno... | 7,559 | 36.058824 | 126 | py |
BrnoLM | BrnoLM-master/scripts/train/train.py | #!/usr/bin/env python
import argparse
import math
import torch
from logger import Logger
from brnolm.data_pipeline.reading import tokens_from_fn
from brnolm.data_pipeline.multistream import batchify
from brnolm.data_pipeline.temporal_splitting import TemporalSplits
from brnolm.runtime.runtime_utils import TransposeW... | 5,079 | 37.484848 | 121 | py |
BrnoLM | BrnoLM-master/scripts/train/train-pero.py | #!/usr/bin/env python
import argparse
import torch
from brnolm.data_pipeline.reading import tokens_from_fn
from brnolm.data_pipeline.multistream import batchify
from brnolm.data_pipeline.temporal_splitting import TemporalSplits
from brnolm.runtime.runtime_utils import TransposeWrapper, init_seeds, epoch_summary
from... | 5,299 | 33.640523 | 121 | py |
BrnoLM | BrnoLM-master/scripts/rescoring/rescore-kaldi-latts-continuous.py | #!/usr/bin/env python3
import argparse
import logging
import torch
import brnolm.language_models.vocab as vocab
from brnolm.rescoring.segment_scoring import SegmentScorer
from safe_gpu.safe_gpu import GPUOwner
import typing
import brnolm.kaldi_itf
def translate_latt_to_model(word_ids, latt_vocab, model_vocab, mod... | 4,702 | 35.742188 | 211 | py |
BrnoLM | BrnoLM-master/scripts/rescoring/rescore-nbest-continuous.py | #!/usr/bin/env python3
import argparse
import logging
import torch
from brnolm.rescoring.segment_scoring import SegmentScorer
from safe_gpu.safe_gpu import GPUOwner
import typing
import brnolm.kaldi_itf
def select_hidden_state_to_pass(hidden_states):
return hidden_states['1']
def spk_sess(segment_name):
... | 3,427 | 31.961538 | 101 | py |
BrnoLM | BrnoLM-master/scripts/model-building/build-shallow-nn.py | #!/usr/bin/env python
import argparse
import torch
from brnolm.language_models import ffnn_models, vocab, language_model
from brnolm.language_models.decoders import FullSoftmaxDecoder
if __name__ == '__main__':
parser = argparse.ArgumentParser(description='PyTorch FFNN Language Model')
parser.add_argument('-... | 2,045 | 40.755102 | 96 | py |
BrnoLM | BrnoLM-master/scripts/model-building/build-transformer.py | #!/usr/bin/env python
import argparse
import torch
from brnolm.language_models import transformer, vocab, language_model
from brnolm.language_models.decoders import FullSoftmaxDecoder
if __name__ == '__main__':
parser = argparse.ArgumentParser(description='PyTorch LSTM Language Model')
parser.add_argument('... | 2,643 | 39.676923 | 113 | py |
BrnoLM | BrnoLM-master/scripts/model-building/build-lstm.py | #!/usr/bin/env python
import argparse
import torch
from brnolm.language_models import lstm_model, vocab, language_model
from brnolm.language_models.decoders import CustomLossFullSoftmaxDecoder
from brnolm.language_models.encoders import FlatEmbedding
if __name__ == '__main__':
parser = argparse.ArgumentParser(d... | 3,111 | 42.830986 | 113 | py |
BrnoLM | BrnoLM-master/scripts/model-building/build-lstmp.py | #!/usr/bin/env python
import argparse
import torch
from brnolm.language_models import lstm_model, vocab, language_model
from brnolm.language_models.decoders import FullSoftmaxDecoder
if __name__ == '__main__':
parser = argparse.ArgumentParser(description='PyTorch LSTM Language Model')
parser.add_argument('-... | 2,290 | 40.654545 | 89 | py |
BrnoLM | BrnoLM-master/scripts/model-building/build-shallow-nn-with-ivec.py | #!/usr/bin/env python
import argparse
import torch
from brnolm.language_models import language_model, vocab, ffnn_models
from brnolm.language_models.decoders import FullSoftmaxDecoder
if __name__ == '__main__':
parser = argparse.ArgumentParser(description='PyTorch FFNN Language Model')
parser.add_argument('-... | 2,187 | 41.901961 | 96 | py |
BrnoLM | BrnoLM-master/brnolm/multifile-ivec-unigram-ppl.py | #!/usr/bin/env python
import argparse
import math
import runtime_utils
import torch
import language_model
import smm_ivec_extractor
def bows_to_ps(bows):
uni_ps = bows.t() / bows.sum(dim=1)
return uni_ps.t()
if __name__ == '__main__':
parser = argparse.ArgumentParser()
parser.add_argument('--file-... | 1,596 | 26.534483 | 68 | py |
BrnoLM | BrnoLM-master/brnolm/analyze-ivec-changes.py | #!/usr/bin/env python
import argparse
import io
import math
import sys
import torch
import split_corpus_dataset
import ivec_appenders
import smm_ivec_extractor
from runtime_utils import filenames_file_to_filenames
class DummyDict:
def __getitem__(self, index):
return 0
def euclidean_distance(a, b):
... | 4,087 | 30.206107 | 127 | py |
BrnoLM | BrnoLM-master/brnolm/rmn-plotter.py | #!/usr/bin/env python
import torch
import argparse
import plotting
import numpy as np
if __name__ == "__main__":
parser = argparse.ArgumentParser()
parser.add_argument("model")
args = parser.parse_args()
with open(args.model, 'rb') as f:
model = torch.load(f)
cs_numpied = [c.weight.dat... | 1,037 | 30.454545 | 102 | py |
BrnoLM | BrnoLM-master/brnolm/investigate-ivecs.py | #!/usr/bin/env python
import argparse
import torch
import smm_ivec_extractor
from runtime_utils import init_seeds, filenames_file_to_filenames
if __name__ == '__main__':
parser = argparse.ArgumentParser(description='PyTorch RNN/LSTM Language Model')
parser.add_argument('--filelist', type=str, required=True,... | 1,516 | 31.276596 | 83 | py |
BrnoLM | BrnoLM-master/brnolm/analysis.py | import torch
def categorical_entropy(p, eps=1e-100):
zeros = p <= eps
log_p = p.log()
log_p.masked_fill_(zeros, 0.0) # eliminates -inf for p[x] = 0.0
H_p = - torch.sum(p*log_p, dim=-1)
return H_p / torch.log(torch.FloatTensor([2]))
def categorical_cross_entropy(p, q, eps=1e-100):
zeros... | 651 | 23.148148 | 67 | py |
BrnoLM | BrnoLM-master/brnolm/srilm-debug2.py | #!/usr/bin/env python
import argparse
import math
import sys
import torch
BATCH_SIZE = 1
def per_word_logprobs(words, lm):
words_tensor = torch.tensor([lm.vocab[w] for w in words], requires_grad=False).view(1, -1)
x = words_tensor[:, :-1]
t = words_tensor[:, 1:]
h0 = lm.model.init_hidden(x.size(0... | 2,668 | 28.655556 | 138 | py |
BrnoLM | BrnoLM-master/brnolm/multifile-ml-unigram-tranfer-ppl.py | #!/usr/bin/env python
import argparse
import math
import runtime_utils
import vocab
import torch
import analysis
def bows_to_ps(bows):
uni_ps = bows.t() / bows.sum(dim=1)
return uni_ps.t()
def bows_to_ent(bows):
uni_ps = bows_to_ps(bows)
entropies = analysis.categorical_entropy(uni_ps)
av... | 1,914 | 26.357143 | 83 | py |
BrnoLM | BrnoLM-master/brnolm/multifile-ml-unigram-ppl.py | #!/usr/bin/env python
import argparse
import math
import runtime_utils
import vocab
import torch
import analysis
def bows_to_ps(bows):
uni_ps = bows.t() / bows.sum(dim=1)
return uni_ps.t()
def bows_to_ent(bows):
uni_ps = bows_to_ps(bows)
entropies = analysis.categorical_entropy(uni_ps)
av... | 1,368 | 24.351852 | 70 | py |
BrnoLM | BrnoLM-master/brnolm/analyze-ivec-distribution.py | #!/usr/bin/env python
import argparse
import io
import math
import sys
import torch
import numpy as np
import split_corpus_dataset
import ivec_appenders
import smm_ivec_extractor
from runtime_utils import filenames_file_to_filenames
if __name__ == '__main__':
parser = argparse.ArgumentParser()
parser.add_arg... | 1,245 | 26.086957 | 63 | py |
BrnoLM | BrnoLM-master/brnolm/lm-info.py | #!/usr/bin/env python
import argparse
import torch
if __name__ == '__main__':
parser = argparse.ArgumentParser(description='PyTorch RNN/LSTM Language Model')
parser.add_argument('--vocab', action='store_true', help='print out the full vocab')
parser.add_argument('load', help='where to load a model from')
... | 523 | 29.823529 | 88 | py |
BrnoLM | BrnoLM-master/brnolm/oov_clustering/embeddings_computation.py | import torch
def tensor_from_words(words, vocab):
return torch.tensor([vocab[w] for w in words]).view(1, -1)
| 115 | 18.333333 | 62 | py |
BrnoLM | BrnoLM-master/brnolm/runtime/runtime_multifile.py | import torch
from .runtime_utils import repackage_hidden
from .tensor_reorganization import TensorReorganizer
def prepare_inputs(inputs, do_transpose, use_ivecs, custom_batches):
X = inputs[0]
batch_size = X.size(0)
if do_transpose:
X = X.t()
targets = inputs[1]
if do_transpose:
... | 3,869 | 25.506849 | 112 | py |
BrnoLM | BrnoLM-master/brnolm/runtime/model_statistics.py | def scaled_int_str(value):
if value < 1000:
return f'{value}'
elif value < 1000000:
return f'{value/1000:.1f}k'
else:
return f'{value/1000000:.1f}M'
class ModelStatistics:
def __init__(self, model):
self.model = model
def total_nb_params(self):
return sum(p... | 1,328 | 39.272727 | 130 | py |
BrnoLM | BrnoLM-master/brnolm/runtime/reporting.py | import pathlib
import torch
import math
class ValidationWatcher:
def __init__(self, val_fn, initial_val_loss, freq_in_tokens, workdir, lm, lr_control=None):
self.val_losses = [initial_val_loss]
self.validation_fn = val_fn
self.lm = lm
pathlib.Path(workdir).mkdir(parents=True, exis... | 1,917 | 29.444444 | 95 | py |
BrnoLM | BrnoLM-master/brnolm/runtime/tensor_reorganization.py | import torch
from typing import Dict, Any
class Singleton(type):
_instances: Dict[Any, Any] = {} # TODO what is the actual type?
def __call__(cls, *args, **kwargs):
if cls not in cls._instances:
cls._instances[cls] = super(Singleton, cls).__call__(*args, **kwargs)
return cls._in... | 1,926 | 26.140845 | 99 | py |
BrnoLM | BrnoLM-master/brnolm/runtime/loggers.py | import sys
import time
import math
import torch
class BaseLogger():
def __init__(self, report_period, output_file=sys.stdout):
self._start_time = time.time()
self._nb_logs = 0
self._report_period = report_period
self._of = output_file
self._construction_time = time.time()
... | 3,926 | 27.664234 | 182 | py |
BrnoLM | BrnoLM-master/brnolm/runtime/evaluation.py | import logging
import math
import numpy as np
from dataclasses import dataclass
from brnolm.data_pipeline.reading import get_independent_lines, tokens_from_fn
from brnolm.data_pipeline.threaded import OndemandDataProvider
from brnolm.data_pipeline.multistream import Batcher, batchify
from brnolm.data_pipeline.temporal... | 10,317 | 36.249097 | 156 | py |
BrnoLM | BrnoLM-master/brnolm/runtime/runtime_utils.py | import random
import torch
from brnolm.data_pipeline import split_corpus_dataset
import sys
import math
class CudaStream():
def __init__(self, source):
self._source = source
def __iter__(self):
for batch in self._source:
yield tuple(x.cuda() for x in batch)
class TransposeWrap... | 3,493 | 27.876033 | 110 | py |
BrnoLM | BrnoLM-master/brnolm/rescoring/segment_scoring.py | from dataclasses import dataclass
import logging
import typing
import torch
from brnolm.language_models.language_model import split_batch_hidden_state, detach_hidden_state
# for LSTMs
def lstm_h0_provider(single_h, batch_size):
h, c = single_h
return (torch.stack([h]*batch_size, axis=1), torch.stack([c]*batc... | 5,188 | 39.539063 | 178 | py |
BrnoLM | BrnoLM-master/brnolm/data_pipeline/masked.py | from typing import List
import torch
def masked_tensor_from_sentences(sentences: List[List[int]], filler=0, device=torch.device('cpu'), target_all=False):
try:
sentences[0][0]
except TypeError:
raise ValueError("masked_tensor_from_sentences() consumes List of Lists (batch X time)")
batch_... | 1,467 | 35.7 | 117 | py |
BrnoLM | BrnoLM-master/brnolm/data_pipeline/split_corpus_dataset.py | import torch
from brnolm.data_pipeline.temporal_splitting import TemporalSplits
class TokenizedSplitFFBase():
def __init__(self, f, vocab, temporal_split_builder):
"""
Args:
f (file): File with a document.
vocab (Vocabulary): Vocabulary for translation word -> ... | 3,756 | 33.787037 | 123 | py |
BrnoLM | BrnoLM-master/brnolm/data_pipeline/multistream.py | import torch
class LineTooLongError(Exception):
pass
def batchify(data, bsz, cuda):
""" For simple rearranging of 'single sentence' data.
"""
# Work out how cleanly we can divide the dataset into bsz parts.
nbatch = data.size(0) // bsz
# Trim off any extra elements that wouldn't cleanly fit ... | 3,518 | 29.336207 | 100 | py |
BrnoLM | BrnoLM-master/brnolm/data_pipeline/augmentation.py | import torch
class Substitutor:
def __init__(self, rate, replacements_range):
self.rate = rate
self.replacements_range = replacements_range
if not isinstance(replacements_range, int) or replacements_range < 0:
raise ValueError(f"Replacements range needs to be a positive integer... | 1,627 | 32.22449 | 108 | py |
BrnoLM | BrnoLM-master/brnolm/data_pipeline/reading.py | import torch
def word_splitter(line):
return line.split()
def char_splitter(line, sentence_end_token=None):
chars = list(line)
if sentence_end_token is None:
return chars
else:
return chars + [sentence_end_token]
class WordIdProvider:
def __init__(self, vocab):
self.vo... | 2,350 | 22.989796 | 119 | py |
BrnoLM | BrnoLM-master/brnolm/data_pipeline/flexible_pipeline.py | import random
import torch
class SequenceReadingHead:
def __init__(self, seq, start=0):
self.seq = seq
self.pos = start
def __next__(self):
val = self.seq[self.pos]
self.pos = (self.pos + 1) % len(self.seq)
return val
class FileReadingHead:
def __init__(self, fn,... | 3,484 | 29.304348 | 226 | py |
BrnoLM | BrnoLM-master/brnolm/data_pipeline/aug_paper_pipeline.py | import numpy as np
import random
import torch
def form_input_targets(stream):
return stream[:-1], stream[1:]
class CleanStreamsProvider:
def __init__(self, stream):
self.stream = stream
def provide(self):
return self.stream[:-1], self.stream[1:]
class Corruptor:
def __init__(self,... | 12,152 | 32.946927 | 276 | py |
BrnoLM | BrnoLM-master/brnolm/language_models/ffnn_models.py | import torch
import torch.nn as nn
import torch.nn.functional as F
class BengioModel(nn.Module):
"""Container module with an encoder, a recurrent module, and a decoder."""
def __init__(self, ntoken, emb_size, in_len, nb_hidden, dropout=0.5):
super().__init__()
self.drop = nn.Dropout(dropout)
... | 3,043 | 35.674699 | 106 | py |
BrnoLM | BrnoLM-master/brnolm/language_models/lstm_model.py | import torch
import torch.nn as nn
from typing import Tuple
class LSTMLanguageModel(nn.Module):
"""Container module with an encoder, a recurrent module, and a decoder."""
def __init__(self, token_encoder, dim_input, dim_lstm, nb_layers, dropout=0.5):
super(LSTMLanguageModel, self).__init__()
... | 3,018 | 31.462366 | 93 | py |
BrnoLM | BrnoLM-master/brnolm/language_models/transformer.py | import torch
from torch import nn
import math
from torch.nn import TransformerEncoder, TransformerEncoderLayer
class PositionalEncoding(nn.Module):
r"""Inject some information about the relative or absolute position of the tokens
in the sequence. The positional encodings have the same dimension as
... | 3,545 | 36.326316 | 98 | py |
BrnoLM | BrnoLM-master/brnolm/language_models/decoders.py | import torch
class FullSoftmaxDecoder(torch.nn.Module):
def __init__(self, nb_hidden, nb_output, init_range=0.1):
super().__init__()
self.projection = torch.nn.Linear(nb_hidden, nb_output)
self.log_softmax = torch.nn.LogSoftmax(dim=-1)
self.projection.weight.data.uniform_(-init_r... | 2,752 | 31.011628 | 104 | py |
BrnoLM | BrnoLM-master/brnolm/language_models/language_model.py | import torch
import pickle
import zipfile
from brnolm.data_pipeline.masked import masked_tensor_from_sentences
# these helper functions are being developed against an LSTM implementation
# It's ugly, but necessary for Chime. To be made nice later :-)
def detach_hidden_state(h):
if isinstance(h, tuple):
... | 5,174 | 31.34375 | 96 | py |
BrnoLM | BrnoLM-master/brnolm/language_models/encoders.py | import torch.nn as nn
class FlatEmbedding(nn.Module):
def __init__(self, nb_tokens, dim_embs, init_range=0.1):
super().__init__()
self.embeddings = nn.Embedding(nb_tokens, dim_embs)
nn.init.uniform_(self.embeddings.weight, -init_range, init_range)
def forward(self, x):
return ... | 339 | 27.333333 | 73 | py |
BrnoLM | BrnoLM-master/brnolm/smm_itf/xtract-ivecs-example.py | #!/usr/bin/env python3
# -*- coding: utf-8 -*-
# author : KarelB
# e-mail : ibenes AT fit.vutbr.cz
import argparse
import torch
import .smm_ivec_extractor
def bow_from_sentence(sentence, vocab):
bow = torch.zeros((1, len(vocab))).float()
for w in sentence.split():
bow[0, vocab[w]] += 1.0
retu... | 1,172 | 25.066667 | 98 | py |
BrnoLM | BrnoLM-master/brnolm/smm_itf/smm_ivec_extractor.py | import io
import tempfile
import pickle
import numpy as np
import torch
from smm import update_ws
class IvecExtractor():
def __init__(self, model, nb_iters, lr, tokenizer):
self._model = model
self._nb_iters = nb_iters
self._lr = lr
self._tokenizer = tokenizer
def __call__(s... | 4,743 | 31.493151 | 111 | py |
robust_overfitting | robust_overfitting-master/generate_validation.py | import torch
import torchvision
import numpy as np
np.random.seed(0)
m = 50000
P = np.random.permutation(m)
n = 1000
def cifar10(root):
train_set = torchvision.datasets.CIFAR10(root=root, train=True, download=True)
test_set = torchvision.datasets.CIFAR10(root=root, train=False, download=True)
return {
... | 967 | 25.162162 | 82 | py |
robust_overfitting | robust_overfitting-master/train_cifar_semisupervised_half.py | import argparse
import logging
import sys
import time
import math
import pickle
import numpy as np
import torch
import torch.nn as nn
import torch.nn.functional as F
from torch.autograd import Variable
import os
from wideresnet import WideResNet
from preactresnet import PreActResNet18
from utils import *
mu = torc... | 19,326 | 39.860465 | 192 | py |
robust_overfitting | robust_overfitting-master/preactresnet.py | '''Pre-activation ResNet in PyTorch.
Reference:
[1] Kaiming He, Xiangyu Zhang, Shaoqing Ren, Jian Sun
Identity Mappings in Deep Residual Networks. arXiv:1603.05027
'''
import torch
import torch.nn as nn
import torch.nn.functional as F
class PreActBlock(nn.Module):
'''Pre-activation version of the BasicBlock.... | 4,208 | 33.785124 | 102 | py |
robust_overfitting | robust_overfitting-master/utils.py | import numpy as np
from collections import namedtuple
import torch
from torch import nn
import torchvision
device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
################################################################
## Components from https://github.com/davidcpage/cifar10-fast ##
###########... | 4,043 | 32.421488 | 122 | py |
robust_overfitting | robust_overfitting-master/train_svhn.py | import argparse
import logging
import sys
import time
import math
import numpy as np
import torch
import torch.nn as nn
import torch.nn.functional as F
from torch.autograd import Variable
from torchvision import datasets, transforms
import os
from wideresnet import WideResNet
from preactresnet import PreActResNet18
... | 13,458 | 37.127479 | 192 | py |
robust_overfitting | robust_overfitting-master/train_cifar.py | import argparse
import logging
import sys
import time
import math
import numpy as np
import torch
import torch.nn as nn
import torch.nn.functional as F
from torch.autograd import Variable
import os
from wideresnet import WideResNet
from preactresnet import PreActResNet18
from utils import *
mu = torch.tensor(cifar... | 18,952 | 40.202174 | 192 | py |
robust_overfitting | robust_overfitting-master/train_cifar100.py |
import argparse
import logging
import sys
import time
import math
import numpy as np
import torch
import torch.nn as nn
import torch.nn.functional as F
from torch.autograd import Variable
from torchvision import datasets, transforms
import os
from wideresnet import WideResNet
from preactresnet import PreActResNet18... | 12,725 | 36.650888 | 192 | py |
robust_overfitting | robust_overfitting-master/wideresnet.py | import math
import torch
import torch.nn as nn
import torch.nn.functional as F
class BasicBlock(nn.Module):
def __init__(self, in_planes, out_planes, stride, dropRate=0.0):
super(BasicBlock, self).__init__()
self.bn1 = nn.BatchNorm2d(in_planes)
self.relu1 = nn.ReLU(inplace=True)
se... | 3,685 | 42.880952 | 116 | py |
lagros | lagros-main/Codes/cartpole.py | #!/usr/bin/env python3
# -*- coding: utf-8 -*-
"""
Created on Fri Dec 11 21:27:54 2020
@author: hiroyasu
"""
import numpy as np
import cvxpy as cp
import scipy as sp
from matplotlib import pyplot as plt
from matplotlib import animation
import control
import time
from tensorflow.keras.models import Sequential
from te... | 30,165 | 33.633754 | 117 | py |
lagros | lagros-main/Codes/test.py | #!/usr/bin/env python3
# -*- coding: utf-8 -*-
"""
Created on Mon Feb 8 16:52:25 2021
@author: hiroyasu
"""
import numpy as np
import cvxpy as cp
from matplotlib import pyplot as plt
from mpl_toolkits.mplot3d import Axes3D
from matplotlib import animation
import control
import time
from tensorflow.keras.models imp... | 65,740 | 37.648442 | 194 | py |
lagros | lagros-main/Codes/multi_agent_leo.py | #!/usr/bin/env python3
# -*- coding: utf-8 -*-
"""
Created on Mon Dec 7 14:46:55 2020
@author: hiroyasu
"""
import numpy as np
import cvxpy as cp
from matplotlib import pyplot as plt
from mpl_toolkits.mplot3d import Axes3D
from matplotlib import animation
import control
import time
from tensorflow.keras.models impo... | 68,117 | 37.902342 | 194 | py |
lagros | lagros-main/Codes/multi_agent_sc_obs.py | #!/usr/bin/env python3
# -*- coding: utf-8 -*-
"""
Created on Thu Dec 10 11:56:53 2020
@author: hiroyasu
"""
import numpy as np
import cvxpy as cp
from matplotlib import pyplot as plt
from matplotlib import animation
import control
import time
from tensorflow.keras.models import Sequential
from tensorflow.keras.laye... | 61,172 | 38.364865 | 135 | py |
lagros | lagros-main/Codes/classncm.py | #!/usr/bin/env python3
# -*- coding: utf-8 -*-
"""
MIT License
Copyright (c) 2020 Hiroyasu Tsukamoto https://hirotsukamoto.com/
Permission is hereby granted, free of charge, to any person obtaining a copy
of this software and associated documentation files (the "Software"), to deal
in the Software without restricti... | 47,298 | 37.865242 | 80 | py |
awesome-align | awesome-align-master/setup.py | from setuptools import setup, find_packages
import codecs
setup(
name='awesome_align',
install_requires=[
'tokenizers>=0.5.2',
'torch>=1.2.0',
'tqdm',
'numpy',
'boto3',
'filelock',
'requests'
],
version='0.1.7',
author='NeuLab',
author_ema... | 811 | 25.193548 | 71 | py |
awesome-align | awesome-align-master/awesome_align/configuration_utils.py | # coding=utf-8
# Copyright 2018 The Google AI Language Team Authors and The HuggingFace Inc. team.
# Copyright (c) 2018, NVIDIA CORPORATION. All rights reserved.
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a cop... | 17,149 | 47.583569 | 193 | py |
awesome-align | awesome-align-master/awesome_align/modeling_utils.py | # coding=utf-8
# Copyright 2018 The Google AI Language Team Authors, Facebook AI Research authors and The HuggingFace Inc. team.
# Copyright (c) 2018, NVIDIA CORPORATION. All rights reserved.
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the L... | 80,266 | 48.578135 | 472 | py |
awesome-align | awesome-align-master/awesome_align/modeling.py | # coding=utf-8
# Copyright 2018 The Google AI Language Team Authors and The HuggingFace Inc. team.
# Copyright (c) 2018, NVIDIA CORPORATION. All rights reserved.
# Modifications copyright (C) 2020 Zi-Yi Dou
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compli... | 31,151 | 41.499318 | 228 | py |
awesome-align | awesome-align-master/awesome_align/tokenization_bert.py | # coding=utf-8
# Copyright 2018 The Google AI Language Team Authors and The HuggingFace Inc. team.
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
# http://www.apache.org/licenses/LICEN... | 27,055 | 41.341158 | 183 | py |
awesome-align | awesome-align-master/awesome_align/train_utils.py | # coding=utf-8
# Copyright 2018 The Google AI Language Team Authors and The HuggingFace Inc. team.
# Modifications copyright (C) 2020 Zi-Yi Dou
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at... | 9,383 | 41.654545 | 130 | py |
awesome-align | awesome-align-master/awesome_align/sparsemax.py | """
An implementation of entmax (Peters et al., 2019). See
https://arxiv.org/pdf/1905.05702 for detailed description.
This builds on previous work with sparsemax (Martins & Astudillo, 2016).
See https://arxiv.org/pdf/1602.02068.
"""
# Author: Ben Peters
# Author: Vlad Niculae <vlad@vene.ro>
# License: MIT
import torc... | 9,917 | 33.922535 | 78 | py |
awesome-align | awesome-align-master/awesome_align/file_utils.py | """
Utilities for working with the local dataset cache.
This file is adapted from the AllenNLP library at https://github.com/allenai/allennlp
Copyright by the AllenNLP authors.
"""
import fnmatch
import json
import logging
import os
import shutil
import sys
import tarfile
import tempfile
from contextlib import context... | 17,924 | 34.921844 | 144 | py |
awesome-align | awesome-align-master/awesome_align/activations.py | import math
import torch
import torch.nn.functional as F
def swish(x):
return x * torch.sigmoid(x)
def _gelu_python(x):
""" Original Implementation of the gelu activation function in Google Bert repo when initially created.
For information: OpenAI GPT's gelu is slightly different (and gives slightl... | 1,415 | 26.230769 | 111 | py |
awesome-align | awesome-align-master/awesome_align/run_train.py | # coding=utf-8
# Copyright 2018 The Google AI Language Team Authors and The HuggingFace Inc. team.
# Copyright (c) 2018, NVIDIA CORPORATION. All rights reserved.
# Modifications copyright (C) 2020, Zi-Yi Dou
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compl... | 42,790 | 47.188063 | 313 | py |
awesome-align | awesome-align-master/awesome_align/run_align.py | # coding=utf-8
# Copyright 2018 The Google AI Language Team Authors and The HuggingFace Inc. team.
# Copyright (c) 2018, NVIDIA CORPORATION. All rights reserved.
# Modifications copyright (C) 2020 Zi-Yi Dou
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compli... | 12,446 | 40.768456 | 297 | py |
awesome-align | awesome-align-master/awesome_align/tokenization_utils.py | # coding=utf-8
# Copyright 2018 The Open AI Team Authors and The HuggingFace Inc. team.
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
# http://www.apache.org/licenses/LICENSE-2.0
#
# ... | 93,064 | 47.270228 | 372 | py |
head2head | head2head-master/test.py | import time
import os
import numpy as np
import torch
from torch.autograd import Variable
from collections import OrderedDict
from options.test_options import TestOptions
from data.data_loader import CreateDataLoader
from models.models import create_model
import util.util as util
from util.visualizer import Visualizer
... | 4,117 | 38.980583 | 135 | py |
head2head | head2head-master/demo.py | import time
import os
import numpy as np
import torch
import torchvision
import cv2
import dlib
from torch.autograd import Variable
from collections import OrderedDict
from PIL import Image
from multiprocessing import Process, Queue
from torch.multiprocessing import Process as torchProcess
from torch.multiprocessing im... | 14,204 | 44.528846 | 116 | py |
head2head | head2head-master/train.py | import time
import os
import numpy as np
import torch
from torch.autograd import Variable
from collections import OrderedDict
import fractions
from options.train_options import TrainOptions
from data.data_loader import CreateDataLoader
from models.models import create_model
import util.util as util
from util.visualizer... | 10,970 | 50.027907 | 156 | py |
head2head | head2head-master/options/base_options.py | import argparse
import os
from util import util
import torch
class BaseOptions():
def __init__(self):
self.parser = argparse.ArgumentParser()
self.initialized = False
def initialize(self):
self.parser.add_argument('--max_n_sequences', type=int, default=None, help='Maximum number of sub... | 5,556 | 65.154762 | 205 | py |
head2head | head2head-master/models/base_model.py | import os
import torch
import sys
class BaseModel(torch.nn.Module):
def name(self):
return 'BaseModel'
def initialize(self, opt):
self.opt = opt
self.gpu_ids = opt.gpu_ids
self.isTrain = opt.isTrain
self.Tensor = torch.cuda.FloatTensor if self.gpu_ids else torch.Tensor
... | 3,151 | 34.022222 | 122 | py |
head2head | head2head-master/models/flownet.py | import numpy as np
import torch
import sys
from .base_model import BaseModel
class FlowNet(BaseModel):
def name(self):
return 'FlowNet'
def initialize(self, opt):
BaseModel.initialize(self, opt)
# flownet 2
from .flownet2_pytorch import models as flownet2_models
from .... | 2,596 | 39.578125 | 105 | py |
head2head | head2head-master/models/networks.py | import torch
import torch.nn as nn
from torch.nn import init
import functools
from torch.nn.parameter import Parameter
from torch.autograd import Variable
import numpy as np
import torch.nn.functional as F
from torchvision import models
def get_norm_layer(norm_type='instance'):
if norm_type == 'batch':
nor... | 15,375 | 39.893617 | 139 | py |
head2head | head2head-master/models/models.py | import torch.nn as nn
from .head2head_model import Head2HeadModelG
def create_model(opt):
modelG = Head2HeadModelG()
modelG.initialize(opt)
if opt.isTrain and len(opt.gpu_ids):
modelG = nn.DataParallel(modelG, device_ids=opt.gpu_ids)
from .head2head_model import Head2HeadModelD
from... | 676 | 32.85 | 66 | py |
head2head | head2head-master/models/head2head_model.py | import numpy as np
import torch
import torch.nn as nn
import torch.nn.functional as F
import os
import sys
import math
from collections import OrderedDict
from torch.autograd import Variable
import util.util as util
from .base_model import BaseModel
from . import networks
from .flownet2_pytorch.networks.resample2d_pac... | 15,353 | 46.09816 | 151 | py |
head2head | head2head-master/models/flownet2_pytorch/main.py | #!/usr/bin/env python
import torch
import torch.nn as nn
from torch.utils.data import DataLoader
from torch.autograd import Variable
from tensorboardX import SummaryWriter
import argparse, os, sys, subprocess
import setproctitle, colorama
import numpy as np
from tqdm import tqdm
from glob import glob
from os.path imp... | 23,190 | 51.114607 | 233 | py |
head2head | head2head-master/models/flownet2_pytorch/losses.py | '''
Portions of this code copyright 2017, Clement Pinard
'''
# freda (todo) : adversarial loss
import torch
import torch.nn as nn
import math
def EPE(input_flow, target_flow):
return torch.norm(target_flow-input_flow,p=2,dim=1).mean()
class L1(nn.Module):
def __init__(self):
super(L1, self).__init_... | 2,739 | 30.136364 | 136 | py |
head2head | head2head-master/models/flownet2_pytorch/datasets.py | import torch
import torch.utils.data as data
import os, math, random
from os.path import *
import numpy as np
from glob import glob
import utils.frame_utils as frame_utils
from scipy.misc import imread, imresize
class StaticRandomCrop(object):
def __init__(self, image_size, crop_size):
self.th, self.tw ... | 13,305 | 32.857506 | 145 | py |
head2head | head2head-master/models/flownet2_pytorch/models.py | import torch
import torch.nn as nn
from torch.nn import init
import math
import numpy as np
from .networks.resample2d_package.resample2d import Resample2d
from .networks.channelnorm_package.channelnorm import ChannelNorm
from .networks import FlowNetC
from .networks import FlowNetS
from .networks import FlowNetSD
fr... | 18,156 | 38.300866 | 165 | py |
head2head | head2head-master/models/flownet2_pytorch/convert.py | #!/usr/bin/env python2.7
import caffe
from caffe.proto import caffe_pb2
import sys, os
import torch
import torch.nn as nn
import argparse, tempfile
import numpy as np
parser = argparse.ArgumentParser()
parser.add_argument('caffe_model', help='input model in hdf5 or caffemodel format')
parser.add_argument('prototxt_... | 4,703 | 33.335766 | 99 | py |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.