code
stringlengths
81
54k
code_codestyle
int64
0
721
style_context
stringlengths
91
41.9k
style_context_codestyle
int64
0
699
label
int64
0
1
A_ : List[Any] = { 'A': '.-', 'B': '-...', 'C': '-.-.', 'D': '-..', 'E': '.', 'F': '..-.', 'G': '--.', 'H': '....', 'I': '..', 'J': '.---', 'K': '-.-', 'L': '.-..', 'M': '--', 'N': '-.', 'O': '---', 'P': '.--.', 'Q': '--.-', 'R': '.-.', 'S': '...', 'T': '-', 'U': '..-', 'V': '...-', 'W': '.--', 'X': '-..-', 'Y': '-.--', 'Z': '--..', '1': '.----', '2': '..---', '3': '...--', '4': '....-', '5': '.....', '6': '-....', '7': '--...', '8': '---..', '9': '----.', '0': '-----', '&': '.-...', '@': '.--.-.', ':': '---...', ',': '--..--', '.': '.-.-.-', '\'': '.----.', '"': '.-..-.', '?': '..--..', '/': '-..-.', '=': '-...-', '+': '.-.-.', '-': '-....-', '(': '-.--.', ')': '-.--.-', '!': '-.-.--', ' ': '/' } # Exclamation mark is not in ITU-R recommendation # fmt: on A_ : str = {value: key for key, value in MORSE_CODE_DICT.items()} def UpperCamelCase (lowercase_: str ) -> str: return " ".join(MORSE_CODE_DICT[char] for char in message.upper() ) def UpperCamelCase (lowercase_: str ) -> str: return "".join(REVERSE_DICT[char] for char in message.split() ) def UpperCamelCase () -> None: A__ : int = """Morse code here!""" print(lowercase_ ) A__ : Optional[Any] = encrypt(lowercase_ ) print(lowercase_ ) A__ : List[Any] = decrypt(lowercase_ ) print(lowercase_ ) if __name__ == "__main__": main()
64
import itertools import random import unittest import numpy as np from transformers import ASTFeatureExtractor from transformers.testing_utils import require_torch, require_torchaudio from transformers.utils.import_utils import is_torch_available from ...test_sequence_feature_extraction_common import SequenceFeatureExtractionTestMixin A_ : Dict = random.Random() if is_torch_available(): import torch def UpperCamelCase (lowercase_: Tuple , lowercase_: Tuple=1.0 , lowercase_: Dict=None , lowercase_: int=None ) -> str: if rng is None: A__ : Optional[Any] = global_rng A__ : List[str] = [] for batch_idx in range(shape[0] ): values.append([] ) for _ in range(shape[1] ): values[-1].append(rng.random() * scale ) return values class _a (unittest.TestCase ): '''simple docstring''' def __init__( self , A__ , A__=7 , A__=400 , A__=2000 , A__=1 , A__=0.0 , A__=1_6000 , A__=True , A__=True , ): A__ : Any = parent A__ : Optional[int] = batch_size A__ : Union[str, Any] = min_seq_length A__ : Dict = max_seq_length A__ : Tuple = (self.max_seq_length - self.min_seq_length) // (self.batch_size - 1) A__ : str = feature_size A__ : Optional[int] = padding_value A__ : List[str] = sampling_rate A__ : List[str] = return_attention_mask A__ : int = do_normalize def __A ( self ): return { "feature_size": self.feature_size, "padding_value": self.padding_value, "sampling_rate": self.sampling_rate, "return_attention_mask": self.return_attention_mask, "do_normalize": self.do_normalize, } def __A ( self , A__=False , A__=False ): def _flatten(A__ ): return list(itertools.chain(*A__ ) ) if equal_length: A__ : Dict = floats_list((self.batch_size, self.max_seq_length) ) else: # make sure that inputs increase in size A__ : Union[str, Any] = [ _flatten(floats_list((x, self.feature_size) ) ) for x in range(self.min_seq_length , self.max_seq_length , self.seq_length_diff ) ] if numpify: A__ : Optional[int] = [np.asarray(A__ ) for x in speech_inputs] return speech_inputs @require_torch @require_torchaudio class _a (__magic_name__ , unittest.TestCase ): '''simple docstring''' UpperCAmelCase__: int = ASTFeatureExtractor def __A ( self ): A__ : Optional[Any] = ASTFeatureExtractionTester(self ) def __A ( self ): # Tests that all call wrap to encode_plus and batch_encode_plus A__ : Optional[int] = self.feature_extraction_class(**self.feat_extract_tester.prepare_feat_extract_dict() ) # create three inputs of length 800, 1000, and 1200 A__ : List[Any] = [floats_list((1, x) )[0] for x in range(800 , 1400 , 200 )] A__ : Optional[Any] = [np.asarray(A__ ) for speech_input in speech_inputs] # Test not batched input A__ : Tuple = feat_extract(speech_inputs[0] , return_tensors="""np""" ).input_values A__ : Tuple = feat_extract(np_speech_inputs[0] , return_tensors="""np""" ).input_values self.assertTrue(np.allclose(A__ , A__ , atol=1e-3 ) ) # Test batched A__ : Tuple = feat_extract(A__ , padding=A__ , return_tensors="""np""" ).input_values A__ : Tuple = feat_extract(A__ , padding=A__ , return_tensors="""np""" ).input_values for enc_seq_a, enc_seq_a in zip(A__ , A__ ): self.assertTrue(np.allclose(A__ , A__ , atol=1e-3 ) ) # Test 2-D numpy arrays are batched. A__ : int = [floats_list((1, x) )[0] for x in (800, 800, 800)] A__ : List[str] = np.asarray(A__ ) A__ : Union[str, Any] = feat_extract(A__ , return_tensors="""np""" ).input_values A__ : Optional[Any] = feat_extract(A__ , return_tensors="""np""" ).input_values for enc_seq_a, enc_seq_a in zip(A__ , A__ ): self.assertTrue(np.allclose(A__ , A__ , atol=1e-3 ) ) @require_torch def __A ( self ): import torch A__ : Union[str, Any] = self.feature_extraction_class(**self.feat_extract_tester.prepare_feat_extract_dict() ) A__ : Tuple = np.random.rand(100 ).astype(np.floataa ) A__ : Tuple = np_speech_inputs.tolist() for inputs in [py_speech_inputs, np_speech_inputs]: A__ : List[str] = feature_extractor.pad([{"""input_values""": inputs}] , return_tensors="""np""" ) self.assertTrue(np_processed.input_values.dtype == np.floataa ) A__ : Any = feature_extractor.pad([{"""input_values""": inputs}] , return_tensors="""pt""" ) self.assertTrue(pt_processed.input_values.dtype == torch.floataa ) def __A ( self , A__ ): from datasets import load_dataset A__ : str = load_dataset("""hf-internal-testing/librispeech_asr_dummy""" , """clean""" , split="""validation""" ) # automatic decoding with librispeech A__ : str = ds.sort("""id""" ).select(range(A__ ) )[:num_samples]["""audio"""] return [x["array"] for x in speech_samples] @require_torch def __A ( self ): # fmt: off A__ : Optional[Any] = torch.tensor( [-0.9_8_9_4, -1.2_7_7_6, -0.9_0_6_6, -1.2_7_7_6, -0.9_3_4_9, -1.2_6_0_9, -1.0_3_8_6, -1.2_7_7_6, -1.1_5_6_1, -1.2_7_7_6, -1.2_0_5_2, -1.2_7_2_3, -1.2_1_9_0, -1.2_1_3_2, -1.2_7_7_6, -1.1_1_3_3, -1.1_9_5_3, -1.1_3_4_3, -1.1_5_8_4, -1.2_2_0_3, -1.1_7_7_0, -1.2_4_7_4, -1.2_3_8_1, -1.1_9_3_6, -0.9_2_7_0, -0.8_3_1_7, -0.8_0_4_9, -0.7_7_0_6, -0.7_5_6_5, -0.7_8_6_9] ) # fmt: on A__ : Any = self._load_datasamples(1 ) A__ : Tuple = ASTFeatureExtractor() A__ : Dict = feature_extractor(A__ , return_tensors="""pt""" ).input_values self.assertEquals(input_values.shape , (1, 1024, 128) ) self.assertTrue(torch.allclose(input_values[0, 0, :30] , A__ , atol=1e-4 ) )
64
1
import itertools from dataclasses import dataclass from typing import Any, Callable, Dict, List, Optional, Union import pandas as pd import pyarrow as pa import datasets import datasets.config from datasets.features.features import require_storage_cast from datasets.table import table_cast from datasets.utils.py_utils import Literal A_ : int = datasets.utils.logging.get_logger(__name__) A_ : str = ['names', 'prefix'] A_ : List[Any] = ['warn_bad_lines', 'error_bad_lines', 'mangle_dupe_cols'] A_ : List[Any] = ['encoding_errors', 'on_bad_lines'] A_ : Optional[int] = ['date_format'] @dataclass class _a (datasets.BuilderConfig ): '''simple docstring''' UpperCAmelCase__: str = "," UpperCAmelCase__: Optional[str] = None UpperCAmelCase__: Optional[Union[int, List[int], str]] = "infer" UpperCAmelCase__: Optional[List[str]] = None UpperCAmelCase__: Optional[List[str]] = None UpperCAmelCase__: Optional[Union[int, str, List[int], List[str]]] = None UpperCAmelCase__: Optional[Union[List[int], List[str]]] = None UpperCAmelCase__: Optional[str] = None UpperCAmelCase__: bool = True UpperCAmelCase__: Optional[Literal["c", "python", "pyarrow"]] = None UpperCAmelCase__: Dict[Union[int, str], Callable[[Any], Any]] = None UpperCAmelCase__: Optional[list] = None UpperCAmelCase__: Optional[list] = None UpperCAmelCase__: bool = False UpperCAmelCase__: Optional[Union[int, List[int]]] = None UpperCAmelCase__: Optional[int] = None UpperCAmelCase__: Optional[Union[str, List[str]]] = None UpperCAmelCase__: bool = True UpperCAmelCase__: bool = True UpperCAmelCase__: bool = False UpperCAmelCase__: bool = True UpperCAmelCase__: Optional[str] = None UpperCAmelCase__: str = "." UpperCAmelCase__: Optional[str] = None UpperCAmelCase__: str = '"' UpperCAmelCase__: int = 0 UpperCAmelCase__: Optional[str] = None UpperCAmelCase__: Optional[str] = None UpperCAmelCase__: Optional[str] = None UpperCAmelCase__: Optional[str] = None UpperCAmelCase__: bool = True UpperCAmelCase__: bool = True UpperCAmelCase__: int = 0 UpperCAmelCase__: bool = True UpperCAmelCase__: bool = False UpperCAmelCase__: Optional[str] = None UpperCAmelCase__: int = 1_00_00 UpperCAmelCase__: Optional[datasets.Features] = None UpperCAmelCase__: Optional[str] = "strict" UpperCAmelCase__: Literal["error", "warn", "skip"] = "error" UpperCAmelCase__: Optional[str] = None def __A ( self ): if self.delimiter is not None: A__ : Any = self.delimiter if self.column_names is not None: A__ : Optional[Any] = self.column_names @property def __A ( self ): A__ : Tuple = { """sep""": self.sep, """header""": self.header, """names""": self.names, """index_col""": self.index_col, """usecols""": self.usecols, """prefix""": self.prefix, """mangle_dupe_cols""": self.mangle_dupe_cols, """engine""": self.engine, """converters""": self.converters, """true_values""": self.true_values, """false_values""": self.false_values, """skipinitialspace""": self.skipinitialspace, """skiprows""": self.skiprows, """nrows""": self.nrows, """na_values""": self.na_values, """keep_default_na""": self.keep_default_na, """na_filter""": self.na_filter, """verbose""": self.verbose, """skip_blank_lines""": self.skip_blank_lines, """thousands""": self.thousands, """decimal""": self.decimal, """lineterminator""": self.lineterminator, """quotechar""": self.quotechar, """quoting""": self.quoting, """escapechar""": self.escapechar, """comment""": self.comment, """encoding""": self.encoding, """dialect""": self.dialect, """error_bad_lines""": self.error_bad_lines, """warn_bad_lines""": self.warn_bad_lines, """skipfooter""": self.skipfooter, """doublequote""": self.doublequote, """memory_map""": self.memory_map, """float_precision""": self.float_precision, """chunksize""": self.chunksize, """encoding_errors""": self.encoding_errors, """on_bad_lines""": self.on_bad_lines, """date_format""": self.date_format, } # some kwargs must not be passed if they don't have a default value # some others are deprecated and we can also not pass them if they are the default value for pd_read_csv_parameter in _PANDAS_READ_CSV_NO_DEFAULT_PARAMETERS + _PANDAS_READ_CSV_DEPRECATED_PARAMETERS: if pd_read_csv_kwargs[pd_read_csv_parameter] == getattr(CsvConfig() , A__ ): del pd_read_csv_kwargs[pd_read_csv_parameter] # Remove 2.0 new arguments if not (datasets.config.PANDAS_VERSION.major >= 2): for pd_read_csv_parameter in _PANDAS_READ_CSV_NEW_2_0_0_PARAMETERS: del pd_read_csv_kwargs[pd_read_csv_parameter] # Remove 1.3 new arguments if not (datasets.config.PANDAS_VERSION.major >= 1 and datasets.config.PANDAS_VERSION.minor >= 3): for pd_read_csv_parameter in _PANDAS_READ_CSV_NEW_1_3_0_PARAMETERS: del pd_read_csv_kwargs[pd_read_csv_parameter] return pd_read_csv_kwargs class _a (datasets.ArrowBasedBuilder ): '''simple docstring''' UpperCAmelCase__: Tuple = CsvConfig def __A ( self ): return datasets.DatasetInfo(features=self.config.features ) def __A ( self , A__ ): if not self.config.data_files: raise ValueError(F"""At least one data file must be specified, but got data_files={self.config.data_files}""" ) A__ : Any = dl_manager.download_and_extract(self.config.data_files ) if isinstance(A__ , (str, list, tuple) ): A__ : Dict = data_files if isinstance(A__ , A__ ): A__ : Any = [files] A__ : Dict = [dl_manager.iter_files(A__ ) for file in files] return [datasets.SplitGenerator(name=datasets.Split.TRAIN , gen_kwargs={"""files""": files} )] A__ : List[Any] = [] for split_name, files in data_files.items(): if isinstance(A__ , A__ ): A__ : Dict = [files] A__ : int = [dl_manager.iter_files(A__ ) for file in files] splits.append(datasets.SplitGenerator(name=A__ , gen_kwargs={"""files""": files} ) ) return splits def __A ( self , A__ ): if self.config.features is not None: A__ : List[str] = self.config.features.arrow_schema if all(not require_storage_cast(A__ ) for feature in self.config.features.values() ): # cheaper cast A__ : Any = pa.Table.from_arrays([pa_table[field.name] for field in schema] , schema=A__ ) else: # more expensive cast; allows str <-> int/float or str to Audio for example A__ : Optional[int] = table_cast(A__ , A__ ) return pa_table def __A ( self , A__ ): A__ : Any = self.config.features.arrow_schema if self.config.features else None # dtype allows reading an int column as str A__ : Optional[int] = ( { name: dtype.to_pandas_dtype() if not require_storage_cast(A__ ) else object for name, dtype, feature in zip(schema.names , schema.types , self.config.features.values() ) } if schema is not None else None ) for file_idx, file in enumerate(itertools.chain.from_iterable(A__ ) ): A__ : Union[str, Any] = pd.read_csv(A__ , iterator=A__ , dtype=A__ , **self.config.pd_read_csv_kwargs ) try: for batch_idx, df in enumerate(A__ ): A__ : Optional[int] = pa.Table.from_pandas(A__ ) # Uncomment for debugging (will print the Arrow table size and elements) # logger.warning(f"pa_table: {pa_table} num rows: {pa_table.num_rows}") # logger.warning('\n'.join(str(pa_table.slice(i, 1).to_pydict()) for i in range(pa_table.num_rows))) yield (file_idx, batch_idx), self._cast_table(A__ ) except ValueError as e: logger.error(F"""Failed to read file '{file}' with error {type(A__ )}: {e}""" ) raise
64
from typing import Optional import numpy as np import torch from torch import nn from transformers import GPTaConfig, GPTaLMHeadModel from transformers.modeling_utils import ModuleUtilsMixin from ...configuration_utils import ConfigMixin, register_to_config from ...models import ModelMixin class _a (__magic_name__ , __magic_name__ , __magic_name__ ): '''simple docstring''' UpperCAmelCase__: str = [r'''h\.\d+\.attn\.bias''', r'''h\.\d+\.attn\.masked_bias'''] @register_to_config def __init__( self , A__ , A__ , A__ = None , A__ = 5_0257 , A__ = 1024 , A__ = 768 , A__ = 12 , A__ = 12 , A__ = None , A__ = "gelu_new" , A__ = 0.1 , A__ = 0.1 , A__ = 0.1 , A__ = 1e-5 , A__ = 0.0_2 , A__ = True , A__ = True , A__ = False , A__ = False , ): super().__init__() A__ : Union[str, Any] = prefix_length if prefix_inner_dim != n_embd and prefix_hidden_dim is None: raise ValueError( F"""`prefix_hidden_dim` cannot be `None` when `prefix_inner_dim`: {prefix_hidden_dim} and""" F""" `n_embd`: {n_embd} are not equal.""" ) A__ : str = prefix_inner_dim A__ : Optional[Any] = prefix_hidden_dim A__ : Tuple = ( nn.Linear(self.prefix_inner_dim , self.prefix_hidden_dim ) if self.prefix_hidden_dim is not None else nn.Identity() ) A__ : int = ( nn.Linear(self.prefix_hidden_dim , A__ ) if self.prefix_hidden_dim is not None else nn.Identity() ) A__ : Tuple = GPTaConfig( vocab_size=A__ , n_positions=A__ , n_embd=A__ , n_layer=A__ , n_head=A__ , n_inner=A__ , activation_function=A__ , resid_pdrop=A__ , embd_pdrop=A__ , attn_pdrop=A__ , layer_norm_epsilon=A__ , initializer_range=A__ , scale_attn_weights=A__ , use_cache=A__ , scale_attn_by_inverse_layer_idx=A__ , reorder_and_upcast_attn=A__ , ) A__ : int = GPTaLMHeadModel(A__ ) def __A ( self , A__ , A__ , A__ = None , A__ = None , ): A__ : List[str] = self.transformer.transformer.wte(A__ ) A__ : int = self.encode_prefix(A__ ) A__ : int = self.decode_prefix(A__ ) A__ : Optional[Any] = torch.cat((prefix_embeds, embedding_text) , dim=1 ) if labels is not None: A__ : Any = self.get_dummy_token(input_ids.shape[0] , input_ids.device ) A__ : List[Any] = torch.cat((dummy_token, input_ids) , dim=1 ) A__ : List[str] = self.transformer(inputs_embeds=A__ , labels=A__ , attention_mask=A__ ) if self.prefix_hidden_dim is not None: return out, hidden else: return out def __A ( self , A__ , A__ ): return torch.zeros(A__ , self.prefix_length , dtype=torch.intaa , device=A__ ) def __A ( self , A__ ): return self.encode_prefix(A__ ) @torch.no_grad() def __A ( self , A__ , A__ , A__ ): A__ : List[Any] = torch.split(A__ , 1 , dim=0 ) A__ : Optional[int] = [] A__ : str = [] for feature in features: A__ : Dict = self.decode_prefix(feature.to(A__ ) ) # back to the clip feature # Only support beam search for now A__ , A__ : Union[str, Any] = self.generate_beam( input_embeds=A__ , device=A__ , eos_token_id=A__ ) generated_tokens.append(output_tokens[0] ) generated_seq_lengths.append(seq_lengths[0] ) A__ : int = torch.stack(A__ ) A__ : List[Any] = torch.stack(A__ ) return generated_tokens, generated_seq_lengths @torch.no_grad() def __A ( self , A__=None , A__=None , A__=None , A__ = 5 , A__ = 67 , A__ = 1.0 , A__ = None , ): A__ : Any = eos_token_id A__ : Any = None A__ : Optional[int] = None A__ : Optional[Any] = torch.ones(A__ , device=A__ , dtype=torch.int ) A__ : Any = torch.zeros(A__ , device=A__ , dtype=torch.bool ) if input_embeds is not None: A__ : Dict = input_embeds else: A__ : str = self.transformer.transformer.wte(A__ ) for i in range(A__ ): A__ : Dict = self.transformer(inputs_embeds=A__ ) A__ : str = outputs.logits A__ : Union[str, Any] = logits[:, -1, :] / (temperature if temperature > 0 else 1.0) A__ : Any = logits.softmax(-1 ).log() if scores is None: A__ , A__ : Optional[int] = logits.topk(A__ , -1 ) A__ : List[Any] = generated.expand(A__ , *generated.shape[1:] ) A__ , A__ : List[Any] = next_tokens.permute(1 , 0 ), scores.squeeze(0 ) if tokens is None: A__ : Optional[Any] = next_tokens else: A__ : List[Any] = tokens.expand(A__ , *tokens.shape[1:] ) A__ : int = torch.cat((tokens, next_tokens) , dim=1 ) else: A__ : Optional[int] = -float(np.inf ) A__ : List[Any] = 0 A__ : str = scores[:, None] + logits seq_lengths[~is_stopped] += 1 A__ : Dict = scores_sum / seq_lengths[:, None] A__ , A__ : List[Any] = scores_sum_average.view(-1 ).topk(A__ , -1 ) A__ : Tuple = next_tokens // scores_sum.shape[1] A__ : Optional[Any] = seq_lengths[next_tokens_source] A__ : List[str] = next_tokens % scores_sum.shape[1] A__ : Optional[int] = next_tokens.unsqueeze(1 ) A__ : int = tokens[next_tokens_source] A__ : List[Any] = torch.cat((tokens, next_tokens) , dim=1 ) A__ : str = generated[next_tokens_source] A__ : Optional[Any] = scores_sum_average * seq_lengths A__ : Union[str, Any] = is_stopped[next_tokens_source] A__ : str = self.transformer.transformer.wte(next_tokens.squeeze() ).view(generated.shape[0] , 1 , -1 ) A__ : Optional[int] = torch.cat((generated, next_token_embed) , dim=1 ) A__ : List[str] = is_stopped + next_tokens.eq(A__ ).squeeze() if is_stopped.all(): break A__ : Dict = scores / seq_lengths A__ : Dict = scores.argsort(descending=A__ ) # tokens tensors are already padded to max_seq_length A__ : Union[str, Any] = [tokens[i] for i in order] A__ : Any = torch.stack(A__ , dim=0 ) A__ : Dict = torch.tensor([seq_lengths[i] for i in order] , dtype=seq_lengths.dtype ) return output_texts, seq_lengths
64
1
import itertools import json import os import unittest from transformers import AddedToken, LongformerTokenizer, LongformerTokenizerFast from transformers.models.longformer.tokenization_longformer import VOCAB_FILES_NAMES from transformers.testing_utils import require_tokenizers, slow from ...test_tokenization_common import TokenizerTesterMixin @require_tokenizers class _a (__magic_name__ , unittest.TestCase ): '''simple docstring''' UpperCAmelCase__: Dict = LongformerTokenizer UpperCAmelCase__: Optional[Any] = True UpperCAmelCase__: Union[str, Any] = LongformerTokenizerFast UpperCAmelCase__: Optional[Any] = True def __A ( self ): super().setUp() # Adapted from Sennrich et al. 2015 and https://github.com/rsennrich/subword-nmt A__ : Optional[int] = [ """l""", """o""", """w""", """e""", """r""", """s""", """t""", """i""", """d""", """n""", """\u0120""", """\u0120l""", """\u0120n""", """\u0120lo""", """\u0120low""", """er""", """\u0120lowest""", """\u0120newer""", """\u0120wider""", """<unk>""", ] A__ : Optional[Any] = dict(zip(A__ , range(len(A__ ) ) ) ) A__ : Any = ["""#version: 0.2""", """\u0120 l""", """\u0120l o""", """\u0120lo w""", """e r""", """"""] A__ : str = {"""unk_token""": """<unk>"""} A__ : Dict = os.path.join(self.tmpdirname , VOCAB_FILES_NAMES["""vocab_file"""] ) A__ : Optional[int] = os.path.join(self.tmpdirname , VOCAB_FILES_NAMES["""merges_file"""] ) with open(self.vocab_file , """w""" , encoding="""utf-8""" ) as fp: fp.write(json.dumps(A__ ) + """\n""" ) with open(self.merges_file , """w""" , encoding="""utf-8""" ) as fp: fp.write("""\n""".join(A__ ) ) def __A ( self , **A__ ): kwargs.update(self.special_tokens_map ) return self.tokenizer_class.from_pretrained(self.tmpdirname , **A__ ) def __A ( self , **A__ ): kwargs.update(self.special_tokens_map ) return self.rust_tokenizer_class.from_pretrained(self.tmpdirname , **A__ ) def __A ( self , A__ ): A__ : Dict = """lower newer""" A__ : Dict = """lower newer""" return input_text, output_text def __A ( self ): A__ : List[Any] = self.tokenizer_class(self.vocab_file , self.merges_file , **self.special_tokens_map ) A__ : Union[str, Any] = """lower newer""" A__ : Dict = ["""l""", """o""", """w""", """er""", """\u0120""", """n""", """e""", """w""", """er"""] A__ : List[Any] = tokenizer.tokenize(A__ ) # , add_prefix_space=True) self.assertListEqual(A__ , A__ ) A__ : Dict = tokens + [tokenizer.unk_token] A__ : List[str] = [0, 1, 2, 15, 10, 9, 3, 2, 15, 19] self.assertListEqual(tokenizer.convert_tokens_to_ids(A__ ) , A__ ) def __A ( self ): A__ : Tuple = self.get_tokenizer() self.assertListEqual(tokenizer.encode("""Hello world!""" , add_special_tokens=A__ ) , [0, 3_1414, 232, 328, 2] ) self.assertListEqual( tokenizer.encode("""Hello world! cécé herlolip 418""" , add_special_tokens=A__ ) , [0, 3_1414, 232, 328, 740, 1140, 1_2695, 69, 4_6078, 1588, 2] , ) @slow def __A ( self ): A__ : Tuple = self.tokenizer_class.from_pretrained("""allenai/longformer-base-4096""" ) A__ : Dict = tokenizer.encode("""sequence builders""" , add_special_tokens=A__ ) A__ : Optional[int] = tokenizer.encode("""multi-sequence build""" , add_special_tokens=A__ ) A__ : Any = tokenizer.encode( """sequence builders""" , add_special_tokens=A__ , add_prefix_space=A__ ) A__ : List[str] = tokenizer.encode( """sequence builders""" , """multi-sequence build""" , add_special_tokens=A__ , add_prefix_space=A__ ) A__ : Tuple = tokenizer.build_inputs_with_special_tokens(A__ ) A__ : List[Any] = tokenizer.build_inputs_with_special_tokens(A__ , A__ ) assert encoded_sentence == encoded_text_from_decode assert encoded_pair == encoded_pair_from_decode def __A ( self ): A__ : Any = self.get_tokenizer() A__ : str = """Encode this sequence.""" A__ : List[Any] = tokenizer.byte_encoder[""" """.encode("""utf-8""" )[0]] # Testing encoder arguments A__ : Optional[Any] = tokenizer.encode(A__ , add_special_tokens=A__ , add_prefix_space=A__ ) A__ : Optional[int] = tokenizer.convert_ids_to_tokens(encoded[0] )[0] self.assertNotEqual(A__ , A__ ) A__ : str = tokenizer.encode(A__ , add_special_tokens=A__ , add_prefix_space=A__ ) A__ : Optional[int] = tokenizer.convert_ids_to_tokens(encoded[0] )[0] self.assertEqual(A__ , A__ ) tokenizer.add_special_tokens({"""bos_token""": """<s>"""} ) A__ : Optional[int] = tokenizer.encode(A__ , add_special_tokens=A__ ) A__ : Any = tokenizer.convert_ids_to_tokens(encoded[1] )[0] self.assertNotEqual(A__ , A__ ) # Testing spaces after special tokens A__ : Dict = """<mask>""" tokenizer.add_special_tokens( {"""mask_token""": AddedToken(A__ , lstrip=A__ , rstrip=A__ )} ) # mask token has a left space A__ : Dict = tokenizer.convert_tokens_to_ids(A__ ) A__ : List[Any] = """Encode <mask> sequence""" A__ : Tuple = """Encode <mask>sequence""" A__ : Optional[int] = tokenizer.encode(A__ ) A__ : Union[str, Any] = encoded.index(A__ ) A__ : Optional[Any] = tokenizer.convert_ids_to_tokens(encoded[mask_loc + 1] )[0] self.assertEqual(A__ , A__ ) A__ : Optional[int] = tokenizer.encode(A__ ) A__ : Union[str, Any] = encoded.index(A__ ) A__ : Union[str, Any] = tokenizer.convert_ids_to_tokens(encoded[mask_loc + 1] )[0] self.assertNotEqual(A__ , A__ ) def __A ( self ): pass def __A ( self ): for tokenizer, pretrained_name, kwargs in self.tokenizers_list: with self.subTest(F"""{tokenizer.__class__.__name__} ({pretrained_name})""" ): A__ : List[Any] = self.rust_tokenizer_class.from_pretrained(A__ , **A__ ) A__ : int = self.tokenizer_class.from_pretrained(A__ , **A__ ) A__ : Optional[Any] = """A, <mask> AllenNLP sentence.""" A__ : Union[str, Any] = tokenizer_r.encode_plus(A__ , add_special_tokens=A__ , return_token_type_ids=A__ ) A__ : Dict = tokenizer_p.encode_plus(A__ , add_special_tokens=A__ , return_token_type_ids=A__ ) # token_type_ids should put 0 everywhere self.assertEqual(sum(tokens_r["""token_type_ids"""] ) , sum(tokens_p["""token_type_ids"""] ) ) # attention_mask should put 1 everywhere, so sum over length should be 1 self.assertEqual( sum(tokens_r["""attention_mask"""] ) / len(tokens_r["""attention_mask"""] ) , sum(tokens_p["""attention_mask"""] ) / len(tokens_p["""attention_mask"""] ) , ) A__ : Any = tokenizer_r.convert_ids_to_tokens(tokens_r["""input_ids"""] ) A__ : int = tokenizer_p.convert_ids_to_tokens(tokens_p["""input_ids"""] ) # Rust correctly handles the space before the mask while python doesnt self.assertSequenceEqual(tokens_p["""input_ids"""] , [0, 250, 6, 5_0264, 3823, 487, 2_1992, 3645, 4, 2] ) self.assertSequenceEqual(tokens_r["""input_ids"""] , [0, 250, 6, 5_0264, 3823, 487, 2_1992, 3645, 4, 2] ) self.assertSequenceEqual( A__ , ["""<s>""", """A""", """,""", """<mask>""", """ĠAllen""", """N""", """LP""", """Ġsentence""", """.""", """</s>"""] ) self.assertSequenceEqual( A__ , ["""<s>""", """A""", """,""", """<mask>""", """ĠAllen""", """N""", """LP""", """Ġsentence""", """.""", """</s>"""] ) def __A ( self ): for trim_offsets, add_prefix_space in itertools.product([True, False] , repeat=2 ): A__ : str = self.rust_tokenizer_class.from_pretrained( self.tmpdirname , use_fast=A__ , add_prefix_space=A__ , trim_offsets=A__ ) A__ : Dict = json.loads(tokenizer_r.backend_tokenizer.pre_tokenizer.__getstate__() ) A__ : Optional[int] = json.loads(tokenizer_r.backend_tokenizer.post_processor.__getstate__() ) self.assertEqual(pre_tokenizer_state["""add_prefix_space"""] , A__ ) self.assertEqual(post_processor_state["""add_prefix_space"""] , A__ ) self.assertEqual(post_processor_state["""trim_offsets"""] , A__ ) def __A ( self ): # Test which aims to verify that the offsets are well adapted to the argument `add_prefix_space` and # `trim_offsets` for tokenizer, pretrained_name, kwargs in self.tokenizers_list: with self.subTest(F"""{tokenizer.__class__.__name__} ({pretrained_name})""" ): A__ : List[Any] = """hello""" # `hello` is a token in the vocabulary of `pretrained_name` A__ : int = F"""{text_of_1_token} {text_of_1_token}""" A__ : Union[str, Any] = self.rust_tokenizer_class.from_pretrained( A__ , use_fast=A__ , add_prefix_space=A__ , trim_offsets=A__ ) A__ : Union[str, Any] = tokenizer_r(A__ , return_offsets_mapping=A__ , add_special_tokens=A__ ) self.assertEqual(encoding.offset_mapping[0] , (0, len(A__ )) ) self.assertEqual( encoding.offset_mapping[1] , (len(A__ ) + 1, len(A__ ) + 1 + len(A__ )) , ) A__ : Optional[Any] = self.rust_tokenizer_class.from_pretrained( A__ , use_fast=A__ , add_prefix_space=A__ , trim_offsets=A__ ) A__ : Union[str, Any] = tokenizer_r(A__ , return_offsets_mapping=A__ , add_special_tokens=A__ ) self.assertEqual(encoding.offset_mapping[0] , (0, len(A__ )) ) self.assertEqual( encoding.offset_mapping[1] , (len(A__ ) + 1, len(A__ ) + 1 + len(A__ )) , ) A__ : List[str] = self.rust_tokenizer_class.from_pretrained( A__ , use_fast=A__ , add_prefix_space=A__ , trim_offsets=A__ ) A__ : str = tokenizer_r(A__ , return_offsets_mapping=A__ , add_special_tokens=A__ ) self.assertEqual(encoding.offset_mapping[0] , (0, len(A__ )) ) self.assertEqual( encoding.offset_mapping[1] , (len(A__ ), len(A__ ) + 1 + len(A__ )) , ) A__ : str = self.rust_tokenizer_class.from_pretrained( A__ , use_fast=A__ , add_prefix_space=A__ , trim_offsets=A__ ) A__ : Tuple = tokenizer_r(A__ , return_offsets_mapping=A__ , add_special_tokens=A__ ) self.assertEqual(encoding.offset_mapping[0] , (0, len(A__ )) ) self.assertEqual( encoding.offset_mapping[1] , (len(A__ ), len(A__ ) + 1 + len(A__ )) , ) A__ : Any = F""" {text}""" # tokenizer_r = self.rust_tokenizer_class.from_pretrained( # pretrained_name, use_fast=True, add_prefix_space=True, trim_offsets=True # ) # encoding = tokenizer_r(text, return_offsets_mapping=True, add_special_tokens=False) # self.assertEqual(encoding.offset_mapping[0], (1, 1 + len(text_of_1_token))) # self.assertEqual( # encoding.offset_mapping[1], # (1 + len(text_of_1_token) + 1, 1 + len(text_of_1_token) + 1 + len(text_of_1_token)), # ) A__ : Optional[Any] = self.rust_tokenizer_class.from_pretrained( A__ , use_fast=A__ , add_prefix_space=A__ , trim_offsets=A__ ) A__ : Optional[Any] = tokenizer_r(A__ , return_offsets_mapping=A__ , add_special_tokens=A__ ) self.assertEqual(encoding.offset_mapping[0] , (1, 1 + len(A__ )) ) self.assertEqual( encoding.offset_mapping[1] , (1 + len(A__ ) + 1, 1 + len(A__ ) + 1 + len(A__ )) , ) A__ : int = self.rust_tokenizer_class.from_pretrained( A__ , use_fast=A__ , add_prefix_space=A__ , trim_offsets=A__ ) A__ : Any = tokenizer_r(A__ , return_offsets_mapping=A__ , add_special_tokens=A__ ) self.assertEqual(encoding.offset_mapping[0] , (0, 1 + len(A__ )) ) self.assertEqual( encoding.offset_mapping[1] , (1 + len(A__ ), 1 + len(A__ ) + 1 + len(A__ )) , ) A__ : int = self.rust_tokenizer_class.from_pretrained( A__ , use_fast=A__ , add_prefix_space=A__ , trim_offsets=A__ ) A__ : List[str] = tokenizer_r(A__ , return_offsets_mapping=A__ , add_special_tokens=A__ ) self.assertEqual(encoding.offset_mapping[0] , (0, 1 + len(A__ )) ) self.assertEqual( encoding.offset_mapping[1] , (1 + len(A__ ), 1 + len(A__ ) + 1 + len(A__ )) , )
64
import io import itertools import json from dataclasses import dataclass from typing import Optional import pyarrow as pa import pyarrow.json as paj import datasets from datasets.table import table_cast from datasets.utils.file_utils import readline A_ : Tuple = datasets.utils.logging.get_logger(__name__) @dataclass class _a (datasets.BuilderConfig ): '''simple docstring''' UpperCAmelCase__: Optional[datasets.Features] = None UpperCAmelCase__: str = "utf-8" UpperCAmelCase__: Optional[str] = None UpperCAmelCase__: Optional[str] = None UpperCAmelCase__: bool = True # deprecated UpperCAmelCase__: Optional[int] = None # deprecated UpperCAmelCase__: int = 10 << 20 # 10MB UpperCAmelCase__: Optional[bool] = None class _a (datasets.ArrowBasedBuilder ): '''simple docstring''' UpperCAmelCase__: List[str] = JsonConfig def __A ( self ): if self.config.block_size is not None: logger.warning("""The JSON loader parameter `block_size` is deprecated. Please use `chunksize` instead""" ) A__ : Union[str, Any] = self.config.block_size if self.config.use_threads is not True: logger.warning( """The JSON loader parameter `use_threads` is deprecated and doesn't have any effect anymore.""" ) if self.config.newlines_in_values is not None: raise ValueError("""The JSON loader parameter `newlines_in_values` is no longer supported""" ) return datasets.DatasetInfo(features=self.config.features ) def __A ( self , A__ ): if not self.config.data_files: raise ValueError(F"""At least one data file must be specified, but got data_files={self.config.data_files}""" ) A__ : int = dl_manager.download_and_extract(self.config.data_files ) if isinstance(A__ , (str, list, tuple) ): A__ : Optional[Any] = data_files if isinstance(A__ , A__ ): A__ : List[str] = [files] A__ : int = [dl_manager.iter_files(A__ ) for file in files] return [datasets.SplitGenerator(name=datasets.Split.TRAIN , gen_kwargs={"""files""": files} )] A__ : List[str] = [] for split_name, files in data_files.items(): if isinstance(A__ , A__ ): A__ : Optional[int] = [files] A__ : Optional[int] = [dl_manager.iter_files(A__ ) for file in files] splits.append(datasets.SplitGenerator(name=A__ , gen_kwargs={"""files""": files} ) ) return splits def __A ( self , A__ ): if self.config.features is not None: # adding missing columns for column_name in set(self.config.features ) - set(pa_table.column_names ): A__ : Optional[Any] = self.config.features.arrow_schema.field(A__ ).type A__ : str = pa_table.append_column(A__ , pa.array([None] * len(A__ ) , type=A__ ) ) # more expensive cast to support nested structures with keys in a different order # allows str <-> int/float or str to Audio for example A__ : Optional[int] = table_cast(A__ , self.config.features.arrow_schema ) return pa_table def __A ( self , A__ ): for file_idx, file in enumerate(itertools.chain.from_iterable(A__ ) ): # If the file is one json object and if we need to look at the list of items in one specific field if self.config.field is not None: with open(A__ , encoding=self.config.encoding , errors=self.config.encoding_errors ) as f: A__ : Optional[Any] = json.load(A__ ) # We keep only the field we are interested in A__ : Optional[int] = dataset[self.config.field] # We accept two format: a list of dicts or a dict of lists if isinstance(A__ , (list, tuple) ): A__ : Union[str, Any] = set().union(*[row.keys() for row in dataset] ) A__ : Any = {col: [row.get(A__ ) for row in dataset] for col in keys} else: A__ : Any = dataset A__ : Any = pa.Table.from_pydict(A__ ) yield file_idx, self._cast_table(A__ ) # If the file has one json object per line else: with open(A__ , """rb""" ) as f: A__ : List[str] = 0 # Use block_size equal to the chunk size divided by 32 to leverage multithreading # Set a default minimum value of 16kB if the chunk size is really small A__ : List[str] = max(self.config.chunksize // 32 , 16 << 10 ) A__ : Any = ( self.config.encoding_errors if self.config.encoding_errors is not None else """strict""" ) while True: A__ : Dict = f.read(self.config.chunksize ) if not batch: break # Finish current line try: batch += f.readline() except (AttributeError, io.UnsupportedOperation): batch += readline(A__ ) # PyArrow only accepts utf-8 encoded bytes if self.config.encoding != "utf-8": A__ : List[Any] = batch.decode(self.config.encoding , errors=A__ ).encode("""utf-8""" ) try: while True: try: A__ : str = paj.read_json( io.BytesIO(A__ ) , read_options=paj.ReadOptions(block_size=A__ ) ) break except (pa.ArrowInvalid, pa.ArrowNotImplementedError) as e: if ( isinstance(A__ , pa.ArrowInvalid ) and "straddling" not in str(A__ ) or block_size > len(A__ ) ): raise else: # Increase the block size in case it was too small. # The block size will be reset for the next file. logger.debug( F"""Batch of {len(A__ )} bytes couldn't be parsed with block_size={block_size}. Retrying with block_size={block_size * 2}.""" ) block_size *= 2 except pa.ArrowInvalid as e: try: with open( A__ , encoding=self.config.encoding , errors=self.config.encoding_errors ) as f: A__ : Optional[Any] = json.load(A__ ) except json.JSONDecodeError: logger.error(F"""Failed to read file '{file}' with error {type(A__ )}: {e}""" ) raise e # If possible, parse the file as a list of json objects and exit the loop if isinstance(A__ , A__ ): # list is the only sequence type supported in JSON try: A__ : str = set().union(*[row.keys() for row in dataset] ) A__ : List[str] = {col: [row.get(A__ ) for row in dataset] for col in keys} A__ : int = pa.Table.from_pydict(A__ ) except (pa.ArrowInvalid, AttributeError) as e: logger.error(F"""Failed to read file '{file}' with error {type(A__ )}: {e}""" ) raise ValueError(F"""Not able to read records in the JSON file at {file}.""" ) from None yield file_idx, self._cast_table(A__ ) break else: logger.error(F"""Failed to read file '{file}' with error {type(A__ )}: {e}""" ) raise ValueError( F"""Not able to read records in the JSON file at {file}. """ F"""You should probably indicate the field of the JSON file containing your records. """ F"""This JSON file contain the following fields: {str(list(dataset.keys() ) )}. """ F"""Select the correct one and provide it as `field='XXX'` to the dataset loading method. """ ) from None # Uncomment for debugging (will print the Arrow table size and elements) # logger.warning(f"pa_table: {pa_table} num rows: {pa_table.num_rows}") # logger.warning('\n'.join(str(pa_table.slice(i, 1).to_pydict()) for i in range(pa_table.num_rows))) yield (file_idx, batch_idx), self._cast_table(A__ ) batch_idx += 1
64
1
import unittest import numpy as np from transformers.testing_utils import require_torch, require_vision from transformers.utils import is_torch_available, is_vision_available from ...test_image_processing_common import ImageProcessingSavingTestMixin, prepare_image_inputs if is_torch_available(): import torch if is_vision_available(): from PIL import Image from transformers import GLPNImageProcessor class _a (unittest.TestCase ): '''simple docstring''' def __init__( self , A__ , A__=7 , A__=3 , A__=18 , A__=30 , A__=400 , A__=True , A__=32 , A__=True , ): A__ : List[Any] = parent A__ : List[str] = batch_size A__ : Tuple = num_channels A__ : int = image_size A__ : Any = min_resolution A__ : str = max_resolution A__ : List[Any] = do_resize A__ : str = size_divisor A__ : Optional[int] = do_rescale def __A ( self ): return { "do_resize": self.do_resize, "size_divisor": self.size_divisor, "do_rescale": self.do_rescale, } @require_torch @require_vision class _a (__magic_name__ , unittest.TestCase ): '''simple docstring''' UpperCAmelCase__: List[str] = GLPNImageProcessor if is_vision_available() else None def __A ( self ): A__ : str = GLPNImageProcessingTester(self ) @property def __A ( self ): return self.image_processor_tester.prepare_image_processor_dict() def __A ( self ): A__ : Union[str, Any] = self.image_processing_class(**self.image_processor_dict ) self.assertTrue(hasattr(A__ , """do_resize""" ) ) self.assertTrue(hasattr(A__ , """size_divisor""" ) ) self.assertTrue(hasattr(A__ , """resample""" ) ) self.assertTrue(hasattr(A__ , """do_rescale""" ) ) def __A ( self ): pass def __A ( self ): # Initialize image_processing A__ : Dict = self.image_processing_class(**self.image_processor_dict ) # create random PIL images A__ : int = prepare_image_inputs(self.image_processor_tester , equal_resolution=A__ ) for image in image_inputs: self.assertIsInstance(A__ , Image.Image ) # Test not batched input (GLPNImageProcessor doesn't support batching) A__ : Optional[int] = image_processing(image_inputs[0] , return_tensors="""pt""" ).pixel_values self.assertTrue(encoded_images.shape[-1] % self.image_processor_tester.size_divisor == 0 ) self.assertTrue(encoded_images.shape[-2] % self.image_processor_tester.size_divisor == 0 ) def __A ( self ): # Initialize image_processing A__ : List[str] = self.image_processing_class(**self.image_processor_dict ) # create random numpy tensors A__ : Tuple = prepare_image_inputs(self.image_processor_tester , equal_resolution=A__ , numpify=A__ ) for image in image_inputs: self.assertIsInstance(A__ , np.ndarray ) # Test not batched input (GLPNImageProcessor doesn't support batching) A__ : Any = image_processing(image_inputs[0] , return_tensors="""pt""" ).pixel_values self.assertTrue(encoded_images.shape[-1] % self.image_processor_tester.size_divisor == 0 ) self.assertTrue(encoded_images.shape[-2] % self.image_processor_tester.size_divisor == 0 ) def __A ( self ): # Initialize image_processing A__ : int = self.image_processing_class(**self.image_processor_dict ) # create random PyTorch tensors A__ : List[Any] = prepare_image_inputs(self.image_processor_tester , equal_resolution=A__ , torchify=A__ ) for image in image_inputs: self.assertIsInstance(A__ , torch.Tensor ) # Test not batched input (GLPNImageProcessor doesn't support batching) A__ : Optional[int] = image_processing(image_inputs[0] , return_tensors="""pt""" ).pixel_values self.assertTrue(encoded_images.shape[-1] % self.image_processor_tester.size_divisor == 0 ) self.assertTrue(encoded_images.shape[-2] % self.image_processor_tester.size_divisor == 0 )
64
import os try: from .build_directory_md import good_file_paths except ImportError: from build_directory_md import good_file_paths # type: ignore A_ : Dict = list(good_file_paths()) assert filepaths, "good_file_paths() failed!" A_ : Optional[Any] = [file for file in filepaths if file != file.lower()] if upper_files: print(f'''{len(upper_files)} files contain uppercase characters:''') print('\n'.join(upper_files) + '\n') A_ : Tuple = [file for file in filepaths if ' ' in file] if space_files: print(f'''{len(space_files)} files contain space characters:''') print('\n'.join(space_files) + '\n') A_ : Any = [file for file in filepaths if '-' in file] if hyphen_files: print(f'''{len(hyphen_files)} files contain hyphen characters:''') print('\n'.join(hyphen_files) + '\n') A_ : List[str] = [file for file in filepaths if os.sep not in file] if nodir_files: print(f'''{len(nodir_files)} files are not in a directory:''') print('\n'.join(nodir_files) + '\n') A_ : Any = len(upper_files + space_files + hyphen_files + nodir_files) if bad_files: import sys sys.exit(bad_files)
64
1
import warnings from ...utils import logging from .image_processing_poolformer import PoolFormerImageProcessor A_ : Union[str, Any] = logging.get_logger(__name__) class _a (__magic_name__ ): '''simple docstring''' def __init__( self , *A__ , **A__ ): warnings.warn( """The class PoolFormerFeatureExtractor is deprecated and will be removed in version 5 of Transformers.""" """ Please use PoolFormerImageProcessor instead.""" , A__ , ) super().__init__(*A__ , **A__ )
64
import inspect import warnings from typing import Any, Dict, Optional, Union from packaging import version def UpperCamelCase (*lowercase_: Optional[int] , lowercase_: Optional[Union[Dict, Any]] = None , lowercase_: Dict=True , lowercase_: Tuple=2 ) -> Dict: from .. import __version__ A__ : Dict = take_from A__ : str = () if not isinstance(args[0] , lowercase_ ): A__ : int = (args,) for attribute, version_name, message in args: if version.parse(version.parse(lowercase_ ).base_version ) >= version.parse(lowercase_ ): raise ValueError( f"""The deprecation tuple {(attribute, version_name, message)} should be removed since diffusers'""" f""" version {__version__} is >= {version_name}""" ) A__ : Any = None if isinstance(lowercase_ , lowercase_ ) and attribute in deprecated_kwargs: values += (deprecated_kwargs.pop(lowercase_ ),) A__ : List[str] = f"""The `{attribute}` argument is deprecated and will be removed in version {version_name}.""" elif hasattr(lowercase_ , lowercase_ ): values += (getattr(lowercase_ , lowercase_ ),) A__ : Optional[Any] = f"""The `{attribute}` attribute is deprecated and will be removed in version {version_name}.""" elif deprecated_kwargs is None: A__ : int = f"""`{attribute}` is deprecated and will be removed in version {version_name}.""" if warning is not None: A__ : int = warning + """ """ if standard_warn else """""" warnings.warn(warning + message , lowercase_ , stacklevel=lowercase_ ) if isinstance(lowercase_ , lowercase_ ) and len(lowercase_ ) > 0: A__ : Union[str, Any] = inspect.getouterframes(inspect.currentframe() )[1] A__ : Optional[Any] = call_frame.filename A__ : Optional[int] = call_frame.lineno A__ : Any = call_frame.function A__ , A__ : List[str] = next(iter(deprecated_kwargs.items() ) ) raise TypeError(f"""{function} in {filename} line {line_number-1} got an unexpected keyword argument `{key}`""" ) if len(lowercase_ ) == 0: return elif len(lowercase_ ) == 1: return values[0] return values
64
1
import hashlib import unittest from transformers import MODEL_FOR_DEPTH_ESTIMATION_MAPPING, is_torch_available, is_vision_available from transformers.pipelines import DepthEstimationPipeline, pipeline from transformers.testing_utils import ( is_pipeline_test, nested_simplify, require_tf, require_timm, require_torch, require_vision, slow, ) from .test_pipelines_common import ANY if is_torch_available(): import torch if is_vision_available(): from PIL import Image else: class _a : '''simple docstring''' @staticmethod def __A ( *A__ , **A__ ): pass def UpperCamelCase (lowercase_: Image ) -> str: A__ : Tuple = hashlib.mda(image.tobytes() ) return m.hexdigest() @is_pipeline_test @require_vision @require_timm @require_torch class _a (unittest.TestCase ): '''simple docstring''' UpperCAmelCase__: Dict = MODEL_FOR_DEPTH_ESTIMATION_MAPPING def __A ( self , A__ , A__ , A__ ): A__ : Any = DepthEstimationPipeline(model=A__ , image_processor=A__ ) return depth_estimator, [ "./tests/fixtures/tests_samples/COCO/000000039769.png", "./tests/fixtures/tests_samples/COCO/000000039769.png", ] def __A ( self , A__ , A__ ): A__ : Optional[Any] = depth_estimator("""./tests/fixtures/tests_samples/COCO/000000039769.png""" ) self.assertEqual({"""predicted_depth""": ANY(torch.Tensor ), """depth""": ANY(Image.Image )} , A__ ) import datasets A__ : Optional[int] = datasets.load_dataset("""hf-internal-testing/fixtures_image_utils""" , """image""" , split="""test""" ) A__ : str = depth_estimator( [ Image.open("""./tests/fixtures/tests_samples/COCO/000000039769.png""" ), """http://images.cocodataset.org/val2017/000000039769.jpg""", # RGBA dataset[0]["""file"""], # LA dataset[1]["""file"""], # L dataset[2]["""file"""], ] ) self.assertEqual( [ {"""predicted_depth""": ANY(torch.Tensor ), """depth""": ANY(Image.Image )}, {"""predicted_depth""": ANY(torch.Tensor ), """depth""": ANY(Image.Image )}, {"""predicted_depth""": ANY(torch.Tensor ), """depth""": ANY(Image.Image )}, {"""predicted_depth""": ANY(torch.Tensor ), """depth""": ANY(Image.Image )}, {"""predicted_depth""": ANY(torch.Tensor ), """depth""": ANY(Image.Image )}, ] , A__ , ) @require_tf @unittest.skip("""Depth estimation is not implemented in TF""" ) def __A ( self ): pass @slow @require_torch def __A ( self ): A__ : Optional[Any] = """Intel/dpt-large""" A__ : Dict = pipeline("""depth-estimation""" , model=A__ ) A__ : List[str] = depth_estimator("""http://images.cocodataset.org/val2017/000000039769.jpg""" ) A__ : List[Any] = hashimage(outputs["""depth"""] ) # This seems flaky. # self.assertEqual(outputs["depth"], "1a39394e282e9f3b0741a90b9f108977") self.assertEqual(nested_simplify(outputs["""predicted_depth"""].max().item() ) , 2_9.3_0_4 ) self.assertEqual(nested_simplify(outputs["""predicted_depth"""].min().item() ) , 2.6_6_2 ) @require_torch def __A ( self ): # This is highly irregular to have no small tests. self.skipTest("""There is not hf-internal-testing tiny model for either GLPN nor DPT""" )
64
import argparse import json import os import tensorstore as ts import torch from flax import serialization from flax.traverse_util import flatten_dict, unflatten_dict from tensorflow.io import gfile from transformers.modeling_utils import dtype_byte_size from transformers.models.switch_transformers.convert_switch_transformers_original_flax_checkpoint_to_pytorch import ( rename_keys, ) from transformers.utils import WEIGHTS_INDEX_NAME, WEIGHTS_NAME from transformers.utils.hub import convert_file_size_to_int def UpperCamelCase (lowercase_: List[str] , lowercase_: str ) -> Optional[Any]: if flax_key_tuple[-1] == "kernel" and flax_tensor.ndim == 3: # expert layer A__ : Union[str, Any] = flax_key_tuple[:-1] + ("""weight""",) A__ : Optional[int] = torch.permute(lowercase_ , (0, 2, 1) ) elif flax_key_tuple[-1] == "kernel" and ".".join(lowercase_ ): # linear layer A__ : Optional[Any] = flax_key_tuple[:-1] + ("""weight""",) A__ : int = flax_tensor.T elif flax_key_tuple[-1] in ["scale", "embedding"]: A__ : Optional[int] = flax_key_tuple[:-1] + ("""weight""",) return flax_key_tuple, flax_tensor def UpperCamelCase (lowercase_: Tuple , lowercase_: Optional[int] , lowercase_: str ) -> Union[str, Any]: if "metadata" in layer: A__ : Tuple = layer.split("""metadata""" ) A__ : Optional[Any] = """""".join(split_layer[0] )[:-1] A__ : Optional[Any] = [tuple(("""metadata""" + split_layer[1]).split("""/""" ) )] elif "kvstore" in layer: A__ : str = layer.split("""kvstore""" ) A__ : int = """""".join(split_layer[0] )[:-1] A__ : Optional[int] = [tuple(("""kvstore""" + split_layer[1]).split("""/""" ) )] else: A__ : Any = layer.split("""/""" ) A__ : int = """/""".join(split_layer[:-1] ) A__ : str = (split_layer[-1],) if "kvstore/path" in layer: A__ : Dict = f"""{switch_checkpoint_path}/{checkpoint_info[layer]}""" elif "kvstore/driver" in layer: A__ : Optional[int] = """file""" else: A__ : str = checkpoint_info[layer] return curr_real_layer_name, split_layer, content def UpperCamelCase (lowercase_: str , lowercase_: List[Any] ) -> int: A__ : int = rename_keys(lowercase_ ) A__ : Any = {} for k, v in current_block.items(): A__ : Dict = v A__ : str = new_current_block torch.save(lowercase_ , lowercase_ ) def UpperCamelCase (lowercase_: Dict , lowercase_: Optional[Any] , lowercase_: Optional[Any] , lowercase_: Optional[int] , lowercase_: str = WEIGHTS_NAME ) -> Tuple: A__ : Optional[int] = convert_file_size_to_int(lowercase_ ) A__ : List[Any] = [] A__ : int = {} A__ : List[str] = 0 A__ : Any = 0 os.makedirs(lowercase_ , exist_ok=lowercase_ ) with gfile.GFile(switch_checkpoint_path + """/checkpoint""" , """rb""" ) as fp: A__ : Optional[Any] = serialization.msgpack_restore(fp.read() )["""optimizer"""]["""target"""] A__ : Dict = flatten_dict(lowercase_ , sep="""/""" ) A__ : Any = {} for layer in checkpoint_info.keys(): A__ , A__ , A__ : Union[str, Any] = get_key_and_tensorstore_dict( lowercase_ , lowercase_ , lowercase_ ) if curr_real_layer_name in all_layers: A__ : Optional[int] = content else: A__ : List[Any] = {split_layer[-1]: content} for key in all_layers.keys(): # open tensorstore file A__ : Optional[Any] = ts.open(unflatten_dict(all_layers[key] ) ).result().read().result() A__ : List[Any] = torch.tensor(lowercase_ ) A__ : List[Any] = raw_weights.numel() * dtype_byte_size(raw_weights.dtype ) # use the renaming pattern from the small conversion scripts A__ , A__ : Any = rename_base_flax_keys(tuple(key.split("""/""" ) ) , lowercase_ ) A__ : Any = """/""".join(lowercase_ ) # If this weight is going to tip up over the maximal size, we split. if current_block_size + weight_size > max_shard_size: A__ : List[Any] = os.path.join( lowercase_ , weights_name.replace(""".bin""" , f"""-{len(lowercase_ )+1:05d}-of-???.bin""" ) ) rename_and_save_block(lowercase_ , lowercase_ ) sharded_state_dicts.append(current_block.keys() ) del current_block A__ : Any = {} A__ : str = 0 A__ : List[str] = raw_weights.to(getattr(lowercase_ , lowercase_ ) ) current_block_size += weight_size total_size += weight_size # Add the last block A__ : Union[str, Any] = os.path.join(lowercase_ , weights_name.replace(""".bin""" , f"""-{len(lowercase_ )+1:05d}-of-???.bin""" ) ) rename_and_save_block(lowercase_ , lowercase_ ) sharded_state_dicts.append(current_block.keys() ) # If we only have one shard, we return it if len(lowercase_ ) == 1: return {weights_name: sharded_state_dicts[0]}, None # Otherwise, let's build the index A__ : str = {} A__ : Any = {} for idx, shard in enumerate(lowercase_ ): A__ : Any = weights_name.replace( """.bin""" , f"""-{idx+1:05d}-of-{len(lowercase_ ):05d}.bin""" ) # len(sharded_state_dicts):05d} A__ : Dict = os.path.join(lowercase_ , weights_name.replace(""".bin""" , f"""-{idx+1:05d}-of-???.bin""" ) ) os.rename(lowercase_ , os.path.join(lowercase_ , lowercase_ ) ) A__ : str = shard for key in shard: A__ : Any = shard_file # Add the metadata A__ : Tuple = {"""total_size""": total_size} A__ : Union[str, Any] = {"""metadata""": metadata, """weight_map""": weight_map} with open(os.path.join(lowercase_ , lowercase_ ) , """w""" , encoding="""utf-8""" ) as f: A__ : Dict = json.dumps(lowercase_ , indent=2 , sort_keys=lowercase_ ) + """\n""" f.write(lowercase_ ) return metadata, index if __name__ == "__main__": A_ : Optional[int] = argparse.ArgumentParser() # Required parameters parser.add_argument( '--switch_t5x_checkpoint_path', default='/mnt/disks/disk_switch/original_checkpoints/switch-xxl-128/checkpoint_634600', type=str, required=False, help='Path to a directory containing a folder per layer. Follows the original Google format.', ) parser.add_argument('--max_shard_size', default='10GB', required=False, help='Max shard size') parser.add_argument('--dtype', default='bfloat16', type=str, required=False, help='dtype of the saved model') parser.add_argument( '--pytorch_dump_folder_path', default='/mnt/disks/disk_switch/original_checkpoints/switch-xxl-128-converted', type=str, required=False, help='Path to the output pytorch model.', ) A_ : Dict = parser.parse_args() shard_on_the_fly( args.switch_tax_checkpoint_path, args.pytorch_dump_folder_path, args.max_shard_size, args.dtype, ) def UpperCamelCase () -> int: from transformers import SwitchTransformersConfig, SwitchTransformersForConditionalGeneration, TaTokenizer A__ : str = SwitchTransformersConfig.from_pretrained("""google/switch-base-8""" ) config.save_pretrained("""/home/arthur_huggingface_co/transformers/switch_converted""" ) A__ : str = SwitchTransformersForConditionalGeneration.from_pretrained( """/home/arthur_huggingface_co/transformers/switch_converted""" , device_map="""auto""" ) A__ : Tuple = TaTokenizer.from_pretrained("""t5-small""" ) A__ : Dict = """A <extra_id_0> walks into a bar a orders a <extra_id_1> with <extra_id_2> pinch of <extra_id_3>.""" A__ : Union[str, Any] = tokenizer(lowercase_ , return_tensors="""pt""" ).input_ids A__ : Tuple = model.generate(lowercase_ , decoder_start_token_id=0 ) print(tokenizer.decode(out[0] ) )
64
1
class _a : # Public class to implement a graph '''simple docstring''' def __init__( self , A__ , A__ , A__ ): A__ : Union[str, Any] = row A__ : int = col A__ : Any = graph def __A ( self , A__ , A__ , A__ ): return ( 0 <= i < self.ROW and 0 <= j < self.COL and not visited[i][j] and self.graph[i][j] ) def __A ( self , A__ , A__ , A__ ): # Checking all 8 elements surrounding nth element A__ : Optional[int] = [-1, -1, -1, 0, 0, 1, 1, 1] # Coordinate order A__ : int = [-1, 0, 1, -1, 1, -1, 0, 1] A__ : List[Any] = True # Make those cells visited for k in range(8 ): if self.is_safe(i + row_nbr[k] , j + col_nbr[k] , A__ ): self.diffs(i + row_nbr[k] , j + col_nbr[k] , A__ ) def __A ( self ): # And finally, count all islands. A__ : Union[str, Any] = [[False for j in range(self.COL )] for i in range(self.ROW )] A__ : Tuple = 0 for i in range(self.ROW ): for j in range(self.COL ): if visited[i][j] is False and self.graph[i][j] == 1: self.diffs(A__ , A__ , A__ ) count += 1 return count
64
from typing import TYPE_CHECKING from ...utils import OptionalDependencyNotAvailable, _LazyModule, is_sentencepiece_available A_ : Optional[Any] = {} try: if not is_sentencepiece_available(): raise OptionalDependencyNotAvailable() except OptionalDependencyNotAvailable: pass else: A_ : str = ['BartphoTokenizer'] if TYPE_CHECKING: try: if not is_sentencepiece_available(): raise OptionalDependencyNotAvailable() except OptionalDependencyNotAvailable: pass else: from .tokenization_bartpho import BartphoTokenizer else: import sys A_ : Optional[int] = _LazyModule(__name__, globals()['__file__'], _import_structure, module_spec=__spec__)
64
1
import argparse import torch from transformers import ( SpeechTaConfig, SpeechTaFeatureExtractor, SpeechTaForSpeechToSpeech, SpeechTaForSpeechToText, SpeechTaForTextToSpeech, SpeechTaProcessor, SpeechTaTokenizer, logging, ) from transformers.tokenization_utils import AddedToken logging.set_verbosity_info() A_ : List[str] = logging.get_logger('transformers.models.speecht5') A_ : List[str] = { 'speech_encoder_prenet.layer_norm': 'speecht5.encoder.prenet.feature_projection.layer_norm', 'speech_encoder_prenet.post_extract_proj': 'speecht5.encoder.prenet.feature_projection.projection', 'speech_encoder_prenet.pos_conv.0': 'speecht5.encoder.prenet.pos_conv_embed.conv', 'speech_encoder_prenet.mask_emb': 'speecht5.encoder.prenet.masked_spec_embed', } A_ : int = { 'text_encoder_prenet.encoder_prenet.0': 'speecht5.encoder.prenet.embed_tokens', 'text_encoder_prenet.encoder_prenet.1.alpha': 'speecht5.encoder.prenet.encode_positions.alpha', } A_ : str = { 'speech_decoder_prenet.decoder_prenet.0.0.prenet.0.0': 'speecht5.decoder.prenet.layers.0', 'speech_decoder_prenet.decoder_prenet.0.0.prenet.1.0': 'speecht5.decoder.prenet.layers.1', 'speech_decoder_prenet.decoder_prenet.0.1': 'speecht5.decoder.prenet.final_layer', 'speech_decoder_prenet.decoder_prenet.1.alpha': 'speecht5.decoder.prenet.encode_positions.alpha', 'speech_decoder_prenet.spkembs_layer.0': 'speecht5.decoder.prenet.speaker_embeds_layer', } A_ : Any = { 'speech_decoder_postnet.feat_out': 'speech_decoder_postnet.feat_out', 'speech_decoder_postnet.prob_out': 'speech_decoder_postnet.prob_out', 'speech_decoder_postnet.postnet.postnet.0.0': 'speech_decoder_postnet.layers.0.conv', 'speech_decoder_postnet.postnet.postnet.0.1': 'speech_decoder_postnet.layers.0.batch_norm', 'speech_decoder_postnet.postnet.postnet.1.0': 'speech_decoder_postnet.layers.1.conv', 'speech_decoder_postnet.postnet.postnet.1.1': 'speech_decoder_postnet.layers.1.batch_norm', 'speech_decoder_postnet.postnet.postnet.2.0': 'speech_decoder_postnet.layers.2.conv', 'speech_decoder_postnet.postnet.postnet.2.1': 'speech_decoder_postnet.layers.2.batch_norm', 'speech_decoder_postnet.postnet.postnet.3.0': 'speech_decoder_postnet.layers.3.conv', 'speech_decoder_postnet.postnet.postnet.3.1': 'speech_decoder_postnet.layers.3.batch_norm', 'speech_decoder_postnet.postnet.postnet.4.0': 'speech_decoder_postnet.layers.4.conv', 'speech_decoder_postnet.postnet.postnet.4.1': 'speech_decoder_postnet.layers.4.batch_norm', } A_ : Any = { 'text_decoder_prenet.embed_tokens': 'speecht5.decoder.prenet.embed_tokens', } A_ : int = { 'text_decoder_postnet.output_projection': 'text_decoder_postnet.lm_head', } A_ : Any = { 'encoder.layers.*.self_attn.k_proj': 'speecht5.encoder.wrapped_encoder.layers.*.attention.k_proj', 'encoder.layers.*.self_attn.v_proj': 'speecht5.encoder.wrapped_encoder.layers.*.attention.v_proj', 'encoder.layers.*.self_attn.q_proj': 'speecht5.encoder.wrapped_encoder.layers.*.attention.q_proj', 'encoder.layers.*.self_attn.out_proj': 'speecht5.encoder.wrapped_encoder.layers.*.attention.out_proj', 'encoder.layers.*.self_attn_layer_norm': 'speecht5.encoder.wrapped_encoder.layers.*.layer_norm', 'encoder.layers.*.fc1': 'speecht5.encoder.wrapped_encoder.layers.*.feed_forward.intermediate_dense', 'encoder.layers.*.fc2': 'speecht5.encoder.wrapped_encoder.layers.*.feed_forward.output_dense', 'encoder.layers.*.final_layer_norm': 'speecht5.encoder.wrapped_encoder.layers.*.final_layer_norm', 'encoder.layer_norm': 'speecht5.encoder.wrapped_encoder.layer_norm', 'encoder.pos_emb.pe_k': 'speecht5.encoder.wrapped_encoder.embed_positions.pe_k', } A_ : Optional[Any] = { 'decoder.layers.*.self_attn.k_proj': 'speecht5.decoder.wrapped_decoder.layers.*.self_attn.k_proj', 'decoder.layers.*.self_attn.v_proj': 'speecht5.decoder.wrapped_decoder.layers.*.self_attn.v_proj', 'decoder.layers.*.self_attn.q_proj': 'speecht5.decoder.wrapped_decoder.layers.*.self_attn.q_proj', 'decoder.layers.*.self_attn.out_proj': 'speecht5.decoder.wrapped_decoder.layers.*.self_attn.out_proj', 'decoder.layers.*.self_attn_layer_norm': 'speecht5.decoder.wrapped_decoder.layers.*.self_attn_layer_norm', 'decoder.layers.*.encoder_attn.k_proj': 'speecht5.decoder.wrapped_decoder.layers.*.encoder_attn.k_proj', 'decoder.layers.*.encoder_attn.v_proj': 'speecht5.decoder.wrapped_decoder.layers.*.encoder_attn.v_proj', 'decoder.layers.*.encoder_attn.q_proj': 'speecht5.decoder.wrapped_decoder.layers.*.encoder_attn.q_proj', 'decoder.layers.*.encoder_attn.out_proj': 'speecht5.decoder.wrapped_decoder.layers.*.encoder_attn.out_proj', 'decoder.layers.*.encoder_attn_layer_norm': 'speecht5.decoder.wrapped_decoder.layers.*.encoder_attn_layer_norm', 'decoder.layers.*.fc1': 'speecht5.decoder.wrapped_decoder.layers.*.feed_forward.intermediate_dense', 'decoder.layers.*.fc2': 'speecht5.decoder.wrapped_decoder.layers.*.feed_forward.output_dense', 'decoder.layers.*.final_layer_norm': 'speecht5.decoder.wrapped_decoder.layers.*.final_layer_norm', } A_ : str = { **MAPPING_SPEECH_ENCODER_PRENET, **MAPPING_ENCODER, **MAPPING_DECODER, **MAPPING_TEXT_DECODER_PRENET, **MAPPING_TEXT_DECODER_POSTNET, } A_ : str = { **MAPPING_TEXT_ENCODER_PRENET, **MAPPING_ENCODER, **MAPPING_DECODER, **MAPPING_SPEECH_DECODER_PRENET, **MAPPING_SPEECH_DECODER_POSTNET, } A_ : Union[str, Any] = { **MAPPING_SPEECH_ENCODER_PRENET, **MAPPING_ENCODER, **MAPPING_DECODER, **MAPPING_SPEECH_DECODER_PRENET, **MAPPING_SPEECH_DECODER_POSTNET, } A_ : int = [] A_ : Optional[Any] = [ 'encoder.version', 'encoder.layers.*.norm_k.weight', 'encoder.layers.*.norm_k.bias', 'decoder.version', 'decoder.layers.*.norm_k.weight', 'decoder.layers.*.norm_k.bias', 'decoder.pos_emb.pe_k', 'speech_encoder_prenet.embed_positions._float_tensor', 'text_decoder_prenet.embed_positions._float_tensor', ] A_ : Optional[Any] = IGNORE_KEYS + [ 'encoder.proj', 'text_encoder_prenet.*', 'speech_decoder_prenet.*', 'speech_decoder_postnet.*', ] A_ : Union[str, Any] = IGNORE_KEYS + [ 'encoder.proj', 'speech_encoder_prenet.*', 'text_decoder_prenet.*', 'text_decoder_postnet.*', ] A_ : Optional[int] = IGNORE_KEYS + [ 'encoder.proj', 'text_encoder_prenet.*', 'text_decoder_prenet.*', 'text_decoder_postnet.*', ] def UpperCamelCase (lowercase_: List[str] , lowercase_: str , lowercase_: str , lowercase_: Union[str, Any] , lowercase_: Optional[int] ) -> int: for attribute in key.split(""".""" ): A__ : Dict = getattr(lowercase_ , lowercase_ ) if weight_type is not None: A__ : Tuple = getattr(lowercase_ , lowercase_ ).shape else: A__ : int = hf_pointer.shape if hf_shape != value.shape: raise ValueError( f"""Shape of hf {key + '.' + weight_type if weight_type is not None else ''} is {hf_shape}, but should be""" f""" {value.shape} for {full_name}""" ) if weight_type == "weight": A__ : Any = value elif weight_type == "weight_g": A__ : Union[str, Any] = value elif weight_type == "weight_v": A__ : Optional[Any] = value elif weight_type == "bias": A__ : str = value elif weight_type == "running_mean": A__ : List[Any] = value elif weight_type == "running_var": A__ : str = value elif weight_type == "num_batches_tracked": A__ : List[Any] = value else: A__ : Optional[Any] = value logger.info(f"""{key + ('.' + weight_type if weight_type is not None else '')} was initialized from {full_name}.""" ) def UpperCamelCase (lowercase_: Optional[Any] , lowercase_: Optional[int] ) -> str: for key in ignore_keys: if key.endswith(""".*""" ): if name.startswith(key[:-1] ): return True elif ".*." in key: A__ , A__ : List[Any] = key.split(""".*.""" ) if prefix in name and suffix in name: return True elif key in name: return True return False def UpperCamelCase (lowercase_: Dict , lowercase_: List[str] , lowercase_: Optional[Any] ) -> Optional[int]: A__ : Any = [] if task == "s2t": A__ : Any = hf_model.speechta.encoder.prenet.feature_encoder A__ : Optional[Any] = MAPPING_S2T A__ : List[Any] = IGNORE_KEYS_S2T elif task == "t2s": A__ : Any = None A__ : Optional[int] = MAPPING_T2S A__ : Optional[int] = IGNORE_KEYS_T2S elif task == "s2s": A__ : Tuple = hf_model.speechta.encoder.prenet.feature_encoder A__ : List[str] = MAPPING_S2S A__ : Dict = IGNORE_KEYS_S2S else: raise ValueError(f"""Unsupported task: {task}""" ) for name, value in fairseq_dict.items(): if should_ignore(lowercase_ , lowercase_ ): logger.info(f"""{name} was ignored""" ) continue A__ : Optional[Any] = False if "conv_layers" in name: load_conv_layer( lowercase_ , lowercase_ , lowercase_ , lowercase_ , hf_model.config.feat_extract_norm == """group""" , ) A__ : Any = True else: for key, mapped_key in MAPPING.items(): # mapped_key = "speecht5." + mapped_key if mapped_key not in TOP_LEVEL_KEYS else mapped_key if "*" in key: A__ , A__ : Tuple = key.split(""".*.""" ) if prefix in name and suffix in name: A__ : Optional[Any] = suffix # if key in name or key.split("w2v_model.")[-1] == name.split(".")[0]: if key in name: A__ : Dict = True if "*" in mapped_key: A__ : int = name.split(lowercase_ )[0].split(""".""" )[-2] A__ : List[str] = mapped_key.replace("""*""" , lowercase_ ) if "weight_g" in name: A__ : Optional[Any] = """weight_g""" elif "weight_v" in name: A__ : List[Any] = """weight_v""" elif "bias" in name: A__ : int = """bias""" elif "weight" in name: A__ : Any = """weight""" elif "running_mean" in name: A__ : Optional[Any] = """running_mean""" elif "running_var" in name: A__ : Tuple = """running_var""" elif "num_batches_tracked" in name: A__ : Dict = """num_batches_tracked""" else: A__ : Union[str, Any] = None set_recursively(lowercase_ , lowercase_ , lowercase_ , lowercase_ , lowercase_ ) continue if not is_used: unused_weights.append(lowercase_ ) logger.warning(f"""Unused weights: {unused_weights}""" ) def UpperCamelCase (lowercase_: Union[str, Any] , lowercase_: Any , lowercase_: Optional[Any] , lowercase_: str , lowercase_: Any ) -> List[str]: A__ : int = full_name.split("""conv_layers.""" )[-1] A__ : Optional[int] = name.split(""".""" ) A__ : Any = int(items[0] ) A__ : Dict = int(items[1] ) if type_id == 0: if "bias" in name: if value.shape != feature_extractor.conv_layers[layer_id].conv.bias.data.shape: raise ValueError( f"""{full_name} has size {value.shape}, but""" f""" {feature_extractor.conv_layers[layer_id].conv.bias.data.shape} was found.""" ) A__ : Optional[int] = value logger.info(f"""Feat extract conv layer {layer_id} was initialized from {full_name}.""" ) elif "weight" in name: if value.shape != feature_extractor.conv_layers[layer_id].conv.weight.data.shape: raise ValueError( f"""{full_name} has size {value.shape}, but""" f""" {feature_extractor.conv_layers[layer_id].conv.weight.data.shape} was found.""" ) A__ : List[str] = value logger.info(f"""Feat extract conv layer {layer_id} was initialized from {full_name}.""" ) elif (type_id == 2 and not use_group_norm) or (type_id == 2 and layer_id == 0 and use_group_norm): if "bias" in name: if value.shape != feature_extractor.conv_layers[layer_id].layer_norm.bias.data.shape: raise ValueError( f"""{full_name} has size {value.shape}, but""" f""" {feature_extractor.conv_layers[layer_id].layer_norm.bias.data.shape} was found.""" ) A__ : int = value logger.info(f"""Feat extract layer norm weight of layer {layer_id} was initialized from {full_name}.""" ) elif "weight" in name: if value.shape != feature_extractor.conv_layers[layer_id].layer_norm.weight.data.shape: raise ValueError( f"""{full_name} has size {value.shape}, but""" f""" {feature_extractor.conv_layers[layer_id].layer_norm.weight.data.shape} was found.""" ) A__ : str = value logger.info(f"""Feat extract layer norm weight of layer {layer_id} was initialized from {full_name}.""" ) else: unused_weights.append(lowercase_ ) @torch.no_grad() def UpperCamelCase (lowercase_: Union[str, Any] , lowercase_: int , lowercase_: str , lowercase_: List[Any]=None , lowercase_: Tuple=None , lowercase_: Dict=None , ) -> Union[str, Any]: if config_path is not None: A__ : Tuple = SpeechTaConfig.from_pretrained(lowercase_ ) else: A__ : Tuple = SpeechTaConfig() if task == "s2t": A__ : Optional[int] = config.max_text_positions A__ : Any = SpeechTaForSpeechToText(lowercase_ ) elif task == "t2s": A__ : List[Any] = 1876 A__ : str = 600 A__ : List[Any] = config.max_speech_positions A__ : Tuple = SpeechTaForTextToSpeech(lowercase_ ) elif task == "s2s": A__ : Dict = 1876 A__ : int = config.max_speech_positions A__ : Any = SpeechTaForSpeechToSpeech(lowercase_ ) else: raise ValueError(f"""Unknown task name: {task}""" ) if vocab_path: A__ : Any = SpeechTaTokenizer(lowercase_ , model_max_length=config.max_text_positions ) # Mask token behaves like a normal word, i.e. include the space before it A__ : Optional[Any] = AddedToken("""<mask>""" , lstrip=lowercase_ , rstrip=lowercase_ ) A__ : List[Any] = mask_token tokenizer.add_special_tokens({"""mask_token""": mask_token} ) tokenizer.add_tokens(["""<ctc_blank>"""] ) A__ : Dict = SpeechTaFeatureExtractor() A__ : List[Any] = SpeechTaProcessor(tokenizer=lowercase_ , feature_extractor=lowercase_ ) processor.save_pretrained(lowercase_ ) A__ : List[Any] = torch.load(lowercase_ ) recursively_load_weights(fairseq_checkpoint["""model"""] , lowercase_ , lowercase_ ) model.save_pretrained(lowercase_ ) if repo_id: print("""Pushing to the hub...""" ) processor.push_to_hub(lowercase_ ) model.push_to_hub(lowercase_ ) if __name__ == "__main__": A_ : Union[str, Any] = argparse.ArgumentParser() parser.add_argument( '--task', default='s2t', type=str, help='Type of the SpeechT5 model you\'d like to convert. Should be one of \'s2t\', \'t2s\', \'s2s\'.', ) parser.add_argument('--checkpoint_path', required=True, default=None, type=str, help='Path to fairseq checkpoint') parser.add_argument('--vocab_path', default=None, type=str, help='Path to SentencePiece model') parser.add_argument('--config_path', default=None, type=str, help='Path to hf config.json of model to convert') parser.add_argument( '--pytorch_dump_folder_path', required=True, default=None, type=str, help='Path to the output PyTorch model.' ) parser.add_argument( '--push_to_hub', default=None, type=str, help='Where to upload the converted model on the 🤗 hub.' ) A_ : List[str] = parser.parse_args() convert_speechta_checkpoint( args.task, args.checkpoint_path, args.pytorch_dump_folder_path, args.config_path, args.vocab_path, args.push_to_hub, )
64
import argparse import hashlib import os import urllib import warnings import torch from torch import nn from tqdm import tqdm from transformers import WhisperConfig, WhisperForConditionalGeneration A_ : Dict = { 'tiny.en': 'https://openaipublic.azureedge.net/main/whisper/models/d3dd57d32accea0b295c96e26691aa14d8822fac7d9d27d5dc00b4ca2826dd03/tiny.en.pt', 'tiny': 'https://openaipublic.azureedge.net/main/whisper/models/65147644a518d12f04e32d6f3b26facc3f8dd46e5390956a9424a650c0ce22b9/tiny.pt', 'base.en': 'https://openaipublic.azureedge.net/main/whisper/models/25a8566e1d0c1e2231d1c762132cd20e0f96a85d16145c3a00adf5d1ac670ead/base.en.pt', 'base': 'https://openaipublic.azureedge.net/main/whisper/models/ed3a0b6b1c0edf879ad9b11b1af5a0e6ab5db9205f891f668f8b0e6c6326e34e/base.pt', 'small.en': 'https://openaipublic.azureedge.net/main/whisper/models/f953ad0fd29cacd07d5a9eda5624af0f6bcf2258be67c92b79389873d91e0872/small.en.pt', 'small': 'https://openaipublic.azureedge.net/main/whisper/models/9ecf779972d90ba49c06d968637d720dd632c55bbf19d441fb42bf17a411e794/small.pt', 'medium.en': 'https://openaipublic.azureedge.net/main/whisper/models/d7440d1dc186f76616474e0ff0b3b6b879abc9d1a4926b7adfa41db2d497ab4f/medium.en.pt', 'medium': 'https://openaipublic.azureedge.net/main/whisper/models/345ae4da62f9b3d59415adc60127b97c714f32e89e936602e85993674d08dcb1/medium.pt', 'large': 'https://openaipublic.azureedge.net/main/whisper/models/e4b87e7e0bf463eb8e6956e646f1e277e901512310def2c24bf0e11bd3c28e9a/large.pt', 'large-v2': 'https://openaipublic.azureedge.net/main/whisper/models/81f7c96c852ee8fc832187b0132e569d6c3065a3252ed18e56effd0b6a73e524/large-v2.pt', } def UpperCamelCase (lowercase_: Optional[Any] ) -> Optional[int]: A__ : List[Any] = ["""layers""", """blocks"""] for k in ignore_keys: state_dict.pop(lowercase_ , lowercase_ ) A_ : Any = { 'blocks': 'layers', 'mlp.0': 'fc1', 'mlp.2': 'fc2', 'mlp_ln': 'final_layer_norm', '.attn.query': '.self_attn.q_proj', '.attn.key': '.self_attn.k_proj', '.attn.value': '.self_attn.v_proj', '.attn_ln': '.self_attn_layer_norm', '.attn.out': '.self_attn.out_proj', '.cross_attn.query': '.encoder_attn.q_proj', '.cross_attn.key': '.encoder_attn.k_proj', '.cross_attn.value': '.encoder_attn.v_proj', '.cross_attn_ln': '.encoder_attn_layer_norm', '.cross_attn.out': '.encoder_attn.out_proj', 'decoder.ln.': 'decoder.layer_norm.', 'encoder.ln.': 'encoder.layer_norm.', 'token_embedding': 'embed_tokens', 'encoder.positional_embedding': 'encoder.embed_positions.weight', 'decoder.positional_embedding': 'decoder.embed_positions.weight', 'ln_post': 'layer_norm', } def UpperCamelCase (lowercase_: str ) -> Any: A__ : Dict = list(s_dict.keys() ) for key in keys: A__ : List[str] = key for k, v in WHISPER_MAPPING.items(): if k in key: A__ : List[Any] = new_key.replace(lowercase_ , lowercase_ ) print(f"""{key} -> {new_key}""" ) A__ : Tuple = s_dict.pop(lowercase_ ) return s_dict def UpperCamelCase (lowercase_: Tuple ) -> Optional[int]: A__ , A__ : Any = emb.weight.shape A__ : str = nn.Linear(lowercase_ , lowercase_ , bias=lowercase_ ) A__ : Union[str, Any] = emb.weight.data return lin_layer def UpperCamelCase (lowercase_: str , lowercase_: str ) -> bytes: os.makedirs(lowercase_ , exist_ok=lowercase_ ) A__ : Tuple = os.path.basename(lowercase_ ) A__ : int = url.split("""/""" )[-2] A__ : Dict = os.path.join(lowercase_ , lowercase_ ) if os.path.exists(lowercase_ ) and not os.path.isfile(lowercase_ ): raise RuntimeError(f"""{download_target} exists and is not a regular file""" ) if os.path.isfile(lowercase_ ): A__ : Optional[Any] = open(lowercase_ , """rb""" ).read() if hashlib.shaaaa(lowercase_ ).hexdigest() == expected_shaaaa: return model_bytes else: warnings.warn(f"""{download_target} exists, but the SHA256 checksum does not match; re-downloading the file""" ) with urllib.request.urlopen(lowercase_ ) as source, open(lowercase_ , """wb""" ) as output: with tqdm( total=int(source.info().get("""Content-Length""" ) ) , ncols=80 , unit="""iB""" , unit_scale=lowercase_ , unit_divisor=1024 ) as loop: while True: A__ : Any = source.read(8192 ) if not buffer: break output.write(lowercase_ ) loop.update(len(lowercase_ ) ) A__ : Dict = open(lowercase_ , """rb""" ).read() if hashlib.shaaaa(lowercase_ ).hexdigest() != expected_shaaaa: raise RuntimeError( """Model has been downloaded but the SHA256 checksum does not not match. Please retry loading the model.""" ) return model_bytes def UpperCamelCase (lowercase_: List[Any] , lowercase_: Tuple ) -> Optional[Any]: if ".pt" not in checkpoint_path: A__ : Tuple = _download(_MODELS[checkpoint_path] ) else: A__ : Optional[int] = torch.load(lowercase_ , map_location="""cpu""" ) A__ : str = original_checkpoint["""dims"""] A__ : List[Any] = original_checkpoint["""model_state_dict"""] A__ : Optional[Any] = state_dict["""decoder.token_embedding.weight"""] remove_ignore_keys_(lowercase_ ) rename_keys(lowercase_ ) A__ : List[str] = True A__ : Optional[Any] = state_dict["""decoder.layers.0.fc1.weight"""].shape[0] A__ : List[Any] = WhisperConfig( vocab_size=dimensions["""n_vocab"""] , encoder_ffn_dim=lowercase_ , decoder_ffn_dim=lowercase_ , num_mel_bins=dimensions["""n_mels"""] , d_model=dimensions["""n_audio_state"""] , max_target_positions=dimensions["""n_text_ctx"""] , encoder_layers=dimensions["""n_audio_layer"""] , encoder_attention_heads=dimensions["""n_audio_head"""] , decoder_layers=dimensions["""n_text_layer"""] , decoder_attention_heads=dimensions["""n_text_state"""] , max_source_positions=dimensions["""n_audio_ctx"""] , ) A__ : Optional[Any] = WhisperForConditionalGeneration(lowercase_ ) A__ , A__ : List[Any] = model.model.load_state_dict(lowercase_ , strict=lowercase_ ) if len(lowercase_ ) > 0 and not set(lowercase_ ) <= { "encoder.embed_positions.weights", "decoder.embed_positions.weights", }: raise ValueError( """Only `encoder.embed_positions.weights` and `decoder.embed_positions.weights` are allowed to be missing,""" f""" but all the following weights are missing {missing}""" ) if tie_embeds: A__ : Any = make_linear_from_emb(model.model.decoder.embed_tokens ) else: A__ : str = proj_out_weights model.save_pretrained(lowercase_ ) if __name__ == "__main__": A_ : Any = argparse.ArgumentParser() # # Required parameters parser.add_argument('--checkpoint_path', type=str, help='Patht to the downloaded checkpoints') parser.add_argument('--pytorch_dump_folder_path', default=None, type=str, help='Path to the output PyTorch model.') A_ : Tuple = parser.parse_args() convert_openai_whisper_to_tfms(args.checkpoint_path, args.pytorch_dump_folder_path)
64
1
from __future__ import annotations def UpperCamelCase (lowercase_: float , lowercase_: float , lowercase_: float ) -> float: if days_between_payments <= 0: raise ValueError("""days_between_payments must be > 0""" ) if daily_interest_rate < 0: raise ValueError("""daily_interest_rate must be >= 0""" ) if principal <= 0: raise ValueError("""principal must be > 0""" ) return principal * daily_interest_rate * days_between_payments def UpperCamelCase (lowercase_: float , lowercase_: float , lowercase_: float , ) -> float: if number_of_compounding_periods <= 0: raise ValueError("""number_of_compounding_periods must be > 0""" ) if nominal_annual_interest_rate_percentage < 0: raise ValueError("""nominal_annual_interest_rate_percentage must be >= 0""" ) if principal <= 0: raise ValueError("""principal must be > 0""" ) return principal * ( (1 + nominal_annual_interest_rate_percentage) ** number_of_compounding_periods - 1 ) def UpperCamelCase (lowercase_: float , lowercase_: float , lowercase_: float , ) -> float: if number_of_years <= 0: raise ValueError("""number_of_years must be > 0""" ) if nominal_annual_percentage_rate < 0: raise ValueError("""nominal_annual_percentage_rate must be >= 0""" ) if principal <= 0: raise ValueError("""principal must be > 0""" ) return compound_interest( lowercase_ , nominal_annual_percentage_rate / 365 , number_of_years * 365 ) if __name__ == "__main__": import doctest doctest.testmod()
64
import unittest import numpy as np import torch from transformers import CLIPTextConfig, CLIPTextModel, CLIPTokenizer from diffusers import ( AutoencoderKL, DDIMScheduler, DPMSolverMultistepScheduler, TextToVideoSDPipeline, UNetaDConditionModel, ) from diffusers.utils import is_xformers_available, load_numpy, skip_mps, slow, torch_device from diffusers.utils.testing_utils import enable_full_determinism from ..pipeline_params import TEXT_TO_IMAGE_BATCH_PARAMS, TEXT_TO_IMAGE_PARAMS from ..test_pipelines_common import PipelineTesterMixin enable_full_determinism() @skip_mps class _a (__magic_name__ , unittest.TestCase ): '''simple docstring''' UpperCAmelCase__: Any = TextToVideoSDPipeline UpperCAmelCase__: Any = TEXT_TO_IMAGE_PARAMS UpperCAmelCase__: Optional[Any] = TEXT_TO_IMAGE_BATCH_PARAMS # No `output_type`. UpperCAmelCase__: Optional[int] = frozenset( [ '''num_inference_steps''', '''generator''', '''latents''', '''return_dict''', '''callback''', '''callback_steps''', ] ) def __A ( self ): torch.manual_seed(0 ) A__ : Optional[int] = UNetaDConditionModel( block_out_channels=(32, 64, 64, 64) , layers_per_block=2 , sample_size=32 , in_channels=4 , out_channels=4 , down_block_types=("""CrossAttnDownBlock3D""", """CrossAttnDownBlock3D""", """CrossAttnDownBlock3D""", """DownBlock3D""") , up_block_types=("""UpBlock3D""", """CrossAttnUpBlock3D""", """CrossAttnUpBlock3D""", """CrossAttnUpBlock3D""") , cross_attention_dim=32 , attention_head_dim=4 , ) A__ : Optional[int] = DDIMScheduler( beta_start=0.0_0_0_8_5 , beta_end=0.0_1_2 , beta_schedule="""scaled_linear""" , clip_sample=A__ , set_alpha_to_one=A__ , ) torch.manual_seed(0 ) A__ : Union[str, Any] = AutoencoderKL( block_out_channels=[32, 64] , in_channels=3 , out_channels=3 , down_block_types=["""DownEncoderBlock2D""", """DownEncoderBlock2D"""] , up_block_types=["""UpDecoderBlock2D""", """UpDecoderBlock2D"""] , latent_channels=4 , sample_size=128 , ) torch.manual_seed(0 ) A__ : Union[str, Any] = CLIPTextConfig( bos_token_id=0 , eos_token_id=2 , hidden_size=32 , intermediate_size=37 , layer_norm_eps=1e-05 , num_attention_heads=4 , num_hidden_layers=5 , pad_token_id=1 , vocab_size=1000 , hidden_act="""gelu""" , projection_dim=512 , ) A__ : Union[str, Any] = CLIPTextModel(A__ ) A__ : Tuple = CLIPTokenizer.from_pretrained("""hf-internal-testing/tiny-random-clip""" ) A__ : Dict = { """unet""": unet, """scheduler""": scheduler, """vae""": vae, """text_encoder""": text_encoder, """tokenizer""": tokenizer, } return components def __A ( self , A__ , A__=0 ): if str(A__ ).startswith("""mps""" ): A__ : Tuple = torch.manual_seed(A__ ) else: A__ : List[str] = torch.Generator(device=A__ ).manual_seed(A__ ) A__ : List[str] = { """prompt""": """A painting of a squirrel eating a burger""", """generator""": generator, """num_inference_steps""": 2, """guidance_scale""": 6.0, """output_type""": """pt""", } return inputs def __A ( self ): A__ : List[str] = """cpu""" # ensure determinism for the device-dependent torch.Generator A__ : Union[str, Any] = self.get_dummy_components() A__ : Union[str, Any] = TextToVideoSDPipeline(**A__ ) A__ : int = sd_pipe.to(A__ ) sd_pipe.set_progress_bar_config(disable=A__ ) A__ : int = self.get_dummy_inputs(A__ ) A__ : int = """np""" A__ : Any = sd_pipe(**A__ ).frames A__ : Dict = frames[0][-3:, -3:, -1] assert frames[0].shape == (64, 64, 3) A__ : Optional[Any] = np.array([1_5_8.0, 1_6_0.0, 1_5_3.0, 1_2_5.0, 1_0_0.0, 1_2_1.0, 1_1_1.0, 9_3.0, 1_1_3.0] ) assert np.abs(image_slice.flatten() - expected_slice ).max() < 1e-2 def __A ( self ): self._test_attention_slicing_forward_pass(test_mean_pixel_difference=A__ , expected_max_diff=3e-3 ) @unittest.skipIf( torch_device != """cuda""" or not is_xformers_available() , reason="""XFormers attention is only available with CUDA and `xformers` installed""" , ) def __A ( self ): self._test_xformers_attention_forwardGenerator_pass(test_mean_pixel_difference=A__ , expected_max_diff=1e-2 ) @unittest.skip(reason="""Batching needs to be properly figured out first for this pipeline.""" ) def __A ( self ): pass @unittest.skip(reason="""Batching needs to be properly figured out first for this pipeline.""" ) def __A ( self ): pass @unittest.skip(reason="""`num_images_per_prompt` argument is not supported for this pipeline.""" ) def __A ( self ): pass def __A ( self ): return super().test_progress_bar() @slow @skip_mps class _a (unittest.TestCase ): '''simple docstring''' def __A ( self ): A__ : Union[str, Any] = load_numpy( """https://huggingface.co/datasets/hf-internal-testing/diffusers-images/resolve/main/text_to_video/video.npy""" ) A__ : Tuple = TextToVideoSDPipeline.from_pretrained("""damo-vilab/text-to-video-ms-1.7b""" ) A__ : Any = DPMSolverMultistepScheduler.from_config(pipe.scheduler.config ) A__ : int = pipe.to("""cuda""" ) A__ : Optional[Any] = """Spiderman is surfing""" A__ : List[str] = torch.Generator(device="""cpu""" ).manual_seed(0 ) A__ : Optional[Any] = pipe(A__ , generator=A__ , num_inference_steps=25 , output_type="""pt""" ).frames A__ : Dict = video_frames.cpu().numpy() assert np.abs(expected_video - video ).mean() < 5e-2 def __A ( self ): A__ : List[Any] = load_numpy( """https://huggingface.co/datasets/hf-internal-testing/diffusers-images/resolve/main/text_to_video/video_2step.npy""" ) A__ : Optional[int] = TextToVideoSDPipeline.from_pretrained("""damo-vilab/text-to-video-ms-1.7b""" ) A__ : List[str] = pipe.to("""cuda""" ) A__ : Dict = """Spiderman is surfing""" A__ : Union[str, Any] = torch.Generator(device="""cpu""" ).manual_seed(0 ) A__ : Optional[int] = pipe(A__ , generator=A__ , num_inference_steps=2 , output_type="""pt""" ).frames A__ : Optional[int] = video_frames.cpu().numpy() assert np.abs(expected_video - video ).mean() < 5e-2
64
1
from typing import TYPE_CHECKING from ...utils import OptionalDependencyNotAvailable, _LazyModule, is_tokenizers_available, is_torch_available A_ : Union[str, Any] = { 'configuration_graphormer': ['GRAPHORMER_PRETRAINED_CONFIG_ARCHIVE_MAP', 'GraphormerConfig'], } try: if not is_torch_available(): raise OptionalDependencyNotAvailable() except OptionalDependencyNotAvailable: pass else: A_ : Dict = [ 'GRAPHORMER_PRETRAINED_MODEL_ARCHIVE_LIST', 'GraphormerForGraphClassification', 'GraphormerModel', 'GraphormerPreTrainedModel', ] if TYPE_CHECKING: from .configuration_graphormer import GRAPHORMER_PRETRAINED_CONFIG_ARCHIVE_MAP, GraphormerConfig try: if not is_torch_available(): raise OptionalDependencyNotAvailable() except OptionalDependencyNotAvailable: pass else: from .modeling_graphormer import ( GRAPHORMER_PRETRAINED_MODEL_ARCHIVE_LIST, GraphormerForGraphClassification, GraphormerModel, GraphormerPreTrainedModel, ) else: import sys A_ : Optional[Any] = _LazyModule(__name__, globals()['__file__'], _import_structure, module_spec=__spec__)
64
def UpperCamelCase (lowercase_: int ) -> int: if not isinstance(lowercase_ , lowercase_ ): raise TypeError("""Input value must be an 'int' type""" ) A__ : int = 0 while number: position += 1 number >>= 1 return position if __name__ == "__main__": import doctest doctest.testmod()
64
1
import argparse import torch from transformers import FunnelBaseModel, FunnelConfig, FunnelModel, load_tf_weights_in_funnel from transformers.utils import logging logging.set_verbosity_info() def UpperCamelCase (lowercase_: List[str] , lowercase_: Union[str, Any] , lowercase_: Optional[Any] , lowercase_: int ) -> Optional[Any]: # Initialise PyTorch model A__ : Dict = FunnelConfig.from_json_file(lowercase_ ) print(f"""Building PyTorch model from configuration: {config}""" ) A__ : Optional[Any] = FunnelBaseModel(lowercase_ ) if base_model else FunnelModel(lowercase_ ) # Load weights from tf checkpoint load_tf_weights_in_funnel(lowercase_ , lowercase_ , lowercase_ ) # Save pytorch-model print(f"""Save PyTorch model to {pytorch_dump_path}""" ) torch.save(model.state_dict() , lowercase_ ) if __name__ == "__main__": A_ : Optional[Any] = argparse.ArgumentParser() # Required parameters parser.add_argument( '--tf_checkpoint_path', default=None, type=str, required=True, help='Path to the TensorFlow checkpoint path.' ) parser.add_argument( '--config_file', default=None, type=str, required=True, help='The config json file corresponding to the pre-trained model. \nThis specifies the model architecture.', ) parser.add_argument( '--pytorch_dump_path', default=None, type=str, required=True, help='Path to the output PyTorch model.' ) parser.add_argument( '--base_model', action='store_true', help='Whether you want just the base model (no decoder) or not.' ) A_ : Optional[Any] = parser.parse_args() convert_tf_checkpoint_to_pytorch( args.tf_checkpoint_path, args.config_file, args.pytorch_dump_path, args.base_model )
64
from pathlib import Path import cva import numpy as np from matplotlib import pyplot as plt def UpperCamelCase (lowercase_: np.ndarray , lowercase_: np.ndarray , lowercase_: np.ndarray , lowercase_: int , lowercase_: int ) -> np.ndarray: A__ : Any = cva.getAffineTransform(lowercase_ , lowercase_ ) return cva.warpAffine(lowercase_ , lowercase_ , (rows, cols) ) if __name__ == "__main__": # read original image A_ : List[Any] = cva.imread( str(Path(__file__).resolve().parent.parent / 'image_data' / 'lena.jpg') ) # turn image in gray scale value A_ : List[Any] = cva.cvtColor(image, cva.COLOR_BGR2GRAY) # get image shape A_ , A_ : Optional[Any] = gray_img.shape # set different points to rotate image A_ : str = np.array([[50, 50], [200, 50], [50, 200]], np.floataa) A_ : Dict = np.array([[10, 100], [200, 50], [100, 250]], np.floataa) A_ : Optional[int] = np.array([[50, 50], [150, 50], [120, 200]], np.floataa) A_ : Optional[int] = np.array([[10, 100], [80, 50], [180, 250]], np.floataa) # add all rotated images in a list A_ : Dict = [ gray_img, get_rotation(gray_img, ptsa, ptsa, img_rows, img_cols), get_rotation(gray_img, ptsa, ptsa, img_rows, img_cols), get_rotation(gray_img, ptsa, ptsa, img_rows, img_cols), ] # plot different image rotations A_ : Union[str, Any] = plt.figure(1) A_ : Union[str, Any] = ['Original', 'Rotation 1', 'Rotation 2', 'Rotation 3'] for i, image in enumerate(images): plt.subplot(2, 2, i + 1), plt.imshow(image, 'gray') plt.title(titles[i]) plt.axis('off') plt.subplots_adjust(left=0.0, bottom=0.05, right=1.0, top=0.95) plt.show()
64
1
# Copyright 2022 The HuggingFace Team. 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 copy of the License at # # http://www.apache.org/licenses/LICENSE-2.0 # # Unless required by applicable law or agreed to in writing, software # distributed under the License is distributed on an "AS IS" BASIS, # WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. # See the License for the specific language governing permissions and # limitations under the License. import argparse import os import subprocess from packaging.version import Version, parse from accelerate.commands.config.config_args import default_config_file, load_config_from_file A_ : List[Any] = 'Run commands across TPU VMs for initial setup before running `accelerate launch`.' def UpperCamelCase (lowercase_: int=None ) -> Any: if subparsers is not None: A__ : int = subparsers.add_parser("""tpu-config""" , description=_description ) else: A__ : Any = argparse.ArgumentParser("""Accelerate tpu-config command""" , description=_description ) # Core arguments A__ : Union[str, Any] = parser.add_argument_group( """Config Arguments""" , """Arguments that can be configured through `accelerate config`.""" ) config_args.add_argument( """--config_file""" , type=lowercase_ , default=lowercase_ , help="""Path to the config file to use for accelerate.""" , ) config_args.add_argument( """--tpu_name""" , default=lowercase_ , help="""The name of the TPU to use. If not specified, will use the TPU specified in the config file.""" , ) config_args.add_argument( """--tpu_zone""" , default=lowercase_ , help="""The zone of the TPU to use. If not specified, will use the zone specified in the config file.""" , ) A__ : int = parser.add_argument_group("""TPU Arguments""" , """Arguments for options ran inside the TPU.""" ) pod_args.add_argument( """--use_alpha""" , action="""store_true""" , help="""Whether to use `gcloud alpha` when running the TPU training script instead of `gcloud`.""" , ) pod_args.add_argument( """--command_file""" , default=lowercase_ , help="""The path to the file containing the commands to run on the pod on startup.""" , ) pod_args.add_argument( """--command""" , action="""append""" , nargs="""+""" , help="""A command to run on the pod. Can be passed multiple times.""" , ) pod_args.add_argument( """--install_accelerate""" , action="""store_true""" , help="""Whether to install accelerate on the pod. Defaults to False.""" , ) pod_args.add_argument( """--accelerate_version""" , default="""latest""" , help="""The version of accelerate to install on the pod. If not specified, will use the latest pypi version. Specify 'dev' to install from GitHub.""" , ) pod_args.add_argument( """--debug""" , action="""store_true""" , help="""If set, will print the command that would be run instead of running it.""" ) if subparsers is not None: parser.set_defaults(func=lowercase_ ) return parser def UpperCamelCase (lowercase_: int ) -> int: A__ : List[str] = None # Get the default from the config file if it exists. if args.config_file is not None or os.path.isfile(lowercase_ ): A__ : str = load_config_from_file(args.config_file ) if not args.command_file and defaults.command_file is not None and not args.command: A__ : List[str] = defaults.command_file if not args.command and defaults.commands is not None: A__ : Tuple = defaults.commands if not args.tpu_name: A__ : int = defaults.tpu_name if not args.tpu_zone: A__ : Optional[int] = defaults.tpu_zone if args.accelerate_version == "dev": A__ : List[Any] = """git+https://github.com/huggingface/accelerate.git""" elif args.accelerate_version == "latest": A__ : int = """accelerate -U""" elif isinstance(parse(args.accelerate_version ) , lowercase_ ): A__ : int = f"""accelerate=={args.accelerate_version}""" if not args.command_file and not args.command: raise ValueError("""You must specify either a command file or a command to run on the pod.""" ) if args.command_file: with open(args.command_file , """r""" ) as f: A__ : List[str] = [f.read().splitlines()] # To turn list of lists into list of strings if isinstance(args.command[0] , lowercase_ ): A__ : List[str] = [line for cmd in args.command for line in cmd] # Default to the shared folder and install accelerate A__ : List[Any] = ["""cd /usr/share"""] if args.install_accelerate: new_cmd += [f"""pip install {args.accelerate_version}"""] new_cmd += args.command A__ : str = """; """.join(lowercase_ ) # Then send it to gcloud # Eventually try to use google-api-core to do this instead of subprocess A__ : Union[str, Any] = ["""gcloud"""] if args.use_alpha: cmd += ["alpha"] cmd += [ "compute", "tpus", "tpu-vm", "ssh", args.tpu_name, "--zone", args.tpu_zone, "--command", args.command, "--worker", "all", ] if args.debug: print(f"""Running {' '.join(lowercase_ )}""" ) return subprocess.run(lowercase_ ) print("""Successfully setup pod.""" ) def UpperCamelCase () -> Optional[Any]: A__ : str = tpu_command_parser() A__ : Dict = parser.parse_args() tpu_command_launcher(lowercase_ )
64
import os import tempfile import unittest from pathlib import Path from transformers import AutoConfig, is_torch_available from transformers.testing_utils import require_torch, torch_device if is_torch_available(): from transformers import PyTorchBenchmark, PyTorchBenchmarkArguments @require_torch class _a (unittest.TestCase ): '''simple docstring''' def __A ( self , A__ ): for model_result in results.values(): for batch_size, sequence_length in zip(model_result["""bs"""] , model_result["""ss"""] ): A__ : str = model_result["""result"""][batch_size][sequence_length] self.assertIsNotNone(A__ ) def __A ( self ): A__ : Dict = """sshleifer/tiny-gpt2""" A__ : Tuple = PyTorchBenchmarkArguments( models=[MODEL_ID] , training=A__ , inference=A__ , sequence_lengths=[8] , batch_sizes=[1] , multi_process=A__ , ) A__ : int = PyTorchBenchmark(A__ ) A__ : List[Any] = benchmark.run() self.check_results_dict_not_empty(results.time_inference_result ) self.check_results_dict_not_empty(results.memory_inference_result ) def __A ( self ): A__ : Dict = """sgugger/tiny-distilbert-classification""" A__ : Dict = PyTorchBenchmarkArguments( models=[MODEL_ID] , training=A__ , inference=A__ , sequence_lengths=[8] , batch_sizes=[1] , multi_process=A__ , only_pretrain_model=A__ , ) A__ : str = PyTorchBenchmark(A__ ) A__ : List[str] = benchmark.run() self.check_results_dict_not_empty(results.time_inference_result ) self.check_results_dict_not_empty(results.memory_inference_result ) def __A ( self ): A__ : Any = """sshleifer/tiny-gpt2""" A__ : List[Any] = PyTorchBenchmarkArguments( models=[MODEL_ID] , training=A__ , inference=A__ , torchscript=A__ , sequence_lengths=[8] , batch_sizes=[1] , multi_process=A__ , ) A__ : Tuple = PyTorchBenchmark(A__ ) A__ : str = benchmark.run() self.check_results_dict_not_empty(results.time_inference_result ) self.check_results_dict_not_empty(results.memory_inference_result ) @unittest.skipIf(torch_device == """cpu""" , """Cant do half precision""" ) def __A ( self ): A__ : Optional[Any] = """sshleifer/tiny-gpt2""" A__ : Optional[Any] = PyTorchBenchmarkArguments( models=[MODEL_ID] , training=A__ , inference=A__ , fpaa=A__ , sequence_lengths=[8] , batch_sizes=[1] , multi_process=A__ , ) A__ : str = PyTorchBenchmark(A__ ) A__ : Any = benchmark.run() self.check_results_dict_not_empty(results.time_inference_result ) self.check_results_dict_not_empty(results.memory_inference_result ) def __A ( self ): A__ : Optional[Any] = """sshleifer/tiny-gpt2""" A__ : Tuple = AutoConfig.from_pretrained(A__ ) # set architectures equal to `None` A__ : List[Any] = None A__ : str = PyTorchBenchmarkArguments( models=[MODEL_ID] , training=A__ , inference=A__ , sequence_lengths=[8] , batch_sizes=[1] , multi_process=A__ , ) A__ : List[str] = PyTorchBenchmark(A__ , configs=[config] ) A__ : Any = benchmark.run() self.check_results_dict_not_empty(results.time_inference_result ) self.check_results_dict_not_empty(results.memory_inference_result ) def __A ( self ): A__ : Optional[int] = """sshleifer/tiny-gpt2""" A__ : Optional[int] = PyTorchBenchmarkArguments( models=[MODEL_ID] , training=A__ , inference=A__ , sequence_lengths=[8] , batch_sizes=[1] , multi_process=A__ , ) A__ : Any = PyTorchBenchmark(A__ ) A__ : Dict = benchmark.run() self.check_results_dict_not_empty(results.time_train_result ) self.check_results_dict_not_empty(results.memory_train_result ) @unittest.skipIf(torch_device == """cpu""" , """Can't do half precision""" ) def __A ( self ): A__ : Optional[int] = """sshleifer/tiny-gpt2""" A__ : List[str] = PyTorchBenchmarkArguments( models=[MODEL_ID] , training=A__ , inference=A__ , sequence_lengths=[8] , batch_sizes=[1] , fpaa=A__ , multi_process=A__ , ) A__ : Dict = PyTorchBenchmark(A__ ) A__ : Optional[int] = benchmark.run() self.check_results_dict_not_empty(results.time_train_result ) self.check_results_dict_not_empty(results.memory_train_result ) def __A ( self ): A__ : int = """sshleifer/tiny-gpt2""" A__ : Optional[int] = AutoConfig.from_pretrained(A__ ) A__ : str = PyTorchBenchmarkArguments( models=[MODEL_ID] , training=A__ , inference=A__ , sequence_lengths=[8] , batch_sizes=[1] , multi_process=A__ , ) A__ : int = PyTorchBenchmark(A__ , configs=[config] ) A__ : Tuple = benchmark.run() self.check_results_dict_not_empty(results.time_inference_result ) self.check_results_dict_not_empty(results.memory_inference_result ) def __A ( self ): A__ : List[str] = """sshleifer/tinier_bart""" A__ : List[str] = AutoConfig.from_pretrained(A__ ) A__ : List[str] = PyTorchBenchmarkArguments( models=[MODEL_ID] , training=A__ , inference=A__ , sequence_lengths=[8] , batch_sizes=[1] , multi_process=A__ , ) A__ : Union[str, Any] = PyTorchBenchmark(A__ , configs=[config] ) A__ : str = benchmark.run() self.check_results_dict_not_empty(results.time_inference_result ) self.check_results_dict_not_empty(results.memory_inference_result ) def __A ( self ): A__ : Optional[int] = """sshleifer/tiny-gpt2""" A__ : Union[str, Any] = AutoConfig.from_pretrained(A__ ) A__ : Tuple = PyTorchBenchmarkArguments( models=[MODEL_ID] , training=A__ , inference=A__ , sequence_lengths=[8] , batch_sizes=[1] , multi_process=A__ , ) A__ : int = PyTorchBenchmark(A__ , configs=[config] ) A__ : Tuple = benchmark.run() self.check_results_dict_not_empty(results.time_train_result ) self.check_results_dict_not_empty(results.memory_train_result ) def __A ( self ): A__ : Dict = """sshleifer/tinier_bart""" A__ : int = AutoConfig.from_pretrained(A__ ) A__ : Union[str, Any] = PyTorchBenchmarkArguments( models=[MODEL_ID] , training=A__ , inference=A__ , sequence_lengths=[8] , batch_sizes=[1] , multi_process=A__ , ) A__ : List[Any] = PyTorchBenchmark(A__ , configs=[config] ) A__ : Optional[Any] = benchmark.run() self.check_results_dict_not_empty(results.time_train_result ) self.check_results_dict_not_empty(results.memory_train_result ) def __A ( self ): A__ : int = """sshleifer/tiny-gpt2""" with tempfile.TemporaryDirectory() as tmp_dir: A__ : Dict = PyTorchBenchmarkArguments( models=[MODEL_ID] , training=A__ , inference=A__ , save_to_csv=A__ , sequence_lengths=[8] , batch_sizes=[1] , inference_time_csv_file=os.path.join(A__ , """inf_time.csv""" ) , train_memory_csv_file=os.path.join(A__ , """train_mem.csv""" ) , inference_memory_csv_file=os.path.join(A__ , """inf_mem.csv""" ) , train_time_csv_file=os.path.join(A__ , """train_time.csv""" ) , env_info_csv_file=os.path.join(A__ , """env.csv""" ) , multi_process=A__ , ) A__ : Optional[Any] = PyTorchBenchmark(A__ ) benchmark.run() self.assertTrue(Path(os.path.join(A__ , """inf_time.csv""" ) ).exists() ) self.assertTrue(Path(os.path.join(A__ , """train_time.csv""" ) ).exists() ) self.assertTrue(Path(os.path.join(A__ , """inf_mem.csv""" ) ).exists() ) self.assertTrue(Path(os.path.join(A__ , """train_mem.csv""" ) ).exists() ) self.assertTrue(Path(os.path.join(A__ , """env.csv""" ) ).exists() ) def __A ( self ): A__ : Optional[int] = """sshleifer/tiny-gpt2""" def _check_summary_is_not_empty(A__ ): self.assertTrue(hasattr(A__ , """sequential""" ) ) self.assertTrue(hasattr(A__ , """cumulative""" ) ) self.assertTrue(hasattr(A__ , """current""" ) ) self.assertTrue(hasattr(A__ , """total""" ) ) with tempfile.TemporaryDirectory() as tmp_dir: A__ : Dict = PyTorchBenchmarkArguments( models=[MODEL_ID] , training=A__ , inference=A__ , sequence_lengths=[8] , batch_sizes=[1] , log_filename=os.path.join(A__ , """log.txt""" ) , log_print=A__ , trace_memory_line_by_line=A__ , multi_process=A__ , ) A__ : Dict = PyTorchBenchmark(A__ ) A__ : str = benchmark.run() _check_summary_is_not_empty(result.inference_summary ) _check_summary_is_not_empty(result.train_summary ) self.assertTrue(Path(os.path.join(A__ , """log.txt""" ) ).exists() )
64
1
from typing import List, Optional, Union from ...image_utils import ImageInput from ...processing_utils import ProcessorMixin from ...tokenization_utils_base import BatchEncoding, PaddingStrategy, PreTokenizedInput, TextInput, TruncationStrategy from ...utils import TensorType class _a (__magic_name__ ): '''simple docstring''' UpperCAmelCase__: Dict = ['''image_processor''', '''tokenizer'''] UpperCAmelCase__: Optional[Any] = '''BlipImageProcessor''' UpperCAmelCase__: str = '''AutoTokenizer''' def __init__( self , A__ , A__ ): A__ : List[Any] = False super().__init__(A__ , A__ ) A__ : Optional[Any] = self.image_processor def __call__( self , A__ = None , A__ = None , A__ = True , A__ = False , A__ = None , A__ = None , A__ = 0 , A__ = None , A__ = None , A__ = False , A__ = False , A__ = False , A__ = False , A__ = False , A__ = True , A__ = None , **A__ , ): if images is None and text is None: raise ValueError("""You have to specify either images or text.""" ) # Get only text if images is None: A__ : Optional[int] = self.tokenizer A__ : int = self.tokenizer( text=A__ , add_special_tokens=A__ , padding=A__ , truncation=A__ , max_length=A__ , stride=A__ , pad_to_multiple_of=A__ , return_attention_mask=A__ , return_overflowing_tokens=A__ , return_special_tokens_mask=A__ , return_offsets_mapping=A__ , return_token_type_ids=A__ , return_length=A__ , verbose=A__ , return_tensors=A__ , **A__ , ) return text_encoding # add pixel_values A__ : Union[str, Any] = self.image_processor(A__ , return_tensors=A__ ) if text is not None: A__ : Tuple = self.tokenizer( text=A__ , add_special_tokens=A__ , padding=A__ , truncation=A__ , max_length=A__ , stride=A__ , pad_to_multiple_of=A__ , return_attention_mask=A__ , return_overflowing_tokens=A__ , return_special_tokens_mask=A__ , return_offsets_mapping=A__ , return_token_type_ids=A__ , return_length=A__ , verbose=A__ , return_tensors=A__ , **A__ , ) else: A__ : List[str] = None if text_encoding is not None: encoding_image_processor.update(A__ ) return encoding_image_processor def __A ( self , *A__ , **A__ ): return self.tokenizer.batch_decode(*A__ , **A__ ) def __A ( self , *A__ , **A__ ): return self.tokenizer.decode(*A__ , **A__ ) @property # Copied from transformers.models.blip.processing_blip.BlipProcessor.model_input_names def __A ( self ): A__ : str = self.tokenizer.model_input_names A__ : Union[str, Any] = self.image_processor.model_input_names return list(dict.fromkeys(tokenizer_input_names + image_processor_input_names ) )
64
# tests directory-specific settings - this file is run automatically # by pytest before any tests are run import doctest import sys import warnings from os.path import abspath, dirname, join import _pytest from transformers.testing_utils import HfDoctestModule, HfDocTestParser # allow having multiple repository checkouts and not needing to remember to rerun # 'pip install -e .[dev]' when switching between checkouts and running tests. A_ : Optional[int] = abspath(join(dirname(__file__), 'src')) sys.path.insert(1, git_repo_path) # silence FutureWarning warnings in tests since often we can't act on them until # they become normal warnings - i.e. the tests still need to test the current functionality warnings.simplefilter(action='ignore', category=FutureWarning) def UpperCamelCase (lowercase_: List[str] ) -> Any: config.addinivalue_line( """markers""" , """is_pt_tf_cross_test: mark test to run only when PT and TF interactions are tested""" ) config.addinivalue_line( """markers""" , """is_pt_flax_cross_test: mark test to run only when PT and FLAX interactions are tested""" ) config.addinivalue_line("""markers""" , """is_pipeline_test: mark test to run only when pipelines are tested""" ) config.addinivalue_line("""markers""" , """is_staging_test: mark test to run only in the staging environment""" ) config.addinivalue_line("""markers""" , """accelerate_tests: mark test that require accelerate""" ) config.addinivalue_line("""markers""" , """tool_tests: mark the tool tests that are run on their specific schedule""" ) def UpperCamelCase (lowercase_: Optional[int] ) -> Optional[Any]: from transformers.testing_utils import pytest_addoption_shared pytest_addoption_shared(lowercase_ ) def UpperCamelCase (lowercase_: List[str] ) -> Optional[Any]: from transformers.testing_utils import pytest_terminal_summary_main A__ : List[Any] = terminalreporter.config.getoption("""--make-reports""" ) if make_reports: pytest_terminal_summary_main(lowercase_ , id=lowercase_ ) def UpperCamelCase (lowercase_: Union[str, Any] , lowercase_: int ) -> List[str]: # If no tests are collected, pytest exists with code 5, which makes the CI fail. if exitstatus == 5: A__ : Tuple = 0 # Doctest custom flag to ignore output. A_ : Tuple = doctest.register_optionflag('IGNORE_RESULT') A_ : Dict = doctest.OutputChecker class _a (__magic_name__ ): '''simple docstring''' def __A ( self , A__ , A__ , A__ ): if IGNORE_RESULT & optionflags: return True return OutputChecker.check_output(self , A__ , A__ , A__ ) A_ : str = CustomOutputChecker A_ : Dict = HfDoctestModule A_ : Optional[int] = HfDocTestParser
64
1
from itertools import permutations def UpperCamelCase (lowercase_: tuple ) -> bool: if num[3] % 2 != 0: return False if (num[2] + num[3] + num[4]) % 3 != 0: return False if num[5] % 5 != 0: return False A__ : Tuple = [7, 11, 13, 17] for i, test in enumerate(lowercase_ ): if (num[i + 4] * 100 + num[i + 5] * 10 + num[i + 6]) % test != 0: return False return True def UpperCamelCase (lowercase_: int = 10 ) -> int: return sum( int("""""".join(map(lowercase_ , lowercase_ ) ) ) for num in permutations(range(lowercase_ ) ) if is_substring_divisible(lowercase_ ) ) if __name__ == "__main__": print(f'''{solution() = }''')
64
from __future__ import annotations import unittest from transformers import AutoTokenizer, PegasusConfig, is_tf_available from transformers.testing_utils import require_sentencepiece, require_tf, require_tokenizers, slow from transformers.utils import cached_property from ...test_configuration_common import ConfigTester from ...test_modeling_tf_common import TFModelTesterMixin, ids_tensor from ...test_pipeline_mixin import PipelineTesterMixin if is_tf_available(): import tensorflow as tf from transformers import TFAutoModelForSeqaSeqLM, TFPegasusForConditionalGeneration, TFPegasusModel @require_tf class _a : '''simple docstring''' UpperCAmelCase__: List[Any] = PegasusConfig UpperCAmelCase__: Optional[int] = {} UpperCAmelCase__: List[str] = '''gelu''' def __init__( self , A__ , A__=13 , A__=7 , A__=True , A__=False , A__=99 , A__=32 , A__=2 , A__=4 , A__=37 , A__=0.1 , A__=0.1 , A__=40 , A__=2 , A__=1 , A__=0 , ): A__ : Dict = parent A__ : Dict = batch_size A__ : Any = seq_length A__ : Optional[Any] = is_training A__ : int = use_labels A__ : Any = vocab_size A__ : Union[str, Any] = hidden_size A__ : Tuple = num_hidden_layers A__ : Tuple = num_attention_heads A__ : List[Any] = intermediate_size A__ : Union[str, Any] = hidden_dropout_prob A__ : Optional[Any] = attention_probs_dropout_prob A__ : List[Any] = max_position_embeddings A__ : Any = eos_token_id A__ : List[Any] = pad_token_id A__ : List[Any] = bos_token_id def __A ( self ): A__ : str = ids_tensor([self.batch_size, self.seq_length - 1] , self.vocab_size ) A__ : Dict = tf.expand_dims(tf.constant([self.eos_token_id] * self.batch_size ) , 1 ) A__ : List[Any] = tf.concat([input_ids, eos_tensor] , axis=1 ) A__ : Union[str, Any] = ids_tensor([self.batch_size, self.seq_length] , self.vocab_size ) A__ : Tuple = self.config_cls( vocab_size=self.vocab_size , d_model=self.hidden_size , encoder_layers=self.num_hidden_layers , decoder_layers=self.num_hidden_layers , encoder_attention_heads=self.num_attention_heads , decoder_attention_heads=self.num_attention_heads , encoder_ffn_dim=self.intermediate_size , decoder_ffn_dim=self.intermediate_size , dropout=self.hidden_dropout_prob , attention_dropout=self.attention_probs_dropout_prob , max_position_embeddings=self.max_position_embeddings , eos_token_ids=[2] , bos_token_id=self.bos_token_id , pad_token_id=self.pad_token_id , decoder_start_token_id=self.pad_token_id , **self.config_updates , ) A__ : str = prepare_pegasus_inputs_dict(A__ , A__ , A__ ) return config, inputs_dict def __A ( self , A__ , A__ ): A__ : int = TFPegasusModel(config=A__ ).get_decoder() A__ : List[Any] = inputs_dict["""input_ids"""] A__ : Any = input_ids[:1, :] A__ : Optional[Any] = inputs_dict["""attention_mask"""][:1, :] A__ : Optional[int] = inputs_dict["""head_mask"""] A__ : Any = 1 # first forward pass A__ : Tuple = model(A__ , attention_mask=A__ , head_mask=A__ , use_cache=A__ ) A__ , A__ : Dict = outputs.to_tuple() # create hypothetical next token and extent to next_input_ids A__ : Tuple = ids_tensor((self.batch_size, 3) , config.vocab_size ) A__ : Optional[Any] = tf.cast(ids_tensor((self.batch_size, 3) , 2 ) , tf.inta ) # append to next input_ids and A__ : List[Any] = tf.concat([input_ids, next_tokens] , axis=-1 ) A__ : Tuple = tf.concat([attention_mask, next_attn_mask] , axis=-1 ) A__ : Optional[Any] = model(A__ , attention_mask=A__ )[0] A__ : Any = model(A__ , attention_mask=A__ , past_key_values=A__ )[0] self.parent.assertEqual(next_tokens.shape[1] , output_from_past.shape[1] ) # select random slice A__ : int = int(ids_tensor((1,) , output_from_past.shape[-1] ) ) A__ : Any = output_from_no_past[:, -3:, random_slice_idx] A__ : Tuple = output_from_past[:, :, random_slice_idx] # test that outputs are equal for slice tf.debugging.assert_near(A__ , A__ , rtol=1e-3 ) def UpperCamelCase (lowercase_: Union[str, Any] , lowercase_: Dict , lowercase_: List[Any] , lowercase_: Dict=None , lowercase_: int=None , lowercase_: List[Any]=None , lowercase_: List[Any]=None , lowercase_: str=None , ) -> int: if attention_mask is None: A__ : List[str] = tf.cast(tf.math.not_equal(lowercase_ , config.pad_token_id ) , tf.inta ) if decoder_attention_mask is None: A__ : Dict = tf.concat( [ tf.ones(decoder_input_ids[:, :1].shape , dtype=tf.inta ), tf.cast(tf.math.not_equal(decoder_input_ids[:, 1:] , config.pad_token_id ) , tf.inta ), ] , axis=-1 , ) if head_mask is None: A__ : Any = tf.ones((config.encoder_layers, config.encoder_attention_heads) ) if decoder_head_mask is None: A__ : Tuple = tf.ones((config.decoder_layers, config.decoder_attention_heads) ) if cross_attn_head_mask is None: A__ : List[str] = tf.ones((config.decoder_layers, config.decoder_attention_heads) ) return { "input_ids": input_ids, "decoder_input_ids": decoder_input_ids, "attention_mask": attention_mask, "decoder_attention_mask": decoder_attention_mask, "head_mask": head_mask, "decoder_head_mask": decoder_head_mask, "cross_attn_head_mask": cross_attn_head_mask, } @require_tf class _a (__magic_name__ , __magic_name__ , unittest.TestCase ): '''simple docstring''' UpperCAmelCase__: List[Any] = (TFPegasusForConditionalGeneration, TFPegasusModel) if is_tf_available() else () UpperCAmelCase__: Tuple = (TFPegasusForConditionalGeneration,) if is_tf_available() else () UpperCAmelCase__: Tuple = ( { '''conversational''': TFPegasusForConditionalGeneration, '''feature-extraction''': TFPegasusModel, '''summarization''': TFPegasusForConditionalGeneration, '''text2text-generation''': TFPegasusForConditionalGeneration, '''translation''': TFPegasusForConditionalGeneration, } if is_tf_available() else {} ) UpperCAmelCase__: int = True UpperCAmelCase__: Union[str, Any] = False UpperCAmelCase__: List[str] = False def __A ( self ): A__ : Optional[Any] = TFPegasusModelTester(self ) A__ : Tuple = ConfigTester(self , config_class=A__ ) def __A ( self ): self.config_tester.run_common_tests() def __A ( self ): A__ : int = self.model_tester.prepare_config_and_inputs_for_common() self.model_tester.check_decoder_model_past_large_inputs(*A__ ) @require_sentencepiece @require_tokenizers @require_tf class _a (unittest.TestCase ): '''simple docstring''' UpperCAmelCase__: Optional[int] = [ ''' PG&E stated it scheduled the blackouts in response to forecasts for high winds amid dry conditions. The aim is to reduce the risk of wildfires. Nearly 800 thousand customers were scheduled to be affected by the shutoffs which were expected to last through at least midday tomorrow.''', ''' The London trio are up for best UK act and best album, as well as getting two nominations in the best song category."We got told like this morning \'Oh I think you\'re nominated\'", said Dappy."And I was like \'Oh yeah, which one?\' And now we\'ve got nominated for four awards. I mean, wow!"Bandmate Fazer added: "We thought it\'s best of us to come down and mingle with everyone and say hello to the cameras. And now we find we\'ve got four nominations."The band have two shots at the best song prize, getting the nod for their Tynchy Stryder collaboration Number One, and single Strong Again.Their album Uncle B will also go up against records by the likes of Beyonce and Kanye West.N-Dubz picked up the best newcomer Mobo in 2007, but female member Tulisa said they wouldn\'t be too disappointed if they didn\'t win this time around."At the end of the day we\'re grateful to be where we are in our careers."If it don\'t happen then it don\'t happen - live to fight another day and keep on making albums and hits for the fans."Dappy also revealed they could be performing live several times on the night.The group will be doing Number One and also a possible rendition of the War Child single, I Got Soul.The charity song is a re-working of The Killers\' All These Things That I\'ve Done and is set to feature artists like Chipmunk, Ironik and Pixie Lott.This year\'s Mobos will be held outside of London for the first time, in Glasgow on 30 September.N-Dubz said they were looking forward to performing for their Scottish fans and boasted about their recent shows north of the border."We just done Edinburgh the other day," said Dappy."We smashed up an N-Dubz show over there. We done Aberdeen about three or four months ago - we smashed up that show over there! Everywhere we go we smash it up!" ''', ] UpperCAmelCase__: Any = [ '''California\'s largest electricity provider has cut power to hundreds of thousands of customers in an effort to''' ''' reduce the risk of wildfires.''', '''N-Dubz have revealed they\'re "grateful" to have been nominated for four Mobo Awards.''', ] # differs slightly from pytorch, likely due to numerical differences in linear layers UpperCAmelCase__: List[str] = '''google/pegasus-xsum''' @cached_property def __A ( self ): return AutoTokenizer.from_pretrained(self.model_name ) @cached_property def __A ( self ): A__ : int = TFAutoModelForSeqaSeqLM.from_pretrained(self.model_name ) return model def __A ( self , **A__ ): A__ : str = self.translate_src_text(**A__ ) assert self.expected_text == generated_words def __A ( self , **A__ ): A__ : List[str] = self.tokenizer(self.src_text , **A__ , padding=A__ , return_tensors="""tf""" ) A__ : Optional[int] = self.model.generate( model_inputs.input_ids , attention_mask=model_inputs.attention_mask , num_beams=2 , use_cache=A__ , ) A__ : Dict = self.tokenizer.batch_decode(generated_ids.numpy() , skip_special_tokens=A__ ) return generated_words @slow def __A ( self ): self._assert_generated_batch_equal_expected()
64
1
import json from typing import List, Optional, Tuple from tokenizers import normalizers from tokenizers.pre_tokenizers import BertPreTokenizer, PreTokenizer from ...tokenization_utils_fast import PreTrainedTokenizerFast from ...utils import logging from .tokenization_roformer import RoFormerTokenizer from .tokenization_utils import JiebaPreTokenizer A_ : Dict = logging.get_logger(__name__) A_ : List[Any] = {'vocab_file': 'vocab.txt', 'tokenizer_file': 'tokenizer.json'} A_ : List[str] = { 'vocab_file': { 'junnyu/roformer_chinese_small': 'https://huggingface.co/junnyu/roformer_chinese_small/resolve/main/vocab.txt', 'junnyu/roformer_chinese_base': 'https://huggingface.co/junnyu/roformer_chinese_base/resolve/main/vocab.txt', 'junnyu/roformer_chinese_char_small': ( 'https://huggingface.co/junnyu/roformer_chinese_char_small/resolve/main/vocab.txt' ), 'junnyu/roformer_chinese_char_base': ( 'https://huggingface.co/junnyu/roformer_chinese_char_base/resolve/main/vocab.txt' ), 'junnyu/roformer_small_discriminator': ( 'https://huggingface.co/junnyu/roformer_small_discriminator/resolve/main/vocab.txt' ), 'junnyu/roformer_small_generator': ( 'https://huggingface.co/junnyu/roformer_small_generator/resolve/main/vocab.txt' ), } } A_ : Any = { 'junnyu/roformer_chinese_small': 1536, 'junnyu/roformer_chinese_base': 1536, 'junnyu/roformer_chinese_char_small': 512, 'junnyu/roformer_chinese_char_base': 512, 'junnyu/roformer_small_discriminator': 128, 'junnyu/roformer_small_generator': 128, } A_ : Union[str, Any] = { 'junnyu/roformer_chinese_small': {'do_lower_case': True}, 'junnyu/roformer_chinese_base': {'do_lower_case': True}, 'junnyu/roformer_chinese_char_small': {'do_lower_case': True}, 'junnyu/roformer_chinese_char_base': {'do_lower_case': True}, 'junnyu/roformer_small_discriminator': {'do_lower_case': True}, 'junnyu/roformer_small_generator': {'do_lower_case': True}, } class _a (__magic_name__ ): '''simple docstring''' UpperCAmelCase__: Union[str, Any] = VOCAB_FILES_NAMES UpperCAmelCase__: List[Any] = PRETRAINED_VOCAB_FILES_MAP UpperCAmelCase__: str = PRETRAINED_POSITIONAL_EMBEDDINGS_SIZES UpperCAmelCase__: str = PRETRAINED_INIT_CONFIGURATION UpperCAmelCase__: List[str] = RoFormerTokenizer def __init__( self , A__=None , A__=None , A__=True , A__="[UNK]" , A__="[SEP]" , A__="[PAD]" , A__="[CLS]" , A__="[MASK]" , A__=True , A__=None , **A__ , ): super().__init__( A__ , tokenizer_file=A__ , do_lower_case=A__ , unk_token=A__ , sep_token=A__ , pad_token=A__ , cls_token=A__ , mask_token=A__ , tokenize_chinese_chars=A__ , strip_accents=A__ , **A__ , ) A__ : str = json.loads(self.backend_tokenizer.normalizer.__getstate__() ) if ( pre_tok_state.get("""lowercase""" , A__ ) != do_lower_case or pre_tok_state.get("""strip_accents""" , A__ ) != strip_accents ): A__ : Dict = getattr(A__ , pre_tok_state.pop("""type""" ) ) A__ : int = do_lower_case A__ : str = strip_accents A__ : int = pre_tok_class(**A__ ) A__ : List[str] = do_lower_case def __getstate__( self ): A__ : Union[str, Any] = self.__dict__.copy() A__ : Tuple = BertPreTokenizer() return state def __setstate__( self , A__ ): A__ : List[Any] = d A__ : str = self.__dict__["""_tokenizer"""].get_vocab() A__ : Optional[int] = PreTokenizer.custom(JiebaPreTokenizer(A__ ) ) def __A ( self , A__ , A__=None ): A__ : Tuple = [self.cls_token_id] + token_ids_a + [self.sep_token_id] if token_ids_a: output += token_ids_a + [self.sep_token_id] return output def __A ( self , A__ , A__ = None ): A__ : Dict = [self.sep_token_id] A__ : Tuple = [self.cls_token_id] if token_ids_a is None: return len(cls + token_ids_a + sep ) * [0] return len(cls + token_ids_a + sep ) * [0] + len(token_ids_a + sep ) * [1] def __A ( self , A__ , A__ = None ): A__ : str = self._tokenizer.model.save(A__ , name=A__ ) return tuple(A__ ) def __A ( self , A__ , A__=None , A__=None , A__=False , **A__ , ): A__ : str = BertPreTokenizer() return super().save_pretrained(A__ , A__ , A__ , A__ , **A__ )
64
class _a : '''simple docstring''' def __init__( self ): A__ : str = """""" A__ : Any = """""" A__ : List[Any] = [] def __A ( self , A__ , A__ ): if m == -1: return n + 1 elif n == -1: return m + 1 elif self.dp[m][n] > -1: return self.dp[m][n] else: if self.worda[m] == self.worda[n]: A__ : Optional[Any] = self.__min_dist_top_down_dp(m - 1 , n - 1 ) else: A__ : Union[str, Any] = self.__min_dist_top_down_dp(A__ , n - 1 ) A__ : Union[str, Any] = self.__min_dist_top_down_dp(m - 1 , A__ ) A__ : Union[str, Any] = self.__min_dist_top_down_dp(m - 1 , n - 1 ) A__ : List[Any] = 1 + min(A__ , A__ , A__ ) return self.dp[m][n] def __A ( self , A__ , A__ ): A__ : Tuple = worda A__ : Dict = worda A__ : Optional[Any] = [[-1 for _ in range(len(A__ ) )] for _ in range(len(A__ ) )] return self.__min_dist_top_down_dp(len(A__ ) - 1 , len(A__ ) - 1 ) def __A ( self , A__ , A__ ): A__ : Optional[Any] = worda A__ : Dict = worda A__ : Union[str, Any] = len(A__ ) A__ : List[str] = len(A__ ) A__ : int = [[0 for _ in range(n + 1 )] for _ in range(m + 1 )] for i in range(m + 1 ): for j in range(n + 1 ): if i == 0: # first string is empty A__ : Tuple = j elif j == 0: # second string is empty A__ : Dict = i elif worda[i - 1] == worda[j - 1]: # last characters are equal A__ : str = self.dp[i - 1][j - 1] else: A__ : Union[str, Any] = self.dp[i][j - 1] A__ : str = self.dp[i - 1][j] A__ : Union[str, Any] = self.dp[i - 1][j - 1] A__ : Tuple = 1 + min(A__ , A__ , A__ ) return self.dp[m][n] if __name__ == "__main__": A_ : Union[str, Any] = EditDistance() print('****************** Testing Edit Distance DP Algorithm ******************') print() A_ : int = input('Enter the first string: ').strip() A_ : List[str] = input('Enter the second string: ').strip() print() print(f'''The minimum edit distance is: {solver.min_dist_top_down(Sa, Sa)}''') print(f'''The minimum edit distance is: {solver.min_dist_bottom_up(Sa, Sa)}''') print() print('*************** End of Testing Edit Distance DP Algorithm ***************')
64
1
import unittest import numpy as np from transformers.testing_utils import is_flaky, require_torch, require_vision from transformers.utils import is_torch_available, is_vision_available from ...test_image_processing_common import ImageProcessingSavingTestMixin, prepare_image_inputs if is_torch_available(): import torch if is_vision_available(): from PIL import Image from transformers import DonutImageProcessor class _a (unittest.TestCase ): '''simple docstring''' def __init__( self , A__ , A__=7 , A__=3 , A__=18 , A__=30 , A__=400 , A__=True , A__=None , A__=True , A__=False , A__=True , A__=True , A__=[0.5, 0.5, 0.5] , A__=[0.5, 0.5, 0.5] , ): A__ : int = parent A__ : Optional[Any] = batch_size A__ : Tuple = num_channels A__ : Dict = image_size A__ : str = min_resolution A__ : List[str] = max_resolution A__ : Tuple = do_resize A__ : Tuple = size if size is not None else {"""height""": 18, """width""": 20} A__ : Optional[Any] = do_thumbnail A__ : int = do_align_axis A__ : List[str] = do_pad A__ : List[str] = do_normalize A__ : Any = image_mean A__ : List[Any] = image_std def __A ( self ): return { "do_resize": self.do_resize, "size": self.size, "do_thumbnail": self.do_thumbnail, "do_align_long_axis": self.do_align_axis, "do_pad": self.do_pad, "do_normalize": self.do_normalize, "image_mean": self.image_mean, "image_std": self.image_std, } @require_torch @require_vision class _a (__magic_name__ , unittest.TestCase ): '''simple docstring''' UpperCAmelCase__: Tuple = DonutImageProcessor if is_vision_available() else None def __A ( self ): A__ : Optional[int] = DonutImageProcessingTester(self ) @property def __A ( self ): return self.image_processor_tester.prepare_image_processor_dict() def __A ( self ): A__ : List[Any] = self.image_processing_class(**self.image_processor_dict ) self.assertTrue(hasattr(A__ , """do_resize""" ) ) self.assertTrue(hasattr(A__ , """size""" ) ) self.assertTrue(hasattr(A__ , """do_thumbnail""" ) ) self.assertTrue(hasattr(A__ , """do_align_long_axis""" ) ) self.assertTrue(hasattr(A__ , """do_pad""" ) ) self.assertTrue(hasattr(A__ , """do_normalize""" ) ) self.assertTrue(hasattr(A__ , """image_mean""" ) ) self.assertTrue(hasattr(A__ , """image_std""" ) ) def __A ( self ): A__ : str = self.image_processing_class.from_dict(self.image_processor_dict ) self.assertEqual(image_processor.size , {"""height""": 18, """width""": 20} ) A__ : int = self.image_processing_class.from_dict(self.image_processor_dict , size=42 ) self.assertEqual(image_processor.size , {"""height""": 42, """width""": 42} ) # Previous config had dimensions in (width, height) order A__ : Tuple = self.image_processing_class.from_dict(self.image_processor_dict , size=(42, 84) ) self.assertEqual(image_processor.size , {"""height""": 84, """width""": 42} ) def __A ( self ): pass @is_flaky() def __A ( self ): # Initialize image_processing A__ : int = self.image_processing_class(**self.image_processor_dict ) # create random PIL images A__ : Union[str, Any] = prepare_image_inputs(self.image_processor_tester , equal_resolution=A__ ) for image in image_inputs: self.assertIsInstance(A__ , Image.Image ) # Test not batched input A__ : Optional[int] = image_processing(image_inputs[0] , return_tensors="""pt""" ).pixel_values self.assertEqual( encoded_images.shape , ( 1, self.image_processor_tester.num_channels, self.image_processor_tester.size["""height"""], self.image_processor_tester.size["""width"""], ) , ) # Test batched A__ : Union[str, Any] = image_processing(A__ , return_tensors="""pt""" ).pixel_values self.assertEqual( encoded_images.shape , ( self.image_processor_tester.batch_size, self.image_processor_tester.num_channels, self.image_processor_tester.size["""height"""], self.image_processor_tester.size["""width"""], ) , ) @is_flaky() def __A ( self ): # Initialize image_processing A__ : Union[str, Any] = self.image_processing_class(**self.image_processor_dict ) # create random numpy tensors A__ : List[str] = prepare_image_inputs(self.image_processor_tester , equal_resolution=A__ , numpify=A__ ) for image in image_inputs: self.assertIsInstance(A__ , np.ndarray ) # Test not batched input A__ : Union[str, Any] = image_processing(image_inputs[0] , return_tensors="""pt""" ).pixel_values self.assertEqual( encoded_images.shape , ( 1, self.image_processor_tester.num_channels, self.image_processor_tester.size["""height"""], self.image_processor_tester.size["""width"""], ) , ) # Test batched A__ : Optional[int] = image_processing(A__ , return_tensors="""pt""" ).pixel_values self.assertEqual( encoded_images.shape , ( self.image_processor_tester.batch_size, self.image_processor_tester.num_channels, self.image_processor_tester.size["""height"""], self.image_processor_tester.size["""width"""], ) , ) @is_flaky() def __A ( self ): # Initialize image_processing A__ : Any = self.image_processing_class(**self.image_processor_dict ) # create random PyTorch tensors A__ : List[str] = prepare_image_inputs(self.image_processor_tester , equal_resolution=A__ , torchify=A__ ) for image in image_inputs: self.assertIsInstance(A__ , torch.Tensor ) # Test not batched input A__ : Optional[Any] = image_processing(image_inputs[0] , return_tensors="""pt""" ).pixel_values self.assertEqual( encoded_images.shape , ( 1, self.image_processor_tester.num_channels, self.image_processor_tester.size["""height"""], self.image_processor_tester.size["""width"""], ) , ) # Test batched A__ : List[Any] = image_processing(A__ , return_tensors="""pt""" ).pixel_values self.assertEqual( encoded_images.shape , ( self.image_processor_tester.batch_size, self.image_processor_tester.num_channels, self.image_processor_tester.size["""height"""], self.image_processor_tester.size["""width"""], ) , )
64
def UpperCamelCase (lowercase_: int , lowercase_: int ) -> int: while second != 0: A__ : int = first & second first ^= second A__ : int = c << 1 return first if __name__ == "__main__": import doctest doctest.testmod() A_ : Optional[Any] = int(input('Enter the first number: ').strip()) A_ : List[str] = int(input('Enter the second number: ').strip()) print(f'''{add(first, second) = }''')
64
1
import functools def UpperCamelCase (lowercase_: list[int] , lowercase_: list[int] ) -> int: # Validation if not isinstance(lowercase_ , lowercase_ ) or not all(isinstance(lowercase_ , lowercase_ ) for day in days ): raise ValueError("""The parameter days should be a list of integers""" ) if len(lowercase_ ) != 3 or not all(isinstance(lowercase_ , lowercase_ ) for cost in costs ): raise ValueError("""The parameter costs should be a list of three integers""" ) if len(lowercase_ ) == 0: return 0 if min(lowercase_ ) <= 0: raise ValueError("""All days elements should be greater than 0""" ) if max(lowercase_ ) >= 366: raise ValueError("""All days elements should be less than 366""" ) A__ : List[str] = set(lowercase_ ) @functools.cache def dynamic_programming(lowercase_: int ) -> int: if index > 365: return 0 if index not in days_set: return dynamic_programming(index + 1 ) return min( costs[0] + dynamic_programming(index + 1 ) , costs[1] + dynamic_programming(index + 7 ) , costs[2] + dynamic_programming(index + 30 ) , ) return dynamic_programming(1 ) if __name__ == "__main__": import doctest doctest.testmod()
64
from __future__ import annotations from collections.abc import Callable A_ : List[Any] = list[list[float | int]] def UpperCamelCase (lowercase_: Matrix , lowercase_: Matrix ) -> Matrix: A__ : int = len(lowercase_ ) A__ : Matrix = [[0 for _ in range(size + 1 )] for _ in range(lowercase_ )] A__ : int A__ : int A__ : int A__ : int A__ : int A__ : float for row in range(lowercase_ ): for col in range(lowercase_ ): A__ : List[str] = matrix[row][col] A__ : int = vector[row][0] A__ : Optional[int] = 0 A__ : str = 0 while row < size and col < size: # pivoting A__ : int = max((abs(augmented[rowa][col] ), rowa) for rowa in range(lowercase_ , lowercase_ ) )[ 1 ] if augmented[pivot_row][col] == 0: col += 1 continue else: A__ , A__ : Union[str, Any] = augmented[pivot_row], augmented[row] for rowa in range(row + 1 , lowercase_ ): A__ : List[Any] = augmented[rowa][col] / augmented[row][col] A__ : Dict = 0 for cola in range(col + 1 , size + 1 ): augmented[rowa][cola] -= augmented[row][cola] * ratio row += 1 col += 1 # back substitution for col in range(1 , lowercase_ ): for row in range(lowercase_ ): A__ : List[str] = augmented[row][col] / augmented[col][col] for cola in range(lowercase_ , size + 1 ): augmented[row][cola] -= augmented[col][cola] * ratio # round to get rid of numbers like 2.000000000000004 return [ [round(augmented[row][size] / augmented[row][row] , 10 )] for row in range(lowercase_ ) ] def UpperCamelCase (lowercase_: list[int] ) -> Callable[[int], int]: A__ : int = len(lowercase_ ) A__ : Matrix = [[0 for _ in range(lowercase_ )] for _ in range(lowercase_ )] A__ : Matrix = [[0] for _ in range(lowercase_ )] A__ : Matrix A__ : int A__ : int A__ : int for x_val, y_val in enumerate(lowercase_ ): for col in range(lowercase_ ): A__ : Dict = (x_val + 1) ** (size - col - 1) A__ : Any = y_val A__ : Union[str, Any] = solve(lowercase_ , lowercase_ ) def interpolated_func(lowercase_: int ) -> int: return sum( round(coeffs[x_val][0] ) * (var ** (size - x_val - 1)) for x_val in range(lowercase_ ) ) return interpolated_func def UpperCamelCase (lowercase_: int ) -> int: return ( 1 - variable + variable**2 - variable**3 + variable**4 - variable**5 + variable**6 - variable**7 + variable**8 - variable**9 + variable**10 ) def UpperCamelCase (lowercase_: Callable[[int], int] = question_function , lowercase_: int = 10 ) -> int: A__ : list[int] = [func(lowercase_ ) for x_val in range(1 , order + 1 )] A__ : list[Callable[[int], int]] = [ interpolate(data_points[:max_coeff] ) for max_coeff in range(1 , order + 1 ) ] A__ : int = 0 A__ : Callable[[int], int] A__ : int for poly in polynomials: A__ : List[str] = 1 while func(lowercase_ ) == poly(lowercase_ ): x_val += 1 ret += poly(lowercase_ ) return ret if __name__ == "__main__": print(f'''{solution() = }''')
64
1
import importlib import sys from argparse import REMAINDER, ArgumentParser from pathlib import Path import torch_xla.distributed.xla_multiprocessing as xmp def UpperCamelCase () -> Any: A__ : int = ArgumentParser( description=( """PyTorch TPU distributed training launch """ """helper utility that will spawn up """ """multiple distributed processes""" ) ) # Optional arguments for the launch helper parser.add_argument("""--num_cores""" , type=lowercase_ , default=1 , help="""Number of TPU cores to use (1 or 8).""" ) # positional parser.add_argument( """training_script""" , type=lowercase_ , help=( """The full path to the single TPU training """ """program/script to be launched in parallel, """ """followed by all the arguments for the """ """training script""" ) , ) # rest from the training program parser.add_argument("""training_script_args""" , nargs=lowercase_ ) return parser.parse_args() def UpperCamelCase () -> List[Any]: A__ : List[str] = parse_args() # Import training_script as a module. A__ : List[str] = Path(args.training_script ) sys.path.append(str(script_fpath.parent.resolve() ) ) A__ : str = script_fpath.stem A__ : str = importlib.import_module(lowercase_ ) # Patch sys.argv A__ : Optional[int] = [args.training_script] + args.training_script_args + ["""--tpu_num_cores""", str(args.num_cores )] xmp.spawn(mod._mp_fn , args=() , nprocs=args.num_cores ) if __name__ == "__main__": main()
64
from functools import lru_cache @lru_cache def UpperCamelCase (lowercase_: int ) -> int: if num < 0: raise ValueError("""Number should not be negative.""" ) return 1 if num in (0, 1) else num * factorial(num - 1 ) if __name__ == "__main__": import doctest doctest.testmod()
64
1
import logging import os import sys from dataclasses import dataclass, field from itertools import chain from typing import Optional, Union import datasets import numpy as np import torch from datasets import load_dataset import transformers from transformers import ( AutoConfig, AutoModelForMultipleChoice, AutoTokenizer, HfArgumentParser, Trainer, TrainingArguments, default_data_collator, set_seed, ) from transformers.tokenization_utils_base import PreTrainedTokenizerBase from transformers.trainer_utils import get_last_checkpoint from transformers.utils import PaddingStrategy, check_min_version, send_example_telemetry # Will error if the minimal version of Transformers is not installed. Remove at your own risks. check_min_version('4.31.0') A_ : Tuple = logging.getLogger(__name__) @dataclass class _a : '''simple docstring''' UpperCAmelCase__: str = field( metadata={'''help''': '''Path to pretrained model or model identifier from huggingface.co/models'''} ) UpperCAmelCase__: Optional[str] = field( default=__magic_name__ , metadata={'''help''': '''Pretrained config name or path if not the same as model_name'''} ) UpperCAmelCase__: Optional[str] = field( default=__magic_name__ , metadata={'''help''': '''Pretrained tokenizer name or path if not the same as model_name'''} ) UpperCAmelCase__: Optional[str] = field( default=__magic_name__ , metadata={'''help''': '''Where do you want to store the pretrained models downloaded from huggingface.co'''} , ) UpperCAmelCase__: bool = field( default=__magic_name__ , metadata={'''help''': '''Whether to use one of the fast tokenizer (backed by the tokenizers library) or not.'''} , ) UpperCAmelCase__: str = field( default='''main''' , metadata={'''help''': '''The specific model version to use (can be a branch name, tag name or commit id).'''} , ) UpperCAmelCase__: bool = field( default=__magic_name__ , metadata={ '''help''': ( '''Will use the token generated when running `huggingface-cli login` (necessary to use this script ''' '''with private models).''' ) } , ) @dataclass class _a : '''simple docstring''' UpperCAmelCase__: Optional[str] = field(default=__magic_name__ , metadata={'''help''': '''The input training data file (a text file).'''} ) UpperCAmelCase__: Optional[str] = field( default=__magic_name__ , metadata={'''help''': '''An optional input evaluation data file to evaluate the perplexity on (a text file).'''} , ) UpperCAmelCase__: bool = field( default=__magic_name__ , metadata={'''help''': '''Overwrite the cached training and evaluation sets'''} ) UpperCAmelCase__: Optional[int] = field( default=__magic_name__ , metadata={'''help''': '''The number of processes to use for the preprocessing.'''} , ) UpperCAmelCase__: Optional[int] = field( default=__magic_name__ , metadata={ '''help''': ( '''The maximum total input sequence length after tokenization. If passed, sequences longer ''' '''than this will be truncated, sequences shorter will be padded.''' ) } , ) UpperCAmelCase__: bool = field( default=__magic_name__ , metadata={ '''help''': ( '''Whether to pad all samples to the maximum sentence length. ''' '''If False, will pad the samples dynamically when batching to the maximum length in the batch. More ''' '''efficient on GPU but very bad for TPU.''' ) } , ) UpperCAmelCase__: Optional[int] = field( default=__magic_name__ , metadata={ '''help''': ( '''For debugging purposes or quicker training, truncate the number of training examples to this ''' '''value if set.''' ) } , ) UpperCAmelCase__: Optional[int] = field( default=__magic_name__ , metadata={ '''help''': ( '''For debugging purposes or quicker training, truncate the number of evaluation examples to this ''' '''value if set.''' ) } , ) def __A ( self ): if self.train_file is not None: A__ : int = self.train_file.split(""".""" )[-1] assert extension in ["csv", "json"], "`train_file` should be a csv or a json file." if self.validation_file is not None: A__ : Any = self.validation_file.split(""".""" )[-1] assert extension in ["csv", "json"], "`validation_file` should be a csv or a json file." @dataclass class _a : '''simple docstring''' UpperCAmelCase__: PreTrainedTokenizerBase UpperCAmelCase__: Union[bool, str, PaddingStrategy] = True UpperCAmelCase__: Optional[int] = None UpperCAmelCase__: Optional[int] = None def __call__( self , A__ ): A__ : Dict = """label""" if """label""" in features[0].keys() else """labels""" A__ : int = [feature.pop(A__ ) for feature in features] A__ : Tuple = len(A__ ) A__ : Any = len(features[0]["""input_ids"""] ) A__ : str = [ [{k: v[i] for k, v in feature.items()} for i in range(A__ )] for feature in features ] A__ : Dict = list(chain(*A__ ) ) A__ : Dict = self.tokenizer.pad( A__ , padding=self.padding , max_length=self.max_length , pad_to_multiple_of=self.pad_to_multiple_of , return_tensors="""pt""" , ) # Un-flatten A__ : List[Any] = {k: v.view(A__ , A__ , -1 ) for k, v in batch.items()} # Add back labels A__ : Any = torch.tensor(A__ , dtype=torch.intaa ) return batch def UpperCamelCase () -> Tuple: # See all possible arguments in src/transformers/training_args.py # or by passing the --help flag to this script. # We now keep distinct sets of args, for a cleaner separation of concerns. A__ : List[str] = HfArgumentParser((ModelArguments, DataTrainingArguments, TrainingArguments) ) if len(sys.argv ) == 2 and sys.argv[1].endswith(""".json""" ): # If we pass only one argument to the script and it's the path to a json file, # let's parse it to get our arguments. A__ , A__ , A__ : str = parser.parse_json_file(json_file=os.path.abspath(sys.argv[1] ) ) else: A__ , A__ , A__ : Any = parser.parse_args_into_dataclasses() # Sending telemetry. Tracking the example usage helps us better allocate resources to maintain them. The # information sent is the one passed as arguments along with your Python/PyTorch versions. send_example_telemetry("""run_swag""" , lowercase_ , lowercase_ ) # Setup logging logging.basicConfig( format="""%(asctime)s - %(levelname)s - %(name)s - %(message)s""" , datefmt="""%m/%d/%Y %H:%M:%S""" , handlers=[logging.StreamHandler(sys.stdout )] , ) if training_args.should_log: # The default of training_args.log_level is passive, so we set log level at info here to have that default. transformers.utils.logging.set_verbosity_info() A__ : Any = training_args.get_process_log_level() logger.setLevel(lowercase_ ) datasets.utils.logging.set_verbosity(lowercase_ ) transformers.utils.logging.set_verbosity(lowercase_ ) transformers.utils.logging.enable_default_handler() transformers.utils.logging.enable_explicit_format() # Log on each process the small summary: logger.warning( f"""Process rank: {training_args.local_rank}, device: {training_args.device}, n_gpu: {training_args.n_gpu}""" + f"""distributed training: {bool(training_args.local_rank != -1 )}, 16-bits training: {training_args.fpaa}""" ) logger.info(f"""Training/evaluation parameters {training_args}""" ) # Detecting last checkpoint. A__ : int = None if os.path.isdir(training_args.output_dir ) and training_args.do_train and not training_args.overwrite_output_dir: A__ : str = get_last_checkpoint(training_args.output_dir ) if last_checkpoint is None and len(os.listdir(training_args.output_dir ) ) > 0: raise ValueError( f"""Output directory ({training_args.output_dir}) already exists and is not empty. """ """Use --overwrite_output_dir to overcome.""" ) elif last_checkpoint is not None and training_args.resume_from_checkpoint is None: logger.info( f"""Checkpoint detected, resuming training at {last_checkpoint}. To avoid this behavior, change """ """the `--output_dir` or add `--overwrite_output_dir` to train from scratch.""" ) # Set seed before initializing model. set_seed(training_args.seed ) # Get the datasets: you can either provide your own CSV/JSON/TXT training and evaluation files (see below) # or just provide the name of one of the public datasets available on the hub at https://huggingface.co/datasets/ # (the dataset will be downloaded automatically from the datasets Hub). # For CSV/JSON files, this script will use the column called 'text' or the first column if no column called # 'text' is found. You can easily tweak this behavior (see below). # In distributed training, the load_dataset function guarantee that only one local process can concurrently # download the dataset. if data_args.train_file is not None or data_args.validation_file is not None: A__ : Union[str, Any] = {} if data_args.train_file is not None: A__ : List[str] = data_args.train_file if data_args.validation_file is not None: A__ : Optional[Any] = data_args.validation_file A__ : List[Any] = data_args.train_file.split(""".""" )[-1] A__ : Any = load_dataset( lowercase_ , data_files=lowercase_ , cache_dir=model_args.cache_dir , use_auth_token=True if model_args.use_auth_token else None , ) else: # Downloading and loading the swag dataset from the hub. A__ : int = load_dataset( """swag""" , """regular""" , cache_dir=model_args.cache_dir , use_auth_token=True if model_args.use_auth_token else None , ) # See more about loading any type of standard or custom dataset (from files, python dict, pandas DataFrame, etc) at # https://huggingface.co/docs/datasets/loading_datasets.html. # Load pretrained model and tokenizer # Distributed training: # The .from_pretrained methods guarantee that only one local process can concurrently # download model & vocab. A__ : Dict = AutoConfig.from_pretrained( model_args.config_name if model_args.config_name else model_args.model_name_or_path , cache_dir=model_args.cache_dir , revision=model_args.model_revision , use_auth_token=True if model_args.use_auth_token else None , ) A__ : int = AutoTokenizer.from_pretrained( model_args.tokenizer_name if model_args.tokenizer_name else model_args.model_name_or_path , cache_dir=model_args.cache_dir , use_fast=model_args.use_fast_tokenizer , revision=model_args.model_revision , use_auth_token=True if model_args.use_auth_token else None , ) A__ : List[Any] = AutoModelForMultipleChoice.from_pretrained( model_args.model_name_or_path , from_tf=bool(""".ckpt""" in model_args.model_name_or_path ) , config=lowercase_ , cache_dir=model_args.cache_dir , revision=model_args.model_revision , use_auth_token=True if model_args.use_auth_token else None , ) # When using your own dataset or a different dataset from swag, you will probably need to change this. A__ : Optional[int] = [f"""ending{i}""" for i in range(4 )] A__ : List[str] = """sent1""" A__ : List[str] = """sent2""" if data_args.max_seq_length is None: A__ : Tuple = tokenizer.model_max_length if max_seq_length > 1024: logger.warning( """The chosen tokenizer supports a `model_max_length` that is longer than the default `block_size` value""" """ of 1024. If you would like to use a longer `block_size` up to `tokenizer.model_max_length` you can""" """ override this default with `--block_size xxx`.""" ) A__ : Optional[Any] = 1024 else: if data_args.max_seq_length > tokenizer.model_max_length: logger.warning( f"""The max_seq_length passed ({data_args.max_seq_length}) is larger than the maximum length for the""" f"""model ({tokenizer.model_max_length}). Using max_seq_length={tokenizer.model_max_length}.""" ) A__ : List[Any] = min(data_args.max_seq_length , tokenizer.model_max_length ) # Preprocessing the datasets. def preprocess_function(lowercase_: Dict ): A__ : Union[str, Any] = [[context] * 4 for context in examples[context_name]] A__ : str = examples[question_header_name] A__ : List[str] = [ [f"""{header} {examples[end][i]}""" for end in ending_names] for i, header in enumerate(lowercase_ ) ] # Flatten out A__ : int = list(chain(*lowercase_ ) ) A__ : Any = list(chain(*lowercase_ ) ) # Tokenize A__ : Optional[Any] = tokenizer( lowercase_ , lowercase_ , truncation=lowercase_ , max_length=lowercase_ , padding="""max_length""" if data_args.pad_to_max_length else False , ) # Un-flatten return {k: [v[i : i + 4] for i in range(0 , len(lowercase_ ) , 4 )] for k, v in tokenized_examples.items()} if training_args.do_train: if "train" not in raw_datasets: raise ValueError("""--do_train requires a train dataset""" ) A__ : Dict = raw_datasets["""train"""] if data_args.max_train_samples is not None: A__ : int = min(len(lowercase_ ) , data_args.max_train_samples ) A__ : str = train_dataset.select(range(lowercase_ ) ) with training_args.main_process_first(desc="""train dataset map pre-processing""" ): A__ : str = train_dataset.map( lowercase_ , batched=lowercase_ , num_proc=data_args.preprocessing_num_workers , load_from_cache_file=not data_args.overwrite_cache , ) if training_args.do_eval: if "validation" not in raw_datasets: raise ValueError("""--do_eval requires a validation dataset""" ) A__ : Dict = raw_datasets["""validation"""] if data_args.max_eval_samples is not None: A__ : List[Any] = min(len(lowercase_ ) , data_args.max_eval_samples ) A__ : int = eval_dataset.select(range(lowercase_ ) ) with training_args.main_process_first(desc="""validation dataset map pre-processing""" ): A__ : Dict = eval_dataset.map( lowercase_ , batched=lowercase_ , num_proc=data_args.preprocessing_num_workers , load_from_cache_file=not data_args.overwrite_cache , ) # Data collator A__ : int = ( default_data_collator if data_args.pad_to_max_length else DataCollatorForMultipleChoice(tokenizer=lowercase_ , pad_to_multiple_of=8 if training_args.fpaa else None ) ) # Metric def compute_metrics(lowercase_: str ): A__ , A__ : Union[str, Any] = eval_predictions A__ : Dict = np.argmax(lowercase_ , axis=1 ) return {"accuracy": (preds == label_ids).astype(np.floataa ).mean().item()} # Initialize our Trainer A__ : Union[str, Any] = Trainer( model=lowercase_ , args=lowercase_ , train_dataset=train_dataset if training_args.do_train else None , eval_dataset=eval_dataset if training_args.do_eval else None , tokenizer=lowercase_ , data_collator=lowercase_ , compute_metrics=lowercase_ , ) # Training if training_args.do_train: A__ : str = None if training_args.resume_from_checkpoint is not None: A__ : Optional[int] = training_args.resume_from_checkpoint elif last_checkpoint is not None: A__ : List[Any] = last_checkpoint A__ : Optional[int] = trainer.train(resume_from_checkpoint=lowercase_ ) trainer.save_model() # Saves the tokenizer too for easy upload A__ : Tuple = train_result.metrics A__ : str = ( data_args.max_train_samples if data_args.max_train_samples is not None else len(lowercase_ ) ) A__ : int = min(lowercase_ , len(lowercase_ ) ) trainer.log_metrics("""train""" , lowercase_ ) trainer.save_metrics("""train""" , lowercase_ ) trainer.save_state() # Evaluation if training_args.do_eval: logger.info("""*** Evaluate ***""" ) A__ : Any = trainer.evaluate() A__ : Dict = data_args.max_eval_samples if data_args.max_eval_samples is not None else len(lowercase_ ) A__ : Union[str, Any] = min(lowercase_ , len(lowercase_ ) ) trainer.log_metrics("""eval""" , lowercase_ ) trainer.save_metrics("""eval""" , lowercase_ ) A__ : Union[str, Any] = { """finetuned_from""": model_args.model_name_or_path, """tasks""": """multiple-choice""", """dataset_tags""": """swag""", """dataset_args""": """regular""", """dataset""": """SWAG""", """language""": """en""", } if training_args.push_to_hub: trainer.push_to_hub(**lowercase_ ) else: trainer.create_model_card(**lowercase_ ) def UpperCamelCase (lowercase_: Optional[Any] ) -> Tuple: # For xla_spawn (TPUs) main() if __name__ == "__main__": main()
64
import os import tempfile from functools import partial from unittest import TestCase from unittest.mock import patch import datasets import datasets.config from .utils import require_beam class _a (datasets.BeamBasedBuilder ): '''simple docstring''' def __A ( self ): return datasets.DatasetInfo( features=datasets.Features({"""content""": datasets.Value("""string""" )} ) , supervised_keys=A__ , ) def __A ( self , A__ , A__ ): return [datasets.SplitGenerator(name=datasets.Split.TRAIN , gen_kwargs={"""examples""": get_test_dummy_examples()} )] def __A ( self , A__ , A__ ): import apache_beam as beam return pipeline | "Load Examples" >> beam.Create(A__ ) class _a (datasets.BeamBasedBuilder ): '''simple docstring''' def __A ( self ): return datasets.DatasetInfo( features=datasets.Features({"""a""": datasets.Sequence({"""b""": datasets.Value("""string""" )} )} ) , supervised_keys=A__ , ) def __A ( self , A__ , A__ ): return [ datasets.SplitGenerator(name=datasets.Split.TRAIN , gen_kwargs={"""examples""": get_test_nested_examples()} ) ] def __A ( self , A__ , A__ ): import apache_beam as beam return pipeline | "Load Examples" >> beam.Create(A__ ) def UpperCamelCase () -> Dict: return [(i, {"content": content}) for i, content in enumerate(["""foo""", """bar""", """foobar"""] )] def UpperCamelCase () -> Tuple: return [(i, {"a": {"b": [content]}}) for i, content in enumerate(["""foo""", """bar""", """foobar"""] )] class _a (__magic_name__ ): '''simple docstring''' @require_beam def __A ( self ): A__ : Dict = len(get_test_dummy_examples() ) with tempfile.TemporaryDirectory() as tmp_cache_dir: A__ : int = DummyBeamDataset(cache_dir=A__ , beam_runner="""DirectRunner""" ) builder.download_and_prepare() self.assertTrue( os.path.exists( os.path.join(A__ , builder.name , """default""" , """0.0.0""" , F"""{builder.name}-train.arrow""" ) ) ) self.assertDictEqual(builder.info.features , datasets.Features({"""content""": datasets.Value("""string""" )} ) ) A__ : int = builder.as_dataset() self.assertEqual(dset["""train"""].num_rows , A__ ) self.assertEqual(dset["""train"""].info.splits["""train"""].num_examples , A__ ) self.assertDictEqual(dset["""train"""][0] , get_test_dummy_examples()[0][1] ) self.assertDictEqual( dset["""train"""][expected_num_examples - 1] , get_test_dummy_examples()[expected_num_examples - 1][1] ) self.assertTrue( os.path.exists(os.path.join(A__ , builder.name , """default""" , """0.0.0""" , """dataset_info.json""" ) ) ) del dset @require_beam def __A ( self ): import apache_beam as beam A__ : int = beam.io.parquetio.WriteToParquet A__ : List[str] = len(get_test_dummy_examples() ) with tempfile.TemporaryDirectory() as tmp_cache_dir: A__ : str = DummyBeamDataset(cache_dir=A__ , beam_runner="""DirectRunner""" ) with patch("""apache_beam.io.parquetio.WriteToParquet""" ) as write_parquet_mock: A__ : Optional[Any] = partial(A__ , num_shards=2 ) builder.download_and_prepare() self.assertTrue( os.path.exists( os.path.join( A__ , builder.name , """default""" , """0.0.0""" , F"""{builder.name}-train-00000-of-00002.arrow""" ) ) ) self.assertTrue( os.path.exists( os.path.join( A__ , builder.name , """default""" , """0.0.0""" , F"""{builder.name}-train-00000-of-00002.arrow""" ) ) ) self.assertDictEqual(builder.info.features , datasets.Features({"""content""": datasets.Value("""string""" )} ) ) A__ : Optional[int] = builder.as_dataset() self.assertEqual(dset["""train"""].num_rows , A__ ) self.assertEqual(dset["""train"""].info.splits["""train"""].num_examples , A__ ) # Order is not preserved when sharding, so we just check that all the elements are there self.assertListEqual(sorted(dset["""train"""]["""content"""] ) , sorted(["""foo""", """bar""", """foobar"""] ) ) self.assertTrue( os.path.exists(os.path.join(A__ , builder.name , """default""" , """0.0.0""" , """dataset_info.json""" ) ) ) del dset @require_beam def __A ( self ): with tempfile.TemporaryDirectory() as tmp_cache_dir: A__ : int = DummyBeamDataset(cache_dir=A__ ) self.assertRaises(datasets.builder.MissingBeamOptions , builder.download_and_prepare ) @require_beam def __A ( self ): A__ : List[Any] = len(get_test_nested_examples() ) with tempfile.TemporaryDirectory() as tmp_cache_dir: A__ : Optional[int] = NestedBeamDataset(cache_dir=A__ , beam_runner="""DirectRunner""" ) builder.download_and_prepare() self.assertTrue( os.path.exists( os.path.join(A__ , builder.name , """default""" , """0.0.0""" , F"""{builder.name}-train.arrow""" ) ) ) self.assertDictEqual( builder.info.features , datasets.Features({"""a""": datasets.Sequence({"""b""": datasets.Value("""string""" )} )} ) ) A__ : Optional[int] = builder.as_dataset() self.assertEqual(dset["""train"""].num_rows , A__ ) self.assertEqual(dset["""train"""].info.splits["""train"""].num_examples , A__ ) self.assertDictEqual(dset["""train"""][0] , get_test_nested_examples()[0][1] ) self.assertDictEqual( dset["""train"""][expected_num_examples - 1] , get_test_nested_examples()[expected_num_examples - 1][1] ) self.assertTrue( os.path.exists(os.path.join(A__ , builder.name , """default""" , """0.0.0""" , """dataset_info.json""" ) ) ) del dset
64
1
def UpperCamelCase (lowercase_: list[int] , lowercase_: str ) -> list[int]: A__ : Optional[int] = int(lowercase_ ) # Initialize Result A__ : Any = [] # Traverse through all denomination for denomination in reversed(lowercase_ ): # Find denominations while int(lowercase_ ) >= int(lowercase_ ): total_value -= int(lowercase_ ) answer.append(lowercase_ ) # Append the "answers" array return answer # Driver Code if __name__ == "__main__": A_ : Optional[Any] = [] A_ : Tuple = '0' if ( input('Do you want to enter your denominations ? (yY/n): ').strip().lower() == "y" ): A_ : str = int(input('Enter the number of denominations you want to add: ').strip()) for i in range(0, n): denominations.append(int(input(f'''Denomination {i}: ''').strip())) A_ : Any = input('Enter the change you want to make in Indian Currency: ').strip() else: # All denominations of Indian Currency if user does not enter A_ : Dict = [1, 2, 5, 10, 20, 50, 100, 500, 2000] A_ : Any = input('Enter the change you want to make: ').strip() if int(value) == 0 or int(value) < 0: print('The total value cannot be zero or negative.') else: print(f'''Following is minimal change for {value}: ''') A_ : List[str] = find_minimum_change(denominations, value) # Print result for i in range(len(answer)): print(answer[i], end=' ')
64
import warnings from ...utils import logging from .image_processing_poolformer import PoolFormerImageProcessor A_ : Union[str, Any] = logging.get_logger(__name__) class _a (__magic_name__ ): '''simple docstring''' def __init__( self , *A__ , **A__ ): warnings.warn( """The class PoolFormerFeatureExtractor is deprecated and will be removed in version 5 of Transformers.""" """ Please use PoolFormerImageProcessor instead.""" , A__ , ) super().__init__(*A__ , **A__ )
64
1
from typing import List, Optional, Tuple from ...tokenization_utils_fast import PreTrainedTokenizerFast from ...utils import logging from .tokenization_herbert import HerbertTokenizer A_ : List[Any] = logging.get_logger(__name__) A_ : Any = {'vocab_file': 'vocab.json', 'merges_file': 'merges.txt', 'tokenizer_file': 'tokenizer.json'} A_ : Union[str, Any] = { 'vocab_file': { 'allegro/herbert-base-cased': 'https://huggingface.co/allegro/herbert-base-cased/resolve/main/vocab.json' }, 'merges_file': { 'allegro/herbert-base-cased': 'https://huggingface.co/allegro/herbert-base-cased/resolve/main/merges.txt' }, } A_ : Dict = {'allegro/herbert-base-cased': 514} A_ : Tuple = {} class _a (__magic_name__ ): '''simple docstring''' UpperCAmelCase__: Union[str, Any] = VOCAB_FILES_NAMES UpperCAmelCase__: Dict = PRETRAINED_VOCAB_FILES_MAP UpperCAmelCase__: Any = PRETRAINED_INIT_CONFIGURATION UpperCAmelCase__: Any = PRETRAINED_POSITIONAL_EMBEDDINGS_SIZES UpperCAmelCase__: Optional[int] = HerbertTokenizer def __init__( self , A__=None , A__=None , A__=None , A__="<s>" , A__="<unk>" , A__="<pad>" , A__="<mask>" , A__="</s>" , **A__ , ): super().__init__( A__ , A__ , tokenizer_file=A__ , cls_token=A__ , unk_token=A__ , pad_token=A__ , mask_token=A__ , sep_token=A__ , **A__ , ) def __A ( self , A__ , A__ = None ): A__ : int = [self.cls_token_id] A__ : Optional[Any] = [self.sep_token_id] if token_ids_a is None: return cls + token_ids_a + sep return cls + token_ids_a + sep + token_ids_a + sep def __A ( self , A__ , A__ = None , A__ = False ): if already_has_special_tokens: return super().get_special_tokens_mask( token_ids_a=A__ , token_ids_a=A__ , already_has_special_tokens=A__ ) if token_ids_a is None: return [1] + ([0] * len(A__ )) + [1] return [1] + ([0] * len(A__ )) + [1] + ([0] * len(A__ )) + [1] def __A ( self , A__ , A__ = None ): A__ : Optional[Any] = [self.sep_token_id] A__ : Optional[int] = [self.cls_token_id] if token_ids_a is None: return len(cls + token_ids_a + sep ) * [0] return len(cls + token_ids_a + sep ) * [0] + len(token_ids_a + sep ) * [1] def __A ( self , A__ , A__ = None ): A__ : List[str] = self._tokenizer.model.save(A__ , name=A__ ) return tuple(A__ )
64
import inspect import logging import os import random import shutil import tempfile import unittest import pytest import torch from torch import nn from torch.utils.data import DataLoader, TensorDataset from accelerate import Accelerator from accelerate.test_utils import execute_subprocess_async, require_cuda from accelerate.utils import ProjectConfiguration, set_seed A_ : Any = logging.getLogger(__name__) def UpperCamelCase (lowercase_: Optional[Any]=2 , lowercase_: Union[str, Any]=3 , lowercase_: int=16 , lowercase_: int = 10 , lowercase_: int = 2 ) -> int: def get_dataset(lowercase_: Optional[int] ): A__ : Optional[Any] = torch.randn(batch_size * n_batches , 1 ) return TensorDataset(lowercase_ , a * x + b + 0.1 * torch.randn(batch_size * n_batches , 1 ) ) A__ : Dict = get_dataset(lowercase_ ) A__ : Any = get_dataset(lowercase_ ) A__ : Dict = DataLoader(lowercase_ , shuffle=lowercase_ , batch_size=lowercase_ , num_workers=4 ) A__ : Optional[Any] = DataLoader(lowercase_ , shuffle=lowercase_ , batch_size=lowercase_ , num_workers=4 ) return (train_dataloader, valid_dataloader) def UpperCamelCase (lowercase_: Optional[Any] , lowercase_: List[str] , lowercase_: int , lowercase_: int , lowercase_: List[str] , lowercase_: Dict=None ) -> List[Any]: A__ : List[Any] = [] for epoch in range(lowercase_ ): # Train quickly model.train() for batch in dataloader: A__ , A__ : Any = batch A__ : Any = model(lowercase_ ) A__ : Any = torch.nn.functional.mse_loss(lowercase_ , lowercase_ ) accelerator.backward(lowercase_ ) optimizer.step() optimizer.zero_grad() rands.append(random.random() ) # Introduce some randomness if scheduler is not None: scheduler.step() return rands class _a (nn.Module ): '''simple docstring''' def __init__( self ): super().__init__() A__ : str = nn.Parameter(torch.randn(1 ) ) A__ : Any = nn.Parameter(torch.randn(1 ) ) def __A ( self , A__ ): return x * self.a + self.b class _a (unittest.TestCase ): '''simple docstring''' def __A ( self ): with tempfile.TemporaryDirectory() as tmpdir: set_seed(42 ) A__ : Optional[Any] = DummyModel() A__ : Optional[Any] = torch.optim.Adam(params=model.parameters() , lr=1e-3 ) A__ , A__ : str = dummy_dataloaders() A__ : Dict = ProjectConfiguration(total_limit=1 , project_dir=A__ , automatic_checkpoint_naming=A__ ) # Train baseline A__ : List[str] = Accelerator(project_config=A__ ) A__ , A__ , A__ , A__ : Any = accelerator.prepare( A__ , A__ , A__ , A__ ) # Save initial accelerator.save_state() # Save second state accelerator.save_state() self.assertEqual(len(os.listdir(accelerator.project_dir ) ) , 1 ) def __A ( self ): with tempfile.TemporaryDirectory() as tmpdir: set_seed(42 ) A__ : str = DummyModel() A__ : Optional[int] = torch.optim.Adam(params=model.parameters() , lr=1e-3 ) A__ , A__ : int = dummy_dataloaders() # Train baseline A__ : str = Accelerator() A__ , A__ , A__ , A__ : List[str] = accelerator.prepare( A__ , A__ , A__ , A__ ) # Save initial A__ : List[Any] = os.path.join(A__ , """initial""" ) accelerator.save_state(A__ ) ((A__) , (A__)) : str = model.a.item(), model.b.item() A__ : Dict = optimizer.state_dict() A__ : List[str] = train(3 , A__ , A__ , A__ , A__ ) ((A__) , (A__)) : str = model.a.item(), model.b.item() A__ : Any = optimizer.state_dict() # Train partially set_seed(42 ) A__ : Optional[int] = DummyModel() A__ : Dict = torch.optim.Adam(params=model.parameters() , lr=1e-3 ) A__ , A__ : Dict = dummy_dataloaders() A__ : List[str] = Accelerator() A__ , A__ , A__ , A__ : Optional[Any] = accelerator.prepare( A__ , A__ , A__ , A__ ) accelerator.load_state(A__ ) ((A__) , (A__)) : Tuple = model.a.item(), model.b.item() A__ : Union[str, Any] = optimizer.state_dict() self.assertEqual(A__ , A__ ) self.assertEqual(A__ , A__ ) self.assertEqual(A__ , A__ ) A__ : List[str] = train(2 , A__ , A__ , A__ , A__ ) # Save everything A__ : Optional[int] = os.path.join(A__ , """checkpoint""" ) accelerator.save_state(A__ ) # Load everything back in and make sure all states work accelerator.load_state(A__ ) test_rands += train(1 , A__ , A__ , A__ , A__ ) ((A__) , (A__)) : Union[str, Any] = model.a.item(), model.b.item() A__ : Optional[int] = optimizer.state_dict() self.assertEqual(A__ , A__ ) self.assertEqual(A__ , A__ ) self.assertEqual(A__ , A__ ) self.assertEqual(A__ , A__ ) def __A ( self ): with tempfile.TemporaryDirectory() as tmpdir: set_seed(42 ) A__ : int = DummyModel() A__ : Optional[int] = torch.optim.Adam(params=model.parameters() , lr=1e-3 ) A__ , A__ : List[str] = dummy_dataloaders() A__ : str = ProjectConfiguration(automatic_checkpoint_naming=A__ ) # Train baseline A__ : Any = Accelerator(project_dir=A__ , project_config=A__ ) A__ , A__ , A__ , A__ : str = accelerator.prepare( A__ , A__ , A__ , A__ ) # Save initial accelerator.save_state() ((A__) , (A__)) : Tuple = model.a.item(), model.b.item() A__ : int = optimizer.state_dict() A__ : int = train(3 , A__ , A__ , A__ , A__ ) ((A__) , (A__)) : Optional[Any] = model.a.item(), model.b.item() A__ : Any = optimizer.state_dict() # Train partially set_seed(42 ) A__ : Dict = DummyModel() A__ : List[Any] = torch.optim.Adam(params=model.parameters() , lr=1e-3 ) A__ , A__ : Union[str, Any] = dummy_dataloaders() A__ : List[Any] = ProjectConfiguration(iteration=1 , automatic_checkpoint_naming=A__ ) A__ : Dict = Accelerator(project_dir=A__ , project_config=A__ ) A__ , A__ , A__ , A__ : Union[str, Any] = accelerator.prepare( A__ , A__ , A__ , A__ ) accelerator.load_state(os.path.join(A__ , """checkpoints""" , """checkpoint_0""" ) ) ((A__) , (A__)) : Optional[int] = model.a.item(), model.b.item() A__ : Tuple = optimizer.state_dict() self.assertEqual(A__ , A__ ) self.assertEqual(A__ , A__ ) self.assertEqual(A__ , A__ ) A__ : str = train(2 , A__ , A__ , A__ , A__ ) # Save everything accelerator.save_state() # Load everything back in and make sure all states work accelerator.load_state(os.path.join(A__ , """checkpoints""" , """checkpoint_1""" ) ) test_rands += train(1 , A__ , A__ , A__ , A__ ) ((A__) , (A__)) : Optional[int] = model.a.item(), model.b.item() A__ : List[Any] = optimizer.state_dict() self.assertEqual(A__ , A__ ) self.assertEqual(A__ , A__ ) self.assertEqual(A__ , A__ ) self.assertEqual(A__ , A__ ) def __A ( self ): A__ : Union[str, Any] = torch.tensor([1, 2, 3] ) A__ : int = torch.tensor([2, 3, 4] ) A__ : List[Any] = DummyModel() A__ : List[Any] = torch.optim.Adam(net.parameters() ) A__ : Tuple = Accelerator() with self.assertRaises(A__ ) as ve: accelerator.register_for_checkpointing(A__ , A__ , A__ , A__ ) A__ : Any = str(ve.exception ) self.assertTrue("""Item at index 0""" in message ) self.assertTrue("""Item at index 1""" in message ) self.assertFalse("""Item at index 2""" in message ) self.assertFalse("""Item at index 3""" in message ) def __A ( self ): with tempfile.TemporaryDirectory() as tmpdir: set_seed(42 ) A__ : Any = DummyModel() A__ : Union[str, Any] = torch.optim.Adam(params=model.parameters() , lr=1e-3 ) A__ : Dict = torch.optim.lr_scheduler.StepLR(A__ , step_size=1 , gamma=0.9_9 ) A__ , A__ : List[Any] = dummy_dataloaders() A__ : Tuple = ProjectConfiguration(automatic_checkpoint_naming=A__ ) # Train baseline A__ : Optional[Any] = Accelerator(project_dir=A__ , project_config=A__ ) A__ , A__ , A__ , A__ , A__ : Union[str, Any] = accelerator.prepare( A__ , A__ , A__ , A__ , A__ ) # Save initial accelerator.save_state() A__ : Tuple = scheduler.state_dict() train(3 , A__ , A__ , A__ , A__ , A__ ) self.assertNotEqual(A__ , scheduler.state_dict() ) # Load everything back in and make sure all states work accelerator.load_state(os.path.join(A__ , """checkpoints""" , """checkpoint_0""" ) ) self.assertEqual(A__ , scheduler.state_dict() ) def __A ( self ): with tempfile.TemporaryDirectory() as tmpdir: set_seed(42 ) A__ : Optional[Any] = DummyModel() A__ : int = ProjectConfiguration(automatic_checkpoint_naming=A__ , total_limit=2 ) # Train baseline A__ : List[str] = Accelerator(project_dir=A__ , project_config=A__ ) A__ : Union[str, Any] = accelerator.prepare(A__ ) # Save 3 states: for _ in range(11 ): accelerator.save_state() self.assertTrue(not os.path.exists(os.path.join(A__ , """checkpoints""" , """checkpoint_0""" ) ) ) self.assertTrue(os.path.exists(os.path.join(A__ , """checkpoints""" , """checkpoint_9""" ) ) ) self.assertTrue(os.path.exists(os.path.join(A__ , """checkpoints""" , """checkpoint_10""" ) ) ) @require_cuda def __A ( self ): A__ : Dict = ["""torchrun""", F"""--nproc_per_node={torch.cuda.device_count()}""", inspect.getfile(self.__class__ )] execute_subprocess_async(A__ , env=os.environ.copy() ) if __name__ == "__main__": A_ : List[str] = '/tmp/accelerate/state_checkpointing' A_ : Optional[Any] = DummyModel() A_ : Union[str, Any] = torch.optim.Adam(params=model.parameters(), lr=1E-3) A_ : str = torch.optim.lr_scheduler.StepLR(optimizer, step_size=1, gamma=0.99) A_ , A_ : List[Any] = dummy_dataloaders() A_ : int = ProjectConfiguration(automatic_checkpoint_naming=True) # Train baseline A_ : List[str] = Accelerator(project_dir=savedir, project_config=project_config, mixed_precision='no') if accelerator.process_index == 0: if os.path.exists(savedir): shutil.rmtree(savedir) os.makedirs(savedir) A_ , A_ , A_ , A_ , A_ : List[Any] = accelerator.prepare( model, optimizer, train_dataloader, valid_dataloader, scheduler ) A_ , A_ : Dict = accelerator.prepare(model, optimizer) train(3, model, train_dataloader, optimizer, accelerator, scheduler) # Check that the intial optimizer is loaded on the GPU for group in optimizer.param_groups: A_ : str = group['params'][0].device break assert param_device.type == accelerator.device.type A_ : Optional[Any] = model.cpu() accelerator.wait_for_everyone() accelerator.save_state() accelerator.wait_for_everyone() # Check CPU state accelerator.load_state(os.path.join(savedir, 'checkpoints', 'checkpoint_0'), map_location='cpu') for group in optimizer.param_groups: A_ : str = group['params'][0].device break assert ( param_device.type == torch.device('cpu').type ), f"Loaded optimizer states did not match, expected to be loaded on the CPU but got {param_device}" # Check device state model.to(accelerator.device) accelerator.load_state(os.path.join(savedir, 'checkpoints', 'checkpoint_0'), map_location='on_device') for group in optimizer.param_groups: A_ : Tuple = group['params'][0].device break assert ( param_device.type == accelerator.device.type ), f"Loaded optimizer states did not match, expected to be loaded on {accelerator.device} but got {param_device}" # Check error with pytest.raises(TypeError, match='Unsupported optimizer map location passed'): accelerator.load_state(os.path.join(savedir, 'checkpoints', 'checkpoint_0'), map_location='invalid') accelerator.wait_for_everyone() if accelerator.process_index == 0: shutil.rmtree(savedir) accelerator.wait_for_everyone()
64
1
from math import factorial def UpperCamelCase (lowercase_: int = 100 ) -> int: return sum(map(lowercase_ , str(factorial(lowercase_ ) ) ) ) if __name__ == "__main__": print(solution(int(input('Enter the Number: ').strip())))
64
def UpperCamelCase (lowercase_: str , lowercase_: str ) -> bool: A__ : Union[str, Any] = len(lowercase_ ) A__ : List[Any] = len(lowercase_ ) A__ : List[Any] = [[False for _ in range(m + 1 )] for _ in range(n + 1 )] A__ : str = True for i in range(lowercase_ ): for j in range(m + 1 ): if dp[i][j]: if j < m and a[i].upper() == b[j]: A__ : int = True if a[i].islower(): A__ : Dict = True return dp[n][m] if __name__ == "__main__": import doctest doctest.testmod()
64
1
import argparse import math import os import torch from neural_compressor.utils.pytorch import load from PIL import Image from transformers import CLIPTextModel, CLIPTokenizer from diffusers import AutoencoderKL, StableDiffusionPipeline, UNetaDConditionModel def UpperCamelCase () -> Any: A__ : List[str] = argparse.ArgumentParser() parser.add_argument( """-m""" , """--pretrained_model_name_or_path""" , type=lowercase_ , default=lowercase_ , required=lowercase_ , help="""Path to pretrained model or model identifier from huggingface.co/models.""" , ) parser.add_argument( """-c""" , """--caption""" , type=lowercase_ , default="""robotic cat with wings""" , help="""Text used to generate images.""" , ) parser.add_argument( """-n""" , """--images_num""" , type=lowercase_ , default=4 , help="""How much images to generate.""" , ) parser.add_argument( """-s""" , """--seed""" , type=lowercase_ , default=42 , help="""Seed for random process.""" , ) parser.add_argument( """-ci""" , """--cuda_id""" , type=lowercase_ , default=0 , help="""cuda_id.""" , ) A__ : Optional[int] = parser.parse_args() return args def UpperCamelCase (lowercase_: Union[str, Any] , lowercase_: int , lowercase_: Union[str, Any] ) -> Tuple: if not len(lowercase_ ) == rows * cols: raise ValueError("""The specified number of rows and columns are not correct.""" ) A__ , A__ : Union[str, Any] = imgs[0].size A__ : Optional[int] = Image.new("""RGB""" , size=(cols * w, rows * h) ) A__ , A__ : Tuple = grid.size for i, img in enumerate(lowercase_ ): grid.paste(lowercase_ , box=(i % cols * w, i // cols * h) ) return grid def UpperCamelCase (lowercase_: str , lowercase_: Optional[Any]="robotic cat with wings" , lowercase_: Tuple=7.5 , lowercase_: Dict=50 , lowercase_: Optional[Any]=1 , lowercase_: Dict=42 , ) -> int: A__ : List[str] = torch.Generator(pipeline.device ).manual_seed(lowercase_ ) A__ : int = pipeline( lowercase_ , guidance_scale=lowercase_ , num_inference_steps=lowercase_ , generator=lowercase_ , num_images_per_prompt=lowercase_ , ).images A__ : Dict = int(math.sqrt(lowercase_ ) ) A__ : Optional[int] = image_grid(lowercase_ , rows=_rows , cols=num_images_per_prompt // _rows ) return grid, images A_ : List[str] = parse_args() # Load models and create wrapper for stable diffusion A_ : List[str] = CLIPTokenizer.from_pretrained(args.pretrained_model_name_or_path, subfolder='tokenizer') A_ : List[str] = CLIPTextModel.from_pretrained(args.pretrained_model_name_or_path, subfolder='text_encoder') A_ : Dict = AutoencoderKL.from_pretrained(args.pretrained_model_name_or_path, subfolder='vae') A_ : int = UNetaDConditionModel.from_pretrained(args.pretrained_model_name_or_path, subfolder='unet') A_ : Dict = StableDiffusionPipeline.from_pretrained( args.pretrained_model_name_or_path, text_encoder=text_encoder, vae=vae, unet=unet, tokenizer=tokenizer ) A_ : str = lambda images, clip_input: (images, False) if os.path.exists(os.path.join(args.pretrained_model_name_or_path, 'best_model.pt')): A_ : int = load(args.pretrained_model_name_or_path, model=unet) unet.eval() setattr(pipeline, 'unet', unet) else: A_ : Union[str, Any] = unet.to(torch.device('cuda', args.cuda_id)) A_ : int = pipeline.to(unet.device) A_ , A_ : List[str] = generate_images(pipeline, prompt=args.caption, num_images_per_prompt=args.images_num, seed=args.seed) grid.save(os.path.join(args.pretrained_model_name_or_path, '{}.png'.format('_'.join(args.caption.split())))) A_ : Optional[Any] = os.path.join(args.pretrained_model_name_or_path, '_'.join(args.caption.split())) os.makedirs(dirname, exist_ok=True) for idx, image in enumerate(images): image.save(os.path.join(dirname, '{}.png'.format(idx + 1)))
64
import itertools import random import unittest import numpy as np from transformers import ASTFeatureExtractor from transformers.testing_utils import require_torch, require_torchaudio from transformers.utils.import_utils import is_torch_available from ...test_sequence_feature_extraction_common import SequenceFeatureExtractionTestMixin A_ : Dict = random.Random() if is_torch_available(): import torch def UpperCamelCase (lowercase_: Tuple , lowercase_: Tuple=1.0 , lowercase_: Dict=None , lowercase_: int=None ) -> str: if rng is None: A__ : Optional[Any] = global_rng A__ : List[str] = [] for batch_idx in range(shape[0] ): values.append([] ) for _ in range(shape[1] ): values[-1].append(rng.random() * scale ) return values class _a (unittest.TestCase ): '''simple docstring''' def __init__( self , A__ , A__=7 , A__=400 , A__=2000 , A__=1 , A__=0.0 , A__=1_6000 , A__=True , A__=True , ): A__ : Any = parent A__ : Optional[int] = batch_size A__ : Union[str, Any] = min_seq_length A__ : Dict = max_seq_length A__ : Tuple = (self.max_seq_length - self.min_seq_length) // (self.batch_size - 1) A__ : str = feature_size A__ : Optional[int] = padding_value A__ : List[str] = sampling_rate A__ : List[str] = return_attention_mask A__ : int = do_normalize def __A ( self ): return { "feature_size": self.feature_size, "padding_value": self.padding_value, "sampling_rate": self.sampling_rate, "return_attention_mask": self.return_attention_mask, "do_normalize": self.do_normalize, } def __A ( self , A__=False , A__=False ): def _flatten(A__ ): return list(itertools.chain(*A__ ) ) if equal_length: A__ : Dict = floats_list((self.batch_size, self.max_seq_length) ) else: # make sure that inputs increase in size A__ : Union[str, Any] = [ _flatten(floats_list((x, self.feature_size) ) ) for x in range(self.min_seq_length , self.max_seq_length , self.seq_length_diff ) ] if numpify: A__ : Optional[int] = [np.asarray(A__ ) for x in speech_inputs] return speech_inputs @require_torch @require_torchaudio class _a (__magic_name__ , unittest.TestCase ): '''simple docstring''' UpperCAmelCase__: int = ASTFeatureExtractor def __A ( self ): A__ : Optional[Any] = ASTFeatureExtractionTester(self ) def __A ( self ): # Tests that all call wrap to encode_plus and batch_encode_plus A__ : Optional[int] = self.feature_extraction_class(**self.feat_extract_tester.prepare_feat_extract_dict() ) # create three inputs of length 800, 1000, and 1200 A__ : List[Any] = [floats_list((1, x) )[0] for x in range(800 , 1400 , 200 )] A__ : Optional[Any] = [np.asarray(A__ ) for speech_input in speech_inputs] # Test not batched input A__ : Tuple = feat_extract(speech_inputs[0] , return_tensors="""np""" ).input_values A__ : Tuple = feat_extract(np_speech_inputs[0] , return_tensors="""np""" ).input_values self.assertTrue(np.allclose(A__ , A__ , atol=1e-3 ) ) # Test batched A__ : Tuple = feat_extract(A__ , padding=A__ , return_tensors="""np""" ).input_values A__ : Tuple = feat_extract(A__ , padding=A__ , return_tensors="""np""" ).input_values for enc_seq_a, enc_seq_a in zip(A__ , A__ ): self.assertTrue(np.allclose(A__ , A__ , atol=1e-3 ) ) # Test 2-D numpy arrays are batched. A__ : int = [floats_list((1, x) )[0] for x in (800, 800, 800)] A__ : List[str] = np.asarray(A__ ) A__ : Union[str, Any] = feat_extract(A__ , return_tensors="""np""" ).input_values A__ : Optional[Any] = feat_extract(A__ , return_tensors="""np""" ).input_values for enc_seq_a, enc_seq_a in zip(A__ , A__ ): self.assertTrue(np.allclose(A__ , A__ , atol=1e-3 ) ) @require_torch def __A ( self ): import torch A__ : Union[str, Any] = self.feature_extraction_class(**self.feat_extract_tester.prepare_feat_extract_dict() ) A__ : Tuple = np.random.rand(100 ).astype(np.floataa ) A__ : Tuple = np_speech_inputs.tolist() for inputs in [py_speech_inputs, np_speech_inputs]: A__ : List[str] = feature_extractor.pad([{"""input_values""": inputs}] , return_tensors="""np""" ) self.assertTrue(np_processed.input_values.dtype == np.floataa ) A__ : Any = feature_extractor.pad([{"""input_values""": inputs}] , return_tensors="""pt""" ) self.assertTrue(pt_processed.input_values.dtype == torch.floataa ) def __A ( self , A__ ): from datasets import load_dataset A__ : str = load_dataset("""hf-internal-testing/librispeech_asr_dummy""" , """clean""" , split="""validation""" ) # automatic decoding with librispeech A__ : str = ds.sort("""id""" ).select(range(A__ ) )[:num_samples]["""audio"""] return [x["array"] for x in speech_samples] @require_torch def __A ( self ): # fmt: off A__ : Optional[Any] = torch.tensor( [-0.9_8_9_4, -1.2_7_7_6, -0.9_0_6_6, -1.2_7_7_6, -0.9_3_4_9, -1.2_6_0_9, -1.0_3_8_6, -1.2_7_7_6, -1.1_5_6_1, -1.2_7_7_6, -1.2_0_5_2, -1.2_7_2_3, -1.2_1_9_0, -1.2_1_3_2, -1.2_7_7_6, -1.1_1_3_3, -1.1_9_5_3, -1.1_3_4_3, -1.1_5_8_4, -1.2_2_0_3, -1.1_7_7_0, -1.2_4_7_4, -1.2_3_8_1, -1.1_9_3_6, -0.9_2_7_0, -0.8_3_1_7, -0.8_0_4_9, -0.7_7_0_6, -0.7_5_6_5, -0.7_8_6_9] ) # fmt: on A__ : Any = self._load_datasamples(1 ) A__ : Tuple = ASTFeatureExtractor() A__ : Dict = feature_extractor(A__ , return_tensors="""pt""" ).input_values self.assertEquals(input_values.shape , (1, 1024, 128) ) self.assertTrue(torch.allclose(input_values[0, 0, :30] , A__ , atol=1e-4 ) )
64
1
import argparse import torch from transformers import ( UniSpeechSatConfig, UniSpeechSatForAudioFrameClassification, UniSpeechSatForSequenceClassification, UniSpeechSatForXVector, WavaVecaFeatureExtractor, logging, ) logging.set_verbosity_info() A_ : Union[str, Any] = logging.get_logger(__name__) def UpperCamelCase (lowercase_: Optional[int] , lowercase_: Dict , lowercase_: Tuple ) -> Dict: A__ : Tuple = UniSpeechSatForSequenceClassification.from_pretrained(lowercase_ , config=lowercase_ ) A__ : Dict = downstream_dict["""projector.weight"""] A__ : str = downstream_dict["""projector.bias"""] A__ : str = downstream_dict["""model.post_net.linear.weight"""] A__ : Optional[int] = downstream_dict["""model.post_net.linear.bias"""] return model def UpperCamelCase (lowercase_: Any , lowercase_: Dict , lowercase_: Optional[int] ) -> Dict: A__ : Any = UniSpeechSatForAudioFrameClassification.from_pretrained(lowercase_ , config=lowercase_ ) A__ : Union[str, Any] = downstream_dict["""model.linear.weight"""] A__ : Tuple = downstream_dict["""model.linear.bias"""] return model def UpperCamelCase (lowercase_: Any , lowercase_: Tuple , lowercase_: int ) -> Optional[Any]: A__ : List[Any] = UniSpeechSatForXVector.from_pretrained(lowercase_ , config=lowercase_ ) A__ : str = downstream_dict["""connector.weight"""] A__ : List[Any] = downstream_dict["""connector.bias"""] for i, kernel_size in enumerate(hf_config.tdnn_kernel ): A__ : Optional[int] = downstream_dict[ f"""model.framelevel_feature_extractor.module.{i}.kernel.weight""" ] A__ : Dict = downstream_dict[f"""model.framelevel_feature_extractor.module.{i}.kernel.bias"""] A__ : List[str] = downstream_dict["""model.utterancelevel_feature_extractor.linear1.weight"""] A__ : List[Any] = downstream_dict["""model.utterancelevel_feature_extractor.linear1.bias"""] A__ : Optional[int] = downstream_dict["""model.utterancelevel_feature_extractor.linear2.weight"""] A__ : List[Any] = downstream_dict["""model.utterancelevel_feature_extractor.linear2.bias"""] A__ : List[Any] = downstream_dict["""objective.W"""] return model @torch.no_grad() def UpperCamelCase (lowercase_: Dict , lowercase_: Tuple , lowercase_: List[Any] , lowercase_: Optional[Any] ) -> int: A__ : List[str] = torch.load(lowercase_ , map_location="""cpu""" ) A__ : Any = checkpoint["""Downstream"""] A__ : Optional[Any] = UniSpeechSatConfig.from_pretrained(lowercase_ ) A__ : Optional[int] = WavaVecaFeatureExtractor.from_pretrained( lowercase_ , return_attention_mask=lowercase_ , do_normalize=lowercase_ ) A__ : str = hf_config.architectures[0] if arch.endswith("""ForSequenceClassification""" ): A__ : List[Any] = convert_classification(lowercase_ , lowercase_ , lowercase_ ) elif arch.endswith("""ForAudioFrameClassification""" ): A__ : List[Any] = convert_diarization(lowercase_ , lowercase_ , lowercase_ ) elif arch.endswith("""ForXVector""" ): A__ : str = convert_xvector(lowercase_ , lowercase_ , lowercase_ ) else: raise NotImplementedError(f"""S3PRL weights conversion is not supported for {arch}""" ) if hf_config.use_weighted_layer_sum: A__ : Optional[Any] = checkpoint["""Featurizer"""]["""weights"""] hf_feature_extractor.save_pretrained(lowercase_ ) hf_model.save_pretrained(lowercase_ ) if __name__ == "__main__": A_ : Optional[Any] = argparse.ArgumentParser() parser.add_argument( '--base_model_name', default=None, type=str, help='Name of the huggingface pretrained base model.' ) parser.add_argument('--config_path', default=None, type=str, help='Path to the huggingface classifier config.') parser.add_argument('--checkpoint_path', default=None, type=str, help='Path to the s3prl checkpoint.') parser.add_argument('--model_dump_path', default=None, type=str, help='Path to the final converted model.') A_ : Optional[Any] = parser.parse_args() convert_saprl_checkpoint(args.base_model_name, args.config_path, args.checkpoint_path, args.model_dump_path)
64
from typing import Optional import numpy as np import torch from torch import nn from transformers import GPTaConfig, GPTaLMHeadModel from transformers.modeling_utils import ModuleUtilsMixin from ...configuration_utils import ConfigMixin, register_to_config from ...models import ModelMixin class _a (__magic_name__ , __magic_name__ , __magic_name__ ): '''simple docstring''' UpperCAmelCase__: str = [r'''h\.\d+\.attn\.bias''', r'''h\.\d+\.attn\.masked_bias'''] @register_to_config def __init__( self , A__ , A__ , A__ = None , A__ = 5_0257 , A__ = 1024 , A__ = 768 , A__ = 12 , A__ = 12 , A__ = None , A__ = "gelu_new" , A__ = 0.1 , A__ = 0.1 , A__ = 0.1 , A__ = 1e-5 , A__ = 0.0_2 , A__ = True , A__ = True , A__ = False , A__ = False , ): super().__init__() A__ : Union[str, Any] = prefix_length if prefix_inner_dim != n_embd and prefix_hidden_dim is None: raise ValueError( F"""`prefix_hidden_dim` cannot be `None` when `prefix_inner_dim`: {prefix_hidden_dim} and""" F""" `n_embd`: {n_embd} are not equal.""" ) A__ : str = prefix_inner_dim A__ : Optional[Any] = prefix_hidden_dim A__ : Tuple = ( nn.Linear(self.prefix_inner_dim , self.prefix_hidden_dim ) if self.prefix_hidden_dim is not None else nn.Identity() ) A__ : int = ( nn.Linear(self.prefix_hidden_dim , A__ ) if self.prefix_hidden_dim is not None else nn.Identity() ) A__ : Tuple = GPTaConfig( vocab_size=A__ , n_positions=A__ , n_embd=A__ , n_layer=A__ , n_head=A__ , n_inner=A__ , activation_function=A__ , resid_pdrop=A__ , embd_pdrop=A__ , attn_pdrop=A__ , layer_norm_epsilon=A__ , initializer_range=A__ , scale_attn_weights=A__ , use_cache=A__ , scale_attn_by_inverse_layer_idx=A__ , reorder_and_upcast_attn=A__ , ) A__ : int = GPTaLMHeadModel(A__ ) def __A ( self , A__ , A__ , A__ = None , A__ = None , ): A__ : List[str] = self.transformer.transformer.wte(A__ ) A__ : int = self.encode_prefix(A__ ) A__ : int = self.decode_prefix(A__ ) A__ : Optional[Any] = torch.cat((prefix_embeds, embedding_text) , dim=1 ) if labels is not None: A__ : Any = self.get_dummy_token(input_ids.shape[0] , input_ids.device ) A__ : List[Any] = torch.cat((dummy_token, input_ids) , dim=1 ) A__ : List[str] = self.transformer(inputs_embeds=A__ , labels=A__ , attention_mask=A__ ) if self.prefix_hidden_dim is not None: return out, hidden else: return out def __A ( self , A__ , A__ ): return torch.zeros(A__ , self.prefix_length , dtype=torch.intaa , device=A__ ) def __A ( self , A__ ): return self.encode_prefix(A__ ) @torch.no_grad() def __A ( self , A__ , A__ , A__ ): A__ : List[Any] = torch.split(A__ , 1 , dim=0 ) A__ : Optional[int] = [] A__ : str = [] for feature in features: A__ : Dict = self.decode_prefix(feature.to(A__ ) ) # back to the clip feature # Only support beam search for now A__ , A__ : Union[str, Any] = self.generate_beam( input_embeds=A__ , device=A__ , eos_token_id=A__ ) generated_tokens.append(output_tokens[0] ) generated_seq_lengths.append(seq_lengths[0] ) A__ : int = torch.stack(A__ ) A__ : List[Any] = torch.stack(A__ ) return generated_tokens, generated_seq_lengths @torch.no_grad() def __A ( self , A__=None , A__=None , A__=None , A__ = 5 , A__ = 67 , A__ = 1.0 , A__ = None , ): A__ : Any = eos_token_id A__ : Any = None A__ : Optional[int] = None A__ : Optional[Any] = torch.ones(A__ , device=A__ , dtype=torch.int ) A__ : Any = torch.zeros(A__ , device=A__ , dtype=torch.bool ) if input_embeds is not None: A__ : Dict = input_embeds else: A__ : str = self.transformer.transformer.wte(A__ ) for i in range(A__ ): A__ : Dict = self.transformer(inputs_embeds=A__ ) A__ : str = outputs.logits A__ : Union[str, Any] = logits[:, -1, :] / (temperature if temperature > 0 else 1.0) A__ : Any = logits.softmax(-1 ).log() if scores is None: A__ , A__ : Optional[int] = logits.topk(A__ , -1 ) A__ : List[Any] = generated.expand(A__ , *generated.shape[1:] ) A__ , A__ : List[Any] = next_tokens.permute(1 , 0 ), scores.squeeze(0 ) if tokens is None: A__ : Optional[Any] = next_tokens else: A__ : List[Any] = tokens.expand(A__ , *tokens.shape[1:] ) A__ : int = torch.cat((tokens, next_tokens) , dim=1 ) else: A__ : Optional[int] = -float(np.inf ) A__ : List[Any] = 0 A__ : str = scores[:, None] + logits seq_lengths[~is_stopped] += 1 A__ : Dict = scores_sum / seq_lengths[:, None] A__ , A__ : List[Any] = scores_sum_average.view(-1 ).topk(A__ , -1 ) A__ : Tuple = next_tokens // scores_sum.shape[1] A__ : Optional[Any] = seq_lengths[next_tokens_source] A__ : List[str] = next_tokens % scores_sum.shape[1] A__ : Optional[int] = next_tokens.unsqueeze(1 ) A__ : int = tokens[next_tokens_source] A__ : List[Any] = torch.cat((tokens, next_tokens) , dim=1 ) A__ : str = generated[next_tokens_source] A__ : Optional[Any] = scores_sum_average * seq_lengths A__ : Union[str, Any] = is_stopped[next_tokens_source] A__ : str = self.transformer.transformer.wte(next_tokens.squeeze() ).view(generated.shape[0] , 1 , -1 ) A__ : Optional[int] = torch.cat((generated, next_token_embed) , dim=1 ) A__ : List[str] = is_stopped + next_tokens.eq(A__ ).squeeze() if is_stopped.all(): break A__ : Dict = scores / seq_lengths A__ : Dict = scores.argsort(descending=A__ ) # tokens tensors are already padded to max_seq_length A__ : Union[str, Any] = [tokens[i] for i in order] A__ : Any = torch.stack(A__ , dim=0 ) A__ : Dict = torch.tensor([seq_lengths[i] for i in order] , dtype=seq_lengths.dtype ) return output_texts, seq_lengths
64
1
import inspect import warnings from typing import Any, Dict, Optional, Union from packaging import version def UpperCamelCase (*lowercase_: Optional[int] , lowercase_: Optional[Union[Dict, Any]] = None , lowercase_: Dict=True , lowercase_: Tuple=2 ) -> Dict: from .. import __version__ A__ : Dict = take_from A__ : str = () if not isinstance(args[0] , lowercase_ ): A__ : int = (args,) for attribute, version_name, message in args: if version.parse(version.parse(lowercase_ ).base_version ) >= version.parse(lowercase_ ): raise ValueError( f"""The deprecation tuple {(attribute, version_name, message)} should be removed since diffusers'""" f""" version {__version__} is >= {version_name}""" ) A__ : Any = None if isinstance(lowercase_ , lowercase_ ) and attribute in deprecated_kwargs: values += (deprecated_kwargs.pop(lowercase_ ),) A__ : List[str] = f"""The `{attribute}` argument is deprecated and will be removed in version {version_name}.""" elif hasattr(lowercase_ , lowercase_ ): values += (getattr(lowercase_ , lowercase_ ),) A__ : Optional[Any] = f"""The `{attribute}` attribute is deprecated and will be removed in version {version_name}.""" elif deprecated_kwargs is None: A__ : int = f"""`{attribute}` is deprecated and will be removed in version {version_name}.""" if warning is not None: A__ : int = warning + """ """ if standard_warn else """""" warnings.warn(warning + message , lowercase_ , stacklevel=lowercase_ ) if isinstance(lowercase_ , lowercase_ ) and len(lowercase_ ) > 0: A__ : Union[str, Any] = inspect.getouterframes(inspect.currentframe() )[1] A__ : Optional[Any] = call_frame.filename A__ : Optional[int] = call_frame.lineno A__ : Any = call_frame.function A__ , A__ : List[str] = next(iter(deprecated_kwargs.items() ) ) raise TypeError(f"""{function} in {filename} line {line_number-1} got an unexpected keyword argument `{key}`""" ) if len(lowercase_ ) == 0: return elif len(lowercase_ ) == 1: return values[0] return values
64
import io import itertools import json from dataclasses import dataclass from typing import Optional import pyarrow as pa import pyarrow.json as paj import datasets from datasets.table import table_cast from datasets.utils.file_utils import readline A_ : Tuple = datasets.utils.logging.get_logger(__name__) @dataclass class _a (datasets.BuilderConfig ): '''simple docstring''' UpperCAmelCase__: Optional[datasets.Features] = None UpperCAmelCase__: str = "utf-8" UpperCAmelCase__: Optional[str] = None UpperCAmelCase__: Optional[str] = None UpperCAmelCase__: bool = True # deprecated UpperCAmelCase__: Optional[int] = None # deprecated UpperCAmelCase__: int = 10 << 20 # 10MB UpperCAmelCase__: Optional[bool] = None class _a (datasets.ArrowBasedBuilder ): '''simple docstring''' UpperCAmelCase__: List[str] = JsonConfig def __A ( self ): if self.config.block_size is not None: logger.warning("""The JSON loader parameter `block_size` is deprecated. Please use `chunksize` instead""" ) A__ : Union[str, Any] = self.config.block_size if self.config.use_threads is not True: logger.warning( """The JSON loader parameter `use_threads` is deprecated and doesn't have any effect anymore.""" ) if self.config.newlines_in_values is not None: raise ValueError("""The JSON loader parameter `newlines_in_values` is no longer supported""" ) return datasets.DatasetInfo(features=self.config.features ) def __A ( self , A__ ): if not self.config.data_files: raise ValueError(F"""At least one data file must be specified, but got data_files={self.config.data_files}""" ) A__ : int = dl_manager.download_and_extract(self.config.data_files ) if isinstance(A__ , (str, list, tuple) ): A__ : Optional[Any] = data_files if isinstance(A__ , A__ ): A__ : List[str] = [files] A__ : int = [dl_manager.iter_files(A__ ) for file in files] return [datasets.SplitGenerator(name=datasets.Split.TRAIN , gen_kwargs={"""files""": files} )] A__ : List[str] = [] for split_name, files in data_files.items(): if isinstance(A__ , A__ ): A__ : Optional[int] = [files] A__ : Optional[int] = [dl_manager.iter_files(A__ ) for file in files] splits.append(datasets.SplitGenerator(name=A__ , gen_kwargs={"""files""": files} ) ) return splits def __A ( self , A__ ): if self.config.features is not None: # adding missing columns for column_name in set(self.config.features ) - set(pa_table.column_names ): A__ : Optional[Any] = self.config.features.arrow_schema.field(A__ ).type A__ : str = pa_table.append_column(A__ , pa.array([None] * len(A__ ) , type=A__ ) ) # more expensive cast to support nested structures with keys in a different order # allows str <-> int/float or str to Audio for example A__ : Optional[int] = table_cast(A__ , self.config.features.arrow_schema ) return pa_table def __A ( self , A__ ): for file_idx, file in enumerate(itertools.chain.from_iterable(A__ ) ): # If the file is one json object and if we need to look at the list of items in one specific field if self.config.field is not None: with open(A__ , encoding=self.config.encoding , errors=self.config.encoding_errors ) as f: A__ : Optional[Any] = json.load(A__ ) # We keep only the field we are interested in A__ : Optional[int] = dataset[self.config.field] # We accept two format: a list of dicts or a dict of lists if isinstance(A__ , (list, tuple) ): A__ : Union[str, Any] = set().union(*[row.keys() for row in dataset] ) A__ : Any = {col: [row.get(A__ ) for row in dataset] for col in keys} else: A__ : Any = dataset A__ : Any = pa.Table.from_pydict(A__ ) yield file_idx, self._cast_table(A__ ) # If the file has one json object per line else: with open(A__ , """rb""" ) as f: A__ : List[str] = 0 # Use block_size equal to the chunk size divided by 32 to leverage multithreading # Set a default minimum value of 16kB if the chunk size is really small A__ : List[str] = max(self.config.chunksize // 32 , 16 << 10 ) A__ : Any = ( self.config.encoding_errors if self.config.encoding_errors is not None else """strict""" ) while True: A__ : Dict = f.read(self.config.chunksize ) if not batch: break # Finish current line try: batch += f.readline() except (AttributeError, io.UnsupportedOperation): batch += readline(A__ ) # PyArrow only accepts utf-8 encoded bytes if self.config.encoding != "utf-8": A__ : List[Any] = batch.decode(self.config.encoding , errors=A__ ).encode("""utf-8""" ) try: while True: try: A__ : str = paj.read_json( io.BytesIO(A__ ) , read_options=paj.ReadOptions(block_size=A__ ) ) break except (pa.ArrowInvalid, pa.ArrowNotImplementedError) as e: if ( isinstance(A__ , pa.ArrowInvalid ) and "straddling" not in str(A__ ) or block_size > len(A__ ) ): raise else: # Increase the block size in case it was too small. # The block size will be reset for the next file. logger.debug( F"""Batch of {len(A__ )} bytes couldn't be parsed with block_size={block_size}. Retrying with block_size={block_size * 2}.""" ) block_size *= 2 except pa.ArrowInvalid as e: try: with open( A__ , encoding=self.config.encoding , errors=self.config.encoding_errors ) as f: A__ : Optional[Any] = json.load(A__ ) except json.JSONDecodeError: logger.error(F"""Failed to read file '{file}' with error {type(A__ )}: {e}""" ) raise e # If possible, parse the file as a list of json objects and exit the loop if isinstance(A__ , A__ ): # list is the only sequence type supported in JSON try: A__ : str = set().union(*[row.keys() for row in dataset] ) A__ : List[str] = {col: [row.get(A__ ) for row in dataset] for col in keys} A__ : int = pa.Table.from_pydict(A__ ) except (pa.ArrowInvalid, AttributeError) as e: logger.error(F"""Failed to read file '{file}' with error {type(A__ )}: {e}""" ) raise ValueError(F"""Not able to read records in the JSON file at {file}.""" ) from None yield file_idx, self._cast_table(A__ ) break else: logger.error(F"""Failed to read file '{file}' with error {type(A__ )}: {e}""" ) raise ValueError( F"""Not able to read records in the JSON file at {file}. """ F"""You should probably indicate the field of the JSON file containing your records. """ F"""This JSON file contain the following fields: {str(list(dataset.keys() ) )}. """ F"""Select the correct one and provide it as `field='XXX'` to the dataset loading method. """ ) from None # Uncomment for debugging (will print the Arrow table size and elements) # logger.warning(f"pa_table: {pa_table} num rows: {pa_table.num_rows}") # logger.warning('\n'.join(str(pa_table.slice(i, 1).to_pydict()) for i in range(pa_table.num_rows))) yield (file_idx, batch_idx), self._cast_table(A__ ) batch_idx += 1
64
1
import numpy as np import torch from torch.utils.data import Dataset from utils import logger class _a (__magic_name__ ): '''simple docstring''' def __init__( self , A__ , A__ ): A__ : int = params A__ : Tuple = np.array(A__ ) A__ : Tuple = np.array([len(A__ ) for t in data] ) self.check() self.remove_long_sequences() self.remove_empty_sequences() self.remove_unknown_sequences() self.check() self.print_statistics() def __getitem__( self , A__ ): return (self.token_ids[index], self.lengths[index]) def __len__( self ): return len(self.lengths ) def __A ( self ): assert len(self.token_ids ) == len(self.lengths ) assert all(self.lengths[i] == len(self.token_ids[i] ) for i in range(len(self.lengths ) ) ) def __A ( self ): A__ : List[str] = self.params.max_model_input_size A__ : Optional[Any] = self.lengths > max_len logger.info(F"""Splitting {sum(A__ )} too long sequences.""" ) def divide_chunks(A__ , A__ ): return [l[i : i + n] for i in range(0 , len(A__ ) , A__ )] A__ : Dict = [] A__ : Optional[Any] = [] if self.params.mlm: A__ , A__ : Any = self.params.special_tok_ids["""cls_token"""], self.params.special_tok_ids["""sep_token"""] else: A__ , A__ : Tuple = self.params.special_tok_ids["""bos_token"""], self.params.special_tok_ids["""eos_token"""] for seq_, len_ in zip(self.token_ids , self.lengths ): assert (seq_[0] == cls_id) and (seq_[-1] == sep_id), seq_ if len_ <= max_len: new_tok_ids.append(seq_ ) new_lengths.append(len_ ) else: A__ : List[Any] = [] for sub_s in divide_chunks(seq_ , max_len - 2 ): if sub_s[0] != cls_id: A__ : List[Any] = np.insert(A__ , 0 , A__ ) if sub_s[-1] != sep_id: A__ : Union[str, Any] = np.insert(A__ , len(A__ ) , A__ ) assert len(A__ ) <= max_len assert (sub_s[0] == cls_id) and (sub_s[-1] == sep_id), sub_s sub_seqs.append(A__ ) new_tok_ids.extend(A__ ) new_lengths.extend([len(A__ ) for l in sub_seqs] ) A__ : int = np.array(A__ ) A__ : List[str] = np.array(A__ ) def __A ( self ): A__ : Any = len(self ) A__ : Tuple = self.lengths > 11 A__ : List[Any] = self.token_ids[indices] A__ : str = self.lengths[indices] A__ : str = len(self ) logger.info(F"""Remove {init_size - new_size} too short (<=11 tokens) sequences.""" ) def __A ( self ): if "unk_token" not in self.params.special_tok_ids: return else: A__ : List[str] = self.params.special_tok_ids["""unk_token"""] A__ : List[str] = len(self ) A__ : Dict = np.array([np.count_nonzero(a == unk_token_id ) for a in self.token_ids] ) A__ : Optional[int] = (unk_occs / self.lengths) < 0.5 A__ : Union[str, Any] = self.token_ids[indices] A__ : Optional[int] = self.lengths[indices] A__ : int = len(self ) logger.info(F"""Remove {init_size - new_size} sequences with a high level of unknown tokens (50%).""" ) def __A ( self ): if not self.params.is_master: return logger.info(F"""{len(self )} sequences""" ) # data_len = sum(self.lengths) # nb_unique_tokens = len(Counter(list(chain(*self.token_ids)))) # logger.info(f'{data_len} tokens ({nb_unique_tokens} unique)') # unk_idx = self.params.special_tok_ids['unk_token'] # nb_unknown = sum([(t==unk_idx).sum() for t in self.token_ids]) # logger.info(f'{nb_unknown} unknown tokens (covering {100*nb_unknown/data_len:.2f}% of the data)') def __A ( self , A__ ): A__ : Tuple = [t[0] for t in batch] A__ : Any = [t[1] for t in batch] assert len(A__ ) == len(A__ ) # Max for paddings A__ : List[str] = max(A__ ) # Pad token ids if self.params.mlm: A__ : str = self.params.special_tok_ids["""pad_token"""] else: A__ : List[str] = self.params.special_tok_ids["""unk_token"""] A__ : int = [list(t.astype(A__ ) ) + [pad_idx] * (max_seq_len_ - len(A__ )) for t in token_ids] assert len(tk_ ) == len(A__ ) assert all(len(A__ ) == max_seq_len_ for t in tk_ ) A__ : Any = torch.tensor(tk_ ) # (bs, max_seq_len_) A__ : Any = torch.tensor(A__ ) # (bs) return tk_t, lg_t
64
import os try: from .build_directory_md import good_file_paths except ImportError: from build_directory_md import good_file_paths # type: ignore A_ : Dict = list(good_file_paths()) assert filepaths, "good_file_paths() failed!" A_ : Optional[Any] = [file for file in filepaths if file != file.lower()] if upper_files: print(f'''{len(upper_files)} files contain uppercase characters:''') print('\n'.join(upper_files) + '\n') A_ : Tuple = [file for file in filepaths if ' ' in file] if space_files: print(f'''{len(space_files)} files contain space characters:''') print('\n'.join(space_files) + '\n') A_ : Any = [file for file in filepaths if '-' in file] if hyphen_files: print(f'''{len(hyphen_files)} files contain hyphen characters:''') print('\n'.join(hyphen_files) + '\n') A_ : List[str] = [file for file in filepaths if os.sep not in file] if nodir_files: print(f'''{len(nodir_files)} files are not in a directory:''') print('\n'.join(nodir_files) + '\n') A_ : Any = len(upper_files + space_files + hyphen_files + nodir_files) if bad_files: import sys sys.exit(bad_files)
64
1
import json import os import shutil import sys import tempfile import unittest import unittest.mock as mock from pathlib import Path from huggingface_hub import HfFolder, delete_repo from requests.exceptions import HTTPError from transformers import AutoConfig, BertConfig, GPTaConfig from transformers.configuration_utils import PretrainedConfig from transformers.testing_utils import TOKEN, USER, is_staging_test sys.path.append(str(Path(__file__).parent.parent / 'utils')) from test_module.custom_configuration import CustomConfig # noqa E402 A_ : Tuple = { 'return_dict': False, 'output_hidden_states': True, 'output_attentions': True, 'torchscript': True, 'torch_dtype': 'float16', 'use_bfloat16': True, 'tf_legacy_loss': True, 'pruned_heads': {'a': 1}, 'tie_word_embeddings': False, 'is_decoder': True, 'cross_attention_hidden_size': 128, 'add_cross_attention': True, 'tie_encoder_decoder': True, 'max_length': 50, 'min_length': 3, 'do_sample': True, 'early_stopping': True, 'num_beams': 3, 'num_beam_groups': 3, 'diversity_penalty': 0.5, 'temperature': 2.0, 'top_k': 10, 'top_p': 0.7, 'typical_p': 0.2, 'repetition_penalty': 0.8, 'length_penalty': 0.8, 'no_repeat_ngram_size': 5, 'encoder_no_repeat_ngram_size': 5, 'bad_words_ids': [1, 2, 3], 'num_return_sequences': 3, 'chunk_size_feed_forward': 5, 'output_scores': True, 'return_dict_in_generate': True, 'forced_bos_token_id': 2, 'forced_eos_token_id': 3, 'remove_invalid_values': True, 'architectures': ['BertModel'], 'finetuning_task': 'translation', 'id2label': {0: 'label'}, 'label2id': {'label': '0'}, 'tokenizer_class': 'BertTokenizerFast', 'prefix': 'prefix', 'bos_token_id': 6, 'pad_token_id': 7, 'eos_token_id': 8, 'sep_token_id': 9, 'decoder_start_token_id': 10, 'exponential_decay_length_penalty': (5, 1.01), 'suppress_tokens': [0, 1], 'begin_suppress_tokens': 2, 'task_specific_params': {'translation': 'some_params'}, 'problem_type': 'regression', } @is_staging_test class _a (unittest.TestCase ): '''simple docstring''' @classmethod def __A ( cls ): A__ : List[str] = TOKEN HfFolder.save_token(A__ ) @classmethod def __A ( cls ): try: delete_repo(token=cls._token , repo_id="""test-config""" ) except HTTPError: pass try: delete_repo(token=cls._token , repo_id="""valid_org/test-config-org""" ) except HTTPError: pass try: delete_repo(token=cls._token , repo_id="""test-dynamic-config""" ) except HTTPError: pass def __A ( self ): A__ : int = BertConfig( vocab_size=99 , hidden_size=32 , num_hidden_layers=5 , num_attention_heads=4 , intermediate_size=37 ) config.push_to_hub("""test-config""" , use_auth_token=self._token ) A__ : Optional[Any] = BertConfig.from_pretrained(F"""{USER}/test-config""" ) for k, v in config.to_dict().items(): if k != "transformers_version": self.assertEqual(A__ , getattr(A__ , A__ ) ) # Reset repo delete_repo(token=self._token , repo_id="""test-config""" ) # Push to hub via save_pretrained with tempfile.TemporaryDirectory() as tmp_dir: config.save_pretrained(A__ , repo_id="""test-config""" , push_to_hub=A__ , use_auth_token=self._token ) A__ : str = BertConfig.from_pretrained(F"""{USER}/test-config""" ) for k, v in config.to_dict().items(): if k != "transformers_version": self.assertEqual(A__ , getattr(A__ , A__ ) ) def __A ( self ): A__ : Union[str, Any] = BertConfig( vocab_size=99 , hidden_size=32 , num_hidden_layers=5 , num_attention_heads=4 , intermediate_size=37 ) config.push_to_hub("""valid_org/test-config-org""" , use_auth_token=self._token ) A__ : Optional[Any] = BertConfig.from_pretrained("""valid_org/test-config-org""" ) for k, v in config.to_dict().items(): if k != "transformers_version": self.assertEqual(A__ , getattr(A__ , A__ ) ) # Reset repo delete_repo(token=self._token , repo_id="""valid_org/test-config-org""" ) # Push to hub via save_pretrained with tempfile.TemporaryDirectory() as tmp_dir: config.save_pretrained( A__ , repo_id="""valid_org/test-config-org""" , push_to_hub=A__ , use_auth_token=self._token ) A__ : str = BertConfig.from_pretrained("""valid_org/test-config-org""" ) for k, v in config.to_dict().items(): if k != "transformers_version": self.assertEqual(A__ , getattr(A__ , A__ ) ) def __A ( self ): CustomConfig.register_for_auto_class() A__ : Optional[Any] = CustomConfig(attribute=42 ) config.push_to_hub("""test-dynamic-config""" , use_auth_token=self._token ) # This has added the proper auto_map field to the config self.assertDictEqual(config.auto_map , {"""AutoConfig""": """custom_configuration.CustomConfig"""} ) A__ : Optional[int] = AutoConfig.from_pretrained(F"""{USER}/test-dynamic-config""" , trust_remote_code=A__ ) # Can't make an isinstance check because the new_config is from the FakeConfig class of a dynamic module self.assertEqual(new_config.__class__.__name__ , """CustomConfig""" ) self.assertEqual(new_config.attribute , 42 ) class _a (unittest.TestCase ): '''simple docstring''' def __A ( self ): A__ : Any = GPTaConfig() # attempt to modify each of int/float/bool/str config records and verify they were updated A__ : Optional[int] = c.n_embd + 1 # int A__ : List[str] = c.resid_pdrop + 1.0 # float A__ : Union[str, Any] = not c.scale_attn_weights # bool A__ : Optional[Any] = c.summary_type + """foo""" # str c.update_from_string( F"""n_embd={n_embd},resid_pdrop={resid_pdrop},scale_attn_weights={scale_attn_weights},summary_type={summary_type}""" ) self.assertEqual(A__ , c.n_embd , """mismatch for key: n_embd""" ) self.assertEqual(A__ , c.resid_pdrop , """mismatch for key: resid_pdrop""" ) self.assertEqual(A__ , c.scale_attn_weights , """mismatch for key: scale_attn_weights""" ) self.assertEqual(A__ , c.summary_type , """mismatch for key: summary_type""" ) def __A ( self ): A__ : Optional[Any] = PretrainedConfig() A__ : Tuple = [key for key in base_config.__dict__ if key not in config_common_kwargs] # If this part of the test fails, you have arguments to addin config_common_kwargs above. self.assertListEqual( A__ , ["""is_encoder_decoder""", """_name_or_path""", """_commit_hash""", """transformers_version"""] ) A__ : int = [key for key, value in config_common_kwargs.items() if value == getattr(A__ , A__ )] if len(A__ ) > 0: raise ValueError( """The following keys are set with the default values in""" """ `test_configuration_common.config_common_kwargs` pick another value for them:""" F""" {', '.join(A__ )}.""" ) def __A ( self ): with self.assertRaises(A__ ): # config is in subfolder, the following should not work without specifying the subfolder A__ : Any = BertConfig.from_pretrained("""hf-internal-testing/tiny-random-bert-subfolder""" ) A__ : int = BertConfig.from_pretrained("""hf-internal-testing/tiny-random-bert-subfolder""" , subfolder="""bert""" ) self.assertIsNotNone(A__ ) def __A ( self ): # A mock response for an HTTP head request to emulate server down A__ : Tuple = mock.Mock() A__ : Tuple = 500 A__ : Tuple = {} A__ : Dict = HTTPError A__ : Optional[Any] = {} # Download this model to make sure it's in the cache. A__ : List[str] = BertConfig.from_pretrained("""hf-internal-testing/tiny-random-bert""" ) # Under the mock environment we get a 500 error when trying to reach the model. with mock.patch("""requests.Session.request""" , return_value=A__ ) as mock_head: A__ : int = BertConfig.from_pretrained("""hf-internal-testing/tiny-random-bert""" ) # This check we did call the fake head request mock_head.assert_called() def __A ( self ): # This test is for deprecated behavior and can be removed in v5 A__ : List[Any] = BertConfig.from_pretrained( """https://huggingface.co/hf-internal-testing/tiny-random-bert/resolve/main/config.json""" ) def __A ( self ): A__ : Any = AutoConfig.from_pretrained("""bert-base-cased""" ) A__ : str = ["""config.4.0.0.json"""] with tempfile.TemporaryDirectory() as tmp_dir: configuration.save_pretrained(A__ ) A__ : List[str] = 2 json.dump(configuration.to_dict() , open(os.path.join(A__ , """config.4.0.0.json""" ) , """w""" ) ) # This should pick the new configuration file as the version of Transformers is > 4.0.0 A__ : Optional[int] = AutoConfig.from_pretrained(A__ ) self.assertEqual(new_configuration.hidden_size , 2 ) # Will need to be adjusted if we reach v42 and this test is still here. # Should pick the old configuration file as the version of Transformers is < 4.42.0 A__ : Dict = ["""config.42.0.0.json"""] A__ : Any = 768 configuration.save_pretrained(A__ ) shutil.move(os.path.join(A__ , """config.4.0.0.json""" ) , os.path.join(A__ , """config.42.0.0.json""" ) ) A__ : List[Any] = AutoConfig.from_pretrained(A__ ) self.assertEqual(new_configuration.hidden_size , 768 ) def __A ( self ): # This repo has two configuration files, one for v4.0.0 and above with a different hidden size. A__ : List[Any] = """hf-internal-testing/test-two-configs""" import transformers as new_transformers A__ : Union[str, Any] = """v4.0.0""" A__ , A__ : List[Any] = new_transformers.models.auto.AutoConfig.from_pretrained( A__ , return_unused_kwargs=A__ ) self.assertEqual(new_configuration.hidden_size , 2 ) # This checks `_configuration_file` ia not kept in the kwargs by mistake. self.assertDictEqual(A__ , {} ) # Testing an older version by monkey-patching the version in the module it's used. import transformers as old_transformers A__ : Tuple = """v3.0.0""" A__ : str = old_transformers.models.auto.AutoConfig.from_pretrained(A__ ) self.assertEqual(old_configuration.hidden_size , 768 )
64
import inspect import warnings from typing import Any, Dict, Optional, Union from packaging import version def UpperCamelCase (*lowercase_: Optional[int] , lowercase_: Optional[Union[Dict, Any]] = None , lowercase_: Dict=True , lowercase_: Tuple=2 ) -> Dict: from .. import __version__ A__ : Dict = take_from A__ : str = () if not isinstance(args[0] , lowercase_ ): A__ : int = (args,) for attribute, version_name, message in args: if version.parse(version.parse(lowercase_ ).base_version ) >= version.parse(lowercase_ ): raise ValueError( f"""The deprecation tuple {(attribute, version_name, message)} should be removed since diffusers'""" f""" version {__version__} is >= {version_name}""" ) A__ : Any = None if isinstance(lowercase_ , lowercase_ ) and attribute in deprecated_kwargs: values += (deprecated_kwargs.pop(lowercase_ ),) A__ : List[str] = f"""The `{attribute}` argument is deprecated and will be removed in version {version_name}.""" elif hasattr(lowercase_ , lowercase_ ): values += (getattr(lowercase_ , lowercase_ ),) A__ : Optional[Any] = f"""The `{attribute}` attribute is deprecated and will be removed in version {version_name}.""" elif deprecated_kwargs is None: A__ : int = f"""`{attribute}` is deprecated and will be removed in version {version_name}.""" if warning is not None: A__ : int = warning + """ """ if standard_warn else """""" warnings.warn(warning + message , lowercase_ , stacklevel=lowercase_ ) if isinstance(lowercase_ , lowercase_ ) and len(lowercase_ ) > 0: A__ : Union[str, Any] = inspect.getouterframes(inspect.currentframe() )[1] A__ : Optional[Any] = call_frame.filename A__ : Optional[int] = call_frame.lineno A__ : Any = call_frame.function A__ , A__ : List[str] = next(iter(deprecated_kwargs.items() ) ) raise TypeError(f"""{function} in {filename} line {line_number-1} got an unexpected keyword argument `{key}`""" ) if len(lowercase_ ) == 0: return elif len(lowercase_ ) == 1: return values[0] return values
64
1
import doctest import glob import importlib import inspect import os import re from contextlib import contextmanager from functools import wraps from unittest.mock import patch import numpy as np import pytest from absl.testing import parameterized import datasets from datasets import load_metric from .utils import for_all_test_methods, local, slow # mark all tests as integration A_ : Tuple = pytest.mark.integration A_ : Union[str, Any] = {'comet'} A_ : Tuple = importlib.util.find_spec('fairseq') is not None A_ : Optional[int] = {'code_eval'} A_ : Union[str, Any] = os.name == 'nt' A_ : int = {'bertscore', 'frugalscore', 'perplexity'} A_ : str = importlib.util.find_spec('transformers') is not None def UpperCamelCase (lowercase_: str ) -> List[str]: @wraps(lowercase_ ) def wrapper(self: int , lowercase_: List[str] ): if not _has_fairseq and metric_name in REQUIRE_FAIRSEQ: self.skipTest("""\"test requires Fairseq\"""" ) else: test_case(self , lowercase_ ) return wrapper def UpperCamelCase (lowercase_: str ) -> Optional[int]: @wraps(lowercase_ ) def wrapper(self: Optional[Any] , lowercase_: Dict ): if not _has_transformers and metric_name in REQUIRE_TRANSFORMERS: self.skipTest("""\"test requires transformers\"""" ) else: test_case(self , lowercase_ ) return wrapper def UpperCamelCase (lowercase_: Tuple ) -> Tuple: @wraps(lowercase_ ) def wrapper(self: List[str] , lowercase_: List[Any] ): if _on_windows and metric_name in UNSUPPORTED_ON_WINDOWS: self.skipTest("""\"test not supported on Windows\"""" ) else: test_case(self , lowercase_ ) return wrapper def UpperCamelCase () -> Optional[Any]: A__ : List[str] = [metric_dir.split(os.sep )[-2] for metric_dir in glob.glob("""./metrics/*/""" )] return [{"testcase_name": x, "metric_name": x} for x in metrics if x != "gleu"] # gleu is unfinished @parameterized.named_parameters(get_local_metric_names() ) @for_all_test_methods( __magic_name__ , __magic_name__ , __magic_name__ ) @local class _a (parameterized.TestCase ): '''simple docstring''' UpperCAmelCase__: Optional[Any] = {} UpperCAmelCase__: Any = None @pytest.mark.filterwarnings("""ignore:metric_module_factory is deprecated:FutureWarning""" ) @pytest.mark.filterwarnings("""ignore:load_metric is deprecated:FutureWarning""" ) def __A ( self , A__ ): A__ : int = """[...]""" A__ : Union[str, Any] = importlib.import_module( datasets.load.metric_module_factory(os.path.join("""metrics""" , A__ ) ).module_path ) A__ : Optional[int] = datasets.load.import_main_class(metric_module.__name__ , dataset=A__ ) # check parameters A__ : int = inspect.signature(metric._compute ).parameters self.assertTrue(all(p.kind != p.VAR_KEYWORD for p in parameters.values() ) ) # no **kwargs # run doctest with self.patch_intensive_calls(A__ , metric_module.__name__ ): with self.use_local_metrics(): try: A__ : Optional[Any] = doctest.testmod(A__ , verbose=A__ , raise_on_error=A__ ) except doctest.UnexpectedException as e: raise e.exc_info[1] # raise the exception that doctest caught self.assertEqual(results.failed , 0 ) self.assertGreater(results.attempted , 1 ) @slow def __A ( self , A__ ): A__ : str = """[...]""" A__ : List[str] = importlib.import_module( datasets.load.metric_module_factory(os.path.join("""metrics""" , A__ ) ).module_path ) # run doctest with self.use_local_metrics(): A__ : int = doctest.testmod(A__ , verbose=A__ , raise_on_error=A__ ) self.assertEqual(results.failed , 0 ) self.assertGreater(results.attempted , 1 ) @contextmanager def __A ( self , A__ , A__ ): if metric_name in self.INTENSIVE_CALLS_PATCHER: with self.INTENSIVE_CALLS_PATCHER[metric_name](A__ ): yield else: yield @contextmanager def __A ( self ): def load_local_metric(A__ , *A__ , **A__ ): return load_metric(os.path.join("""metrics""" , A__ ) , *A__ , **A__ ) with patch("""datasets.load_metric""" ) as mock_load_metric: A__ : Optional[Any] = load_local_metric yield @classmethod def __A ( cls , A__ ): def wrapper(A__ ): A__ : Any = contextmanager(A__ ) A__ : int = patcher return patcher return wrapper @LocalMetricTest.register_intensive_calls_patcher("""bleurt""" ) def UpperCamelCase (lowercase_: Optional[Any] ) -> int: import tensorflow.compat.va as tf from bleurt.score import Predictor tf.flags.DEFINE_string("""sv""" , """""" , """""" ) # handle pytest cli flags class _a (__magic_name__ ): '''simple docstring''' def __A ( self , A__ ): assert len(input_dict["""input_ids"""] ) == 2 return np.array([1.0_3, 1.0_4] ) # mock predict_fn which is supposed to do a forward pass with a bleurt model with patch("""bleurt.score._create_predictor""" ) as mock_create_predictor: A__ : List[Any] = MockedPredictor() yield @LocalMetricTest.register_intensive_calls_patcher("""bertscore""" ) def UpperCamelCase (lowercase_: List[Any] ) -> List[Any]: import torch def bert_cos_score_idf(lowercase_: Tuple , lowercase_: Union[str, Any] , *lowercase_: Union[str, Any] , **lowercase_: List[str] ): return torch.tensor([[1.0, 1.0, 1.0]] * len(lowercase_ ) ) # mock get_model which is supposed to do download a bert model # mock bert_cos_score_idf which is supposed to do a forward pass with a bert model with patch("""bert_score.scorer.get_model""" ), patch( """bert_score.scorer.bert_cos_score_idf""" ) as mock_bert_cos_score_idf: A__ : Union[str, Any] = bert_cos_score_idf yield @LocalMetricTest.register_intensive_calls_patcher("""comet""" ) def UpperCamelCase (lowercase_: str ) -> Optional[Any]: def load_from_checkpoint(lowercase_: List[Any] ): class _a : '''simple docstring''' def __A ( self , A__ , *A__ , **A__ ): assert len(A__ ) == 2 A__ : str = [0.1_9, 0.9_2] return scores, sum(A__ ) / len(A__ ) return Model() # mock load_from_checkpoint which is supposed to do download a bert model # mock load_from_checkpoint which is supposed to do download a bert model with patch("""comet.download_model""" ) as mock_download_model: A__ : Optional[Any] = None with patch("""comet.load_from_checkpoint""" ) as mock_load_from_checkpoint: A__ : Optional[Any] = load_from_checkpoint yield def UpperCamelCase () -> List[str]: A__ : List[Any] = load_metric(os.path.join("""metrics""" , """seqeval""" ) ) A__ : List[Any] = """ERROR""" A__ : List[str] = f"""Scheme should be one of [IOB1, IOB2, IOE1, IOE2, IOBES, BILOU], got {wrong_scheme}""" with pytest.raises(lowercase_ , match=re.escape(lowercase_ ) ): metric.compute(predictions=[] , references=[] , scheme=lowercase_ )
64
import argparse import json import os import tensorstore as ts import torch from flax import serialization from flax.traverse_util import flatten_dict, unflatten_dict from tensorflow.io import gfile from transformers.modeling_utils import dtype_byte_size from transformers.models.switch_transformers.convert_switch_transformers_original_flax_checkpoint_to_pytorch import ( rename_keys, ) from transformers.utils import WEIGHTS_INDEX_NAME, WEIGHTS_NAME from transformers.utils.hub import convert_file_size_to_int def UpperCamelCase (lowercase_: List[str] , lowercase_: str ) -> Optional[Any]: if flax_key_tuple[-1] == "kernel" and flax_tensor.ndim == 3: # expert layer A__ : Union[str, Any] = flax_key_tuple[:-1] + ("""weight""",) A__ : Optional[int] = torch.permute(lowercase_ , (0, 2, 1) ) elif flax_key_tuple[-1] == "kernel" and ".".join(lowercase_ ): # linear layer A__ : Optional[Any] = flax_key_tuple[:-1] + ("""weight""",) A__ : int = flax_tensor.T elif flax_key_tuple[-1] in ["scale", "embedding"]: A__ : Optional[int] = flax_key_tuple[:-1] + ("""weight""",) return flax_key_tuple, flax_tensor def UpperCamelCase (lowercase_: Tuple , lowercase_: Optional[int] , lowercase_: str ) -> Union[str, Any]: if "metadata" in layer: A__ : Tuple = layer.split("""metadata""" ) A__ : Optional[Any] = """""".join(split_layer[0] )[:-1] A__ : Optional[Any] = [tuple(("""metadata""" + split_layer[1]).split("""/""" ) )] elif "kvstore" in layer: A__ : str = layer.split("""kvstore""" ) A__ : int = """""".join(split_layer[0] )[:-1] A__ : Optional[int] = [tuple(("""kvstore""" + split_layer[1]).split("""/""" ) )] else: A__ : Any = layer.split("""/""" ) A__ : int = """/""".join(split_layer[:-1] ) A__ : str = (split_layer[-1],) if "kvstore/path" in layer: A__ : Dict = f"""{switch_checkpoint_path}/{checkpoint_info[layer]}""" elif "kvstore/driver" in layer: A__ : Optional[int] = """file""" else: A__ : str = checkpoint_info[layer] return curr_real_layer_name, split_layer, content def UpperCamelCase (lowercase_: str , lowercase_: List[Any] ) -> int: A__ : int = rename_keys(lowercase_ ) A__ : Any = {} for k, v in current_block.items(): A__ : Dict = v A__ : str = new_current_block torch.save(lowercase_ , lowercase_ ) def UpperCamelCase (lowercase_: Dict , lowercase_: Optional[Any] , lowercase_: Optional[Any] , lowercase_: Optional[int] , lowercase_: str = WEIGHTS_NAME ) -> Tuple: A__ : Optional[int] = convert_file_size_to_int(lowercase_ ) A__ : List[Any] = [] A__ : int = {} A__ : List[str] = 0 A__ : Any = 0 os.makedirs(lowercase_ , exist_ok=lowercase_ ) with gfile.GFile(switch_checkpoint_path + """/checkpoint""" , """rb""" ) as fp: A__ : Optional[Any] = serialization.msgpack_restore(fp.read() )["""optimizer"""]["""target"""] A__ : Dict = flatten_dict(lowercase_ , sep="""/""" ) A__ : Any = {} for layer in checkpoint_info.keys(): A__ , A__ , A__ : Union[str, Any] = get_key_and_tensorstore_dict( lowercase_ , lowercase_ , lowercase_ ) if curr_real_layer_name in all_layers: A__ : Optional[int] = content else: A__ : List[Any] = {split_layer[-1]: content} for key in all_layers.keys(): # open tensorstore file A__ : Optional[Any] = ts.open(unflatten_dict(all_layers[key] ) ).result().read().result() A__ : List[Any] = torch.tensor(lowercase_ ) A__ : List[Any] = raw_weights.numel() * dtype_byte_size(raw_weights.dtype ) # use the renaming pattern from the small conversion scripts A__ , A__ : Any = rename_base_flax_keys(tuple(key.split("""/""" ) ) , lowercase_ ) A__ : Any = """/""".join(lowercase_ ) # If this weight is going to tip up over the maximal size, we split. if current_block_size + weight_size > max_shard_size: A__ : List[Any] = os.path.join( lowercase_ , weights_name.replace(""".bin""" , f"""-{len(lowercase_ )+1:05d}-of-???.bin""" ) ) rename_and_save_block(lowercase_ , lowercase_ ) sharded_state_dicts.append(current_block.keys() ) del current_block A__ : Any = {} A__ : str = 0 A__ : List[str] = raw_weights.to(getattr(lowercase_ , lowercase_ ) ) current_block_size += weight_size total_size += weight_size # Add the last block A__ : Union[str, Any] = os.path.join(lowercase_ , weights_name.replace(""".bin""" , f"""-{len(lowercase_ )+1:05d}-of-???.bin""" ) ) rename_and_save_block(lowercase_ , lowercase_ ) sharded_state_dicts.append(current_block.keys() ) # If we only have one shard, we return it if len(lowercase_ ) == 1: return {weights_name: sharded_state_dicts[0]}, None # Otherwise, let's build the index A__ : str = {} A__ : Any = {} for idx, shard in enumerate(lowercase_ ): A__ : Any = weights_name.replace( """.bin""" , f"""-{idx+1:05d}-of-{len(lowercase_ ):05d}.bin""" ) # len(sharded_state_dicts):05d} A__ : Dict = os.path.join(lowercase_ , weights_name.replace(""".bin""" , f"""-{idx+1:05d}-of-???.bin""" ) ) os.rename(lowercase_ , os.path.join(lowercase_ , lowercase_ ) ) A__ : str = shard for key in shard: A__ : Any = shard_file # Add the metadata A__ : Tuple = {"""total_size""": total_size} A__ : Union[str, Any] = {"""metadata""": metadata, """weight_map""": weight_map} with open(os.path.join(lowercase_ , lowercase_ ) , """w""" , encoding="""utf-8""" ) as f: A__ : Dict = json.dumps(lowercase_ , indent=2 , sort_keys=lowercase_ ) + """\n""" f.write(lowercase_ ) return metadata, index if __name__ == "__main__": A_ : Optional[int] = argparse.ArgumentParser() # Required parameters parser.add_argument( '--switch_t5x_checkpoint_path', default='/mnt/disks/disk_switch/original_checkpoints/switch-xxl-128/checkpoint_634600', type=str, required=False, help='Path to a directory containing a folder per layer. Follows the original Google format.', ) parser.add_argument('--max_shard_size', default='10GB', required=False, help='Max shard size') parser.add_argument('--dtype', default='bfloat16', type=str, required=False, help='dtype of the saved model') parser.add_argument( '--pytorch_dump_folder_path', default='/mnt/disks/disk_switch/original_checkpoints/switch-xxl-128-converted', type=str, required=False, help='Path to the output pytorch model.', ) A_ : Dict = parser.parse_args() shard_on_the_fly( args.switch_tax_checkpoint_path, args.pytorch_dump_folder_path, args.max_shard_size, args.dtype, ) def UpperCamelCase () -> int: from transformers import SwitchTransformersConfig, SwitchTransformersForConditionalGeneration, TaTokenizer A__ : str = SwitchTransformersConfig.from_pretrained("""google/switch-base-8""" ) config.save_pretrained("""/home/arthur_huggingface_co/transformers/switch_converted""" ) A__ : str = SwitchTransformersForConditionalGeneration.from_pretrained( """/home/arthur_huggingface_co/transformers/switch_converted""" , device_map="""auto""" ) A__ : Tuple = TaTokenizer.from_pretrained("""t5-small""" ) A__ : Dict = """A <extra_id_0> walks into a bar a orders a <extra_id_1> with <extra_id_2> pinch of <extra_id_3>.""" A__ : Union[str, Any] = tokenizer(lowercase_ , return_tensors="""pt""" ).input_ids A__ : Tuple = model.generate(lowercase_ , decoder_start_token_id=0 ) print(tokenizer.decode(out[0] ) )
64
1
import pickle import shutil import tempfile import unittest from transformers import SPIECE_UNDERLINE, XGLMTokenizer, XGLMTokenizerFast from transformers.testing_utils import get_tests_dir, require_sentencepiece, require_tokenizers, slow from transformers.utils import cached_property from ...test_tokenization_common import TokenizerTesterMixin A_ : List[Any] = get_tests_dir('fixtures/test_sentencepiece.model') @require_sentencepiece @require_tokenizers class _a (__magic_name__ , unittest.TestCase ): '''simple docstring''' UpperCAmelCase__: Dict = XGLMTokenizer UpperCAmelCase__: List[Any] = XGLMTokenizerFast UpperCAmelCase__: Tuple = True UpperCAmelCase__: str = True def __A ( self ): super().setUp() # We have a SentencePiece fixture for testing A__ : Any = XGLMTokenizer(A__ , keep_accents=A__ ) tokenizer.save_pretrained(self.tmpdirname ) def __A ( self ): A__ : Tuple = """<pad>""" A__ : Tuple = 1 self.assertEqual(self.get_tokenizer()._convert_token_to_id(A__ ) , A__ ) self.assertEqual(self.get_tokenizer()._convert_id_to_token(A__ ) , A__ ) def __A ( self ): A__ : Dict = list(self.get_tokenizer().get_vocab().keys() ) self.assertEqual(vocab_keys[0] , """<s>""" ) self.assertEqual(vocab_keys[1] , """<pad>""" ) self.assertEqual(len(A__ ) , 1008 ) def __A ( self ): self.assertEqual(self.get_tokenizer().vocab_size , 1008 ) def __A ( self ): A__ : Optional[int] = XGLMTokenizer(A__ , keep_accents=A__ ) A__ : Tuple = tokenizer.tokenize("""This is a test""" ) self.assertListEqual(A__ , ["""▁This""", """▁is""", """▁a""", """▁t""", """est"""] ) self.assertListEqual( tokenizer.convert_tokens_to_ids(A__ ) , [value + tokenizer.fairseq_offset for value in [285, 46, 10, 170, 382]] , ) A__ : List[Any] = tokenizer.tokenize("""I was born in 92000, and this is falsé.""" ) self.assertListEqual( A__ , [ SPIECE_UNDERLINE + """I""", SPIECE_UNDERLINE + """was""", SPIECE_UNDERLINE + """b""", """or""", """n""", SPIECE_UNDERLINE + """in""", SPIECE_UNDERLINE + """""", """9""", """2""", """0""", """0""", """0""", """,""", SPIECE_UNDERLINE + """and""", SPIECE_UNDERLINE + """this""", SPIECE_UNDERLINE + """is""", SPIECE_UNDERLINE + """f""", """al""", """s""", """é""", """.""", ] , ) A__ : List[str] = tokenizer.convert_tokens_to_ids(A__ ) self.assertListEqual( A__ , [ value + tokenizer.fairseq_offset for value in [8, 21, 84, 55, 24, 19, 7, 2, 602, 347, 347, 347, 3, 12, 66, 46, 72, 80, 6, 2, 4] ] , ) A__ : List[Any] = tokenizer.convert_ids_to_tokens(A__ ) self.assertListEqual( A__ , [ SPIECE_UNDERLINE + """I""", SPIECE_UNDERLINE + """was""", SPIECE_UNDERLINE + """b""", """or""", """n""", SPIECE_UNDERLINE + """in""", SPIECE_UNDERLINE + """""", """<unk>""", """2""", """0""", """0""", """0""", """,""", SPIECE_UNDERLINE + """and""", SPIECE_UNDERLINE + """this""", SPIECE_UNDERLINE + """is""", SPIECE_UNDERLINE + """f""", """al""", """s""", """<unk>""", """.""", ] , ) @cached_property def __A ( self ): return XGLMTokenizer.from_pretrained("""facebook/xglm-564M""" ) def __A ( self ): with tempfile.NamedTemporaryFile() as f: shutil.copyfile(A__ , f.name ) A__ : List[str] = XGLMTokenizer(f.name , keep_accents=A__ ) A__ : List[Any] = pickle.dumps(A__ ) pickle.loads(A__ ) def __A ( self ): if not self.test_rust_tokenizer: return A__ : List[str] = self.get_tokenizer() A__ : Dict = self.get_rust_tokenizer() A__ : int = """I was born in 92000, and this is falsé.""" A__ : List[Any] = tokenizer.tokenize(A__ ) A__ : str = rust_tokenizer.tokenize(A__ ) self.assertListEqual(A__ , A__ ) A__ : Union[str, Any] = tokenizer.encode(A__ , add_special_tokens=A__ ) A__ : List[str] = rust_tokenizer.encode(A__ , add_special_tokens=A__ ) self.assertListEqual(A__ , A__ ) A__ : int = self.get_rust_tokenizer() A__ : Optional[int] = tokenizer.encode(A__ ) A__ : List[str] = rust_tokenizer.encode(A__ ) self.assertListEqual(A__ , A__ ) @slow def __A ( self ): A__ : Optional[int] = """Hello World!""" A__ : Union[str, Any] = [2, 3_1227, 4447, 35] self.assertListEqual(A__ , self.big_tokenizer.encode(A__ ) ) @slow def __A ( self ): A__ : Dict = ( """This is a very long text with a lot of weird characters, such as: . , ~ ? ( ) \" [ ] ! : - . Also we will""" """ add words that should not exsist and be tokenized to unk, such as saoneuhaoesuth""" ) # fmt: off A__ : str = [2, 1018, 67, 11, 1988, 2617, 5631, 278, 11, 3407, 48, 7_1630, 2_8085, 4, 3234, 157, 13, 6, 5, 6, 4, 3526, 768, 15, 659, 57, 298, 3983, 864, 129, 21, 6, 5, 1_3675, 377, 652, 7580, 1_0341, 155, 2817, 422, 1666, 7, 1674, 53, 113, 20_2277, 1_7892, 33, 60, 87, 4, 3234, 157, 61, 2667, 5_2376, 19, 88, 23, 735] # fmt: on self.assertListEqual(A__ , self.big_tokenizer.encode(A__ ) ) @slow def __A ( self ): # fmt: off A__ : Tuple = { """input_ids""": [[2, 10_8825, 1163, 15, 8_8010, 473, 1_5898, 157, 1_3672, 1857, 312, 8, 23_8021, 1163, 53, 1_3672, 1857, 312, 8, 5_3283, 18_2396, 8, 1_8566, 16, 3_6733, 4101, 8, 230, 24_4017, 12_2553, 7, 15, 13_2597, 4, 293, 1_2511, 7610, 4, 3414, 13_2597, 9, 4, 3_2361, 362, 4, 734, 2_8512, 3_2569, 18, 4, 3_2361, 2_6096, 1_4982, 73, 1_8715, 2_1433, 23_5261, 15, 492, 1_2427, 16, 53, 1_8715, 2_1433, 6_5454, 15, 2_3659, 563, 16, 278, 597, 2843, 595, 7931, 18_2396, 6_4186, 22, 886, 595, 13_2981, 53, 2_5540, 3449, 4_3982, 3_9901, 5951, 878, 330, 4, 2_7694, 8_0269, 312, 53, 6517, 1_1780, 611, 2_0408, 5], [2, 6, 13_2597, 67, 4_2897, 33, 592, 8, 16_3729, 2_5540, 361, 13_6997, 10_9514, 17_3230, 7, 501, 60, 10_2913, 196, 5631, 235, 6_3243, 473, 6, 23_1757, 74, 5277, 7905, 53, 3095, 3_7317, 22, 454, 18_3874, 5], [2, 268, 3_1298, 4_6530, 6, 13_2935, 4_3831, 7, 597, 32, 24, 3688, 9865, 5]], """attention_mask""": [[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1], [1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1], [1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]] } # noqa: E501 # fmt: on self.tokenizer_integration_test_util( expected_encoding=A__ , model_name="""facebook/xglm-564M""" , padding=A__ , )
64
from typing import TYPE_CHECKING from ...utils import OptionalDependencyNotAvailable, _LazyModule, is_sentencepiece_available A_ : Optional[Any] = {} try: if not is_sentencepiece_available(): raise OptionalDependencyNotAvailable() except OptionalDependencyNotAvailable: pass else: A_ : str = ['BartphoTokenizer'] if TYPE_CHECKING: try: if not is_sentencepiece_available(): raise OptionalDependencyNotAvailable() except OptionalDependencyNotAvailable: pass else: from .tokenization_bartpho import BartphoTokenizer else: import sys A_ : Optional[int] = _LazyModule(__name__, globals()['__file__'], _import_structure, module_spec=__spec__)
64
1
import argparse import json import logging import os import sys from unittest.mock import patch from transformers.testing_utils import TestCasePlus, get_gpu_count, slow A_ : Dict = [ os.path.join(os.path.dirname(__file__), dirname) for dirname in [ 'text-classification', 'language-modeling', 'summarization', 'token-classification', 'question-answering', ] ] sys.path.extend(SRC_DIRS) if SRC_DIRS is not None: import run_clm_flax import run_flax_glue import run_flax_ner import run_mlm_flax import run_qa import run_summarization_flax import run_ta_mlm_flax logging.basicConfig(level=logging.DEBUG) A_ : Optional[Any] = logging.getLogger() def UpperCamelCase () -> Optional[Any]: A__ : Optional[Any] = argparse.ArgumentParser() parser.add_argument("""-f""" ) A__ : Tuple = parser.parse_args() return args.f def UpperCamelCase (lowercase_: Optional[int] , lowercase_: int="eval" ) -> Dict: A__ : List[Any] = os.path.join(lowercase_ , f"""{split}_results.json""" ) if os.path.exists(lowercase_ ): with open(lowercase_ , """r""" ) as f: return json.load(lowercase_ ) raise ValueError(f"""can't find {path}""" ) A_ : str = logging.StreamHandler(sys.stdout) logger.addHandler(stream_handler) class _a (__magic_name__ ): '''simple docstring''' def __A ( self ): A__ : Tuple = self.get_auto_remove_tmp_dir() A__ : str = F""" run_glue.py --model_name_or_path distilbert-base-uncased --output_dir {tmp_dir} --train_file ./tests/fixtures/tests_samples/MRPC/train.csv --validation_file ./tests/fixtures/tests_samples/MRPC/dev.csv --per_device_train_batch_size=2 --per_device_eval_batch_size=1 --learning_rate=1e-4 --eval_steps=2 --warmup_steps=2 --seed=42 --max_seq_length=128 """.split() with patch.object(A__ , """argv""" , A__ ): run_flax_glue.main() A__ : Tuple = get_results(A__ ) self.assertGreaterEqual(result["""eval_accuracy"""] , 0.7_5 ) @slow def __A ( self ): A__ : List[str] = self.get_auto_remove_tmp_dir() A__ : Tuple = F""" run_clm_flax.py --model_name_or_path distilgpt2 --train_file ./tests/fixtures/sample_text.txt --validation_file ./tests/fixtures/sample_text.txt --do_train --do_eval --block_size 128 --per_device_train_batch_size 4 --per_device_eval_batch_size 4 --num_train_epochs 2 --logging_steps 2 --eval_steps 2 --output_dir {tmp_dir} --overwrite_output_dir """.split() with patch.object(A__ , """argv""" , A__ ): run_clm_flax.main() A__ : int = get_results(A__ ) self.assertLess(result["""eval_perplexity"""] , 100 ) @slow def __A ( self ): A__ : Tuple = self.get_auto_remove_tmp_dir() A__ : Union[str, Any] = F""" run_summarization.py --model_name_or_path t5-small --train_file tests/fixtures/tests_samples/xsum/sample.json --validation_file tests/fixtures/tests_samples/xsum/sample.json --test_file tests/fixtures/tests_samples/xsum/sample.json --output_dir {tmp_dir} --overwrite_output_dir --num_train_epochs=3 --warmup_steps=8 --do_train --do_eval --do_predict --learning_rate=2e-4 --per_device_train_batch_size=2 --per_device_eval_batch_size=1 --predict_with_generate """.split() with patch.object(A__ , """argv""" , A__ ): run_summarization_flax.main() A__ : List[Any] = get_results(A__ , split="""test""" ) self.assertGreaterEqual(result["""test_rouge1"""] , 10 ) self.assertGreaterEqual(result["""test_rouge2"""] , 2 ) self.assertGreaterEqual(result["""test_rougeL"""] , 7 ) self.assertGreaterEqual(result["""test_rougeLsum"""] , 7 ) @slow def __A ( self ): A__ : int = self.get_auto_remove_tmp_dir() A__ : int = F""" run_mlm.py --model_name_or_path distilroberta-base --train_file ./tests/fixtures/sample_text.txt --validation_file ./tests/fixtures/sample_text.txt --output_dir {tmp_dir} --overwrite_output_dir --max_seq_length 128 --per_device_train_batch_size 4 --per_device_eval_batch_size 4 --logging_steps 2 --eval_steps 2 --do_train --do_eval --num_train_epochs=1 """.split() with patch.object(A__ , """argv""" , A__ ): run_mlm_flax.main() A__ : str = get_results(A__ ) self.assertLess(result["""eval_perplexity"""] , 42 ) @slow def __A ( self ): A__ : List[str] = self.get_auto_remove_tmp_dir() A__ : Union[str, Any] = F""" run_t5_mlm_flax.py --model_name_or_path t5-small --train_file ./tests/fixtures/sample_text.txt --validation_file ./tests/fixtures/sample_text.txt --do_train --do_eval --max_seq_length 128 --per_device_train_batch_size 4 --per_device_eval_batch_size 4 --num_train_epochs 2 --logging_steps 2 --eval_steps 2 --output_dir {tmp_dir} --overwrite_output_dir """.split() with patch.object(A__ , """argv""" , A__ ): run_ta_mlm_flax.main() A__ : str = get_results(A__ ) self.assertGreaterEqual(result["""eval_accuracy"""] , 0.4_2 ) @slow def __A ( self ): # with so little data distributed training needs more epochs to get the score on par with 0/1 gpu A__ : Any = 7 if get_gpu_count() > 1 else 2 A__ : List[Any] = self.get_auto_remove_tmp_dir() A__ : int = F""" run_flax_ner.py --model_name_or_path bert-base-uncased --train_file tests/fixtures/tests_samples/conll/sample.json --validation_file tests/fixtures/tests_samples/conll/sample.json --output_dir {tmp_dir} --overwrite_output_dir --do_train --do_eval --warmup_steps=2 --learning_rate=2e-4 --logging_steps 2 --eval_steps 2 --per_device_train_batch_size=2 --per_device_eval_batch_size=2 --num_train_epochs={epochs} --seed 7 """.split() with patch.object(A__ , """argv""" , A__ ): run_flax_ner.main() A__ : Union[str, Any] = get_results(A__ ) self.assertGreaterEqual(result["""eval_accuracy"""] , 0.7_5 ) self.assertGreaterEqual(result["""eval_f1"""] , 0.3 ) @slow def __A ( self ): A__ : Optional[Any] = self.get_auto_remove_tmp_dir() A__ : Tuple = F""" run_qa.py --model_name_or_path bert-base-uncased --version_2_with_negative --train_file tests/fixtures/tests_samples/SQUAD/sample.json --validation_file tests/fixtures/tests_samples/SQUAD/sample.json --output_dir {tmp_dir} --overwrite_output_dir --num_train_epochs=3 --warmup_steps=2 --do_train --do_eval --logging_steps 2 --eval_steps 2 --learning_rate=2e-4 --per_device_train_batch_size=2 --per_device_eval_batch_size=1 """.split() with patch.object(A__ , """argv""" , A__ ): run_qa.main() A__ : Tuple = get_results(A__ ) self.assertGreaterEqual(result["""eval_f1"""] , 30 ) self.assertGreaterEqual(result["""eval_exact"""] , 30 )
64
import argparse import hashlib import os import urllib import warnings import torch from torch import nn from tqdm import tqdm from transformers import WhisperConfig, WhisperForConditionalGeneration A_ : Dict = { 'tiny.en': 'https://openaipublic.azureedge.net/main/whisper/models/d3dd57d32accea0b295c96e26691aa14d8822fac7d9d27d5dc00b4ca2826dd03/tiny.en.pt', 'tiny': 'https://openaipublic.azureedge.net/main/whisper/models/65147644a518d12f04e32d6f3b26facc3f8dd46e5390956a9424a650c0ce22b9/tiny.pt', 'base.en': 'https://openaipublic.azureedge.net/main/whisper/models/25a8566e1d0c1e2231d1c762132cd20e0f96a85d16145c3a00adf5d1ac670ead/base.en.pt', 'base': 'https://openaipublic.azureedge.net/main/whisper/models/ed3a0b6b1c0edf879ad9b11b1af5a0e6ab5db9205f891f668f8b0e6c6326e34e/base.pt', 'small.en': 'https://openaipublic.azureedge.net/main/whisper/models/f953ad0fd29cacd07d5a9eda5624af0f6bcf2258be67c92b79389873d91e0872/small.en.pt', 'small': 'https://openaipublic.azureedge.net/main/whisper/models/9ecf779972d90ba49c06d968637d720dd632c55bbf19d441fb42bf17a411e794/small.pt', 'medium.en': 'https://openaipublic.azureedge.net/main/whisper/models/d7440d1dc186f76616474e0ff0b3b6b879abc9d1a4926b7adfa41db2d497ab4f/medium.en.pt', 'medium': 'https://openaipublic.azureedge.net/main/whisper/models/345ae4da62f9b3d59415adc60127b97c714f32e89e936602e85993674d08dcb1/medium.pt', 'large': 'https://openaipublic.azureedge.net/main/whisper/models/e4b87e7e0bf463eb8e6956e646f1e277e901512310def2c24bf0e11bd3c28e9a/large.pt', 'large-v2': 'https://openaipublic.azureedge.net/main/whisper/models/81f7c96c852ee8fc832187b0132e569d6c3065a3252ed18e56effd0b6a73e524/large-v2.pt', } def UpperCamelCase (lowercase_: Optional[Any] ) -> Optional[int]: A__ : List[Any] = ["""layers""", """blocks"""] for k in ignore_keys: state_dict.pop(lowercase_ , lowercase_ ) A_ : Any = { 'blocks': 'layers', 'mlp.0': 'fc1', 'mlp.2': 'fc2', 'mlp_ln': 'final_layer_norm', '.attn.query': '.self_attn.q_proj', '.attn.key': '.self_attn.k_proj', '.attn.value': '.self_attn.v_proj', '.attn_ln': '.self_attn_layer_norm', '.attn.out': '.self_attn.out_proj', '.cross_attn.query': '.encoder_attn.q_proj', '.cross_attn.key': '.encoder_attn.k_proj', '.cross_attn.value': '.encoder_attn.v_proj', '.cross_attn_ln': '.encoder_attn_layer_norm', '.cross_attn.out': '.encoder_attn.out_proj', 'decoder.ln.': 'decoder.layer_norm.', 'encoder.ln.': 'encoder.layer_norm.', 'token_embedding': 'embed_tokens', 'encoder.positional_embedding': 'encoder.embed_positions.weight', 'decoder.positional_embedding': 'decoder.embed_positions.weight', 'ln_post': 'layer_norm', } def UpperCamelCase (lowercase_: str ) -> Any: A__ : Dict = list(s_dict.keys() ) for key in keys: A__ : List[str] = key for k, v in WHISPER_MAPPING.items(): if k in key: A__ : List[Any] = new_key.replace(lowercase_ , lowercase_ ) print(f"""{key} -> {new_key}""" ) A__ : Tuple = s_dict.pop(lowercase_ ) return s_dict def UpperCamelCase (lowercase_: Tuple ) -> Optional[int]: A__ , A__ : Any = emb.weight.shape A__ : str = nn.Linear(lowercase_ , lowercase_ , bias=lowercase_ ) A__ : Union[str, Any] = emb.weight.data return lin_layer def UpperCamelCase (lowercase_: str , lowercase_: str ) -> bytes: os.makedirs(lowercase_ , exist_ok=lowercase_ ) A__ : Tuple = os.path.basename(lowercase_ ) A__ : int = url.split("""/""" )[-2] A__ : Dict = os.path.join(lowercase_ , lowercase_ ) if os.path.exists(lowercase_ ) and not os.path.isfile(lowercase_ ): raise RuntimeError(f"""{download_target} exists and is not a regular file""" ) if os.path.isfile(lowercase_ ): A__ : Optional[Any] = open(lowercase_ , """rb""" ).read() if hashlib.shaaaa(lowercase_ ).hexdigest() == expected_shaaaa: return model_bytes else: warnings.warn(f"""{download_target} exists, but the SHA256 checksum does not match; re-downloading the file""" ) with urllib.request.urlopen(lowercase_ ) as source, open(lowercase_ , """wb""" ) as output: with tqdm( total=int(source.info().get("""Content-Length""" ) ) , ncols=80 , unit="""iB""" , unit_scale=lowercase_ , unit_divisor=1024 ) as loop: while True: A__ : Any = source.read(8192 ) if not buffer: break output.write(lowercase_ ) loop.update(len(lowercase_ ) ) A__ : Dict = open(lowercase_ , """rb""" ).read() if hashlib.shaaaa(lowercase_ ).hexdigest() != expected_shaaaa: raise RuntimeError( """Model has been downloaded but the SHA256 checksum does not not match. Please retry loading the model.""" ) return model_bytes def UpperCamelCase (lowercase_: List[Any] , lowercase_: Tuple ) -> Optional[Any]: if ".pt" not in checkpoint_path: A__ : Tuple = _download(_MODELS[checkpoint_path] ) else: A__ : Optional[int] = torch.load(lowercase_ , map_location="""cpu""" ) A__ : str = original_checkpoint["""dims"""] A__ : List[Any] = original_checkpoint["""model_state_dict"""] A__ : Optional[Any] = state_dict["""decoder.token_embedding.weight"""] remove_ignore_keys_(lowercase_ ) rename_keys(lowercase_ ) A__ : List[str] = True A__ : Optional[Any] = state_dict["""decoder.layers.0.fc1.weight"""].shape[0] A__ : List[Any] = WhisperConfig( vocab_size=dimensions["""n_vocab"""] , encoder_ffn_dim=lowercase_ , decoder_ffn_dim=lowercase_ , num_mel_bins=dimensions["""n_mels"""] , d_model=dimensions["""n_audio_state"""] , max_target_positions=dimensions["""n_text_ctx"""] , encoder_layers=dimensions["""n_audio_layer"""] , encoder_attention_heads=dimensions["""n_audio_head"""] , decoder_layers=dimensions["""n_text_layer"""] , decoder_attention_heads=dimensions["""n_text_state"""] , max_source_positions=dimensions["""n_audio_ctx"""] , ) A__ : Optional[Any] = WhisperForConditionalGeneration(lowercase_ ) A__ , A__ : List[Any] = model.model.load_state_dict(lowercase_ , strict=lowercase_ ) if len(lowercase_ ) > 0 and not set(lowercase_ ) <= { "encoder.embed_positions.weights", "decoder.embed_positions.weights", }: raise ValueError( """Only `encoder.embed_positions.weights` and `decoder.embed_positions.weights` are allowed to be missing,""" f""" but all the following weights are missing {missing}""" ) if tie_embeds: A__ : Any = make_linear_from_emb(model.model.decoder.embed_tokens ) else: A__ : str = proj_out_weights model.save_pretrained(lowercase_ ) if __name__ == "__main__": A_ : Any = argparse.ArgumentParser() # # Required parameters parser.add_argument('--checkpoint_path', type=str, help='Patht to the downloaded checkpoints') parser.add_argument('--pytorch_dump_folder_path', default=None, type=str, help='Path to the output PyTorch model.') A_ : Tuple = parser.parse_args() convert_openai_whisper_to_tfms(args.checkpoint_path, args.pytorch_dump_folder_path)
64
1
from .integrations import ( is_optuna_available, is_ray_available, is_sigopt_available, is_wandb_available, run_hp_search_optuna, run_hp_search_ray, run_hp_search_sigopt, run_hp_search_wandb, ) from .trainer_utils import ( HPSearchBackend, default_hp_space_optuna, default_hp_space_ray, default_hp_space_sigopt, default_hp_space_wandb, ) from .utils import logging A_ : Tuple = logging.get_logger(__name__) class _a : '''simple docstring''' UpperCAmelCase__: str UpperCAmelCase__: str = None @staticmethod def __A ( ): raise NotImplementedError def __A ( self , A__ , A__ , A__ , **A__ ): raise NotImplementedError def __A ( self , A__ ): raise NotImplementedError def __A ( self ): if not self.is_available(): raise RuntimeError( F"""You picked the {self.name} backend, but it is not installed. Run {self.pip_install()}.""" ) @classmethod def __A ( cls ): return F"""`pip install {cls.pip_package or cls.name}`""" class _a (__magic_name__ ): '''simple docstring''' UpperCAmelCase__: Tuple = '''optuna''' @staticmethod def __A ( ): return is_optuna_available() def __A ( self , A__ , A__ , A__ , **A__ ): return run_hp_search_optuna(A__ , A__ , A__ , **A__ ) def __A ( self , A__ ): return default_hp_space_optuna(A__ ) class _a (__magic_name__ ): '''simple docstring''' UpperCAmelCase__: Union[str, Any] = '''ray''' UpperCAmelCase__: Any = '''\'ray[tune]\'''' @staticmethod def __A ( ): return is_ray_available() def __A ( self , A__ , A__ , A__ , **A__ ): return run_hp_search_ray(A__ , A__ , A__ , **A__ ) def __A ( self , A__ ): return default_hp_space_ray(A__ ) class _a (__magic_name__ ): '''simple docstring''' UpperCAmelCase__: Tuple = '''sigopt''' @staticmethod def __A ( ): return is_sigopt_available() def __A ( self , A__ , A__ , A__ , **A__ ): return run_hp_search_sigopt(A__ , A__ , A__ , **A__ ) def __A ( self , A__ ): return default_hp_space_sigopt(A__ ) class _a (__magic_name__ ): '''simple docstring''' UpperCAmelCase__: str = '''wandb''' @staticmethod def __A ( ): return is_wandb_available() def __A ( self , A__ , A__ , A__ , **A__ ): return run_hp_search_wandb(A__ , A__ , A__ , **A__ ) def __A ( self , A__ ): return default_hp_space_wandb(A__ ) A_ : List[str] = { HPSearchBackend(backend.name): backend for backend in [OptunaBackend, RayTuneBackend, SigOptBackend, WandbBackend] } def UpperCamelCase () -> str: A__ : Union[str, Any] = [backend for backend in ALL_HYPERPARAMETER_SEARCH_BACKENDS.values() if backend.is_available()] if len(lowercase_ ) > 0: A__ : Optional[Any] = available_backends[0].name if len(lowercase_ ) > 1: logger.info( f"""{len(lowercase_ )} hyperparameter search backends available. Using {name} as the default.""" ) return name raise RuntimeError( """No hyperparameter search backend available.\n""" + """\n""".join( f""" - To install {backend.name} run {backend.pip_install()}""" for backend in ALL_HYPERPARAMETER_SEARCH_BACKENDS.values() ) )
64
import unittest import numpy as np import torch from transformers import CLIPTextConfig, CLIPTextModel, CLIPTokenizer from diffusers import ( AutoencoderKL, DDIMScheduler, DPMSolverMultistepScheduler, TextToVideoSDPipeline, UNetaDConditionModel, ) from diffusers.utils import is_xformers_available, load_numpy, skip_mps, slow, torch_device from diffusers.utils.testing_utils import enable_full_determinism from ..pipeline_params import TEXT_TO_IMAGE_BATCH_PARAMS, TEXT_TO_IMAGE_PARAMS from ..test_pipelines_common import PipelineTesterMixin enable_full_determinism() @skip_mps class _a (__magic_name__ , unittest.TestCase ): '''simple docstring''' UpperCAmelCase__: Any = TextToVideoSDPipeline UpperCAmelCase__: Any = TEXT_TO_IMAGE_PARAMS UpperCAmelCase__: Optional[Any] = TEXT_TO_IMAGE_BATCH_PARAMS # No `output_type`. UpperCAmelCase__: Optional[int] = frozenset( [ '''num_inference_steps''', '''generator''', '''latents''', '''return_dict''', '''callback''', '''callback_steps''', ] ) def __A ( self ): torch.manual_seed(0 ) A__ : Optional[int] = UNetaDConditionModel( block_out_channels=(32, 64, 64, 64) , layers_per_block=2 , sample_size=32 , in_channels=4 , out_channels=4 , down_block_types=("""CrossAttnDownBlock3D""", """CrossAttnDownBlock3D""", """CrossAttnDownBlock3D""", """DownBlock3D""") , up_block_types=("""UpBlock3D""", """CrossAttnUpBlock3D""", """CrossAttnUpBlock3D""", """CrossAttnUpBlock3D""") , cross_attention_dim=32 , attention_head_dim=4 , ) A__ : Optional[int] = DDIMScheduler( beta_start=0.0_0_0_8_5 , beta_end=0.0_1_2 , beta_schedule="""scaled_linear""" , clip_sample=A__ , set_alpha_to_one=A__ , ) torch.manual_seed(0 ) A__ : Union[str, Any] = AutoencoderKL( block_out_channels=[32, 64] , in_channels=3 , out_channels=3 , down_block_types=["""DownEncoderBlock2D""", """DownEncoderBlock2D"""] , up_block_types=["""UpDecoderBlock2D""", """UpDecoderBlock2D"""] , latent_channels=4 , sample_size=128 , ) torch.manual_seed(0 ) A__ : Union[str, Any] = CLIPTextConfig( bos_token_id=0 , eos_token_id=2 , hidden_size=32 , intermediate_size=37 , layer_norm_eps=1e-05 , num_attention_heads=4 , num_hidden_layers=5 , pad_token_id=1 , vocab_size=1000 , hidden_act="""gelu""" , projection_dim=512 , ) A__ : Union[str, Any] = CLIPTextModel(A__ ) A__ : Tuple = CLIPTokenizer.from_pretrained("""hf-internal-testing/tiny-random-clip""" ) A__ : Dict = { """unet""": unet, """scheduler""": scheduler, """vae""": vae, """text_encoder""": text_encoder, """tokenizer""": tokenizer, } return components def __A ( self , A__ , A__=0 ): if str(A__ ).startswith("""mps""" ): A__ : Tuple = torch.manual_seed(A__ ) else: A__ : List[str] = torch.Generator(device=A__ ).manual_seed(A__ ) A__ : List[str] = { """prompt""": """A painting of a squirrel eating a burger""", """generator""": generator, """num_inference_steps""": 2, """guidance_scale""": 6.0, """output_type""": """pt""", } return inputs def __A ( self ): A__ : List[str] = """cpu""" # ensure determinism for the device-dependent torch.Generator A__ : Union[str, Any] = self.get_dummy_components() A__ : Union[str, Any] = TextToVideoSDPipeline(**A__ ) A__ : int = sd_pipe.to(A__ ) sd_pipe.set_progress_bar_config(disable=A__ ) A__ : int = self.get_dummy_inputs(A__ ) A__ : int = """np""" A__ : Any = sd_pipe(**A__ ).frames A__ : Dict = frames[0][-3:, -3:, -1] assert frames[0].shape == (64, 64, 3) A__ : Optional[Any] = np.array([1_5_8.0, 1_6_0.0, 1_5_3.0, 1_2_5.0, 1_0_0.0, 1_2_1.0, 1_1_1.0, 9_3.0, 1_1_3.0] ) assert np.abs(image_slice.flatten() - expected_slice ).max() < 1e-2 def __A ( self ): self._test_attention_slicing_forward_pass(test_mean_pixel_difference=A__ , expected_max_diff=3e-3 ) @unittest.skipIf( torch_device != """cuda""" or not is_xformers_available() , reason="""XFormers attention is only available with CUDA and `xformers` installed""" , ) def __A ( self ): self._test_xformers_attention_forwardGenerator_pass(test_mean_pixel_difference=A__ , expected_max_diff=1e-2 ) @unittest.skip(reason="""Batching needs to be properly figured out first for this pipeline.""" ) def __A ( self ): pass @unittest.skip(reason="""Batching needs to be properly figured out first for this pipeline.""" ) def __A ( self ): pass @unittest.skip(reason="""`num_images_per_prompt` argument is not supported for this pipeline.""" ) def __A ( self ): pass def __A ( self ): return super().test_progress_bar() @slow @skip_mps class _a (unittest.TestCase ): '''simple docstring''' def __A ( self ): A__ : Union[str, Any] = load_numpy( """https://huggingface.co/datasets/hf-internal-testing/diffusers-images/resolve/main/text_to_video/video.npy""" ) A__ : Tuple = TextToVideoSDPipeline.from_pretrained("""damo-vilab/text-to-video-ms-1.7b""" ) A__ : Any = DPMSolverMultistepScheduler.from_config(pipe.scheduler.config ) A__ : int = pipe.to("""cuda""" ) A__ : Optional[Any] = """Spiderman is surfing""" A__ : List[str] = torch.Generator(device="""cpu""" ).manual_seed(0 ) A__ : Optional[Any] = pipe(A__ , generator=A__ , num_inference_steps=25 , output_type="""pt""" ).frames A__ : Dict = video_frames.cpu().numpy() assert np.abs(expected_video - video ).mean() < 5e-2 def __A ( self ): A__ : List[Any] = load_numpy( """https://huggingface.co/datasets/hf-internal-testing/diffusers-images/resolve/main/text_to_video/video_2step.npy""" ) A__ : Optional[int] = TextToVideoSDPipeline.from_pretrained("""damo-vilab/text-to-video-ms-1.7b""" ) A__ : List[str] = pipe.to("""cuda""" ) A__ : Dict = """Spiderman is surfing""" A__ : Union[str, Any] = torch.Generator(device="""cpu""" ).manual_seed(0 ) A__ : Optional[int] = pipe(A__ , generator=A__ , num_inference_steps=2 , output_type="""pt""" ).frames A__ : Optional[int] = video_frames.cpu().numpy() assert np.abs(expected_video - video ).mean() < 5e-2
64
1
def UpperCamelCase (lowercase_: int ) -> list: A__ : str = int(lowercase_ ) if n_element < 1: A__ : Optional[int] = ValueError("""a should be a positive number""" ) raise my_error A__ : int = [1] A__ , A__ , A__ : str = (0, 0, 0) A__ : Union[str, Any] = 1 while index < n_element: while hamming_list[i] * 2 <= hamming_list[-1]: i += 1 while hamming_list[j] * 3 <= hamming_list[-1]: j += 1 while hamming_list[k] * 5 <= hamming_list[-1]: k += 1 hamming_list.append( min(hamming_list[i] * 2 , hamming_list[j] * 3 , hamming_list[k] * 5 ) ) index += 1 return hamming_list if __name__ == "__main__": A_ : Optional[int] = input('Enter the last number (nth term) of the Hamming Number Series: ') print('Formula of Hamming Number Series => 2^i * 3^j * 5^k') A_ : str = hamming(int(n)) print('-----------------------------------------------------') print(f'''The list with nth numbers is: {hamming_numbers}''') print('-----------------------------------------------------')
64
def UpperCamelCase (lowercase_: int ) -> int: if not isinstance(lowercase_ , lowercase_ ): raise TypeError("""Input value must be an 'int' type""" ) A__ : int = 0 while number: position += 1 number >>= 1 return position if __name__ == "__main__": import doctest doctest.testmod()
64
1
import os try: from .build_directory_md import good_file_paths except ImportError: from build_directory_md import good_file_paths # type: ignore A_ : Dict = list(good_file_paths()) assert filepaths, "good_file_paths() failed!" A_ : Optional[Any] = [file for file in filepaths if file != file.lower()] if upper_files: print(f'''{len(upper_files)} files contain uppercase characters:''') print('\n'.join(upper_files) + '\n') A_ : Tuple = [file for file in filepaths if ' ' in file] if space_files: print(f'''{len(space_files)} files contain space characters:''') print('\n'.join(space_files) + '\n') A_ : Any = [file for file in filepaths if '-' in file] if hyphen_files: print(f'''{len(hyphen_files)} files contain hyphen characters:''') print('\n'.join(hyphen_files) + '\n') A_ : List[str] = [file for file in filepaths if os.sep not in file] if nodir_files: print(f'''{len(nodir_files)} files are not in a directory:''') print('\n'.join(nodir_files) + '\n') A_ : Any = len(upper_files + space_files + hyphen_files + nodir_files) if bad_files: import sys sys.exit(bad_files)
64
from pathlib import Path import cva import numpy as np from matplotlib import pyplot as plt def UpperCamelCase (lowercase_: np.ndarray , lowercase_: np.ndarray , lowercase_: np.ndarray , lowercase_: int , lowercase_: int ) -> np.ndarray: A__ : Any = cva.getAffineTransform(lowercase_ , lowercase_ ) return cva.warpAffine(lowercase_ , lowercase_ , (rows, cols) ) if __name__ == "__main__": # read original image A_ : List[Any] = cva.imread( str(Path(__file__).resolve().parent.parent / 'image_data' / 'lena.jpg') ) # turn image in gray scale value A_ : List[Any] = cva.cvtColor(image, cva.COLOR_BGR2GRAY) # get image shape A_ , A_ : Optional[Any] = gray_img.shape # set different points to rotate image A_ : str = np.array([[50, 50], [200, 50], [50, 200]], np.floataa) A_ : Dict = np.array([[10, 100], [200, 50], [100, 250]], np.floataa) A_ : Optional[int] = np.array([[50, 50], [150, 50], [120, 200]], np.floataa) A_ : Optional[int] = np.array([[10, 100], [80, 50], [180, 250]], np.floataa) # add all rotated images in a list A_ : Dict = [ gray_img, get_rotation(gray_img, ptsa, ptsa, img_rows, img_cols), get_rotation(gray_img, ptsa, ptsa, img_rows, img_cols), get_rotation(gray_img, ptsa, ptsa, img_rows, img_cols), ] # plot different image rotations A_ : Union[str, Any] = plt.figure(1) A_ : Union[str, Any] = ['Original', 'Rotation 1', 'Rotation 2', 'Rotation 3'] for i, image in enumerate(images): plt.subplot(2, 2, i + 1), plt.imshow(image, 'gray') plt.title(titles[i]) plt.axis('off') plt.subplots_adjust(left=0.0, bottom=0.05, right=1.0, top=0.95) plt.show()
64
1
def UpperCamelCase (lowercase_: int = 1000000 ) -> int: A__ : Optional[int] = set(range(3 , lowercase_ , 2 ) ) primes.add(2 ) for p in range(3 , lowercase_ , 2 ): if p not in primes: continue primes.difference_update(set(range(p * p , lowercase_ , lowercase_ ) ) ) A__ : str = [float(lowercase_ ) for n in range(limit + 1 )] for p in primes: for n in range(lowercase_ , limit + 1 , lowercase_ ): phi[n] *= 1 - 1 / p return int(sum(phi[2:] ) ) if __name__ == "__main__": print(f'''{solution() = }''')
64
import os import tempfile import unittest from pathlib import Path from transformers import AutoConfig, is_torch_available from transformers.testing_utils import require_torch, torch_device if is_torch_available(): from transformers import PyTorchBenchmark, PyTorchBenchmarkArguments @require_torch class _a (unittest.TestCase ): '''simple docstring''' def __A ( self , A__ ): for model_result in results.values(): for batch_size, sequence_length in zip(model_result["""bs"""] , model_result["""ss"""] ): A__ : str = model_result["""result"""][batch_size][sequence_length] self.assertIsNotNone(A__ ) def __A ( self ): A__ : Dict = """sshleifer/tiny-gpt2""" A__ : Tuple = PyTorchBenchmarkArguments( models=[MODEL_ID] , training=A__ , inference=A__ , sequence_lengths=[8] , batch_sizes=[1] , multi_process=A__ , ) A__ : int = PyTorchBenchmark(A__ ) A__ : List[Any] = benchmark.run() self.check_results_dict_not_empty(results.time_inference_result ) self.check_results_dict_not_empty(results.memory_inference_result ) def __A ( self ): A__ : Dict = """sgugger/tiny-distilbert-classification""" A__ : Dict = PyTorchBenchmarkArguments( models=[MODEL_ID] , training=A__ , inference=A__ , sequence_lengths=[8] , batch_sizes=[1] , multi_process=A__ , only_pretrain_model=A__ , ) A__ : str = PyTorchBenchmark(A__ ) A__ : List[str] = benchmark.run() self.check_results_dict_not_empty(results.time_inference_result ) self.check_results_dict_not_empty(results.memory_inference_result ) def __A ( self ): A__ : Any = """sshleifer/tiny-gpt2""" A__ : List[Any] = PyTorchBenchmarkArguments( models=[MODEL_ID] , training=A__ , inference=A__ , torchscript=A__ , sequence_lengths=[8] , batch_sizes=[1] , multi_process=A__ , ) A__ : Tuple = PyTorchBenchmark(A__ ) A__ : str = benchmark.run() self.check_results_dict_not_empty(results.time_inference_result ) self.check_results_dict_not_empty(results.memory_inference_result ) @unittest.skipIf(torch_device == """cpu""" , """Cant do half precision""" ) def __A ( self ): A__ : Optional[Any] = """sshleifer/tiny-gpt2""" A__ : Optional[Any] = PyTorchBenchmarkArguments( models=[MODEL_ID] , training=A__ , inference=A__ , fpaa=A__ , sequence_lengths=[8] , batch_sizes=[1] , multi_process=A__ , ) A__ : str = PyTorchBenchmark(A__ ) A__ : Any = benchmark.run() self.check_results_dict_not_empty(results.time_inference_result ) self.check_results_dict_not_empty(results.memory_inference_result ) def __A ( self ): A__ : Optional[Any] = """sshleifer/tiny-gpt2""" A__ : Tuple = AutoConfig.from_pretrained(A__ ) # set architectures equal to `None` A__ : List[Any] = None A__ : str = PyTorchBenchmarkArguments( models=[MODEL_ID] , training=A__ , inference=A__ , sequence_lengths=[8] , batch_sizes=[1] , multi_process=A__ , ) A__ : List[str] = PyTorchBenchmark(A__ , configs=[config] ) A__ : Any = benchmark.run() self.check_results_dict_not_empty(results.time_inference_result ) self.check_results_dict_not_empty(results.memory_inference_result ) def __A ( self ): A__ : Optional[int] = """sshleifer/tiny-gpt2""" A__ : Optional[int] = PyTorchBenchmarkArguments( models=[MODEL_ID] , training=A__ , inference=A__ , sequence_lengths=[8] , batch_sizes=[1] , multi_process=A__ , ) A__ : Any = PyTorchBenchmark(A__ ) A__ : Dict = benchmark.run() self.check_results_dict_not_empty(results.time_train_result ) self.check_results_dict_not_empty(results.memory_train_result ) @unittest.skipIf(torch_device == """cpu""" , """Can't do half precision""" ) def __A ( self ): A__ : Optional[int] = """sshleifer/tiny-gpt2""" A__ : List[str] = PyTorchBenchmarkArguments( models=[MODEL_ID] , training=A__ , inference=A__ , sequence_lengths=[8] , batch_sizes=[1] , fpaa=A__ , multi_process=A__ , ) A__ : Dict = PyTorchBenchmark(A__ ) A__ : Optional[int] = benchmark.run() self.check_results_dict_not_empty(results.time_train_result ) self.check_results_dict_not_empty(results.memory_train_result ) def __A ( self ): A__ : int = """sshleifer/tiny-gpt2""" A__ : Optional[int] = AutoConfig.from_pretrained(A__ ) A__ : str = PyTorchBenchmarkArguments( models=[MODEL_ID] , training=A__ , inference=A__ , sequence_lengths=[8] , batch_sizes=[1] , multi_process=A__ , ) A__ : int = PyTorchBenchmark(A__ , configs=[config] ) A__ : Tuple = benchmark.run() self.check_results_dict_not_empty(results.time_inference_result ) self.check_results_dict_not_empty(results.memory_inference_result ) def __A ( self ): A__ : List[str] = """sshleifer/tinier_bart""" A__ : List[str] = AutoConfig.from_pretrained(A__ ) A__ : List[str] = PyTorchBenchmarkArguments( models=[MODEL_ID] , training=A__ , inference=A__ , sequence_lengths=[8] , batch_sizes=[1] , multi_process=A__ , ) A__ : Union[str, Any] = PyTorchBenchmark(A__ , configs=[config] ) A__ : str = benchmark.run() self.check_results_dict_not_empty(results.time_inference_result ) self.check_results_dict_not_empty(results.memory_inference_result ) def __A ( self ): A__ : Optional[int] = """sshleifer/tiny-gpt2""" A__ : Union[str, Any] = AutoConfig.from_pretrained(A__ ) A__ : Tuple = PyTorchBenchmarkArguments( models=[MODEL_ID] , training=A__ , inference=A__ , sequence_lengths=[8] , batch_sizes=[1] , multi_process=A__ , ) A__ : int = PyTorchBenchmark(A__ , configs=[config] ) A__ : Tuple = benchmark.run() self.check_results_dict_not_empty(results.time_train_result ) self.check_results_dict_not_empty(results.memory_train_result ) def __A ( self ): A__ : Dict = """sshleifer/tinier_bart""" A__ : int = AutoConfig.from_pretrained(A__ ) A__ : Union[str, Any] = PyTorchBenchmarkArguments( models=[MODEL_ID] , training=A__ , inference=A__ , sequence_lengths=[8] , batch_sizes=[1] , multi_process=A__ , ) A__ : List[Any] = PyTorchBenchmark(A__ , configs=[config] ) A__ : Optional[Any] = benchmark.run() self.check_results_dict_not_empty(results.time_train_result ) self.check_results_dict_not_empty(results.memory_train_result ) def __A ( self ): A__ : int = """sshleifer/tiny-gpt2""" with tempfile.TemporaryDirectory() as tmp_dir: A__ : Dict = PyTorchBenchmarkArguments( models=[MODEL_ID] , training=A__ , inference=A__ , save_to_csv=A__ , sequence_lengths=[8] , batch_sizes=[1] , inference_time_csv_file=os.path.join(A__ , """inf_time.csv""" ) , train_memory_csv_file=os.path.join(A__ , """train_mem.csv""" ) , inference_memory_csv_file=os.path.join(A__ , """inf_mem.csv""" ) , train_time_csv_file=os.path.join(A__ , """train_time.csv""" ) , env_info_csv_file=os.path.join(A__ , """env.csv""" ) , multi_process=A__ , ) A__ : Optional[Any] = PyTorchBenchmark(A__ ) benchmark.run() self.assertTrue(Path(os.path.join(A__ , """inf_time.csv""" ) ).exists() ) self.assertTrue(Path(os.path.join(A__ , """train_time.csv""" ) ).exists() ) self.assertTrue(Path(os.path.join(A__ , """inf_mem.csv""" ) ).exists() ) self.assertTrue(Path(os.path.join(A__ , """train_mem.csv""" ) ).exists() ) self.assertTrue(Path(os.path.join(A__ , """env.csv""" ) ).exists() ) def __A ( self ): A__ : Optional[int] = """sshleifer/tiny-gpt2""" def _check_summary_is_not_empty(A__ ): self.assertTrue(hasattr(A__ , """sequential""" ) ) self.assertTrue(hasattr(A__ , """cumulative""" ) ) self.assertTrue(hasattr(A__ , """current""" ) ) self.assertTrue(hasattr(A__ , """total""" ) ) with tempfile.TemporaryDirectory() as tmp_dir: A__ : Dict = PyTorchBenchmarkArguments( models=[MODEL_ID] , training=A__ , inference=A__ , sequence_lengths=[8] , batch_sizes=[1] , log_filename=os.path.join(A__ , """log.txt""" ) , log_print=A__ , trace_memory_line_by_line=A__ , multi_process=A__ , ) A__ : Dict = PyTorchBenchmark(A__ ) A__ : str = benchmark.run() _check_summary_is_not_empty(result.inference_summary ) _check_summary_is_not_empty(result.train_summary ) self.assertTrue(Path(os.path.join(A__ , """log.txt""" ) ).exists() )
64
1
import io import json import fsspec import pytest from datasets import Dataset, DatasetDict, Features, NamedSplit, Value from datasets.io.json import JsonDatasetReader, JsonDatasetWriter from ..utils import assert_arrow_memory_doesnt_increase, assert_arrow_memory_increases def UpperCamelCase (lowercase_: Dict , lowercase_: Union[str, Any] ) -> Union[str, Any]: assert isinstance(lowercase_ , lowercase_ ) assert dataset.num_rows == 4 assert dataset.num_columns == 3 assert dataset.column_names == ["col_1", "col_2", "col_3"] for feature, expected_dtype in expected_features.items(): assert dataset.features[feature].dtype == expected_dtype @pytest.mark.parametrize("""keep_in_memory""" , [False, True] ) def UpperCamelCase (lowercase_: Dict , lowercase_: Optional[int] , lowercase_: Tuple ) -> Optional[int]: A__ : Dict = tmp_path / """cache""" A__ : str = {"""col_1""": """string""", """col_2""": """int64""", """col_3""": """float64"""} with assert_arrow_memory_increases() if keep_in_memory else assert_arrow_memory_doesnt_increase(): A__ : Union[str, Any] = JsonDatasetReader(lowercase_ , cache_dir=lowercase_ , keep_in_memory=lowercase_ ).read() _check_json_dataset(lowercase_ , lowercase_ ) @pytest.mark.parametrize( """features""" , [ None, {"""col_1""": """string""", """col_2""": """int64""", """col_3""": """float64"""}, {"""col_1""": """string""", """col_2""": """string""", """col_3""": """string"""}, {"""col_1""": """int32""", """col_2""": """int32""", """col_3""": """int32"""}, {"""col_1""": """float32""", """col_2""": """float32""", """col_3""": """float32"""}, ] , ) def UpperCamelCase (lowercase_: int , lowercase_: Optional[int] , lowercase_: List[str] ) -> int: A__ : Union[str, Any] = tmp_path / """cache""" A__ : Union[str, Any] = {"""col_1""": """string""", """col_2""": """int64""", """col_3""": """float64"""} A__ : Any = features.copy() if features else default_expected_features A__ : List[Any] = ( Features({feature: Value(lowercase_ ) for feature, dtype in features.items()} ) if features is not None else None ) A__ : List[Any] = JsonDatasetReader(lowercase_ , features=lowercase_ , cache_dir=lowercase_ ).read() _check_json_dataset(lowercase_ , lowercase_ ) @pytest.mark.parametrize( """features""" , [ None, {"""col_3""": """float64""", """col_1""": """string""", """col_2""": """int64"""}, ] , ) def UpperCamelCase (lowercase_: str , lowercase_: List[Any] , lowercase_: List[str] ) -> Dict: A__ : Optional[int] = tmp_path / """cache""" A__ : Optional[int] = {"""col_3""": """float64""", """col_1""": """string""", """col_2""": """int64"""} A__ : Optional[int] = features.copy() if features else default_expected_features A__ : int = ( Features({feature: Value(lowercase_ ) for feature, dtype in features.items()} ) if features is not None else None ) A__ : Union[str, Any] = JsonDatasetReader(lowercase_ , features=lowercase_ , cache_dir=lowercase_ ).read() assert isinstance(lowercase_ , lowercase_ ) assert dataset.num_rows == 2 assert dataset.num_columns == 3 assert dataset.column_names == ["col_3", "col_1", "col_2"] for feature, expected_dtype in expected_features.items(): assert dataset.features[feature].dtype == expected_dtype def UpperCamelCase (lowercase_: int , lowercase_: Tuple ) -> int: # jsonl_312_path features are {"col_3": "float64", "col_1": "string", "col_2": "int64"} A__ : str = {"""col_2""": """int64""", """col_3""": """float64""", """col_1""": """string"""} A__ : Optional[int] = features.copy() A__ : Optional[int] = ( Features({feature: Value(lowercase_ ) for feature, dtype in features.items()} ) if features is not None else None ) A__ : Union[str, Any] = tmp_path / """cache""" A__ : Dict = JsonDatasetReader(lowercase_ , features=lowercase_ , cache_dir=lowercase_ ).read() assert isinstance(lowercase_ , lowercase_ ) assert dataset.num_rows == 2 assert dataset.num_columns == 3 assert dataset.column_names == ["col_2", "col_3", "col_1"] for feature, expected_dtype in expected_features.items(): assert dataset.features[feature].dtype == expected_dtype @pytest.mark.parametrize("""split""" , [None, NamedSplit("""train""" ), """train""", """test"""] ) def UpperCamelCase (lowercase_: Dict , lowercase_: Dict , lowercase_: Optional[Any] ) -> Optional[Any]: A__ : Union[str, Any] = tmp_path / """cache""" A__ : Tuple = {"""col_1""": """string""", """col_2""": """int64""", """col_3""": """float64"""} A__ : List[str] = JsonDatasetReader(lowercase_ , cache_dir=lowercase_ , split=lowercase_ ).read() _check_json_dataset(lowercase_ , lowercase_ ) assert dataset.split == split if split else "train" @pytest.mark.parametrize("""path_type""" , [str, list] ) def UpperCamelCase (lowercase_: str , lowercase_: Union[str, Any] , lowercase_: List[str] ) -> List[Any]: if issubclass(lowercase_ , lowercase_ ): A__ : Optional[int] = jsonl_path elif issubclass(lowercase_ , lowercase_ ): A__ : Tuple = [jsonl_path] A__ : int = tmp_path / """cache""" A__ : Tuple = {"""col_1""": """string""", """col_2""": """int64""", """col_3""": """float64"""} A__ : Tuple = JsonDatasetReader(lowercase_ , cache_dir=lowercase_ ).read() _check_json_dataset(lowercase_ , lowercase_ ) def UpperCamelCase (lowercase_: int , lowercase_: Dict , lowercase_: Any=("train",) ) -> Optional[int]: assert isinstance(lowercase_ , lowercase_ ) for split in splits: A__ : Optional[Any] = dataset_dict[split] assert dataset.num_rows == 4 assert dataset.num_columns == 3 assert dataset.column_names == ["col_1", "col_2", "col_3"] for feature, expected_dtype in expected_features.items(): assert dataset.features[feature].dtype == expected_dtype @pytest.mark.parametrize("""keep_in_memory""" , [False, True] ) def UpperCamelCase (lowercase_: Optional[Any] , lowercase_: Any , lowercase_: int ) -> Dict: A__ : int = tmp_path / """cache""" A__ : Optional[int] = {"""col_1""": """string""", """col_2""": """int64""", """col_3""": """float64"""} with assert_arrow_memory_increases() if keep_in_memory else assert_arrow_memory_doesnt_increase(): A__ : List[str] = JsonDatasetReader({"""train""": jsonl_path} , cache_dir=lowercase_ , keep_in_memory=lowercase_ ).read() _check_json_datasetdict(lowercase_ , lowercase_ ) @pytest.mark.parametrize( """features""" , [ None, {"""col_1""": """string""", """col_2""": """int64""", """col_3""": """float64"""}, {"""col_1""": """string""", """col_2""": """string""", """col_3""": """string"""}, {"""col_1""": """int32""", """col_2""": """int32""", """col_3""": """int32"""}, {"""col_1""": """float32""", """col_2""": """float32""", """col_3""": """float32"""}, ] , ) def UpperCamelCase (lowercase_: str , lowercase_: Dict , lowercase_: Optional[int] ) -> int: A__ : List[Any] = tmp_path / """cache""" A__ : List[str] = {"""col_1""": """string""", """col_2""": """int64""", """col_3""": """float64"""} A__ : str = features.copy() if features else default_expected_features A__ : Optional[int] = ( Features({feature: Value(lowercase_ ) for feature, dtype in features.items()} ) if features is not None else None ) A__ : str = JsonDatasetReader({"""train""": jsonl_path} , features=lowercase_ , cache_dir=lowercase_ ).read() _check_json_datasetdict(lowercase_ , lowercase_ ) @pytest.mark.parametrize("""split""" , [None, NamedSplit("""train""" ), """train""", """test"""] ) def UpperCamelCase (lowercase_: Union[str, Any] , lowercase_: Any , lowercase_: int ) -> List[str]: if split: A__ : List[Any] = {split: jsonl_path} else: A__ : List[str] = """train""" A__ : Optional[Any] = {"""train""": jsonl_path, """test""": jsonl_path} A__ : Any = tmp_path / """cache""" A__ : Any = {"""col_1""": """string""", """col_2""": """int64""", """col_3""": """float64"""} A__ : List[Any] = JsonDatasetReader(lowercase_ , cache_dir=lowercase_ ).read() _check_json_datasetdict(lowercase_ , lowercase_ , splits=list(path.keys() ) ) assert all(dataset[split].split == split for split in path.keys() ) def UpperCamelCase (lowercase_: Optional[int] ) -> Optional[Any]: return json.load(lowercase_ ) def UpperCamelCase (lowercase_: Union[str, Any] ) -> Optional[int]: return [json.loads(lowercase_ ) for line in buffer] class _a : '''simple docstring''' @pytest.mark.parametrize("""lines, load_json_function""" , [(True, load_json_lines), (False, load_json)] ) def __A ( self , A__ , A__ , A__ ): with io.BytesIO() as buffer: JsonDatasetWriter(A__ , A__ , lines=A__ ).write() buffer.seek(0 ) A__ : Union[str, Any] = load_json_function(A__ ) assert isinstance(A__ , A__ ) assert isinstance(exported_content[0] , A__ ) assert len(A__ ) == 10 @pytest.mark.parametrize( """orient, container, keys, len_at""" , [ ("""records""", list, {"""tokens""", """labels""", """answers""", """id"""}, None), ("""split""", dict, {"""columns""", """data"""}, """data"""), ("""index""", dict, set("""0123456789""" ), None), ("""columns""", dict, {"""tokens""", """labels""", """answers""", """id"""}, """tokens"""), ("""values""", list, None, None), ("""table""", dict, {"""schema""", """data"""}, """data"""), ] , ) def __A ( self , A__ , A__ , A__ , A__ , A__ ): with io.BytesIO() as buffer: JsonDatasetWriter(A__ , A__ , lines=A__ , orient=A__ ).write() buffer.seek(0 ) A__ : Any = load_json(A__ ) assert isinstance(A__ , A__ ) if keys: if container is dict: assert exported_content.keys() == keys else: assert exported_content[0].keys() == keys else: assert not hasattr(A__ , """keys""" ) and not hasattr(exported_content[0] , """keys""" ) if len_at: assert len(exported_content[len_at] ) == 10 else: assert len(A__ ) == 10 @pytest.mark.parametrize("""lines, load_json_function""" , [(True, load_json_lines), (False, load_json)] ) def __A ( self , A__ , A__ , A__ ): with io.BytesIO() as buffer: JsonDatasetWriter(A__ , A__ , lines=A__ , num_proc=2 ).write() buffer.seek(0 ) A__ : List[str] = load_json_function(A__ ) assert isinstance(A__ , A__ ) assert isinstance(exported_content[0] , A__ ) assert len(A__ ) == 10 @pytest.mark.parametrize( """orient, container, keys, len_at""" , [ ("""records""", list, {"""tokens""", """labels""", """answers""", """id"""}, None), ("""split""", dict, {"""columns""", """data"""}, """data"""), ("""index""", dict, set("""0123456789""" ), None), ("""columns""", dict, {"""tokens""", """labels""", """answers""", """id"""}, """tokens"""), ("""values""", list, None, None), ("""table""", dict, {"""schema""", """data"""}, """data"""), ] , ) def __A ( self , A__ , A__ , A__ , A__ , A__ ): with io.BytesIO() as buffer: JsonDatasetWriter(A__ , A__ , lines=A__ , orient=A__ , num_proc=2 ).write() buffer.seek(0 ) A__ : List[Any] = load_json(A__ ) assert isinstance(A__ , A__ ) if keys: if container is dict: assert exported_content.keys() == keys else: assert exported_content[0].keys() == keys else: assert not hasattr(A__ , """keys""" ) and not hasattr(exported_content[0] , """keys""" ) if len_at: assert len(exported_content[len_at] ) == 10 else: assert len(A__ ) == 10 def __A ( self , A__ ): with pytest.raises(A__ ): with io.BytesIO() as buffer: JsonDatasetWriter(A__ , A__ , num_proc=0 ) @pytest.mark.parametrize("""compression, extension""" , [("""gzip""", """gz"""), ("""bz2""", """bz2"""), ("""xz""", """xz""")] ) def __A ( self , A__ , A__ , A__ , A__ , A__ ): A__ : Dict = tmp_path_factory.mktemp("""data""" ) / F"""test.json.{extension}""" A__ : Optional[int] = str(shared_datadir / F"""test_file.json.{extension}""" ) JsonDatasetWriter(A__ , A__ , compression=A__ ).write() with fsspec.open(A__ , """rb""" , compression="""infer""" ) as f: A__ : Dict = f.read() with fsspec.open(A__ , """rb""" , compression="""infer""" ) as f: A__ : Optional[Any] = f.read() assert exported_content == original_content
64
# tests directory-specific settings - this file is run automatically # by pytest before any tests are run import doctest import sys import warnings from os.path import abspath, dirname, join import _pytest from transformers.testing_utils import HfDoctestModule, HfDocTestParser # allow having multiple repository checkouts and not needing to remember to rerun # 'pip install -e .[dev]' when switching between checkouts and running tests. A_ : Optional[int] = abspath(join(dirname(__file__), 'src')) sys.path.insert(1, git_repo_path) # silence FutureWarning warnings in tests since often we can't act on them until # they become normal warnings - i.e. the tests still need to test the current functionality warnings.simplefilter(action='ignore', category=FutureWarning) def UpperCamelCase (lowercase_: List[str] ) -> Any: config.addinivalue_line( """markers""" , """is_pt_tf_cross_test: mark test to run only when PT and TF interactions are tested""" ) config.addinivalue_line( """markers""" , """is_pt_flax_cross_test: mark test to run only when PT and FLAX interactions are tested""" ) config.addinivalue_line("""markers""" , """is_pipeline_test: mark test to run only when pipelines are tested""" ) config.addinivalue_line("""markers""" , """is_staging_test: mark test to run only in the staging environment""" ) config.addinivalue_line("""markers""" , """accelerate_tests: mark test that require accelerate""" ) config.addinivalue_line("""markers""" , """tool_tests: mark the tool tests that are run on their specific schedule""" ) def UpperCamelCase (lowercase_: Optional[int] ) -> Optional[Any]: from transformers.testing_utils import pytest_addoption_shared pytest_addoption_shared(lowercase_ ) def UpperCamelCase (lowercase_: List[str] ) -> Optional[Any]: from transformers.testing_utils import pytest_terminal_summary_main A__ : List[Any] = terminalreporter.config.getoption("""--make-reports""" ) if make_reports: pytest_terminal_summary_main(lowercase_ , id=lowercase_ ) def UpperCamelCase (lowercase_: Union[str, Any] , lowercase_: int ) -> List[str]: # If no tests are collected, pytest exists with code 5, which makes the CI fail. if exitstatus == 5: A__ : Tuple = 0 # Doctest custom flag to ignore output. A_ : Tuple = doctest.register_optionflag('IGNORE_RESULT') A_ : Dict = doctest.OutputChecker class _a (__magic_name__ ): '''simple docstring''' def __A ( self , A__ , A__ , A__ ): if IGNORE_RESULT & optionflags: return True return OutputChecker.check_output(self , A__ , A__ , A__ ) A_ : str = CustomOutputChecker A_ : Dict = HfDoctestModule A_ : Optional[int] = HfDocTestParser
64
1
# This model implementation is heavily inspired by https://github.com/haofanwang/ControlNet-for-Diffusers/ import gc import random import tempfile import unittest import numpy as np import torch from PIL import Image from transformers import CLIPTextConfig, CLIPTextModel, CLIPTokenizer from diffusers import ( AutoencoderKL, ControlNetModel, DDIMScheduler, StableDiffusionControlNetImgaImgPipeline, UNetaDConditionModel, ) from diffusers.pipelines.stable_diffusion.pipeline_stable_diffusion_controlnet import MultiControlNetModel from diffusers.utils import floats_tensor, load_image, load_numpy, randn_tensor, slow, torch_device from diffusers.utils.import_utils import is_xformers_available from diffusers.utils.testing_utils import enable_full_determinism, require_torch_gpu from ..pipeline_params import ( IMAGE_TO_IMAGE_IMAGE_PARAMS, TEXT_GUIDED_IMAGE_VARIATION_BATCH_PARAMS, TEXT_GUIDED_IMAGE_VARIATION_PARAMS, ) from ..test_pipelines_common import ( PipelineKarrasSchedulerTesterMixin, PipelineLatentTesterMixin, PipelineTesterMixin, ) enable_full_determinism() class _a (__magic_name__ , __magic_name__ , __magic_name__ , unittest.TestCase ): '''simple docstring''' UpperCAmelCase__: int = StableDiffusionControlNetImgaImgPipeline UpperCAmelCase__: int = TEXT_GUIDED_IMAGE_VARIATION_PARAMS - {'''height''', '''width'''} UpperCAmelCase__: Any = TEXT_GUIDED_IMAGE_VARIATION_BATCH_PARAMS UpperCAmelCase__: Optional[Any] = IMAGE_TO_IMAGE_IMAGE_PARAMS.union({'''control_image'''} ) UpperCAmelCase__: Union[str, Any] = IMAGE_TO_IMAGE_IMAGE_PARAMS def __A ( self ): torch.manual_seed(0 ) A__ : List[Any] = UNetaDConditionModel( block_out_channels=(32, 64) , layers_per_block=2 , sample_size=32 , in_channels=4 , out_channels=4 , down_block_types=("""DownBlock2D""", """CrossAttnDownBlock2D""") , up_block_types=("""CrossAttnUpBlock2D""", """UpBlock2D""") , cross_attention_dim=32 , ) torch.manual_seed(0 ) A__ : List[str] = ControlNetModel( block_out_channels=(32, 64) , layers_per_block=2 , in_channels=4 , down_block_types=("""DownBlock2D""", """CrossAttnDownBlock2D""") , cross_attention_dim=32 , conditioning_embedding_out_channels=(16, 32) , ) torch.manual_seed(0 ) A__ : Optional[Any] = DDIMScheduler( beta_start=0.0_0_0_8_5 , beta_end=0.0_1_2 , beta_schedule="""scaled_linear""" , clip_sample=A__ , set_alpha_to_one=A__ , ) torch.manual_seed(0 ) A__ : Dict = AutoencoderKL( block_out_channels=[32, 64] , in_channels=3 , out_channels=3 , down_block_types=["""DownEncoderBlock2D""", """DownEncoderBlock2D"""] , up_block_types=["""UpDecoderBlock2D""", """UpDecoderBlock2D"""] , latent_channels=4 , ) torch.manual_seed(0 ) A__ : List[str] = CLIPTextConfig( bos_token_id=0 , eos_token_id=2 , hidden_size=32 , intermediate_size=37 , layer_norm_eps=1e-05 , num_attention_heads=4 , num_hidden_layers=5 , pad_token_id=1 , vocab_size=1000 , ) A__ : Tuple = CLIPTextModel(A__ ) A__ : Dict = CLIPTokenizer.from_pretrained("""hf-internal-testing/tiny-random-clip""" ) A__ : Tuple = { """unet""": unet, """controlnet""": controlnet, """scheduler""": scheduler, """vae""": vae, """text_encoder""": text_encoder, """tokenizer""": tokenizer, """safety_checker""": None, """feature_extractor""": None, } return components def __A ( self , A__ , A__=0 ): if str(A__ ).startswith("""mps""" ): A__ : Union[str, Any] = torch.manual_seed(A__ ) else: A__ : Optional[Any] = torch.Generator(device=A__ ).manual_seed(A__ ) A__ : Dict = 2 A__ : Tuple = randn_tensor( (1, 3, 32 * controlnet_embedder_scale_factor, 32 * controlnet_embedder_scale_factor) , generator=A__ , device=torch.device(A__ ) , ) A__ : List[str] = floats_tensor(control_image.shape , rng=random.Random(A__ ) ).to(A__ ) A__ : Union[str, Any] = image.cpu().permute(0 , 2 , 3 , 1 )[0] A__ : str = Image.fromarray(np.uinta(A__ ) ).convert("""RGB""" ).resize((64, 64) ) A__ : Union[str, Any] = { """prompt""": """A painting of a squirrel eating a burger""", """generator""": generator, """num_inference_steps""": 2, """guidance_scale""": 6.0, """output_type""": """numpy""", """image""": image, """control_image""": control_image, } return inputs def __A ( self ): return self._test_attention_slicing_forward_pass(expected_max_diff=2e-3 ) @unittest.skipIf( torch_device != """cuda""" or not is_xformers_available() , reason="""XFormers attention is only available with CUDA and `xformers` installed""" , ) def __A ( self ): self._test_xformers_attention_forwardGenerator_pass(expected_max_diff=2e-3 ) def __A ( self ): self._test_inference_batch_single_identical(expected_max_diff=2e-3 ) class _a (__magic_name__ , __magic_name__ , unittest.TestCase ): '''simple docstring''' UpperCAmelCase__: Dict = StableDiffusionControlNetImgaImgPipeline UpperCAmelCase__: str = TEXT_GUIDED_IMAGE_VARIATION_PARAMS - {'''height''', '''width'''} UpperCAmelCase__: str = TEXT_GUIDED_IMAGE_VARIATION_BATCH_PARAMS UpperCAmelCase__: Tuple = frozenset([] ) # TO_DO: add image_params once refactored VaeImageProcessor.preprocess def __A ( self ): torch.manual_seed(0 ) A__ : str = UNetaDConditionModel( block_out_channels=(32, 64) , layers_per_block=2 , sample_size=32 , in_channels=4 , out_channels=4 , down_block_types=("""DownBlock2D""", """CrossAttnDownBlock2D""") , up_block_types=("""CrossAttnUpBlock2D""", """UpBlock2D""") , cross_attention_dim=32 , ) torch.manual_seed(0 ) def init_weights(A__ ): if isinstance(A__ , torch.nn.Convad ): torch.nn.init.normal(m.weight ) m.bias.data.fill_(1.0 ) A__ : Optional[int] = ControlNetModel( block_out_channels=(32, 64) , layers_per_block=2 , in_channels=4 , down_block_types=("""DownBlock2D""", """CrossAttnDownBlock2D""") , cross_attention_dim=32 , conditioning_embedding_out_channels=(16, 32) , ) controlneta.controlnet_down_blocks.apply(A__ ) torch.manual_seed(0 ) A__ : Tuple = ControlNetModel( block_out_channels=(32, 64) , layers_per_block=2 , in_channels=4 , down_block_types=("""DownBlock2D""", """CrossAttnDownBlock2D""") , cross_attention_dim=32 , conditioning_embedding_out_channels=(16, 32) , ) controlneta.controlnet_down_blocks.apply(A__ ) torch.manual_seed(0 ) A__ : Optional[int] = DDIMScheduler( beta_start=0.0_0_0_8_5 , beta_end=0.0_1_2 , beta_schedule="""scaled_linear""" , clip_sample=A__ , set_alpha_to_one=A__ , ) torch.manual_seed(0 ) A__ : Any = AutoencoderKL( block_out_channels=[32, 64] , in_channels=3 , out_channels=3 , down_block_types=["""DownEncoderBlock2D""", """DownEncoderBlock2D"""] , up_block_types=["""UpDecoderBlock2D""", """UpDecoderBlock2D"""] , latent_channels=4 , ) torch.manual_seed(0 ) A__ : List[Any] = CLIPTextConfig( bos_token_id=0 , eos_token_id=2 , hidden_size=32 , intermediate_size=37 , layer_norm_eps=1e-05 , num_attention_heads=4 , num_hidden_layers=5 , pad_token_id=1 , vocab_size=1000 , ) A__ : Any = CLIPTextModel(A__ ) A__ : Optional[Any] = CLIPTokenizer.from_pretrained("""hf-internal-testing/tiny-random-clip""" ) A__ : Optional[Any] = MultiControlNetModel([controlneta, controlneta] ) A__ : Any = { """unet""": unet, """controlnet""": controlnet, """scheduler""": scheduler, """vae""": vae, """text_encoder""": text_encoder, """tokenizer""": tokenizer, """safety_checker""": None, """feature_extractor""": None, } return components def __A ( self , A__ , A__=0 ): if str(A__ ).startswith("""mps""" ): A__ : List[Any] = torch.manual_seed(A__ ) else: A__ : List[Any] = torch.Generator(device=A__ ).manual_seed(A__ ) A__ : List[str] = 2 A__ : Optional[int] = [ randn_tensor( (1, 3, 32 * controlnet_embedder_scale_factor, 32 * controlnet_embedder_scale_factor) , generator=A__ , device=torch.device(A__ ) , ), randn_tensor( (1, 3, 32 * controlnet_embedder_scale_factor, 32 * controlnet_embedder_scale_factor) , generator=A__ , device=torch.device(A__ ) , ), ] A__ : List[Any] = floats_tensor(control_image[0].shape , rng=random.Random(A__ ) ).to(A__ ) A__ : Optional[Any] = image.cpu().permute(0 , 2 , 3 , 1 )[0] A__ : Union[str, Any] = Image.fromarray(np.uinta(A__ ) ).convert("""RGB""" ).resize((64, 64) ) A__ : List[str] = { """prompt""": """A painting of a squirrel eating a burger""", """generator""": generator, """num_inference_steps""": 2, """guidance_scale""": 6.0, """output_type""": """numpy""", """image""": image, """control_image""": control_image, } return inputs def __A ( self ): A__ : List[str] = self.get_dummy_components() A__ : Tuple = self.pipeline_class(**A__ ) pipe.to(A__ ) A__ : List[str] = 1_0.0 A__ : Union[str, Any] = 4 A__ : Union[str, Any] = self.get_dummy_inputs(A__ ) A__ : str = steps A__ : Dict = scale A__ : Optional[Any] = pipe(**A__ )[0] A__ : Union[str, Any] = self.get_dummy_inputs(A__ ) A__ : Dict = steps A__ : Dict = scale A__ : List[str] = pipe(**A__ , control_guidance_start=0.1 , control_guidance_end=0.2 )[0] A__ : str = self.get_dummy_inputs(A__ ) A__ : Dict = steps A__ : Optional[int] = scale A__ : Dict = pipe(**A__ , control_guidance_start=[0.1, 0.3] , control_guidance_end=[0.2, 0.7] )[0] A__ : List[str] = self.get_dummy_inputs(A__ ) A__ : List[str] = steps A__ : Optional[int] = scale A__ : int = pipe(**A__ , control_guidance_start=0.4 , control_guidance_end=[0.5, 0.8] )[0] # make sure that all outputs are different assert np.sum(np.abs(output_a - output_a ) ) > 1e-3 assert np.sum(np.abs(output_a - output_a ) ) > 1e-3 assert np.sum(np.abs(output_a - output_a ) ) > 1e-3 def __A ( self ): return self._test_attention_slicing_forward_pass(expected_max_diff=2e-3 ) @unittest.skipIf( torch_device != """cuda""" or not is_xformers_available() , reason="""XFormers attention is only available with CUDA and `xformers` installed""" , ) def __A ( self ): self._test_xformers_attention_forwardGenerator_pass(expected_max_diff=2e-3 ) def __A ( self ): self._test_inference_batch_single_identical(expected_max_diff=2e-3 ) def __A ( self ): A__ : Optional[Any] = self.get_dummy_components() A__ : Optional[int] = self.pipeline_class(**A__ ) pipe.to(A__ ) pipe.set_progress_bar_config(disable=A__ ) with tempfile.TemporaryDirectory() as tmpdir: try: # save_pretrained is not implemented for Multi-ControlNet pipe.save_pretrained(A__ ) except NotImplementedError: pass @slow @require_torch_gpu class _a (unittest.TestCase ): '''simple docstring''' def __A ( self ): super().tearDown() gc.collect() torch.cuda.empty_cache() def __A ( self ): A__ : Dict = ControlNetModel.from_pretrained("""lllyasviel/sd-controlnet-canny""" ) A__ : Union[str, Any] = StableDiffusionControlNetImgaImgPipeline.from_pretrained( """runwayml/stable-diffusion-v1-5""" , safety_checker=A__ , controlnet=A__ ) pipe.enable_model_cpu_offload() pipe.set_progress_bar_config(disable=A__ ) A__ : str = torch.Generator(device="""cpu""" ).manual_seed(0 ) A__ : List[Any] = """evil space-punk bird""" A__ : Union[str, Any] = load_image( """https://huggingface.co/datasets/hf-internal-testing/diffusers-images/resolve/main/sd_controlnet/bird_canny.png""" ).resize((512, 512) ) A__ : Optional[int] = load_image( """https://huggingface.co/lllyasviel/sd-controlnet-canny/resolve/main/images/bird.png""" ).resize((512, 512) ) A__ : List[str] = pipe( A__ , A__ , control_image=A__ , generator=A__ , output_type="""np""" , num_inference_steps=50 , strength=0.6 , ) A__ : Optional[Any] = output.images[0] assert image.shape == (512, 512, 3) A__ : Union[str, Any] = load_numpy( """https://huggingface.co/datasets/hf-internal-testing/diffusers-images/resolve/main/sd_controlnet/img2img.npy""" ) assert np.abs(expected_image - image ).max() < 9e-2
64
from __future__ import annotations import unittest from transformers import AutoTokenizer, PegasusConfig, is_tf_available from transformers.testing_utils import require_sentencepiece, require_tf, require_tokenizers, slow from transformers.utils import cached_property from ...test_configuration_common import ConfigTester from ...test_modeling_tf_common import TFModelTesterMixin, ids_tensor from ...test_pipeline_mixin import PipelineTesterMixin if is_tf_available(): import tensorflow as tf from transformers import TFAutoModelForSeqaSeqLM, TFPegasusForConditionalGeneration, TFPegasusModel @require_tf class _a : '''simple docstring''' UpperCAmelCase__: List[Any] = PegasusConfig UpperCAmelCase__: Optional[int] = {} UpperCAmelCase__: List[str] = '''gelu''' def __init__( self , A__ , A__=13 , A__=7 , A__=True , A__=False , A__=99 , A__=32 , A__=2 , A__=4 , A__=37 , A__=0.1 , A__=0.1 , A__=40 , A__=2 , A__=1 , A__=0 , ): A__ : Dict = parent A__ : Dict = batch_size A__ : Any = seq_length A__ : Optional[Any] = is_training A__ : int = use_labels A__ : Any = vocab_size A__ : Union[str, Any] = hidden_size A__ : Tuple = num_hidden_layers A__ : Tuple = num_attention_heads A__ : List[Any] = intermediate_size A__ : Union[str, Any] = hidden_dropout_prob A__ : Optional[Any] = attention_probs_dropout_prob A__ : List[Any] = max_position_embeddings A__ : Any = eos_token_id A__ : List[Any] = pad_token_id A__ : List[Any] = bos_token_id def __A ( self ): A__ : str = ids_tensor([self.batch_size, self.seq_length - 1] , self.vocab_size ) A__ : Dict = tf.expand_dims(tf.constant([self.eos_token_id] * self.batch_size ) , 1 ) A__ : List[Any] = tf.concat([input_ids, eos_tensor] , axis=1 ) A__ : Union[str, Any] = ids_tensor([self.batch_size, self.seq_length] , self.vocab_size ) A__ : Tuple = self.config_cls( vocab_size=self.vocab_size , d_model=self.hidden_size , encoder_layers=self.num_hidden_layers , decoder_layers=self.num_hidden_layers , encoder_attention_heads=self.num_attention_heads , decoder_attention_heads=self.num_attention_heads , encoder_ffn_dim=self.intermediate_size , decoder_ffn_dim=self.intermediate_size , dropout=self.hidden_dropout_prob , attention_dropout=self.attention_probs_dropout_prob , max_position_embeddings=self.max_position_embeddings , eos_token_ids=[2] , bos_token_id=self.bos_token_id , pad_token_id=self.pad_token_id , decoder_start_token_id=self.pad_token_id , **self.config_updates , ) A__ : str = prepare_pegasus_inputs_dict(A__ , A__ , A__ ) return config, inputs_dict def __A ( self , A__ , A__ ): A__ : int = TFPegasusModel(config=A__ ).get_decoder() A__ : List[Any] = inputs_dict["""input_ids"""] A__ : Any = input_ids[:1, :] A__ : Optional[Any] = inputs_dict["""attention_mask"""][:1, :] A__ : Optional[int] = inputs_dict["""head_mask"""] A__ : Any = 1 # first forward pass A__ : Tuple = model(A__ , attention_mask=A__ , head_mask=A__ , use_cache=A__ ) A__ , A__ : Dict = outputs.to_tuple() # create hypothetical next token and extent to next_input_ids A__ : Tuple = ids_tensor((self.batch_size, 3) , config.vocab_size ) A__ : Optional[Any] = tf.cast(ids_tensor((self.batch_size, 3) , 2 ) , tf.inta ) # append to next input_ids and A__ : List[Any] = tf.concat([input_ids, next_tokens] , axis=-1 ) A__ : Tuple = tf.concat([attention_mask, next_attn_mask] , axis=-1 ) A__ : Optional[Any] = model(A__ , attention_mask=A__ )[0] A__ : Any = model(A__ , attention_mask=A__ , past_key_values=A__ )[0] self.parent.assertEqual(next_tokens.shape[1] , output_from_past.shape[1] ) # select random slice A__ : int = int(ids_tensor((1,) , output_from_past.shape[-1] ) ) A__ : Any = output_from_no_past[:, -3:, random_slice_idx] A__ : Tuple = output_from_past[:, :, random_slice_idx] # test that outputs are equal for slice tf.debugging.assert_near(A__ , A__ , rtol=1e-3 ) def UpperCamelCase (lowercase_: Union[str, Any] , lowercase_: Dict , lowercase_: List[Any] , lowercase_: Dict=None , lowercase_: int=None , lowercase_: List[Any]=None , lowercase_: List[Any]=None , lowercase_: str=None , ) -> int: if attention_mask is None: A__ : List[str] = tf.cast(tf.math.not_equal(lowercase_ , config.pad_token_id ) , tf.inta ) if decoder_attention_mask is None: A__ : Dict = tf.concat( [ tf.ones(decoder_input_ids[:, :1].shape , dtype=tf.inta ), tf.cast(tf.math.not_equal(decoder_input_ids[:, 1:] , config.pad_token_id ) , tf.inta ), ] , axis=-1 , ) if head_mask is None: A__ : Any = tf.ones((config.encoder_layers, config.encoder_attention_heads) ) if decoder_head_mask is None: A__ : Tuple = tf.ones((config.decoder_layers, config.decoder_attention_heads) ) if cross_attn_head_mask is None: A__ : List[str] = tf.ones((config.decoder_layers, config.decoder_attention_heads) ) return { "input_ids": input_ids, "decoder_input_ids": decoder_input_ids, "attention_mask": attention_mask, "decoder_attention_mask": decoder_attention_mask, "head_mask": head_mask, "decoder_head_mask": decoder_head_mask, "cross_attn_head_mask": cross_attn_head_mask, } @require_tf class _a (__magic_name__ , __magic_name__ , unittest.TestCase ): '''simple docstring''' UpperCAmelCase__: List[Any] = (TFPegasusForConditionalGeneration, TFPegasusModel) if is_tf_available() else () UpperCAmelCase__: Tuple = (TFPegasusForConditionalGeneration,) if is_tf_available() else () UpperCAmelCase__: Tuple = ( { '''conversational''': TFPegasusForConditionalGeneration, '''feature-extraction''': TFPegasusModel, '''summarization''': TFPegasusForConditionalGeneration, '''text2text-generation''': TFPegasusForConditionalGeneration, '''translation''': TFPegasusForConditionalGeneration, } if is_tf_available() else {} ) UpperCAmelCase__: int = True UpperCAmelCase__: Union[str, Any] = False UpperCAmelCase__: List[str] = False def __A ( self ): A__ : Optional[Any] = TFPegasusModelTester(self ) A__ : Tuple = ConfigTester(self , config_class=A__ ) def __A ( self ): self.config_tester.run_common_tests() def __A ( self ): A__ : int = self.model_tester.prepare_config_and_inputs_for_common() self.model_tester.check_decoder_model_past_large_inputs(*A__ ) @require_sentencepiece @require_tokenizers @require_tf class _a (unittest.TestCase ): '''simple docstring''' UpperCAmelCase__: Optional[int] = [ ''' PG&E stated it scheduled the blackouts in response to forecasts for high winds amid dry conditions. The aim is to reduce the risk of wildfires. Nearly 800 thousand customers were scheduled to be affected by the shutoffs which were expected to last through at least midday tomorrow.''', ''' The London trio are up for best UK act and best album, as well as getting two nominations in the best song category."We got told like this morning \'Oh I think you\'re nominated\'", said Dappy."And I was like \'Oh yeah, which one?\' And now we\'ve got nominated for four awards. I mean, wow!"Bandmate Fazer added: "We thought it\'s best of us to come down and mingle with everyone and say hello to the cameras. And now we find we\'ve got four nominations."The band have two shots at the best song prize, getting the nod for their Tynchy Stryder collaboration Number One, and single Strong Again.Their album Uncle B will also go up against records by the likes of Beyonce and Kanye West.N-Dubz picked up the best newcomer Mobo in 2007, but female member Tulisa said they wouldn\'t be too disappointed if they didn\'t win this time around."At the end of the day we\'re grateful to be where we are in our careers."If it don\'t happen then it don\'t happen - live to fight another day and keep on making albums and hits for the fans."Dappy also revealed they could be performing live several times on the night.The group will be doing Number One and also a possible rendition of the War Child single, I Got Soul.The charity song is a re-working of The Killers\' All These Things That I\'ve Done and is set to feature artists like Chipmunk, Ironik and Pixie Lott.This year\'s Mobos will be held outside of London for the first time, in Glasgow on 30 September.N-Dubz said they were looking forward to performing for their Scottish fans and boasted about their recent shows north of the border."We just done Edinburgh the other day," said Dappy."We smashed up an N-Dubz show over there. We done Aberdeen about three or four months ago - we smashed up that show over there! Everywhere we go we smash it up!" ''', ] UpperCAmelCase__: Any = [ '''California\'s largest electricity provider has cut power to hundreds of thousands of customers in an effort to''' ''' reduce the risk of wildfires.''', '''N-Dubz have revealed they\'re "grateful" to have been nominated for four Mobo Awards.''', ] # differs slightly from pytorch, likely due to numerical differences in linear layers UpperCAmelCase__: List[str] = '''google/pegasus-xsum''' @cached_property def __A ( self ): return AutoTokenizer.from_pretrained(self.model_name ) @cached_property def __A ( self ): A__ : int = TFAutoModelForSeqaSeqLM.from_pretrained(self.model_name ) return model def __A ( self , **A__ ): A__ : str = self.translate_src_text(**A__ ) assert self.expected_text == generated_words def __A ( self , **A__ ): A__ : List[str] = self.tokenizer(self.src_text , **A__ , padding=A__ , return_tensors="""tf""" ) A__ : Optional[int] = self.model.generate( model_inputs.input_ids , attention_mask=model_inputs.attention_mask , num_beams=2 , use_cache=A__ , ) A__ : Dict = self.tokenizer.batch_decode(generated_ids.numpy() , skip_special_tokens=A__ ) return generated_words @slow def __A ( self ): self._assert_generated_batch_equal_expected()
64
1
from ...configuration_utils import PretrainedConfig from ...utils import logging A_ : Optional[Any] = logging.get_logger(__name__) A_ : List[Any] = { 'microsoft/biogpt': 'https://huggingface.co/microsoft/biogpt/resolve/main/config.json', # See all BioGPT models at https://huggingface.co/models?filter=biogpt } class _a (__magic_name__ ): '''simple docstring''' UpperCAmelCase__: List[str] = '''biogpt''' def __init__( self , A__=4_2384 , A__=1024 , A__=24 , A__=16 , A__=4096 , A__="gelu" , A__=0.1 , A__=0.1 , A__=1024 , A__=0.0_2 , A__=1e-12 , A__=True , A__=True , A__=0.0 , A__=0.0 , A__=1 , A__=0 , A__=2 , **A__ , ): A__ : Optional[int] = vocab_size A__ : List[str] = max_position_embeddings A__ : Optional[int] = hidden_size A__ : Tuple = num_hidden_layers A__ : Optional[Any] = num_attention_heads A__ : Optional[int] = intermediate_size A__ : Dict = hidden_act A__ : List[Any] = hidden_dropout_prob A__ : Union[str, Any] = attention_probs_dropout_prob A__ : Union[str, Any] = initializer_range A__ : Optional[int] = layer_norm_eps A__ : Tuple = scale_embedding A__ : Tuple = use_cache A__ : Dict = layerdrop A__ : Optional[Any] = activation_dropout super().__init__(pad_token_id=A__ , bos_token_id=A__ , eos_token_id=A__ , **A__ )
64
class _a : '''simple docstring''' def __init__( self ): A__ : str = """""" A__ : Any = """""" A__ : List[Any] = [] def __A ( self , A__ , A__ ): if m == -1: return n + 1 elif n == -1: return m + 1 elif self.dp[m][n] > -1: return self.dp[m][n] else: if self.worda[m] == self.worda[n]: A__ : Optional[Any] = self.__min_dist_top_down_dp(m - 1 , n - 1 ) else: A__ : Union[str, Any] = self.__min_dist_top_down_dp(A__ , n - 1 ) A__ : Union[str, Any] = self.__min_dist_top_down_dp(m - 1 , A__ ) A__ : Union[str, Any] = self.__min_dist_top_down_dp(m - 1 , n - 1 ) A__ : List[Any] = 1 + min(A__ , A__ , A__ ) return self.dp[m][n] def __A ( self , A__ , A__ ): A__ : Tuple = worda A__ : Dict = worda A__ : Optional[Any] = [[-1 for _ in range(len(A__ ) )] for _ in range(len(A__ ) )] return self.__min_dist_top_down_dp(len(A__ ) - 1 , len(A__ ) - 1 ) def __A ( self , A__ , A__ ): A__ : Optional[Any] = worda A__ : Dict = worda A__ : Union[str, Any] = len(A__ ) A__ : List[str] = len(A__ ) A__ : int = [[0 for _ in range(n + 1 )] for _ in range(m + 1 )] for i in range(m + 1 ): for j in range(n + 1 ): if i == 0: # first string is empty A__ : Tuple = j elif j == 0: # second string is empty A__ : Dict = i elif worda[i - 1] == worda[j - 1]: # last characters are equal A__ : str = self.dp[i - 1][j - 1] else: A__ : Union[str, Any] = self.dp[i][j - 1] A__ : str = self.dp[i - 1][j] A__ : Union[str, Any] = self.dp[i - 1][j - 1] A__ : Tuple = 1 + min(A__ , A__ , A__ ) return self.dp[m][n] if __name__ == "__main__": A_ : Union[str, Any] = EditDistance() print('****************** Testing Edit Distance DP Algorithm ******************') print() A_ : int = input('Enter the first string: ').strip() A_ : List[str] = input('Enter the second string: ').strip() print() print(f'''The minimum edit distance is: {solver.min_dist_top_down(Sa, Sa)}''') print(f'''The minimum edit distance is: {solver.min_dist_bottom_up(Sa, Sa)}''') print() print('*************** End of Testing Edit Distance DP Algorithm ***************')
64
1
# tests directory-specific settings - this file is run automatically # by pytest before any tests are run import sys import warnings from os.path import abspath, dirname, join # allow having multiple repository checkouts and not needing to remember to rerun # 'pip install -e .[dev]' when switching between checkouts and running tests. A_ : Dict = abspath(join(dirname(dirname(__file__)), 'src')) sys.path.insert(1, git_repo_path) # silence FutureWarning warnings in tests since often we can't act on them until # they become normal warnings - i.e. the tests still need to test the current functionality warnings.simplefilter(action='ignore', category=FutureWarning) def UpperCamelCase (lowercase_: Optional[Any] ) -> str: from diffusers.utils.testing_utils import pytest_addoption_shared pytest_addoption_shared(lowercase_ ) def UpperCamelCase (lowercase_: Any ) -> Any: from diffusers.utils.testing_utils import pytest_terminal_summary_main A__ : List[str] = terminalreporter.config.getoption("""--make-reports""" ) if make_reports: pytest_terminal_summary_main(lowercase_ , id=lowercase_ )
64
def UpperCamelCase (lowercase_: int , lowercase_: int ) -> int: while second != 0: A__ : int = first & second first ^= second A__ : int = c << 1 return first if __name__ == "__main__": import doctest doctest.testmod() A_ : Optional[Any] = int(input('Enter the first number: ').strip()) A_ : List[str] = int(input('Enter the second number: ').strip()) print(f'''{add(first, second) = }''')
64
1
import argparse import json import os import tensorstore as ts import torch from flax import serialization from flax.traverse_util import flatten_dict, unflatten_dict from tensorflow.io import gfile from transformers.modeling_utils import dtype_byte_size from transformers.models.switch_transformers.convert_switch_transformers_original_flax_checkpoint_to_pytorch import ( rename_keys, ) from transformers.utils import WEIGHTS_INDEX_NAME, WEIGHTS_NAME from transformers.utils.hub import convert_file_size_to_int def UpperCamelCase (lowercase_: List[str] , lowercase_: str ) -> Optional[Any]: if flax_key_tuple[-1] == "kernel" and flax_tensor.ndim == 3: # expert layer A__ : Union[str, Any] = flax_key_tuple[:-1] + ("""weight""",) A__ : Optional[int] = torch.permute(lowercase_ , (0, 2, 1) ) elif flax_key_tuple[-1] == "kernel" and ".".join(lowercase_ ): # linear layer A__ : Optional[Any] = flax_key_tuple[:-1] + ("""weight""",) A__ : int = flax_tensor.T elif flax_key_tuple[-1] in ["scale", "embedding"]: A__ : Optional[int] = flax_key_tuple[:-1] + ("""weight""",) return flax_key_tuple, flax_tensor def UpperCamelCase (lowercase_: Tuple , lowercase_: Optional[int] , lowercase_: str ) -> Union[str, Any]: if "metadata" in layer: A__ : Tuple = layer.split("""metadata""" ) A__ : Optional[Any] = """""".join(split_layer[0] )[:-1] A__ : Optional[Any] = [tuple(("""metadata""" + split_layer[1]).split("""/""" ) )] elif "kvstore" in layer: A__ : str = layer.split("""kvstore""" ) A__ : int = """""".join(split_layer[0] )[:-1] A__ : Optional[int] = [tuple(("""kvstore""" + split_layer[1]).split("""/""" ) )] else: A__ : Any = layer.split("""/""" ) A__ : int = """/""".join(split_layer[:-1] ) A__ : str = (split_layer[-1],) if "kvstore/path" in layer: A__ : Dict = f"""{switch_checkpoint_path}/{checkpoint_info[layer]}""" elif "kvstore/driver" in layer: A__ : Optional[int] = """file""" else: A__ : str = checkpoint_info[layer] return curr_real_layer_name, split_layer, content def UpperCamelCase (lowercase_: str , lowercase_: List[Any] ) -> int: A__ : int = rename_keys(lowercase_ ) A__ : Any = {} for k, v in current_block.items(): A__ : Dict = v A__ : str = new_current_block torch.save(lowercase_ , lowercase_ ) def UpperCamelCase (lowercase_: Dict , lowercase_: Optional[Any] , lowercase_: Optional[Any] , lowercase_: Optional[int] , lowercase_: str = WEIGHTS_NAME ) -> Tuple: A__ : Optional[int] = convert_file_size_to_int(lowercase_ ) A__ : List[Any] = [] A__ : int = {} A__ : List[str] = 0 A__ : Any = 0 os.makedirs(lowercase_ , exist_ok=lowercase_ ) with gfile.GFile(switch_checkpoint_path + """/checkpoint""" , """rb""" ) as fp: A__ : Optional[Any] = serialization.msgpack_restore(fp.read() )["""optimizer"""]["""target"""] A__ : Dict = flatten_dict(lowercase_ , sep="""/""" ) A__ : Any = {} for layer in checkpoint_info.keys(): A__ , A__ , A__ : Union[str, Any] = get_key_and_tensorstore_dict( lowercase_ , lowercase_ , lowercase_ ) if curr_real_layer_name in all_layers: A__ : Optional[int] = content else: A__ : List[Any] = {split_layer[-1]: content} for key in all_layers.keys(): # open tensorstore file A__ : Optional[Any] = ts.open(unflatten_dict(all_layers[key] ) ).result().read().result() A__ : List[Any] = torch.tensor(lowercase_ ) A__ : List[Any] = raw_weights.numel() * dtype_byte_size(raw_weights.dtype ) # use the renaming pattern from the small conversion scripts A__ , A__ : Any = rename_base_flax_keys(tuple(key.split("""/""" ) ) , lowercase_ ) A__ : Any = """/""".join(lowercase_ ) # If this weight is going to tip up over the maximal size, we split. if current_block_size + weight_size > max_shard_size: A__ : List[Any] = os.path.join( lowercase_ , weights_name.replace(""".bin""" , f"""-{len(lowercase_ )+1:05d}-of-???.bin""" ) ) rename_and_save_block(lowercase_ , lowercase_ ) sharded_state_dicts.append(current_block.keys() ) del current_block A__ : Any = {} A__ : str = 0 A__ : List[str] = raw_weights.to(getattr(lowercase_ , lowercase_ ) ) current_block_size += weight_size total_size += weight_size # Add the last block A__ : Union[str, Any] = os.path.join(lowercase_ , weights_name.replace(""".bin""" , f"""-{len(lowercase_ )+1:05d}-of-???.bin""" ) ) rename_and_save_block(lowercase_ , lowercase_ ) sharded_state_dicts.append(current_block.keys() ) # If we only have one shard, we return it if len(lowercase_ ) == 1: return {weights_name: sharded_state_dicts[0]}, None # Otherwise, let's build the index A__ : str = {} A__ : Any = {} for idx, shard in enumerate(lowercase_ ): A__ : Any = weights_name.replace( """.bin""" , f"""-{idx+1:05d}-of-{len(lowercase_ ):05d}.bin""" ) # len(sharded_state_dicts):05d} A__ : Dict = os.path.join(lowercase_ , weights_name.replace(""".bin""" , f"""-{idx+1:05d}-of-???.bin""" ) ) os.rename(lowercase_ , os.path.join(lowercase_ , lowercase_ ) ) A__ : str = shard for key in shard: A__ : Any = shard_file # Add the metadata A__ : Tuple = {"""total_size""": total_size} A__ : Union[str, Any] = {"""metadata""": metadata, """weight_map""": weight_map} with open(os.path.join(lowercase_ , lowercase_ ) , """w""" , encoding="""utf-8""" ) as f: A__ : Dict = json.dumps(lowercase_ , indent=2 , sort_keys=lowercase_ ) + """\n""" f.write(lowercase_ ) return metadata, index if __name__ == "__main__": A_ : Optional[int] = argparse.ArgumentParser() # Required parameters parser.add_argument( '--switch_t5x_checkpoint_path', default='/mnt/disks/disk_switch/original_checkpoints/switch-xxl-128/checkpoint_634600', type=str, required=False, help='Path to a directory containing a folder per layer. Follows the original Google format.', ) parser.add_argument('--max_shard_size', default='10GB', required=False, help='Max shard size') parser.add_argument('--dtype', default='bfloat16', type=str, required=False, help='dtype of the saved model') parser.add_argument( '--pytorch_dump_folder_path', default='/mnt/disks/disk_switch/original_checkpoints/switch-xxl-128-converted', type=str, required=False, help='Path to the output pytorch model.', ) A_ : Dict = parser.parse_args() shard_on_the_fly( args.switch_tax_checkpoint_path, args.pytorch_dump_folder_path, args.max_shard_size, args.dtype, ) def UpperCamelCase () -> int: from transformers import SwitchTransformersConfig, SwitchTransformersForConditionalGeneration, TaTokenizer A__ : str = SwitchTransformersConfig.from_pretrained("""google/switch-base-8""" ) config.save_pretrained("""/home/arthur_huggingface_co/transformers/switch_converted""" ) A__ : str = SwitchTransformersForConditionalGeneration.from_pretrained( """/home/arthur_huggingface_co/transformers/switch_converted""" , device_map="""auto""" ) A__ : Tuple = TaTokenizer.from_pretrained("""t5-small""" ) A__ : Dict = """A <extra_id_0> walks into a bar a orders a <extra_id_1> with <extra_id_2> pinch of <extra_id_3>.""" A__ : Union[str, Any] = tokenizer(lowercase_ , return_tensors="""pt""" ).input_ids A__ : Tuple = model.generate(lowercase_ , decoder_start_token_id=0 ) print(tokenizer.decode(out[0] ) )
64
from __future__ import annotations from collections.abc import Callable A_ : List[Any] = list[list[float | int]] def UpperCamelCase (lowercase_: Matrix , lowercase_: Matrix ) -> Matrix: A__ : int = len(lowercase_ ) A__ : Matrix = [[0 for _ in range(size + 1 )] for _ in range(lowercase_ )] A__ : int A__ : int A__ : int A__ : int A__ : int A__ : float for row in range(lowercase_ ): for col in range(lowercase_ ): A__ : List[str] = matrix[row][col] A__ : int = vector[row][0] A__ : Optional[int] = 0 A__ : str = 0 while row < size and col < size: # pivoting A__ : int = max((abs(augmented[rowa][col] ), rowa) for rowa in range(lowercase_ , lowercase_ ) )[ 1 ] if augmented[pivot_row][col] == 0: col += 1 continue else: A__ , A__ : Union[str, Any] = augmented[pivot_row], augmented[row] for rowa in range(row + 1 , lowercase_ ): A__ : List[Any] = augmented[rowa][col] / augmented[row][col] A__ : Dict = 0 for cola in range(col + 1 , size + 1 ): augmented[rowa][cola] -= augmented[row][cola] * ratio row += 1 col += 1 # back substitution for col in range(1 , lowercase_ ): for row in range(lowercase_ ): A__ : List[str] = augmented[row][col] / augmented[col][col] for cola in range(lowercase_ , size + 1 ): augmented[row][cola] -= augmented[col][cola] * ratio # round to get rid of numbers like 2.000000000000004 return [ [round(augmented[row][size] / augmented[row][row] , 10 )] for row in range(lowercase_ ) ] def UpperCamelCase (lowercase_: list[int] ) -> Callable[[int], int]: A__ : int = len(lowercase_ ) A__ : Matrix = [[0 for _ in range(lowercase_ )] for _ in range(lowercase_ )] A__ : Matrix = [[0] for _ in range(lowercase_ )] A__ : Matrix A__ : int A__ : int A__ : int for x_val, y_val in enumerate(lowercase_ ): for col in range(lowercase_ ): A__ : Dict = (x_val + 1) ** (size - col - 1) A__ : Any = y_val A__ : Union[str, Any] = solve(lowercase_ , lowercase_ ) def interpolated_func(lowercase_: int ) -> int: return sum( round(coeffs[x_val][0] ) * (var ** (size - x_val - 1)) for x_val in range(lowercase_ ) ) return interpolated_func def UpperCamelCase (lowercase_: int ) -> int: return ( 1 - variable + variable**2 - variable**3 + variable**4 - variable**5 + variable**6 - variable**7 + variable**8 - variable**9 + variable**10 ) def UpperCamelCase (lowercase_: Callable[[int], int] = question_function , lowercase_: int = 10 ) -> int: A__ : list[int] = [func(lowercase_ ) for x_val in range(1 , order + 1 )] A__ : list[Callable[[int], int]] = [ interpolate(data_points[:max_coeff] ) for max_coeff in range(1 , order + 1 ) ] A__ : int = 0 A__ : Callable[[int], int] A__ : int for poly in polynomials: A__ : List[str] = 1 while func(lowercase_ ) == poly(lowercase_ ): x_val += 1 ret += poly(lowercase_ ) return ret if __name__ == "__main__": print(f'''{solution() = }''')
64
1
import json import os import subprocess import unittest from ast import literal_eval import pytest from parameterized import parameterized, parameterized_class from . import is_sagemaker_available if is_sagemaker_available(): from sagemaker import Session, TrainingJobAnalytics from sagemaker.huggingface import HuggingFace @pytest.mark.skipif( literal_eval(os.getenv('''TEST_SAGEMAKER''' , '''False''' ) ) is not True , reason='''Skipping test because should only be run when releasing minor transformers version''' , ) @pytest.mark.usefixtures('''sm_env''' ) @parameterized_class( [ { '''framework''': '''pytorch''', '''script''': '''run_glue_model_parallelism.py''', '''model_name_or_path''': '''roberta-large''', '''instance_type''': '''ml.p3dn.24xlarge''', '''results''': {'''train_runtime''': 16_00, '''eval_accuracy''': 0.3, '''eval_loss''': 1.2}, }, { '''framework''': '''pytorch''', '''script''': '''run_glue.py''', '''model_name_or_path''': '''roberta-large''', '''instance_type''': '''ml.p3dn.24xlarge''', '''results''': {'''train_runtime''': 16_00, '''eval_accuracy''': 0.3, '''eval_loss''': 1.2}, }, ] ) class _a (unittest.TestCase ): '''simple docstring''' def __A ( self ): if self.framework == "pytorch": subprocess.run( F"""cp ./examples/pytorch/text-classification/run_glue.py {self.env.test_path}/run_glue.py""".split() , encoding="""utf-8""" , check=A__ , ) assert hasattr(self , """env""" ) def __A ( self , A__ ): # configuration for running training on smdistributed Model Parallel A__ : List[str] = { """enabled""": True, """processes_per_host""": 8, } A__ : Union[str, Any] = { """enabled""": True, """parameters""": { """microbatches""": 4, """placement_strategy""": """spread""", """pipeline""": """interleaved""", """optimize""": """speed""", """partitions""": 4, """ddp""": True, }, } A__ : Dict = {"""smdistributed""": {"""modelparallel""": smp_options}, """mpi""": mpi_options} A__ : Tuple = """trainer""" if self.script == """run_glue.py""" else """smtrainer""" # creates estimator return HuggingFace( entry_point=self.script , source_dir=self.env.test_path , role=self.env.role , image_uri=self.env.image_uri , base_job_name=F"""{self.env.base_job_name}-{instance_count}-smp-{name_extension}""" , instance_count=A__ , instance_type=self.instance_type , debugger_hook_config=A__ , hyperparameters={ **self.env.hyperparameters, """model_name_or_path""": self.model_name_or_path, """max_steps""": 500, } , metric_definitions=self.env.metric_definitions , distribution=A__ , py_version="""py36""" , ) def __A ( self , A__ ): TrainingJobAnalytics(A__ ).export_csv(F"""{self.env.test_path}/{job_name}_metrics.csv""" ) @parameterized.expand([(1,)] ) def __A ( self , A__ ): # create estimator A__ : int = self.create_estimator(A__ ) # run training estimator.fit() # result dataframe A__ : str = TrainingJobAnalytics(estimator.latest_training_job.name ).dataframe() # extract kpis A__ : Tuple = list(result_metrics_df[result_metrics_df.metric_name == """eval_accuracy"""]["""value"""] ) A__ : Union[str, Any] = list(result_metrics_df[result_metrics_df.metric_name == """eval_loss"""]["""value"""] ) # get train time from SageMaker job, this includes starting, preprocessing, stopping A__ : Optional[Any] = ( Session().describe_training_job(estimator.latest_training_job.name ).get("""TrainingTimeInSeconds""" , 99_9999 ) ) # assert kpis assert train_runtime <= self.results["train_runtime"] assert all(t >= self.results["""eval_accuracy"""] for t in eval_accuracy ) assert all(t <= self.results["""eval_loss"""] for t in eval_loss ) # dump tests result into json file to share in PR with open(F"""{estimator.latest_training_job.name}.json""" , """w""" ) as outfile: json.dump({"""train_time""": train_runtime, """eval_accuracy""": eval_accuracy, """eval_loss""": eval_loss} , A__ )
64
from functools import lru_cache @lru_cache def UpperCamelCase (lowercase_: int ) -> int: if num < 0: raise ValueError("""Number should not be negative.""" ) return 1 if num in (0, 1) else num * factorial(num - 1 ) if __name__ == "__main__": import doctest doctest.testmod()
64
1
from __future__ import annotations from collections.abc import Callable from typing import Any, Generic, TypeVar A_ : Dict = TypeVar('T') class _a (Generic[T] ): '''simple docstring''' def __init__( self , A__ , A__ ): A__ : Any | T = None A__ : int = len(A__ ) A__ : list[T] = [any_type for _ in range(self.N )] + arr A__ : Any = fnc self.build() def __A ( self ): for p in range(self.N - 1 , 0 , -1 ): A__ : str = self.fn(self.st[p * 2] , self.st[p * 2 + 1] ) def __A ( self , A__ , A__ ): p += self.N A__ : Dict = v while p > 1: A__ : List[str] = p // 2 A__ : Dict = self.fn(self.st[p * 2] , self.st[p * 2 + 1] ) def __A ( self , A__ , A__ ): # noqa: E741 A__ , A__ : List[Any] = l + self.N, r + self.N A__ : T | None = None while l <= r: if l % 2 == 1: A__ : List[str] = self.st[l] if res is None else self.fn(A__ , self.st[l] ) if r % 2 == 0: A__ : List[str] = self.st[r] if res is None else self.fn(A__ , self.st[r] ) A__ , A__ : Union[str, Any] = (l + 1) // 2, (r - 1) // 2 return res if __name__ == "__main__": from functools import reduce A_ : Optional[int] = [1, 10, -2, 9, -3, 8, 4, -7, 5, 6, 11, -12] A_ : List[Any] = { 0: 7, 1: 2, 2: 6, 3: -14, 4: 5, 5: 4, 6: 7, 7: -10, 8: 9, 9: 10, 10: 12, 11: 1, } A_ : Dict = SegmentTree(test_array, min) A_ : str = SegmentTree(test_array, max) A_ : str = SegmentTree(test_array, lambda a, b: a + b) def UpperCamelCase () -> None: for i in range(len(lowercase_ ) ): for j in range(lowercase_ , len(lowercase_ ) ): A__ : List[str] = reduce(lowercase_ , test_array[i : j + 1] ) A__ : Any = reduce(lowercase_ , test_array[i : j + 1] ) A__ : Tuple = reduce(lambda lowercase_ , lowercase_ : a + b , test_array[i : j + 1] ) assert min_range == min_segment_tree.query(lowercase_ , lowercase_ ) assert max_range == max_segment_tree.query(lowercase_ , lowercase_ ) assert sum_range == sum_segment_tree.query(lowercase_ , lowercase_ ) test_all_segments() for index, value in test_updates.items(): A_ : Optional[int] = value min_segment_tree.update(index, value) max_segment_tree.update(index, value) sum_segment_tree.update(index, value) test_all_segments()
64
import os import tempfile from functools import partial from unittest import TestCase from unittest.mock import patch import datasets import datasets.config from .utils import require_beam class _a (datasets.BeamBasedBuilder ): '''simple docstring''' def __A ( self ): return datasets.DatasetInfo( features=datasets.Features({"""content""": datasets.Value("""string""" )} ) , supervised_keys=A__ , ) def __A ( self , A__ , A__ ): return [datasets.SplitGenerator(name=datasets.Split.TRAIN , gen_kwargs={"""examples""": get_test_dummy_examples()} )] def __A ( self , A__ , A__ ): import apache_beam as beam return pipeline | "Load Examples" >> beam.Create(A__ ) class _a (datasets.BeamBasedBuilder ): '''simple docstring''' def __A ( self ): return datasets.DatasetInfo( features=datasets.Features({"""a""": datasets.Sequence({"""b""": datasets.Value("""string""" )} )} ) , supervised_keys=A__ , ) def __A ( self , A__ , A__ ): return [ datasets.SplitGenerator(name=datasets.Split.TRAIN , gen_kwargs={"""examples""": get_test_nested_examples()} ) ] def __A ( self , A__ , A__ ): import apache_beam as beam return pipeline | "Load Examples" >> beam.Create(A__ ) def UpperCamelCase () -> Dict: return [(i, {"content": content}) for i, content in enumerate(["""foo""", """bar""", """foobar"""] )] def UpperCamelCase () -> Tuple: return [(i, {"a": {"b": [content]}}) for i, content in enumerate(["""foo""", """bar""", """foobar"""] )] class _a (__magic_name__ ): '''simple docstring''' @require_beam def __A ( self ): A__ : Dict = len(get_test_dummy_examples() ) with tempfile.TemporaryDirectory() as tmp_cache_dir: A__ : int = DummyBeamDataset(cache_dir=A__ , beam_runner="""DirectRunner""" ) builder.download_and_prepare() self.assertTrue( os.path.exists( os.path.join(A__ , builder.name , """default""" , """0.0.0""" , F"""{builder.name}-train.arrow""" ) ) ) self.assertDictEqual(builder.info.features , datasets.Features({"""content""": datasets.Value("""string""" )} ) ) A__ : int = builder.as_dataset() self.assertEqual(dset["""train"""].num_rows , A__ ) self.assertEqual(dset["""train"""].info.splits["""train"""].num_examples , A__ ) self.assertDictEqual(dset["""train"""][0] , get_test_dummy_examples()[0][1] ) self.assertDictEqual( dset["""train"""][expected_num_examples - 1] , get_test_dummy_examples()[expected_num_examples - 1][1] ) self.assertTrue( os.path.exists(os.path.join(A__ , builder.name , """default""" , """0.0.0""" , """dataset_info.json""" ) ) ) del dset @require_beam def __A ( self ): import apache_beam as beam A__ : int = beam.io.parquetio.WriteToParquet A__ : List[str] = len(get_test_dummy_examples() ) with tempfile.TemporaryDirectory() as tmp_cache_dir: A__ : str = DummyBeamDataset(cache_dir=A__ , beam_runner="""DirectRunner""" ) with patch("""apache_beam.io.parquetio.WriteToParquet""" ) as write_parquet_mock: A__ : Optional[Any] = partial(A__ , num_shards=2 ) builder.download_and_prepare() self.assertTrue( os.path.exists( os.path.join( A__ , builder.name , """default""" , """0.0.0""" , F"""{builder.name}-train-00000-of-00002.arrow""" ) ) ) self.assertTrue( os.path.exists( os.path.join( A__ , builder.name , """default""" , """0.0.0""" , F"""{builder.name}-train-00000-of-00002.arrow""" ) ) ) self.assertDictEqual(builder.info.features , datasets.Features({"""content""": datasets.Value("""string""" )} ) ) A__ : Optional[int] = builder.as_dataset() self.assertEqual(dset["""train"""].num_rows , A__ ) self.assertEqual(dset["""train"""].info.splits["""train"""].num_examples , A__ ) # Order is not preserved when sharding, so we just check that all the elements are there self.assertListEqual(sorted(dset["""train"""]["""content"""] ) , sorted(["""foo""", """bar""", """foobar"""] ) ) self.assertTrue( os.path.exists(os.path.join(A__ , builder.name , """default""" , """0.0.0""" , """dataset_info.json""" ) ) ) del dset @require_beam def __A ( self ): with tempfile.TemporaryDirectory() as tmp_cache_dir: A__ : int = DummyBeamDataset(cache_dir=A__ ) self.assertRaises(datasets.builder.MissingBeamOptions , builder.download_and_prepare ) @require_beam def __A ( self ): A__ : List[Any] = len(get_test_nested_examples() ) with tempfile.TemporaryDirectory() as tmp_cache_dir: A__ : Optional[int] = NestedBeamDataset(cache_dir=A__ , beam_runner="""DirectRunner""" ) builder.download_and_prepare() self.assertTrue( os.path.exists( os.path.join(A__ , builder.name , """default""" , """0.0.0""" , F"""{builder.name}-train.arrow""" ) ) ) self.assertDictEqual( builder.info.features , datasets.Features({"""a""": datasets.Sequence({"""b""": datasets.Value("""string""" )} )} ) ) A__ : Optional[int] = builder.as_dataset() self.assertEqual(dset["""train"""].num_rows , A__ ) self.assertEqual(dset["""train"""].info.splits["""train"""].num_examples , A__ ) self.assertDictEqual(dset["""train"""][0] , get_test_nested_examples()[0][1] ) self.assertDictEqual( dset["""train"""][expected_num_examples - 1] , get_test_nested_examples()[expected_num_examples - 1][1] ) self.assertTrue( os.path.exists(os.path.join(A__ , builder.name , """default""" , """0.0.0""" , """dataset_info.json""" ) ) ) del dset
64
1
from typing import Dict, Iterable, Optional, Union import numpy as np from ...image_processing_utils import BaseImageProcessor, BatchFeature, get_size_dict from ...image_transforms import normalize, rescale, resize, to_channel_dimension_format, to_pil_image from ...image_utils import ( IMAGENET_STANDARD_MEAN, IMAGENET_STANDARD_STD, ChannelDimension, ImageInput, PILImageResampling, make_list_of_images, to_numpy_array, valid_images, ) from ...utils import TensorType, is_pytesseract_available, is_vision_available, logging, requires_backends if is_vision_available(): import PIL # soft dependency if is_pytesseract_available(): import pytesseract A_ : str = logging.get_logger(__name__) def UpperCamelCase (lowercase_: Dict , lowercase_: Any , lowercase_: Optional[Any] ) -> Optional[Any]: return [ int(1000 * (box[0] / width) ), int(1000 * (box[1] / height) ), int(1000 * (box[2] / width) ), int(1000 * (box[3] / height) ), ] def UpperCamelCase (lowercase_: np.ndarray , lowercase_: Optional[str] , lowercase_: Optional[str] ) -> List[Any]: A__ : Tuple = to_pil_image(lowercase_ ) A__ , A__ : int = pil_image.size A__ : Dict = pytesseract.image_to_data(lowercase_ , lang=lowercase_ , output_type="""dict""" , config=lowercase_ ) A__ , A__ , A__ , A__ , A__ : Dict = data["""text"""], data["""left"""], data["""top"""], data["""width"""], data["""height"""] # filter empty words and corresponding coordinates A__ : List[Any] = [idx for idx, word in enumerate(lowercase_ ) if not word.strip()] A__ : Union[str, Any] = [word for idx, word in enumerate(lowercase_ ) if idx not in irrelevant_indices] A__ : Union[str, Any] = [coord for idx, coord in enumerate(lowercase_ ) if idx not in irrelevant_indices] A__ : Dict = [coord for idx, coord in enumerate(lowercase_ ) if idx not in irrelevant_indices] A__ : Any = [coord for idx, coord in enumerate(lowercase_ ) if idx not in irrelevant_indices] A__ : List[str] = [coord for idx, coord in enumerate(lowercase_ ) if idx not in irrelevant_indices] # turn coordinates into (left, top, left+width, top+height) format A__ : int = [] for x, y, w, h in zip(lowercase_ , lowercase_ , lowercase_ , lowercase_ ): A__ : Optional[Any] = [x, y, x + w, y + h] actual_boxes.append(lowercase_ ) # finally, normalize the bounding boxes A__ : Union[str, Any] = [] for box in actual_boxes: normalized_boxes.append(normalize_box(lowercase_ , lowercase_ , lowercase_ ) ) assert len(lowercase_ ) == len(lowercase_ ), "Not as many words as there are bounding boxes" return words, normalized_boxes class _a (__magic_name__ ): '''simple docstring''' UpperCAmelCase__: List[Any] = ['''pixel_values'''] def __init__( self , A__ = True , A__ = None , A__ = PILImageResampling.BILINEAR , A__ = True , A__ = 1 / 255 , A__ = True , A__ = None , A__ = None , A__ = True , A__ = None , A__ = "" , **A__ , ): super().__init__(**A__ ) A__ : Union[str, Any] = size if size is not None else {"""height""": 224, """width""": 224} A__ : Tuple = get_size_dict(A__ ) A__ : int = do_resize A__ : Union[str, Any] = size A__ : List[str] = resample A__ : List[str] = do_rescale A__ : Optional[int] = rescale_value A__ : Union[str, Any] = do_normalize A__ : List[Any] = image_mean if image_mean is not None else IMAGENET_STANDARD_MEAN A__ : Dict = image_std if image_std is not None else IMAGENET_STANDARD_STD A__ : List[str] = apply_ocr A__ : Optional[int] = ocr_lang A__ : int = tesseract_config def __A ( self , A__ , A__ , A__ = PILImageResampling.BILINEAR , A__ = None , **A__ , ): A__ : Dict = get_size_dict(A__ ) if "height" not in size or "width" not in size: raise ValueError(F"""The size dictionary must contain the keys 'height' and 'width'. Got {size.keys()}""" ) A__ : Tuple = (size["""height"""], size["""width"""]) return resize(A__ , size=A__ , resample=A__ , data_format=A__ , **A__ ) def __A ( self , A__ , A__ , A__ = None , **A__ , ): return rescale(A__ , scale=A__ , data_format=A__ , **A__ ) def __A ( self , A__ , A__ , A__ , A__ = None , **A__ , ): return normalize(A__ , mean=A__ , std=A__ , data_format=A__ , **A__ ) def __A ( self , A__ , A__ = None , A__ = None , A__=None , A__ = None , A__ = None , A__ = None , A__ = None , A__ = None , A__ = None , A__ = None , A__ = None , A__ = None , A__ = ChannelDimension.FIRST , **A__ , ): A__ : Tuple = do_resize if do_resize is not None else self.do_resize A__ : Any = size if size is not None else self.size A__ : Tuple = get_size_dict(A__ ) A__ : str = resample if resample is not None else self.resample A__ : Union[str, Any] = do_rescale if do_rescale is not None else self.do_rescale A__ : Optional[int] = rescale_factor if rescale_factor is not None else self.rescale_factor A__ : int = do_normalize if do_normalize is not None else self.do_normalize A__ : str = image_mean if image_mean is not None else self.image_mean A__ : List[Any] = image_std if image_std is not None else self.image_std A__ : Tuple = apply_ocr if apply_ocr is not None else self.apply_ocr A__ : List[str] = ocr_lang if ocr_lang is not None else self.ocr_lang A__ : Union[str, Any] = tesseract_config if tesseract_config is not None else self.tesseract_config A__ : Optional[int] = make_list_of_images(A__ ) if not valid_images(A__ ): raise ValueError( """Invalid image type. Must be of type PIL.Image.Image, numpy.ndarray, """ """torch.Tensor, tf.Tensor or jax.ndarray.""" ) if do_resize and size is None: raise ValueError("""Size must be specified if do_resize is True.""" ) if do_rescale and rescale_factor is None: raise ValueError("""Rescale factor must be specified if do_rescale is True.""" ) if do_normalize and (image_mean is None or image_std is None): raise ValueError("""If do_normalize is True, image_mean and image_std must be specified.""" ) # All transformations expect numpy arrays. A__ : List[Any] = [to_numpy_array(A__ ) for image in images] # Tesseract OCR to get words + normalized bounding boxes if apply_ocr: requires_backends(self , """pytesseract""" ) A__ : List[Any] = [] A__ : int = [] for image in images: A__ , A__ : List[Any] = apply_tesseract(A__ , A__ , A__ ) words_batch.append(A__ ) boxes_batch.append(A__ ) if do_resize: A__ : int = [self.resize(image=A__ , size=A__ , resample=A__ ) for image in images] if do_rescale: A__ : Dict = [self.rescale(image=A__ , scale=A__ ) for image in images] if do_normalize: A__ : List[Any] = [self.normalize(image=A__ , mean=A__ , std=A__ ) for image in images] A__ : str = [to_channel_dimension_format(A__ , A__ ) for image in images] A__ : Optional[int] = BatchFeature(data={"""pixel_values""": images} , tensor_type=A__ ) if apply_ocr: A__ : str = words_batch A__ : Any = boxes_batch return data
64
import warnings from ...utils import logging from .image_processing_poolformer import PoolFormerImageProcessor A_ : Union[str, Any] = logging.get_logger(__name__) class _a (__magic_name__ ): '''simple docstring''' def __init__( self , *A__ , **A__ ): warnings.warn( """The class PoolFormerFeatureExtractor is deprecated and will be removed in version 5 of Transformers.""" """ Please use PoolFormerImageProcessor instead.""" , A__ , ) super().__init__(*A__ , **A__ )
64
1
from collections import OrderedDict from typing import Mapping from ...configuration_utils import PretrainedConfig from ...onnx import OnnxConfig A_ : List[Any] = { 'albert-base-v1': 'https://huggingface.co/albert-base-v1/resolve/main/config.json', 'albert-large-v1': 'https://huggingface.co/albert-large-v1/resolve/main/config.json', 'albert-xlarge-v1': 'https://huggingface.co/albert-xlarge-v1/resolve/main/config.json', 'albert-xxlarge-v1': 'https://huggingface.co/albert-xxlarge-v1/resolve/main/config.json', 'albert-base-v2': 'https://huggingface.co/albert-base-v2/resolve/main/config.json', 'albert-large-v2': 'https://huggingface.co/albert-large-v2/resolve/main/config.json', 'albert-xlarge-v2': 'https://huggingface.co/albert-xlarge-v2/resolve/main/config.json', 'albert-xxlarge-v2': 'https://huggingface.co/albert-xxlarge-v2/resolve/main/config.json', } class _a (__magic_name__ ): '''simple docstring''' UpperCAmelCase__: List[Any] = '''albert''' def __init__( self , A__=3_0000 , A__=128 , A__=4096 , A__=12 , A__=1 , A__=64 , A__=1_6384 , A__=1 , A__="gelu_new" , A__=0 , A__=0 , A__=512 , A__=2 , A__=0.0_2 , A__=1e-12 , A__=0.1 , A__="absolute" , A__=0 , A__=2 , A__=3 , **A__ , ): super().__init__(pad_token_id=A__ , bos_token_id=A__ , eos_token_id=A__ , **A__ ) A__ : Tuple = vocab_size A__ : str = embedding_size A__ : List[Any] = hidden_size A__ : Union[str, Any] = num_hidden_layers A__ : Dict = num_hidden_groups A__ : List[str] = num_attention_heads A__ : List[Any] = inner_group_num A__ : Optional[int] = hidden_act A__ : Any = intermediate_size A__ : Tuple = hidden_dropout_prob A__ : Dict = attention_probs_dropout_prob A__ : str = max_position_embeddings A__ : Union[str, Any] = type_vocab_size A__ : str = initializer_range A__ : Tuple = layer_norm_eps A__ : Optional[Any] = classifier_dropout_prob A__ : Any = position_embedding_type class _a (__magic_name__ ): '''simple docstring''' @property def __A ( self ): if self.task == "multiple-choice": A__ : Optional[Any] = {0: """batch""", 1: """choice""", 2: """sequence"""} else: A__ : int = {0: """batch""", 1: """sequence"""} return OrderedDict( [ ("""input_ids""", dynamic_axis), ("""attention_mask""", dynamic_axis), ("""token_type_ids""", dynamic_axis), ] )
64
import inspect import logging import os import random import shutil import tempfile import unittest import pytest import torch from torch import nn from torch.utils.data import DataLoader, TensorDataset from accelerate import Accelerator from accelerate.test_utils import execute_subprocess_async, require_cuda from accelerate.utils import ProjectConfiguration, set_seed A_ : Any = logging.getLogger(__name__) def UpperCamelCase (lowercase_: Optional[Any]=2 , lowercase_: Union[str, Any]=3 , lowercase_: int=16 , lowercase_: int = 10 , lowercase_: int = 2 ) -> int: def get_dataset(lowercase_: Optional[int] ): A__ : Optional[Any] = torch.randn(batch_size * n_batches , 1 ) return TensorDataset(lowercase_ , a * x + b + 0.1 * torch.randn(batch_size * n_batches , 1 ) ) A__ : Dict = get_dataset(lowercase_ ) A__ : Any = get_dataset(lowercase_ ) A__ : Dict = DataLoader(lowercase_ , shuffle=lowercase_ , batch_size=lowercase_ , num_workers=4 ) A__ : Optional[Any] = DataLoader(lowercase_ , shuffle=lowercase_ , batch_size=lowercase_ , num_workers=4 ) return (train_dataloader, valid_dataloader) def UpperCamelCase (lowercase_: Optional[Any] , lowercase_: List[str] , lowercase_: int , lowercase_: int , lowercase_: List[str] , lowercase_: Dict=None ) -> List[Any]: A__ : List[Any] = [] for epoch in range(lowercase_ ): # Train quickly model.train() for batch in dataloader: A__ , A__ : Any = batch A__ : Any = model(lowercase_ ) A__ : Any = torch.nn.functional.mse_loss(lowercase_ , lowercase_ ) accelerator.backward(lowercase_ ) optimizer.step() optimizer.zero_grad() rands.append(random.random() ) # Introduce some randomness if scheduler is not None: scheduler.step() return rands class _a (nn.Module ): '''simple docstring''' def __init__( self ): super().__init__() A__ : str = nn.Parameter(torch.randn(1 ) ) A__ : Any = nn.Parameter(torch.randn(1 ) ) def __A ( self , A__ ): return x * self.a + self.b class _a (unittest.TestCase ): '''simple docstring''' def __A ( self ): with tempfile.TemporaryDirectory() as tmpdir: set_seed(42 ) A__ : Optional[Any] = DummyModel() A__ : Optional[Any] = torch.optim.Adam(params=model.parameters() , lr=1e-3 ) A__ , A__ : str = dummy_dataloaders() A__ : Dict = ProjectConfiguration(total_limit=1 , project_dir=A__ , automatic_checkpoint_naming=A__ ) # Train baseline A__ : List[str] = Accelerator(project_config=A__ ) A__ , A__ , A__ , A__ : Any = accelerator.prepare( A__ , A__ , A__ , A__ ) # Save initial accelerator.save_state() # Save second state accelerator.save_state() self.assertEqual(len(os.listdir(accelerator.project_dir ) ) , 1 ) def __A ( self ): with tempfile.TemporaryDirectory() as tmpdir: set_seed(42 ) A__ : str = DummyModel() A__ : Optional[int] = torch.optim.Adam(params=model.parameters() , lr=1e-3 ) A__ , A__ : int = dummy_dataloaders() # Train baseline A__ : str = Accelerator() A__ , A__ , A__ , A__ : List[str] = accelerator.prepare( A__ , A__ , A__ , A__ ) # Save initial A__ : List[Any] = os.path.join(A__ , """initial""" ) accelerator.save_state(A__ ) ((A__) , (A__)) : str = model.a.item(), model.b.item() A__ : Dict = optimizer.state_dict() A__ : List[str] = train(3 , A__ , A__ , A__ , A__ ) ((A__) , (A__)) : str = model.a.item(), model.b.item() A__ : Any = optimizer.state_dict() # Train partially set_seed(42 ) A__ : Optional[int] = DummyModel() A__ : Dict = torch.optim.Adam(params=model.parameters() , lr=1e-3 ) A__ , A__ : Dict = dummy_dataloaders() A__ : List[str] = Accelerator() A__ , A__ , A__ , A__ : Optional[Any] = accelerator.prepare( A__ , A__ , A__ , A__ ) accelerator.load_state(A__ ) ((A__) , (A__)) : Tuple = model.a.item(), model.b.item() A__ : Union[str, Any] = optimizer.state_dict() self.assertEqual(A__ , A__ ) self.assertEqual(A__ , A__ ) self.assertEqual(A__ , A__ ) A__ : List[str] = train(2 , A__ , A__ , A__ , A__ ) # Save everything A__ : Optional[int] = os.path.join(A__ , """checkpoint""" ) accelerator.save_state(A__ ) # Load everything back in and make sure all states work accelerator.load_state(A__ ) test_rands += train(1 , A__ , A__ , A__ , A__ ) ((A__) , (A__)) : Union[str, Any] = model.a.item(), model.b.item() A__ : Optional[int] = optimizer.state_dict() self.assertEqual(A__ , A__ ) self.assertEqual(A__ , A__ ) self.assertEqual(A__ , A__ ) self.assertEqual(A__ , A__ ) def __A ( self ): with tempfile.TemporaryDirectory() as tmpdir: set_seed(42 ) A__ : int = DummyModel() A__ : Optional[int] = torch.optim.Adam(params=model.parameters() , lr=1e-3 ) A__ , A__ : List[str] = dummy_dataloaders() A__ : str = ProjectConfiguration(automatic_checkpoint_naming=A__ ) # Train baseline A__ : Any = Accelerator(project_dir=A__ , project_config=A__ ) A__ , A__ , A__ , A__ : str = accelerator.prepare( A__ , A__ , A__ , A__ ) # Save initial accelerator.save_state() ((A__) , (A__)) : Tuple = model.a.item(), model.b.item() A__ : int = optimizer.state_dict() A__ : int = train(3 , A__ , A__ , A__ , A__ ) ((A__) , (A__)) : Optional[Any] = model.a.item(), model.b.item() A__ : Any = optimizer.state_dict() # Train partially set_seed(42 ) A__ : Dict = DummyModel() A__ : List[Any] = torch.optim.Adam(params=model.parameters() , lr=1e-3 ) A__ , A__ : Union[str, Any] = dummy_dataloaders() A__ : List[Any] = ProjectConfiguration(iteration=1 , automatic_checkpoint_naming=A__ ) A__ : Dict = Accelerator(project_dir=A__ , project_config=A__ ) A__ , A__ , A__ , A__ : Union[str, Any] = accelerator.prepare( A__ , A__ , A__ , A__ ) accelerator.load_state(os.path.join(A__ , """checkpoints""" , """checkpoint_0""" ) ) ((A__) , (A__)) : Optional[int] = model.a.item(), model.b.item() A__ : Tuple = optimizer.state_dict() self.assertEqual(A__ , A__ ) self.assertEqual(A__ , A__ ) self.assertEqual(A__ , A__ ) A__ : str = train(2 , A__ , A__ , A__ , A__ ) # Save everything accelerator.save_state() # Load everything back in and make sure all states work accelerator.load_state(os.path.join(A__ , """checkpoints""" , """checkpoint_1""" ) ) test_rands += train(1 , A__ , A__ , A__ , A__ ) ((A__) , (A__)) : Optional[int] = model.a.item(), model.b.item() A__ : List[Any] = optimizer.state_dict() self.assertEqual(A__ , A__ ) self.assertEqual(A__ , A__ ) self.assertEqual(A__ , A__ ) self.assertEqual(A__ , A__ ) def __A ( self ): A__ : Union[str, Any] = torch.tensor([1, 2, 3] ) A__ : int = torch.tensor([2, 3, 4] ) A__ : List[Any] = DummyModel() A__ : List[Any] = torch.optim.Adam(net.parameters() ) A__ : Tuple = Accelerator() with self.assertRaises(A__ ) as ve: accelerator.register_for_checkpointing(A__ , A__ , A__ , A__ ) A__ : Any = str(ve.exception ) self.assertTrue("""Item at index 0""" in message ) self.assertTrue("""Item at index 1""" in message ) self.assertFalse("""Item at index 2""" in message ) self.assertFalse("""Item at index 3""" in message ) def __A ( self ): with tempfile.TemporaryDirectory() as tmpdir: set_seed(42 ) A__ : Any = DummyModel() A__ : Union[str, Any] = torch.optim.Adam(params=model.parameters() , lr=1e-3 ) A__ : Dict = torch.optim.lr_scheduler.StepLR(A__ , step_size=1 , gamma=0.9_9 ) A__ , A__ : List[Any] = dummy_dataloaders() A__ : Tuple = ProjectConfiguration(automatic_checkpoint_naming=A__ ) # Train baseline A__ : Optional[Any] = Accelerator(project_dir=A__ , project_config=A__ ) A__ , A__ , A__ , A__ , A__ : Union[str, Any] = accelerator.prepare( A__ , A__ , A__ , A__ , A__ ) # Save initial accelerator.save_state() A__ : Tuple = scheduler.state_dict() train(3 , A__ , A__ , A__ , A__ , A__ ) self.assertNotEqual(A__ , scheduler.state_dict() ) # Load everything back in and make sure all states work accelerator.load_state(os.path.join(A__ , """checkpoints""" , """checkpoint_0""" ) ) self.assertEqual(A__ , scheduler.state_dict() ) def __A ( self ): with tempfile.TemporaryDirectory() as tmpdir: set_seed(42 ) A__ : Optional[Any] = DummyModel() A__ : int = ProjectConfiguration(automatic_checkpoint_naming=A__ , total_limit=2 ) # Train baseline A__ : List[str] = Accelerator(project_dir=A__ , project_config=A__ ) A__ : Union[str, Any] = accelerator.prepare(A__ ) # Save 3 states: for _ in range(11 ): accelerator.save_state() self.assertTrue(not os.path.exists(os.path.join(A__ , """checkpoints""" , """checkpoint_0""" ) ) ) self.assertTrue(os.path.exists(os.path.join(A__ , """checkpoints""" , """checkpoint_9""" ) ) ) self.assertTrue(os.path.exists(os.path.join(A__ , """checkpoints""" , """checkpoint_10""" ) ) ) @require_cuda def __A ( self ): A__ : Dict = ["""torchrun""", F"""--nproc_per_node={torch.cuda.device_count()}""", inspect.getfile(self.__class__ )] execute_subprocess_async(A__ , env=os.environ.copy() ) if __name__ == "__main__": A_ : List[str] = '/tmp/accelerate/state_checkpointing' A_ : Optional[Any] = DummyModel() A_ : Union[str, Any] = torch.optim.Adam(params=model.parameters(), lr=1E-3) A_ : str = torch.optim.lr_scheduler.StepLR(optimizer, step_size=1, gamma=0.99) A_ , A_ : List[Any] = dummy_dataloaders() A_ : int = ProjectConfiguration(automatic_checkpoint_naming=True) # Train baseline A_ : List[str] = Accelerator(project_dir=savedir, project_config=project_config, mixed_precision='no') if accelerator.process_index == 0: if os.path.exists(savedir): shutil.rmtree(savedir) os.makedirs(savedir) A_ , A_ , A_ , A_ , A_ : List[Any] = accelerator.prepare( model, optimizer, train_dataloader, valid_dataloader, scheduler ) A_ , A_ : Dict = accelerator.prepare(model, optimizer) train(3, model, train_dataloader, optimizer, accelerator, scheduler) # Check that the intial optimizer is loaded on the GPU for group in optimizer.param_groups: A_ : str = group['params'][0].device break assert param_device.type == accelerator.device.type A_ : Optional[Any] = model.cpu() accelerator.wait_for_everyone() accelerator.save_state() accelerator.wait_for_everyone() # Check CPU state accelerator.load_state(os.path.join(savedir, 'checkpoints', 'checkpoint_0'), map_location='cpu') for group in optimizer.param_groups: A_ : str = group['params'][0].device break assert ( param_device.type == torch.device('cpu').type ), f"Loaded optimizer states did not match, expected to be loaded on the CPU but got {param_device}" # Check device state model.to(accelerator.device) accelerator.load_state(os.path.join(savedir, 'checkpoints', 'checkpoint_0'), map_location='on_device') for group in optimizer.param_groups: A_ : Tuple = group['params'][0].device break assert ( param_device.type == accelerator.device.type ), f"Loaded optimizer states did not match, expected to be loaded on {accelerator.device} but got {param_device}" # Check error with pytest.raises(TypeError, match='Unsupported optimizer map location passed'): accelerator.load_state(os.path.join(savedir, 'checkpoints', 'checkpoint_0'), map_location='invalid') accelerator.wait_for_everyone() if accelerator.process_index == 0: shutil.rmtree(savedir) accelerator.wait_for_everyone()
64
1
def UpperCamelCase (lowercase_: int ) -> int: if not isinstance(lowercase_ , lowercase_ ): raise TypeError("""Input value must be an 'int' type""" ) A__ : int = 0 while number: position += 1 number >>= 1 return position if __name__ == "__main__": import doctest doctest.testmod()
64
def UpperCamelCase (lowercase_: str , lowercase_: str ) -> bool: A__ : Union[str, Any] = len(lowercase_ ) A__ : List[Any] = len(lowercase_ ) A__ : List[Any] = [[False for _ in range(m + 1 )] for _ in range(n + 1 )] A__ : str = True for i in range(lowercase_ ): for j in range(m + 1 ): if dp[i][j]: if j < m and a[i].upper() == b[j]: A__ : int = True if a[i].islower(): A__ : Dict = True return dp[n][m] if __name__ == "__main__": import doctest doctest.testmod()
64
1
import json from typing import TYPE_CHECKING, List, Optional, Tuple from tokenizers import pre_tokenizers from ...tokenization_utils_base import BatchEncoding from ...tokenization_utils_fast import PreTrainedTokenizerFast from ...utils import logging from .tokenization_gpta import GPTaTokenizer if TYPE_CHECKING: from transformers.pipelines.conversational import Conversation A_ : Tuple = logging.get_logger(__name__) A_ : Optional[int] = {'vocab_file': 'vocab.json', 'merges_file': 'merges.txt', 'tokenizer_file': 'tokenizer.json'} A_ : List[str] = { 'vocab_file': { 'gpt2': 'https://huggingface.co/gpt2/resolve/main/vocab.json', 'gpt2-medium': 'https://huggingface.co/gpt2-medium/resolve/main/vocab.json', 'gpt2-large': 'https://huggingface.co/gpt2-large/resolve/main/vocab.json', 'gpt2-xl': 'https://huggingface.co/gpt2-xl/resolve/main/vocab.json', 'distilgpt2': 'https://huggingface.co/distilgpt2/resolve/main/vocab.json', }, 'merges_file': { 'gpt2': 'https://huggingface.co/gpt2/resolve/main/merges.txt', 'gpt2-medium': 'https://huggingface.co/gpt2-medium/resolve/main/merges.txt', 'gpt2-large': 'https://huggingface.co/gpt2-large/resolve/main/merges.txt', 'gpt2-xl': 'https://huggingface.co/gpt2-xl/resolve/main/merges.txt', 'distilgpt2': 'https://huggingface.co/distilgpt2/resolve/main/merges.txt', }, 'tokenizer_file': { 'gpt2': 'https://huggingface.co/gpt2/resolve/main/tokenizer.json', 'gpt2-medium': 'https://huggingface.co/gpt2-medium/resolve/main/tokenizer.json', 'gpt2-large': 'https://huggingface.co/gpt2-large/resolve/main/tokenizer.json', 'gpt2-xl': 'https://huggingface.co/gpt2-xl/resolve/main/tokenizer.json', 'distilgpt2': 'https://huggingface.co/distilgpt2/resolve/main/tokenizer.json', }, } A_ : int = { 'gpt2': 1024, 'gpt2-medium': 1024, 'gpt2-large': 1024, 'gpt2-xl': 1024, 'distilgpt2': 1024, } class _a (__magic_name__ ): '''simple docstring''' UpperCAmelCase__: str = VOCAB_FILES_NAMES UpperCAmelCase__: List[Any] = PRETRAINED_VOCAB_FILES_MAP UpperCAmelCase__: Union[str, Any] = PRETRAINED_POSITIONAL_EMBEDDINGS_SIZES UpperCAmelCase__: List[Any] = ['''input_ids''', '''attention_mask'''] UpperCAmelCase__: Tuple = GPTaTokenizer def __init__( self , A__=None , A__=None , A__=None , A__="<|endoftext|>" , A__="<|endoftext|>" , A__="<|endoftext|>" , A__=False , **A__ , ): super().__init__( A__ , A__ , tokenizer_file=A__ , unk_token=A__ , bos_token=A__ , eos_token=A__ , add_prefix_space=A__ , **A__ , ) A__ : Union[str, Any] = kwargs.pop("""add_bos_token""" , A__ ) A__ : Union[str, Any] = json.loads(self.backend_tokenizer.pre_tokenizer.__getstate__() ) if pre_tok_state.get("""add_prefix_space""" , A__ ) != add_prefix_space: A__ : Union[str, Any] = getattr(A__ , pre_tok_state.pop("""type""" ) ) A__ : Tuple = add_prefix_space A__ : Dict = pre_tok_class(**A__ ) A__ : Dict = add_prefix_space def __A ( self , *A__ , **A__ ): A__ : Any = kwargs.get("""is_split_into_words""" , A__ ) assert self.add_prefix_space or not is_split_into_words, ( F"""You need to instantiate {self.__class__.__name__} with add_prefix_space=True """ "to use it with pretokenized inputs." ) return super()._batch_encode_plus(*A__ , **A__ ) def __A ( self , *A__ , **A__ ): A__ : Optional[Any] = kwargs.get("""is_split_into_words""" , A__ ) assert self.add_prefix_space or not is_split_into_words, ( F"""You need to instantiate {self.__class__.__name__} with add_prefix_space=True """ "to use it with pretokenized inputs." ) return super()._encode_plus(*A__ , **A__ ) def __A ( self , A__ , A__ = None ): A__ : Optional[int] = self._tokenizer.model.save(A__ , name=A__ ) return tuple(A__ ) def __A ( self , A__ ): A__ : Optional[Any] = [] for is_user, text in conversation.iter_texts(): input_ids.extend(self.encode(A__ , add_special_tokens=A__ ) + [self.eos_token_id] ) if len(A__ ) > self.model_max_length: A__ : str = input_ids[-self.model_max_length :] return input_ids
64
import itertools import random import unittest import numpy as np from transformers import ASTFeatureExtractor from transformers.testing_utils import require_torch, require_torchaudio from transformers.utils.import_utils import is_torch_available from ...test_sequence_feature_extraction_common import SequenceFeatureExtractionTestMixin A_ : Dict = random.Random() if is_torch_available(): import torch def UpperCamelCase (lowercase_: Tuple , lowercase_: Tuple=1.0 , lowercase_: Dict=None , lowercase_: int=None ) -> str: if rng is None: A__ : Optional[Any] = global_rng A__ : List[str] = [] for batch_idx in range(shape[0] ): values.append([] ) for _ in range(shape[1] ): values[-1].append(rng.random() * scale ) return values class _a (unittest.TestCase ): '''simple docstring''' def __init__( self , A__ , A__=7 , A__=400 , A__=2000 , A__=1 , A__=0.0 , A__=1_6000 , A__=True , A__=True , ): A__ : Any = parent A__ : Optional[int] = batch_size A__ : Union[str, Any] = min_seq_length A__ : Dict = max_seq_length A__ : Tuple = (self.max_seq_length - self.min_seq_length) // (self.batch_size - 1) A__ : str = feature_size A__ : Optional[int] = padding_value A__ : List[str] = sampling_rate A__ : List[str] = return_attention_mask A__ : int = do_normalize def __A ( self ): return { "feature_size": self.feature_size, "padding_value": self.padding_value, "sampling_rate": self.sampling_rate, "return_attention_mask": self.return_attention_mask, "do_normalize": self.do_normalize, } def __A ( self , A__=False , A__=False ): def _flatten(A__ ): return list(itertools.chain(*A__ ) ) if equal_length: A__ : Dict = floats_list((self.batch_size, self.max_seq_length) ) else: # make sure that inputs increase in size A__ : Union[str, Any] = [ _flatten(floats_list((x, self.feature_size) ) ) for x in range(self.min_seq_length , self.max_seq_length , self.seq_length_diff ) ] if numpify: A__ : Optional[int] = [np.asarray(A__ ) for x in speech_inputs] return speech_inputs @require_torch @require_torchaudio class _a (__magic_name__ , unittest.TestCase ): '''simple docstring''' UpperCAmelCase__: int = ASTFeatureExtractor def __A ( self ): A__ : Optional[Any] = ASTFeatureExtractionTester(self ) def __A ( self ): # Tests that all call wrap to encode_plus and batch_encode_plus A__ : Optional[int] = self.feature_extraction_class(**self.feat_extract_tester.prepare_feat_extract_dict() ) # create three inputs of length 800, 1000, and 1200 A__ : List[Any] = [floats_list((1, x) )[0] for x in range(800 , 1400 , 200 )] A__ : Optional[Any] = [np.asarray(A__ ) for speech_input in speech_inputs] # Test not batched input A__ : Tuple = feat_extract(speech_inputs[0] , return_tensors="""np""" ).input_values A__ : Tuple = feat_extract(np_speech_inputs[0] , return_tensors="""np""" ).input_values self.assertTrue(np.allclose(A__ , A__ , atol=1e-3 ) ) # Test batched A__ : Tuple = feat_extract(A__ , padding=A__ , return_tensors="""np""" ).input_values A__ : Tuple = feat_extract(A__ , padding=A__ , return_tensors="""np""" ).input_values for enc_seq_a, enc_seq_a in zip(A__ , A__ ): self.assertTrue(np.allclose(A__ , A__ , atol=1e-3 ) ) # Test 2-D numpy arrays are batched. A__ : int = [floats_list((1, x) )[0] for x in (800, 800, 800)] A__ : List[str] = np.asarray(A__ ) A__ : Union[str, Any] = feat_extract(A__ , return_tensors="""np""" ).input_values A__ : Optional[Any] = feat_extract(A__ , return_tensors="""np""" ).input_values for enc_seq_a, enc_seq_a in zip(A__ , A__ ): self.assertTrue(np.allclose(A__ , A__ , atol=1e-3 ) ) @require_torch def __A ( self ): import torch A__ : Union[str, Any] = self.feature_extraction_class(**self.feat_extract_tester.prepare_feat_extract_dict() ) A__ : Tuple = np.random.rand(100 ).astype(np.floataa ) A__ : Tuple = np_speech_inputs.tolist() for inputs in [py_speech_inputs, np_speech_inputs]: A__ : List[str] = feature_extractor.pad([{"""input_values""": inputs}] , return_tensors="""np""" ) self.assertTrue(np_processed.input_values.dtype == np.floataa ) A__ : Any = feature_extractor.pad([{"""input_values""": inputs}] , return_tensors="""pt""" ) self.assertTrue(pt_processed.input_values.dtype == torch.floataa ) def __A ( self , A__ ): from datasets import load_dataset A__ : str = load_dataset("""hf-internal-testing/librispeech_asr_dummy""" , """clean""" , split="""validation""" ) # automatic decoding with librispeech A__ : str = ds.sort("""id""" ).select(range(A__ ) )[:num_samples]["""audio"""] return [x["array"] for x in speech_samples] @require_torch def __A ( self ): # fmt: off A__ : Optional[Any] = torch.tensor( [-0.9_8_9_4, -1.2_7_7_6, -0.9_0_6_6, -1.2_7_7_6, -0.9_3_4_9, -1.2_6_0_9, -1.0_3_8_6, -1.2_7_7_6, -1.1_5_6_1, -1.2_7_7_6, -1.2_0_5_2, -1.2_7_2_3, -1.2_1_9_0, -1.2_1_3_2, -1.2_7_7_6, -1.1_1_3_3, -1.1_9_5_3, -1.1_3_4_3, -1.1_5_8_4, -1.2_2_0_3, -1.1_7_7_0, -1.2_4_7_4, -1.2_3_8_1, -1.1_9_3_6, -0.9_2_7_0, -0.8_3_1_7, -0.8_0_4_9, -0.7_7_0_6, -0.7_5_6_5, -0.7_8_6_9] ) # fmt: on A__ : Any = self._load_datasamples(1 ) A__ : Tuple = ASTFeatureExtractor() A__ : Dict = feature_extractor(A__ , return_tensors="""pt""" ).input_values self.assertEquals(input_values.shape , (1, 1024, 128) ) self.assertTrue(torch.allclose(input_values[0, 0, :30] , A__ , atol=1e-4 ) )
64
1
import json import os import tempfile from transformers.testing_utils import check_json_file_has_correct_format class _a : '''simple docstring''' UpperCAmelCase__: List[Any] = None def __A ( self ): A__ : List[str] = self.feature_extraction_class(**self.feat_extract_dict ) A__ : Optional[Any] = json.loads(feat_extract.to_json_string() ) for key, value in self.feat_extract_dict.items(): self.assertEqual(obj[key] , A__ ) def __A ( self ): A__ : int = self.feature_extraction_class(**self.feat_extract_dict ) with tempfile.TemporaryDirectory() as tmpdirname: A__ : str = os.path.join(A__ , """feat_extract.json""" ) feat_extract_first.to_json_file(A__ ) A__ : Optional[Any] = self.feature_extraction_class.from_json_file(A__ ) self.assertEqual(feat_extract_second.to_dict() , feat_extract_first.to_dict() ) def __A ( self ): A__ : Dict = self.feature_extraction_class(**self.feat_extract_dict ) with tempfile.TemporaryDirectory() as tmpdirname: A__ : Optional[Any] = feat_extract_first.save_pretrained(A__ )[0] check_json_file_has_correct_format(A__ ) A__ : Optional[Any] = self.feature_extraction_class.from_pretrained(A__ ) self.assertEqual(feat_extract_second.to_dict() , feat_extract_first.to_dict() ) def __A ( self ): A__ : int = self.feature_extraction_class() self.assertIsNotNone(A__ )
64
from typing import Optional import numpy as np import torch from torch import nn from transformers import GPTaConfig, GPTaLMHeadModel from transformers.modeling_utils import ModuleUtilsMixin from ...configuration_utils import ConfigMixin, register_to_config from ...models import ModelMixin class _a (__magic_name__ , __magic_name__ , __magic_name__ ): '''simple docstring''' UpperCAmelCase__: str = [r'''h\.\d+\.attn\.bias''', r'''h\.\d+\.attn\.masked_bias'''] @register_to_config def __init__( self , A__ , A__ , A__ = None , A__ = 5_0257 , A__ = 1024 , A__ = 768 , A__ = 12 , A__ = 12 , A__ = None , A__ = "gelu_new" , A__ = 0.1 , A__ = 0.1 , A__ = 0.1 , A__ = 1e-5 , A__ = 0.0_2 , A__ = True , A__ = True , A__ = False , A__ = False , ): super().__init__() A__ : Union[str, Any] = prefix_length if prefix_inner_dim != n_embd and prefix_hidden_dim is None: raise ValueError( F"""`prefix_hidden_dim` cannot be `None` when `prefix_inner_dim`: {prefix_hidden_dim} and""" F""" `n_embd`: {n_embd} are not equal.""" ) A__ : str = prefix_inner_dim A__ : Optional[Any] = prefix_hidden_dim A__ : Tuple = ( nn.Linear(self.prefix_inner_dim , self.prefix_hidden_dim ) if self.prefix_hidden_dim is not None else nn.Identity() ) A__ : int = ( nn.Linear(self.prefix_hidden_dim , A__ ) if self.prefix_hidden_dim is not None else nn.Identity() ) A__ : Tuple = GPTaConfig( vocab_size=A__ , n_positions=A__ , n_embd=A__ , n_layer=A__ , n_head=A__ , n_inner=A__ , activation_function=A__ , resid_pdrop=A__ , embd_pdrop=A__ , attn_pdrop=A__ , layer_norm_epsilon=A__ , initializer_range=A__ , scale_attn_weights=A__ , use_cache=A__ , scale_attn_by_inverse_layer_idx=A__ , reorder_and_upcast_attn=A__ , ) A__ : int = GPTaLMHeadModel(A__ ) def __A ( self , A__ , A__ , A__ = None , A__ = None , ): A__ : List[str] = self.transformer.transformer.wte(A__ ) A__ : int = self.encode_prefix(A__ ) A__ : int = self.decode_prefix(A__ ) A__ : Optional[Any] = torch.cat((prefix_embeds, embedding_text) , dim=1 ) if labels is not None: A__ : Any = self.get_dummy_token(input_ids.shape[0] , input_ids.device ) A__ : List[Any] = torch.cat((dummy_token, input_ids) , dim=1 ) A__ : List[str] = self.transformer(inputs_embeds=A__ , labels=A__ , attention_mask=A__ ) if self.prefix_hidden_dim is not None: return out, hidden else: return out def __A ( self , A__ , A__ ): return torch.zeros(A__ , self.prefix_length , dtype=torch.intaa , device=A__ ) def __A ( self , A__ ): return self.encode_prefix(A__ ) @torch.no_grad() def __A ( self , A__ , A__ , A__ ): A__ : List[Any] = torch.split(A__ , 1 , dim=0 ) A__ : Optional[int] = [] A__ : str = [] for feature in features: A__ : Dict = self.decode_prefix(feature.to(A__ ) ) # back to the clip feature # Only support beam search for now A__ , A__ : Union[str, Any] = self.generate_beam( input_embeds=A__ , device=A__ , eos_token_id=A__ ) generated_tokens.append(output_tokens[0] ) generated_seq_lengths.append(seq_lengths[0] ) A__ : int = torch.stack(A__ ) A__ : List[Any] = torch.stack(A__ ) return generated_tokens, generated_seq_lengths @torch.no_grad() def __A ( self , A__=None , A__=None , A__=None , A__ = 5 , A__ = 67 , A__ = 1.0 , A__ = None , ): A__ : Any = eos_token_id A__ : Any = None A__ : Optional[int] = None A__ : Optional[Any] = torch.ones(A__ , device=A__ , dtype=torch.int ) A__ : Any = torch.zeros(A__ , device=A__ , dtype=torch.bool ) if input_embeds is not None: A__ : Dict = input_embeds else: A__ : str = self.transformer.transformer.wte(A__ ) for i in range(A__ ): A__ : Dict = self.transformer(inputs_embeds=A__ ) A__ : str = outputs.logits A__ : Union[str, Any] = logits[:, -1, :] / (temperature if temperature > 0 else 1.0) A__ : Any = logits.softmax(-1 ).log() if scores is None: A__ , A__ : Optional[int] = logits.topk(A__ , -1 ) A__ : List[Any] = generated.expand(A__ , *generated.shape[1:] ) A__ , A__ : List[Any] = next_tokens.permute(1 , 0 ), scores.squeeze(0 ) if tokens is None: A__ : Optional[Any] = next_tokens else: A__ : List[Any] = tokens.expand(A__ , *tokens.shape[1:] ) A__ : int = torch.cat((tokens, next_tokens) , dim=1 ) else: A__ : Optional[int] = -float(np.inf ) A__ : List[Any] = 0 A__ : str = scores[:, None] + logits seq_lengths[~is_stopped] += 1 A__ : Dict = scores_sum / seq_lengths[:, None] A__ , A__ : List[Any] = scores_sum_average.view(-1 ).topk(A__ , -1 ) A__ : Tuple = next_tokens // scores_sum.shape[1] A__ : Optional[Any] = seq_lengths[next_tokens_source] A__ : List[str] = next_tokens % scores_sum.shape[1] A__ : Optional[int] = next_tokens.unsqueeze(1 ) A__ : int = tokens[next_tokens_source] A__ : List[Any] = torch.cat((tokens, next_tokens) , dim=1 ) A__ : str = generated[next_tokens_source] A__ : Optional[Any] = scores_sum_average * seq_lengths A__ : Union[str, Any] = is_stopped[next_tokens_source] A__ : str = self.transformer.transformer.wte(next_tokens.squeeze() ).view(generated.shape[0] , 1 , -1 ) A__ : Optional[int] = torch.cat((generated, next_token_embed) , dim=1 ) A__ : List[str] = is_stopped + next_tokens.eq(A__ ).squeeze() if is_stopped.all(): break A__ : Dict = scores / seq_lengths A__ : Dict = scores.argsort(descending=A__ ) # tokens tensors are already padded to max_seq_length A__ : Union[str, Any] = [tokens[i] for i in order] A__ : Any = torch.stack(A__ , dim=0 ) A__ : Dict = torch.tensor([seq_lengths[i] for i in order] , dtype=seq_lengths.dtype ) return output_texts, seq_lengths
64
1
import logging import os from typing import Dict, List, Optional, Union import torch import torch.nn as nn from accelerate.utils.imports import ( is_abit_bnb_available, is_abit_bnb_available, is_bnb_available, ) from ..big_modeling import dispatch_model, init_empty_weights from .dataclasses import BnbQuantizationConfig from .modeling import ( find_tied_parameters, get_balanced_memory, infer_auto_device_map, load_checkpoint_in_model, offload_weight, set_module_tensor_to_device, ) if is_bnb_available(): import bitsandbytes as bnb from copy import deepcopy A_ : Union[str, Any] = logging.getLogger(__name__) def UpperCamelCase (lowercase_: torch.nn.Module , lowercase_: BnbQuantizationConfig , lowercase_: Union[str, os.PathLike] = None , lowercase_: Optional[Dict[str, Union[int, str, torch.device]]] = None , lowercase_: Optional[List[str]] = None , lowercase_: Optional[Dict[Union[int, str], Union[int, str]]] = None , lowercase_: Optional[Union[str, os.PathLike]] = None , lowercase_: bool = False , ) -> int: A__ : Union[str, Any] = bnb_quantization_config.load_in_abit A__ : Optional[int] = bnb_quantization_config.load_in_abit if load_in_abit and not is_abit_bnb_available(): raise ImportError( """You have a version of `bitsandbytes` that is not compatible with 8bit quantization,""" """ make sure you have the latest version of `bitsandbytes` installed.""" ) if load_in_abit and not is_abit_bnb_available(): raise ValueError( """You have a version of `bitsandbytes` that is not compatible with 4bit quantization,""" """make sure you have the latest version of `bitsandbytes` installed.""" ) A__ : Any = [] # custom device map if isinstance(lowercase_ , lowercase_ ) and len(device_map.keys() ) > 1: A__ : Dict = [key for key, value in device_map.items() if value in ["""disk""", """cpu"""]] # We keep some modules such as the lm_head in their original dtype for numerical stability reasons if bnb_quantization_config.skip_modules is None: A__ : str = get_keys_to_not_convert(lowercase_ ) # add cpu modules to skip modules only for 4-bit modules if load_in_abit: bnb_quantization_config.skip_modules.extend(lowercase_ ) A__ : Any = bnb_quantization_config.skip_modules # We add the modules we want to keep in full precision if bnb_quantization_config.keep_in_fpaa_modules is None: A__ : Optional[int] = [] A__ : List[str] = bnb_quantization_config.keep_in_fpaa_modules modules_to_not_convert.extend(lowercase_ ) # compatibility with peft A__ : Optional[Any] = load_in_abit A__ : Any = load_in_abit A__ : Tuple = get_parameter_device(lowercase_ ) if model_device.type != "meta": # quantization of an already loaded model logger.warning( """It is not recommended to quantize a loaded model. """ """The model should be instantiated under the `init_empty_weights` context manager.""" ) A__ : List[str] = replace_with_bnb_layers(lowercase_ , lowercase_ , modules_to_not_convert=lowercase_ ) # convert param to the right dtype A__ : Dict = bnb_quantization_config.torch_dtype for name, param in model.state_dict().items(): if any(module_to_keep_in_fpaa in name for module_to_keep_in_fpaa in keep_in_fpaa_modules ): param.to(torch.floataa ) if param.dtype != torch.floataa: A__ : Optional[Any] = name.replace(""".weight""" , """""" ).replace(""".bias""" , """""" ) A__ : Union[str, Any] = getattr(lowercase_ , lowercase_ , lowercase_ ) if param is not None: param.to(torch.floataa ) elif torch.is_floating_point(lowercase_ ): param.to(lowercase_ ) if model_device.type == "cuda": # move everything to cpu in the first place because we can't do quantization if the weights are already on cuda model.cuda(torch.cuda.current_device() ) torch.cuda.empty_cache() elif torch.cuda.is_available(): model.to(torch.cuda.current_device() ) else: raise RuntimeError("""No GPU found. A GPU is needed for quantization.""" ) logger.info( f"""The model device type is {model_device.type}. However, cuda is needed for quantization.""" """We move the model to cuda.""" ) return model elif weights_location is None: raise RuntimeError( f"""`weights_location` needs to be the folder path containing the weights of the model, but we found {weights_location} """ ) else: with init_empty_weights(): A__ : str = replace_with_bnb_layers( lowercase_ , lowercase_ , modules_to_not_convert=lowercase_ ) A__ : int = get_quantized_model_device_map( lowercase_ , lowercase_ , lowercase_ , max_memory=lowercase_ , no_split_module_classes=lowercase_ , ) if offload_state_dict is None and device_map is not None and "disk" in device_map.values(): A__ : Tuple = True A__ : Tuple = any(x in list(device_map.values() ) for x in ["""cpu""", """disk"""] ) load_checkpoint_in_model( lowercase_ , lowercase_ , lowercase_ , dtype=bnb_quantization_config.torch_dtype , offload_folder=lowercase_ , offload_state_dict=lowercase_ , keep_in_fpaa_modules=bnb_quantization_config.keep_in_fpaa_modules , offload_abit_bnb=load_in_abit and offload , ) return dispatch_model(lowercase_ , device_map=lowercase_ , offload_dir=lowercase_ ) def UpperCamelCase (lowercase_: Optional[int] , lowercase_: Any , lowercase_: List[Any]=None , lowercase_: int=None , lowercase_: List[str]=None ) -> Union[str, Any]: if device_map is None: if torch.cuda.is_available(): A__ : Union[str, Any] = {"""""": torch.cuda.current_device()} else: raise RuntimeError("""No GPU found. A GPU is needed for quantization.""" ) logger.info("""The device_map was not initialized.""" """Setting device_map to `{'':torch.cuda.current_device()}`.""" ) if isinstance(lowercase_ , lowercase_ ): if device_map not in ["auto", "balanced", "balanced_low_0", "sequential"]: raise ValueError( """If passing a string for `device_map`, please choose 'auto', 'balanced', 'balanced_low_0' or """ """'sequential'.""" ) A__ : List[Any] = {} special_dtypes.update( { name: bnb_quantization_config.torch_dtype for name, _ in model.named_parameters() if any(m in name for m in bnb_quantization_config.skip_modules ) } ) special_dtypes.update( { name: torch.floataa for name, _ in model.named_parameters() if any(m in name for m in bnb_quantization_config.keep_in_fpaa_modules ) } ) A__ : str = {} A__ : Tuple = special_dtypes A__ : List[Any] = no_split_module_classes A__ : List[Any] = bnb_quantization_config.target_dtype # get max_memory for each device. if device_map != "sequential": A__ : str = get_balanced_memory( lowercase_ , low_zero=(device_map == """balanced_low_0""") , max_memory=lowercase_ , **lowercase_ , ) A__ : List[Any] = max_memory A__ : Union[str, Any] = infer_auto_device_map(lowercase_ , **lowercase_ ) if isinstance(lowercase_ , lowercase_ ): # check if don't have any quantized module on the cpu A__ : List[str] = bnb_quantization_config.skip_modules + bnb_quantization_config.keep_in_fpaa_modules A__ : Union[str, Any] = { key: device_map[key] for key in device_map.keys() if key not in modules_not_to_convert } for device in ["cpu", "disk"]: if device in device_map_without_some_modules.values(): if bnb_quantization_config.load_in_abit: raise ValueError( """ Some modules are dispatched on the CPU or the disk. Make sure you have enough GPU RAM to fit the quantized model. If you want to dispatch the model on the CPU or the disk while keeping these modules in `torch_dtype`, you need to pass a custom `device_map` to `load_and_quantize_model`. Check https://huggingface.co/docs/accelerate/main/en/usage_guides/quantization#offload-modules-to-cpu-and-disk for more details. """ ) else: logger.info( """Some modules are are offloaded to the CPU or the disk. Note that these modules will be converted to 8-bit""" ) del device_map_without_some_modules return device_map def UpperCamelCase (lowercase_: Tuple , lowercase_: Optional[int] , lowercase_: List[str]=None , lowercase_: Any=None ) -> Optional[int]: if modules_to_not_convert is None: A__ : str = [] A__ , A__ : Tuple = _replace_with_bnb_layers( lowercase_ , lowercase_ , lowercase_ , lowercase_ ) if not has_been_replaced: logger.warning( """You are loading your model in 8bit or 4bit but no linear modules were found in your model.""" """ this can happen for some architectures such as gpt2 that uses Conv1D instead of Linear layers.""" """ Please double check your model architecture, or submit an issue on github if you think this is""" """ a bug.""" ) return model def UpperCamelCase (lowercase_: Tuple , lowercase_: Optional[Any] , lowercase_: List[Any]=None , lowercase_: Tuple=None , ) -> Dict: A__ : List[str] = False for name, module in model.named_children(): if current_key_name is None: A__ : Optional[int] = [] current_key_name.append(lowercase_ ) if isinstance(lowercase_ , nn.Linear ) and name not in modules_to_not_convert: # Check if the current key is not in the `modules_to_not_convert` A__ : List[Any] = """.""".join(lowercase_ ) A__ : Optional[int] = True for key in modules_to_not_convert: if ( (key in current_key_name_str) and (key + "." in current_key_name_str) ) or key == current_key_name_str: A__ : int = False break if proceed: # Load bnb module with empty weight and replace ``nn.Linear` module if bnb_quantization_config.load_in_abit: A__ : List[str] = bnb.nn.LinearabitLt( module.in_features , module.out_features , module.bias is not None , has_fpaa_weights=lowercase_ , threshold=bnb_quantization_config.llm_inta_threshold , ) elif bnb_quantization_config.load_in_abit: A__ : Optional[Any] = bnb.nn.Linearabit( module.in_features , module.out_features , module.bias is not None , bnb_quantization_config.bnb_abit_compute_dtype , compress_statistics=bnb_quantization_config.bnb_abit_use_double_quant , quant_type=bnb_quantization_config.bnb_abit_quant_type , ) else: raise ValueError("""load_in_8bit and load_in_4bit can't be both False""" ) A__ : Dict = module.weight.data if module.bias is not None: A__ : Union[str, Any] = module.bias.data bnb_module.requires_grad_(lowercase_ ) setattr(lowercase_ , lowercase_ , lowercase_ ) A__ : Any = True if len(list(module.children() ) ) > 0: A__ , A__ : List[str] = _replace_with_bnb_layers( lowercase_ , lowercase_ , lowercase_ , lowercase_ ) A__ : int = has_been_replaced | _has_been_replaced # Remove the last key for recursion current_key_name.pop(-1 ) return model, has_been_replaced def UpperCamelCase (lowercase_: Dict ) -> Tuple: # Create a copy of the model with init_empty_weights(): A__ : List[Any] = deepcopy(lowercase_ ) # this has 0 cost since it is done inside `init_empty_weights` context manager` A__ : Any = find_tied_parameters(lowercase_ ) # For compatibility with Accelerate < 0.18 if isinstance(lowercase_ , lowercase_ ): A__ : Union[str, Any] = sum(list(tied_params.values() ) , [] ) + list(tied_params.keys() ) else: A__ : Optional[Any] = sum(lowercase_ , [] ) A__ : str = len(lowercase_ ) > 0 # Check if it is a base model A__ : List[str] = False if hasattr(lowercase_ , """base_model_prefix""" ): A__ : List[Any] = not hasattr(lowercase_ , model.base_model_prefix ) # Ignore this for base models (BertModel, GPT2Model, etc.) if (not has_tied_params) and is_base_model: return [] # otherwise they have an attached head A__ : Optional[Any] = list(model.named_children() ) A__ : Any = [list_modules[-1][0]] # add last module together with tied weights A__ : List[Any] = set(lowercase_ ) - set(lowercase_ ) A__ : Optional[Any] = list(set(lowercase_ ) ) + list(lowercase_ ) # remove ".weight" from the keys A__ : List[Any] = [""".weight""", """.bias"""] A__ : int = [] for name in list_untouched: for name_to_remove in names_to_remove: if name_to_remove in name: A__ : List[Any] = name.replace(lowercase_ , """""" ) filtered_module_names.append(lowercase_ ) return filtered_module_names def UpperCamelCase (lowercase_: str ) -> List[Any]: for m in model.modules(): if isinstance(lowercase_ , bnb.nn.Linearabit ): return True return False def UpperCamelCase (lowercase_: nn.Module ) -> Optional[int]: return next(parameter.parameters() ).device def UpperCamelCase (lowercase_: Dict , lowercase_: str , lowercase_: str , lowercase_: List[Any] , lowercase_: Tuple , lowercase_: Tuple , lowercase_: List[Any] ) -> int: # if it is not quantized, we quantize and offload the quantized weights and the SCB stats if fpaa_statistics is None: set_module_tensor_to_device(lowercase_ , lowercase_ , 0 , dtype=lowercase_ , value=lowercase_ ) A__ : Optional[Any] = param_name A__ : Optional[int] = model if "." in tensor_name: A__ : Union[str, Any] = tensor_name.split(""".""" ) for split in splits[:-1]: A__ : List[Any] = getattr(lowercase_ , lowercase_ ) if new_module is None: raise ValueError(f"""{module} has no attribute {split}.""" ) A__ : Any = new_module A__ : List[Any] = splits[-1] # offload weights A__ : List[Any] = False offload_weight(module._parameters[tensor_name] , lowercase_ , lowercase_ , index=lowercase_ ) if hasattr(module._parameters[tensor_name] , """SCB""" ): offload_weight( module._parameters[tensor_name].SCB , param_name.replace("""weight""" , """SCB""" ) , lowercase_ , index=lowercase_ , ) else: offload_weight(lowercase_ , lowercase_ , lowercase_ , index=lowercase_ ) offload_weight(lowercase_ , param_name.replace("""weight""" , """SCB""" ) , lowercase_ , index=lowercase_ ) set_module_tensor_to_device(lowercase_ , lowercase_ , """meta""" , dtype=lowercase_ , value=torch.empty(*param.size() ) )
64
import io import itertools import json from dataclasses import dataclass from typing import Optional import pyarrow as pa import pyarrow.json as paj import datasets from datasets.table import table_cast from datasets.utils.file_utils import readline A_ : Tuple = datasets.utils.logging.get_logger(__name__) @dataclass class _a (datasets.BuilderConfig ): '''simple docstring''' UpperCAmelCase__: Optional[datasets.Features] = None UpperCAmelCase__: str = "utf-8" UpperCAmelCase__: Optional[str] = None UpperCAmelCase__: Optional[str] = None UpperCAmelCase__: bool = True # deprecated UpperCAmelCase__: Optional[int] = None # deprecated UpperCAmelCase__: int = 10 << 20 # 10MB UpperCAmelCase__: Optional[bool] = None class _a (datasets.ArrowBasedBuilder ): '''simple docstring''' UpperCAmelCase__: List[str] = JsonConfig def __A ( self ): if self.config.block_size is not None: logger.warning("""The JSON loader parameter `block_size` is deprecated. Please use `chunksize` instead""" ) A__ : Union[str, Any] = self.config.block_size if self.config.use_threads is not True: logger.warning( """The JSON loader parameter `use_threads` is deprecated and doesn't have any effect anymore.""" ) if self.config.newlines_in_values is not None: raise ValueError("""The JSON loader parameter `newlines_in_values` is no longer supported""" ) return datasets.DatasetInfo(features=self.config.features ) def __A ( self , A__ ): if not self.config.data_files: raise ValueError(F"""At least one data file must be specified, but got data_files={self.config.data_files}""" ) A__ : int = dl_manager.download_and_extract(self.config.data_files ) if isinstance(A__ , (str, list, tuple) ): A__ : Optional[Any] = data_files if isinstance(A__ , A__ ): A__ : List[str] = [files] A__ : int = [dl_manager.iter_files(A__ ) for file in files] return [datasets.SplitGenerator(name=datasets.Split.TRAIN , gen_kwargs={"""files""": files} )] A__ : List[str] = [] for split_name, files in data_files.items(): if isinstance(A__ , A__ ): A__ : Optional[int] = [files] A__ : Optional[int] = [dl_manager.iter_files(A__ ) for file in files] splits.append(datasets.SplitGenerator(name=A__ , gen_kwargs={"""files""": files} ) ) return splits def __A ( self , A__ ): if self.config.features is not None: # adding missing columns for column_name in set(self.config.features ) - set(pa_table.column_names ): A__ : Optional[Any] = self.config.features.arrow_schema.field(A__ ).type A__ : str = pa_table.append_column(A__ , pa.array([None] * len(A__ ) , type=A__ ) ) # more expensive cast to support nested structures with keys in a different order # allows str <-> int/float or str to Audio for example A__ : Optional[int] = table_cast(A__ , self.config.features.arrow_schema ) return pa_table def __A ( self , A__ ): for file_idx, file in enumerate(itertools.chain.from_iterable(A__ ) ): # If the file is one json object and if we need to look at the list of items in one specific field if self.config.field is not None: with open(A__ , encoding=self.config.encoding , errors=self.config.encoding_errors ) as f: A__ : Optional[Any] = json.load(A__ ) # We keep only the field we are interested in A__ : Optional[int] = dataset[self.config.field] # We accept two format: a list of dicts or a dict of lists if isinstance(A__ , (list, tuple) ): A__ : Union[str, Any] = set().union(*[row.keys() for row in dataset] ) A__ : Any = {col: [row.get(A__ ) for row in dataset] for col in keys} else: A__ : Any = dataset A__ : Any = pa.Table.from_pydict(A__ ) yield file_idx, self._cast_table(A__ ) # If the file has one json object per line else: with open(A__ , """rb""" ) as f: A__ : List[str] = 0 # Use block_size equal to the chunk size divided by 32 to leverage multithreading # Set a default minimum value of 16kB if the chunk size is really small A__ : List[str] = max(self.config.chunksize // 32 , 16 << 10 ) A__ : Any = ( self.config.encoding_errors if self.config.encoding_errors is not None else """strict""" ) while True: A__ : Dict = f.read(self.config.chunksize ) if not batch: break # Finish current line try: batch += f.readline() except (AttributeError, io.UnsupportedOperation): batch += readline(A__ ) # PyArrow only accepts utf-8 encoded bytes if self.config.encoding != "utf-8": A__ : List[Any] = batch.decode(self.config.encoding , errors=A__ ).encode("""utf-8""" ) try: while True: try: A__ : str = paj.read_json( io.BytesIO(A__ ) , read_options=paj.ReadOptions(block_size=A__ ) ) break except (pa.ArrowInvalid, pa.ArrowNotImplementedError) as e: if ( isinstance(A__ , pa.ArrowInvalid ) and "straddling" not in str(A__ ) or block_size > len(A__ ) ): raise else: # Increase the block size in case it was too small. # The block size will be reset for the next file. logger.debug( F"""Batch of {len(A__ )} bytes couldn't be parsed with block_size={block_size}. Retrying with block_size={block_size * 2}.""" ) block_size *= 2 except pa.ArrowInvalid as e: try: with open( A__ , encoding=self.config.encoding , errors=self.config.encoding_errors ) as f: A__ : Optional[Any] = json.load(A__ ) except json.JSONDecodeError: logger.error(F"""Failed to read file '{file}' with error {type(A__ )}: {e}""" ) raise e # If possible, parse the file as a list of json objects and exit the loop if isinstance(A__ , A__ ): # list is the only sequence type supported in JSON try: A__ : str = set().union(*[row.keys() for row in dataset] ) A__ : List[str] = {col: [row.get(A__ ) for row in dataset] for col in keys} A__ : int = pa.Table.from_pydict(A__ ) except (pa.ArrowInvalid, AttributeError) as e: logger.error(F"""Failed to read file '{file}' with error {type(A__ )}: {e}""" ) raise ValueError(F"""Not able to read records in the JSON file at {file}.""" ) from None yield file_idx, self._cast_table(A__ ) break else: logger.error(F"""Failed to read file '{file}' with error {type(A__ )}: {e}""" ) raise ValueError( F"""Not able to read records in the JSON file at {file}. """ F"""You should probably indicate the field of the JSON file containing your records. """ F"""This JSON file contain the following fields: {str(list(dataset.keys() ) )}. """ F"""Select the correct one and provide it as `field='XXX'` to the dataset loading method. """ ) from None # Uncomment for debugging (will print the Arrow table size and elements) # logger.warning(f"pa_table: {pa_table} num rows: {pa_table.num_rows}") # logger.warning('\n'.join(str(pa_table.slice(i, 1).to_pydict()) for i in range(pa_table.num_rows))) yield (file_idx, batch_idx), self._cast_table(A__ ) batch_idx += 1
64
1
from functools import lru_cache @lru_cache def UpperCamelCase (lowercase_: int ) -> int: if num < 0: raise ValueError("""Number should not be negative.""" ) return 1 if num in (0, 1) else num * factorial(num - 1 ) if __name__ == "__main__": import doctest doctest.testmod()
64
import os try: from .build_directory_md import good_file_paths except ImportError: from build_directory_md import good_file_paths # type: ignore A_ : Dict = list(good_file_paths()) assert filepaths, "good_file_paths() failed!" A_ : Optional[Any] = [file for file in filepaths if file != file.lower()] if upper_files: print(f'''{len(upper_files)} files contain uppercase characters:''') print('\n'.join(upper_files) + '\n') A_ : Tuple = [file for file in filepaths if ' ' in file] if space_files: print(f'''{len(space_files)} files contain space characters:''') print('\n'.join(space_files) + '\n') A_ : Any = [file for file in filepaths if '-' in file] if hyphen_files: print(f'''{len(hyphen_files)} files contain hyphen characters:''') print('\n'.join(hyphen_files) + '\n') A_ : List[str] = [file for file in filepaths if os.sep not in file] if nodir_files: print(f'''{len(nodir_files)} files are not in a directory:''') print('\n'.join(nodir_files) + '\n') A_ : Any = len(upper_files + space_files + hyphen_files + nodir_files) if bad_files: import sys sys.exit(bad_files)
64
1
import os from typing import Any, Callable, Dict, List, Optional, Tuple, Union import torch from torch import nn from ...models.controlnet import ControlNetModel, ControlNetOutput from ...models.modeling_utils import ModelMixin from ...utils import logging A_ : Tuple = logging.get_logger(__name__) class _a (__magic_name__ ): '''simple docstring''' def __init__( self , A__ ): super().__init__() A__ : Any = nn.ModuleList(A__ ) def __A ( self , A__ , A__ , A__ , A__ , A__ , A__ = None , A__ = None , A__ = None , A__ = None , A__ = False , A__ = True , ): for i, (image, scale, controlnet) in enumerate(zip(A__ , A__ , self.nets ) ): A__ , A__ : List[str] = controlnet( A__ , A__ , A__ , A__ , A__ , A__ , A__ , A__ , A__ , A__ , A__ , ) # merge samples if i == 0: A__ , A__ : List[Any] = down_samples, mid_sample else: A__ : List[str] = [ samples_prev + samples_curr for samples_prev, samples_curr in zip(A__ , A__ ) ] mid_block_res_sample += mid_sample return down_block_res_samples, mid_block_res_sample def __A ( self , A__ , A__ = True , A__ = None , A__ = False , A__ = None , ): A__ : Optional[int] = 0 A__ : Union[str, Any] = save_directory for controlnet in self.nets: controlnet.save_pretrained( A__ , is_main_process=A__ , save_function=A__ , safe_serialization=A__ , variant=A__ , ) idx += 1 A__ : Union[str, Any] = model_path_to_save + F"""_{idx}""" @classmethod def __A ( cls , A__ , **A__ ): A__ : Union[str, Any] = 0 A__ : Dict = [] # load controlnet and append to list until no controlnet directory exists anymore # first controlnet has to be saved under `./mydirectory/controlnet` to be compliant with `DiffusionPipeline.from_prertained` # second, third, ... controlnets have to be saved under `./mydirectory/controlnet_1`, `./mydirectory/controlnet_2`, ... A__ : Union[str, Any] = pretrained_model_path while os.path.isdir(A__ ): A__ : Any = ControlNetModel.from_pretrained(A__ , **A__ ) controlnets.append(A__ ) idx += 1 A__ : List[str] = pretrained_model_path + F"""_{idx}""" logger.info(F"""{len(A__ )} controlnets loaded from {pretrained_model_path}.""" ) if len(A__ ) == 0: raise ValueError( F"""No ControlNets found under {os.path.dirname(A__ )}. Expected at least {pretrained_model_path + '_0'}.""" ) return cls(A__ )
64
import inspect import warnings from typing import Any, Dict, Optional, Union from packaging import version def UpperCamelCase (*lowercase_: Optional[int] , lowercase_: Optional[Union[Dict, Any]] = None , lowercase_: Dict=True , lowercase_: Tuple=2 ) -> Dict: from .. import __version__ A__ : Dict = take_from A__ : str = () if not isinstance(args[0] , lowercase_ ): A__ : int = (args,) for attribute, version_name, message in args: if version.parse(version.parse(lowercase_ ).base_version ) >= version.parse(lowercase_ ): raise ValueError( f"""The deprecation tuple {(attribute, version_name, message)} should be removed since diffusers'""" f""" version {__version__} is >= {version_name}""" ) A__ : Any = None if isinstance(lowercase_ , lowercase_ ) and attribute in deprecated_kwargs: values += (deprecated_kwargs.pop(lowercase_ ),) A__ : List[str] = f"""The `{attribute}` argument is deprecated and will be removed in version {version_name}.""" elif hasattr(lowercase_ , lowercase_ ): values += (getattr(lowercase_ , lowercase_ ),) A__ : Optional[Any] = f"""The `{attribute}` attribute is deprecated and will be removed in version {version_name}.""" elif deprecated_kwargs is None: A__ : int = f"""`{attribute}` is deprecated and will be removed in version {version_name}.""" if warning is not None: A__ : int = warning + """ """ if standard_warn else """""" warnings.warn(warning + message , lowercase_ , stacklevel=lowercase_ ) if isinstance(lowercase_ , lowercase_ ) and len(lowercase_ ) > 0: A__ : Union[str, Any] = inspect.getouterframes(inspect.currentframe() )[1] A__ : Optional[Any] = call_frame.filename A__ : Optional[int] = call_frame.lineno A__ : Any = call_frame.function A__ , A__ : List[str] = next(iter(deprecated_kwargs.items() ) ) raise TypeError(f"""{function} in {filename} line {line_number-1} got an unexpected keyword argument `{key}`""" ) if len(lowercase_ ) == 0: return elif len(lowercase_ ) == 1: return values[0] return values
64
1
import unittest from transformers import BertGenerationConfig, is_torch_available from transformers.testing_utils import require_torch, slow, torch_device from ...generation.test_utils import GenerationTesterMixin from ...test_configuration_common import ConfigTester from ...test_modeling_common import ModelTesterMixin, floats_tensor, ids_tensor, random_attention_mask from ...test_pipeline_mixin import PipelineTesterMixin if is_torch_available(): import torch from transformers import BertGenerationDecoder, BertGenerationEncoder class _a : '''simple docstring''' def __init__( self , A__ , A__=13 , A__=7 , A__=True , A__=True , A__=99 , A__=32 , A__=5 , A__=4 , A__=37 , A__="gelu" , A__=0.1 , A__=0.1 , A__=50 , A__=0.0_2 , A__=True , A__=None , ): A__ : Optional[Any] = parent A__ : Optional[Any] = batch_size A__ : Any = seq_length A__ : Any = is_training A__ : List[Any] = use_input_mask A__ : int = vocab_size A__ : int = hidden_size A__ : Union[str, Any] = num_hidden_layers A__ : Dict = num_attention_heads A__ : Optional[int] = intermediate_size A__ : Any = hidden_act A__ : Any = hidden_dropout_prob A__ : Optional[Any] = attention_probs_dropout_prob A__ : int = max_position_embeddings A__ : Optional[int] = initializer_range A__ : Optional[int] = use_labels A__ : List[Any] = scope def __A ( self ): A__ : Any = ids_tensor([self.batch_size, self.seq_length] , self.vocab_size ) A__ : Union[str, Any] = None if self.use_input_mask: A__ : Optional[Any] = random_attention_mask([self.batch_size, self.seq_length] ) if self.use_labels: A__ : List[Any] = ids_tensor([self.batch_size, self.seq_length] , self.vocab_size ) A__ : Any = self.get_config() return config, input_ids, input_mask, token_labels def __A ( self ): return BertGenerationConfig( vocab_size=self.vocab_size , hidden_size=self.hidden_size , num_hidden_layers=self.num_hidden_layers , num_attention_heads=self.num_attention_heads , intermediate_size=self.intermediate_size , hidden_act=self.hidden_act , hidden_dropout_prob=self.hidden_dropout_prob , attention_probs_dropout_prob=self.attention_probs_dropout_prob , max_position_embeddings=self.max_position_embeddings , is_decoder=A__ , initializer_range=self.initializer_range , ) def __A ( self ): ( ( A__ ) , ( A__ ) , ( A__ ) , ( A__ ) , ) : List[str] = self.prepare_config_and_inputs() A__ : List[str] = True A__ : str = floats_tensor([self.batch_size, self.seq_length, self.hidden_size] ) A__ : Optional[int] = ids_tensor([self.batch_size, self.seq_length] , vocab_size=2 ) return ( config, input_ids, input_mask, token_labels, encoder_hidden_states, encoder_attention_mask, ) def __A ( self , A__ , A__ , A__ , A__ , **A__ , ): A__ : List[str] = BertGenerationEncoder(config=A__ ) model.to(A__ ) model.eval() A__ : Optional[int] = model(A__ , attention_mask=A__ ) A__ : List[str] = model(A__ ) self.parent.assertEqual(result.last_hidden_state.shape , (self.batch_size, self.seq_length, self.hidden_size) ) def __A ( self , A__ , A__ , A__ , A__ , A__ , A__ , **A__ , ): A__ : List[str] = True A__ : List[Any] = BertGenerationEncoder(config=A__ ) model.to(A__ ) model.eval() A__ : List[str] = model( A__ , attention_mask=A__ , encoder_hidden_states=A__ , encoder_attention_mask=A__ , ) A__ : Any = model( A__ , attention_mask=A__ , encoder_hidden_states=A__ , ) self.parent.assertEqual(result.last_hidden_state.shape , (self.batch_size, self.seq_length, self.hidden_size) ) def __A ( self , A__ , A__ , A__ , A__ , A__ , A__ , **A__ , ): A__ : str = True A__ : Dict = True A__ : str = BertGenerationDecoder(config=A__ ).to(A__ ).eval() # first forward pass A__ : Any = model( A__ , attention_mask=A__ , encoder_hidden_states=A__ , encoder_attention_mask=A__ , use_cache=A__ , ) A__ : List[Any] = outputs.past_key_values # create hypothetical multiple next token and extent to next_input_ids A__ : int = ids_tensor((self.batch_size, 3) , config.vocab_size ) A__ : List[Any] = ids_tensor((self.batch_size, 3) , vocab_size=2 ) # append to next input_ids and A__ : Union[str, Any] = torch.cat([input_ids, next_tokens] , dim=-1 ) A__ : Dict = torch.cat([input_mask, next_mask] , dim=-1 ) A__ : Optional[int] = model( A__ , attention_mask=A__ , encoder_hidden_states=A__ , encoder_attention_mask=A__ , output_hidden_states=A__ , )["""hidden_states"""][0] A__ : Dict = model( A__ , attention_mask=A__ , encoder_hidden_states=A__ , encoder_attention_mask=A__ , past_key_values=A__ , output_hidden_states=A__ , )["""hidden_states"""][0] # select random slice A__ : str = ids_tensor((1,) , output_from_past.shape[-1] ).item() A__ : int = output_from_no_past[:, -3:, random_slice_idx].detach() A__ : Dict = output_from_past[:, :, random_slice_idx].detach() self.parent.assertTrue(output_from_past_slice.shape[1] == next_tokens.shape[1] ) # test that outputs are equal for slice self.parent.assertTrue(torch.allclose(A__ , A__ , atol=1e-3 ) ) def __A ( self , A__ , A__ , A__ , A__ , *A__ , ): A__ : Dict = BertGenerationDecoder(A__ ) model.to(A__ ) model.eval() A__ : Tuple = model(A__ , attention_mask=A__ , labels=A__ ) self.parent.assertEqual(result.logits.shape , (self.batch_size, self.seq_length, self.vocab_size) ) def __A ( self ): A__ , A__ , A__ , A__ : str = self.prepare_config_and_inputs() A__ : Dict = {"""input_ids""": input_ids, """attention_mask""": input_mask} return config, inputs_dict @require_torch class _a (__magic_name__ , __magic_name__ , __magic_name__ , unittest.TestCase ): '''simple docstring''' UpperCAmelCase__: Union[str, Any] = (BertGenerationEncoder, BertGenerationDecoder) if is_torch_available() else () UpperCAmelCase__: Union[str, Any] = (BertGenerationDecoder,) if is_torch_available() else () UpperCAmelCase__: Optional[Any] = ( {'''feature-extraction''': BertGenerationEncoder, '''text-generation''': BertGenerationDecoder} if is_torch_available() else {} ) def __A ( self ): A__ : Union[str, Any] = BertGenerationEncoderTester(self ) A__ : str = ConfigTester(self , config_class=A__ , hidden_size=37 ) def __A ( self ): self.config_tester.run_common_tests() def __A ( self ): A__ : Optional[Any] = self.model_tester.prepare_config_and_inputs() self.model_tester.create_and_check_model(*A__ ) def __A ( self ): A__ , A__ , A__ , A__ : Dict = self.model_tester.prepare_config_and_inputs() A__ : int = """bert""" self.model_tester.create_and_check_model(A__ , A__ , A__ , A__ ) def __A ( self ): A__ : Union[str, Any] = self.model_tester.prepare_config_and_inputs_for_decoder() self.model_tester.create_and_check_model_as_decoder(*A__ ) def __A ( self ): A__ : Dict = self.model_tester.prepare_config_and_inputs_for_decoder() self.model_tester.create_and_check_decoder_model_past_large_inputs(*A__ ) def __A ( self ): # This regression test was failing with PyTorch < 1.3 ( ( A__ ) , ( A__ ) , ( A__ ) , ( A__ ) , ( A__ ) , ( A__ ) , ) : Union[str, Any] = self.model_tester.prepare_config_and_inputs_for_decoder() A__ : List[str] = None self.model_tester.create_and_check_model_as_decoder( A__ , A__ , A__ , A__ , A__ , A__ , ) def __A ( self ): A__ : List[Any] = self.model_tester.prepare_config_and_inputs_for_decoder() self.model_tester.create_and_check_for_causal_lm(*A__ ) @slow def __A ( self ): A__ : int = BertGenerationEncoder.from_pretrained("""google/bert_for_seq_generation_L-24_bbc_encoder""" ) self.assertIsNotNone(A__ ) @require_torch class _a (unittest.TestCase ): '''simple docstring''' @slow def __A ( self ): A__ : Tuple = BertGenerationEncoder.from_pretrained("""google/bert_for_seq_generation_L-24_bbc_encoder""" ) A__ : Optional[Any] = torch.tensor([[101, 7592, 1010, 2026, 3899, 2003, 1_0140, 102]] ) with torch.no_grad(): A__ : Union[str, Any] = model(A__ )[0] A__ : str = torch.Size([1, 8, 1024] ) self.assertEqual(output.shape , A__ ) A__ : Tuple = torch.tensor( [[[0.1_7_7_5, 0.0_0_8_3, -0.0_3_2_1], [1.6_0_0_2, 0.1_2_8_7, 0.3_9_1_2], [2.1_4_7_3, 0.5_7_9_1, 0.6_0_6_6]]] ) self.assertTrue(torch.allclose(output[:, :3, :3] , A__ , atol=1e-4 ) ) @require_torch class _a (unittest.TestCase ): '''simple docstring''' @slow def __A ( self ): A__ : Optional[int] = BertGenerationDecoder.from_pretrained("""google/bert_for_seq_generation_L-24_bbc_encoder""" ) A__ : Union[str, Any] = torch.tensor([[101, 7592, 1010, 2026, 3899, 2003, 1_0140, 102]] ) with torch.no_grad(): A__ : str = model(A__ )[0] A__ : str = torch.Size([1, 8, 5_0358] ) self.assertEqual(output.shape , A__ ) A__ : Optional[Any] = torch.tensor( [[[-0.5_7_8_8, -2.5_9_9_4, -3.7_0_5_4], [0.0_4_3_8, 4.7_9_9_7, 1.8_7_9_5], [1.5_8_6_2, 6.6_4_0_9, 4.4_6_3_8]]] ) self.assertTrue(torch.allclose(output[:, :3, :3] , A__ , atol=1e-4 ) )
64
import argparse import json import os import tensorstore as ts import torch from flax import serialization from flax.traverse_util import flatten_dict, unflatten_dict from tensorflow.io import gfile from transformers.modeling_utils import dtype_byte_size from transformers.models.switch_transformers.convert_switch_transformers_original_flax_checkpoint_to_pytorch import ( rename_keys, ) from transformers.utils import WEIGHTS_INDEX_NAME, WEIGHTS_NAME from transformers.utils.hub import convert_file_size_to_int def UpperCamelCase (lowercase_: List[str] , lowercase_: str ) -> Optional[Any]: if flax_key_tuple[-1] == "kernel" and flax_tensor.ndim == 3: # expert layer A__ : Union[str, Any] = flax_key_tuple[:-1] + ("""weight""",) A__ : Optional[int] = torch.permute(lowercase_ , (0, 2, 1) ) elif flax_key_tuple[-1] == "kernel" and ".".join(lowercase_ ): # linear layer A__ : Optional[Any] = flax_key_tuple[:-1] + ("""weight""",) A__ : int = flax_tensor.T elif flax_key_tuple[-1] in ["scale", "embedding"]: A__ : Optional[int] = flax_key_tuple[:-1] + ("""weight""",) return flax_key_tuple, flax_tensor def UpperCamelCase (lowercase_: Tuple , lowercase_: Optional[int] , lowercase_: str ) -> Union[str, Any]: if "metadata" in layer: A__ : Tuple = layer.split("""metadata""" ) A__ : Optional[Any] = """""".join(split_layer[0] )[:-1] A__ : Optional[Any] = [tuple(("""metadata""" + split_layer[1]).split("""/""" ) )] elif "kvstore" in layer: A__ : str = layer.split("""kvstore""" ) A__ : int = """""".join(split_layer[0] )[:-1] A__ : Optional[int] = [tuple(("""kvstore""" + split_layer[1]).split("""/""" ) )] else: A__ : Any = layer.split("""/""" ) A__ : int = """/""".join(split_layer[:-1] ) A__ : str = (split_layer[-1],) if "kvstore/path" in layer: A__ : Dict = f"""{switch_checkpoint_path}/{checkpoint_info[layer]}""" elif "kvstore/driver" in layer: A__ : Optional[int] = """file""" else: A__ : str = checkpoint_info[layer] return curr_real_layer_name, split_layer, content def UpperCamelCase (lowercase_: str , lowercase_: List[Any] ) -> int: A__ : int = rename_keys(lowercase_ ) A__ : Any = {} for k, v in current_block.items(): A__ : Dict = v A__ : str = new_current_block torch.save(lowercase_ , lowercase_ ) def UpperCamelCase (lowercase_: Dict , lowercase_: Optional[Any] , lowercase_: Optional[Any] , lowercase_: Optional[int] , lowercase_: str = WEIGHTS_NAME ) -> Tuple: A__ : Optional[int] = convert_file_size_to_int(lowercase_ ) A__ : List[Any] = [] A__ : int = {} A__ : List[str] = 0 A__ : Any = 0 os.makedirs(lowercase_ , exist_ok=lowercase_ ) with gfile.GFile(switch_checkpoint_path + """/checkpoint""" , """rb""" ) as fp: A__ : Optional[Any] = serialization.msgpack_restore(fp.read() )["""optimizer"""]["""target"""] A__ : Dict = flatten_dict(lowercase_ , sep="""/""" ) A__ : Any = {} for layer in checkpoint_info.keys(): A__ , A__ , A__ : Union[str, Any] = get_key_and_tensorstore_dict( lowercase_ , lowercase_ , lowercase_ ) if curr_real_layer_name in all_layers: A__ : Optional[int] = content else: A__ : List[Any] = {split_layer[-1]: content} for key in all_layers.keys(): # open tensorstore file A__ : Optional[Any] = ts.open(unflatten_dict(all_layers[key] ) ).result().read().result() A__ : List[Any] = torch.tensor(lowercase_ ) A__ : List[Any] = raw_weights.numel() * dtype_byte_size(raw_weights.dtype ) # use the renaming pattern from the small conversion scripts A__ , A__ : Any = rename_base_flax_keys(tuple(key.split("""/""" ) ) , lowercase_ ) A__ : Any = """/""".join(lowercase_ ) # If this weight is going to tip up over the maximal size, we split. if current_block_size + weight_size > max_shard_size: A__ : List[Any] = os.path.join( lowercase_ , weights_name.replace(""".bin""" , f"""-{len(lowercase_ )+1:05d}-of-???.bin""" ) ) rename_and_save_block(lowercase_ , lowercase_ ) sharded_state_dicts.append(current_block.keys() ) del current_block A__ : Any = {} A__ : str = 0 A__ : List[str] = raw_weights.to(getattr(lowercase_ , lowercase_ ) ) current_block_size += weight_size total_size += weight_size # Add the last block A__ : Union[str, Any] = os.path.join(lowercase_ , weights_name.replace(""".bin""" , f"""-{len(lowercase_ )+1:05d}-of-???.bin""" ) ) rename_and_save_block(lowercase_ , lowercase_ ) sharded_state_dicts.append(current_block.keys() ) # If we only have one shard, we return it if len(lowercase_ ) == 1: return {weights_name: sharded_state_dicts[0]}, None # Otherwise, let's build the index A__ : str = {} A__ : Any = {} for idx, shard in enumerate(lowercase_ ): A__ : Any = weights_name.replace( """.bin""" , f"""-{idx+1:05d}-of-{len(lowercase_ ):05d}.bin""" ) # len(sharded_state_dicts):05d} A__ : Dict = os.path.join(lowercase_ , weights_name.replace(""".bin""" , f"""-{idx+1:05d}-of-???.bin""" ) ) os.rename(lowercase_ , os.path.join(lowercase_ , lowercase_ ) ) A__ : str = shard for key in shard: A__ : Any = shard_file # Add the metadata A__ : Tuple = {"""total_size""": total_size} A__ : Union[str, Any] = {"""metadata""": metadata, """weight_map""": weight_map} with open(os.path.join(lowercase_ , lowercase_ ) , """w""" , encoding="""utf-8""" ) as f: A__ : Dict = json.dumps(lowercase_ , indent=2 , sort_keys=lowercase_ ) + """\n""" f.write(lowercase_ ) return metadata, index if __name__ == "__main__": A_ : Optional[int] = argparse.ArgumentParser() # Required parameters parser.add_argument( '--switch_t5x_checkpoint_path', default='/mnt/disks/disk_switch/original_checkpoints/switch-xxl-128/checkpoint_634600', type=str, required=False, help='Path to a directory containing a folder per layer. Follows the original Google format.', ) parser.add_argument('--max_shard_size', default='10GB', required=False, help='Max shard size') parser.add_argument('--dtype', default='bfloat16', type=str, required=False, help='dtype of the saved model') parser.add_argument( '--pytorch_dump_folder_path', default='/mnt/disks/disk_switch/original_checkpoints/switch-xxl-128-converted', type=str, required=False, help='Path to the output pytorch model.', ) A_ : Dict = parser.parse_args() shard_on_the_fly( args.switch_tax_checkpoint_path, args.pytorch_dump_folder_path, args.max_shard_size, args.dtype, ) def UpperCamelCase () -> int: from transformers import SwitchTransformersConfig, SwitchTransformersForConditionalGeneration, TaTokenizer A__ : str = SwitchTransformersConfig.from_pretrained("""google/switch-base-8""" ) config.save_pretrained("""/home/arthur_huggingface_co/transformers/switch_converted""" ) A__ : str = SwitchTransformersForConditionalGeneration.from_pretrained( """/home/arthur_huggingface_co/transformers/switch_converted""" , device_map="""auto""" ) A__ : Tuple = TaTokenizer.from_pretrained("""t5-small""" ) A__ : Dict = """A <extra_id_0> walks into a bar a orders a <extra_id_1> with <extra_id_2> pinch of <extra_id_3>.""" A__ : Union[str, Any] = tokenizer(lowercase_ , return_tensors="""pt""" ).input_ids A__ : Tuple = model.generate(lowercase_ , decoder_start_token_id=0 ) print(tokenizer.decode(out[0] ) )
64
1
import math import unittest from transformers import BioGptConfig, is_torch_available from transformers.testing_utils import require_torch, slow, torch_device from ...generation.test_utils import GenerationTesterMixin from ...test_configuration_common import ConfigTester from ...test_modeling_common import ModelTesterMixin, ids_tensor, random_attention_mask from ...test_pipeline_mixin import PipelineTesterMixin if is_torch_available(): import torch from transformers import ( BioGptForCausalLM, BioGptForSequenceClassification, BioGptForTokenClassification, BioGptModel, BioGptTokenizer, ) from transformers.models.biogpt.modeling_biogpt import BIOGPT_PRETRAINED_MODEL_ARCHIVE_LIST class _a : '''simple docstring''' def __init__( self , A__ , A__=13 , A__=7 , A__=True , A__=True , A__=False , A__=True , A__=99 , A__=32 , A__=5 , A__=4 , A__=37 , A__="gelu" , A__=0.1 , A__=0.1 , A__=512 , A__=16 , A__=2 , A__=0.0_2 , A__=3 , A__=4 , A__=None , ): A__ : List[Any] = parent A__ : str = batch_size A__ : List[Any] = seq_length A__ : str = is_training A__ : Tuple = use_input_mask A__ : List[Any] = use_token_type_ids A__ : Tuple = use_labels A__ : Dict = vocab_size A__ : List[str] = hidden_size A__ : Union[str, Any] = num_hidden_layers A__ : Optional[Any] = num_attention_heads A__ : Any = intermediate_size A__ : Optional[int] = hidden_act A__ : Dict = hidden_dropout_prob A__ : Dict = attention_probs_dropout_prob A__ : Optional[Any] = max_position_embeddings A__ : List[str] = type_vocab_size A__ : Optional[int] = type_sequence_label_size A__ : Dict = initializer_range A__ : Union[str, Any] = num_labels A__ : List[str] = num_choices A__ : Union[str, Any] = scope def __A ( self ): A__ : Tuple = ids_tensor([self.batch_size, self.seq_length] , self.vocab_size ) A__ : List[Any] = None if self.use_input_mask: A__ : List[Any] = random_attention_mask([self.batch_size, self.seq_length] ) A__ : Optional[Any] = None if self.use_token_type_ids: A__ : List[Any] = ids_tensor([self.batch_size, self.seq_length] , self.type_vocab_size ) A__ : Tuple = None A__ : Optional[Any] = None A__ : str = None if self.use_labels: A__ : List[str] = ids_tensor([self.batch_size] , self.type_sequence_label_size ) A__ : Optional[int] = ids_tensor([self.batch_size, self.seq_length] , self.num_labels ) A__ : str = ids_tensor([self.batch_size] , self.num_choices ) A__ : Dict = self.get_config() return config, input_ids, token_type_ids, input_mask, sequence_labels, token_labels, choice_labels def __A ( self ): return BioGptConfig( vocab_size=self.vocab_size , hidden_size=self.hidden_size , num_hidden_layers=self.num_hidden_layers , num_attention_heads=self.num_attention_heads , intermediate_size=self.intermediate_size , hidden_act=self.hidden_act , hidden_dropout_prob=self.hidden_dropout_prob , attention_probs_dropout_prob=self.attention_probs_dropout_prob , max_position_embeddings=self.max_position_embeddings , type_vocab_size=self.type_vocab_size , is_decoder=A__ , initializer_range=self.initializer_range , ) def __A ( self , A__ , A__ , A__ , A__ , A__ , A__ , A__ ): A__ : Optional[Any] = BioGptModel(config=A__ ) model.to(A__ ) model.eval() A__ : Optional[Any] = model(A__ , attention_mask=A__ ) A__ : Optional[Any] = model(A__ ) self.parent.assertEqual(result.last_hidden_state.shape , (self.batch_size, self.seq_length, self.hidden_size) ) def __A ( self , A__ , A__ , A__ , A__ , A__ , A__ , A__ , A__ , A__ , ): A__ : List[str] = BioGptForCausalLM(config=A__ ) model.to(A__ ) model.eval() A__ : Optional[Any] = model(A__ , attention_mask=A__ , token_type_ids=A__ , labels=A__ ) self.parent.assertEqual(result.logits.shape , (self.batch_size, self.seq_length, self.vocab_size) ) def __A ( self , A__ , A__ , A__ , A__ , A__ , *A__ ): A__ : str = BioGptModel(config=A__ ) model.to(A__ ) model.eval() # create attention mask A__ : str = torch.ones(input_ids.shape , dtype=torch.long , device=A__ ) A__ : str = self.seq_length // 2 A__ : Any = 0 # first forward pass A__ , A__ : List[str] = model(A__ , attention_mask=A__ ).to_tuple() # create hypothetical next token and extent to next_input_ids A__ : str = ids_tensor((self.batch_size, 1) , config.vocab_size ) # change a random masked slice from input_ids A__ : str = ids_tensor((1,) , A__ ).item() + 1 A__ : List[str] = ids_tensor((self.batch_size, 1) , config.vocab_size ).squeeze(-1 ) A__ : Dict = random_other_next_tokens # append to next input_ids and attn_mask A__ : Tuple = torch.cat([input_ids, next_tokens] , dim=-1 ) A__ : int = torch.cat( [attn_mask, torch.ones((attn_mask.shape[0], 1) , dtype=torch.long , device=A__ )] , dim=1 , ) # get two different outputs A__ : Any = model(A__ , attention_mask=A__ )["""last_hidden_state"""] A__ : Optional[int] = model(A__ , past_key_values=A__ , attention_mask=A__ )["""last_hidden_state"""] # select random slice A__ : Optional[Any] = ids_tensor((1,) , output_from_past.shape[-1] ).item() A__ : Optional[int] = output_from_no_past[:, -1, random_slice_idx].detach() A__ : List[str] = output_from_past[:, 0, random_slice_idx].detach() # test that outputs are equal for slice self.parent.assertTrue(torch.allclose(A__ , A__ , atol=1e-3 ) ) def __A ( self , A__ , A__ , A__ , A__ , A__ , *A__ ): A__ : Optional[Any] = BioGptModel(config=A__ ).to(A__ ).eval() A__ : int = torch.ones(input_ids.shape , dtype=torch.long , device=A__ ) # first forward pass A__ : int = model(A__ , attention_mask=A__ , use_cache=A__ ) A__ , A__ : Union[str, Any] = outputs.to_tuple() # create hypothetical multiple next token and extent to next_input_ids A__ : int = ids_tensor((self.batch_size, 3) , config.vocab_size ) A__ : List[str] = ids_tensor((self.batch_size, 3) , 2 ) # append to next input_ids and A__ : Optional[int] = torch.cat([input_ids, next_tokens] , dim=-1 ) A__ : Union[str, Any] = torch.cat([attention_mask, next_attn_mask] , dim=-1 ) A__ : str = model(A__ , attention_mask=A__ )["""last_hidden_state"""] A__ : Optional[int] = model(A__ , attention_mask=A__ , past_key_values=A__ )[ """last_hidden_state""" ] # select random slice A__ : List[str] = ids_tensor((1,) , output_from_past.shape[-1] ).item() A__ : int = output_from_no_past[:, -3:, random_slice_idx].detach() A__ : List[Any] = output_from_past[:, :, random_slice_idx].detach() self.parent.assertTrue(output_from_past_slice.shape[1] == next_tokens.shape[1] ) # test that outputs are equal for slice self.parent.assertTrue(torch.allclose(A__ , A__ , atol=1e-3 ) ) def __A ( self , A__ , A__ , A__ , A__ , A__ , *A__ , A__=False ): A__ : Optional[int] = BioGptForCausalLM(A__ ) model.to(A__ ) if gradient_checkpointing: model.gradient_checkpointing_enable() A__ : str = model(A__ , labels=A__ ) self.parent.assertEqual(result.loss.shape , () ) self.parent.assertEqual(result.logits.shape , (self.batch_size, self.seq_length, self.vocab_size) ) result.loss.backward() def __A ( self , A__ , *A__ ): A__ : Optional[Any] = BioGptModel(A__ ) A__ : str = model.config.initializer_range / math.sqrt(2 * model.config.num_hidden_layers ) for key in model.state_dict().keys(): if "c_proj" in key and "weight" in key: self.parent.assertLessEqual(abs(torch.std(model.state_dict()[key] ) - model_std ) , 0.0_0_1 ) self.parent.assertLessEqual(abs(torch.mean(model.state_dict()[key] ) - 0.0 ) , 0.0_1 ) def __A ( self , A__ , A__ , A__ , A__ , A__ , *A__ ): A__ : List[str] = self.num_labels A__ : str = BioGptForTokenClassification(A__ ) model.to(A__ ) model.eval() A__ : Any = model(A__ , attention_mask=A__ , token_type_ids=A__ ) self.parent.assertEqual(result.logits.shape , (self.batch_size, self.seq_length, self.num_labels) ) def __A ( self ): A__ : Dict = self.prepare_config_and_inputs() ( ( A__ ) , ( A__ ) , ( A__ ) , ( A__ ) , ( A__ ) , ( A__ ) , ( A__ ) , ) : List[str] = config_and_inputs A__ : int = {"""input_ids""": input_ids, """attention_mask""": input_mask} return config, inputs_dict @require_torch class _a (__magic_name__ , __magic_name__ , __magic_name__ , unittest.TestCase ): '''simple docstring''' UpperCAmelCase__: int = ( (BioGptModel, BioGptForCausalLM, BioGptForSequenceClassification, BioGptForTokenClassification) if is_torch_available() else () ) UpperCAmelCase__: Dict = (BioGptForCausalLM,) if is_torch_available() else () UpperCAmelCase__: List[Any] = ( { '''feature-extraction''': BioGptModel, '''text-classification''': BioGptForSequenceClassification, '''text-generation''': BioGptForCausalLM, '''token-classification''': BioGptForTokenClassification, '''zero-shot''': BioGptForSequenceClassification, } if is_torch_available() else {} ) UpperCAmelCase__: int = False def __A ( self ): A__ : Optional[int] = BioGptModelTester(self ) A__ : Optional[int] = ConfigTester(self , config_class=A__ , hidden_size=37 ) def __A ( self ): self.config_tester.run_common_tests() def __A ( self ): A__ : Tuple = self.model_tester.prepare_config_and_inputs() self.model_tester.create_and_check_model(*A__ ) def __A ( self ): A__ : Optional[int] = self.model_tester.prepare_config_and_inputs() for type in ["absolute", "relative_key", "relative_key_query"]: A__ : Optional[Any] = type self.model_tester.create_and_check_model(*A__ ) def __A ( self ): A__ : Optional[Any] = self.model_tester.prepare_config_and_inputs() self.model_tester.create_and_check_biogpt_model_attention_mask_past(*A__ ) def __A ( self ): A__ : int = self.model_tester.prepare_config_and_inputs() self.model_tester.create_and_check_forward_and_backwards(*A__ , gradient_checkpointing=A__ ) def __A ( self ): A__ : Tuple = self.model_tester.prepare_config_and_inputs() self.model_tester.create_and_check_biogpt_model_past_large_inputs(*A__ ) def __A ( self ): A__ : int = self.model_tester.prepare_config_and_inputs() self.model_tester.create_and_check_biogpt_weight_initialization(*A__ ) def __A ( self ): A__ : Dict = self.model_tester.prepare_config_and_inputs() self.model_tester.create_and_check_biogpt_for_token_classification(*A__ ) @slow def __A ( self ): A__ : Union[str, Any] = BioGptForCausalLM.from_pretrained("""microsoft/biogpt""" ) model.to(A__ ) A__ : Tuple = BioGptTokenizer.from_pretrained("""microsoft/biogpt""" ) A__ : Optional[Any] = """left""" # Define PAD Token = EOS Token = 50256 A__ : Dict = tokenizer.eos_token A__ : Dict = model.config.eos_token_id # use different length sentences to test batching A__ : Union[str, Any] = [ """Hello, my dog is a little""", """Today, I""", ] A__ : int = tokenizer(A__ , return_tensors="""pt""" , padding=A__ ) A__ : Union[str, Any] = inputs["""input_ids"""].to(A__ ) A__ : Optional[Any] = model.generate( input_ids=A__ , attention_mask=inputs["""attention_mask"""].to(A__ ) , ) A__ : str = tokenizer(sentences[0] , return_tensors="""pt""" ).input_ids.to(A__ ) A__ : Dict = model.generate(input_ids=A__ ) A__ : int = inputs_non_padded.shape[-1] - inputs["""attention_mask"""][-1].long().sum().cpu().item() A__ : int = tokenizer(sentences[1] , return_tensors="""pt""" ).input_ids.to(A__ ) A__ : List[Any] = model.generate(input_ids=A__ , max_length=model.config.max_length - num_paddings ) A__ : Union[str, Any] = tokenizer.batch_decode(A__ , skip_special_tokens=A__ ) A__ : Any = tokenizer.decode(output_non_padded[0] , skip_special_tokens=A__ ) A__ : int = tokenizer.decode(output_padded[0] , skip_special_tokens=A__ ) A__ : Optional[Any] = [ """Hello, my dog is a little bit bigger than a little bit.""", """Today, I have a good idea of how to use the information""", ] self.assertListEqual(A__ , A__ ) self.assertListEqual(A__ , [non_padded_sentence, padded_sentence] ) @slow def __A ( self ): for model_name in BIOGPT_PRETRAINED_MODEL_ARCHIVE_LIST[:1]: A__ : int = BioGptModel.from_pretrained(A__ ) self.assertIsNotNone(A__ ) def __A ( self ): A__ , A__ : int = self.model_tester.prepare_config_and_inputs_for_common() A__ : str = 3 A__ : Union[str, Any] = input_dict["""input_ids"""] A__ : List[str] = input_ids.ne(1 ).to(A__ ) A__ : List[Any] = ids_tensor([self.model_tester.batch_size] , self.model_tester.type_sequence_label_size ) A__ : str = BioGptForSequenceClassification(A__ ) model.to(A__ ) model.eval() A__ : List[Any] = model(A__ , attention_mask=A__ , labels=A__ ) self.assertEqual(result.logits.shape , (self.model_tester.batch_size, self.model_tester.num_labels) ) def __A ( self ): A__ , A__ : Tuple = self.model_tester.prepare_config_and_inputs_for_common() A__ : Any = 3 A__ : Tuple = """multi_label_classification""" A__ : Tuple = input_dict["""input_ids"""] A__ : Optional[int] = input_ids.ne(1 ).to(A__ ) A__ : List[Any] = ids_tensor( [self.model_tester.batch_size, config.num_labels] , self.model_tester.type_sequence_label_size ).to(torch.float ) A__ : List[str] = BioGptForSequenceClassification(A__ ) model.to(A__ ) model.eval() A__ : Any = model(A__ , attention_mask=A__ , labels=A__ ) self.assertEqual(result.logits.shape , (self.model_tester.batch_size, self.model_tester.num_labels) ) @require_torch class _a (unittest.TestCase ): '''simple docstring''' @slow def __A ( self ): A__ : int = BioGptForCausalLM.from_pretrained("""microsoft/biogpt""" ) A__ : List[str] = torch.tensor([[2, 4805, 9, 656, 21]] ) A__ : Union[str, Any] = model(A__ )[0] A__ : List[str] = 4_2384 A__ : Tuple = torch.Size((1, 5, vocab_size) ) self.assertEqual(output.shape , A__ ) A__ : List[str] = torch.tensor( [[[-9.5_2_3_6, -9.8_9_1_8, 1_0.4_5_5_7], [-1_1.0_4_6_9, -9.6_4_2_3, 8.1_0_2_2], [-8.8_6_6_4, -7.8_8_2_6, 5.5_3_2_5]]] ) self.assertTrue(torch.allclose(output[:, :3, :3] , A__ , atol=1e-4 ) ) @slow def __A ( self ): A__ : List[str] = BioGptTokenizer.from_pretrained("""microsoft/biogpt""" ) A__ : int = BioGptForCausalLM.from_pretrained("""microsoft/biogpt""" ) model.to(A__ ) torch.manual_seed(0 ) A__ : Dict = tokenizer("""COVID-19 is""" , return_tensors="""pt""" ).to(A__ ) A__ : Any = model.generate( **A__ , min_length=100 , max_length=1024 , num_beams=5 , early_stopping=A__ , ) A__ : Union[str, Any] = tokenizer.decode(output_ids[0] , skip_special_tokens=A__ ) A__ : Union[str, Any] = ( """COVID-19 is a global pandemic caused by severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), the""" """ causative agent of coronavirus disease 2019 (COVID-19), which has spread to more than 200 countries and""" """ territories, including the United States (US), Canada, Australia, New Zealand, the United Kingdom (UK),""" """ and the United States of America (USA), as of March 11, 2020, with more than 800,000 confirmed cases and""" """ more than 800,000 deaths.""" ) self.assertEqual(A__ , A__ )
64
from typing import TYPE_CHECKING from ...utils import OptionalDependencyNotAvailable, _LazyModule, is_sentencepiece_available A_ : Optional[Any] = {} try: if not is_sentencepiece_available(): raise OptionalDependencyNotAvailable() except OptionalDependencyNotAvailable: pass else: A_ : str = ['BartphoTokenizer'] if TYPE_CHECKING: try: if not is_sentencepiece_available(): raise OptionalDependencyNotAvailable() except OptionalDependencyNotAvailable: pass else: from .tokenization_bartpho import BartphoTokenizer else: import sys A_ : Optional[int] = _LazyModule(__name__, globals()['__file__'], _import_structure, module_spec=__spec__)
64
1
import numpy as np def UpperCamelCase (lowercase_: Tuple , lowercase_: int , lowercase_: str , lowercase_: Any , lowercase_: Any ) -> List[Any]: A__ : int = int(np.ceil((x_end - xa) / h ) ) A__ : str = np.zeros((n + 1,) ) A__ : int = ya A__ : List[Any] = xa for k in range(lowercase_ ): A__ : int = f(lowercase_ , y[k] ) A__ : int = f(x + 0.5 * h , y[k] + 0.5 * h * ka ) A__ : int = f(x + 0.5 * h , y[k] + 0.5 * h * ka ) A__ : int = f(x + h , y[k] + h * ka ) A__ : Optional[Any] = y[k] + (1 / 6) * h * (ka + 2 * ka + 2 * ka + ka) x += h return y if __name__ == "__main__": import doctest doctest.testmod()
64
import argparse import hashlib import os import urllib import warnings import torch from torch import nn from tqdm import tqdm from transformers import WhisperConfig, WhisperForConditionalGeneration A_ : Dict = { 'tiny.en': 'https://openaipublic.azureedge.net/main/whisper/models/d3dd57d32accea0b295c96e26691aa14d8822fac7d9d27d5dc00b4ca2826dd03/tiny.en.pt', 'tiny': 'https://openaipublic.azureedge.net/main/whisper/models/65147644a518d12f04e32d6f3b26facc3f8dd46e5390956a9424a650c0ce22b9/tiny.pt', 'base.en': 'https://openaipublic.azureedge.net/main/whisper/models/25a8566e1d0c1e2231d1c762132cd20e0f96a85d16145c3a00adf5d1ac670ead/base.en.pt', 'base': 'https://openaipublic.azureedge.net/main/whisper/models/ed3a0b6b1c0edf879ad9b11b1af5a0e6ab5db9205f891f668f8b0e6c6326e34e/base.pt', 'small.en': 'https://openaipublic.azureedge.net/main/whisper/models/f953ad0fd29cacd07d5a9eda5624af0f6bcf2258be67c92b79389873d91e0872/small.en.pt', 'small': 'https://openaipublic.azureedge.net/main/whisper/models/9ecf779972d90ba49c06d968637d720dd632c55bbf19d441fb42bf17a411e794/small.pt', 'medium.en': 'https://openaipublic.azureedge.net/main/whisper/models/d7440d1dc186f76616474e0ff0b3b6b879abc9d1a4926b7adfa41db2d497ab4f/medium.en.pt', 'medium': 'https://openaipublic.azureedge.net/main/whisper/models/345ae4da62f9b3d59415adc60127b97c714f32e89e936602e85993674d08dcb1/medium.pt', 'large': 'https://openaipublic.azureedge.net/main/whisper/models/e4b87e7e0bf463eb8e6956e646f1e277e901512310def2c24bf0e11bd3c28e9a/large.pt', 'large-v2': 'https://openaipublic.azureedge.net/main/whisper/models/81f7c96c852ee8fc832187b0132e569d6c3065a3252ed18e56effd0b6a73e524/large-v2.pt', } def UpperCamelCase (lowercase_: Optional[Any] ) -> Optional[int]: A__ : List[Any] = ["""layers""", """blocks"""] for k in ignore_keys: state_dict.pop(lowercase_ , lowercase_ ) A_ : Any = { 'blocks': 'layers', 'mlp.0': 'fc1', 'mlp.2': 'fc2', 'mlp_ln': 'final_layer_norm', '.attn.query': '.self_attn.q_proj', '.attn.key': '.self_attn.k_proj', '.attn.value': '.self_attn.v_proj', '.attn_ln': '.self_attn_layer_norm', '.attn.out': '.self_attn.out_proj', '.cross_attn.query': '.encoder_attn.q_proj', '.cross_attn.key': '.encoder_attn.k_proj', '.cross_attn.value': '.encoder_attn.v_proj', '.cross_attn_ln': '.encoder_attn_layer_norm', '.cross_attn.out': '.encoder_attn.out_proj', 'decoder.ln.': 'decoder.layer_norm.', 'encoder.ln.': 'encoder.layer_norm.', 'token_embedding': 'embed_tokens', 'encoder.positional_embedding': 'encoder.embed_positions.weight', 'decoder.positional_embedding': 'decoder.embed_positions.weight', 'ln_post': 'layer_norm', } def UpperCamelCase (lowercase_: str ) -> Any: A__ : Dict = list(s_dict.keys() ) for key in keys: A__ : List[str] = key for k, v in WHISPER_MAPPING.items(): if k in key: A__ : List[Any] = new_key.replace(lowercase_ , lowercase_ ) print(f"""{key} -> {new_key}""" ) A__ : Tuple = s_dict.pop(lowercase_ ) return s_dict def UpperCamelCase (lowercase_: Tuple ) -> Optional[int]: A__ , A__ : Any = emb.weight.shape A__ : str = nn.Linear(lowercase_ , lowercase_ , bias=lowercase_ ) A__ : Union[str, Any] = emb.weight.data return lin_layer def UpperCamelCase (lowercase_: str , lowercase_: str ) -> bytes: os.makedirs(lowercase_ , exist_ok=lowercase_ ) A__ : Tuple = os.path.basename(lowercase_ ) A__ : int = url.split("""/""" )[-2] A__ : Dict = os.path.join(lowercase_ , lowercase_ ) if os.path.exists(lowercase_ ) and not os.path.isfile(lowercase_ ): raise RuntimeError(f"""{download_target} exists and is not a regular file""" ) if os.path.isfile(lowercase_ ): A__ : Optional[Any] = open(lowercase_ , """rb""" ).read() if hashlib.shaaaa(lowercase_ ).hexdigest() == expected_shaaaa: return model_bytes else: warnings.warn(f"""{download_target} exists, but the SHA256 checksum does not match; re-downloading the file""" ) with urllib.request.urlopen(lowercase_ ) as source, open(lowercase_ , """wb""" ) as output: with tqdm( total=int(source.info().get("""Content-Length""" ) ) , ncols=80 , unit="""iB""" , unit_scale=lowercase_ , unit_divisor=1024 ) as loop: while True: A__ : Any = source.read(8192 ) if not buffer: break output.write(lowercase_ ) loop.update(len(lowercase_ ) ) A__ : Dict = open(lowercase_ , """rb""" ).read() if hashlib.shaaaa(lowercase_ ).hexdigest() != expected_shaaaa: raise RuntimeError( """Model has been downloaded but the SHA256 checksum does not not match. Please retry loading the model.""" ) return model_bytes def UpperCamelCase (lowercase_: List[Any] , lowercase_: Tuple ) -> Optional[Any]: if ".pt" not in checkpoint_path: A__ : Tuple = _download(_MODELS[checkpoint_path] ) else: A__ : Optional[int] = torch.load(lowercase_ , map_location="""cpu""" ) A__ : str = original_checkpoint["""dims"""] A__ : List[Any] = original_checkpoint["""model_state_dict"""] A__ : Optional[Any] = state_dict["""decoder.token_embedding.weight"""] remove_ignore_keys_(lowercase_ ) rename_keys(lowercase_ ) A__ : List[str] = True A__ : Optional[Any] = state_dict["""decoder.layers.0.fc1.weight"""].shape[0] A__ : List[Any] = WhisperConfig( vocab_size=dimensions["""n_vocab"""] , encoder_ffn_dim=lowercase_ , decoder_ffn_dim=lowercase_ , num_mel_bins=dimensions["""n_mels"""] , d_model=dimensions["""n_audio_state"""] , max_target_positions=dimensions["""n_text_ctx"""] , encoder_layers=dimensions["""n_audio_layer"""] , encoder_attention_heads=dimensions["""n_audio_head"""] , decoder_layers=dimensions["""n_text_layer"""] , decoder_attention_heads=dimensions["""n_text_state"""] , max_source_positions=dimensions["""n_audio_ctx"""] , ) A__ : Optional[Any] = WhisperForConditionalGeneration(lowercase_ ) A__ , A__ : List[Any] = model.model.load_state_dict(lowercase_ , strict=lowercase_ ) if len(lowercase_ ) > 0 and not set(lowercase_ ) <= { "encoder.embed_positions.weights", "decoder.embed_positions.weights", }: raise ValueError( """Only `encoder.embed_positions.weights` and `decoder.embed_positions.weights` are allowed to be missing,""" f""" but all the following weights are missing {missing}""" ) if tie_embeds: A__ : Any = make_linear_from_emb(model.model.decoder.embed_tokens ) else: A__ : str = proj_out_weights model.save_pretrained(lowercase_ ) if __name__ == "__main__": A_ : Any = argparse.ArgumentParser() # # Required parameters parser.add_argument('--checkpoint_path', type=str, help='Patht to the downloaded checkpoints') parser.add_argument('--pytorch_dump_folder_path', default=None, type=str, help='Path to the output PyTorch model.') A_ : Tuple = parser.parse_args() convert_openai_whisper_to_tfms(args.checkpoint_path, args.pytorch_dump_folder_path)
64
1
import inspect import unittest class _a (unittest.TestCase ): '''simple docstring''' def __A ( self ): try: import diffusers # noqa: F401 except ImportError: assert False def __A ( self ): import diffusers from diffusers.dependency_versions_table import deps A__ : Tuple = inspect.getmembers(A__ , inspect.isclass ) for cls_name, cls_module in all_classes: if "dummy_" in cls_module.__module__: for backend in cls_module._backends: if backend == "k_diffusion": A__ : Dict = """k-diffusion""" elif backend == "invisible_watermark": A__ : List[Any] = """invisible-watermark""" assert backend in deps, F"""{backend} is not in the deps table!"""
64
import unittest import numpy as np import torch from transformers import CLIPTextConfig, CLIPTextModel, CLIPTokenizer from diffusers import ( AutoencoderKL, DDIMScheduler, DPMSolverMultistepScheduler, TextToVideoSDPipeline, UNetaDConditionModel, ) from diffusers.utils import is_xformers_available, load_numpy, skip_mps, slow, torch_device from diffusers.utils.testing_utils import enable_full_determinism from ..pipeline_params import TEXT_TO_IMAGE_BATCH_PARAMS, TEXT_TO_IMAGE_PARAMS from ..test_pipelines_common import PipelineTesterMixin enable_full_determinism() @skip_mps class _a (__magic_name__ , unittest.TestCase ): '''simple docstring''' UpperCAmelCase__: Any = TextToVideoSDPipeline UpperCAmelCase__: Any = TEXT_TO_IMAGE_PARAMS UpperCAmelCase__: Optional[Any] = TEXT_TO_IMAGE_BATCH_PARAMS # No `output_type`. UpperCAmelCase__: Optional[int] = frozenset( [ '''num_inference_steps''', '''generator''', '''latents''', '''return_dict''', '''callback''', '''callback_steps''', ] ) def __A ( self ): torch.manual_seed(0 ) A__ : Optional[int] = UNetaDConditionModel( block_out_channels=(32, 64, 64, 64) , layers_per_block=2 , sample_size=32 , in_channels=4 , out_channels=4 , down_block_types=("""CrossAttnDownBlock3D""", """CrossAttnDownBlock3D""", """CrossAttnDownBlock3D""", """DownBlock3D""") , up_block_types=("""UpBlock3D""", """CrossAttnUpBlock3D""", """CrossAttnUpBlock3D""", """CrossAttnUpBlock3D""") , cross_attention_dim=32 , attention_head_dim=4 , ) A__ : Optional[int] = DDIMScheduler( beta_start=0.0_0_0_8_5 , beta_end=0.0_1_2 , beta_schedule="""scaled_linear""" , clip_sample=A__ , set_alpha_to_one=A__ , ) torch.manual_seed(0 ) A__ : Union[str, Any] = AutoencoderKL( block_out_channels=[32, 64] , in_channels=3 , out_channels=3 , down_block_types=["""DownEncoderBlock2D""", """DownEncoderBlock2D"""] , up_block_types=["""UpDecoderBlock2D""", """UpDecoderBlock2D"""] , latent_channels=4 , sample_size=128 , ) torch.manual_seed(0 ) A__ : Union[str, Any] = CLIPTextConfig( bos_token_id=0 , eos_token_id=2 , hidden_size=32 , intermediate_size=37 , layer_norm_eps=1e-05 , num_attention_heads=4 , num_hidden_layers=5 , pad_token_id=1 , vocab_size=1000 , hidden_act="""gelu""" , projection_dim=512 , ) A__ : Union[str, Any] = CLIPTextModel(A__ ) A__ : Tuple = CLIPTokenizer.from_pretrained("""hf-internal-testing/tiny-random-clip""" ) A__ : Dict = { """unet""": unet, """scheduler""": scheduler, """vae""": vae, """text_encoder""": text_encoder, """tokenizer""": tokenizer, } return components def __A ( self , A__ , A__=0 ): if str(A__ ).startswith("""mps""" ): A__ : Tuple = torch.manual_seed(A__ ) else: A__ : List[str] = torch.Generator(device=A__ ).manual_seed(A__ ) A__ : List[str] = { """prompt""": """A painting of a squirrel eating a burger""", """generator""": generator, """num_inference_steps""": 2, """guidance_scale""": 6.0, """output_type""": """pt""", } return inputs def __A ( self ): A__ : List[str] = """cpu""" # ensure determinism for the device-dependent torch.Generator A__ : Union[str, Any] = self.get_dummy_components() A__ : Union[str, Any] = TextToVideoSDPipeline(**A__ ) A__ : int = sd_pipe.to(A__ ) sd_pipe.set_progress_bar_config(disable=A__ ) A__ : int = self.get_dummy_inputs(A__ ) A__ : int = """np""" A__ : Any = sd_pipe(**A__ ).frames A__ : Dict = frames[0][-3:, -3:, -1] assert frames[0].shape == (64, 64, 3) A__ : Optional[Any] = np.array([1_5_8.0, 1_6_0.0, 1_5_3.0, 1_2_5.0, 1_0_0.0, 1_2_1.0, 1_1_1.0, 9_3.0, 1_1_3.0] ) assert np.abs(image_slice.flatten() - expected_slice ).max() < 1e-2 def __A ( self ): self._test_attention_slicing_forward_pass(test_mean_pixel_difference=A__ , expected_max_diff=3e-3 ) @unittest.skipIf( torch_device != """cuda""" or not is_xformers_available() , reason="""XFormers attention is only available with CUDA and `xformers` installed""" , ) def __A ( self ): self._test_xformers_attention_forwardGenerator_pass(test_mean_pixel_difference=A__ , expected_max_diff=1e-2 ) @unittest.skip(reason="""Batching needs to be properly figured out first for this pipeline.""" ) def __A ( self ): pass @unittest.skip(reason="""Batching needs to be properly figured out first for this pipeline.""" ) def __A ( self ): pass @unittest.skip(reason="""`num_images_per_prompt` argument is not supported for this pipeline.""" ) def __A ( self ): pass def __A ( self ): return super().test_progress_bar() @slow @skip_mps class _a (unittest.TestCase ): '''simple docstring''' def __A ( self ): A__ : Union[str, Any] = load_numpy( """https://huggingface.co/datasets/hf-internal-testing/diffusers-images/resolve/main/text_to_video/video.npy""" ) A__ : Tuple = TextToVideoSDPipeline.from_pretrained("""damo-vilab/text-to-video-ms-1.7b""" ) A__ : Any = DPMSolverMultistepScheduler.from_config(pipe.scheduler.config ) A__ : int = pipe.to("""cuda""" ) A__ : Optional[Any] = """Spiderman is surfing""" A__ : List[str] = torch.Generator(device="""cpu""" ).manual_seed(0 ) A__ : Optional[Any] = pipe(A__ , generator=A__ , num_inference_steps=25 , output_type="""pt""" ).frames A__ : Dict = video_frames.cpu().numpy() assert np.abs(expected_video - video ).mean() < 5e-2 def __A ( self ): A__ : List[Any] = load_numpy( """https://huggingface.co/datasets/hf-internal-testing/diffusers-images/resolve/main/text_to_video/video_2step.npy""" ) A__ : Optional[int] = TextToVideoSDPipeline.from_pretrained("""damo-vilab/text-to-video-ms-1.7b""" ) A__ : List[str] = pipe.to("""cuda""" ) A__ : Dict = """Spiderman is surfing""" A__ : Union[str, Any] = torch.Generator(device="""cpu""" ).manual_seed(0 ) A__ : Optional[int] = pipe(A__ , generator=A__ , num_inference_steps=2 , output_type="""pt""" ).frames A__ : Optional[int] = video_frames.cpu().numpy() assert np.abs(expected_video - video ).mean() < 5e-2
64
1
import argparse import json from pathlib import Path import requests import timm import torch from huggingface_hub import hf_hub_download from PIL import Image from timm.data import resolve_data_config from timm.data.transforms_factory import create_transform from transformers import ( BitConfig, ViTHybridConfig, ViTHybridForImageClassification, ViTHybridImageProcessor, ViTHybridModel, ) from transformers.image_utils import PILImageResampling from transformers.utils import logging logging.set_verbosity_info() A_ : Tuple = logging.get_logger(__name__) def UpperCamelCase (lowercase_: Dict , lowercase_: Tuple=False ) -> str: A__ : Optional[int] = [] # fmt: off # stem: rename_keys.append(("""cls_token""", """vit.embeddings.cls_token""") ) rename_keys.append(("""pos_embed""", """vit.embeddings.position_embeddings""") ) rename_keys.append(("""patch_embed.proj.weight""", """vit.embeddings.patch_embeddings.projection.weight""") ) rename_keys.append(("""patch_embed.proj.bias""", """vit.embeddings.patch_embeddings.projection.bias""") ) # backbone rename_keys.append(("""patch_embed.backbone.stem.conv.weight""", """vit.embeddings.patch_embeddings.backbone.bit.embedder.convolution.weight""") ) rename_keys.append(("""patch_embed.backbone.stem.norm.weight""", """vit.embeddings.patch_embeddings.backbone.bit.embedder.norm.weight""") ) rename_keys.append(("""patch_embed.backbone.stem.norm.bias""", """vit.embeddings.patch_embeddings.backbone.bit.embedder.norm.bias""") ) for stage_idx in range(len(config.backbone_config.depths ) ): for layer_idx in range(config.backbone_config.depths[stage_idx] ): rename_keys.append((f"""patch_embed.backbone.stages.{stage_idx}.blocks.{layer_idx}.conv1.weight""", f"""vit.embeddings.patch_embeddings.backbone.bit.encoder.stages.{stage_idx}.layers.{layer_idx}.conv1.weight""") ) rename_keys.append((f"""patch_embed.backbone.stages.{stage_idx}.blocks.{layer_idx}.norm1.weight""", f"""vit.embeddings.patch_embeddings.backbone.bit.encoder.stages.{stage_idx}.layers.{layer_idx}.norm1.weight""") ) rename_keys.append((f"""patch_embed.backbone.stages.{stage_idx}.blocks.{layer_idx}.norm1.bias""", f"""vit.embeddings.patch_embeddings.backbone.bit.encoder.stages.{stage_idx}.layers.{layer_idx}.norm1.bias""") ) rename_keys.append((f"""patch_embed.backbone.stages.{stage_idx}.blocks.{layer_idx}.conv2.weight""", f"""vit.embeddings.patch_embeddings.backbone.bit.encoder.stages.{stage_idx}.layers.{layer_idx}.conv2.weight""") ) rename_keys.append((f"""patch_embed.backbone.stages.{stage_idx}.blocks.{layer_idx}.norm2.weight""", f"""vit.embeddings.patch_embeddings.backbone.bit.encoder.stages.{stage_idx}.layers.{layer_idx}.norm2.weight""") ) rename_keys.append((f"""patch_embed.backbone.stages.{stage_idx}.blocks.{layer_idx}.norm2.bias""", f"""vit.embeddings.patch_embeddings.backbone.bit.encoder.stages.{stage_idx}.layers.{layer_idx}.norm2.bias""") ) rename_keys.append((f"""patch_embed.backbone.stages.{stage_idx}.blocks.{layer_idx}.conv3.weight""", f"""vit.embeddings.patch_embeddings.backbone.bit.encoder.stages.{stage_idx}.layers.{layer_idx}.conv3.weight""") ) rename_keys.append((f"""patch_embed.backbone.stages.{stage_idx}.blocks.{layer_idx}.norm3.weight""", f"""vit.embeddings.patch_embeddings.backbone.bit.encoder.stages.{stage_idx}.layers.{layer_idx}.norm3.weight""") ) rename_keys.append((f"""patch_embed.backbone.stages.{stage_idx}.blocks.{layer_idx}.norm3.bias""", f"""vit.embeddings.patch_embeddings.backbone.bit.encoder.stages.{stage_idx}.layers.{layer_idx}.norm3.bias""") ) rename_keys.append((f"""patch_embed.backbone.stages.{stage_idx}.blocks.0.downsample.conv.weight""", f"""vit.embeddings.patch_embeddings.backbone.bit.encoder.stages.{stage_idx}.layers.0.downsample.conv.weight""") ) rename_keys.append((f"""patch_embed.backbone.stages.{stage_idx}.blocks.0.downsample.norm.weight""", f"""vit.embeddings.patch_embeddings.backbone.bit.encoder.stages.{stage_idx}.layers.0.downsample.norm.weight""") ) rename_keys.append((f"""patch_embed.backbone.stages.{stage_idx}.blocks.0.downsample.norm.bias""", f"""vit.embeddings.patch_embeddings.backbone.bit.encoder.stages.{stage_idx}.layers.0.downsample.norm.bias""") ) # transformer encoder for i in range(config.num_hidden_layers ): # encoder layers: output projection, 2 feedforward neural networks and 2 layernorms rename_keys.append((f"""blocks.{i}.norm1.weight""", f"""vit.encoder.layer.{i}.layernorm_before.weight""") ) rename_keys.append((f"""blocks.{i}.norm1.bias""", f"""vit.encoder.layer.{i}.layernorm_before.bias""") ) rename_keys.append((f"""blocks.{i}.attn.proj.weight""", f"""vit.encoder.layer.{i}.attention.output.dense.weight""") ) rename_keys.append((f"""blocks.{i}.attn.proj.bias""", f"""vit.encoder.layer.{i}.attention.output.dense.bias""") ) rename_keys.append((f"""blocks.{i}.norm2.weight""", f"""vit.encoder.layer.{i}.layernorm_after.weight""") ) rename_keys.append((f"""blocks.{i}.norm2.bias""", f"""vit.encoder.layer.{i}.layernorm_after.bias""") ) rename_keys.append((f"""blocks.{i}.mlp.fc1.weight""", f"""vit.encoder.layer.{i}.intermediate.dense.weight""") ) rename_keys.append((f"""blocks.{i}.mlp.fc1.bias""", f"""vit.encoder.layer.{i}.intermediate.dense.bias""") ) rename_keys.append((f"""blocks.{i}.mlp.fc2.weight""", f"""vit.encoder.layer.{i}.output.dense.weight""") ) rename_keys.append((f"""blocks.{i}.mlp.fc2.bias""", f"""vit.encoder.layer.{i}.output.dense.bias""") ) if base_model: # layernorm + pooler rename_keys.extend( [ ("""norm.weight""", """layernorm.weight"""), ("""norm.bias""", """layernorm.bias"""), ("""pre_logits.fc.weight""", """pooler.dense.weight"""), ("""pre_logits.fc.bias""", """pooler.dense.bias"""), ] ) # if just the base model, we should remove "vit" from all keys that start with "vit" A__ : Tuple = [(pair[0], pair[1][4:]) if pair[1].startswith("""vit""" ) else pair for pair in rename_keys] else: # layernorm + classification head rename_keys.extend( [ ("""norm.weight""", """vit.layernorm.weight"""), ("""norm.bias""", """vit.layernorm.bias"""), ("""head.weight""", """classifier.weight"""), ("""head.bias""", """classifier.bias"""), ] ) # fmt: on return rename_keys def UpperCamelCase (lowercase_: Dict , lowercase_: Tuple , lowercase_: Optional[int]=False ) -> str: for i in range(config.num_hidden_layers ): if base_model: A__ : Optional[int] = """""" else: A__ : List[str] = """vit.""" # read in weights + bias of input projection layer (in timm, this is a single matrix + bias) A__ : Tuple = state_dict.pop(f"""blocks.{i}.attn.qkv.weight""" ) A__ : Tuple = state_dict.pop(f"""blocks.{i}.attn.qkv.bias""" ) # next, add query, keys and values (in that order) to the state dict A__ : str = in_proj_weight[ : config.hidden_size, : ] A__ : Any = in_proj_bias[: config.hidden_size] A__ : str = in_proj_weight[ config.hidden_size : config.hidden_size * 2, : ] A__ : Any = in_proj_bias[ config.hidden_size : config.hidden_size * 2 ] A__ : str = in_proj_weight[ -config.hidden_size :, : ] A__ : Tuple = in_proj_bias[-config.hidden_size :] def UpperCamelCase (lowercase_: List[Any] ) -> Optional[Any]: A__ : int = ["""head.weight""", """head.bias"""] for k in ignore_keys: state_dict.pop(lowercase_ , lowercase_ ) def UpperCamelCase (lowercase_: int , lowercase_: Any , lowercase_: str ) -> Dict: A__ : Optional[Any] = dct.pop(lowercase_ ) A__ : Tuple = val def UpperCamelCase () -> Tuple: A__ : List[str] = """http://images.cocodataset.org/val2017/000000039769.jpg""" A__ : str = Image.open(requests.get(lowercase_ , stream=lowercase_ ).raw ) return im @torch.no_grad() def UpperCamelCase (lowercase_: Union[str, Any] , lowercase_: Dict , lowercase_: str=False ) -> str: A__ : Dict = BitConfig( global_padding="""same""" , layer_type="""bottleneck""" , depths=(3, 4, 9) , out_features=["""stage3"""] , embedding_dynamic_padding=lowercase_ , ) A__ : List[Any] = ViTHybridConfig(backbone_config=lowercase_ , image_size=384 , num_labels=1000 ) A__ : Any = False # load original model from timm A__ : List[Any] = timm.create_model(lowercase_ , pretrained=lowercase_ ) timm_model.eval() # load state_dict of original model, remove and rename some keys A__ : Union[str, Any] = timm_model.state_dict() if base_model: remove_classification_head_(lowercase_ ) A__ : Any = create_rename_keys(lowercase_ , lowercase_ ) for src, dest in rename_keys: rename_key(lowercase_ , lowercase_ , lowercase_ ) read_in_q_k_v(lowercase_ , lowercase_ , lowercase_ ) A__ : Any = """huggingface/label-files""" A__ : Optional[int] = """imagenet-1k-id2label.json""" A__ : List[str] = json.load(open(hf_hub_download(lowercase_ , lowercase_ , repo_type="""dataset""" ) , """r""" ) ) A__ : Union[str, Any] = {int(lowercase_ ): v for k, v in idalabel.items()} A__ : Dict = idalabel A__ : Tuple = {v: k for k, v in idalabel.items()} # load HuggingFace model if vit_name[-5:] == "in21k": A__ : Union[str, Any] = ViTHybridModel(lowercase_ ).eval() else: A__ : Dict = ViTHybridForImageClassification(lowercase_ ).eval() model.load_state_dict(lowercase_ ) # create image processor A__ : List[Any] = create_transform(**resolve_data_config({} , model=lowercase_ ) ) A__ : int = transform.transforms A__ : str = { """bilinear""": PILImageResampling.BILINEAR, """bicubic""": PILImageResampling.BICUBIC, """nearest""": PILImageResampling.NEAREST, } A__ : Tuple = ViTHybridImageProcessor( do_resize=lowercase_ , size={"""shortest_edge""": timm_transforms[0].size} , resample=pillow_resamplings[timm_transforms[0].interpolation.value] , do_center_crop=lowercase_ , crop_size={"""height""": timm_transforms[1].size[0], """width""": timm_transforms[1].size[1]} , do_normalize=lowercase_ , image_mean=timm_transforms[-1].mean.tolist() , image_std=timm_transforms[-1].std.tolist() , ) A__ : Any = prepare_img() A__ : List[str] = transform(lowercase_ ).unsqueeze(0 ) A__ : List[Any] = processor(lowercase_ , return_tensors="""pt""" ).pixel_values # verify pixel values assert torch.allclose(lowercase_ , lowercase_ ) # verify logits with torch.no_grad(): A__ : Optional[int] = model(lowercase_ ) A__ : str = outputs.logits print("""Predicted class:""" , logits.argmax(-1 ).item() ) if base_model: A__ : Any = timm_model.forward_features(lowercase_ ) assert timm_pooled_output.shape == outputs.pooler_output.shape assert torch.allclose(lowercase_ , outputs.pooler_output , atol=1E-3 ) else: A__ : Optional[int] = timm_model(lowercase_ ) assert timm_logits.shape == outputs.logits.shape assert torch.allclose(lowercase_ , outputs.logits , atol=1E-3 ) print("""Looks ok!""" ) if pytorch_dump_folder_path is not None: Path(lowercase_ ).mkdir(exist_ok=lowercase_ ) print(f"""Saving model {vit_name} to {pytorch_dump_folder_path}""" ) model.save_pretrained(lowercase_ ) print(f"""Saving processor to {pytorch_dump_folder_path}""" ) processor.save_pretrained(lowercase_ ) if push_to_hub: print(f"""Pushing model and processor to the hub {vit_name}""" ) model.push_to_hub(f"""ybelkada/{vit_name}""" ) processor.push_to_hub(f"""ybelkada/{vit_name}""" ) if __name__ == "__main__": A_ : Tuple = argparse.ArgumentParser() # Required parameters parser.add_argument( '--vit_name', default='vit_base_r50_s16_384', type=str, help='Name of the hybrid ViT timm model you\'d like to convert.', ) parser.add_argument( '--pytorch_dump_folder_path', default=None, type=str, help='Path to the output PyTorch model directory.' ) parser.add_argument( '--push_to_hub', action='store_true', help='Whether to upload the model to the HuggingFace hub.' ) A_ : int = parser.parse_args() convert_vit_checkpoint(args.vit_name, args.pytorch_dump_folder_path, args.push_to_hub)
64
def UpperCamelCase (lowercase_: int ) -> int: if not isinstance(lowercase_ , lowercase_ ): raise TypeError("""Input value must be an 'int' type""" ) A__ : int = 0 while number: position += 1 number >>= 1 return position if __name__ == "__main__": import doctest doctest.testmod()
64
1
import math import numpy as np import qiskit from qiskit import Aer, ClassicalRegister, QuantumCircuit, QuantumRegister, execute def UpperCamelCase (lowercase_: int = 3 ) -> qiskit.result.counts.Counts: if isinstance(lowercase_ , lowercase_ ): raise TypeError("""number of qubits must be a integer.""" ) if number_of_qubits <= 0: raise ValueError("""number of qubits must be > 0.""" ) if math.floor(lowercase_ ) != number_of_qubits: raise ValueError("""number of qubits must be exact integer.""" ) if number_of_qubits > 10: raise ValueError("""number of qubits too large to simulate(>10).""" ) A__ : Any = QuantumRegister(lowercase_ , """qr""" ) A__ : Any = ClassicalRegister(lowercase_ , """cr""" ) A__ : Optional[Any] = QuantumCircuit(lowercase_ , lowercase_ ) A__ : Optional[Any] = number_of_qubits for i in range(lowercase_ ): quantum_circuit.h(number_of_qubits - i - 1 ) counter -= 1 for j in range(lowercase_ ): quantum_circuit.cp(np.pi / 2 ** (counter - j) , lowercase_ , lowercase_ ) for k in range(number_of_qubits // 2 ): quantum_circuit.swap(lowercase_ , number_of_qubits - k - 1 ) # measure all the qubits quantum_circuit.measure(lowercase_ , lowercase_ ) # simulate with 10000 shots A__ : int = Aer.get_backend("""qasm_simulator""" ) A__ : List[Any] = execute(lowercase_ , lowercase_ , shots=10000 ) return job.result().get_counts(lowercase_ ) if __name__ == "__main__": print( f'''Total count for quantum fourier transform state is: \ {quantum_fourier_transform(3)}''' )
64
from pathlib import Path import cva import numpy as np from matplotlib import pyplot as plt def UpperCamelCase (lowercase_: np.ndarray , lowercase_: np.ndarray , lowercase_: np.ndarray , lowercase_: int , lowercase_: int ) -> np.ndarray: A__ : Any = cva.getAffineTransform(lowercase_ , lowercase_ ) return cva.warpAffine(lowercase_ , lowercase_ , (rows, cols) ) if __name__ == "__main__": # read original image A_ : List[Any] = cva.imread( str(Path(__file__).resolve().parent.parent / 'image_data' / 'lena.jpg') ) # turn image in gray scale value A_ : List[Any] = cva.cvtColor(image, cva.COLOR_BGR2GRAY) # get image shape A_ , A_ : Optional[Any] = gray_img.shape # set different points to rotate image A_ : str = np.array([[50, 50], [200, 50], [50, 200]], np.floataa) A_ : Dict = np.array([[10, 100], [200, 50], [100, 250]], np.floataa) A_ : Optional[int] = np.array([[50, 50], [150, 50], [120, 200]], np.floataa) A_ : Optional[int] = np.array([[10, 100], [80, 50], [180, 250]], np.floataa) # add all rotated images in a list A_ : Dict = [ gray_img, get_rotation(gray_img, ptsa, ptsa, img_rows, img_cols), get_rotation(gray_img, ptsa, ptsa, img_rows, img_cols), get_rotation(gray_img, ptsa, ptsa, img_rows, img_cols), ] # plot different image rotations A_ : Union[str, Any] = plt.figure(1) A_ : Union[str, Any] = ['Original', 'Rotation 1', 'Rotation 2', 'Rotation 3'] for i, image in enumerate(images): plt.subplot(2, 2, i + 1), plt.imshow(image, 'gray') plt.title(titles[i]) plt.axis('off') plt.subplots_adjust(left=0.0, bottom=0.05, right=1.0, top=0.95) plt.show()
64
1
from __future__ import annotations from collections.abc import Callable A_ : List[Any] = list[list[float | int]] def UpperCamelCase (lowercase_: Matrix , lowercase_: Matrix ) -> Matrix: A__ : int = len(lowercase_ ) A__ : Matrix = [[0 for _ in range(size + 1 )] for _ in range(lowercase_ )] A__ : int A__ : int A__ : int A__ : int A__ : int A__ : float for row in range(lowercase_ ): for col in range(lowercase_ ): A__ : List[str] = matrix[row][col] A__ : int = vector[row][0] A__ : Optional[int] = 0 A__ : str = 0 while row < size and col < size: # pivoting A__ : int = max((abs(augmented[rowa][col] ), rowa) for rowa in range(lowercase_ , lowercase_ ) )[ 1 ] if augmented[pivot_row][col] == 0: col += 1 continue else: A__ , A__ : Union[str, Any] = augmented[pivot_row], augmented[row] for rowa in range(row + 1 , lowercase_ ): A__ : List[Any] = augmented[rowa][col] / augmented[row][col] A__ : Dict = 0 for cola in range(col + 1 , size + 1 ): augmented[rowa][cola] -= augmented[row][cola] * ratio row += 1 col += 1 # back substitution for col in range(1 , lowercase_ ): for row in range(lowercase_ ): A__ : List[str] = augmented[row][col] / augmented[col][col] for cola in range(lowercase_ , size + 1 ): augmented[row][cola] -= augmented[col][cola] * ratio # round to get rid of numbers like 2.000000000000004 return [ [round(augmented[row][size] / augmented[row][row] , 10 )] for row in range(lowercase_ ) ] def UpperCamelCase (lowercase_: list[int] ) -> Callable[[int], int]: A__ : int = len(lowercase_ ) A__ : Matrix = [[0 for _ in range(lowercase_ )] for _ in range(lowercase_ )] A__ : Matrix = [[0] for _ in range(lowercase_ )] A__ : Matrix A__ : int A__ : int A__ : int for x_val, y_val in enumerate(lowercase_ ): for col in range(lowercase_ ): A__ : Dict = (x_val + 1) ** (size - col - 1) A__ : Any = y_val A__ : Union[str, Any] = solve(lowercase_ , lowercase_ ) def interpolated_func(lowercase_: int ) -> int: return sum( round(coeffs[x_val][0] ) * (var ** (size - x_val - 1)) for x_val in range(lowercase_ ) ) return interpolated_func def UpperCamelCase (lowercase_: int ) -> int: return ( 1 - variable + variable**2 - variable**3 + variable**4 - variable**5 + variable**6 - variable**7 + variable**8 - variable**9 + variable**10 ) def UpperCamelCase (lowercase_: Callable[[int], int] = question_function , lowercase_: int = 10 ) -> int: A__ : list[int] = [func(lowercase_ ) for x_val in range(1 , order + 1 )] A__ : list[Callable[[int], int]] = [ interpolate(data_points[:max_coeff] ) for max_coeff in range(1 , order + 1 ) ] A__ : int = 0 A__ : Callable[[int], int] A__ : int for poly in polynomials: A__ : List[str] = 1 while func(lowercase_ ) == poly(lowercase_ ): x_val += 1 ret += poly(lowercase_ ) return ret if __name__ == "__main__": print(f'''{solution() = }''')
64
import os import tempfile import unittest from pathlib import Path from transformers import AutoConfig, is_torch_available from transformers.testing_utils import require_torch, torch_device if is_torch_available(): from transformers import PyTorchBenchmark, PyTorchBenchmarkArguments @require_torch class _a (unittest.TestCase ): '''simple docstring''' def __A ( self , A__ ): for model_result in results.values(): for batch_size, sequence_length in zip(model_result["""bs"""] , model_result["""ss"""] ): A__ : str = model_result["""result"""][batch_size][sequence_length] self.assertIsNotNone(A__ ) def __A ( self ): A__ : Dict = """sshleifer/tiny-gpt2""" A__ : Tuple = PyTorchBenchmarkArguments( models=[MODEL_ID] , training=A__ , inference=A__ , sequence_lengths=[8] , batch_sizes=[1] , multi_process=A__ , ) A__ : int = PyTorchBenchmark(A__ ) A__ : List[Any] = benchmark.run() self.check_results_dict_not_empty(results.time_inference_result ) self.check_results_dict_not_empty(results.memory_inference_result ) def __A ( self ): A__ : Dict = """sgugger/tiny-distilbert-classification""" A__ : Dict = PyTorchBenchmarkArguments( models=[MODEL_ID] , training=A__ , inference=A__ , sequence_lengths=[8] , batch_sizes=[1] , multi_process=A__ , only_pretrain_model=A__ , ) A__ : str = PyTorchBenchmark(A__ ) A__ : List[str] = benchmark.run() self.check_results_dict_not_empty(results.time_inference_result ) self.check_results_dict_not_empty(results.memory_inference_result ) def __A ( self ): A__ : Any = """sshleifer/tiny-gpt2""" A__ : List[Any] = PyTorchBenchmarkArguments( models=[MODEL_ID] , training=A__ , inference=A__ , torchscript=A__ , sequence_lengths=[8] , batch_sizes=[1] , multi_process=A__ , ) A__ : Tuple = PyTorchBenchmark(A__ ) A__ : str = benchmark.run() self.check_results_dict_not_empty(results.time_inference_result ) self.check_results_dict_not_empty(results.memory_inference_result ) @unittest.skipIf(torch_device == """cpu""" , """Cant do half precision""" ) def __A ( self ): A__ : Optional[Any] = """sshleifer/tiny-gpt2""" A__ : Optional[Any] = PyTorchBenchmarkArguments( models=[MODEL_ID] , training=A__ , inference=A__ , fpaa=A__ , sequence_lengths=[8] , batch_sizes=[1] , multi_process=A__ , ) A__ : str = PyTorchBenchmark(A__ ) A__ : Any = benchmark.run() self.check_results_dict_not_empty(results.time_inference_result ) self.check_results_dict_not_empty(results.memory_inference_result ) def __A ( self ): A__ : Optional[Any] = """sshleifer/tiny-gpt2""" A__ : Tuple = AutoConfig.from_pretrained(A__ ) # set architectures equal to `None` A__ : List[Any] = None A__ : str = PyTorchBenchmarkArguments( models=[MODEL_ID] , training=A__ , inference=A__ , sequence_lengths=[8] , batch_sizes=[1] , multi_process=A__ , ) A__ : List[str] = PyTorchBenchmark(A__ , configs=[config] ) A__ : Any = benchmark.run() self.check_results_dict_not_empty(results.time_inference_result ) self.check_results_dict_not_empty(results.memory_inference_result ) def __A ( self ): A__ : Optional[int] = """sshleifer/tiny-gpt2""" A__ : Optional[int] = PyTorchBenchmarkArguments( models=[MODEL_ID] , training=A__ , inference=A__ , sequence_lengths=[8] , batch_sizes=[1] , multi_process=A__ , ) A__ : Any = PyTorchBenchmark(A__ ) A__ : Dict = benchmark.run() self.check_results_dict_not_empty(results.time_train_result ) self.check_results_dict_not_empty(results.memory_train_result ) @unittest.skipIf(torch_device == """cpu""" , """Can't do half precision""" ) def __A ( self ): A__ : Optional[int] = """sshleifer/tiny-gpt2""" A__ : List[str] = PyTorchBenchmarkArguments( models=[MODEL_ID] , training=A__ , inference=A__ , sequence_lengths=[8] , batch_sizes=[1] , fpaa=A__ , multi_process=A__ , ) A__ : Dict = PyTorchBenchmark(A__ ) A__ : Optional[int] = benchmark.run() self.check_results_dict_not_empty(results.time_train_result ) self.check_results_dict_not_empty(results.memory_train_result ) def __A ( self ): A__ : int = """sshleifer/tiny-gpt2""" A__ : Optional[int] = AutoConfig.from_pretrained(A__ ) A__ : str = PyTorchBenchmarkArguments( models=[MODEL_ID] , training=A__ , inference=A__ , sequence_lengths=[8] , batch_sizes=[1] , multi_process=A__ , ) A__ : int = PyTorchBenchmark(A__ , configs=[config] ) A__ : Tuple = benchmark.run() self.check_results_dict_not_empty(results.time_inference_result ) self.check_results_dict_not_empty(results.memory_inference_result ) def __A ( self ): A__ : List[str] = """sshleifer/tinier_bart""" A__ : List[str] = AutoConfig.from_pretrained(A__ ) A__ : List[str] = PyTorchBenchmarkArguments( models=[MODEL_ID] , training=A__ , inference=A__ , sequence_lengths=[8] , batch_sizes=[1] , multi_process=A__ , ) A__ : Union[str, Any] = PyTorchBenchmark(A__ , configs=[config] ) A__ : str = benchmark.run() self.check_results_dict_not_empty(results.time_inference_result ) self.check_results_dict_not_empty(results.memory_inference_result ) def __A ( self ): A__ : Optional[int] = """sshleifer/tiny-gpt2""" A__ : Union[str, Any] = AutoConfig.from_pretrained(A__ ) A__ : Tuple = PyTorchBenchmarkArguments( models=[MODEL_ID] , training=A__ , inference=A__ , sequence_lengths=[8] , batch_sizes=[1] , multi_process=A__ , ) A__ : int = PyTorchBenchmark(A__ , configs=[config] ) A__ : Tuple = benchmark.run() self.check_results_dict_not_empty(results.time_train_result ) self.check_results_dict_not_empty(results.memory_train_result ) def __A ( self ): A__ : Dict = """sshleifer/tinier_bart""" A__ : int = AutoConfig.from_pretrained(A__ ) A__ : Union[str, Any] = PyTorchBenchmarkArguments( models=[MODEL_ID] , training=A__ , inference=A__ , sequence_lengths=[8] , batch_sizes=[1] , multi_process=A__ , ) A__ : List[Any] = PyTorchBenchmark(A__ , configs=[config] ) A__ : Optional[Any] = benchmark.run() self.check_results_dict_not_empty(results.time_train_result ) self.check_results_dict_not_empty(results.memory_train_result ) def __A ( self ): A__ : int = """sshleifer/tiny-gpt2""" with tempfile.TemporaryDirectory() as tmp_dir: A__ : Dict = PyTorchBenchmarkArguments( models=[MODEL_ID] , training=A__ , inference=A__ , save_to_csv=A__ , sequence_lengths=[8] , batch_sizes=[1] , inference_time_csv_file=os.path.join(A__ , """inf_time.csv""" ) , train_memory_csv_file=os.path.join(A__ , """train_mem.csv""" ) , inference_memory_csv_file=os.path.join(A__ , """inf_mem.csv""" ) , train_time_csv_file=os.path.join(A__ , """train_time.csv""" ) , env_info_csv_file=os.path.join(A__ , """env.csv""" ) , multi_process=A__ , ) A__ : Optional[Any] = PyTorchBenchmark(A__ ) benchmark.run() self.assertTrue(Path(os.path.join(A__ , """inf_time.csv""" ) ).exists() ) self.assertTrue(Path(os.path.join(A__ , """train_time.csv""" ) ).exists() ) self.assertTrue(Path(os.path.join(A__ , """inf_mem.csv""" ) ).exists() ) self.assertTrue(Path(os.path.join(A__ , """train_mem.csv""" ) ).exists() ) self.assertTrue(Path(os.path.join(A__ , """env.csv""" ) ).exists() ) def __A ( self ): A__ : Optional[int] = """sshleifer/tiny-gpt2""" def _check_summary_is_not_empty(A__ ): self.assertTrue(hasattr(A__ , """sequential""" ) ) self.assertTrue(hasattr(A__ , """cumulative""" ) ) self.assertTrue(hasattr(A__ , """current""" ) ) self.assertTrue(hasattr(A__ , """total""" ) ) with tempfile.TemporaryDirectory() as tmp_dir: A__ : Dict = PyTorchBenchmarkArguments( models=[MODEL_ID] , training=A__ , inference=A__ , sequence_lengths=[8] , batch_sizes=[1] , log_filename=os.path.join(A__ , """log.txt""" ) , log_print=A__ , trace_memory_line_by_line=A__ , multi_process=A__ , ) A__ : Dict = PyTorchBenchmark(A__ ) A__ : str = benchmark.run() _check_summary_is_not_empty(result.inference_summary ) _check_summary_is_not_empty(result.train_summary ) self.assertTrue(Path(os.path.join(A__ , """log.txt""" ) ).exists() )
64
1
from ...configuration_utils import PretrainedConfig from ...utils import logging A_ : Union[str, Any] = logging.get_logger(__name__) A_ : Optional[int] = { 'sayakpaul/vit-msn-base': 'https://huggingface.co/sayakpaul/vit-msn-base/resolve/main/config.json', # See all ViT MSN models at https://huggingface.co/models?filter=vit_msn } class _a (__magic_name__ ): '''simple docstring''' UpperCAmelCase__: Optional[int] = '''vit_msn''' def __init__( self , A__=768 , A__=12 , A__=12 , A__=3072 , A__="gelu" , A__=0.0 , A__=0.0 , A__=0.0_2 , A__=1e-06 , A__=224 , A__=16 , A__=3 , A__=True , **A__ , ): super().__init__(**A__ ) A__ : Optional[int] = hidden_size A__ : List[str] = num_hidden_layers A__ : Any = num_attention_heads A__ : Optional[Any] = intermediate_size A__ : str = hidden_act A__ : Union[str, Any] = hidden_dropout_prob A__ : Optional[int] = attention_probs_dropout_prob A__ : Tuple = initializer_range A__ : Optional[Any] = layer_norm_eps A__ : Optional[Any] = image_size A__ : List[str] = patch_size A__ : Tuple = num_channels A__ : str = qkv_bias
64
# tests directory-specific settings - this file is run automatically # by pytest before any tests are run import doctest import sys import warnings from os.path import abspath, dirname, join import _pytest from transformers.testing_utils import HfDoctestModule, HfDocTestParser # allow having multiple repository checkouts and not needing to remember to rerun # 'pip install -e .[dev]' when switching between checkouts and running tests. A_ : Optional[int] = abspath(join(dirname(__file__), 'src')) sys.path.insert(1, git_repo_path) # silence FutureWarning warnings in tests since often we can't act on them until # they become normal warnings - i.e. the tests still need to test the current functionality warnings.simplefilter(action='ignore', category=FutureWarning) def UpperCamelCase (lowercase_: List[str] ) -> Any: config.addinivalue_line( """markers""" , """is_pt_tf_cross_test: mark test to run only when PT and TF interactions are tested""" ) config.addinivalue_line( """markers""" , """is_pt_flax_cross_test: mark test to run only when PT and FLAX interactions are tested""" ) config.addinivalue_line("""markers""" , """is_pipeline_test: mark test to run only when pipelines are tested""" ) config.addinivalue_line("""markers""" , """is_staging_test: mark test to run only in the staging environment""" ) config.addinivalue_line("""markers""" , """accelerate_tests: mark test that require accelerate""" ) config.addinivalue_line("""markers""" , """tool_tests: mark the tool tests that are run on their specific schedule""" ) def UpperCamelCase (lowercase_: Optional[int] ) -> Optional[Any]: from transformers.testing_utils import pytest_addoption_shared pytest_addoption_shared(lowercase_ ) def UpperCamelCase (lowercase_: List[str] ) -> Optional[Any]: from transformers.testing_utils import pytest_terminal_summary_main A__ : List[Any] = terminalreporter.config.getoption("""--make-reports""" ) if make_reports: pytest_terminal_summary_main(lowercase_ , id=lowercase_ ) def UpperCamelCase (lowercase_: Union[str, Any] , lowercase_: int ) -> List[str]: # If no tests are collected, pytest exists with code 5, which makes the CI fail. if exitstatus == 5: A__ : Tuple = 0 # Doctest custom flag to ignore output. A_ : Tuple = doctest.register_optionflag('IGNORE_RESULT') A_ : Dict = doctest.OutputChecker class _a (__magic_name__ ): '''simple docstring''' def __A ( self , A__ , A__ , A__ ): if IGNORE_RESULT & optionflags: return True return OutputChecker.check_output(self , A__ , A__ , A__ ) A_ : str = CustomOutputChecker A_ : Dict = HfDoctestModule A_ : Optional[int] = HfDocTestParser
64
1
import json import os import unittest from transformers.models.ctrl.tokenization_ctrl import VOCAB_FILES_NAMES, CTRLTokenizer from ...test_tokenization_common import TokenizerTesterMixin class _a (__magic_name__ , unittest.TestCase ): '''simple docstring''' UpperCAmelCase__: int = CTRLTokenizer UpperCAmelCase__: int = False UpperCAmelCase__: Optional[Any] = False def __A ( self ): super().setUp() # Adapted from Sennrich et al. 2015 and https://github.com/rsennrich/subword-nmt A__ : List[Any] = ["""adapt""", """re@@""", """a@@""", """apt""", """c@@""", """t""", """<unk>"""] A__ : Union[str, Any] = dict(zip(A__ , range(len(A__ ) ) ) ) A__ : Optional[int] = ["""#version: 0.2""", """a p""", """ap t</w>""", """r e""", """a d""", """ad apt</w>""", """"""] A__ : str = {"""unk_token""": """<unk>"""} A__ : str = os.path.join(self.tmpdirname , VOCAB_FILES_NAMES["""vocab_file"""] ) A__ : Dict = os.path.join(self.tmpdirname , VOCAB_FILES_NAMES["""merges_file"""] ) with open(self.vocab_file , """w""" , encoding="""utf-8""" ) as fp: fp.write(json.dumps(A__ ) + """\n""" ) with open(self.merges_file , """w""" , encoding="""utf-8""" ) as fp: fp.write("""\n""".join(A__ ) ) def __A ( self , **A__ ): kwargs.update(self.special_tokens_map ) return CTRLTokenizer.from_pretrained(self.tmpdirname , **A__ ) def __A ( self , A__ ): A__ : Optional[int] = """adapt react readapt apt""" A__ : Tuple = """adapt react readapt apt""" return input_text, output_text def __A ( self ): A__ : Any = CTRLTokenizer(self.vocab_file , self.merges_file , **self.special_tokens_map ) A__ : str = """adapt react readapt apt""" A__ : Optional[int] = """adapt re@@ a@@ c@@ t re@@ adapt apt""".split() A__ : List[str] = tokenizer.tokenize(A__ ) self.assertListEqual(A__ , A__ ) A__ : List[str] = tokens + [tokenizer.unk_token] A__ : Any = [0, 1, 2, 4, 5, 1, 0, 3, 6] self.assertListEqual(tokenizer.convert_tokens_to_ids(A__ ) , A__ )
64
from __future__ import annotations import unittest from transformers import AutoTokenizer, PegasusConfig, is_tf_available from transformers.testing_utils import require_sentencepiece, require_tf, require_tokenizers, slow from transformers.utils import cached_property from ...test_configuration_common import ConfigTester from ...test_modeling_tf_common import TFModelTesterMixin, ids_tensor from ...test_pipeline_mixin import PipelineTesterMixin if is_tf_available(): import tensorflow as tf from transformers import TFAutoModelForSeqaSeqLM, TFPegasusForConditionalGeneration, TFPegasusModel @require_tf class _a : '''simple docstring''' UpperCAmelCase__: List[Any] = PegasusConfig UpperCAmelCase__: Optional[int] = {} UpperCAmelCase__: List[str] = '''gelu''' def __init__( self , A__ , A__=13 , A__=7 , A__=True , A__=False , A__=99 , A__=32 , A__=2 , A__=4 , A__=37 , A__=0.1 , A__=0.1 , A__=40 , A__=2 , A__=1 , A__=0 , ): A__ : Dict = parent A__ : Dict = batch_size A__ : Any = seq_length A__ : Optional[Any] = is_training A__ : int = use_labels A__ : Any = vocab_size A__ : Union[str, Any] = hidden_size A__ : Tuple = num_hidden_layers A__ : Tuple = num_attention_heads A__ : List[Any] = intermediate_size A__ : Union[str, Any] = hidden_dropout_prob A__ : Optional[Any] = attention_probs_dropout_prob A__ : List[Any] = max_position_embeddings A__ : Any = eos_token_id A__ : List[Any] = pad_token_id A__ : List[Any] = bos_token_id def __A ( self ): A__ : str = ids_tensor([self.batch_size, self.seq_length - 1] , self.vocab_size ) A__ : Dict = tf.expand_dims(tf.constant([self.eos_token_id] * self.batch_size ) , 1 ) A__ : List[Any] = tf.concat([input_ids, eos_tensor] , axis=1 ) A__ : Union[str, Any] = ids_tensor([self.batch_size, self.seq_length] , self.vocab_size ) A__ : Tuple = self.config_cls( vocab_size=self.vocab_size , d_model=self.hidden_size , encoder_layers=self.num_hidden_layers , decoder_layers=self.num_hidden_layers , encoder_attention_heads=self.num_attention_heads , decoder_attention_heads=self.num_attention_heads , encoder_ffn_dim=self.intermediate_size , decoder_ffn_dim=self.intermediate_size , dropout=self.hidden_dropout_prob , attention_dropout=self.attention_probs_dropout_prob , max_position_embeddings=self.max_position_embeddings , eos_token_ids=[2] , bos_token_id=self.bos_token_id , pad_token_id=self.pad_token_id , decoder_start_token_id=self.pad_token_id , **self.config_updates , ) A__ : str = prepare_pegasus_inputs_dict(A__ , A__ , A__ ) return config, inputs_dict def __A ( self , A__ , A__ ): A__ : int = TFPegasusModel(config=A__ ).get_decoder() A__ : List[Any] = inputs_dict["""input_ids"""] A__ : Any = input_ids[:1, :] A__ : Optional[Any] = inputs_dict["""attention_mask"""][:1, :] A__ : Optional[int] = inputs_dict["""head_mask"""] A__ : Any = 1 # first forward pass A__ : Tuple = model(A__ , attention_mask=A__ , head_mask=A__ , use_cache=A__ ) A__ , A__ : Dict = outputs.to_tuple() # create hypothetical next token and extent to next_input_ids A__ : Tuple = ids_tensor((self.batch_size, 3) , config.vocab_size ) A__ : Optional[Any] = tf.cast(ids_tensor((self.batch_size, 3) , 2 ) , tf.inta ) # append to next input_ids and A__ : List[Any] = tf.concat([input_ids, next_tokens] , axis=-1 ) A__ : Tuple = tf.concat([attention_mask, next_attn_mask] , axis=-1 ) A__ : Optional[Any] = model(A__ , attention_mask=A__ )[0] A__ : Any = model(A__ , attention_mask=A__ , past_key_values=A__ )[0] self.parent.assertEqual(next_tokens.shape[1] , output_from_past.shape[1] ) # select random slice A__ : int = int(ids_tensor((1,) , output_from_past.shape[-1] ) ) A__ : Any = output_from_no_past[:, -3:, random_slice_idx] A__ : Tuple = output_from_past[:, :, random_slice_idx] # test that outputs are equal for slice tf.debugging.assert_near(A__ , A__ , rtol=1e-3 ) def UpperCamelCase (lowercase_: Union[str, Any] , lowercase_: Dict , lowercase_: List[Any] , lowercase_: Dict=None , lowercase_: int=None , lowercase_: List[Any]=None , lowercase_: List[Any]=None , lowercase_: str=None , ) -> int: if attention_mask is None: A__ : List[str] = tf.cast(tf.math.not_equal(lowercase_ , config.pad_token_id ) , tf.inta ) if decoder_attention_mask is None: A__ : Dict = tf.concat( [ tf.ones(decoder_input_ids[:, :1].shape , dtype=tf.inta ), tf.cast(tf.math.not_equal(decoder_input_ids[:, 1:] , config.pad_token_id ) , tf.inta ), ] , axis=-1 , ) if head_mask is None: A__ : Any = tf.ones((config.encoder_layers, config.encoder_attention_heads) ) if decoder_head_mask is None: A__ : Tuple = tf.ones((config.decoder_layers, config.decoder_attention_heads) ) if cross_attn_head_mask is None: A__ : List[str] = tf.ones((config.decoder_layers, config.decoder_attention_heads) ) return { "input_ids": input_ids, "decoder_input_ids": decoder_input_ids, "attention_mask": attention_mask, "decoder_attention_mask": decoder_attention_mask, "head_mask": head_mask, "decoder_head_mask": decoder_head_mask, "cross_attn_head_mask": cross_attn_head_mask, } @require_tf class _a (__magic_name__ , __magic_name__ , unittest.TestCase ): '''simple docstring''' UpperCAmelCase__: List[Any] = (TFPegasusForConditionalGeneration, TFPegasusModel) if is_tf_available() else () UpperCAmelCase__: Tuple = (TFPegasusForConditionalGeneration,) if is_tf_available() else () UpperCAmelCase__: Tuple = ( { '''conversational''': TFPegasusForConditionalGeneration, '''feature-extraction''': TFPegasusModel, '''summarization''': TFPegasusForConditionalGeneration, '''text2text-generation''': TFPegasusForConditionalGeneration, '''translation''': TFPegasusForConditionalGeneration, } if is_tf_available() else {} ) UpperCAmelCase__: int = True UpperCAmelCase__: Union[str, Any] = False UpperCAmelCase__: List[str] = False def __A ( self ): A__ : Optional[Any] = TFPegasusModelTester(self ) A__ : Tuple = ConfigTester(self , config_class=A__ ) def __A ( self ): self.config_tester.run_common_tests() def __A ( self ): A__ : int = self.model_tester.prepare_config_and_inputs_for_common() self.model_tester.check_decoder_model_past_large_inputs(*A__ ) @require_sentencepiece @require_tokenizers @require_tf class _a (unittest.TestCase ): '''simple docstring''' UpperCAmelCase__: Optional[int] = [ ''' PG&E stated it scheduled the blackouts in response to forecasts for high winds amid dry conditions. The aim is to reduce the risk of wildfires. Nearly 800 thousand customers were scheduled to be affected by the shutoffs which were expected to last through at least midday tomorrow.''', ''' The London trio are up for best UK act and best album, as well as getting two nominations in the best song category."We got told like this morning \'Oh I think you\'re nominated\'", said Dappy."And I was like \'Oh yeah, which one?\' And now we\'ve got nominated for four awards. I mean, wow!"Bandmate Fazer added: "We thought it\'s best of us to come down and mingle with everyone and say hello to the cameras. And now we find we\'ve got four nominations."The band have two shots at the best song prize, getting the nod for their Tynchy Stryder collaboration Number One, and single Strong Again.Their album Uncle B will also go up against records by the likes of Beyonce and Kanye West.N-Dubz picked up the best newcomer Mobo in 2007, but female member Tulisa said they wouldn\'t be too disappointed if they didn\'t win this time around."At the end of the day we\'re grateful to be where we are in our careers."If it don\'t happen then it don\'t happen - live to fight another day and keep on making albums and hits for the fans."Dappy also revealed they could be performing live several times on the night.The group will be doing Number One and also a possible rendition of the War Child single, I Got Soul.The charity song is a re-working of The Killers\' All These Things That I\'ve Done and is set to feature artists like Chipmunk, Ironik and Pixie Lott.This year\'s Mobos will be held outside of London for the first time, in Glasgow on 30 September.N-Dubz said they were looking forward to performing for their Scottish fans and boasted about their recent shows north of the border."We just done Edinburgh the other day," said Dappy."We smashed up an N-Dubz show over there. We done Aberdeen about three or four months ago - we smashed up that show over there! Everywhere we go we smash it up!" ''', ] UpperCAmelCase__: Any = [ '''California\'s largest electricity provider has cut power to hundreds of thousands of customers in an effort to''' ''' reduce the risk of wildfires.''', '''N-Dubz have revealed they\'re "grateful" to have been nominated for four Mobo Awards.''', ] # differs slightly from pytorch, likely due to numerical differences in linear layers UpperCAmelCase__: List[str] = '''google/pegasus-xsum''' @cached_property def __A ( self ): return AutoTokenizer.from_pretrained(self.model_name ) @cached_property def __A ( self ): A__ : int = TFAutoModelForSeqaSeqLM.from_pretrained(self.model_name ) return model def __A ( self , **A__ ): A__ : str = self.translate_src_text(**A__ ) assert self.expected_text == generated_words def __A ( self , **A__ ): A__ : List[str] = self.tokenizer(self.src_text , **A__ , padding=A__ , return_tensors="""tf""" ) A__ : Optional[int] = self.model.generate( model_inputs.input_ids , attention_mask=model_inputs.attention_mask , num_beams=2 , use_cache=A__ , ) A__ : Dict = self.tokenizer.batch_decode(generated_ids.numpy() , skip_special_tokens=A__ ) return generated_words @slow def __A ( self ): self._assert_generated_batch_equal_expected()
64
1
import tempfile import torch from diffusers import ( DEISMultistepScheduler, DPMSolverMultistepScheduler, DPMSolverSinglestepScheduler, UniPCMultistepScheduler, ) from .test_schedulers import SchedulerCommonTest class _a (__magic_name__ ): '''simple docstring''' UpperCAmelCase__: Optional[int] = (DPMSolverSinglestepScheduler,) UpperCAmelCase__: Optional[int] = (('''num_inference_steps''', 25),) def __A ( self , **A__ ): A__ : Optional[Any] = { """num_train_timesteps""": 1000, """beta_start""": 0.0_0_0_1, """beta_end""": 0.0_2, """beta_schedule""": """linear""", """solver_order""": 2, """prediction_type""": """epsilon""", """thresholding""": False, """sample_max_value""": 1.0, """algorithm_type""": """dpmsolver++""", """solver_type""": """midpoint""", """lambda_min_clipped""": -float("""inf""" ), """variance_type""": None, } config.update(**A__ ) return config def __A ( self , A__=0 , **A__ ): A__ : Tuple = dict(self.forward_default_kwargs ) A__ : str = kwargs.pop("""num_inference_steps""" , A__ ) A__ : Optional[int] = self.dummy_sample A__ : Any = 0.1 * sample A__ : List[Any] = [residual + 0.2, residual + 0.1_5, residual + 0.1_0] for scheduler_class in self.scheduler_classes: A__ : List[str] = self.get_scheduler_config(**A__ ) A__ : Tuple = scheduler_class(**A__ ) scheduler.set_timesteps(A__ ) # copy over dummy past residuals A__ : List[Any] = dummy_past_residuals[: scheduler.config.solver_order] with tempfile.TemporaryDirectory() as tmpdirname: scheduler.save_config(A__ ) A__ : Optional[int] = scheduler_class.from_pretrained(A__ ) new_scheduler.set_timesteps(A__ ) # copy over dummy past residuals A__ : Union[str, Any] = dummy_past_residuals[: new_scheduler.config.solver_order] A__ , A__ : Optional[int] = sample, sample for t in range(A__ , time_step + scheduler.config.solver_order + 1 ): A__ : int = scheduler.step(A__ , A__ , A__ , **A__ ).prev_sample A__ : List[str] = new_scheduler.step(A__ , A__ , A__ , **A__ ).prev_sample assert torch.sum(torch.abs(output - new_output ) ) < 1e-5, "Scheduler outputs are not identical" def __A ( self ): pass def __A ( self , A__=0 , **A__ ): A__ : Any = dict(self.forward_default_kwargs ) A__ : List[str] = kwargs.pop("""num_inference_steps""" , A__ ) A__ : Union[str, Any] = self.dummy_sample A__ : int = 0.1 * sample A__ : Optional[Any] = [residual + 0.2, residual + 0.1_5, residual + 0.1_0] for scheduler_class in self.scheduler_classes: A__ : Optional[int] = self.get_scheduler_config() A__ : Tuple = scheduler_class(**A__ ) scheduler.set_timesteps(A__ ) # copy over dummy past residuals (must be after setting timesteps) A__ : Union[str, Any] = dummy_past_residuals[: scheduler.config.solver_order] with tempfile.TemporaryDirectory() as tmpdirname: scheduler.save_config(A__ ) A__ : Any = scheduler_class.from_pretrained(A__ ) # copy over dummy past residuals new_scheduler.set_timesteps(A__ ) # copy over dummy past residual (must be after setting timesteps) A__ : Any = dummy_past_residuals[: new_scheduler.config.solver_order] A__ : List[Any] = scheduler.step(A__ , A__ , A__ , **A__ ).prev_sample A__ : List[Any] = new_scheduler.step(A__ , A__ , A__ , **A__ ).prev_sample assert torch.sum(torch.abs(output - new_output ) ) < 1e-5, "Scheduler outputs are not identical" def __A ( self , A__=None , **A__ ): if scheduler is None: A__ : int = self.scheduler_classes[0] A__ : List[Any] = self.get_scheduler_config(**A__ ) A__ : Optional[Any] = scheduler_class(**A__ ) A__ : Union[str, Any] = self.scheduler_classes[0] A__ : Optional[Any] = self.get_scheduler_config(**A__ ) A__ : List[str] = scheduler_class(**A__ ) A__ : Union[str, Any] = 10 A__ : Tuple = self.dummy_model() A__ : Dict = self.dummy_sample_deter scheduler.set_timesteps(A__ ) for i, t in enumerate(scheduler.timesteps ): A__ : Union[str, Any] = model(A__ , A__ ) A__ : Dict = scheduler.step(A__ , A__ , A__ ).prev_sample return sample def __A ( self ): A__ : Union[str, Any] = DPMSolverSinglestepScheduler(**self.get_scheduler_config() ) A__ : int = 50 A__ : Dict = self.dummy_model() A__ : Tuple = self.dummy_sample_deter scheduler.set_timesteps(A__ ) # make sure that the first t is uneven for i, t in enumerate(scheduler.timesteps[3:] ): A__ : int = model(A__ , A__ ) A__ : List[Any] = scheduler.step(A__ , A__ , A__ ).prev_sample A__ : str = torch.mean(torch.abs(A__ ) ) assert abs(result_mean.item() - 0.2_5_7_4 ) < 1e-3 def __A ( self ): for timesteps in [25, 50, 100, 999, 1000]: self.check_over_configs(num_train_timesteps=A__ ) def __A ( self ): # make sure that iterating over schedulers with same config names gives same results # for defaults A__ : Dict = DPMSolverSinglestepScheduler(**self.get_scheduler_config() ) A__ : Tuple = self.full_loop(scheduler=A__ ) A__ : Optional[int] = torch.mean(torch.abs(A__ ) ) assert abs(result_mean.item() - 0.2_7_9_1 ) < 1e-3 A__ : Optional[int] = DEISMultistepScheduler.from_config(scheduler.config ) A__ : Optional[Any] = DPMSolverMultistepScheduler.from_config(scheduler.config ) A__ : List[Any] = UniPCMultistepScheduler.from_config(scheduler.config ) A__ : int = DPMSolverSinglestepScheduler.from_config(scheduler.config ) A__ : Any = self.full_loop(scheduler=A__ ) A__ : str = torch.mean(torch.abs(A__ ) ) assert abs(result_mean.item() - 0.2_7_9_1 ) < 1e-3 def __A ( self ): self.check_over_configs(thresholding=A__ ) for order in [1, 2, 3]: for solver_type in ["midpoint", "heun"]: for threshold in [0.5, 1.0, 2.0]: for prediction_type in ["epsilon", "sample"]: self.check_over_configs( thresholding=A__ , prediction_type=A__ , sample_max_value=A__ , algorithm_type="""dpmsolver++""" , solver_order=A__ , solver_type=A__ , ) def __A ( self ): for prediction_type in ["epsilon", "v_prediction"]: self.check_over_configs(prediction_type=A__ ) def __A ( self ): for algorithm_type in ["dpmsolver", "dpmsolver++"]: for solver_type in ["midpoint", "heun"]: for order in [1, 2, 3]: for prediction_type in ["epsilon", "sample"]: self.check_over_configs( solver_order=A__ , solver_type=A__ , prediction_type=A__ , algorithm_type=A__ , ) A__ : Optional[Any] = self.full_loop( solver_order=A__ , solver_type=A__ , prediction_type=A__ , algorithm_type=A__ , ) assert not torch.isnan(A__ ).any(), "Samples have nan numbers" def __A ( self ): self.check_over_configs(lower_order_final=A__ ) self.check_over_configs(lower_order_final=A__ ) def __A ( self ): self.check_over_configs(lambda_min_clipped=-float("""inf""" ) ) self.check_over_configs(lambda_min_clipped=-5.1 ) def __A ( self ): self.check_over_configs(variance_type=A__ ) self.check_over_configs(variance_type="""learned_range""" ) def __A ( self ): for num_inference_steps in [1, 2, 3, 5, 10, 50, 100, 999, 1000]: self.check_over_forward(num_inference_steps=A__ , time_step=0 ) def __A ( self ): A__ : str = self.full_loop() A__ : Dict = torch.mean(torch.abs(A__ ) ) assert abs(result_mean.item() - 0.2_7_9_1 ) < 1e-3 def __A ( self ): A__ : List[str] = self.full_loop(use_karras_sigmas=A__ ) A__ : List[Any] = torch.mean(torch.abs(A__ ) ) assert abs(result_mean.item() - 0.2_2_4_8 ) < 1e-3 def __A ( self ): A__ : Tuple = self.full_loop(prediction_type="""v_prediction""" ) A__ : Any = torch.mean(torch.abs(A__ ) ) assert abs(result_mean.item() - 0.1_4_5_3 ) < 1e-3 def __A ( self ): A__ : List[str] = self.full_loop(prediction_type="""v_prediction""" , use_karras_sigmas=A__ ) A__ : str = torch.mean(torch.abs(A__ ) ) assert abs(result_mean.item() - 0.0_6_4_9 ) < 1e-3 def __A ( self ): A__ : Any = self.scheduler_classes[0] A__ : Union[str, Any] = self.get_scheduler_config(thresholding=A__ , dynamic_thresholding_ratio=0 ) A__ : str = scheduler_class(**A__ ) A__ : Dict = 10 A__ : Union[str, Any] = self.dummy_model() A__ : Optional[int] = self.dummy_sample_deter.half() scheduler.set_timesteps(A__ ) for i, t in enumerate(scheduler.timesteps ): A__ : str = model(A__ , A__ ) A__ : Optional[Any] = scheduler.step(A__ , A__ , A__ ).prev_sample assert sample.dtype == torch.floataa
64
class _a : '''simple docstring''' def __init__( self ): A__ : str = """""" A__ : Any = """""" A__ : List[Any] = [] def __A ( self , A__ , A__ ): if m == -1: return n + 1 elif n == -1: return m + 1 elif self.dp[m][n] > -1: return self.dp[m][n] else: if self.worda[m] == self.worda[n]: A__ : Optional[Any] = self.__min_dist_top_down_dp(m - 1 , n - 1 ) else: A__ : Union[str, Any] = self.__min_dist_top_down_dp(A__ , n - 1 ) A__ : Union[str, Any] = self.__min_dist_top_down_dp(m - 1 , A__ ) A__ : Union[str, Any] = self.__min_dist_top_down_dp(m - 1 , n - 1 ) A__ : List[Any] = 1 + min(A__ , A__ , A__ ) return self.dp[m][n] def __A ( self , A__ , A__ ): A__ : Tuple = worda A__ : Dict = worda A__ : Optional[Any] = [[-1 for _ in range(len(A__ ) )] for _ in range(len(A__ ) )] return self.__min_dist_top_down_dp(len(A__ ) - 1 , len(A__ ) - 1 ) def __A ( self , A__ , A__ ): A__ : Optional[Any] = worda A__ : Dict = worda A__ : Union[str, Any] = len(A__ ) A__ : List[str] = len(A__ ) A__ : int = [[0 for _ in range(n + 1 )] for _ in range(m + 1 )] for i in range(m + 1 ): for j in range(n + 1 ): if i == 0: # first string is empty A__ : Tuple = j elif j == 0: # second string is empty A__ : Dict = i elif worda[i - 1] == worda[j - 1]: # last characters are equal A__ : str = self.dp[i - 1][j - 1] else: A__ : Union[str, Any] = self.dp[i][j - 1] A__ : str = self.dp[i - 1][j] A__ : Union[str, Any] = self.dp[i - 1][j - 1] A__ : Tuple = 1 + min(A__ , A__ , A__ ) return self.dp[m][n] if __name__ == "__main__": A_ : Union[str, Any] = EditDistance() print('****************** Testing Edit Distance DP Algorithm ******************') print() A_ : int = input('Enter the first string: ').strip() A_ : List[str] = input('Enter the second string: ').strip() print() print(f'''The minimum edit distance is: {solver.min_dist_top_down(Sa, Sa)}''') print(f'''The minimum edit distance is: {solver.min_dist_bottom_up(Sa, Sa)}''') print() print('*************** End of Testing Edit Distance DP Algorithm ***************')
64
1
import argparse import gdown import numpy as np import torch from huggingface_hub import hf_hub_download from transformers import ( CLIPTokenizer, CLIPTokenizerFast, VideoMAEImageProcessor, XCLIPConfig, XCLIPModel, XCLIPProcessor, XCLIPTextConfig, XCLIPVisionConfig, ) def UpperCamelCase (lowercase_: List[Any] , lowercase_: Union[str, Any] ) -> Union[str, Any]: A__ : Optional[Any] = XCLIPTextConfig() # derive patch size from model name A__ : Any = model_name.find("""patch""" ) A__ : List[str] = int(model_name[start_idx + len("""patch""" ) : start_idx + len("""patch""" ) + 2] ) A__ : List[Any] = XCLIPVisionConfig(patch_size=lowercase_ , num_frames=lowercase_ ) if "large" in model_name: A__ : str = 768 A__ : str = 3072 A__ : Any = 12 A__ : Tuple = 1024 A__ : str = 4096 A__ : Any = 16 A__ : List[Any] = 24 A__ : Optional[Any] = 768 A__ : Any = 3072 if model_name == "xclip-large-patch14-16-frames": A__ : List[Any] = 336 A__ : List[str] = XCLIPConfig.from_text_vision_configs(lowercase_ , lowercase_ ) if "large" in model_name: A__ : Dict = 768 return config def UpperCamelCase (lowercase_: Dict ) -> Dict: # text encoder if name == "token_embedding.weight": A__ : Dict = name.replace("""token_embedding.weight""" , """text_model.embeddings.token_embedding.weight""" ) if name == "positional_embedding": A__ : Union[str, Any] = name.replace("""positional_embedding""" , """text_model.embeddings.position_embedding.weight""" ) if "ln_1" in name: A__ : Optional[Any] = name.replace("""ln_1""" , """layer_norm1""" ) if "ln_2" in name: A__ : Optional[int] = name.replace("""ln_2""" , """layer_norm2""" ) if "c_fc" in name: A__ : Optional[Any] = name.replace("""c_fc""" , """fc1""" ) if "c_proj" in name: A__ : str = name.replace("""c_proj""" , """fc2""" ) if name.startswith("""transformer.resblocks""" ): A__ : Dict = name.replace("""transformer.resblocks""" , """text_model.encoder.layers""" ) if "attn.out_proj" in name and "message" not in name: A__ : Optional[Any] = name.replace("""attn.out_proj""" , """self_attn.out_proj""" ) if "ln_final" in name: A__ : List[Any] = name.replace("""ln_final""" , """text_model.final_layer_norm""" ) # visual encoder if name == "visual.class_embedding": A__ : List[str] = name.replace("""visual.class_embedding""" , """vision_model.embeddings.class_embedding""" ) if name == "visual.positional_embedding": A__ : Union[str, Any] = name.replace("""visual.positional_embedding""" , """vision_model.embeddings.position_embedding.weight""" ) if name.startswith("""visual.transformer.resblocks""" ): A__ : Any = name.replace("""visual.transformer.resblocks""" , """vision_model.encoder.layers""" ) if "visual.conv1" in name: A__ : str = name.replace("""visual.conv1""" , """vision_model.embeddings.patch_embedding""" ) if "visual.ln_pre" in name: A__ : List[str] = name.replace("""visual.ln_pre""" , """vision_model.pre_layernorm""" ) if "visual.ln_post" in name: A__ : str = name.replace("""visual.ln_post""" , """vision_model.post_layernorm""" ) if "visual.proj" in name: A__ : Dict = name.replace("""visual.proj""" , """visual_projection.weight""" ) if "text_projection" in name: A__ : Any = name.replace("""text_projection""" , """text_projection.weight""" ) # things on top if "prompts_visual_proj" in name: A__ : str = name.replace("""prompts_visual_proj""" , """prompts_visual_projection""" ) if "prompts_visual_ln" in name: A__ : List[Any] = name.replace("""prompts_visual_ln""" , """prompts_visual_layernorm""" ) # mit if name == "mit.positional_embedding": A__ : List[str] = name.replace("""positional""" , """position""" ) if name.startswith("""mit.resblocks""" ): A__ : Union[str, Any] = name.replace("""mit.resblocks""" , """mit.encoder.layers""" ) # prompts generator if name.startswith("""prompts_generator.norm""" ): A__ : List[Any] = name.replace("""prompts_generator.norm""" , """prompts_generator.layernorm""" ) return name def UpperCamelCase (lowercase_: str , lowercase_: int ) -> Optional[Any]: for key in orig_state_dict.copy().keys(): A__ : Dict = orig_state_dict.pop(lowercase_ ) if "attn.in_proj" in key: A__ : Union[str, Any] = key.split(""".""" ) if key.startswith("""visual""" ): A__ : Any = key_split[3] A__ : Optional[int] = config.vision_config.hidden_size if "message_attn" in key: if "weight" in key: A__ : List[Any] = val[ :dim, : ] A__ : Optional[Any] = val[ dim : dim * 2, : ] A__ : Tuple = val[ -dim:, : ] else: A__ : List[str] = val[ :dim ] A__ : Union[str, Any] = val[ dim : dim * 2 ] A__ : List[Any] = val[ -dim: ] else: if "weight" in key: A__ : List[Any] = val[ :dim, : ] A__ : Union[str, Any] = val[ dim : dim * 2, : ] A__ : List[str] = val[ -dim:, : ] else: A__ : Optional[Any] = val[:dim] A__ : Dict = val[ dim : dim * 2 ] A__ : str = val[-dim:] elif key.startswith("""mit""" ): A__ : Optional[Any] = key_split[2] A__ : str = config.vision_config.mit_hidden_size if "weight" in key: A__ : Dict = val[:dim, :] A__ : int = val[dim : dim * 2, :] A__ : List[str] = val[-dim:, :] else: A__ : List[str] = val[:dim] A__ : List[Any] = val[dim : dim * 2] A__ : Optional[Any] = val[-dim:] else: A__ : Optional[int] = key_split[2] A__ : Dict = config.text_config.hidden_size if "weight" in key: A__ : int = val[:dim, :] A__ : Any = val[ dim : dim * 2, : ] A__ : List[str] = val[-dim:, :] else: A__ : Any = val[:dim] A__ : Union[str, Any] = val[ dim : dim * 2 ] A__ : str = val[-dim:] else: A__ : List[Any] = rename_key(lowercase_ ) if new_key_name in ["visual_projection.weight", "text_projection.weight"]: A__ : List[Any] = val.T A__ : List[Any] = val return orig_state_dict def UpperCamelCase (lowercase_: Optional[int] ) -> List[str]: if num_frames == 8: A__ : Tuple = """eating_spaghetti_8_frames.npy""" elif num_frames == 16: A__ : Tuple = """eating_spaghetti.npy""" elif num_frames == 32: A__ : Optional[int] = """eating_spaghetti_32_frames.npy""" A__ : Any = hf_hub_download( repo_id="""hf-internal-testing/spaghetti-video""" , filename=lowercase_ , repo_type="""dataset""" , ) A__ : str = np.load(lowercase_ ) return list(lowercase_ ) def UpperCamelCase (lowercase_: Union[str, Any] , lowercase_: Dict=None , lowercase_: Optional[Any]=False ) -> Union[str, Any]: A__ : int = { # fully supervised kinetics-400 checkpoints """xclip-base-patch32""": """https://github.com/nbl97/X-CLIP_Model_Zoo/releases/download/v1.0/k400_32_8.pth""", """xclip-base-patch32-16-frames""": ( """https://github.com/nbl97/X-CLIP_Model_Zoo/releases/download/v1.0/k400_32_16.pth""" ), """xclip-base-patch16""": """https://github.com/nbl97/X-CLIP_Model_Zoo/releases/download/v1.0/k400_16_8.pth""", """xclip-base-patch16-16-frames""": ( """https://github.com/nbl97/X-CLIP_Model_Zoo/releases/download/v1.0/k400_16_16.pth""" ), """xclip-large-patch14""": """https://drive.google.com/u/0/uc?id=1NUOImq0o5DlQTST17iIP3vG7DgmHQuCx&amp;export=download&amp;confirm=t&amp;uuid=b26caedc-88e2-473e-830a-9d158b653cdb""", """xclip-large-patch14-16-frames""": """https://drive.google.com/u/0/uc?id=1FOYgnJc097OJ4lGwtRCCydQyVPJEOH7d&amp;export=download&amp;confirm=t&amp;uuid=538fa810-e671-4050-b385-9a623f89804f""", # fully supervised kinetics-600 checkpoints """xclip-base-patch16-kinetics-600""": ( """https://github.com/nbl97/X-CLIP_Model_Zoo/releases/download/v1.0/k600_16_8.pth""" ), """xclip-base-patch16-kinetics-600-16-frames""": ( """https://github.com/nbl97/X-CLIP_Model_Zoo/releases/download/v1.0/k600_16_16.pth""" ), """xclip-large-patch14-kinetics-600""": """https://drive.google.com/u/0/uc?id=1FV8C1INuM91sLAN4ImjzePLIlpMSihwV&amp;export=download&amp;confirm=t&amp;uuid=141d4977-4a65-44ae-864f-4b0c19f838be""", # few shot """xclip-base-patch16-hmdb-2-shot""": ( """https://github.com/nbl97/X-CLIP_Model_Zoo/releases/download/v1.0/few_hmdb_2.pth""" ), """xclip-base-patch16-hmdb-4-shot""": ( """https://github.com/nbl97/X-CLIP_Model_Zoo/releases/download/v1.0/few_hmdb_4.pth""" ), """xclip-base-patch16-hmdb-8-shot""": ( """https://github.com/nbl97/X-CLIP_Model_Zoo/releases/download/v1.0/few_hmdb_8.pth""" ), """xclip-base-patch16-hmdb-16-shot""": ( """https://github.com/nbl97/X-CLIP_Model_Zoo/releases/download/v1.0/few_hmdb_16.pth""" ), """xclip-base-patch16-ucf-2-shot""": ( """https://github.com/nbl97/X-CLIP_Model_Zoo/releases/download/v1.0/few_ucf_2.pth""" ), """xclip-base-patch16-ucf-4-shot""": ( """https://github.com/nbl97/X-CLIP_Model_Zoo/releases/download/v1.0/few_ucf_4.pth""" ), """xclip-base-patch16-ucf-8-shot""": ( """https://github.com/nbl97/X-CLIP_Model_Zoo/releases/download/v1.0/few_ucf_8.pth""" ), """xclip-base-patch16-ucf-16-shot""": ( """https://github.com/nbl97/X-CLIP_Model_Zoo/releases/download/v1.0/few_ucf_16.pth""" ), # zero shot """xclip-base-patch16-zero-shot""": """https://github.com/nbl97/X-CLIP_Model_Zoo/releases/download/v1.0/zero.pth""", } A__ : List[Any] = model_to_url[model_name] A__ : List[str] = 8 if "16-frames" in model_name: A__ : Any = 16 elif "shot" in model_name: A__ : Dict = 32 A__ : Any = get_xclip_config(lowercase_ , lowercase_ ) A__ : str = XCLIPModel(lowercase_ ) model.eval() if "drive" in checkpoint_url: A__ : Any = """pytorch_model.bin""" gdown.cached_download(lowercase_ , lowercase_ , quiet=lowercase_ ) A__ : int = torch.load(lowercase_ , map_location="""cpu""" )["""model"""] else: A__ : Dict = torch.hub.load_state_dict_from_url(lowercase_ )["""model"""] A__ : List[Any] = convert_state_dict(lowercase_ , lowercase_ ) A__ : Tuple = XCLIPModel(lowercase_ ) A__ , A__ : List[Any] = model.load_state_dict(lowercase_ , strict=lowercase_ ) assert missing_keys == ["text_model.embeddings.position_ids", "vision_model.embeddings.position_ids"] model.eval() A__ : List[Any] = 336 if model_name == """xclip-large-patch14-16-frames""" else 224 A__ : Optional[int] = VideoMAEImageProcessor(size=lowercase_ ) A__ : Union[str, Any] = CLIPTokenizer.from_pretrained("""openai/clip-vit-base-patch32""" ) A__ : Tuple = CLIPTokenizerFast.from_pretrained("""openai/clip-vit-base-patch32""" ) A__ : Dict = XCLIPProcessor(image_processor=lowercase_ , tokenizer=lowercase_ ) A__ : str = prepare_video(lowercase_ ) A__ : Tuple = processor( text=["""playing sports""", """eating spaghetti""", """go shopping"""] , videos=lowercase_ , return_tensors="""pt""" , padding=lowercase_ ) print("""Shape of pixel values:""" , inputs.pixel_values.shape ) with torch.no_grad(): A__ : List[str] = model(**lowercase_ ) # Verify outputs A__ : int = outputs.logits_per_video A__ : str = logits_per_video.softmax(dim=1 ) print("""Probs:""" , lowercase_ ) # kinetics-400 if model_name == "xclip-base-patch32": A__ : Optional[Any] = torch.tensor([[0.0019, 0.9951, 0.0030]] ) elif model_name == "xclip-base-patch32-16-frames": A__ : Union[str, Any] = torch.tensor([[7.09_99E-04, 9.98_83E-01, 4.55_80E-04]] ) elif model_name == "xclip-base-patch16": A__ : Tuple = torch.tensor([[0.0083, 0.9681, 0.0236]] ) elif model_name == "xclip-base-patch16-16-frames": A__ : Union[str, Any] = torch.tensor([[7.69_37E-04, 9.97_28E-01, 1.94_73E-03]] ) elif model_name == "xclip-large-patch14": A__ : Any = torch.tensor([[0.0062, 0.9864, 0.0075]] ) elif model_name == "xclip-large-patch14-16-frames": A__ : List[Any] = torch.tensor([[3.38_77E-04, 9.99_37E-01, 2.88_88E-04]] ) # kinetics-600 elif model_name == "xclip-base-patch16-kinetics-600": A__ : Tuple = torch.tensor([[0.0555, 0.8914, 0.0531]] ) elif model_name == "xclip-base-patch16-kinetics-600-16-frames": A__ : Tuple = torch.tensor([[3.85_54E-04, 9.99_29E-01, 3.27_54E-04]] ) elif model_name == "xclip-large-patch14-kinetics-600": A__ : List[Any] = torch.tensor([[0.0036, 0.9920, 0.0045]] ) # few shot elif model_name == "xclip-base-patch16-hmdb-2-shot": A__ : List[Any] = torch.tensor([[7.18_90E-06, 9.99_94E-01, 5.65_59E-05]] ) elif model_name == "xclip-base-patch16-hmdb-4-shot": A__ : Optional[int] = torch.tensor([[1.03_20E-05, 9.99_93E-01, 6.24_35E-05]] ) elif model_name == "xclip-base-patch16-hmdb-8-shot": A__ : Union[str, Any] = torch.tensor([[4.13_77E-06, 9.99_90E-01, 9.83_86E-05]] ) elif model_name == "xclip-base-patch16-hmdb-16-shot": A__ : Any = torch.tensor([[4.13_47E-05, 9.99_62E-01, 3.34_11E-04]] ) elif model_name == "xclip-base-patch16-ucf-2-shot": A__ : Any = torch.tensor([[8.58_57E-05, 9.99_28E-01, 6.32_91E-04]] ) elif model_name == "xclip-base-patch16-ucf-4-shot": A__ : Optional[int] = torch.tensor([[8.58_57E-05, 9.99_28E-01, 6.32_91E-04]] ) elif model_name == "xclip-base-patch16-ucf-8-shot": A__ : Union[str, Any] = torch.tensor([[0.0027, 0.9904, 0.0070]] ) elif model_name == "xclip-base-patch16-ucf-16-shot": A__ : List[Any] = torch.tensor([[9.82_19E-04, 9.95_93E-01, 3.08_63E-03]] ) # zero shot elif model_name == "xclip-base-patch16-zero-shot": A__ : Union[str, Any] = torch.tensor([[3.50_82E-04, 9.97_85E-01, 1.79_66E-03]] ) else: raise ValueError(f"""Model name {model_name} not supported""" ) assert torch.allclose(lowercase_ , lowercase_ , atol=1E-3 ) print("""Looks ok!""" ) if pytorch_dump_folder_path is not None: print(f"""Saving model {model_name} to {pytorch_dump_folder_path}""" ) model.save_pretrained(lowercase_ ) if push_to_hub: print("""Pushing model, processor and slow tokenizer files to the hub...""" ) model.push_to_hub(lowercase_ , organization="""nielsr""" ) processor.push_to_hub(lowercase_ , organization="""nielsr""" ) slow_tokenizer.push_to_hub(lowercase_ , organization="""nielsr""" ) if __name__ == "__main__": A_ : List[str] = argparse.ArgumentParser() # Required parameters parser.add_argument( '--model_name', default='xclip-base-patch32', type=str, help='Name of the model.', ) parser.add_argument( '--pytorch_dump_folder_path', default=None, type=str, help='Path to the output PyTorch model directory.' ) parser.add_argument( '--push_to_hub', action='store_true', help='Whether or not to push the converted model to the 🤗 hub.' ) A_ : str = parser.parse_args() convert_xclip_checkpoint(args.model_name, args.pytorch_dump_folder_path, args.push_to_hub)
64
def UpperCamelCase (lowercase_: int , lowercase_: int ) -> int: while second != 0: A__ : int = first & second first ^= second A__ : int = c << 1 return first if __name__ == "__main__": import doctest doctest.testmod() A_ : Optional[Any] = int(input('Enter the first number: ').strip()) A_ : List[str] = int(input('Enter the second number: ').strip()) print(f'''{add(first, second) = }''')
64
1
import time from contextlib import contextmanager from pathlib import Path import pytest import requests from huggingface_hub.hf_api import HfApi, HfFolder A_ : Any = '__DUMMY_TRANSFORMERS_USER__' A_ : Tuple = 'Dummy User' A_ : Any = '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' A_ : str = 'https://hub-ci.huggingface.co' A_ : Any = CI_HUB_ENDPOINT + '/datasets/{repo_id}/resolve/{revision}/{path}' A_ : List[str] = CI_HUB_ENDPOINT + '/{repo_id}/resolve/{revision}/{filename}' A_ : Any = Path('~/.huggingface/hub_ci_token').expanduser() @pytest.fixture def UpperCamelCase (lowercase_: Union[str, Any] ) -> List[str]: monkeypatch.setattr( """huggingface_hub.file_download.HUGGINGFACE_CO_URL_TEMPLATE""" , lowercase_ ) @pytest.fixture def UpperCamelCase (lowercase_: Union[str, Any] ) -> List[str]: monkeypatch.setattr("""datasets.config.HF_ENDPOINT""" , lowercase_ ) monkeypatch.setattr("""datasets.config.HUB_DATASETS_URL""" , lowercase_ ) @pytest.fixture def UpperCamelCase (lowercase_: Dict ) -> List[str]: monkeypatch.setattr("""huggingface_hub.hf_api.HfFolder.path_token""" , lowercase_ ) @pytest.fixture def UpperCamelCase (lowercase_: Dict , lowercase_: str ) -> Tuple: HfFolder.save_token(lowercase_ ) yield HfFolder.delete_token() @pytest.fixture(scope="""session""" ) def UpperCamelCase () -> str: return HfApi(endpoint=lowercase_ ) @pytest.fixture(scope="""session""" ) def UpperCamelCase (lowercase_: HfApi ) -> Optional[int]: A__ : Optional[int] = HfFolder.get_token() HfFolder.save_token(lowercase_ ) yield CI_HUB_USER_TOKEN if previous_token is not None: HfFolder.save_token(lowercase_ ) @pytest.fixture def UpperCamelCase (lowercase_: Any ) -> int: def _cleanup_repo(lowercase_: List[str] ): hf_api.delete_repo(lowercase_ , token=lowercase_ , repo_type="""dataset""" ) return _cleanup_repo @pytest.fixture def UpperCamelCase (lowercase_: str ) -> Dict: @contextmanager def _temporary_repo(lowercase_: Tuple ): try: yield repo_id finally: cleanup_repo(lowercase_ ) return _temporary_repo @pytest.fixture(scope="""session""" ) def UpperCamelCase (lowercase_: HfApi , lowercase_: Optional[Any] , lowercase_: List[str] ) -> Optional[int]: A__ : Dict = f"""repo_txt_data-{int(time.time() * 10E3 )}""" A__ : Optional[int] = f"""{CI_HUB_USER}/{repo_name}""" hf_api.create_repo(lowercase_ , token=lowercase_ , repo_type="""dataset""" , private=lowercase_ ) hf_api.upload_file( token=lowercase_ , path_or_fileobj=str(lowercase_ ) , path_in_repo="""data/text_data.txt""" , repo_id=lowercase_ , repo_type="""dataset""" , ) yield repo_id try: hf_api.delete_repo(lowercase_ , token=lowercase_ , repo_type="""dataset""" ) except (requests.exceptions.HTTPError, ValueError): # catch http error and token invalid error pass @pytest.fixture() def UpperCamelCase (lowercase_: Union[str, Any] , lowercase_: Union[str, Any] , lowercase_: int ) -> int: return hf_private_dataset_repo_txt_data_ @pytest.fixture(scope="""session""" ) def UpperCamelCase (lowercase_: HfApi , lowercase_: List[str] , lowercase_: int ) -> int: A__ : Optional[int] = f"""repo_zipped_txt_data-{int(time.time() * 10E3 )}""" A__ : str = f"""{CI_HUB_USER}/{repo_name}""" hf_api.create_repo(lowercase_ , token=lowercase_ , repo_type="""dataset""" , private=lowercase_ ) hf_api.upload_file( token=lowercase_ , path_or_fileobj=str(lowercase_ ) , path_in_repo="""data.zip""" , repo_id=lowercase_ , repo_type="""dataset""" , ) yield repo_id try: hf_api.delete_repo(lowercase_ , token=lowercase_ , repo_type="""dataset""" ) except (requests.exceptions.HTTPError, ValueError): # catch http error and token invalid error pass @pytest.fixture() def UpperCamelCase (lowercase_: Optional[Any] , lowercase_: List[Any] , lowercase_: Dict ) -> List[Any]: return hf_private_dataset_repo_zipped_txt_data_ @pytest.fixture(scope="""session""" ) def UpperCamelCase (lowercase_: HfApi , lowercase_: List[Any] , lowercase_: List[str] ) -> int: A__ : Dict = f"""repo_zipped_img_data-{int(time.time() * 10E3 )}""" A__ : Tuple = f"""{CI_HUB_USER}/{repo_name}""" hf_api.create_repo(lowercase_ , token=lowercase_ , repo_type="""dataset""" , private=lowercase_ ) hf_api.upload_file( token=lowercase_ , path_or_fileobj=str(lowercase_ ) , path_in_repo="""data.zip""" , repo_id=lowercase_ , repo_type="""dataset""" , ) yield repo_id try: hf_api.delete_repo(lowercase_ , token=lowercase_ , repo_type="""dataset""" ) except (requests.exceptions.HTTPError, ValueError): # catch http error and token invalid error pass @pytest.fixture() def UpperCamelCase (lowercase_: List[str] , lowercase_: List[str] , lowercase_: Any ) -> Optional[int]: return hf_private_dataset_repo_zipped_img_data_
64
from __future__ import annotations from collections.abc import Callable A_ : List[Any] = list[list[float | int]] def UpperCamelCase (lowercase_: Matrix , lowercase_: Matrix ) -> Matrix: A__ : int = len(lowercase_ ) A__ : Matrix = [[0 for _ in range(size + 1 )] for _ in range(lowercase_ )] A__ : int A__ : int A__ : int A__ : int A__ : int A__ : float for row in range(lowercase_ ): for col in range(lowercase_ ): A__ : List[str] = matrix[row][col] A__ : int = vector[row][0] A__ : Optional[int] = 0 A__ : str = 0 while row < size and col < size: # pivoting A__ : int = max((abs(augmented[rowa][col] ), rowa) for rowa in range(lowercase_ , lowercase_ ) )[ 1 ] if augmented[pivot_row][col] == 0: col += 1 continue else: A__ , A__ : Union[str, Any] = augmented[pivot_row], augmented[row] for rowa in range(row + 1 , lowercase_ ): A__ : List[Any] = augmented[rowa][col] / augmented[row][col] A__ : Dict = 0 for cola in range(col + 1 , size + 1 ): augmented[rowa][cola] -= augmented[row][cola] * ratio row += 1 col += 1 # back substitution for col in range(1 , lowercase_ ): for row in range(lowercase_ ): A__ : List[str] = augmented[row][col] / augmented[col][col] for cola in range(lowercase_ , size + 1 ): augmented[row][cola] -= augmented[col][cola] * ratio # round to get rid of numbers like 2.000000000000004 return [ [round(augmented[row][size] / augmented[row][row] , 10 )] for row in range(lowercase_ ) ] def UpperCamelCase (lowercase_: list[int] ) -> Callable[[int], int]: A__ : int = len(lowercase_ ) A__ : Matrix = [[0 for _ in range(lowercase_ )] for _ in range(lowercase_ )] A__ : Matrix = [[0] for _ in range(lowercase_ )] A__ : Matrix A__ : int A__ : int A__ : int for x_val, y_val in enumerate(lowercase_ ): for col in range(lowercase_ ): A__ : Dict = (x_val + 1) ** (size - col - 1) A__ : Any = y_val A__ : Union[str, Any] = solve(lowercase_ , lowercase_ ) def interpolated_func(lowercase_: int ) -> int: return sum( round(coeffs[x_val][0] ) * (var ** (size - x_val - 1)) for x_val in range(lowercase_ ) ) return interpolated_func def UpperCamelCase (lowercase_: int ) -> int: return ( 1 - variable + variable**2 - variable**3 + variable**4 - variable**5 + variable**6 - variable**7 + variable**8 - variable**9 + variable**10 ) def UpperCamelCase (lowercase_: Callable[[int], int] = question_function , lowercase_: int = 10 ) -> int: A__ : list[int] = [func(lowercase_ ) for x_val in range(1 , order + 1 )] A__ : list[Callable[[int], int]] = [ interpolate(data_points[:max_coeff] ) for max_coeff in range(1 , order + 1 ) ] A__ : int = 0 A__ : Callable[[int], int] A__ : int for poly in polynomials: A__ : List[str] = 1 while func(lowercase_ ) == poly(lowercase_ ): x_val += 1 ret += poly(lowercase_ ) return ret if __name__ == "__main__": print(f'''{solution() = }''')
64
1
import gc import unittest import torch from transformers import CLIPTextConfig, CLIPTextModel, CLIPTextModelWithProjection, CLIPTokenizer from diffusers import ( AutoencoderKL, DDIMScheduler, DDPMScheduler, PriorTransformer, StableUnCLIPPipeline, UNetaDConditionModel, ) from diffusers.pipelines.stable_diffusion.stable_unclip_image_normalizer import StableUnCLIPImageNormalizer from diffusers.utils.testing_utils import enable_full_determinism, load_numpy, require_torch_gpu, slow, torch_device from ..pipeline_params import TEXT_TO_IMAGE_BATCH_PARAMS, TEXT_TO_IMAGE_IMAGE_PARAMS, TEXT_TO_IMAGE_PARAMS from ..test_pipelines_common import ( PipelineKarrasSchedulerTesterMixin, PipelineLatentTesterMixin, PipelineTesterMixin, assert_mean_pixel_difference, ) enable_full_determinism() class _a (__magic_name__ , __magic_name__ , __magic_name__ , unittest.TestCase ): '''simple docstring''' UpperCAmelCase__: Dict = StableUnCLIPPipeline UpperCAmelCase__: Any = TEXT_TO_IMAGE_PARAMS UpperCAmelCase__: List[str] = TEXT_TO_IMAGE_BATCH_PARAMS UpperCAmelCase__: Dict = TEXT_TO_IMAGE_IMAGE_PARAMS UpperCAmelCase__: str = TEXT_TO_IMAGE_IMAGE_PARAMS # TODO(will) Expected attn_bias.stride(1) == 0 to be true, but got false UpperCAmelCase__: Union[str, Any] = False def __A ( self ): A__ : Dict = 32 A__ : Any = embedder_hidden_size # prior components torch.manual_seed(0 ) A__ : str = CLIPTokenizer.from_pretrained("""hf-internal-testing/tiny-random-clip""" ) torch.manual_seed(0 ) A__ : int = CLIPTextModelWithProjection( CLIPTextConfig( bos_token_id=0 , eos_token_id=2 , hidden_size=A__ , projection_dim=A__ , intermediate_size=37 , layer_norm_eps=1e-05 , num_attention_heads=4 , num_hidden_layers=5 , pad_token_id=1 , vocab_size=1000 , ) ) torch.manual_seed(0 ) A__ : List[Any] = PriorTransformer( num_attention_heads=2 , attention_head_dim=12 , embedding_dim=A__ , num_layers=1 , ) torch.manual_seed(0 ) A__ : Optional[int] = DDPMScheduler( variance_type="""fixed_small_log""" , prediction_type="""sample""" , num_train_timesteps=1000 , clip_sample=A__ , clip_sample_range=5.0 , beta_schedule="""squaredcos_cap_v2""" , ) # regular denoising components torch.manual_seed(0 ) A__ : Dict = StableUnCLIPImageNormalizer(embedding_dim=A__ ) A__ : int = DDPMScheduler(beta_schedule="""squaredcos_cap_v2""" ) torch.manual_seed(0 ) A__ : str = CLIPTokenizer.from_pretrained("""hf-internal-testing/tiny-random-clip""" ) torch.manual_seed(0 ) A__ : List[str] = CLIPTextModel( CLIPTextConfig( bos_token_id=0 , eos_token_id=2 , hidden_size=A__ , projection_dim=32 , intermediate_size=37 , layer_norm_eps=1e-05 , num_attention_heads=4 , num_hidden_layers=5 , pad_token_id=1 , vocab_size=1000 , ) ) torch.manual_seed(0 ) A__ : Union[str, Any] = UNetaDConditionModel( sample_size=32 , in_channels=4 , out_channels=4 , down_block_types=("""CrossAttnDownBlock2D""", """DownBlock2D""") , up_block_types=("""UpBlock2D""", """CrossAttnUpBlock2D""") , block_out_channels=(32, 64) , attention_head_dim=(2, 4) , class_embed_type="""projection""" , projection_class_embeddings_input_dim=embedder_projection_dim * 2 , cross_attention_dim=A__ , layers_per_block=1 , upcast_attention=A__ , use_linear_projection=A__ , ) torch.manual_seed(0 ) A__ : Union[str, Any] = DDIMScheduler( beta_schedule="""scaled_linear""" , beta_start=0.0_0_0_8_5 , beta_end=0.0_1_2 , prediction_type="""v_prediction""" , set_alpha_to_one=A__ , steps_offset=1 , ) torch.manual_seed(0 ) A__ : List[Any] = AutoencoderKL() A__ : Union[str, Any] = { # prior components """prior_tokenizer""": prior_tokenizer, """prior_text_encoder""": prior_text_encoder, """prior""": prior, """prior_scheduler""": prior_scheduler, # image noising components """image_normalizer""": image_normalizer, """image_noising_scheduler""": image_noising_scheduler, # regular denoising components """tokenizer""": tokenizer, """text_encoder""": text_encoder, """unet""": unet, """scheduler""": scheduler, """vae""": vae, } return components def __A ( self , A__ , A__=0 ): if str(A__ ).startswith("""mps""" ): A__ : Optional[int] = torch.manual_seed(A__ ) else: A__ : int = torch.Generator(device=A__ ).manual_seed(A__ ) A__ : Dict = { """prompt""": """A painting of a squirrel eating a burger""", """generator""": generator, """num_inference_steps""": 2, """prior_num_inference_steps""": 2, """output_type""": """numpy""", } return inputs def __A ( self ): A__ : int = torch_device == """cpu""" self._test_attention_slicing_forward_pass(test_max_difference=A__ ) def __A ( self ): A__ : str = torch_device in ["""cpu""", """mps"""] self._test_inference_batch_single_identical(test_max_difference=A__ ) @slow @require_torch_gpu class _a (unittest.TestCase ): '''simple docstring''' def __A ( self ): # clean up the VRAM after each test super().tearDown() gc.collect() torch.cuda.empty_cache() def __A ( self ): A__ : Any = load_numpy( """https://huggingface.co/datasets/hf-internal-testing/diffusers-images/resolve/main/stable_unclip/stable_unclip_2_1_l_anime_turtle_fp16.npy""" ) A__ : Optional[Any] = StableUnCLIPPipeline.from_pretrained("""fusing/stable-unclip-2-1-l""" , torch_dtype=torch.floataa ) pipe.to(A__ ) pipe.set_progress_bar_config(disable=A__ ) # stable unclip will oom when integration tests are run on a V100, # so turn on memory savings pipe.enable_attention_slicing() pipe.enable_sequential_cpu_offload() A__ : Optional[int] = torch.Generator(device="""cpu""" ).manual_seed(0 ) A__ : List[str] = pipe("""anime turle""" , generator=A__ , output_type="""np""" ) A__ : Tuple = output.images[0] assert image.shape == (768, 768, 3) assert_mean_pixel_difference(A__ , A__ ) def __A ( self ): torch.cuda.empty_cache() torch.cuda.reset_max_memory_allocated() torch.cuda.reset_peak_memory_stats() A__ : Optional[int] = StableUnCLIPPipeline.from_pretrained("""fusing/stable-unclip-2-1-l""" , torch_dtype=torch.floataa ) A__ : List[Any] = pipe.to(A__ ) pipe.set_progress_bar_config(disable=A__ ) pipe.enable_attention_slicing() pipe.enable_sequential_cpu_offload() A__ : Any = pipe( """anime turtle""" , prior_num_inference_steps=2 , num_inference_steps=2 , output_type="""np""" , ) A__ : Tuple = torch.cuda.max_memory_allocated() # make sure that less than 7 GB is allocated assert mem_bytes < 7 * 10**9
64
from functools import lru_cache @lru_cache def UpperCamelCase (lowercase_: int ) -> int: if num < 0: raise ValueError("""Number should not be negative.""" ) return 1 if num in (0, 1) else num * factorial(num - 1 ) if __name__ == "__main__": import doctest doctest.testmod()
64
1
import unittest from knapsack import greedy_knapsack as kp class _a (unittest.TestCase ): '''simple docstring''' def __A ( self ): A__ : Optional[int] = [10, 20, 30, 40, 50, 60] A__ : str = [2, 4, 6, 8, 10, 12] A__ : Union[str, Any] = 100 self.assertEqual(kp.calc_profit(A__ , A__ , A__ ) , 210 ) def __A ( self ): self.assertRaisesRegex(A__ , """max_weight must greater than zero.""" ) def __A ( self ): self.assertRaisesRegex(A__ , """Weight can not be negative.""" ) def __A ( self ): self.assertRaisesRegex(A__ , """Profit can not be negative.""" ) def __A ( self ): self.assertRaisesRegex(A__ , """max_weight must greater than zero.""" ) def __A ( self ): self.assertRaisesRegex( A__ , """The length of profit and weight must be same.""" ) if __name__ == "__main__": unittest.main()
64
import os import tempfile from functools import partial from unittest import TestCase from unittest.mock import patch import datasets import datasets.config from .utils import require_beam class _a (datasets.BeamBasedBuilder ): '''simple docstring''' def __A ( self ): return datasets.DatasetInfo( features=datasets.Features({"""content""": datasets.Value("""string""" )} ) , supervised_keys=A__ , ) def __A ( self , A__ , A__ ): return [datasets.SplitGenerator(name=datasets.Split.TRAIN , gen_kwargs={"""examples""": get_test_dummy_examples()} )] def __A ( self , A__ , A__ ): import apache_beam as beam return pipeline | "Load Examples" >> beam.Create(A__ ) class _a (datasets.BeamBasedBuilder ): '''simple docstring''' def __A ( self ): return datasets.DatasetInfo( features=datasets.Features({"""a""": datasets.Sequence({"""b""": datasets.Value("""string""" )} )} ) , supervised_keys=A__ , ) def __A ( self , A__ , A__ ): return [ datasets.SplitGenerator(name=datasets.Split.TRAIN , gen_kwargs={"""examples""": get_test_nested_examples()} ) ] def __A ( self , A__ , A__ ): import apache_beam as beam return pipeline | "Load Examples" >> beam.Create(A__ ) def UpperCamelCase () -> Dict: return [(i, {"content": content}) for i, content in enumerate(["""foo""", """bar""", """foobar"""] )] def UpperCamelCase () -> Tuple: return [(i, {"a": {"b": [content]}}) for i, content in enumerate(["""foo""", """bar""", """foobar"""] )] class _a (__magic_name__ ): '''simple docstring''' @require_beam def __A ( self ): A__ : Dict = len(get_test_dummy_examples() ) with tempfile.TemporaryDirectory() as tmp_cache_dir: A__ : int = DummyBeamDataset(cache_dir=A__ , beam_runner="""DirectRunner""" ) builder.download_and_prepare() self.assertTrue( os.path.exists( os.path.join(A__ , builder.name , """default""" , """0.0.0""" , F"""{builder.name}-train.arrow""" ) ) ) self.assertDictEqual(builder.info.features , datasets.Features({"""content""": datasets.Value("""string""" )} ) ) A__ : int = builder.as_dataset() self.assertEqual(dset["""train"""].num_rows , A__ ) self.assertEqual(dset["""train"""].info.splits["""train"""].num_examples , A__ ) self.assertDictEqual(dset["""train"""][0] , get_test_dummy_examples()[0][1] ) self.assertDictEqual( dset["""train"""][expected_num_examples - 1] , get_test_dummy_examples()[expected_num_examples - 1][1] ) self.assertTrue( os.path.exists(os.path.join(A__ , builder.name , """default""" , """0.0.0""" , """dataset_info.json""" ) ) ) del dset @require_beam def __A ( self ): import apache_beam as beam A__ : int = beam.io.parquetio.WriteToParquet A__ : List[str] = len(get_test_dummy_examples() ) with tempfile.TemporaryDirectory() as tmp_cache_dir: A__ : str = DummyBeamDataset(cache_dir=A__ , beam_runner="""DirectRunner""" ) with patch("""apache_beam.io.parquetio.WriteToParquet""" ) as write_parquet_mock: A__ : Optional[Any] = partial(A__ , num_shards=2 ) builder.download_and_prepare() self.assertTrue( os.path.exists( os.path.join( A__ , builder.name , """default""" , """0.0.0""" , F"""{builder.name}-train-00000-of-00002.arrow""" ) ) ) self.assertTrue( os.path.exists( os.path.join( A__ , builder.name , """default""" , """0.0.0""" , F"""{builder.name}-train-00000-of-00002.arrow""" ) ) ) self.assertDictEqual(builder.info.features , datasets.Features({"""content""": datasets.Value("""string""" )} ) ) A__ : Optional[int] = builder.as_dataset() self.assertEqual(dset["""train"""].num_rows , A__ ) self.assertEqual(dset["""train"""].info.splits["""train"""].num_examples , A__ ) # Order is not preserved when sharding, so we just check that all the elements are there self.assertListEqual(sorted(dset["""train"""]["""content"""] ) , sorted(["""foo""", """bar""", """foobar"""] ) ) self.assertTrue( os.path.exists(os.path.join(A__ , builder.name , """default""" , """0.0.0""" , """dataset_info.json""" ) ) ) del dset @require_beam def __A ( self ): with tempfile.TemporaryDirectory() as tmp_cache_dir: A__ : int = DummyBeamDataset(cache_dir=A__ ) self.assertRaises(datasets.builder.MissingBeamOptions , builder.download_and_prepare ) @require_beam def __A ( self ): A__ : List[Any] = len(get_test_nested_examples() ) with tempfile.TemporaryDirectory() as tmp_cache_dir: A__ : Optional[int] = NestedBeamDataset(cache_dir=A__ , beam_runner="""DirectRunner""" ) builder.download_and_prepare() self.assertTrue( os.path.exists( os.path.join(A__ , builder.name , """default""" , """0.0.0""" , F"""{builder.name}-train.arrow""" ) ) ) self.assertDictEqual( builder.info.features , datasets.Features({"""a""": datasets.Sequence({"""b""": datasets.Value("""string""" )} )} ) ) A__ : Optional[int] = builder.as_dataset() self.assertEqual(dset["""train"""].num_rows , A__ ) self.assertEqual(dset["""train"""].info.splits["""train"""].num_examples , A__ ) self.assertDictEqual(dset["""train"""][0] , get_test_nested_examples()[0][1] ) self.assertDictEqual( dset["""train"""][expected_num_examples - 1] , get_test_nested_examples()[expected_num_examples - 1][1] ) self.assertTrue( os.path.exists(os.path.join(A__ , builder.name , """default""" , """0.0.0""" , """dataset_info.json""" ) ) ) del dset
64
1
def UpperCamelCase (lowercase_: list ) -> list: if len(lowercase_ ) <= 1: return [tuple(lowercase_ )] A__ : Dict = [] def generate(lowercase_: int , lowercase_: list ): if k == 1: res.append(tuple(arr[:] ) ) return generate(k - 1 , lowercase_ ) for i in range(k - 1 ): if k % 2 == 0: # k is even A__ , A__ : Optional[int] = arr[k - 1], arr[i] else: # k is odd A__ , A__ : List[str] = arr[k - 1], arr[0] generate(k - 1 , lowercase_ ) generate(len(lowercase_ ) , lowercase_ ) return res if __name__ == "__main__": A_ : int = input('Enter numbers separated by a comma:\n').strip() A_ : Any = [int(item) for item in user_input.split(',')] print(heaps(arr))
64
import warnings from ...utils import logging from .image_processing_poolformer import PoolFormerImageProcessor A_ : Union[str, Any] = logging.get_logger(__name__) class _a (__magic_name__ ): '''simple docstring''' def __init__( self , *A__ , **A__ ): warnings.warn( """The class PoolFormerFeatureExtractor is deprecated and will be removed in version 5 of Transformers.""" """ Please use PoolFormerImageProcessor instead.""" , A__ , ) super().__init__(*A__ , **A__ )
64
1
import copy from dataclasses import dataclass, field from typing import ClassVar, Dict from ..features import ClassLabel, Features, Value from .base import TaskTemplate @dataclass(frozen=__magic_name__ ) class _a (__magic_name__ ): '''simple docstring''' UpperCAmelCase__: str = field(default='''text-classification''' , metadata={'''include_in_asdict_even_if_is_default''': True} ) UpperCAmelCase__: ClassVar[Features] = Features({'''text''': Value('''string''' )} ) UpperCAmelCase__: ClassVar[Features] = Features({'''labels''': ClassLabel} ) UpperCAmelCase__: str = "text" UpperCAmelCase__: str = "labels" def __A ( self , A__ ): if self.label_column not in features: raise ValueError(F"""Column {self.label_column} is not present in features.""" ) if not isinstance(features[self.label_column] , A__ ): raise ValueError(F"""Column {self.label_column} is not a ClassLabel.""" ) A__ : Optional[Any] = copy.deepcopy(self ) A__ : Tuple = self.label_schema.copy() A__ : List[Any] = features[self.label_column] A__ : str = label_schema return task_template @property def __A ( self ): return { self.text_column: "text", self.label_column: "labels", }
64
import inspect import logging import os import random import shutil import tempfile import unittest import pytest import torch from torch import nn from torch.utils.data import DataLoader, TensorDataset from accelerate import Accelerator from accelerate.test_utils import execute_subprocess_async, require_cuda from accelerate.utils import ProjectConfiguration, set_seed A_ : Any = logging.getLogger(__name__) def UpperCamelCase (lowercase_: Optional[Any]=2 , lowercase_: Union[str, Any]=3 , lowercase_: int=16 , lowercase_: int = 10 , lowercase_: int = 2 ) -> int: def get_dataset(lowercase_: Optional[int] ): A__ : Optional[Any] = torch.randn(batch_size * n_batches , 1 ) return TensorDataset(lowercase_ , a * x + b + 0.1 * torch.randn(batch_size * n_batches , 1 ) ) A__ : Dict = get_dataset(lowercase_ ) A__ : Any = get_dataset(lowercase_ ) A__ : Dict = DataLoader(lowercase_ , shuffle=lowercase_ , batch_size=lowercase_ , num_workers=4 ) A__ : Optional[Any] = DataLoader(lowercase_ , shuffle=lowercase_ , batch_size=lowercase_ , num_workers=4 ) return (train_dataloader, valid_dataloader) def UpperCamelCase (lowercase_: Optional[Any] , lowercase_: List[str] , lowercase_: int , lowercase_: int , lowercase_: List[str] , lowercase_: Dict=None ) -> List[Any]: A__ : List[Any] = [] for epoch in range(lowercase_ ): # Train quickly model.train() for batch in dataloader: A__ , A__ : Any = batch A__ : Any = model(lowercase_ ) A__ : Any = torch.nn.functional.mse_loss(lowercase_ , lowercase_ ) accelerator.backward(lowercase_ ) optimizer.step() optimizer.zero_grad() rands.append(random.random() ) # Introduce some randomness if scheduler is not None: scheduler.step() return rands class _a (nn.Module ): '''simple docstring''' def __init__( self ): super().__init__() A__ : str = nn.Parameter(torch.randn(1 ) ) A__ : Any = nn.Parameter(torch.randn(1 ) ) def __A ( self , A__ ): return x * self.a + self.b class _a (unittest.TestCase ): '''simple docstring''' def __A ( self ): with tempfile.TemporaryDirectory() as tmpdir: set_seed(42 ) A__ : Optional[Any] = DummyModel() A__ : Optional[Any] = torch.optim.Adam(params=model.parameters() , lr=1e-3 ) A__ , A__ : str = dummy_dataloaders() A__ : Dict = ProjectConfiguration(total_limit=1 , project_dir=A__ , automatic_checkpoint_naming=A__ ) # Train baseline A__ : List[str] = Accelerator(project_config=A__ ) A__ , A__ , A__ , A__ : Any = accelerator.prepare( A__ , A__ , A__ , A__ ) # Save initial accelerator.save_state() # Save second state accelerator.save_state() self.assertEqual(len(os.listdir(accelerator.project_dir ) ) , 1 ) def __A ( self ): with tempfile.TemporaryDirectory() as tmpdir: set_seed(42 ) A__ : str = DummyModel() A__ : Optional[int] = torch.optim.Adam(params=model.parameters() , lr=1e-3 ) A__ , A__ : int = dummy_dataloaders() # Train baseline A__ : str = Accelerator() A__ , A__ , A__ , A__ : List[str] = accelerator.prepare( A__ , A__ , A__ , A__ ) # Save initial A__ : List[Any] = os.path.join(A__ , """initial""" ) accelerator.save_state(A__ ) ((A__) , (A__)) : str = model.a.item(), model.b.item() A__ : Dict = optimizer.state_dict() A__ : List[str] = train(3 , A__ , A__ , A__ , A__ ) ((A__) , (A__)) : str = model.a.item(), model.b.item() A__ : Any = optimizer.state_dict() # Train partially set_seed(42 ) A__ : Optional[int] = DummyModel() A__ : Dict = torch.optim.Adam(params=model.parameters() , lr=1e-3 ) A__ , A__ : Dict = dummy_dataloaders() A__ : List[str] = Accelerator() A__ , A__ , A__ , A__ : Optional[Any] = accelerator.prepare( A__ , A__ , A__ , A__ ) accelerator.load_state(A__ ) ((A__) , (A__)) : Tuple = model.a.item(), model.b.item() A__ : Union[str, Any] = optimizer.state_dict() self.assertEqual(A__ , A__ ) self.assertEqual(A__ , A__ ) self.assertEqual(A__ , A__ ) A__ : List[str] = train(2 , A__ , A__ , A__ , A__ ) # Save everything A__ : Optional[int] = os.path.join(A__ , """checkpoint""" ) accelerator.save_state(A__ ) # Load everything back in and make sure all states work accelerator.load_state(A__ ) test_rands += train(1 , A__ , A__ , A__ , A__ ) ((A__) , (A__)) : Union[str, Any] = model.a.item(), model.b.item() A__ : Optional[int] = optimizer.state_dict() self.assertEqual(A__ , A__ ) self.assertEqual(A__ , A__ ) self.assertEqual(A__ , A__ ) self.assertEqual(A__ , A__ ) def __A ( self ): with tempfile.TemporaryDirectory() as tmpdir: set_seed(42 ) A__ : int = DummyModel() A__ : Optional[int] = torch.optim.Adam(params=model.parameters() , lr=1e-3 ) A__ , A__ : List[str] = dummy_dataloaders() A__ : str = ProjectConfiguration(automatic_checkpoint_naming=A__ ) # Train baseline A__ : Any = Accelerator(project_dir=A__ , project_config=A__ ) A__ , A__ , A__ , A__ : str = accelerator.prepare( A__ , A__ , A__ , A__ ) # Save initial accelerator.save_state() ((A__) , (A__)) : Tuple = model.a.item(), model.b.item() A__ : int = optimizer.state_dict() A__ : int = train(3 , A__ , A__ , A__ , A__ ) ((A__) , (A__)) : Optional[Any] = model.a.item(), model.b.item() A__ : Any = optimizer.state_dict() # Train partially set_seed(42 ) A__ : Dict = DummyModel() A__ : List[Any] = torch.optim.Adam(params=model.parameters() , lr=1e-3 ) A__ , A__ : Union[str, Any] = dummy_dataloaders() A__ : List[Any] = ProjectConfiguration(iteration=1 , automatic_checkpoint_naming=A__ ) A__ : Dict = Accelerator(project_dir=A__ , project_config=A__ ) A__ , A__ , A__ , A__ : Union[str, Any] = accelerator.prepare( A__ , A__ , A__ , A__ ) accelerator.load_state(os.path.join(A__ , """checkpoints""" , """checkpoint_0""" ) ) ((A__) , (A__)) : Optional[int] = model.a.item(), model.b.item() A__ : Tuple = optimizer.state_dict() self.assertEqual(A__ , A__ ) self.assertEqual(A__ , A__ ) self.assertEqual(A__ , A__ ) A__ : str = train(2 , A__ , A__ , A__ , A__ ) # Save everything accelerator.save_state() # Load everything back in and make sure all states work accelerator.load_state(os.path.join(A__ , """checkpoints""" , """checkpoint_1""" ) ) test_rands += train(1 , A__ , A__ , A__ , A__ ) ((A__) , (A__)) : Optional[int] = model.a.item(), model.b.item() A__ : List[Any] = optimizer.state_dict() self.assertEqual(A__ , A__ ) self.assertEqual(A__ , A__ ) self.assertEqual(A__ , A__ ) self.assertEqual(A__ , A__ ) def __A ( self ): A__ : Union[str, Any] = torch.tensor([1, 2, 3] ) A__ : int = torch.tensor([2, 3, 4] ) A__ : List[Any] = DummyModel() A__ : List[Any] = torch.optim.Adam(net.parameters() ) A__ : Tuple = Accelerator() with self.assertRaises(A__ ) as ve: accelerator.register_for_checkpointing(A__ , A__ , A__ , A__ ) A__ : Any = str(ve.exception ) self.assertTrue("""Item at index 0""" in message ) self.assertTrue("""Item at index 1""" in message ) self.assertFalse("""Item at index 2""" in message ) self.assertFalse("""Item at index 3""" in message ) def __A ( self ): with tempfile.TemporaryDirectory() as tmpdir: set_seed(42 ) A__ : Any = DummyModel() A__ : Union[str, Any] = torch.optim.Adam(params=model.parameters() , lr=1e-3 ) A__ : Dict = torch.optim.lr_scheduler.StepLR(A__ , step_size=1 , gamma=0.9_9 ) A__ , A__ : List[Any] = dummy_dataloaders() A__ : Tuple = ProjectConfiguration(automatic_checkpoint_naming=A__ ) # Train baseline A__ : Optional[Any] = Accelerator(project_dir=A__ , project_config=A__ ) A__ , A__ , A__ , A__ , A__ : Union[str, Any] = accelerator.prepare( A__ , A__ , A__ , A__ , A__ ) # Save initial accelerator.save_state() A__ : Tuple = scheduler.state_dict() train(3 , A__ , A__ , A__ , A__ , A__ ) self.assertNotEqual(A__ , scheduler.state_dict() ) # Load everything back in and make sure all states work accelerator.load_state(os.path.join(A__ , """checkpoints""" , """checkpoint_0""" ) ) self.assertEqual(A__ , scheduler.state_dict() ) def __A ( self ): with tempfile.TemporaryDirectory() as tmpdir: set_seed(42 ) A__ : Optional[Any] = DummyModel() A__ : int = ProjectConfiguration(automatic_checkpoint_naming=A__ , total_limit=2 ) # Train baseline A__ : List[str] = Accelerator(project_dir=A__ , project_config=A__ ) A__ : Union[str, Any] = accelerator.prepare(A__ ) # Save 3 states: for _ in range(11 ): accelerator.save_state() self.assertTrue(not os.path.exists(os.path.join(A__ , """checkpoints""" , """checkpoint_0""" ) ) ) self.assertTrue(os.path.exists(os.path.join(A__ , """checkpoints""" , """checkpoint_9""" ) ) ) self.assertTrue(os.path.exists(os.path.join(A__ , """checkpoints""" , """checkpoint_10""" ) ) ) @require_cuda def __A ( self ): A__ : Dict = ["""torchrun""", F"""--nproc_per_node={torch.cuda.device_count()}""", inspect.getfile(self.__class__ )] execute_subprocess_async(A__ , env=os.environ.copy() ) if __name__ == "__main__": A_ : List[str] = '/tmp/accelerate/state_checkpointing' A_ : Optional[Any] = DummyModel() A_ : Union[str, Any] = torch.optim.Adam(params=model.parameters(), lr=1E-3) A_ : str = torch.optim.lr_scheduler.StepLR(optimizer, step_size=1, gamma=0.99) A_ , A_ : List[Any] = dummy_dataloaders() A_ : int = ProjectConfiguration(automatic_checkpoint_naming=True) # Train baseline A_ : List[str] = Accelerator(project_dir=savedir, project_config=project_config, mixed_precision='no') if accelerator.process_index == 0: if os.path.exists(savedir): shutil.rmtree(savedir) os.makedirs(savedir) A_ , A_ , A_ , A_ , A_ : List[Any] = accelerator.prepare( model, optimizer, train_dataloader, valid_dataloader, scheduler ) A_ , A_ : Dict = accelerator.prepare(model, optimizer) train(3, model, train_dataloader, optimizer, accelerator, scheduler) # Check that the intial optimizer is loaded on the GPU for group in optimizer.param_groups: A_ : str = group['params'][0].device break assert param_device.type == accelerator.device.type A_ : Optional[Any] = model.cpu() accelerator.wait_for_everyone() accelerator.save_state() accelerator.wait_for_everyone() # Check CPU state accelerator.load_state(os.path.join(savedir, 'checkpoints', 'checkpoint_0'), map_location='cpu') for group in optimizer.param_groups: A_ : str = group['params'][0].device break assert ( param_device.type == torch.device('cpu').type ), f"Loaded optimizer states did not match, expected to be loaded on the CPU but got {param_device}" # Check device state model.to(accelerator.device) accelerator.load_state(os.path.join(savedir, 'checkpoints', 'checkpoint_0'), map_location='on_device') for group in optimizer.param_groups: A_ : Tuple = group['params'][0].device break assert ( param_device.type == accelerator.device.type ), f"Loaded optimizer states did not match, expected to be loaded on {accelerator.device} but got {param_device}" # Check error with pytest.raises(TypeError, match='Unsupported optimizer map location passed'): accelerator.load_state(os.path.join(savedir, 'checkpoints', 'checkpoint_0'), map_location='invalid') accelerator.wait_for_everyone() if accelerator.process_index == 0: shutil.rmtree(savedir) accelerator.wait_for_everyone()
64
1
from __future__ import annotations from dataclasses import dataclass @dataclass class _a : '''simple docstring''' UpperCAmelCase__: float UpperCAmelCase__: TreeNode | None = None UpperCAmelCase__: TreeNode | None = None def UpperCamelCase (lowercase_: TreeNode | None ) -> bool: # Validation def is_valid_tree(lowercase_: TreeNode | None ) -> bool: if node is None: return True if not isinstance(lowercase_ , lowercase_ ): return False try: float(node.data ) except (TypeError, ValueError): return False return is_valid_tree(node.left ) and is_valid_tree(node.right ) if not is_valid_tree(lowercase_ ): raise ValueError( """Each node should be type of TreeNode and data should be float.""" ) def is_binary_search_tree_recursive_check( lowercase_: TreeNode | None , lowercase_: float , lowercase_: float ) -> bool: if node is None: return True return ( left_bound < node.data < right_bound and is_binary_search_tree_recursive_check(node.left , lowercase_ , node.data ) and is_binary_search_tree_recursive_check( node.right , node.data , lowercase_ ) ) return is_binary_search_tree_recursive_check(lowercase_ , -float("""inf""" ) , float("""inf""" ) ) if __name__ == "__main__": import doctest doctest.testmod()
64
def UpperCamelCase (lowercase_: str , lowercase_: str ) -> bool: A__ : Union[str, Any] = len(lowercase_ ) A__ : List[Any] = len(lowercase_ ) A__ : List[Any] = [[False for _ in range(m + 1 )] for _ in range(n + 1 )] A__ : str = True for i in range(lowercase_ ): for j in range(m + 1 ): if dp[i][j]: if j < m and a[i].upper() == b[j]: A__ : int = True if a[i].islower(): A__ : Dict = True return dp[n][m] if __name__ == "__main__": import doctest doctest.testmod()
64
1
def UpperCamelCase (lowercase_: int ) -> "list[int]": if upper_limit < 0: raise ValueError("""Limit for the Catalan sequence must be ≥ 0""" ) A__ : str = [0] * (upper_limit + 1) # Base case: C(0) = C(1) = 1 A__ : Dict = 1 if upper_limit > 0: A__ : str = 1 # Recurrence relation: C(i) = sum(C(j).C(i-j-1)), from j = 0 to i for i in range(2 , upper_limit + 1 ): for j in range(lowercase_ ): catalan_list[i] += catalan_list[j] * catalan_list[i - j - 1] return catalan_list if __name__ == "__main__": print('\n********* Catalan Numbers Using Dynamic Programming ************\n') print('\n*** Enter -1 at any time to quit ***') print('\nEnter the upper limit (≥ 0) for the Catalan number sequence: ', end='') try: while True: A_ : List[Any] = int(input().strip()) if N < 0: print('\n********* Goodbye!! ************') break else: print(f'''The Catalan numbers from 0 through {N} are:''') print(catalan_numbers(N)) print('Try another upper limit for the sequence: ', end='') except (NameError, ValueError): print('\n********* Invalid input, goodbye! ************\n') import doctest doctest.testmod()
64
import itertools import random import unittest import numpy as np from transformers import ASTFeatureExtractor from transformers.testing_utils import require_torch, require_torchaudio from transformers.utils.import_utils import is_torch_available from ...test_sequence_feature_extraction_common import SequenceFeatureExtractionTestMixin A_ : Dict = random.Random() if is_torch_available(): import torch def UpperCamelCase (lowercase_: Tuple , lowercase_: Tuple=1.0 , lowercase_: Dict=None , lowercase_: int=None ) -> str: if rng is None: A__ : Optional[Any] = global_rng A__ : List[str] = [] for batch_idx in range(shape[0] ): values.append([] ) for _ in range(shape[1] ): values[-1].append(rng.random() * scale ) return values class _a (unittest.TestCase ): '''simple docstring''' def __init__( self , A__ , A__=7 , A__=400 , A__=2000 , A__=1 , A__=0.0 , A__=1_6000 , A__=True , A__=True , ): A__ : Any = parent A__ : Optional[int] = batch_size A__ : Union[str, Any] = min_seq_length A__ : Dict = max_seq_length A__ : Tuple = (self.max_seq_length - self.min_seq_length) // (self.batch_size - 1) A__ : str = feature_size A__ : Optional[int] = padding_value A__ : List[str] = sampling_rate A__ : List[str] = return_attention_mask A__ : int = do_normalize def __A ( self ): return { "feature_size": self.feature_size, "padding_value": self.padding_value, "sampling_rate": self.sampling_rate, "return_attention_mask": self.return_attention_mask, "do_normalize": self.do_normalize, } def __A ( self , A__=False , A__=False ): def _flatten(A__ ): return list(itertools.chain(*A__ ) ) if equal_length: A__ : Dict = floats_list((self.batch_size, self.max_seq_length) ) else: # make sure that inputs increase in size A__ : Union[str, Any] = [ _flatten(floats_list((x, self.feature_size) ) ) for x in range(self.min_seq_length , self.max_seq_length , self.seq_length_diff ) ] if numpify: A__ : Optional[int] = [np.asarray(A__ ) for x in speech_inputs] return speech_inputs @require_torch @require_torchaudio class _a (__magic_name__ , unittest.TestCase ): '''simple docstring''' UpperCAmelCase__: int = ASTFeatureExtractor def __A ( self ): A__ : Optional[Any] = ASTFeatureExtractionTester(self ) def __A ( self ): # Tests that all call wrap to encode_plus and batch_encode_plus A__ : Optional[int] = self.feature_extraction_class(**self.feat_extract_tester.prepare_feat_extract_dict() ) # create three inputs of length 800, 1000, and 1200 A__ : List[Any] = [floats_list((1, x) )[0] for x in range(800 , 1400 , 200 )] A__ : Optional[Any] = [np.asarray(A__ ) for speech_input in speech_inputs] # Test not batched input A__ : Tuple = feat_extract(speech_inputs[0] , return_tensors="""np""" ).input_values A__ : Tuple = feat_extract(np_speech_inputs[0] , return_tensors="""np""" ).input_values self.assertTrue(np.allclose(A__ , A__ , atol=1e-3 ) ) # Test batched A__ : Tuple = feat_extract(A__ , padding=A__ , return_tensors="""np""" ).input_values A__ : Tuple = feat_extract(A__ , padding=A__ , return_tensors="""np""" ).input_values for enc_seq_a, enc_seq_a in zip(A__ , A__ ): self.assertTrue(np.allclose(A__ , A__ , atol=1e-3 ) ) # Test 2-D numpy arrays are batched. A__ : int = [floats_list((1, x) )[0] for x in (800, 800, 800)] A__ : List[str] = np.asarray(A__ ) A__ : Union[str, Any] = feat_extract(A__ , return_tensors="""np""" ).input_values A__ : Optional[Any] = feat_extract(A__ , return_tensors="""np""" ).input_values for enc_seq_a, enc_seq_a in zip(A__ , A__ ): self.assertTrue(np.allclose(A__ , A__ , atol=1e-3 ) ) @require_torch def __A ( self ): import torch A__ : Union[str, Any] = self.feature_extraction_class(**self.feat_extract_tester.prepare_feat_extract_dict() ) A__ : Tuple = np.random.rand(100 ).astype(np.floataa ) A__ : Tuple = np_speech_inputs.tolist() for inputs in [py_speech_inputs, np_speech_inputs]: A__ : List[str] = feature_extractor.pad([{"""input_values""": inputs}] , return_tensors="""np""" ) self.assertTrue(np_processed.input_values.dtype == np.floataa ) A__ : Any = feature_extractor.pad([{"""input_values""": inputs}] , return_tensors="""pt""" ) self.assertTrue(pt_processed.input_values.dtype == torch.floataa ) def __A ( self , A__ ): from datasets import load_dataset A__ : str = load_dataset("""hf-internal-testing/librispeech_asr_dummy""" , """clean""" , split="""validation""" ) # automatic decoding with librispeech A__ : str = ds.sort("""id""" ).select(range(A__ ) )[:num_samples]["""audio"""] return [x["array"] for x in speech_samples] @require_torch def __A ( self ): # fmt: off A__ : Optional[Any] = torch.tensor( [-0.9_8_9_4, -1.2_7_7_6, -0.9_0_6_6, -1.2_7_7_6, -0.9_3_4_9, -1.2_6_0_9, -1.0_3_8_6, -1.2_7_7_6, -1.1_5_6_1, -1.2_7_7_6, -1.2_0_5_2, -1.2_7_2_3, -1.2_1_9_0, -1.2_1_3_2, -1.2_7_7_6, -1.1_1_3_3, -1.1_9_5_3, -1.1_3_4_3, -1.1_5_8_4, -1.2_2_0_3, -1.1_7_7_0, -1.2_4_7_4, -1.2_3_8_1, -1.1_9_3_6, -0.9_2_7_0, -0.8_3_1_7, -0.8_0_4_9, -0.7_7_0_6, -0.7_5_6_5, -0.7_8_6_9] ) # fmt: on A__ : Any = self._load_datasamples(1 ) A__ : Tuple = ASTFeatureExtractor() A__ : Dict = feature_extractor(A__ , return_tensors="""pt""" ).input_values self.assertEquals(input_values.shape , (1, 1024, 128) ) self.assertTrue(torch.allclose(input_values[0, 0, :30] , A__ , atol=1e-4 ) )
64
1
import math class _a : '''simple docstring''' def __init__( self , A__=0 ): # a graph with Node 0,1,...,N-1 A__ : int = n A__ : Union[str, Any] = [ [math.inf for j in range(0 , A__ )] for i in range(0 , A__ ) ] # adjacency matrix for weight A__ : str = [ [math.inf for j in range(0 , A__ )] for i in range(0 , A__ ) ] # dp[i][j] stores minimum distance from i to j def __A ( self , A__ , A__ , A__ ): A__ : str = w def __A ( self ): for k in range(0 , self.n ): for i in range(0 , self.n ): for j in range(0 , self.n ): A__ : Union[str, Any] = min(self.dp[i][j] , self.dp[i][k] + self.dp[k][j] ) def __A ( self , A__ , A__ ): return self.dp[u][v] if __name__ == "__main__": A_ : List[Any] = Graph(5) graph.add_edge(0, 2, 9) graph.add_edge(0, 4, 10) graph.add_edge(1, 3, 5) graph.add_edge(2, 3, 7) graph.add_edge(3, 0, 10) graph.add_edge(3, 1, 2) graph.add_edge(3, 2, 1) graph.add_edge(3, 4, 6) graph.add_edge(4, 1, 3) graph.add_edge(4, 2, 4) graph.add_edge(4, 3, 9) graph.floyd_warshall() graph.show_min(1, 4) graph.show_min(0, 3)
64
from typing import Optional import numpy as np import torch from torch import nn from transformers import GPTaConfig, GPTaLMHeadModel from transformers.modeling_utils import ModuleUtilsMixin from ...configuration_utils import ConfigMixin, register_to_config from ...models import ModelMixin class _a (__magic_name__ , __magic_name__ , __magic_name__ ): '''simple docstring''' UpperCAmelCase__: str = [r'''h\.\d+\.attn\.bias''', r'''h\.\d+\.attn\.masked_bias'''] @register_to_config def __init__( self , A__ , A__ , A__ = None , A__ = 5_0257 , A__ = 1024 , A__ = 768 , A__ = 12 , A__ = 12 , A__ = None , A__ = "gelu_new" , A__ = 0.1 , A__ = 0.1 , A__ = 0.1 , A__ = 1e-5 , A__ = 0.0_2 , A__ = True , A__ = True , A__ = False , A__ = False , ): super().__init__() A__ : Union[str, Any] = prefix_length if prefix_inner_dim != n_embd and prefix_hidden_dim is None: raise ValueError( F"""`prefix_hidden_dim` cannot be `None` when `prefix_inner_dim`: {prefix_hidden_dim} and""" F""" `n_embd`: {n_embd} are not equal.""" ) A__ : str = prefix_inner_dim A__ : Optional[Any] = prefix_hidden_dim A__ : Tuple = ( nn.Linear(self.prefix_inner_dim , self.prefix_hidden_dim ) if self.prefix_hidden_dim is not None else nn.Identity() ) A__ : int = ( nn.Linear(self.prefix_hidden_dim , A__ ) if self.prefix_hidden_dim is not None else nn.Identity() ) A__ : Tuple = GPTaConfig( vocab_size=A__ , n_positions=A__ , n_embd=A__ , n_layer=A__ , n_head=A__ , n_inner=A__ , activation_function=A__ , resid_pdrop=A__ , embd_pdrop=A__ , attn_pdrop=A__ , layer_norm_epsilon=A__ , initializer_range=A__ , scale_attn_weights=A__ , use_cache=A__ , scale_attn_by_inverse_layer_idx=A__ , reorder_and_upcast_attn=A__ , ) A__ : int = GPTaLMHeadModel(A__ ) def __A ( self , A__ , A__ , A__ = None , A__ = None , ): A__ : List[str] = self.transformer.transformer.wte(A__ ) A__ : int = self.encode_prefix(A__ ) A__ : int = self.decode_prefix(A__ ) A__ : Optional[Any] = torch.cat((prefix_embeds, embedding_text) , dim=1 ) if labels is not None: A__ : Any = self.get_dummy_token(input_ids.shape[0] , input_ids.device ) A__ : List[Any] = torch.cat((dummy_token, input_ids) , dim=1 ) A__ : List[str] = self.transformer(inputs_embeds=A__ , labels=A__ , attention_mask=A__ ) if self.prefix_hidden_dim is not None: return out, hidden else: return out def __A ( self , A__ , A__ ): return torch.zeros(A__ , self.prefix_length , dtype=torch.intaa , device=A__ ) def __A ( self , A__ ): return self.encode_prefix(A__ ) @torch.no_grad() def __A ( self , A__ , A__ , A__ ): A__ : List[Any] = torch.split(A__ , 1 , dim=0 ) A__ : Optional[int] = [] A__ : str = [] for feature in features: A__ : Dict = self.decode_prefix(feature.to(A__ ) ) # back to the clip feature # Only support beam search for now A__ , A__ : Union[str, Any] = self.generate_beam( input_embeds=A__ , device=A__ , eos_token_id=A__ ) generated_tokens.append(output_tokens[0] ) generated_seq_lengths.append(seq_lengths[0] ) A__ : int = torch.stack(A__ ) A__ : List[Any] = torch.stack(A__ ) return generated_tokens, generated_seq_lengths @torch.no_grad() def __A ( self , A__=None , A__=None , A__=None , A__ = 5 , A__ = 67 , A__ = 1.0 , A__ = None , ): A__ : Any = eos_token_id A__ : Any = None A__ : Optional[int] = None A__ : Optional[Any] = torch.ones(A__ , device=A__ , dtype=torch.int ) A__ : Any = torch.zeros(A__ , device=A__ , dtype=torch.bool ) if input_embeds is not None: A__ : Dict = input_embeds else: A__ : str = self.transformer.transformer.wte(A__ ) for i in range(A__ ): A__ : Dict = self.transformer(inputs_embeds=A__ ) A__ : str = outputs.logits A__ : Union[str, Any] = logits[:, -1, :] / (temperature if temperature > 0 else 1.0) A__ : Any = logits.softmax(-1 ).log() if scores is None: A__ , A__ : Optional[int] = logits.topk(A__ , -1 ) A__ : List[Any] = generated.expand(A__ , *generated.shape[1:] ) A__ , A__ : List[Any] = next_tokens.permute(1 , 0 ), scores.squeeze(0 ) if tokens is None: A__ : Optional[Any] = next_tokens else: A__ : List[Any] = tokens.expand(A__ , *tokens.shape[1:] ) A__ : int = torch.cat((tokens, next_tokens) , dim=1 ) else: A__ : Optional[int] = -float(np.inf ) A__ : List[Any] = 0 A__ : str = scores[:, None] + logits seq_lengths[~is_stopped] += 1 A__ : Dict = scores_sum / seq_lengths[:, None] A__ , A__ : List[Any] = scores_sum_average.view(-1 ).topk(A__ , -1 ) A__ : Tuple = next_tokens // scores_sum.shape[1] A__ : Optional[Any] = seq_lengths[next_tokens_source] A__ : List[str] = next_tokens % scores_sum.shape[1] A__ : Optional[int] = next_tokens.unsqueeze(1 ) A__ : int = tokens[next_tokens_source] A__ : List[Any] = torch.cat((tokens, next_tokens) , dim=1 ) A__ : str = generated[next_tokens_source] A__ : Optional[Any] = scores_sum_average * seq_lengths A__ : Union[str, Any] = is_stopped[next_tokens_source] A__ : str = self.transformer.transformer.wte(next_tokens.squeeze() ).view(generated.shape[0] , 1 , -1 ) A__ : Optional[int] = torch.cat((generated, next_token_embed) , dim=1 ) A__ : List[str] = is_stopped + next_tokens.eq(A__ ).squeeze() if is_stopped.all(): break A__ : Dict = scores / seq_lengths A__ : Dict = scores.argsort(descending=A__ ) # tokens tensors are already padded to max_seq_length A__ : Union[str, Any] = [tokens[i] for i in order] A__ : Any = torch.stack(A__ , dim=0 ) A__ : Dict = torch.tensor([seq_lengths[i] for i in order] , dtype=seq_lengths.dtype ) return output_texts, seq_lengths
64
1
from typing import TYPE_CHECKING from ...utils import OptionalDependencyNotAvailable, _LazyModule, is_sentencepiece_available A_ : Optional[Any] = {} try: if not is_sentencepiece_available(): raise OptionalDependencyNotAvailable() except OptionalDependencyNotAvailable: pass else: A_ : str = ['BartphoTokenizer'] if TYPE_CHECKING: try: if not is_sentencepiece_available(): raise OptionalDependencyNotAvailable() except OptionalDependencyNotAvailable: pass else: from .tokenization_bartpho import BartphoTokenizer else: import sys A_ : Optional[int] = _LazyModule(__name__, globals()['__file__'], _import_structure, module_spec=__spec__)
64
import io import itertools import json from dataclasses import dataclass from typing import Optional import pyarrow as pa import pyarrow.json as paj import datasets from datasets.table import table_cast from datasets.utils.file_utils import readline A_ : Tuple = datasets.utils.logging.get_logger(__name__) @dataclass class _a (datasets.BuilderConfig ): '''simple docstring''' UpperCAmelCase__: Optional[datasets.Features] = None UpperCAmelCase__: str = "utf-8" UpperCAmelCase__: Optional[str] = None UpperCAmelCase__: Optional[str] = None UpperCAmelCase__: bool = True # deprecated UpperCAmelCase__: Optional[int] = None # deprecated UpperCAmelCase__: int = 10 << 20 # 10MB UpperCAmelCase__: Optional[bool] = None class _a (datasets.ArrowBasedBuilder ): '''simple docstring''' UpperCAmelCase__: List[str] = JsonConfig def __A ( self ): if self.config.block_size is not None: logger.warning("""The JSON loader parameter `block_size` is deprecated. Please use `chunksize` instead""" ) A__ : Union[str, Any] = self.config.block_size if self.config.use_threads is not True: logger.warning( """The JSON loader parameter `use_threads` is deprecated and doesn't have any effect anymore.""" ) if self.config.newlines_in_values is not None: raise ValueError("""The JSON loader parameter `newlines_in_values` is no longer supported""" ) return datasets.DatasetInfo(features=self.config.features ) def __A ( self , A__ ): if not self.config.data_files: raise ValueError(F"""At least one data file must be specified, but got data_files={self.config.data_files}""" ) A__ : int = dl_manager.download_and_extract(self.config.data_files ) if isinstance(A__ , (str, list, tuple) ): A__ : Optional[Any] = data_files if isinstance(A__ , A__ ): A__ : List[str] = [files] A__ : int = [dl_manager.iter_files(A__ ) for file in files] return [datasets.SplitGenerator(name=datasets.Split.TRAIN , gen_kwargs={"""files""": files} )] A__ : List[str] = [] for split_name, files in data_files.items(): if isinstance(A__ , A__ ): A__ : Optional[int] = [files] A__ : Optional[int] = [dl_manager.iter_files(A__ ) for file in files] splits.append(datasets.SplitGenerator(name=A__ , gen_kwargs={"""files""": files} ) ) return splits def __A ( self , A__ ): if self.config.features is not None: # adding missing columns for column_name in set(self.config.features ) - set(pa_table.column_names ): A__ : Optional[Any] = self.config.features.arrow_schema.field(A__ ).type A__ : str = pa_table.append_column(A__ , pa.array([None] * len(A__ ) , type=A__ ) ) # more expensive cast to support nested structures with keys in a different order # allows str <-> int/float or str to Audio for example A__ : Optional[int] = table_cast(A__ , self.config.features.arrow_schema ) return pa_table def __A ( self , A__ ): for file_idx, file in enumerate(itertools.chain.from_iterable(A__ ) ): # If the file is one json object and if we need to look at the list of items in one specific field if self.config.field is not None: with open(A__ , encoding=self.config.encoding , errors=self.config.encoding_errors ) as f: A__ : Optional[Any] = json.load(A__ ) # We keep only the field we are interested in A__ : Optional[int] = dataset[self.config.field] # We accept two format: a list of dicts or a dict of lists if isinstance(A__ , (list, tuple) ): A__ : Union[str, Any] = set().union(*[row.keys() for row in dataset] ) A__ : Any = {col: [row.get(A__ ) for row in dataset] for col in keys} else: A__ : Any = dataset A__ : Any = pa.Table.from_pydict(A__ ) yield file_idx, self._cast_table(A__ ) # If the file has one json object per line else: with open(A__ , """rb""" ) as f: A__ : List[str] = 0 # Use block_size equal to the chunk size divided by 32 to leverage multithreading # Set a default minimum value of 16kB if the chunk size is really small A__ : List[str] = max(self.config.chunksize // 32 , 16 << 10 ) A__ : Any = ( self.config.encoding_errors if self.config.encoding_errors is not None else """strict""" ) while True: A__ : Dict = f.read(self.config.chunksize ) if not batch: break # Finish current line try: batch += f.readline() except (AttributeError, io.UnsupportedOperation): batch += readline(A__ ) # PyArrow only accepts utf-8 encoded bytes if self.config.encoding != "utf-8": A__ : List[Any] = batch.decode(self.config.encoding , errors=A__ ).encode("""utf-8""" ) try: while True: try: A__ : str = paj.read_json( io.BytesIO(A__ ) , read_options=paj.ReadOptions(block_size=A__ ) ) break except (pa.ArrowInvalid, pa.ArrowNotImplementedError) as e: if ( isinstance(A__ , pa.ArrowInvalid ) and "straddling" not in str(A__ ) or block_size > len(A__ ) ): raise else: # Increase the block size in case it was too small. # The block size will be reset for the next file. logger.debug( F"""Batch of {len(A__ )} bytes couldn't be parsed with block_size={block_size}. Retrying with block_size={block_size * 2}.""" ) block_size *= 2 except pa.ArrowInvalid as e: try: with open( A__ , encoding=self.config.encoding , errors=self.config.encoding_errors ) as f: A__ : Optional[Any] = json.load(A__ ) except json.JSONDecodeError: logger.error(F"""Failed to read file '{file}' with error {type(A__ )}: {e}""" ) raise e # If possible, parse the file as a list of json objects and exit the loop if isinstance(A__ , A__ ): # list is the only sequence type supported in JSON try: A__ : str = set().union(*[row.keys() for row in dataset] ) A__ : List[str] = {col: [row.get(A__ ) for row in dataset] for col in keys} A__ : int = pa.Table.from_pydict(A__ ) except (pa.ArrowInvalid, AttributeError) as e: logger.error(F"""Failed to read file '{file}' with error {type(A__ )}: {e}""" ) raise ValueError(F"""Not able to read records in the JSON file at {file}.""" ) from None yield file_idx, self._cast_table(A__ ) break else: logger.error(F"""Failed to read file '{file}' with error {type(A__ )}: {e}""" ) raise ValueError( F"""Not able to read records in the JSON file at {file}. """ F"""You should probably indicate the field of the JSON file containing your records. """ F"""This JSON file contain the following fields: {str(list(dataset.keys() ) )}. """ F"""Select the correct one and provide it as `field='XXX'` to the dataset loading method. """ ) from None # Uncomment for debugging (will print the Arrow table size and elements) # logger.warning(f"pa_table: {pa_table} num rows: {pa_table.num_rows}") # logger.warning('\n'.join(str(pa_table.slice(i, 1).to_pydict()) for i in range(pa_table.num_rows))) yield (file_idx, batch_idx), self._cast_table(A__ ) batch_idx += 1
64
1
import unittest from transformers import PegasusTokenizer, PegasusTokenizerFast from transformers.testing_utils import get_tests_dir, require_sentencepiece, require_tokenizers, require_torch, slow from transformers.utils import cached_property from ...test_tokenization_common import TokenizerTesterMixin A_ : Union[str, Any] = get_tests_dir('fixtures/test_sentencepiece_no_bos.model') @require_sentencepiece @require_tokenizers class _a (__magic_name__ , unittest.TestCase ): '''simple docstring''' UpperCAmelCase__: Any = PegasusTokenizer UpperCAmelCase__: Optional[int] = PegasusTokenizerFast UpperCAmelCase__: Optional[Any] = True UpperCAmelCase__: Any = True def __A ( self ): super().setUp() # We have a SentencePiece fixture for testing A__ : List[str] = PegasusTokenizer(A__ ) tokenizer.save_pretrained(self.tmpdirname ) @cached_property def __A ( self ): return PegasusTokenizer.from_pretrained("""google/pegasus-large""" ) def __A ( self , **A__ ): return PegasusTokenizer.from_pretrained(self.tmpdirname , **A__ ) def __A ( self , A__ ): return ("This is a test", "This is a test") def __A ( self ): A__ : Dict = """</s>""" A__ : Union[str, Any] = 1 self.assertEqual(self.get_tokenizer()._convert_token_to_id(A__ ) , A__ ) self.assertEqual(self.get_tokenizer()._convert_id_to_token(A__ ) , A__ ) def __A ( self ): A__ : Union[str, Any] = list(self.get_tokenizer().get_vocab().keys() ) self.assertEqual(vocab_keys[0] , """<pad>""" ) self.assertEqual(vocab_keys[1] , """</s>""" ) self.assertEqual(vocab_keys[-1] , """v""" ) self.assertEqual(len(A__ ) , 1103 ) def __A ( self ): self.assertEqual(self.get_tokenizer().vocab_size , 1103 ) def __A ( self ): A__ : Union[str, Any] = self.rust_tokenizer_class.from_pretrained(self.tmpdirname ) A__ : int = self.tokenizer_class.from_pretrained(self.tmpdirname ) A__ : int = ( """Let's see which <unk> is the better <unk_token_11> one <mask_1> It seems like this <mask_2> was important""" """ </s> <pad> <pad> <pad>""" ) A__ : Any = rust_tokenizer([raw_input_str] , return_tensors=A__ , add_special_tokens=A__ ).input_ids[0] A__ : Any = py_tokenizer([raw_input_str] , return_tensors=A__ , add_special_tokens=A__ ).input_ids[0] self.assertListEqual(A__ , A__ ) def __A ( self ): A__ : Tuple = self._large_tokenizer # <mask_1> masks whole sentence while <mask_2> masks single word A__ : List[str] = """<mask_1> To ensure a <mask_2> flow of bank resolutions.""" A__ : int = [2, 413, 615, 114, 3, 1971, 113, 1679, 1_0710, 107, 1] A__ : Tuple = tokenizer([raw_input_str] , return_tensors=A__ ).input_ids[0] self.assertListEqual(A__ , A__ ) def __A ( self ): A__ : Optional[Any] = self._large_tokenizer # The tracebacks for the following asserts are **better** without messages or self.assertEqual assert tokenizer.vocab_size == 9_6103 assert tokenizer.pad_token_id == 0 assert tokenizer.eos_token_id == 1 assert tokenizer.offset == 103 assert tokenizer.unk_token_id == tokenizer.offset + 2 == 105 assert tokenizer.unk_token == "<unk>" assert tokenizer.model_max_length == 1024 A__ : str = """To ensure a smooth flow of bank resolutions.""" A__ : Union[str, Any] = [413, 615, 114, 2291, 1971, 113, 1679, 1_0710, 107, 1] A__ : int = tokenizer([raw_input_str] , return_tensors=A__ ).input_ids[0] self.assertListEqual(A__ , A__ ) assert tokenizer.convert_ids_to_tokens([0, 1, 2, 3] ) == ["<pad>", "</s>", "<mask_1>", "<mask_2>"] @require_torch def __A ( self ): A__ : List[str] = ["""This is going to be way too long.""" * 150, """short example"""] A__ : Optional[int] = ["""not super long but more than 5 tokens""", """tiny"""] A__ : Dict = self._large_tokenizer(A__ , padding=A__ , truncation=A__ , return_tensors="""pt""" ) A__ : List[Any] = self._large_tokenizer( text_target=A__ , max_length=5 , padding=A__ , truncation=A__ , return_tensors="""pt""" ) assert batch.input_ids.shape == (2, 1024) assert batch.attention_mask.shape == (2, 1024) assert targets["input_ids"].shape == (2, 5) assert len(A__ ) == 2 # input_ids, attention_mask. @slow def __A ( self ): # fmt: off A__ : Any = {"""input_ids""": [[3_8979, 143, 1_8485, 606, 130, 2_6669, 8_7686, 121, 5_4189, 1129, 111, 2_6669, 8_7686, 121, 9114, 1_4787, 121, 1_3249, 158, 592, 956, 121, 1_4621, 3_1576, 143, 6_2613, 108, 9688, 930, 4_3430, 1_1562, 6_2613, 304, 108, 1_1443, 897, 108, 9314, 1_7415, 6_3399, 108, 1_1443, 7614, 1_8316, 118, 4284, 7148, 1_2430, 143, 1400, 2_5703, 158, 111, 4284, 7148, 1_1772, 143, 2_1297, 1064, 158, 122, 204, 3506, 1754, 1133, 1_4787, 1581, 115, 3_3224, 4482, 111, 1355, 110, 2_9173, 317, 5_0833, 108, 2_0147, 9_4665, 111, 7_7198, 107, 1], [110, 6_2613, 117, 638, 112, 1133, 121, 2_0098, 1355, 7_9050, 1_3872, 135, 1596, 5_3541, 1352, 141, 1_3039, 5542, 124, 302, 518, 111, 268, 2956, 115, 149, 4427, 107, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0], [139, 1235, 2799, 1_8289, 1_7780, 204, 109, 9474, 1296, 107, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0]], """attention_mask""": [[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1], [1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0], [1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0]]} # noqa: E501 # fmt: on self.tokenizer_integration_test_util( expected_encoding=A__ , model_name="""google/bigbird-pegasus-large-arxiv""" , revision="""ba85d0851d708441f91440d509690f1ab6353415""" , ) @require_sentencepiece @require_tokenizers class _a (__magic_name__ , unittest.TestCase ): '''simple docstring''' UpperCAmelCase__: Union[str, Any] = PegasusTokenizer UpperCAmelCase__: List[Any] = PegasusTokenizerFast UpperCAmelCase__: int = True UpperCAmelCase__: Optional[Any] = True def __A ( self ): super().setUp() # We have a SentencePiece fixture for testing A__ : Optional[int] = PegasusTokenizer(A__ , offset=0 , mask_token_sent=A__ , mask_token="""[MASK]""" ) tokenizer.save_pretrained(self.tmpdirname ) @cached_property def __A ( self ): return PegasusTokenizer.from_pretrained("""google/bigbird-pegasus-large-arxiv""" ) def __A ( self , **A__ ): return PegasusTokenizer.from_pretrained(self.tmpdirname , **A__ ) def __A ( self , A__ ): return ("This is a test", "This is a test") def __A ( self ): A__ : int = self.rust_tokenizer_class.from_pretrained(self.tmpdirname ) A__ : Optional[int] = self.tokenizer_class.from_pretrained(self.tmpdirname ) A__ : Optional[int] = ( """Let's see which <unk> is the better <unk_token> one [MASK] It seems like this [MASK] was important </s>""" """ <pad> <pad> <pad>""" ) A__ : List[Any] = rust_tokenizer([raw_input_str] , return_tensors=A__ , add_special_tokens=A__ ).input_ids[0] A__ : List[str] = py_tokenizer([raw_input_str] , return_tensors=A__ , add_special_tokens=A__ ).input_ids[0] self.assertListEqual(A__ , A__ ) @require_torch def __A ( self ): A__ : int = ["""This is going to be way too long.""" * 1000, """short example"""] A__ : Any = ["""not super long but more than 5 tokens""", """tiny"""] A__ : Dict = self._large_tokenizer(A__ , padding=A__ , truncation=A__ , return_tensors="""pt""" ) A__ : Any = self._large_tokenizer( text_target=A__ , max_length=5 , padding=A__ , truncation=A__ , return_tensors="""pt""" ) assert batch.input_ids.shape == (2, 4096) assert batch.attention_mask.shape == (2, 4096) assert targets["input_ids"].shape == (2, 5) assert len(A__ ) == 2 # input_ids, attention_mask. def __A ( self ): A__ : Dict = ( """This is an example string that is used to test the original TF implementation against the HF""" """ implementation""" ) A__ : List[Any] = self._large_tokenizer(A__ ).input_ids self.assertListEqual( A__ , [182, 117, 142, 587, 4211, 120, 117, 263, 112, 804, 109, 856, 2_5016, 3137, 464, 109, 2_6955, 3137, 1] , )
64
import os try: from .build_directory_md import good_file_paths except ImportError: from build_directory_md import good_file_paths # type: ignore A_ : Dict = list(good_file_paths()) assert filepaths, "good_file_paths() failed!" A_ : Optional[Any] = [file for file in filepaths if file != file.lower()] if upper_files: print(f'''{len(upper_files)} files contain uppercase characters:''') print('\n'.join(upper_files) + '\n') A_ : Tuple = [file for file in filepaths if ' ' in file] if space_files: print(f'''{len(space_files)} files contain space characters:''') print('\n'.join(space_files) + '\n') A_ : Any = [file for file in filepaths if '-' in file] if hyphen_files: print(f'''{len(hyphen_files)} files contain hyphen characters:''') print('\n'.join(hyphen_files) + '\n') A_ : List[str] = [file for file in filepaths if os.sep not in file] if nodir_files: print(f'''{len(nodir_files)} files are not in a directory:''') print('\n'.join(nodir_files) + '\n') A_ : Any = len(upper_files + space_files + hyphen_files + nodir_files) if bad_files: import sys sys.exit(bad_files)
64
1
from __future__ import annotations from typing import Any def UpperCamelCase (lowercase_: list[Any] ) -> None: create_state_space_tree(lowercase_ , [] , 0 ) def UpperCamelCase (lowercase_: list[Any] , lowercase_: list[Any] , lowercase_: int ) -> None: if index == len(lowercase_ ): print(lowercase_ ) return create_state_space_tree(lowercase_ , lowercase_ , index + 1 ) current_subsequence.append(sequence[index] ) create_state_space_tree(lowercase_ , lowercase_ , index + 1 ) current_subsequence.pop() if __name__ == "__main__": A_ : list[Any] = [3, 1, 2, 4] generate_all_subsequences(seq) seq.clear() seq.extend(['A', 'B', 'C']) generate_all_subsequences(seq)
64
import inspect import warnings from typing import Any, Dict, Optional, Union from packaging import version def UpperCamelCase (*lowercase_: Optional[int] , lowercase_: Optional[Union[Dict, Any]] = None , lowercase_: Dict=True , lowercase_: Tuple=2 ) -> Dict: from .. import __version__ A__ : Dict = take_from A__ : str = () if not isinstance(args[0] , lowercase_ ): A__ : int = (args,) for attribute, version_name, message in args: if version.parse(version.parse(lowercase_ ).base_version ) >= version.parse(lowercase_ ): raise ValueError( f"""The deprecation tuple {(attribute, version_name, message)} should be removed since diffusers'""" f""" version {__version__} is >= {version_name}""" ) A__ : Any = None if isinstance(lowercase_ , lowercase_ ) and attribute in deprecated_kwargs: values += (deprecated_kwargs.pop(lowercase_ ),) A__ : List[str] = f"""The `{attribute}` argument is deprecated and will be removed in version {version_name}.""" elif hasattr(lowercase_ , lowercase_ ): values += (getattr(lowercase_ , lowercase_ ),) A__ : Optional[Any] = f"""The `{attribute}` attribute is deprecated and will be removed in version {version_name}.""" elif deprecated_kwargs is None: A__ : int = f"""`{attribute}` is deprecated and will be removed in version {version_name}.""" if warning is not None: A__ : int = warning + """ """ if standard_warn else """""" warnings.warn(warning + message , lowercase_ , stacklevel=lowercase_ ) if isinstance(lowercase_ , lowercase_ ) and len(lowercase_ ) > 0: A__ : Union[str, Any] = inspect.getouterframes(inspect.currentframe() )[1] A__ : Optional[Any] = call_frame.filename A__ : Optional[int] = call_frame.lineno A__ : Any = call_frame.function A__ , A__ : List[str] = next(iter(deprecated_kwargs.items() ) ) raise TypeError(f"""{function} in {filename} line {line_number-1} got an unexpected keyword argument `{key}`""" ) if len(lowercase_ ) == 0: return elif len(lowercase_ ) == 1: return values[0] return values
64
1
import warnings from ...processing_utils import ProcessorMixin from ...tokenization_utils_base import BatchEncoding class _a (__magic_name__ ): '''simple docstring''' UpperCAmelCase__: Any = ['''image_processor''', '''tokenizer'''] UpperCAmelCase__: Tuple = '''CLIPImageProcessor''' UpperCAmelCase__: Union[str, Any] = ('''CLIPTokenizer''', '''CLIPTokenizerFast''') def __init__( self , A__=None , A__=None , **A__ ): A__ : List[Any] = None if "feature_extractor" in kwargs: warnings.warn( """The `feature_extractor` argument is deprecated and will be removed in v5, use `image_processor`""" """ instead.""" , A__ , ) A__ : Any = kwargs.pop("""feature_extractor""" ) A__ : Union[str, Any] = image_processor if image_processor is not None else feature_extractor if image_processor is None: raise ValueError("""You need to specify an `image_processor`.""" ) if tokenizer is None: raise ValueError("""You need to specify a `tokenizer`.""" ) super().__init__(A__ , A__ ) def __call__( self , A__=None , A__=None , A__=None , **A__ ): if text is None and images is None: raise ValueError("""You have to specify either text or images. Both cannot be none.""" ) if text is not None: A__ : Tuple = self.tokenizer(A__ , return_tensors=A__ , **A__ ) if images is not None: A__ : List[Any] = self.image_processor(A__ , return_tensors=A__ , **A__ ) if text is not None and images is not None: A__ : List[str] = image_features.pixel_values return encoding elif text is not None: return encoding else: return BatchEncoding(data=dict(**A__ ) , tensor_type=A__ ) def __A ( self , *A__ , **A__ ): return self.tokenizer.batch_decode(*A__ , **A__ ) def __A ( self , *A__ , **A__ ): return self.tokenizer.decode(*A__ , **A__ ) @property def __A ( self ): A__ : Optional[int] = self.tokenizer.model_input_names A__ : Optional[Any] = self.image_processor.model_input_names return list(dict.fromkeys(tokenizer_input_names + image_processor_input_names ) ) @property def __A ( self ): warnings.warn( """`feature_extractor_class` is deprecated and will be removed in v5. Use `image_processor_class` instead.""" , A__ , ) return self.image_processor_class @property def __A ( self ): warnings.warn( """`feature_extractor` is deprecated and will be removed in v5. Use `image_processor` instead.""" , A__ , ) return self.image_processor
64
import argparse import json import os import tensorstore as ts import torch from flax import serialization from flax.traverse_util import flatten_dict, unflatten_dict from tensorflow.io import gfile from transformers.modeling_utils import dtype_byte_size from transformers.models.switch_transformers.convert_switch_transformers_original_flax_checkpoint_to_pytorch import ( rename_keys, ) from transformers.utils import WEIGHTS_INDEX_NAME, WEIGHTS_NAME from transformers.utils.hub import convert_file_size_to_int def UpperCamelCase (lowercase_: List[str] , lowercase_: str ) -> Optional[Any]: if flax_key_tuple[-1] == "kernel" and flax_tensor.ndim == 3: # expert layer A__ : Union[str, Any] = flax_key_tuple[:-1] + ("""weight""",) A__ : Optional[int] = torch.permute(lowercase_ , (0, 2, 1) ) elif flax_key_tuple[-1] == "kernel" and ".".join(lowercase_ ): # linear layer A__ : Optional[Any] = flax_key_tuple[:-1] + ("""weight""",) A__ : int = flax_tensor.T elif flax_key_tuple[-1] in ["scale", "embedding"]: A__ : Optional[int] = flax_key_tuple[:-1] + ("""weight""",) return flax_key_tuple, flax_tensor def UpperCamelCase (lowercase_: Tuple , lowercase_: Optional[int] , lowercase_: str ) -> Union[str, Any]: if "metadata" in layer: A__ : Tuple = layer.split("""metadata""" ) A__ : Optional[Any] = """""".join(split_layer[0] )[:-1] A__ : Optional[Any] = [tuple(("""metadata""" + split_layer[1]).split("""/""" ) )] elif "kvstore" in layer: A__ : str = layer.split("""kvstore""" ) A__ : int = """""".join(split_layer[0] )[:-1] A__ : Optional[int] = [tuple(("""kvstore""" + split_layer[1]).split("""/""" ) )] else: A__ : Any = layer.split("""/""" ) A__ : int = """/""".join(split_layer[:-1] ) A__ : str = (split_layer[-1],) if "kvstore/path" in layer: A__ : Dict = f"""{switch_checkpoint_path}/{checkpoint_info[layer]}""" elif "kvstore/driver" in layer: A__ : Optional[int] = """file""" else: A__ : str = checkpoint_info[layer] return curr_real_layer_name, split_layer, content def UpperCamelCase (lowercase_: str , lowercase_: List[Any] ) -> int: A__ : int = rename_keys(lowercase_ ) A__ : Any = {} for k, v in current_block.items(): A__ : Dict = v A__ : str = new_current_block torch.save(lowercase_ , lowercase_ ) def UpperCamelCase (lowercase_: Dict , lowercase_: Optional[Any] , lowercase_: Optional[Any] , lowercase_: Optional[int] , lowercase_: str = WEIGHTS_NAME ) -> Tuple: A__ : Optional[int] = convert_file_size_to_int(lowercase_ ) A__ : List[Any] = [] A__ : int = {} A__ : List[str] = 0 A__ : Any = 0 os.makedirs(lowercase_ , exist_ok=lowercase_ ) with gfile.GFile(switch_checkpoint_path + """/checkpoint""" , """rb""" ) as fp: A__ : Optional[Any] = serialization.msgpack_restore(fp.read() )["""optimizer"""]["""target"""] A__ : Dict = flatten_dict(lowercase_ , sep="""/""" ) A__ : Any = {} for layer in checkpoint_info.keys(): A__ , A__ , A__ : Union[str, Any] = get_key_and_tensorstore_dict( lowercase_ , lowercase_ , lowercase_ ) if curr_real_layer_name in all_layers: A__ : Optional[int] = content else: A__ : List[Any] = {split_layer[-1]: content} for key in all_layers.keys(): # open tensorstore file A__ : Optional[Any] = ts.open(unflatten_dict(all_layers[key] ) ).result().read().result() A__ : List[Any] = torch.tensor(lowercase_ ) A__ : List[Any] = raw_weights.numel() * dtype_byte_size(raw_weights.dtype ) # use the renaming pattern from the small conversion scripts A__ , A__ : Any = rename_base_flax_keys(tuple(key.split("""/""" ) ) , lowercase_ ) A__ : Any = """/""".join(lowercase_ ) # If this weight is going to tip up over the maximal size, we split. if current_block_size + weight_size > max_shard_size: A__ : List[Any] = os.path.join( lowercase_ , weights_name.replace(""".bin""" , f"""-{len(lowercase_ )+1:05d}-of-???.bin""" ) ) rename_and_save_block(lowercase_ , lowercase_ ) sharded_state_dicts.append(current_block.keys() ) del current_block A__ : Any = {} A__ : str = 0 A__ : List[str] = raw_weights.to(getattr(lowercase_ , lowercase_ ) ) current_block_size += weight_size total_size += weight_size # Add the last block A__ : Union[str, Any] = os.path.join(lowercase_ , weights_name.replace(""".bin""" , f"""-{len(lowercase_ )+1:05d}-of-???.bin""" ) ) rename_and_save_block(lowercase_ , lowercase_ ) sharded_state_dicts.append(current_block.keys() ) # If we only have one shard, we return it if len(lowercase_ ) == 1: return {weights_name: sharded_state_dicts[0]}, None # Otherwise, let's build the index A__ : str = {} A__ : Any = {} for idx, shard in enumerate(lowercase_ ): A__ : Any = weights_name.replace( """.bin""" , f"""-{idx+1:05d}-of-{len(lowercase_ ):05d}.bin""" ) # len(sharded_state_dicts):05d} A__ : Dict = os.path.join(lowercase_ , weights_name.replace(""".bin""" , f"""-{idx+1:05d}-of-???.bin""" ) ) os.rename(lowercase_ , os.path.join(lowercase_ , lowercase_ ) ) A__ : str = shard for key in shard: A__ : Any = shard_file # Add the metadata A__ : Tuple = {"""total_size""": total_size} A__ : Union[str, Any] = {"""metadata""": metadata, """weight_map""": weight_map} with open(os.path.join(lowercase_ , lowercase_ ) , """w""" , encoding="""utf-8""" ) as f: A__ : Dict = json.dumps(lowercase_ , indent=2 , sort_keys=lowercase_ ) + """\n""" f.write(lowercase_ ) return metadata, index if __name__ == "__main__": A_ : Optional[int] = argparse.ArgumentParser() # Required parameters parser.add_argument( '--switch_t5x_checkpoint_path', default='/mnt/disks/disk_switch/original_checkpoints/switch-xxl-128/checkpoint_634600', type=str, required=False, help='Path to a directory containing a folder per layer. Follows the original Google format.', ) parser.add_argument('--max_shard_size', default='10GB', required=False, help='Max shard size') parser.add_argument('--dtype', default='bfloat16', type=str, required=False, help='dtype of the saved model') parser.add_argument( '--pytorch_dump_folder_path', default='/mnt/disks/disk_switch/original_checkpoints/switch-xxl-128-converted', type=str, required=False, help='Path to the output pytorch model.', ) A_ : Dict = parser.parse_args() shard_on_the_fly( args.switch_tax_checkpoint_path, args.pytorch_dump_folder_path, args.max_shard_size, args.dtype, ) def UpperCamelCase () -> int: from transformers import SwitchTransformersConfig, SwitchTransformersForConditionalGeneration, TaTokenizer A__ : str = SwitchTransformersConfig.from_pretrained("""google/switch-base-8""" ) config.save_pretrained("""/home/arthur_huggingface_co/transformers/switch_converted""" ) A__ : str = SwitchTransformersForConditionalGeneration.from_pretrained( """/home/arthur_huggingface_co/transformers/switch_converted""" , device_map="""auto""" ) A__ : Tuple = TaTokenizer.from_pretrained("""t5-small""" ) A__ : Dict = """A <extra_id_0> walks into a bar a orders a <extra_id_1> with <extra_id_2> pinch of <extra_id_3>.""" A__ : Union[str, Any] = tokenizer(lowercase_ , return_tensors="""pt""" ).input_ids A__ : Tuple = model.generate(lowercase_ , decoder_start_token_id=0 ) print(tokenizer.decode(out[0] ) )
64
1
import fire from transformers import AutoConfig, AutoModelForSeqaSeqLM, AutoTokenizer def UpperCamelCase (lowercase_: str , lowercase_: str , **lowercase_: str ) -> Optional[int]: A__ : str = AutoConfig.from_pretrained(lowercase_ , **lowercase_ ) A__ : List[str] = AutoModelForSeqaSeqLM.from_config(lowercase_ ) model.save_pretrained(lowercase_ ) AutoTokenizer.from_pretrained(lowercase_ ).save_pretrained(lowercase_ ) return model if __name__ == "__main__": fire.Fire(save_randomly_initialized_version)
64
from typing import TYPE_CHECKING from ...utils import OptionalDependencyNotAvailable, _LazyModule, is_sentencepiece_available A_ : Optional[Any] = {} try: if not is_sentencepiece_available(): raise OptionalDependencyNotAvailable() except OptionalDependencyNotAvailable: pass else: A_ : str = ['BartphoTokenizer'] if TYPE_CHECKING: try: if not is_sentencepiece_available(): raise OptionalDependencyNotAvailable() except OptionalDependencyNotAvailable: pass else: from .tokenization_bartpho import BartphoTokenizer else: import sys A_ : Optional[int] = _LazyModule(__name__, globals()['__file__'], _import_structure, module_spec=__spec__)
64
1
import inspect import os import unittest import torch import accelerate from accelerate import Accelerator from accelerate.test_utils import execute_subprocess_async, require_multi_gpu from accelerate.utils import patch_environment class _a (unittest.TestCase ): '''simple docstring''' def __A ( self ): A__ : int = inspect.getfile(accelerate.test_utils ) A__ : str = os.path.sep.join(mod_file.split(os.path.sep )[:-1] + ["""scripts""", """test_script.py"""] ) A__ : List[Any] = os.path.sep.join( mod_file.split(os.path.sep )[:-1] + ["""scripts""", """test_distributed_data_loop.py"""] ) A__ : Dict = os.path.sep.join(mod_file.split(os.path.sep )[:-1] + ["""scripts""", """test_ops.py"""] ) @require_multi_gpu def __A ( self ): print(F"""Found {torch.cuda.device_count()} devices.""" ) A__ : Optional[int] = ["""torchrun""", F"""--nproc_per_node={torch.cuda.device_count()}""", self.test_file_path] with patch_environment(omp_num_threads=1 ): execute_subprocess_async(A__ , env=os.environ.copy() ) @require_multi_gpu def __A ( self ): print(F"""Found {torch.cuda.device_count()} devices.""" ) A__ : Any = ["""torchrun""", F"""--nproc_per_node={torch.cuda.device_count()}""", self.operation_file_path] print(F"""Command: {cmd}""" ) with patch_environment(omp_num_threads=1 ): execute_subprocess_async(A__ , env=os.environ.copy() ) @require_multi_gpu def __A ( self ): A__ : List[Any] = ["""torchrun""", F"""--nproc_per_node={torch.cuda.device_count()}""", inspect.getfile(self.__class__ )] with patch_environment(omp_num_threads=1 ): execute_subprocess_async(A__ , env=os.environ.copy() ) @require_multi_gpu def __A ( self ): print(F"""Found {torch.cuda.device_count()} devices, using 2 devices only""" ) A__ : Optional[int] = ["""torchrun""", F"""--nproc_per_node={torch.cuda.device_count()}""", self.data_loop_file_path] with patch_environment(omp_num_threads=1 , cuda_visible_devices="""0,1""" ): execute_subprocess_async(A__ , env=os.environ.copy() ) if __name__ == "__main__": A_ : Optional[Any] = Accelerator() A_ : Tuple = (accelerator.state.process_index + 2, 10) A_ : Any = torch.randint(0, 10, shape).to(accelerator.device) A_ : Any = '' A_ : List[str] = accelerator.pad_across_processes(tensor) if tensora.shape[0] != accelerator.state.num_processes + 1: error_msg += f"Found shape {tensora.shape} but should have {accelerator.state.num_processes + 1} at dim 0." if not torch.equal(tensora[: accelerator.state.process_index + 2], tensor): error_msg += "Tensors have different values." if not torch.all(tensora[accelerator.state.process_index + 2 :] == 0): error_msg += "Padding was not done with the right value (0)." A_ : Tuple = accelerator.pad_across_processes(tensor, pad_first=True) if tensora.shape[0] != accelerator.state.num_processes + 1: error_msg += f"Found shape {tensora.shape} but should have {accelerator.state.num_processes + 1} at dim 0." A_ : List[str] = accelerator.state.num_processes - accelerator.state.process_index - 1 if not torch.equal(tensora[index:], tensor): error_msg += "Tensors have different values." if not torch.all(tensora[:index] == 0): error_msg += "Padding was not done with the right value (0)." # Raise error at the end to make sure we don't stop at the first failure. if len(error_msg) > 0: raise ValueError(error_msg)
64
import argparse import hashlib import os import urllib import warnings import torch from torch import nn from tqdm import tqdm from transformers import WhisperConfig, WhisperForConditionalGeneration A_ : Dict = { 'tiny.en': 'https://openaipublic.azureedge.net/main/whisper/models/d3dd57d32accea0b295c96e26691aa14d8822fac7d9d27d5dc00b4ca2826dd03/tiny.en.pt', 'tiny': 'https://openaipublic.azureedge.net/main/whisper/models/65147644a518d12f04e32d6f3b26facc3f8dd46e5390956a9424a650c0ce22b9/tiny.pt', 'base.en': 'https://openaipublic.azureedge.net/main/whisper/models/25a8566e1d0c1e2231d1c762132cd20e0f96a85d16145c3a00adf5d1ac670ead/base.en.pt', 'base': 'https://openaipublic.azureedge.net/main/whisper/models/ed3a0b6b1c0edf879ad9b11b1af5a0e6ab5db9205f891f668f8b0e6c6326e34e/base.pt', 'small.en': 'https://openaipublic.azureedge.net/main/whisper/models/f953ad0fd29cacd07d5a9eda5624af0f6bcf2258be67c92b79389873d91e0872/small.en.pt', 'small': 'https://openaipublic.azureedge.net/main/whisper/models/9ecf779972d90ba49c06d968637d720dd632c55bbf19d441fb42bf17a411e794/small.pt', 'medium.en': 'https://openaipublic.azureedge.net/main/whisper/models/d7440d1dc186f76616474e0ff0b3b6b879abc9d1a4926b7adfa41db2d497ab4f/medium.en.pt', 'medium': 'https://openaipublic.azureedge.net/main/whisper/models/345ae4da62f9b3d59415adc60127b97c714f32e89e936602e85993674d08dcb1/medium.pt', 'large': 'https://openaipublic.azureedge.net/main/whisper/models/e4b87e7e0bf463eb8e6956e646f1e277e901512310def2c24bf0e11bd3c28e9a/large.pt', 'large-v2': 'https://openaipublic.azureedge.net/main/whisper/models/81f7c96c852ee8fc832187b0132e569d6c3065a3252ed18e56effd0b6a73e524/large-v2.pt', } def UpperCamelCase (lowercase_: Optional[Any] ) -> Optional[int]: A__ : List[Any] = ["""layers""", """blocks"""] for k in ignore_keys: state_dict.pop(lowercase_ , lowercase_ ) A_ : Any = { 'blocks': 'layers', 'mlp.0': 'fc1', 'mlp.2': 'fc2', 'mlp_ln': 'final_layer_norm', '.attn.query': '.self_attn.q_proj', '.attn.key': '.self_attn.k_proj', '.attn.value': '.self_attn.v_proj', '.attn_ln': '.self_attn_layer_norm', '.attn.out': '.self_attn.out_proj', '.cross_attn.query': '.encoder_attn.q_proj', '.cross_attn.key': '.encoder_attn.k_proj', '.cross_attn.value': '.encoder_attn.v_proj', '.cross_attn_ln': '.encoder_attn_layer_norm', '.cross_attn.out': '.encoder_attn.out_proj', 'decoder.ln.': 'decoder.layer_norm.', 'encoder.ln.': 'encoder.layer_norm.', 'token_embedding': 'embed_tokens', 'encoder.positional_embedding': 'encoder.embed_positions.weight', 'decoder.positional_embedding': 'decoder.embed_positions.weight', 'ln_post': 'layer_norm', } def UpperCamelCase (lowercase_: str ) -> Any: A__ : Dict = list(s_dict.keys() ) for key in keys: A__ : List[str] = key for k, v in WHISPER_MAPPING.items(): if k in key: A__ : List[Any] = new_key.replace(lowercase_ , lowercase_ ) print(f"""{key} -> {new_key}""" ) A__ : Tuple = s_dict.pop(lowercase_ ) return s_dict def UpperCamelCase (lowercase_: Tuple ) -> Optional[int]: A__ , A__ : Any = emb.weight.shape A__ : str = nn.Linear(lowercase_ , lowercase_ , bias=lowercase_ ) A__ : Union[str, Any] = emb.weight.data return lin_layer def UpperCamelCase (lowercase_: str , lowercase_: str ) -> bytes: os.makedirs(lowercase_ , exist_ok=lowercase_ ) A__ : Tuple = os.path.basename(lowercase_ ) A__ : int = url.split("""/""" )[-2] A__ : Dict = os.path.join(lowercase_ , lowercase_ ) if os.path.exists(lowercase_ ) and not os.path.isfile(lowercase_ ): raise RuntimeError(f"""{download_target} exists and is not a regular file""" ) if os.path.isfile(lowercase_ ): A__ : Optional[Any] = open(lowercase_ , """rb""" ).read() if hashlib.shaaaa(lowercase_ ).hexdigest() == expected_shaaaa: return model_bytes else: warnings.warn(f"""{download_target} exists, but the SHA256 checksum does not match; re-downloading the file""" ) with urllib.request.urlopen(lowercase_ ) as source, open(lowercase_ , """wb""" ) as output: with tqdm( total=int(source.info().get("""Content-Length""" ) ) , ncols=80 , unit="""iB""" , unit_scale=lowercase_ , unit_divisor=1024 ) as loop: while True: A__ : Any = source.read(8192 ) if not buffer: break output.write(lowercase_ ) loop.update(len(lowercase_ ) ) A__ : Dict = open(lowercase_ , """rb""" ).read() if hashlib.shaaaa(lowercase_ ).hexdigest() != expected_shaaaa: raise RuntimeError( """Model has been downloaded but the SHA256 checksum does not not match. Please retry loading the model.""" ) return model_bytes def UpperCamelCase (lowercase_: List[Any] , lowercase_: Tuple ) -> Optional[Any]: if ".pt" not in checkpoint_path: A__ : Tuple = _download(_MODELS[checkpoint_path] ) else: A__ : Optional[int] = torch.load(lowercase_ , map_location="""cpu""" ) A__ : str = original_checkpoint["""dims"""] A__ : List[Any] = original_checkpoint["""model_state_dict"""] A__ : Optional[Any] = state_dict["""decoder.token_embedding.weight"""] remove_ignore_keys_(lowercase_ ) rename_keys(lowercase_ ) A__ : List[str] = True A__ : Optional[Any] = state_dict["""decoder.layers.0.fc1.weight"""].shape[0] A__ : List[Any] = WhisperConfig( vocab_size=dimensions["""n_vocab"""] , encoder_ffn_dim=lowercase_ , decoder_ffn_dim=lowercase_ , num_mel_bins=dimensions["""n_mels"""] , d_model=dimensions["""n_audio_state"""] , max_target_positions=dimensions["""n_text_ctx"""] , encoder_layers=dimensions["""n_audio_layer"""] , encoder_attention_heads=dimensions["""n_audio_head"""] , decoder_layers=dimensions["""n_text_layer"""] , decoder_attention_heads=dimensions["""n_text_state"""] , max_source_positions=dimensions["""n_audio_ctx"""] , ) A__ : Optional[Any] = WhisperForConditionalGeneration(lowercase_ ) A__ , A__ : List[Any] = model.model.load_state_dict(lowercase_ , strict=lowercase_ ) if len(lowercase_ ) > 0 and not set(lowercase_ ) <= { "encoder.embed_positions.weights", "decoder.embed_positions.weights", }: raise ValueError( """Only `encoder.embed_positions.weights` and `decoder.embed_positions.weights` are allowed to be missing,""" f""" but all the following weights are missing {missing}""" ) if tie_embeds: A__ : Any = make_linear_from_emb(model.model.decoder.embed_tokens ) else: A__ : str = proj_out_weights model.save_pretrained(lowercase_ ) if __name__ == "__main__": A_ : Any = argparse.ArgumentParser() # # Required parameters parser.add_argument('--checkpoint_path', type=str, help='Patht to the downloaded checkpoints') parser.add_argument('--pytorch_dump_folder_path', default=None, type=str, help='Path to the output PyTorch model.') A_ : Tuple = parser.parse_args() convert_openai_whisper_to_tfms(args.checkpoint_path, args.pytorch_dump_folder_path)
64
1
import torch from torch import nn from transformers import CLIPPreTrainedModel, CLIPVisionModel from ...models.attention import BasicTransformerBlock from ...utils import logging A_ : Optional[Any] = logging.get_logger(__name__) # pylint: disable=invalid-name class _a (__magic_name__ ): '''simple docstring''' def __init__( self , A__ , A__=768 ): super().__init__(A__ ) A__ : Tuple = proj_size A__ : int = CLIPVisionModel(A__ ) A__ : Union[str, Any] = PaintByExampleMapper(A__ ) A__ : Any = nn.LayerNorm(config.hidden_size ) A__ : List[str] = nn.Linear(config.hidden_size , self.proj_size ) # uncondition for scaling A__ : str = nn.Parameter(torch.randn((1, 1, self.proj_size) ) ) def __A ( self , A__ , A__=False ): A__ : Optional[int] = self.model(pixel_values=A__ ) A__ : Dict = clip_output.pooler_output A__ : Union[str, Any] = self.mapper(latent_states[:, None] ) A__ : Optional[Any] = self.final_layer_norm(A__ ) A__ : Any = self.proj_out(A__ ) if return_uncond_vector: return latent_states, self.uncond_vector return latent_states class _a (nn.Module ): '''simple docstring''' def __init__( self , A__ ): super().__init__() A__ : Optional[int] = (config.num_hidden_layers + 1) // 5 A__ : List[Any] = config.hidden_size A__ : Optional[int] = 1 A__ : Tuple = nn.ModuleList( [ BasicTransformerBlock(A__ , A__ , A__ , activation_fn="""gelu""" , attention_bias=A__ ) for _ in range(A__ ) ] ) def __A ( self , A__ ): for block in self.blocks: A__ : str = block(A__ ) return hidden_states
64
import unittest import numpy as np import torch from transformers import CLIPTextConfig, CLIPTextModel, CLIPTokenizer from diffusers import ( AutoencoderKL, DDIMScheduler, DPMSolverMultistepScheduler, TextToVideoSDPipeline, UNetaDConditionModel, ) from diffusers.utils import is_xformers_available, load_numpy, skip_mps, slow, torch_device from diffusers.utils.testing_utils import enable_full_determinism from ..pipeline_params import TEXT_TO_IMAGE_BATCH_PARAMS, TEXT_TO_IMAGE_PARAMS from ..test_pipelines_common import PipelineTesterMixin enable_full_determinism() @skip_mps class _a (__magic_name__ , unittest.TestCase ): '''simple docstring''' UpperCAmelCase__: Any = TextToVideoSDPipeline UpperCAmelCase__: Any = TEXT_TO_IMAGE_PARAMS UpperCAmelCase__: Optional[Any] = TEXT_TO_IMAGE_BATCH_PARAMS # No `output_type`. UpperCAmelCase__: Optional[int] = frozenset( [ '''num_inference_steps''', '''generator''', '''latents''', '''return_dict''', '''callback''', '''callback_steps''', ] ) def __A ( self ): torch.manual_seed(0 ) A__ : Optional[int] = UNetaDConditionModel( block_out_channels=(32, 64, 64, 64) , layers_per_block=2 , sample_size=32 , in_channels=4 , out_channels=4 , down_block_types=("""CrossAttnDownBlock3D""", """CrossAttnDownBlock3D""", """CrossAttnDownBlock3D""", """DownBlock3D""") , up_block_types=("""UpBlock3D""", """CrossAttnUpBlock3D""", """CrossAttnUpBlock3D""", """CrossAttnUpBlock3D""") , cross_attention_dim=32 , attention_head_dim=4 , ) A__ : Optional[int] = DDIMScheduler( beta_start=0.0_0_0_8_5 , beta_end=0.0_1_2 , beta_schedule="""scaled_linear""" , clip_sample=A__ , set_alpha_to_one=A__ , ) torch.manual_seed(0 ) A__ : Union[str, Any] = AutoencoderKL( block_out_channels=[32, 64] , in_channels=3 , out_channels=3 , down_block_types=["""DownEncoderBlock2D""", """DownEncoderBlock2D"""] , up_block_types=["""UpDecoderBlock2D""", """UpDecoderBlock2D"""] , latent_channels=4 , sample_size=128 , ) torch.manual_seed(0 ) A__ : Union[str, Any] = CLIPTextConfig( bos_token_id=0 , eos_token_id=2 , hidden_size=32 , intermediate_size=37 , layer_norm_eps=1e-05 , num_attention_heads=4 , num_hidden_layers=5 , pad_token_id=1 , vocab_size=1000 , hidden_act="""gelu""" , projection_dim=512 , ) A__ : Union[str, Any] = CLIPTextModel(A__ ) A__ : Tuple = CLIPTokenizer.from_pretrained("""hf-internal-testing/tiny-random-clip""" ) A__ : Dict = { """unet""": unet, """scheduler""": scheduler, """vae""": vae, """text_encoder""": text_encoder, """tokenizer""": tokenizer, } return components def __A ( self , A__ , A__=0 ): if str(A__ ).startswith("""mps""" ): A__ : Tuple = torch.manual_seed(A__ ) else: A__ : List[str] = torch.Generator(device=A__ ).manual_seed(A__ ) A__ : List[str] = { """prompt""": """A painting of a squirrel eating a burger""", """generator""": generator, """num_inference_steps""": 2, """guidance_scale""": 6.0, """output_type""": """pt""", } return inputs def __A ( self ): A__ : List[str] = """cpu""" # ensure determinism for the device-dependent torch.Generator A__ : Union[str, Any] = self.get_dummy_components() A__ : Union[str, Any] = TextToVideoSDPipeline(**A__ ) A__ : int = sd_pipe.to(A__ ) sd_pipe.set_progress_bar_config(disable=A__ ) A__ : int = self.get_dummy_inputs(A__ ) A__ : int = """np""" A__ : Any = sd_pipe(**A__ ).frames A__ : Dict = frames[0][-3:, -3:, -1] assert frames[0].shape == (64, 64, 3) A__ : Optional[Any] = np.array([1_5_8.0, 1_6_0.0, 1_5_3.0, 1_2_5.0, 1_0_0.0, 1_2_1.0, 1_1_1.0, 9_3.0, 1_1_3.0] ) assert np.abs(image_slice.flatten() - expected_slice ).max() < 1e-2 def __A ( self ): self._test_attention_slicing_forward_pass(test_mean_pixel_difference=A__ , expected_max_diff=3e-3 ) @unittest.skipIf( torch_device != """cuda""" or not is_xformers_available() , reason="""XFormers attention is only available with CUDA and `xformers` installed""" , ) def __A ( self ): self._test_xformers_attention_forwardGenerator_pass(test_mean_pixel_difference=A__ , expected_max_diff=1e-2 ) @unittest.skip(reason="""Batching needs to be properly figured out first for this pipeline.""" ) def __A ( self ): pass @unittest.skip(reason="""Batching needs to be properly figured out first for this pipeline.""" ) def __A ( self ): pass @unittest.skip(reason="""`num_images_per_prompt` argument is not supported for this pipeline.""" ) def __A ( self ): pass def __A ( self ): return super().test_progress_bar() @slow @skip_mps class _a (unittest.TestCase ): '''simple docstring''' def __A ( self ): A__ : Union[str, Any] = load_numpy( """https://huggingface.co/datasets/hf-internal-testing/diffusers-images/resolve/main/text_to_video/video.npy""" ) A__ : Tuple = TextToVideoSDPipeline.from_pretrained("""damo-vilab/text-to-video-ms-1.7b""" ) A__ : Any = DPMSolverMultistepScheduler.from_config(pipe.scheduler.config ) A__ : int = pipe.to("""cuda""" ) A__ : Optional[Any] = """Spiderman is surfing""" A__ : List[str] = torch.Generator(device="""cpu""" ).manual_seed(0 ) A__ : Optional[Any] = pipe(A__ , generator=A__ , num_inference_steps=25 , output_type="""pt""" ).frames A__ : Dict = video_frames.cpu().numpy() assert np.abs(expected_video - video ).mean() < 5e-2 def __A ( self ): A__ : List[Any] = load_numpy( """https://huggingface.co/datasets/hf-internal-testing/diffusers-images/resolve/main/text_to_video/video_2step.npy""" ) A__ : Optional[int] = TextToVideoSDPipeline.from_pretrained("""damo-vilab/text-to-video-ms-1.7b""" ) A__ : List[str] = pipe.to("""cuda""" ) A__ : Dict = """Spiderman is surfing""" A__ : Union[str, Any] = torch.Generator(device="""cpu""" ).manual_seed(0 ) A__ : Optional[int] = pipe(A__ , generator=A__ , num_inference_steps=2 , output_type="""pt""" ).frames A__ : Optional[int] = video_frames.cpu().numpy() assert np.abs(expected_video - video ).mean() < 5e-2
64
1
A_ : Optional[int] = { 'Pillow': 'Pillow', 'accelerate': 'accelerate>=0.11.0', 'compel': 'compel==0.1.8', 'black': 'black~=23.1', 'datasets': 'datasets', 'filelock': 'filelock', 'flax': 'flax>=0.4.1', 'hf-doc-builder': 'hf-doc-builder>=0.3.0', 'huggingface-hub': 'huggingface-hub>=0.13.2', 'requests-mock': 'requests-mock==1.10.0', 'importlib_metadata': 'importlib_metadata', 'invisible-watermark': 'invisible-watermark', 'isort': 'isort>=5.5.4', 'jax': 'jax>=0.2.8,!=0.3.2', 'jaxlib': 'jaxlib>=0.1.65', 'Jinja2': 'Jinja2', 'k-diffusion': 'k-diffusion>=0.0.12', 'torchsde': 'torchsde', 'note_seq': 'note_seq', 'librosa': 'librosa', 'numpy': 'numpy', 'omegaconf': 'omegaconf', 'parameterized': 'parameterized', 'protobuf': 'protobuf>=3.20.3,<4', 'pytest': 'pytest', 'pytest-timeout': 'pytest-timeout', 'pytest-xdist': 'pytest-xdist', 'ruff': 'ruff>=0.0.241', 'safetensors': 'safetensors', 'sentencepiece': 'sentencepiece>=0.1.91,!=0.1.92', 'scipy': 'scipy', 'onnx': 'onnx', 'regex': 'regex!=2019.12.17', 'requests': 'requests', 'tensorboard': 'tensorboard', 'torch': 'torch>=1.4', 'torchvision': 'torchvision', 'transformers': 'transformers>=4.25.1', 'urllib3': 'urllib3<=2.0.0', }
64
def UpperCamelCase (lowercase_: int ) -> int: if not isinstance(lowercase_ , lowercase_ ): raise TypeError("""Input value must be an 'int' type""" ) A__ : int = 0 while number: position += 1 number >>= 1 return position if __name__ == "__main__": import doctest doctest.testmod()
64
1
import copy import inspect import unittest from transformers import PretrainedConfig, SwiftFormerConfig from transformers.testing_utils import ( require_torch, require_vision, slow, torch_device, ) from transformers.utils import cached_property, is_torch_available, is_vision_available from ...test_configuration_common import ConfigTester from ...test_modeling_common import ModelTesterMixin, floats_tensor, ids_tensor from ...test_pipeline_mixin import PipelineTesterMixin if is_torch_available(): import torch from torch import nn from transformers import SwiftFormerForImageClassification, SwiftFormerModel from transformers.models.swiftformer.modeling_swiftformer import SWIFTFORMER_PRETRAINED_MODEL_ARCHIVE_LIST if is_vision_available(): from PIL import Image from transformers import ViTImageProcessor class _a : '''simple docstring''' def __init__( self , A__ , A__=13 , A__=3 , A__=True , A__=True , A__=0.1 , A__=0.1 , A__=224 , A__=1000 , A__=[3, 3, 6, 4] , A__=[48, 56, 112, 220] , ): A__ : Tuple = parent A__ : str = batch_size A__ : int = num_channels A__ : Union[str, Any] = is_training A__ : Tuple = use_labels A__ : List[str] = hidden_dropout_prob A__ : Any = attention_probs_dropout_prob A__ : Dict = num_labels A__ : List[Any] = image_size A__ : List[Any] = layer_depths A__ : str = embed_dims def __A ( self ): A__ : Tuple = floats_tensor([self.batch_size, self.num_channels, self.image_size, self.image_size] ) A__ : int = None if self.use_labels: A__ : int = ids_tensor([self.batch_size] , self.num_labels ) A__ : Tuple = self.get_config() return config, pixel_values, labels def __A ( self ): return SwiftFormerConfig( depths=self.layer_depths , embed_dims=self.embed_dims , mlp_ratio=4 , downsamples=[True, True, True, True] , hidden_act="""gelu""" , num_labels=self.num_labels , down_patch_size=3 , down_stride=2 , down_pad=1 , drop_rate=0.0 , drop_path_rate=0.0 , use_layer_scale=A__ , layer_scale_init_value=1e-5 , ) def __A ( self , A__ , A__ , A__ ): A__ : List[Any] = SwiftFormerModel(config=A__ ) model.to(A__ ) model.eval() A__ : List[Any] = model(A__ ) self.parent.assertEqual(result.last_hidden_state.shape , (self.batch_size, self.embed_dims[-1], 7, 7) ) def __A ( self , A__ , A__ , A__ ): A__ : Tuple = self.num_labels A__ : Tuple = SwiftFormerForImageClassification(A__ ) model.to(A__ ) model.eval() A__ : int = model(A__ , labels=A__ ) self.parent.assertEqual(result.logits.shape , (self.batch_size, self.num_labels) ) A__ : List[str] = SwiftFormerForImageClassification(A__ ) model.to(A__ ) model.eval() A__ : Any = floats_tensor([self.batch_size, self.num_channels, self.image_size, self.image_size] ) A__ : Optional[int] = model(A__ ) self.parent.assertEqual(result.logits.shape , (self.batch_size, self.num_labels) ) def __A ( self ): ((A__) , (A__) , (A__)) : Tuple = self.prepare_config_and_inputs() A__ : Optional[int] = {"""pixel_values""": pixel_values} return config, inputs_dict @require_torch class _a (__magic_name__ , __magic_name__ , unittest.TestCase ): '''simple docstring''' UpperCAmelCase__: Dict = (SwiftFormerModel, SwiftFormerForImageClassification) if is_torch_available() else () UpperCAmelCase__: List[Any] = ( {'''feature-extraction''': SwiftFormerModel, '''image-classification''': SwiftFormerForImageClassification} if is_torch_available() else {} ) UpperCAmelCase__: Dict = False UpperCAmelCase__: Union[str, Any] = False UpperCAmelCase__: Any = False UpperCAmelCase__: int = False UpperCAmelCase__: Any = False def __A ( self ): A__ : List[str] = SwiftFormerModelTester(self ) A__ : int = ConfigTester( self , config_class=A__ , has_text_modality=A__ , hidden_size=37 , num_attention_heads=12 , num_hidden_layers=12 , ) def __A ( self ): self.config_tester.run_common_tests() @unittest.skip(reason="""SwiftFormer does not use inputs_embeds""" ) def __A ( self ): pass def __A ( self ): A__ , A__ : Dict = self.model_tester.prepare_config_and_inputs_for_common() for model_class in self.all_model_classes: A__ : List[str] = model_class(A__ ) A__ : Dict = model.get_output_embeddings() self.assertTrue(x is None or isinstance(A__ , nn.Linear ) ) def __A ( self ): A__ , A__ : Tuple = self.model_tester.prepare_config_and_inputs_for_common() for model_class in self.all_model_classes: A__ : Any = model_class(A__ ) A__ : Tuple = inspect.signature(model.forward ) # signature.parameters is an OrderedDict => so arg_names order is deterministic A__ : Dict = [*signature.parameters.keys()] A__ : List[str] = ["""pixel_values"""] self.assertListEqual(arg_names[:1] , A__ ) def __A ( self ): A__ : Dict = self.model_tester.prepare_config_and_inputs() self.model_tester.create_and_check_model(*A__ ) def __A ( self ): A__ : Any = self.model_tester.prepare_config_and_inputs() self.model_tester.create_and_check_for_image_classification(*A__ ) @slow def __A ( self ): for model_name in SWIFTFORMER_PRETRAINED_MODEL_ARCHIVE_LIST[:1]: A__ : Tuple = SwiftFormerModel.from_pretrained(A__ ) self.assertIsNotNone(A__ ) @unittest.skip(reason="""SwiftFormer does not output attentions""" ) def __A ( self ): pass def __A ( self ): def check_hidden_states_output(A__ , A__ , A__ ): A__ : Any = model_class(A__ ) model.to(A__ ) model.eval() with torch.no_grad(): A__ : Any = model(**self._prepare_for_class(A__ , A__ ) ) A__ : str = outputs.hidden_states A__ : Tuple = 8 self.assertEqual(len(A__ ) , A__ ) # TODO # SwiftFormer's feature maps are of shape (batch_size, embed_dims, height, width) # with the width and height being successively divided by 2, after every 2 blocks for i in range(len(A__ ) ): self.assertEqual( hidden_states[i].shape , torch.Size( [ self.model_tester.batch_size, self.model_tester.embed_dims[i // 2], (self.model_tester.image_size // 4) // 2 ** (i // 2), (self.model_tester.image_size // 4) // 2 ** (i // 2), ] ) , ) A__ , A__ : Dict = self.model_tester.prepare_config_and_inputs_for_common() for model_class in self.all_model_classes: A__ : Any = True check_hidden_states_output(A__ , A__ , A__ ) # check that output_hidden_states also work using config del inputs_dict["output_hidden_states"] A__ : List[Any] = True check_hidden_states_output(A__ , A__ , A__ ) def __A ( self ): def _config_zero_init(A__ ): A__ : str = copy.deepcopy(A__ ) for key in configs_no_init.__dict__.keys(): if "_range" in key or "_std" in key or "initializer_factor" in key or "layer_scale" in key: setattr(A__ , A__ , 1e-10 ) if isinstance(getattr(A__ , A__ , A__ ) , A__ ): A__ : Optional[Any] = _config_zero_init(getattr(A__ , A__ ) ) setattr(A__ , A__ , A__ ) return configs_no_init A__ , A__ : str = self.model_tester.prepare_config_and_inputs_for_common() A__ : List[Any] = _config_zero_init(A__ ) for model_class in self.all_model_classes: A__ : str = model_class(config=A__ ) for name, param in model.named_parameters(): if param.requires_grad: self.assertIn( ((param.data.mean() * 1e9) / 1e9).round().item() , [0.0, 1.0] , msg=F"""Parameter {name} of model {model_class} seems not properly initialized""" , ) @unittest.skip("""Will be fixed soon by reducing the size of the model used for common tests.""" ) def __A ( self ): pass def UpperCamelCase () -> Optional[Any]: A__ : Union[str, Any] = Image.open("""./tests/fixtures/tests_samples/COCO/000000039769.png""" ) return image @require_torch @require_vision class _a (unittest.TestCase ): '''simple docstring''' @cached_property def __A ( self ): return ViTImageProcessor.from_pretrained("""MBZUAI/swiftformer-xs""" ) if is_vision_available() else None @slow def __A ( self ): A__ : List[Any] = SwiftFormerForImageClassification.from_pretrained("""MBZUAI/swiftformer-xs""" ).to(A__ ) A__ : Any = self.default_image_processor A__ : List[Any] = prepare_img() A__ : Optional[Any] = image_processor(images=A__ , return_tensors="""pt""" ).to(A__ ) # forward pass with torch.no_grad(): A__ : Dict = model(**A__ ) # verify the logits A__ : Union[str, Any] = torch.Size((1, 1000) ) self.assertEqual(outputs.logits.shape , A__ ) A__ : Any = torch.tensor([[-2.1_703e00, 2.1_107e00, -2.0_811e00]] ).to(A__ ) self.assertTrue(torch.allclose(outputs.logits[0, :3] , A__ , atol=1e-4 ) )
64
from pathlib import Path import cva import numpy as np from matplotlib import pyplot as plt def UpperCamelCase (lowercase_: np.ndarray , lowercase_: np.ndarray , lowercase_: np.ndarray , lowercase_: int , lowercase_: int ) -> np.ndarray: A__ : Any = cva.getAffineTransform(lowercase_ , lowercase_ ) return cva.warpAffine(lowercase_ , lowercase_ , (rows, cols) ) if __name__ == "__main__": # read original image A_ : List[Any] = cva.imread( str(Path(__file__).resolve().parent.parent / 'image_data' / 'lena.jpg') ) # turn image in gray scale value A_ : List[Any] = cva.cvtColor(image, cva.COLOR_BGR2GRAY) # get image shape A_ , A_ : Optional[Any] = gray_img.shape # set different points to rotate image A_ : str = np.array([[50, 50], [200, 50], [50, 200]], np.floataa) A_ : Dict = np.array([[10, 100], [200, 50], [100, 250]], np.floataa) A_ : Optional[int] = np.array([[50, 50], [150, 50], [120, 200]], np.floataa) A_ : Optional[int] = np.array([[10, 100], [80, 50], [180, 250]], np.floataa) # add all rotated images in a list A_ : Dict = [ gray_img, get_rotation(gray_img, ptsa, ptsa, img_rows, img_cols), get_rotation(gray_img, ptsa, ptsa, img_rows, img_cols), get_rotation(gray_img, ptsa, ptsa, img_rows, img_cols), ] # plot different image rotations A_ : Union[str, Any] = plt.figure(1) A_ : Union[str, Any] = ['Original', 'Rotation 1', 'Rotation 2', 'Rotation 3'] for i, image in enumerate(images): plt.subplot(2, 2, i + 1), plt.imshow(image, 'gray') plt.title(titles[i]) plt.axis('off') plt.subplots_adjust(left=0.0, bottom=0.05, right=1.0, top=0.95) plt.show()
64
1
import gc import tempfile import unittest import numpy as np import torch from diffusers import VersatileDiffusionTextToImagePipeline from diffusers.utils.testing_utils import nightly, require_torch_gpu, torch_device A_ : Dict = False class _a (unittest.TestCase ): '''simple docstring''' pass @nightly @require_torch_gpu class _a (unittest.TestCase ): '''simple docstring''' def __A ( self ): # clean up the VRAM after each test super().tearDown() gc.collect() torch.cuda.empty_cache() def __A ( self ): A__ : Union[str, Any] = VersatileDiffusionTextToImagePipeline.from_pretrained("""shi-labs/versatile-diffusion""" ) # remove text_unet pipe.remove_unused_weights() pipe.to(A__ ) pipe.set_progress_bar_config(disable=A__ ) A__ : str = """A painting of a squirrel eating a burger """ A__ : int = torch.manual_seed(0 ) A__ : int = pipe( prompt=A__ , generator=A__ , guidance_scale=7.5 , num_inference_steps=2 , output_type="""numpy""" ).images with tempfile.TemporaryDirectory() as tmpdirname: pipe.save_pretrained(A__ ) A__ : List[str] = VersatileDiffusionTextToImagePipeline.from_pretrained(A__ ) pipe.to(A__ ) pipe.set_progress_bar_config(disable=A__ ) A__ : List[str] = generator.manual_seed(0 ) A__ : List[Any] = pipe( prompt=A__ , generator=A__ , guidance_scale=7.5 , num_inference_steps=2 , output_type="""numpy""" ).images assert np.abs(image - new_image ).sum() < 1e-5, "Models don't have the same forward pass" def __A ( self ): A__ : List[Any] = VersatileDiffusionTextToImagePipeline.from_pretrained( """shi-labs/versatile-diffusion""" , torch_dtype=torch.floataa ) pipe.to(A__ ) pipe.set_progress_bar_config(disable=A__ ) A__ : List[str] = """A painting of a squirrel eating a burger """ A__ : List[str] = torch.manual_seed(0 ) A__ : Optional[int] = pipe( prompt=A__ , generator=A__ , guidance_scale=7.5 , num_inference_steps=50 , output_type="""numpy""" ).images A__ : Union[str, Any] = image[0, 253:256, 253:256, -1] assert image.shape == (1, 512, 512, 3) A__ : Optional[Any] = np.array([0.3_3_6_7, 0.3_1_6_9, 0.2_6_5_6, 0.3_8_7_0, 0.4_7_9_0, 0.3_7_9_6, 0.4_0_0_9, 0.4_8_7_8, 0.4_7_7_8] ) assert np.abs(image_slice.flatten() - expected_slice ).max() < 1e-2
64
import os import tempfile import unittest from pathlib import Path from transformers import AutoConfig, is_torch_available from transformers.testing_utils import require_torch, torch_device if is_torch_available(): from transformers import PyTorchBenchmark, PyTorchBenchmarkArguments @require_torch class _a (unittest.TestCase ): '''simple docstring''' def __A ( self , A__ ): for model_result in results.values(): for batch_size, sequence_length in zip(model_result["""bs"""] , model_result["""ss"""] ): A__ : str = model_result["""result"""][batch_size][sequence_length] self.assertIsNotNone(A__ ) def __A ( self ): A__ : Dict = """sshleifer/tiny-gpt2""" A__ : Tuple = PyTorchBenchmarkArguments( models=[MODEL_ID] , training=A__ , inference=A__ , sequence_lengths=[8] , batch_sizes=[1] , multi_process=A__ , ) A__ : int = PyTorchBenchmark(A__ ) A__ : List[Any] = benchmark.run() self.check_results_dict_not_empty(results.time_inference_result ) self.check_results_dict_not_empty(results.memory_inference_result ) def __A ( self ): A__ : Dict = """sgugger/tiny-distilbert-classification""" A__ : Dict = PyTorchBenchmarkArguments( models=[MODEL_ID] , training=A__ , inference=A__ , sequence_lengths=[8] , batch_sizes=[1] , multi_process=A__ , only_pretrain_model=A__ , ) A__ : str = PyTorchBenchmark(A__ ) A__ : List[str] = benchmark.run() self.check_results_dict_not_empty(results.time_inference_result ) self.check_results_dict_not_empty(results.memory_inference_result ) def __A ( self ): A__ : Any = """sshleifer/tiny-gpt2""" A__ : List[Any] = PyTorchBenchmarkArguments( models=[MODEL_ID] , training=A__ , inference=A__ , torchscript=A__ , sequence_lengths=[8] , batch_sizes=[1] , multi_process=A__ , ) A__ : Tuple = PyTorchBenchmark(A__ ) A__ : str = benchmark.run() self.check_results_dict_not_empty(results.time_inference_result ) self.check_results_dict_not_empty(results.memory_inference_result ) @unittest.skipIf(torch_device == """cpu""" , """Cant do half precision""" ) def __A ( self ): A__ : Optional[Any] = """sshleifer/tiny-gpt2""" A__ : Optional[Any] = PyTorchBenchmarkArguments( models=[MODEL_ID] , training=A__ , inference=A__ , fpaa=A__ , sequence_lengths=[8] , batch_sizes=[1] , multi_process=A__ , ) A__ : str = PyTorchBenchmark(A__ ) A__ : Any = benchmark.run() self.check_results_dict_not_empty(results.time_inference_result ) self.check_results_dict_not_empty(results.memory_inference_result ) def __A ( self ): A__ : Optional[Any] = """sshleifer/tiny-gpt2""" A__ : Tuple = AutoConfig.from_pretrained(A__ ) # set architectures equal to `None` A__ : List[Any] = None A__ : str = PyTorchBenchmarkArguments( models=[MODEL_ID] , training=A__ , inference=A__ , sequence_lengths=[8] , batch_sizes=[1] , multi_process=A__ , ) A__ : List[str] = PyTorchBenchmark(A__ , configs=[config] ) A__ : Any = benchmark.run() self.check_results_dict_not_empty(results.time_inference_result ) self.check_results_dict_not_empty(results.memory_inference_result ) def __A ( self ): A__ : Optional[int] = """sshleifer/tiny-gpt2""" A__ : Optional[int] = PyTorchBenchmarkArguments( models=[MODEL_ID] , training=A__ , inference=A__ , sequence_lengths=[8] , batch_sizes=[1] , multi_process=A__ , ) A__ : Any = PyTorchBenchmark(A__ ) A__ : Dict = benchmark.run() self.check_results_dict_not_empty(results.time_train_result ) self.check_results_dict_not_empty(results.memory_train_result ) @unittest.skipIf(torch_device == """cpu""" , """Can't do half precision""" ) def __A ( self ): A__ : Optional[int] = """sshleifer/tiny-gpt2""" A__ : List[str] = PyTorchBenchmarkArguments( models=[MODEL_ID] , training=A__ , inference=A__ , sequence_lengths=[8] , batch_sizes=[1] , fpaa=A__ , multi_process=A__ , ) A__ : Dict = PyTorchBenchmark(A__ ) A__ : Optional[int] = benchmark.run() self.check_results_dict_not_empty(results.time_train_result ) self.check_results_dict_not_empty(results.memory_train_result ) def __A ( self ): A__ : int = """sshleifer/tiny-gpt2""" A__ : Optional[int] = AutoConfig.from_pretrained(A__ ) A__ : str = PyTorchBenchmarkArguments( models=[MODEL_ID] , training=A__ , inference=A__ , sequence_lengths=[8] , batch_sizes=[1] , multi_process=A__ , ) A__ : int = PyTorchBenchmark(A__ , configs=[config] ) A__ : Tuple = benchmark.run() self.check_results_dict_not_empty(results.time_inference_result ) self.check_results_dict_not_empty(results.memory_inference_result ) def __A ( self ): A__ : List[str] = """sshleifer/tinier_bart""" A__ : List[str] = AutoConfig.from_pretrained(A__ ) A__ : List[str] = PyTorchBenchmarkArguments( models=[MODEL_ID] , training=A__ , inference=A__ , sequence_lengths=[8] , batch_sizes=[1] , multi_process=A__ , ) A__ : Union[str, Any] = PyTorchBenchmark(A__ , configs=[config] ) A__ : str = benchmark.run() self.check_results_dict_not_empty(results.time_inference_result ) self.check_results_dict_not_empty(results.memory_inference_result ) def __A ( self ): A__ : Optional[int] = """sshleifer/tiny-gpt2""" A__ : Union[str, Any] = AutoConfig.from_pretrained(A__ ) A__ : Tuple = PyTorchBenchmarkArguments( models=[MODEL_ID] , training=A__ , inference=A__ , sequence_lengths=[8] , batch_sizes=[1] , multi_process=A__ , ) A__ : int = PyTorchBenchmark(A__ , configs=[config] ) A__ : Tuple = benchmark.run() self.check_results_dict_not_empty(results.time_train_result ) self.check_results_dict_not_empty(results.memory_train_result ) def __A ( self ): A__ : Dict = """sshleifer/tinier_bart""" A__ : int = AutoConfig.from_pretrained(A__ ) A__ : Union[str, Any] = PyTorchBenchmarkArguments( models=[MODEL_ID] , training=A__ , inference=A__ , sequence_lengths=[8] , batch_sizes=[1] , multi_process=A__ , ) A__ : List[Any] = PyTorchBenchmark(A__ , configs=[config] ) A__ : Optional[Any] = benchmark.run() self.check_results_dict_not_empty(results.time_train_result ) self.check_results_dict_not_empty(results.memory_train_result ) def __A ( self ): A__ : int = """sshleifer/tiny-gpt2""" with tempfile.TemporaryDirectory() as tmp_dir: A__ : Dict = PyTorchBenchmarkArguments( models=[MODEL_ID] , training=A__ , inference=A__ , save_to_csv=A__ , sequence_lengths=[8] , batch_sizes=[1] , inference_time_csv_file=os.path.join(A__ , """inf_time.csv""" ) , train_memory_csv_file=os.path.join(A__ , """train_mem.csv""" ) , inference_memory_csv_file=os.path.join(A__ , """inf_mem.csv""" ) , train_time_csv_file=os.path.join(A__ , """train_time.csv""" ) , env_info_csv_file=os.path.join(A__ , """env.csv""" ) , multi_process=A__ , ) A__ : Optional[Any] = PyTorchBenchmark(A__ ) benchmark.run() self.assertTrue(Path(os.path.join(A__ , """inf_time.csv""" ) ).exists() ) self.assertTrue(Path(os.path.join(A__ , """train_time.csv""" ) ).exists() ) self.assertTrue(Path(os.path.join(A__ , """inf_mem.csv""" ) ).exists() ) self.assertTrue(Path(os.path.join(A__ , """train_mem.csv""" ) ).exists() ) self.assertTrue(Path(os.path.join(A__ , """env.csv""" ) ).exists() ) def __A ( self ): A__ : Optional[int] = """sshleifer/tiny-gpt2""" def _check_summary_is_not_empty(A__ ): self.assertTrue(hasattr(A__ , """sequential""" ) ) self.assertTrue(hasattr(A__ , """cumulative""" ) ) self.assertTrue(hasattr(A__ , """current""" ) ) self.assertTrue(hasattr(A__ , """total""" ) ) with tempfile.TemporaryDirectory() as tmp_dir: A__ : Dict = PyTorchBenchmarkArguments( models=[MODEL_ID] , training=A__ , inference=A__ , sequence_lengths=[8] , batch_sizes=[1] , log_filename=os.path.join(A__ , """log.txt""" ) , log_print=A__ , trace_memory_line_by_line=A__ , multi_process=A__ , ) A__ : Dict = PyTorchBenchmark(A__ ) A__ : str = benchmark.run() _check_summary_is_not_empty(result.inference_summary ) _check_summary_is_not_empty(result.train_summary ) self.assertTrue(Path(os.path.join(A__ , """log.txt""" ) ).exists() )
64
1
def UpperCamelCase (lowercase_: str , lowercase_: str ) -> list: A__ : List[str] = len(lowercase_ ) A__ : Optional[int] = [] for i in range(len(lowercase_ ) - pat_len + 1 ): A__ : int = True for j in range(lowercase_ ): if s[i + j] != pattern[j]: A__ : List[str] = False break if match_found: position.append(lowercase_ ) return position if __name__ == "__main__": assert naive_pattern_search('ABCDEFG', 'DE') == [3] print(naive_pattern_search('ABAAABCDBBABCDDEBCABC', 'ABC'))
64
# tests directory-specific settings - this file is run automatically # by pytest before any tests are run import doctest import sys import warnings from os.path import abspath, dirname, join import _pytest from transformers.testing_utils import HfDoctestModule, HfDocTestParser # allow having multiple repository checkouts and not needing to remember to rerun # 'pip install -e .[dev]' when switching between checkouts and running tests. A_ : Optional[int] = abspath(join(dirname(__file__), 'src')) sys.path.insert(1, git_repo_path) # silence FutureWarning warnings in tests since often we can't act on them until # they become normal warnings - i.e. the tests still need to test the current functionality warnings.simplefilter(action='ignore', category=FutureWarning) def UpperCamelCase (lowercase_: List[str] ) -> Any: config.addinivalue_line( """markers""" , """is_pt_tf_cross_test: mark test to run only when PT and TF interactions are tested""" ) config.addinivalue_line( """markers""" , """is_pt_flax_cross_test: mark test to run only when PT and FLAX interactions are tested""" ) config.addinivalue_line("""markers""" , """is_pipeline_test: mark test to run only when pipelines are tested""" ) config.addinivalue_line("""markers""" , """is_staging_test: mark test to run only in the staging environment""" ) config.addinivalue_line("""markers""" , """accelerate_tests: mark test that require accelerate""" ) config.addinivalue_line("""markers""" , """tool_tests: mark the tool tests that are run on their specific schedule""" ) def UpperCamelCase (lowercase_: Optional[int] ) -> Optional[Any]: from transformers.testing_utils import pytest_addoption_shared pytest_addoption_shared(lowercase_ ) def UpperCamelCase (lowercase_: List[str] ) -> Optional[Any]: from transformers.testing_utils import pytest_terminal_summary_main A__ : List[Any] = terminalreporter.config.getoption("""--make-reports""" ) if make_reports: pytest_terminal_summary_main(lowercase_ , id=lowercase_ ) def UpperCamelCase (lowercase_: Union[str, Any] , lowercase_: int ) -> List[str]: # If no tests are collected, pytest exists with code 5, which makes the CI fail. if exitstatus == 5: A__ : Tuple = 0 # Doctest custom flag to ignore output. A_ : Tuple = doctest.register_optionflag('IGNORE_RESULT') A_ : Dict = doctest.OutputChecker class _a (__magic_name__ ): '''simple docstring''' def __A ( self , A__ , A__ , A__ ): if IGNORE_RESULT & optionflags: return True return OutputChecker.check_output(self , A__ , A__ , A__ ) A_ : str = CustomOutputChecker A_ : Dict = HfDoctestModule A_ : Optional[int] = HfDocTestParser
64
1
import os def UpperCamelCase () -> Union[str, Any]: A__ : Dict = os.path.dirname(os.path.realpath(lowercase_ ) ) A__ : Union[str, Any] = os.path.join(lowercase_ , """triangle.txt""" ) with open(lowercase_ ) as f: A__ : List[Any] = f.readlines() A__ : Optional[Any] = [] for line in triangle: A__ : Optional[Any] = [] for number in line.strip().split(""" """ ): numbers_from_line.append(int(lowercase_ ) ) a.append(lowercase_ ) for i in range(1 , len(lowercase_ ) ): for j in range(len(a[i] ) ): A__ : int = a[i - 1][j] if j != len(a[i - 1] ) else 0 A__ : Optional[int] = a[i - 1][j - 1] if j > 0 else 0 a[i][j] += max(lowercase_ , lowercase_ ) return max(a[-1] ) if __name__ == "__main__": print(solution())
64
from __future__ import annotations import unittest from transformers import AutoTokenizer, PegasusConfig, is_tf_available from transformers.testing_utils import require_sentencepiece, require_tf, require_tokenizers, slow from transformers.utils import cached_property from ...test_configuration_common import ConfigTester from ...test_modeling_tf_common import TFModelTesterMixin, ids_tensor from ...test_pipeline_mixin import PipelineTesterMixin if is_tf_available(): import tensorflow as tf from transformers import TFAutoModelForSeqaSeqLM, TFPegasusForConditionalGeneration, TFPegasusModel @require_tf class _a : '''simple docstring''' UpperCAmelCase__: List[Any] = PegasusConfig UpperCAmelCase__: Optional[int] = {} UpperCAmelCase__: List[str] = '''gelu''' def __init__( self , A__ , A__=13 , A__=7 , A__=True , A__=False , A__=99 , A__=32 , A__=2 , A__=4 , A__=37 , A__=0.1 , A__=0.1 , A__=40 , A__=2 , A__=1 , A__=0 , ): A__ : Dict = parent A__ : Dict = batch_size A__ : Any = seq_length A__ : Optional[Any] = is_training A__ : int = use_labels A__ : Any = vocab_size A__ : Union[str, Any] = hidden_size A__ : Tuple = num_hidden_layers A__ : Tuple = num_attention_heads A__ : List[Any] = intermediate_size A__ : Union[str, Any] = hidden_dropout_prob A__ : Optional[Any] = attention_probs_dropout_prob A__ : List[Any] = max_position_embeddings A__ : Any = eos_token_id A__ : List[Any] = pad_token_id A__ : List[Any] = bos_token_id def __A ( self ): A__ : str = ids_tensor([self.batch_size, self.seq_length - 1] , self.vocab_size ) A__ : Dict = tf.expand_dims(tf.constant([self.eos_token_id] * self.batch_size ) , 1 ) A__ : List[Any] = tf.concat([input_ids, eos_tensor] , axis=1 ) A__ : Union[str, Any] = ids_tensor([self.batch_size, self.seq_length] , self.vocab_size ) A__ : Tuple = self.config_cls( vocab_size=self.vocab_size , d_model=self.hidden_size , encoder_layers=self.num_hidden_layers , decoder_layers=self.num_hidden_layers , encoder_attention_heads=self.num_attention_heads , decoder_attention_heads=self.num_attention_heads , encoder_ffn_dim=self.intermediate_size , decoder_ffn_dim=self.intermediate_size , dropout=self.hidden_dropout_prob , attention_dropout=self.attention_probs_dropout_prob , max_position_embeddings=self.max_position_embeddings , eos_token_ids=[2] , bos_token_id=self.bos_token_id , pad_token_id=self.pad_token_id , decoder_start_token_id=self.pad_token_id , **self.config_updates , ) A__ : str = prepare_pegasus_inputs_dict(A__ , A__ , A__ ) return config, inputs_dict def __A ( self , A__ , A__ ): A__ : int = TFPegasusModel(config=A__ ).get_decoder() A__ : List[Any] = inputs_dict["""input_ids"""] A__ : Any = input_ids[:1, :] A__ : Optional[Any] = inputs_dict["""attention_mask"""][:1, :] A__ : Optional[int] = inputs_dict["""head_mask"""] A__ : Any = 1 # first forward pass A__ : Tuple = model(A__ , attention_mask=A__ , head_mask=A__ , use_cache=A__ ) A__ , A__ : Dict = outputs.to_tuple() # create hypothetical next token and extent to next_input_ids A__ : Tuple = ids_tensor((self.batch_size, 3) , config.vocab_size ) A__ : Optional[Any] = tf.cast(ids_tensor((self.batch_size, 3) , 2 ) , tf.inta ) # append to next input_ids and A__ : List[Any] = tf.concat([input_ids, next_tokens] , axis=-1 ) A__ : Tuple = tf.concat([attention_mask, next_attn_mask] , axis=-1 ) A__ : Optional[Any] = model(A__ , attention_mask=A__ )[0] A__ : Any = model(A__ , attention_mask=A__ , past_key_values=A__ )[0] self.parent.assertEqual(next_tokens.shape[1] , output_from_past.shape[1] ) # select random slice A__ : int = int(ids_tensor((1,) , output_from_past.shape[-1] ) ) A__ : Any = output_from_no_past[:, -3:, random_slice_idx] A__ : Tuple = output_from_past[:, :, random_slice_idx] # test that outputs are equal for slice tf.debugging.assert_near(A__ , A__ , rtol=1e-3 ) def UpperCamelCase (lowercase_: Union[str, Any] , lowercase_: Dict , lowercase_: List[Any] , lowercase_: Dict=None , lowercase_: int=None , lowercase_: List[Any]=None , lowercase_: List[Any]=None , lowercase_: str=None , ) -> int: if attention_mask is None: A__ : List[str] = tf.cast(tf.math.not_equal(lowercase_ , config.pad_token_id ) , tf.inta ) if decoder_attention_mask is None: A__ : Dict = tf.concat( [ tf.ones(decoder_input_ids[:, :1].shape , dtype=tf.inta ), tf.cast(tf.math.not_equal(decoder_input_ids[:, 1:] , config.pad_token_id ) , tf.inta ), ] , axis=-1 , ) if head_mask is None: A__ : Any = tf.ones((config.encoder_layers, config.encoder_attention_heads) ) if decoder_head_mask is None: A__ : Tuple = tf.ones((config.decoder_layers, config.decoder_attention_heads) ) if cross_attn_head_mask is None: A__ : List[str] = tf.ones((config.decoder_layers, config.decoder_attention_heads) ) return { "input_ids": input_ids, "decoder_input_ids": decoder_input_ids, "attention_mask": attention_mask, "decoder_attention_mask": decoder_attention_mask, "head_mask": head_mask, "decoder_head_mask": decoder_head_mask, "cross_attn_head_mask": cross_attn_head_mask, } @require_tf class _a (__magic_name__ , __magic_name__ , unittest.TestCase ): '''simple docstring''' UpperCAmelCase__: List[Any] = (TFPegasusForConditionalGeneration, TFPegasusModel) if is_tf_available() else () UpperCAmelCase__: Tuple = (TFPegasusForConditionalGeneration,) if is_tf_available() else () UpperCAmelCase__: Tuple = ( { '''conversational''': TFPegasusForConditionalGeneration, '''feature-extraction''': TFPegasusModel, '''summarization''': TFPegasusForConditionalGeneration, '''text2text-generation''': TFPegasusForConditionalGeneration, '''translation''': TFPegasusForConditionalGeneration, } if is_tf_available() else {} ) UpperCAmelCase__: int = True UpperCAmelCase__: Union[str, Any] = False UpperCAmelCase__: List[str] = False def __A ( self ): A__ : Optional[Any] = TFPegasusModelTester(self ) A__ : Tuple = ConfigTester(self , config_class=A__ ) def __A ( self ): self.config_tester.run_common_tests() def __A ( self ): A__ : int = self.model_tester.prepare_config_and_inputs_for_common() self.model_tester.check_decoder_model_past_large_inputs(*A__ ) @require_sentencepiece @require_tokenizers @require_tf class _a (unittest.TestCase ): '''simple docstring''' UpperCAmelCase__: Optional[int] = [ ''' PG&E stated it scheduled the blackouts in response to forecasts for high winds amid dry conditions. The aim is to reduce the risk of wildfires. Nearly 800 thousand customers were scheduled to be affected by the shutoffs which were expected to last through at least midday tomorrow.''', ''' The London trio are up for best UK act and best album, as well as getting two nominations in the best song category."We got told like this morning \'Oh I think you\'re nominated\'", said Dappy."And I was like \'Oh yeah, which one?\' And now we\'ve got nominated for four awards. I mean, wow!"Bandmate Fazer added: "We thought it\'s best of us to come down and mingle with everyone and say hello to the cameras. And now we find we\'ve got four nominations."The band have two shots at the best song prize, getting the nod for their Tynchy Stryder collaboration Number One, and single Strong Again.Their album Uncle B will also go up against records by the likes of Beyonce and Kanye West.N-Dubz picked up the best newcomer Mobo in 2007, but female member Tulisa said they wouldn\'t be too disappointed if they didn\'t win this time around."At the end of the day we\'re grateful to be where we are in our careers."If it don\'t happen then it don\'t happen - live to fight another day and keep on making albums and hits for the fans."Dappy also revealed they could be performing live several times on the night.The group will be doing Number One and also a possible rendition of the War Child single, I Got Soul.The charity song is a re-working of The Killers\' All These Things That I\'ve Done and is set to feature artists like Chipmunk, Ironik and Pixie Lott.This year\'s Mobos will be held outside of London for the first time, in Glasgow on 30 September.N-Dubz said they were looking forward to performing for their Scottish fans and boasted about their recent shows north of the border."We just done Edinburgh the other day," said Dappy."We smashed up an N-Dubz show over there. We done Aberdeen about three or four months ago - we smashed up that show over there! Everywhere we go we smash it up!" ''', ] UpperCAmelCase__: Any = [ '''California\'s largest electricity provider has cut power to hundreds of thousands of customers in an effort to''' ''' reduce the risk of wildfires.''', '''N-Dubz have revealed they\'re "grateful" to have been nominated for four Mobo Awards.''', ] # differs slightly from pytorch, likely due to numerical differences in linear layers UpperCAmelCase__: List[str] = '''google/pegasus-xsum''' @cached_property def __A ( self ): return AutoTokenizer.from_pretrained(self.model_name ) @cached_property def __A ( self ): A__ : int = TFAutoModelForSeqaSeqLM.from_pretrained(self.model_name ) return model def __A ( self , **A__ ): A__ : str = self.translate_src_text(**A__ ) assert self.expected_text == generated_words def __A ( self , **A__ ): A__ : List[str] = self.tokenizer(self.src_text , **A__ , padding=A__ , return_tensors="""tf""" ) A__ : Optional[int] = self.model.generate( model_inputs.input_ids , attention_mask=model_inputs.attention_mask , num_beams=2 , use_cache=A__ , ) A__ : Dict = self.tokenizer.batch_decode(generated_ids.numpy() , skip_special_tokens=A__ ) return generated_words @slow def __A ( self ): self._assert_generated_batch_equal_expected()
64
1
from datetime import datetime import requests from bsa import BeautifulSoup if __name__ == "__main__": A_ : Dict = input('Enter image url: ').strip() print(f'''Downloading image from {url} ...''') A_ : Any = BeautifulSoup(requests.get(url).content, 'html.parser') # The image URL is in the content field of the first meta tag with property og:image A_ : Dict = soup.find('meta', {'property': 'og:image'})['content'] A_ : int = requests.get(image_url).content A_ : Dict = f'''{datetime.now():%Y-%m-%d_%H:%M:%S}.jpg''' with open(file_name, 'wb') as fp: fp.write(image_data) print(f'''Done. Image saved to disk as {file_name}.''')
64
class _a : '''simple docstring''' def __init__( self ): A__ : str = """""" A__ : Any = """""" A__ : List[Any] = [] def __A ( self , A__ , A__ ): if m == -1: return n + 1 elif n == -1: return m + 1 elif self.dp[m][n] > -1: return self.dp[m][n] else: if self.worda[m] == self.worda[n]: A__ : Optional[Any] = self.__min_dist_top_down_dp(m - 1 , n - 1 ) else: A__ : Union[str, Any] = self.__min_dist_top_down_dp(A__ , n - 1 ) A__ : Union[str, Any] = self.__min_dist_top_down_dp(m - 1 , A__ ) A__ : Union[str, Any] = self.__min_dist_top_down_dp(m - 1 , n - 1 ) A__ : List[Any] = 1 + min(A__ , A__ , A__ ) return self.dp[m][n] def __A ( self , A__ , A__ ): A__ : Tuple = worda A__ : Dict = worda A__ : Optional[Any] = [[-1 for _ in range(len(A__ ) )] for _ in range(len(A__ ) )] return self.__min_dist_top_down_dp(len(A__ ) - 1 , len(A__ ) - 1 ) def __A ( self , A__ , A__ ): A__ : Optional[Any] = worda A__ : Dict = worda A__ : Union[str, Any] = len(A__ ) A__ : List[str] = len(A__ ) A__ : int = [[0 for _ in range(n + 1 )] for _ in range(m + 1 )] for i in range(m + 1 ): for j in range(n + 1 ): if i == 0: # first string is empty A__ : Tuple = j elif j == 0: # second string is empty A__ : Dict = i elif worda[i - 1] == worda[j - 1]: # last characters are equal A__ : str = self.dp[i - 1][j - 1] else: A__ : Union[str, Any] = self.dp[i][j - 1] A__ : str = self.dp[i - 1][j] A__ : Union[str, Any] = self.dp[i - 1][j - 1] A__ : Tuple = 1 + min(A__ , A__ , A__ ) return self.dp[m][n] if __name__ == "__main__": A_ : Union[str, Any] = EditDistance() print('****************** Testing Edit Distance DP Algorithm ******************') print() A_ : int = input('Enter the first string: ').strip() A_ : List[str] = input('Enter the second string: ').strip() print() print(f'''The minimum edit distance is: {solver.min_dist_top_down(Sa, Sa)}''') print(f'''The minimum edit distance is: {solver.min_dist_bottom_up(Sa, Sa)}''') print() print('*************** End of Testing Edit Distance DP Algorithm ***************')
64
1
# This script creates a super tiny model that is useful inside tests, when we just want to test that # the machinery works, without needing to the check the quality of the outcomes. # # This version creates a tiny model through reduction of a normal pre-trained model, but keeping the # full vocab, merges file, and thus also resulting in a larger model due to a large vocab size. # This gives ~3MB in total for all files. # # If you want a 50 times smaller than this see `fsmt-make-super-tiny-model.py`, which is slightly more complicated # # # It will be used then as "stas/tiny-wmt19-en-de" # Build from transformers import FSMTTokenizer, FSMTConfig, FSMTForConditionalGeneration A_ : Tuple = 'facebook/wmt19-en-de' A_ : int = FSMTTokenizer.from_pretrained(mname) # get the correct vocab sizes, etc. from the master model A_ : Dict = FSMTConfig.from_pretrained(mname) config.update( dict( d_model=4, encoder_layers=1, decoder_layers=1, encoder_ffn_dim=4, decoder_ffn_dim=4, encoder_attention_heads=1, decoder_attention_heads=1, ) ) A_ : List[str] = FSMTForConditionalGeneration(config) print(f'''num of params {tiny_model.num_parameters()}''') # Test A_ : int = tokenizer(['Making tiny model'], return_tensors='pt') A_ : Optional[int] = tiny_model(**batch) print('test output:', len(outputs.logits[0])) # Save A_ : Optional[Any] = 'tiny-wmt19-en-de' tiny_model.half() # makes it smaller tiny_model.save_pretrained(mname_tiny) tokenizer.save_pretrained(mname_tiny) print(f'''Generated {mname_tiny}''') # Upload # transformers-cli upload tiny-wmt19-en-de
64
def UpperCamelCase (lowercase_: int , lowercase_: int ) -> int: while second != 0: A__ : int = first & second first ^= second A__ : int = c << 1 return first if __name__ == "__main__": import doctest doctest.testmod() A_ : Optional[Any] = int(input('Enter the first number: ').strip()) A_ : List[str] = int(input('Enter the second number: ').strip()) print(f'''{add(first, second) = }''')
64
1
import json import logging import os import re import sys from dataclasses import dataclass, field from typing import Any, Dict, List, Optional, Union import datasets import numpy as np import torch import torchaudio from packaging import version from torch import nn import transformers from transformers import ( HfArgumentParser, Trainer, TrainingArguments, WavaVecaCTCTokenizer, WavaVecaFeatureExtractor, WavaVecaForCTC, WavaVecaProcessor, is_apex_available, set_seed, ) from transformers.trainer_utils import get_last_checkpoint, is_main_process if is_apex_available(): from apex import amp if version.parse(version.parse(torch.__version__).base_version) >= version.parse('1.6'): A_ : Optional[Any] = True from torch.cuda.amp import autocast A_ : int = logging.getLogger(__name__) def UpperCamelCase (lowercase_: Union[str, Any]=None , lowercase_: Tuple=None ) -> Union[str, Any]: return field(default_factory=lambda: default , metadata=lowercase_ ) @dataclass class _a : '''simple docstring''' UpperCAmelCase__: str = field( metadata={'''help''': '''Path to pretrained model or model identifier from huggingface.co/models'''} ) UpperCAmelCase__: Optional[str] = field( default=__magic_name__ , metadata={'''help''': '''Where do you want to store the pretrained models downloaded from huggingface.co'''} , ) UpperCAmelCase__: Optional[bool] = field( default=__magic_name__ , metadata={'''help''': '''Whether to freeze the feature extractor layers of the model.'''} ) UpperCAmelCase__: Optional[float] = field( default=0.1 , metadata={'''help''': '''The dropout ratio for the attention probabilities.'''} ) UpperCAmelCase__: Optional[float] = field( default=0.1 , metadata={'''help''': '''The dropout ratio for activations inside the fully connected layer.'''} ) UpperCAmelCase__: Optional[float] = field( default=0.1 , metadata={ '''help''': '''The dropout probabilitiy for all fully connected layers in the embeddings, encoder, and pooler.''' } , ) UpperCAmelCase__: Optional[float] = field( default=0.1 , metadata={'''help''': '''The dropout probabilitiy for all 1D convolutional layers in feature extractor.'''} , ) UpperCAmelCase__: Optional[float] = field( default=0.05 , metadata={ '''help''': ( '''Propability of each feature vector along the time axis to be chosen as the start of the vector''' '''span to be masked. Approximately ``mask_time_prob * sequence_length // mask_time_length`` feature''' '''vectors will be masked along the time axis. This is only relevant if ``apply_spec_augment is True``.''' ) } , ) UpperCAmelCase__: Optional[float] = field(default=0.0 , metadata={'''help''': '''The LayerDrop probability.'''} ) @dataclass class _a : '''simple docstring''' UpperCAmelCase__: Optional[str] = field( default=__magic_name__ , metadata={'''help''': '''The configuration name of the dataset to use (via the datasets library).'''} ) UpperCAmelCase__: Optional[str] = field( default='''train+validation''' , metadata={ '''help''': '''The name of the training data set split to use (via the datasets library). Defaults to \'train\'''' } , ) UpperCAmelCase__: bool = field( default=__magic_name__ , metadata={'''help''': '''Overwrite the cached preprocessed datasets or not.'''} ) UpperCAmelCase__: Optional[int] = field( default=__magic_name__ , metadata={'''help''': '''The number of processes to use for the preprocessing.'''} , ) UpperCAmelCase__: Optional[int] = field( default=__magic_name__ , metadata={ '''help''': ( '''For debugging purposes or quicker training, truncate the number of training examples to this ''' '''value if set.''' ) } , ) UpperCAmelCase__: Optional[int] = field( default=__magic_name__ , metadata={ '''help''': ( '''For debugging purposes or quicker training, truncate the number of validation examples to this ''' '''value if set.''' ) } , ) UpperCAmelCase__: List[str] = list_field( default=[''',''', '''?''', '''.''', '''!''', '''-''', ''';''', ''':''', '''""''', '''%''', '''\'''', '''"''', '''�'''] , metadata={'''help''': '''A list of characters to remove from the transcripts.'''} , ) @dataclass class _a : '''simple docstring''' UpperCAmelCase__: WavaVecaProcessor UpperCAmelCase__: Union[bool, str] = True UpperCAmelCase__: Optional[int] = None UpperCAmelCase__: Optional[int] = None UpperCAmelCase__: Optional[int] = None UpperCAmelCase__: Optional[int] = None def __call__( self , A__ ): # split inputs and labels since they have to be of different lenghts and need # different padding methods A__ : Tuple = [{"""input_values""": feature["""input_values"""]} for feature in features] A__ : List[str] = [{"""input_ids""": feature["""labels"""]} for feature in features] A__ : List[Any] = self.processor.pad( A__ , padding=self.padding , max_length=self.max_length , pad_to_multiple_of=self.pad_to_multiple_of , return_tensors="""pt""" , ) A__ : Optional[Any] = self.processor.pad( labels=A__ , padding=self.padding , max_length=self.max_length_labels , pad_to_multiple_of=self.pad_to_multiple_of_labels , return_tensors="""pt""" , ) # replace padding with -100 to ignore loss correctly A__ : Optional[Any] = labels_batch["""input_ids"""].masked_fill(labels_batch.attention_mask.ne(1 ) , -100 ) A__ : str = labels return batch class _a (__magic_name__ ): '''simple docstring''' def __A ( self , A__ , A__ ): model.train() A__ : Optional[Any] = self._prepare_inputs(A__ ) if self.use_amp: with autocast(): A__ : Any = self.compute_loss(A__ , A__ ) else: A__ : Union[str, Any] = self.compute_loss(A__ , A__ ) if self.args.n_gpu > 1: if model.module.config.ctc_loss_reduction == "mean": A__ : int = loss.mean() elif model.module.config.ctc_loss_reduction == "sum": A__ : List[Any] = loss.sum() / (inputs["""labels"""] >= 0).sum() else: raise ValueError(F"""{model.config.ctc_loss_reduction} is not valid. Choose one of ['mean', 'sum']""" ) if self.args.gradient_accumulation_steps > 1: A__ : Optional[Any] = loss / self.args.gradient_accumulation_steps if self.use_amp: self.scaler.scale(A__ ).backward() elif self.use_apex: with amp.scale_loss(A__ , self.optimizer ) as scaled_loss: scaled_loss.backward() elif self.deepspeed: self.deepspeed.backward(A__ ) else: loss.backward() return loss.detach() def UpperCamelCase () -> Any: # See all possible arguments in src/transformers/training_args.py # or by passing the --help flag to this script. # We now keep distinct sets of args, for a cleaner separation of concerns. A__ : str = HfArgumentParser((ModelArguments, DataTrainingArguments, TrainingArguments) ) if len(sys.argv ) == 2 and sys.argv[1].endswith(""".json""" ): # If we pass only one argument to the script and it's the path to a json file, # let's parse it to get our arguments. A__ , A__ , A__ : Dict = parser.parse_json_file(json_file=os.path.abspath(sys.argv[1] ) ) else: A__ , A__ , A__ : Dict = parser.parse_args_into_dataclasses() # Detecting last checkpoint. A__ : List[Any] = None if os.path.isdir(training_args.output_dir ) and training_args.do_train and not training_args.overwrite_output_dir: A__ : Tuple = get_last_checkpoint(training_args.output_dir ) if last_checkpoint is None and len(os.listdir(training_args.output_dir ) ) > 0: raise ValueError( f"""Output directory ({training_args.output_dir}) already exists and is not empty. """ """Use --overwrite_output_dir to overcome.""" ) elif last_checkpoint is not None: logger.info( f"""Checkpoint detected, resuming training at {last_checkpoint}. To avoid this behavior, change """ """the `--output_dir` or add `--overwrite_output_dir` to train from scratch.""" ) # Setup logging logging.basicConfig( format="""%(asctime)s - %(levelname)s - %(name)s - %(message)s""" , datefmt="""%m/%d/%Y %H:%M:%S""" , handlers=[logging.StreamHandler(sys.stdout )] , ) logger.setLevel(logging.INFO if is_main_process(training_args.local_rank ) else logging.WARN ) # Log on each process the small summary: logger.warning( f"""Process rank: {training_args.local_rank}, device: {training_args.device}, n_gpu: {training_args.n_gpu}""" + f"""distributed training: {bool(training_args.local_rank != -1 )}, 16-bits training: {training_args.fpaa}""" ) # Set the verbosity to info of the Transformers logger (on main process only): if is_main_process(training_args.local_rank ): transformers.utils.logging.set_verbosity_info() logger.info("""Training/evaluation parameters %s""" , lowercase_ ) # Set seed before initializing model. set_seed(training_args.seed ) # Get the datasets: A__ : Optional[int] = datasets.load_dataset( """common_voice""" , data_args.dataset_config_name , split=data_args.train_split_name ) A__ : Optional[Any] = datasets.load_dataset("""common_voice""" , data_args.dataset_config_name , split="""test""" ) # Create and save tokenizer A__ : Dict = f"""[{''.join(data_args.chars_to_ignore )}]""" def remove_special_characters(lowercase_: List[str] ): A__ : int = re.sub(lowercase_ , """""" , batch["""sentence"""] ).lower() + """ """ return batch A__ : Any = train_dataset.map(lowercase_ , remove_columns=["""sentence"""] ) A__ : Tuple = eval_dataset.map(lowercase_ , remove_columns=["""sentence"""] ) def extract_all_chars(lowercase_: Optional[Any] ): A__ : Optional[int] = """ """.join(batch["""text"""] ) A__ : Union[str, Any] = list(set(lowercase_ ) ) return {"vocab": [vocab], "all_text": [all_text]} A__ : List[str] = train_dataset.map( lowercase_ , batched=lowercase_ , batch_size=-1 , keep_in_memory=lowercase_ , remove_columns=train_dataset.column_names , ) A__ : Optional[Any] = train_dataset.map( lowercase_ , batched=lowercase_ , batch_size=-1 , keep_in_memory=lowercase_ , remove_columns=eval_dataset.column_names , ) A__ : List[Any] = list(set(vocab_train["""vocab"""][0] ) | set(vocab_test["""vocab"""][0] ) ) A__ : Optional[int] = {v: k for k, v in enumerate(lowercase_ )} A__ : List[Any] = vocab_dict[""" """] del vocab_dict[" "] A__ : List[str] = len(lowercase_ ) A__ : Optional[int] = len(lowercase_ ) with open("""vocab.json""" , """w""" ) as vocab_file: json.dump(lowercase_ , lowercase_ ) # Load pretrained model and tokenizer # # Distributed training: # The .from_pretrained methods guarantee that only one local process can concurrently # download model & vocab. A__ : Optional[int] = WavaVecaCTCTokenizer( """vocab.json""" , unk_token="""[UNK]""" , pad_token="""[PAD]""" , word_delimiter_token="""|""" , ) A__ : Union[str, Any] = WavaVecaFeatureExtractor( feature_size=1 , sampling_rate=16000 , padding_value=0.0 , do_normalize=lowercase_ , return_attention_mask=lowercase_ ) A__ : Optional[int] = WavaVecaProcessor(feature_extractor=lowercase_ , tokenizer=lowercase_ ) A__ : Optional[int] = WavaVecaForCTC.from_pretrained( model_args.model_name_or_path , cache_dir=model_args.cache_dir , activation_dropout=model_args.activation_dropout , attention_dropout=model_args.attention_dropout , hidden_dropout=model_args.hidden_dropout , feat_proj_dropout=model_args.feat_proj_dropout , mask_time_prob=model_args.mask_time_prob , gradient_checkpointing=training_args.gradient_checkpointing , layerdrop=model_args.layerdrop , ctc_loss_reduction="""mean""" , pad_token_id=processor.tokenizer.pad_token_id , vocab_size=len(processor.tokenizer ) , ) if data_args.max_train_samples is not None: A__ : Tuple = min(len(lowercase_ ) , data_args.max_train_samples ) A__ : Dict = train_dataset.select(range(lowercase_ ) ) if data_args.max_val_samples is not None: A__ : str = eval_dataset.select(range(data_args.max_val_samples ) ) A__ : Union[str, Any] = torchaudio.transforms.Resample(48000 , 16000 ) # Preprocessing the datasets. # We need to read the aduio files as arrays and tokenize the targets. def speech_file_to_array_fn(lowercase_: Any ): A__ , A__ : Any = torchaudio.load(batch["""path"""] ) A__ : List[Any] = resampler(lowercase_ ).squeeze().numpy() A__ : Tuple = 16000 A__ : Any = batch["""text"""] return batch A__ : Optional[Any] = train_dataset.map( lowercase_ , remove_columns=train_dataset.column_names , num_proc=data_args.preprocessing_num_workers , ) A__ : Optional[Any] = eval_dataset.map( lowercase_ , remove_columns=eval_dataset.column_names , num_proc=data_args.preprocessing_num_workers , ) def prepare_dataset(lowercase_: str ): # check that all files have the correct sampling rate assert ( len(set(batch["""sampling_rate"""] ) ) == 1 ), f"""Make sure all inputs have the same sampling rate of {processor.feature_extractor.sampling_rate}.""" A__ : int = processor( audio=batch["""speech"""] , text=batch["""target_text"""] , sampling_rate=batch["""sampling_rate"""][0] ) batch.update(lowercase_ ) return batch A__ : Any = train_dataset.map( lowercase_ , remove_columns=train_dataset.column_names , batch_size=training_args.per_device_train_batch_size , batched=lowercase_ , num_proc=data_args.preprocessing_num_workers , ) A__ : Union[str, Any] = eval_dataset.map( lowercase_ , remove_columns=eval_dataset.column_names , batch_size=training_args.per_device_train_batch_size , batched=lowercase_ , num_proc=data_args.preprocessing_num_workers , ) # Metric A__ : int = datasets.load_metric("""wer""" ) def compute_metrics(lowercase_: str ): A__ : Any = pred.predictions A__ : Optional[Any] = np.argmax(lowercase_ , axis=-1 ) A__ : List[str] = processor.tokenizer.pad_token_id A__ : List[Any] = processor.batch_decode(lowercase_ ) # we do not want to group tokens when computing the metrics A__ : List[str] = processor.batch_decode(pred.label_ids , group_tokens=lowercase_ ) A__ : int = wer_metric.compute(predictions=lowercase_ , references=lowercase_ ) return {"wer": wer} if model_args.freeze_feature_extractor: model.freeze_feature_extractor() # Data collator A__ : int = DataCollatorCTCWithPadding(processor=lowercase_ , padding=lowercase_ ) # Initialize our Trainer A__ : int = CTCTrainer( model=lowercase_ , data_collator=lowercase_ , args=lowercase_ , compute_metrics=lowercase_ , train_dataset=train_dataset if training_args.do_train else None , eval_dataset=eval_dataset if training_args.do_eval else None , tokenizer=processor.feature_extractor , ) # Training if training_args.do_train: if last_checkpoint is not None: A__ : int = last_checkpoint elif os.path.isdir(model_args.model_name_or_path ): A__ : Tuple = model_args.model_name_or_path else: A__ : str = None # Save the feature_extractor and the tokenizer if is_main_process(training_args.local_rank ): processor.save_pretrained(training_args.output_dir ) A__ : List[Any] = trainer.train(resume_from_checkpoint=lowercase_ ) trainer.save_model() A__ : Tuple = train_result.metrics A__ : Dict = ( data_args.max_train_samples if data_args.max_train_samples is not None else len(lowercase_ ) ) A__ : List[Any] = min(lowercase_ , len(lowercase_ ) ) trainer.log_metrics("""train""" , lowercase_ ) trainer.save_metrics("""train""" , lowercase_ ) trainer.save_state() # Evaluation A__ : str = {} if training_args.do_eval: logger.info("""*** Evaluate ***""" ) A__ : Optional[int] = trainer.evaluate() A__ : str = data_args.max_val_samples if data_args.max_val_samples is not None else len(lowercase_ ) A__ : Tuple = min(lowercase_ , len(lowercase_ ) ) trainer.log_metrics("""eval""" , lowercase_ ) trainer.save_metrics("""eval""" , lowercase_ ) return results if __name__ == "__main__": main()
64
from __future__ import annotations from collections.abc import Callable A_ : List[Any] = list[list[float | int]] def UpperCamelCase (lowercase_: Matrix , lowercase_: Matrix ) -> Matrix: A__ : int = len(lowercase_ ) A__ : Matrix = [[0 for _ in range(size + 1 )] for _ in range(lowercase_ )] A__ : int A__ : int A__ : int A__ : int A__ : int A__ : float for row in range(lowercase_ ): for col in range(lowercase_ ): A__ : List[str] = matrix[row][col] A__ : int = vector[row][0] A__ : Optional[int] = 0 A__ : str = 0 while row < size and col < size: # pivoting A__ : int = max((abs(augmented[rowa][col] ), rowa) for rowa in range(lowercase_ , lowercase_ ) )[ 1 ] if augmented[pivot_row][col] == 0: col += 1 continue else: A__ , A__ : Union[str, Any] = augmented[pivot_row], augmented[row] for rowa in range(row + 1 , lowercase_ ): A__ : List[Any] = augmented[rowa][col] / augmented[row][col] A__ : Dict = 0 for cola in range(col + 1 , size + 1 ): augmented[rowa][cola] -= augmented[row][cola] * ratio row += 1 col += 1 # back substitution for col in range(1 , lowercase_ ): for row in range(lowercase_ ): A__ : List[str] = augmented[row][col] / augmented[col][col] for cola in range(lowercase_ , size + 1 ): augmented[row][cola] -= augmented[col][cola] * ratio # round to get rid of numbers like 2.000000000000004 return [ [round(augmented[row][size] / augmented[row][row] , 10 )] for row in range(lowercase_ ) ] def UpperCamelCase (lowercase_: list[int] ) -> Callable[[int], int]: A__ : int = len(lowercase_ ) A__ : Matrix = [[0 for _ in range(lowercase_ )] for _ in range(lowercase_ )] A__ : Matrix = [[0] for _ in range(lowercase_ )] A__ : Matrix A__ : int A__ : int A__ : int for x_val, y_val in enumerate(lowercase_ ): for col in range(lowercase_ ): A__ : Dict = (x_val + 1) ** (size - col - 1) A__ : Any = y_val A__ : Union[str, Any] = solve(lowercase_ , lowercase_ ) def interpolated_func(lowercase_: int ) -> int: return sum( round(coeffs[x_val][0] ) * (var ** (size - x_val - 1)) for x_val in range(lowercase_ ) ) return interpolated_func def UpperCamelCase (lowercase_: int ) -> int: return ( 1 - variable + variable**2 - variable**3 + variable**4 - variable**5 + variable**6 - variable**7 + variable**8 - variable**9 + variable**10 ) def UpperCamelCase (lowercase_: Callable[[int], int] = question_function , lowercase_: int = 10 ) -> int: A__ : list[int] = [func(lowercase_ ) for x_val in range(1 , order + 1 )] A__ : list[Callable[[int], int]] = [ interpolate(data_points[:max_coeff] ) for max_coeff in range(1 , order + 1 ) ] A__ : int = 0 A__ : Callable[[int], int] A__ : int for poly in polynomials: A__ : List[str] = 1 while func(lowercase_ ) == poly(lowercase_ ): x_val += 1 ret += poly(lowercase_ ) return ret if __name__ == "__main__": print(f'''{solution() = }''')
64
1
import random import torch from huggingface_hub import HfApi from diffusers import UNetaDModel A_ : Tuple = HfApi() A_ : Any = {} # fmt: off A_ : Any = torch.tensor([ -0.7515, -1.6883, 0.2420, 0.0300, 0.6347, 1.3433, -1.1743, -3.7467, 1.2342, -2.2485, 0.4636, 0.8076, -0.7991, 0.3969, 0.8498, 0.9189, -1.8887, -3.3522, 0.7639, 0.2040, 0.6271, -2.7148, -1.6316, 3.0839, 0.3186, 0.2721, -0.9759, -1.2461, 2.6257, 1.3557 ]) A_ : Optional[int] = torch.tensor([ -2.3639, -2.5344, 0.0054, -0.6674, 1.5990, 1.0158, 0.3124, -2.1436, 1.8795, -2.5429, -0.1566, -0.3973, 1.2490, 2.6447, 1.2283, -0.5208, -2.8154, -3.5119, 2.3838, 1.2033, 1.7201, -2.1256, -1.4576, 2.7948, 2.4204, -0.9752, -1.2546, 0.8027, 3.2758, 3.1365 ]) A_ : Optional[Any] = torch.tensor([ -0.6531, -0.6891, -0.3172, -0.5375, -0.9140, -0.5367, -0.1175, -0.7869, -0.3808, -0.4513, -0.2098, -0.0083, 0.3183, 0.5140, 0.2247, -0.1304, -0.1302, -0.2802, -0.2084, -0.2025, -0.4967, -0.4873, -0.0861, 0.6925, 0.0250, 0.1290, -0.1543, 0.6316, 1.0460, 1.4943 ]) A_ : Tuple = torch.tensor([ 0.0911, 0.1107, 0.0182, 0.0435, -0.0805, -0.0608, 0.0381, 0.2172, -0.0280, 0.1327, -0.0299, -0.0255, -0.0050, -0.1170, -0.1046, 0.0309, 0.1367, 0.1728, -0.0533, -0.0748, -0.0534, 0.1624, 0.0384, -0.1805, -0.0707, 0.0642, 0.0220, -0.0134, -0.1333, -0.1505 ]) A_ : Optional[Any] = torch.tensor([ 0.1321, 0.1337, 0.0440, 0.0622, -0.0591, -0.0370, 0.0503, 0.2133, -0.0177, 0.1415, -0.0116, -0.0112, 0.0044, -0.0980, -0.0789, 0.0395, 0.1502, 0.1785, -0.0488, -0.0514, -0.0404, 0.1539, 0.0454, -0.1559, -0.0665, 0.0659, 0.0383, -0.0005, -0.1266, -0.1386 ]) A_ : Optional[int] = torch.tensor([ 0.1154, 0.1218, 0.0307, 0.0526, -0.0711, -0.0541, 0.0366, 0.2078, -0.0267, 0.1317, -0.0226, -0.0193, -0.0014, -0.1055, -0.0902, 0.0330, 0.1391, 0.1709, -0.0562, -0.0693, -0.0560, 0.1482, 0.0381, -0.1683, -0.0681, 0.0661, 0.0331, -0.0046, -0.1268, -0.1431 ]) A_ : Any = torch.tensor([ 0.1192, 0.1240, 0.0414, 0.0606, -0.0557, -0.0412, 0.0430, 0.2042, -0.0200, 0.1385, -0.0115, -0.0132, 0.0017, -0.0965, -0.0802, 0.0398, 0.1433, 0.1747, -0.0458, -0.0533, -0.0407, 0.1545, 0.0419, -0.1574, -0.0645, 0.0626, 0.0341, -0.0010, -0.1199, -0.1390 ]) A_ : Dict = torch.tensor([ 0.1075, 0.1074, 0.0205, 0.0431, -0.0774, -0.0607, 0.0298, 0.2042, -0.0320, 0.1267, -0.0281, -0.0250, -0.0064, -0.1091, -0.0946, 0.0290, 0.1328, 0.1650, -0.0580, -0.0738, -0.0586, 0.1440, 0.0337, -0.1746, -0.0712, 0.0605, 0.0250, -0.0099, -0.1316, -0.1473 ]) A_ : str = torch.tensor([ -1.4572, -2.0481, -0.0414, -0.6005, 1.4136, 0.5848, 0.4028, -2.7330, 1.2212, -2.1228, 0.2155, 0.4039, 0.7662, 2.0535, 0.7477, -0.3243, -2.1758, -2.7648, 1.6947, 0.7026, 1.2338, -1.6078, -0.8682, 2.2810, 1.8574, -0.5718, -0.5586, -0.0186, 2.3415, 2.1251]) A_ : Optional[Any] = torch.tensor([ -1.3690, -1.9720, -0.4090, -0.6966, 1.4660, 0.9938, -0.1385, -2.7324, 0.7736, -1.8917, 0.2923, 0.4293, 0.1693, 1.4112, 1.1887, -0.3181, -2.2160, -2.6381, 1.3170, 0.8163, 0.9240, -1.6544, -0.6099, 2.5259, 1.6430, -0.9090, -0.9392, -0.0126, 2.4268, 2.3266 ]) A_ : List[Any] = torch.tensor([ -1.3525, -1.9628, -0.3956, -0.6860, 1.4664, 1.0014, -0.1259, -2.7212, 0.7772, -1.8811, 0.2996, 0.4388, 0.1704, 1.4029, 1.1701, -0.3027, -2.2053, -2.6287, 1.3350, 0.8131, 0.9274, -1.6292, -0.6098, 2.5131, 1.6505, -0.8958, -0.9298, -0.0151, 2.4257, 2.3355 ]) A_ : Tuple = torch.tensor([ -2.0585, -2.7897, -0.2850, -0.8940, 1.9052, 0.5702, 0.6345, -3.8959, 1.5932, -3.2319, 0.1974, 0.0287, 1.7566, 2.6543, 0.8387, -0.5351, -3.2736, -4.3375, 2.9029, 1.6390, 1.4640, -2.1701, -1.9013, 2.9341, 3.4981, -0.6255, -1.1644, -0.1591, 3.7097, 3.2066 ]) A_ : Any = torch.tensor([ -2.3139, -2.5594, -0.0197, -0.6785, 1.7001, 1.1606, 0.3075, -2.1740, 1.8071, -2.5630, -0.0926, -0.3811, 1.2116, 2.6246, 1.2731, -0.5398, -2.8153, -3.6140, 2.3893, 1.3262, 1.6258, -2.1856, -1.3267, 2.8395, 2.3779, -1.0623, -1.2468, 0.8959, 3.3367, 3.2243 ]) A_ : Tuple = torch.tensor([ -2.0628, -2.7667, -0.2089, -0.8263, 2.0539, 0.5992, 0.6495, -3.8336, 1.6025, -3.2817, 0.1721, -0.0633, 1.7516, 2.7039, 0.8100, -0.5908, -3.2113, -4.4343, 2.9257, 1.3632, 1.5562, -2.1489, -1.9894, 3.0560, 3.3396, -0.7328, -1.0417, 0.0383, 3.7093, 3.2343 ]) A_ : str = torch.tensor([ -1.4574, -2.0569, -0.0473, -0.6117, 1.4018, 0.5769, 0.4129, -2.7344, 1.2241, -2.1397, 0.2000, 0.3937, 0.7616, 2.0453, 0.7324, -0.3391, -2.1746, -2.7744, 1.6963, 0.6921, 1.2187, -1.6172, -0.8877, 2.2439, 1.8471, -0.5839, -0.5605, -0.0464, 2.3250, 2.1219 ]) # fmt: on A_ : Optional[Any] = api.list_models(filter='diffusers') for mod in models: if "google" in mod.author or mod.modelId == "CompVis/ldm-celebahq-256": A_ : int = '/home/patrick/google_checkpoints/' + mod.modelId.split('/')[-1] print(f'''Started running {mod.modelId}!!!''') if mod.modelId.startswith('CompVis'): A_ : Any = UNetaDModel.from_pretrained(local_checkpoint, subfolder='unet') else: A_ : Any = UNetaDModel.from_pretrained(local_checkpoint) torch.manual_seed(0) random.seed(0) A_ : Tuple = torch.randn(1, model.config.in_channels, model.config.sample_size, model.config.sample_size) A_ : List[Any] = torch.tensor([10] * noise.shape[0]) with torch.no_grad(): A_ : Dict = model(noise, time_step).sample assert torch.allclose( logits[0, 0, 0, :30], results['_'.join('_'.join(mod.modelId.split('/')).split('-'))], atol=1E-3 ) print(f'''{mod.modelId} has passed successfully!!!''')
64
from functools import lru_cache @lru_cache def UpperCamelCase (lowercase_: int ) -> int: if num < 0: raise ValueError("""Number should not be negative.""" ) return 1 if num in (0, 1) else num * factorial(num - 1 ) if __name__ == "__main__": import doctest doctest.testmod()
64
1
import enum import os from hashlib import shaaaa from typing import Optional from .. import config from .logging import get_logger A_ : Any = get_logger(__name__) class _a (enum.Enum ): '''simple docstring''' UpperCAmelCase__: Dict = '''all_checks''' UpperCAmelCase__: Optional[int] = '''basic_checks''' UpperCAmelCase__: Optional[Any] = '''no_checks''' class _a (__magic_name__ ): '''simple docstring''' class _a (__magic_name__ ): '''simple docstring''' class _a (__magic_name__ ): '''simple docstring''' class _a (__magic_name__ ): '''simple docstring''' def UpperCamelCase (lowercase_: Optional[dict] , lowercase_: dict , lowercase_: List[Any]=None ) -> Dict: if expected_checksums is None: logger.info("""Unable to verify checksums.""" ) return if len(set(lowercase_ ) - set(lowercase_ ) ) > 0: raise ExpectedMoreDownloadedFiles(str(set(lowercase_ ) - set(lowercase_ ) ) ) if len(set(lowercase_ ) - set(lowercase_ ) ) > 0: raise UnexpectedDownloadedFile(str(set(lowercase_ ) - set(lowercase_ ) ) ) A__ : List[str] = [url for url in expected_checksums if expected_checksums[url] != recorded_checksums[url]] A__ : Union[str, Any] = """ for """ + verification_name if verification_name is not None else """""" if len(lowercase_ ) > 0: raise NonMatchingChecksumError( f"""Checksums didn't match{for_verification_name}:\n""" f"""{bad_urls}\n""" """Set `verification_mode='no_checks'` to skip checksums verification and ignore this error""" ) logger.info("""All the checksums matched successfully""" + for_verification_name ) class _a (__magic_name__ ): '''simple docstring''' class _a (__magic_name__ ): '''simple docstring''' class _a (__magic_name__ ): '''simple docstring''' class _a (__magic_name__ ): '''simple docstring''' def UpperCamelCase (lowercase_: Optional[dict] , lowercase_: dict ) -> int: if expected_splits is None: logger.info("""Unable to verify splits sizes.""" ) return if len(set(lowercase_ ) - set(lowercase_ ) ) > 0: raise ExpectedMoreSplits(str(set(lowercase_ ) - set(lowercase_ ) ) ) if len(set(lowercase_ ) - set(lowercase_ ) ) > 0: raise UnexpectedSplits(str(set(lowercase_ ) - set(lowercase_ ) ) ) A__ : int = [ {"""expected""": expected_splits[name], """recorded""": recorded_splits[name]} for name in expected_splits if expected_splits[name].num_examples != recorded_splits[name].num_examples ] if len(lowercase_ ) > 0: raise NonMatchingSplitsSizesError(str(lowercase_ ) ) logger.info("""All the splits matched successfully.""" ) def UpperCamelCase (lowercase_: str , lowercase_: bool = True ) -> dict: if record_checksum: A__ : List[Any] = shaaaa() with open(lowercase_ , """rb""" ) as f: for chunk in iter(lambda: f.read(1 << 20 ) , b"""""" ): m.update(lowercase_ ) A__ : Any = m.hexdigest() else: A__ : Union[str, Any] = None return {"num_bytes": os.path.getsize(lowercase_ ), "checksum": checksum} def UpperCamelCase (lowercase_: Optional[Any] ) -> List[str]: if dataset_size and config.IN_MEMORY_MAX_SIZE: return dataset_size < config.IN_MEMORY_MAX_SIZE else: return False
64
import os import tempfile from functools import partial from unittest import TestCase from unittest.mock import patch import datasets import datasets.config from .utils import require_beam class _a (datasets.BeamBasedBuilder ): '''simple docstring''' def __A ( self ): return datasets.DatasetInfo( features=datasets.Features({"""content""": datasets.Value("""string""" )} ) , supervised_keys=A__ , ) def __A ( self , A__ , A__ ): return [datasets.SplitGenerator(name=datasets.Split.TRAIN , gen_kwargs={"""examples""": get_test_dummy_examples()} )] def __A ( self , A__ , A__ ): import apache_beam as beam return pipeline | "Load Examples" >> beam.Create(A__ ) class _a (datasets.BeamBasedBuilder ): '''simple docstring''' def __A ( self ): return datasets.DatasetInfo( features=datasets.Features({"""a""": datasets.Sequence({"""b""": datasets.Value("""string""" )} )} ) , supervised_keys=A__ , ) def __A ( self , A__ , A__ ): return [ datasets.SplitGenerator(name=datasets.Split.TRAIN , gen_kwargs={"""examples""": get_test_nested_examples()} ) ] def __A ( self , A__ , A__ ): import apache_beam as beam return pipeline | "Load Examples" >> beam.Create(A__ ) def UpperCamelCase () -> Dict: return [(i, {"content": content}) for i, content in enumerate(["""foo""", """bar""", """foobar"""] )] def UpperCamelCase () -> Tuple: return [(i, {"a": {"b": [content]}}) for i, content in enumerate(["""foo""", """bar""", """foobar"""] )] class _a (__magic_name__ ): '''simple docstring''' @require_beam def __A ( self ): A__ : Dict = len(get_test_dummy_examples() ) with tempfile.TemporaryDirectory() as tmp_cache_dir: A__ : int = DummyBeamDataset(cache_dir=A__ , beam_runner="""DirectRunner""" ) builder.download_and_prepare() self.assertTrue( os.path.exists( os.path.join(A__ , builder.name , """default""" , """0.0.0""" , F"""{builder.name}-train.arrow""" ) ) ) self.assertDictEqual(builder.info.features , datasets.Features({"""content""": datasets.Value("""string""" )} ) ) A__ : int = builder.as_dataset() self.assertEqual(dset["""train"""].num_rows , A__ ) self.assertEqual(dset["""train"""].info.splits["""train"""].num_examples , A__ ) self.assertDictEqual(dset["""train"""][0] , get_test_dummy_examples()[0][1] ) self.assertDictEqual( dset["""train"""][expected_num_examples - 1] , get_test_dummy_examples()[expected_num_examples - 1][1] ) self.assertTrue( os.path.exists(os.path.join(A__ , builder.name , """default""" , """0.0.0""" , """dataset_info.json""" ) ) ) del dset @require_beam def __A ( self ): import apache_beam as beam A__ : int = beam.io.parquetio.WriteToParquet A__ : List[str] = len(get_test_dummy_examples() ) with tempfile.TemporaryDirectory() as tmp_cache_dir: A__ : str = DummyBeamDataset(cache_dir=A__ , beam_runner="""DirectRunner""" ) with patch("""apache_beam.io.parquetio.WriteToParquet""" ) as write_parquet_mock: A__ : Optional[Any] = partial(A__ , num_shards=2 ) builder.download_and_prepare() self.assertTrue( os.path.exists( os.path.join( A__ , builder.name , """default""" , """0.0.0""" , F"""{builder.name}-train-00000-of-00002.arrow""" ) ) ) self.assertTrue( os.path.exists( os.path.join( A__ , builder.name , """default""" , """0.0.0""" , F"""{builder.name}-train-00000-of-00002.arrow""" ) ) ) self.assertDictEqual(builder.info.features , datasets.Features({"""content""": datasets.Value("""string""" )} ) ) A__ : Optional[int] = builder.as_dataset() self.assertEqual(dset["""train"""].num_rows , A__ ) self.assertEqual(dset["""train"""].info.splits["""train"""].num_examples , A__ ) # Order is not preserved when sharding, so we just check that all the elements are there self.assertListEqual(sorted(dset["""train"""]["""content"""] ) , sorted(["""foo""", """bar""", """foobar"""] ) ) self.assertTrue( os.path.exists(os.path.join(A__ , builder.name , """default""" , """0.0.0""" , """dataset_info.json""" ) ) ) del dset @require_beam def __A ( self ): with tempfile.TemporaryDirectory() as tmp_cache_dir: A__ : int = DummyBeamDataset(cache_dir=A__ ) self.assertRaises(datasets.builder.MissingBeamOptions , builder.download_and_prepare ) @require_beam def __A ( self ): A__ : List[Any] = len(get_test_nested_examples() ) with tempfile.TemporaryDirectory() as tmp_cache_dir: A__ : Optional[int] = NestedBeamDataset(cache_dir=A__ , beam_runner="""DirectRunner""" ) builder.download_and_prepare() self.assertTrue( os.path.exists( os.path.join(A__ , builder.name , """default""" , """0.0.0""" , F"""{builder.name}-train.arrow""" ) ) ) self.assertDictEqual( builder.info.features , datasets.Features({"""a""": datasets.Sequence({"""b""": datasets.Value("""string""" )} )} ) ) A__ : Optional[int] = builder.as_dataset() self.assertEqual(dset["""train"""].num_rows , A__ ) self.assertEqual(dset["""train"""].info.splits["""train"""].num_examples , A__ ) self.assertDictEqual(dset["""train"""][0] , get_test_nested_examples()[0][1] ) self.assertDictEqual( dset["""train"""][expected_num_examples - 1] , get_test_nested_examples()[expected_num_examples - 1][1] ) self.assertTrue( os.path.exists(os.path.join(A__ , builder.name , """default""" , """0.0.0""" , """dataset_info.json""" ) ) ) del dset
64
1
from .imports import is_rich_available if is_rich_available(): from rich.traceback import install install(show_locals=False) else: raise ModuleNotFoundError('To use the rich extension, install rich with `pip install rich`')
64
import warnings from ...utils import logging from .image_processing_poolformer import PoolFormerImageProcessor A_ : Union[str, Any] = logging.get_logger(__name__) class _a (__magic_name__ ): '''simple docstring''' def __init__( self , *A__ , **A__ ): warnings.warn( """The class PoolFormerFeatureExtractor is deprecated and will be removed in version 5 of Transformers.""" """ Please use PoolFormerImageProcessor instead.""" , A__ , ) super().__init__(*A__ , **A__ )
64
1
from __future__ import annotations import typing from collections.abc import Iterable import numpy as np A_ : Optional[Any] = typing.Union[Iterable[float], Iterable[int], np.ndarray] # noqa: UP007 A_ : Optional[Any] = typing.Union[np.floataa, int, float] # noqa: UP007 def UpperCamelCase (lowercase_: Vector , lowercase_: Vector ) -> VectorOut: return np.sqrt(np.sum((np.asarray(lowercase_ ) - np.asarray(lowercase_ )) ** 2 ) ) def UpperCamelCase (lowercase_: Vector , lowercase_: Vector ) -> VectorOut: return sum((va - va) ** 2 for va, va in zip(lowercase_ , lowercase_ ) ) ** (1 / 2) if __name__ == "__main__": def UpperCamelCase () -> None: from timeit import timeit print("""Without Numpy""" ) print( timeit( """euclidean_distance_no_np([1, 2, 3], [4, 5, 6])""" , number=10000 , globals=globals() , ) ) print("""With Numpy""" ) print( timeit( """euclidean_distance([1, 2, 3], [4, 5, 6])""" , number=10000 , globals=globals() , ) ) benchmark()
64
import inspect import logging import os import random import shutil import tempfile import unittest import pytest import torch from torch import nn from torch.utils.data import DataLoader, TensorDataset from accelerate import Accelerator from accelerate.test_utils import execute_subprocess_async, require_cuda from accelerate.utils import ProjectConfiguration, set_seed A_ : Any = logging.getLogger(__name__) def UpperCamelCase (lowercase_: Optional[Any]=2 , lowercase_: Union[str, Any]=3 , lowercase_: int=16 , lowercase_: int = 10 , lowercase_: int = 2 ) -> int: def get_dataset(lowercase_: Optional[int] ): A__ : Optional[Any] = torch.randn(batch_size * n_batches , 1 ) return TensorDataset(lowercase_ , a * x + b + 0.1 * torch.randn(batch_size * n_batches , 1 ) ) A__ : Dict = get_dataset(lowercase_ ) A__ : Any = get_dataset(lowercase_ ) A__ : Dict = DataLoader(lowercase_ , shuffle=lowercase_ , batch_size=lowercase_ , num_workers=4 ) A__ : Optional[Any] = DataLoader(lowercase_ , shuffle=lowercase_ , batch_size=lowercase_ , num_workers=4 ) return (train_dataloader, valid_dataloader) def UpperCamelCase (lowercase_: Optional[Any] , lowercase_: List[str] , lowercase_: int , lowercase_: int , lowercase_: List[str] , lowercase_: Dict=None ) -> List[Any]: A__ : List[Any] = [] for epoch in range(lowercase_ ): # Train quickly model.train() for batch in dataloader: A__ , A__ : Any = batch A__ : Any = model(lowercase_ ) A__ : Any = torch.nn.functional.mse_loss(lowercase_ , lowercase_ ) accelerator.backward(lowercase_ ) optimizer.step() optimizer.zero_grad() rands.append(random.random() ) # Introduce some randomness if scheduler is not None: scheduler.step() return rands class _a (nn.Module ): '''simple docstring''' def __init__( self ): super().__init__() A__ : str = nn.Parameter(torch.randn(1 ) ) A__ : Any = nn.Parameter(torch.randn(1 ) ) def __A ( self , A__ ): return x * self.a + self.b class _a (unittest.TestCase ): '''simple docstring''' def __A ( self ): with tempfile.TemporaryDirectory() as tmpdir: set_seed(42 ) A__ : Optional[Any] = DummyModel() A__ : Optional[Any] = torch.optim.Adam(params=model.parameters() , lr=1e-3 ) A__ , A__ : str = dummy_dataloaders() A__ : Dict = ProjectConfiguration(total_limit=1 , project_dir=A__ , automatic_checkpoint_naming=A__ ) # Train baseline A__ : List[str] = Accelerator(project_config=A__ ) A__ , A__ , A__ , A__ : Any = accelerator.prepare( A__ , A__ , A__ , A__ ) # Save initial accelerator.save_state() # Save second state accelerator.save_state() self.assertEqual(len(os.listdir(accelerator.project_dir ) ) , 1 ) def __A ( self ): with tempfile.TemporaryDirectory() as tmpdir: set_seed(42 ) A__ : str = DummyModel() A__ : Optional[int] = torch.optim.Adam(params=model.parameters() , lr=1e-3 ) A__ , A__ : int = dummy_dataloaders() # Train baseline A__ : str = Accelerator() A__ , A__ , A__ , A__ : List[str] = accelerator.prepare( A__ , A__ , A__ , A__ ) # Save initial A__ : List[Any] = os.path.join(A__ , """initial""" ) accelerator.save_state(A__ ) ((A__) , (A__)) : str = model.a.item(), model.b.item() A__ : Dict = optimizer.state_dict() A__ : List[str] = train(3 , A__ , A__ , A__ , A__ ) ((A__) , (A__)) : str = model.a.item(), model.b.item() A__ : Any = optimizer.state_dict() # Train partially set_seed(42 ) A__ : Optional[int] = DummyModel() A__ : Dict = torch.optim.Adam(params=model.parameters() , lr=1e-3 ) A__ , A__ : Dict = dummy_dataloaders() A__ : List[str] = Accelerator() A__ , A__ , A__ , A__ : Optional[Any] = accelerator.prepare( A__ , A__ , A__ , A__ ) accelerator.load_state(A__ ) ((A__) , (A__)) : Tuple = model.a.item(), model.b.item() A__ : Union[str, Any] = optimizer.state_dict() self.assertEqual(A__ , A__ ) self.assertEqual(A__ , A__ ) self.assertEqual(A__ , A__ ) A__ : List[str] = train(2 , A__ , A__ , A__ , A__ ) # Save everything A__ : Optional[int] = os.path.join(A__ , """checkpoint""" ) accelerator.save_state(A__ ) # Load everything back in and make sure all states work accelerator.load_state(A__ ) test_rands += train(1 , A__ , A__ , A__ , A__ ) ((A__) , (A__)) : Union[str, Any] = model.a.item(), model.b.item() A__ : Optional[int] = optimizer.state_dict() self.assertEqual(A__ , A__ ) self.assertEqual(A__ , A__ ) self.assertEqual(A__ , A__ ) self.assertEqual(A__ , A__ ) def __A ( self ): with tempfile.TemporaryDirectory() as tmpdir: set_seed(42 ) A__ : int = DummyModel() A__ : Optional[int] = torch.optim.Adam(params=model.parameters() , lr=1e-3 ) A__ , A__ : List[str] = dummy_dataloaders() A__ : str = ProjectConfiguration(automatic_checkpoint_naming=A__ ) # Train baseline A__ : Any = Accelerator(project_dir=A__ , project_config=A__ ) A__ , A__ , A__ , A__ : str = accelerator.prepare( A__ , A__ , A__ , A__ ) # Save initial accelerator.save_state() ((A__) , (A__)) : Tuple = model.a.item(), model.b.item() A__ : int = optimizer.state_dict() A__ : int = train(3 , A__ , A__ , A__ , A__ ) ((A__) , (A__)) : Optional[Any] = model.a.item(), model.b.item() A__ : Any = optimizer.state_dict() # Train partially set_seed(42 ) A__ : Dict = DummyModel() A__ : List[Any] = torch.optim.Adam(params=model.parameters() , lr=1e-3 ) A__ , A__ : Union[str, Any] = dummy_dataloaders() A__ : List[Any] = ProjectConfiguration(iteration=1 , automatic_checkpoint_naming=A__ ) A__ : Dict = Accelerator(project_dir=A__ , project_config=A__ ) A__ , A__ , A__ , A__ : Union[str, Any] = accelerator.prepare( A__ , A__ , A__ , A__ ) accelerator.load_state(os.path.join(A__ , """checkpoints""" , """checkpoint_0""" ) ) ((A__) , (A__)) : Optional[int] = model.a.item(), model.b.item() A__ : Tuple = optimizer.state_dict() self.assertEqual(A__ , A__ ) self.assertEqual(A__ , A__ ) self.assertEqual(A__ , A__ ) A__ : str = train(2 , A__ , A__ , A__ , A__ ) # Save everything accelerator.save_state() # Load everything back in and make sure all states work accelerator.load_state(os.path.join(A__ , """checkpoints""" , """checkpoint_1""" ) ) test_rands += train(1 , A__ , A__ , A__ , A__ ) ((A__) , (A__)) : Optional[int] = model.a.item(), model.b.item() A__ : List[Any] = optimizer.state_dict() self.assertEqual(A__ , A__ ) self.assertEqual(A__ , A__ ) self.assertEqual(A__ , A__ ) self.assertEqual(A__ , A__ ) def __A ( self ): A__ : Union[str, Any] = torch.tensor([1, 2, 3] ) A__ : int = torch.tensor([2, 3, 4] ) A__ : List[Any] = DummyModel() A__ : List[Any] = torch.optim.Adam(net.parameters() ) A__ : Tuple = Accelerator() with self.assertRaises(A__ ) as ve: accelerator.register_for_checkpointing(A__ , A__ , A__ , A__ ) A__ : Any = str(ve.exception ) self.assertTrue("""Item at index 0""" in message ) self.assertTrue("""Item at index 1""" in message ) self.assertFalse("""Item at index 2""" in message ) self.assertFalse("""Item at index 3""" in message ) def __A ( self ): with tempfile.TemporaryDirectory() as tmpdir: set_seed(42 ) A__ : Any = DummyModel() A__ : Union[str, Any] = torch.optim.Adam(params=model.parameters() , lr=1e-3 ) A__ : Dict = torch.optim.lr_scheduler.StepLR(A__ , step_size=1 , gamma=0.9_9 ) A__ , A__ : List[Any] = dummy_dataloaders() A__ : Tuple = ProjectConfiguration(automatic_checkpoint_naming=A__ ) # Train baseline A__ : Optional[Any] = Accelerator(project_dir=A__ , project_config=A__ ) A__ , A__ , A__ , A__ , A__ : Union[str, Any] = accelerator.prepare( A__ , A__ , A__ , A__ , A__ ) # Save initial accelerator.save_state() A__ : Tuple = scheduler.state_dict() train(3 , A__ , A__ , A__ , A__ , A__ ) self.assertNotEqual(A__ , scheduler.state_dict() ) # Load everything back in and make sure all states work accelerator.load_state(os.path.join(A__ , """checkpoints""" , """checkpoint_0""" ) ) self.assertEqual(A__ , scheduler.state_dict() ) def __A ( self ): with tempfile.TemporaryDirectory() as tmpdir: set_seed(42 ) A__ : Optional[Any] = DummyModel() A__ : int = ProjectConfiguration(automatic_checkpoint_naming=A__ , total_limit=2 ) # Train baseline A__ : List[str] = Accelerator(project_dir=A__ , project_config=A__ ) A__ : Union[str, Any] = accelerator.prepare(A__ ) # Save 3 states: for _ in range(11 ): accelerator.save_state() self.assertTrue(not os.path.exists(os.path.join(A__ , """checkpoints""" , """checkpoint_0""" ) ) ) self.assertTrue(os.path.exists(os.path.join(A__ , """checkpoints""" , """checkpoint_9""" ) ) ) self.assertTrue(os.path.exists(os.path.join(A__ , """checkpoints""" , """checkpoint_10""" ) ) ) @require_cuda def __A ( self ): A__ : Dict = ["""torchrun""", F"""--nproc_per_node={torch.cuda.device_count()}""", inspect.getfile(self.__class__ )] execute_subprocess_async(A__ , env=os.environ.copy() ) if __name__ == "__main__": A_ : List[str] = '/tmp/accelerate/state_checkpointing' A_ : Optional[Any] = DummyModel() A_ : Union[str, Any] = torch.optim.Adam(params=model.parameters(), lr=1E-3) A_ : str = torch.optim.lr_scheduler.StepLR(optimizer, step_size=1, gamma=0.99) A_ , A_ : List[Any] = dummy_dataloaders() A_ : int = ProjectConfiguration(automatic_checkpoint_naming=True) # Train baseline A_ : List[str] = Accelerator(project_dir=savedir, project_config=project_config, mixed_precision='no') if accelerator.process_index == 0: if os.path.exists(savedir): shutil.rmtree(savedir) os.makedirs(savedir) A_ , A_ , A_ , A_ , A_ : List[Any] = accelerator.prepare( model, optimizer, train_dataloader, valid_dataloader, scheduler ) A_ , A_ : Dict = accelerator.prepare(model, optimizer) train(3, model, train_dataloader, optimizer, accelerator, scheduler) # Check that the intial optimizer is loaded on the GPU for group in optimizer.param_groups: A_ : str = group['params'][0].device break assert param_device.type == accelerator.device.type A_ : Optional[Any] = model.cpu() accelerator.wait_for_everyone() accelerator.save_state() accelerator.wait_for_everyone() # Check CPU state accelerator.load_state(os.path.join(savedir, 'checkpoints', 'checkpoint_0'), map_location='cpu') for group in optimizer.param_groups: A_ : str = group['params'][0].device break assert ( param_device.type == torch.device('cpu').type ), f"Loaded optimizer states did not match, expected to be loaded on the CPU but got {param_device}" # Check device state model.to(accelerator.device) accelerator.load_state(os.path.join(savedir, 'checkpoints', 'checkpoint_0'), map_location='on_device') for group in optimizer.param_groups: A_ : Tuple = group['params'][0].device break assert ( param_device.type == accelerator.device.type ), f"Loaded optimizer states did not match, expected to be loaded on {accelerator.device} but got {param_device}" # Check error with pytest.raises(TypeError, match='Unsupported optimizer map location passed'): accelerator.load_state(os.path.join(savedir, 'checkpoints', 'checkpoint_0'), map_location='invalid') accelerator.wait_for_everyone() if accelerator.process_index == 0: shutil.rmtree(savedir) accelerator.wait_for_everyone()
64
1
import unittest from typing import Tuple import torch from diffusers.utils import floats_tensor, randn_tensor, torch_all_close, torch_device from diffusers.utils.testing_utils import require_torch @require_torch class _a : '''simple docstring''' @property def __A ( self ): return self.get_dummy_input() @property def __A ( self ): if self.block_type == "down": return (4, 32, 16, 16) elif self.block_type == "mid": return (4, 32, 32, 32) elif self.block_type == "up": return (4, 32, 64, 64) raise ValueError(F"""'{self.block_type}' is not a supported block_type. Set it to 'up', 'mid', or 'down'.""" ) def __A ( self , A__=True , A__=False , A__=False , A__=False , ): A__ : str = 4 A__ : List[str] = 32 A__ : str = (32, 32) A__ : List[str] = torch.manual_seed(0 ) A__ : Tuple = torch.device(A__ ) A__ : str = (batch_size, num_channels) + sizes A__ : int = randn_tensor(A__ , generator=A__ , device=A__ ) A__ : Union[str, Any] = {"""hidden_states""": hidden_states} if include_temb: A__ : List[Any] = 128 A__ : Dict = randn_tensor((batch_size, temb_channels) , generator=A__ , device=A__ ) if include_res_hidden_states_tuple: A__ : Optional[int] = torch.manual_seed(1 ) A__ : List[Any] = (randn_tensor(A__ , generator=A__ , device=A__ ),) if include_encoder_hidden_states: A__ : List[Any] = floats_tensor((batch_size, 32, 32) ).to(A__ ) if include_skip_sample: A__ : Dict = randn_tensor(((batch_size, 3) + sizes) , generator=A__ , device=A__ ) return dummy_input def __A ( self ): A__ : Dict = { """in_channels""": 32, """out_channels""": 32, """temb_channels""": 128, } if self.block_type == "up": A__ : Union[str, Any] = 32 if self.block_type == "mid": init_dict.pop("""out_channels""" ) A__ : Any = self.dummy_input return init_dict, inputs_dict def __A ( self , A__ ): A__ , A__ : Optional[Any] = self.prepare_init_args_and_inputs_for_common() A__ : Optional[Any] = self.block_class(**A__ ) unet_block.to(A__ ) unet_block.eval() with torch.no_grad(): A__ : Union[str, Any] = unet_block(**A__ ) if isinstance(A__ , A__ ): A__ : Any = output[0] self.assertEqual(output.shape , self.output_shape ) A__ : str = output[0, -1, -3:, -3:] A__ : List[Any] = torch.tensor(A__ ).to(A__ ) assert torch_all_close(output_slice.flatten() , A__ , atol=5e-3 ) @unittest.skipIf(torch_device == """mps""" , """Training is not supported in mps""" ) def __A ( self ): A__ , A__ : Optional[int] = self.prepare_init_args_and_inputs_for_common() A__ : str = self.block_class(**A__ ) model.to(A__ ) model.train() A__ : Optional[int] = model(**A__ ) if isinstance(A__ , A__ ): A__ : Tuple = output[0] A__ : Tuple = torch.device(A__ ) A__ : str = randn_tensor(output.shape , device=A__ ) A__ : Union[str, Any] = torch.nn.functional.mse_loss(A__ , A__ ) loss.backward()
64
def UpperCamelCase (lowercase_: str , lowercase_: str ) -> bool: A__ : Union[str, Any] = len(lowercase_ ) A__ : List[Any] = len(lowercase_ ) A__ : List[Any] = [[False for _ in range(m + 1 )] for _ in range(n + 1 )] A__ : str = True for i in range(lowercase_ ): for j in range(m + 1 ): if dp[i][j]: if j < m and a[i].upper() == b[j]: A__ : int = True if a[i].islower(): A__ : Dict = True return dp[n][m] if __name__ == "__main__": import doctest doctest.testmod()
64
1
# Copyright 2023 The HuggingFace Inc. team. 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 copy of the License at # # http://www.apache.org/licenses/LICENSE-2.0 # # Unless required by applicable law or agreed to in writing, software # distributed under the License is distributed on an "AS IS" BASIS, # WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. # See the License for the specific language governing permissions and # limitations under the License. import re from ..models.auto import AutoProcessor from ..models.vision_encoder_decoder import VisionEncoderDecoderModel from ..utils import is_vision_available from .base import PipelineTool if is_vision_available(): from PIL import Image class _a (__magic_name__ ): '''simple docstring''' UpperCAmelCase__: int = '''naver-clova-ix/donut-base-finetuned-docvqa''' UpperCAmelCase__: Optional[int] = ( '''This is a tool that answers a question about an document (pdf). It takes an input named `document` which ''' '''should be the document containing the information, as well as a `question` that is the question about the ''' '''document. It returns a text that contains the answer to the question.''' ) UpperCAmelCase__: int = '''document_qa''' UpperCAmelCase__: Union[str, Any] = AutoProcessor UpperCAmelCase__: Optional[Any] = VisionEncoderDecoderModel UpperCAmelCase__: Union[str, Any] = ['''image''', '''text'''] UpperCAmelCase__: Union[str, Any] = ['''text'''] def __init__( self , *A__ , **A__ ): if not is_vision_available(): raise ValueError("""Pillow must be installed to use the DocumentQuestionAnsweringTool.""" ) super().__init__(*A__ , **A__ ) def __A ( self , A__ , A__ ): A__ : Optional[int] = """<s_docvqa><s_question>{user_input}</s_question><s_answer>""" A__ : List[str] = task_prompt.replace("""{user_input}""" , A__ ) A__ : Optional[Any] = self.pre_processor.tokenizer( A__ , add_special_tokens=A__ , return_tensors="""pt""" ).input_ids A__ : int = self.pre_processor(A__ , return_tensors="""pt""" ).pixel_values return {"decoder_input_ids": decoder_input_ids, "pixel_values": pixel_values} def __A ( self , A__ ): return self.model.generate( inputs["""pixel_values"""].to(self.device ) , decoder_input_ids=inputs["""decoder_input_ids"""].to(self.device ) , max_length=self.model.decoder.config.max_position_embeddings , early_stopping=A__ , pad_token_id=self.pre_processor.tokenizer.pad_token_id , eos_token_id=self.pre_processor.tokenizer.eos_token_id , use_cache=A__ , num_beams=1 , bad_words_ids=[[self.pre_processor.tokenizer.unk_token_id]] , return_dict_in_generate=A__ , ).sequences def __A ( self , A__ ): A__ : Optional[int] = self.pre_processor.batch_decode(A__ )[0] A__ : Optional[Any] = sequence.replace(self.pre_processor.tokenizer.eos_token , """""" ) A__ : Optional[int] = sequence.replace(self.pre_processor.tokenizer.pad_token , """""" ) A__ : List[Any] = re.sub(r"""<.*?>""" , """""" , A__ , count=1 ).strip() # remove first task start token A__ : Tuple = self.pre_processor.tokenajson(A__ ) return sequence["answer"]
64
import itertools import random import unittest import numpy as np from transformers import ASTFeatureExtractor from transformers.testing_utils import require_torch, require_torchaudio from transformers.utils.import_utils import is_torch_available from ...test_sequence_feature_extraction_common import SequenceFeatureExtractionTestMixin A_ : Dict = random.Random() if is_torch_available(): import torch def UpperCamelCase (lowercase_: Tuple , lowercase_: Tuple=1.0 , lowercase_: Dict=None , lowercase_: int=None ) -> str: if rng is None: A__ : Optional[Any] = global_rng A__ : List[str] = [] for batch_idx in range(shape[0] ): values.append([] ) for _ in range(shape[1] ): values[-1].append(rng.random() * scale ) return values class _a (unittest.TestCase ): '''simple docstring''' def __init__( self , A__ , A__=7 , A__=400 , A__=2000 , A__=1 , A__=0.0 , A__=1_6000 , A__=True , A__=True , ): A__ : Any = parent A__ : Optional[int] = batch_size A__ : Union[str, Any] = min_seq_length A__ : Dict = max_seq_length A__ : Tuple = (self.max_seq_length - self.min_seq_length) // (self.batch_size - 1) A__ : str = feature_size A__ : Optional[int] = padding_value A__ : List[str] = sampling_rate A__ : List[str] = return_attention_mask A__ : int = do_normalize def __A ( self ): return { "feature_size": self.feature_size, "padding_value": self.padding_value, "sampling_rate": self.sampling_rate, "return_attention_mask": self.return_attention_mask, "do_normalize": self.do_normalize, } def __A ( self , A__=False , A__=False ): def _flatten(A__ ): return list(itertools.chain(*A__ ) ) if equal_length: A__ : Dict = floats_list((self.batch_size, self.max_seq_length) ) else: # make sure that inputs increase in size A__ : Union[str, Any] = [ _flatten(floats_list((x, self.feature_size) ) ) for x in range(self.min_seq_length , self.max_seq_length , self.seq_length_diff ) ] if numpify: A__ : Optional[int] = [np.asarray(A__ ) for x in speech_inputs] return speech_inputs @require_torch @require_torchaudio class _a (__magic_name__ , unittest.TestCase ): '''simple docstring''' UpperCAmelCase__: int = ASTFeatureExtractor def __A ( self ): A__ : Optional[Any] = ASTFeatureExtractionTester(self ) def __A ( self ): # Tests that all call wrap to encode_plus and batch_encode_plus A__ : Optional[int] = self.feature_extraction_class(**self.feat_extract_tester.prepare_feat_extract_dict() ) # create three inputs of length 800, 1000, and 1200 A__ : List[Any] = [floats_list((1, x) )[0] for x in range(800 , 1400 , 200 )] A__ : Optional[Any] = [np.asarray(A__ ) for speech_input in speech_inputs] # Test not batched input A__ : Tuple = feat_extract(speech_inputs[0] , return_tensors="""np""" ).input_values A__ : Tuple = feat_extract(np_speech_inputs[0] , return_tensors="""np""" ).input_values self.assertTrue(np.allclose(A__ , A__ , atol=1e-3 ) ) # Test batched A__ : Tuple = feat_extract(A__ , padding=A__ , return_tensors="""np""" ).input_values A__ : Tuple = feat_extract(A__ , padding=A__ , return_tensors="""np""" ).input_values for enc_seq_a, enc_seq_a in zip(A__ , A__ ): self.assertTrue(np.allclose(A__ , A__ , atol=1e-3 ) ) # Test 2-D numpy arrays are batched. A__ : int = [floats_list((1, x) )[0] for x in (800, 800, 800)] A__ : List[str] = np.asarray(A__ ) A__ : Union[str, Any] = feat_extract(A__ , return_tensors="""np""" ).input_values A__ : Optional[Any] = feat_extract(A__ , return_tensors="""np""" ).input_values for enc_seq_a, enc_seq_a in zip(A__ , A__ ): self.assertTrue(np.allclose(A__ , A__ , atol=1e-3 ) ) @require_torch def __A ( self ): import torch A__ : Union[str, Any] = self.feature_extraction_class(**self.feat_extract_tester.prepare_feat_extract_dict() ) A__ : Tuple = np.random.rand(100 ).astype(np.floataa ) A__ : Tuple = np_speech_inputs.tolist() for inputs in [py_speech_inputs, np_speech_inputs]: A__ : List[str] = feature_extractor.pad([{"""input_values""": inputs}] , return_tensors="""np""" ) self.assertTrue(np_processed.input_values.dtype == np.floataa ) A__ : Any = feature_extractor.pad([{"""input_values""": inputs}] , return_tensors="""pt""" ) self.assertTrue(pt_processed.input_values.dtype == torch.floataa ) def __A ( self , A__ ): from datasets import load_dataset A__ : str = load_dataset("""hf-internal-testing/librispeech_asr_dummy""" , """clean""" , split="""validation""" ) # automatic decoding with librispeech A__ : str = ds.sort("""id""" ).select(range(A__ ) )[:num_samples]["""audio"""] return [x["array"] for x in speech_samples] @require_torch def __A ( self ): # fmt: off A__ : Optional[Any] = torch.tensor( [-0.9_8_9_4, -1.2_7_7_6, -0.9_0_6_6, -1.2_7_7_6, -0.9_3_4_9, -1.2_6_0_9, -1.0_3_8_6, -1.2_7_7_6, -1.1_5_6_1, -1.2_7_7_6, -1.2_0_5_2, -1.2_7_2_3, -1.2_1_9_0, -1.2_1_3_2, -1.2_7_7_6, -1.1_1_3_3, -1.1_9_5_3, -1.1_3_4_3, -1.1_5_8_4, -1.2_2_0_3, -1.1_7_7_0, -1.2_4_7_4, -1.2_3_8_1, -1.1_9_3_6, -0.9_2_7_0, -0.8_3_1_7, -0.8_0_4_9, -0.7_7_0_6, -0.7_5_6_5, -0.7_8_6_9] ) # fmt: on A__ : Any = self._load_datasamples(1 ) A__ : Tuple = ASTFeatureExtractor() A__ : Dict = feature_extractor(A__ , return_tensors="""pt""" ).input_values self.assertEquals(input_values.shape , (1, 1024, 128) ) self.assertTrue(torch.allclose(input_values[0, 0, :30] , A__ , atol=1e-4 ) )
64
1
import gc import importlib.metadata import tempfile import unittest from packaging import version from transformers import ( AutoModel, AutoModelForCausalLM, AutoModelForSeqaSeqLM, AutoModelForSequenceClassification, AutoTokenizer, BitsAndBytesConfig, pipeline, ) from transformers.testing_utils import ( is_torch_available, require_accelerate, require_bitsandbytes, require_torch, require_torch_gpu, require_torch_multi_gpu, slow, ) def UpperCamelCase (lowercase_: int ) -> Optional[int]: if model.config.model_type == "gpt2": return model.transformer.h[0].mlp.c_fc return model.transformer.h[0].mlp.dense_ah_to_h if is_torch_available(): import torch import torch.nn as nn class _a (nn.Module ): '''simple docstring''' def __init__( self , A__ , A__ ): super().__init__() A__ : Optional[int] = module A__ : List[Any] = nn.Sequential( nn.Linear(module.in_features , A__ , bias=A__ ) , nn.Linear(A__ , module.out_features , bias=A__ ) , ) A__ : List[Any] = (2.0 / (5 * min(module.in_features , module.out_features ))) ** 0.5 nn.init.normal_(self.adapter[0].weight , std=A__ ) nn.init.zeros_(self.adapter[1].weight ) self.adapter.to(module.weight.device ) def __A ( self , A__ , *A__ , **A__ ): return self.module(A__ , *A__ , **A__ ) + self.adapter(A__ ) @require_bitsandbytes @require_accelerate @require_torch @require_torch_gpu @slow class _a (unittest.TestCase ): '''simple docstring''' UpperCAmelCase__: int = '''bigscience/bloom-1b7''' # Constant values UpperCAmelCase__: List[Any] = 2.1_09_65_95_52_69_25_74 UpperCAmelCase__: str = '''Hello my name is''' UpperCAmelCase__: List[Any] = set() EXPECTED_OUTPUTS.add('''Hello my name is John and I am a professional photographer. I''' ) EXPECTED_OUTPUTS.add('''Hello my name is John.\nI am a friend of your father.\n''' ) EXPECTED_OUTPUTS.add('''Hello my name is John Doe, I am a student at the University''' ) UpperCAmelCase__: List[str] = 10 def __A ( self ): # Models and tokenizer A__ : Any = AutoTokenizer.from_pretrained(self.model_name ) class _a (__magic_name__ ): '''simple docstring''' def __A ( self ): super().setUp() # Models and tokenizer A__ : Any = AutoModelForCausalLM.from_pretrained( self.model_name , torch_dtype=torch.floataa , device_map="""auto""" ) A__ : Dict = AutoModelForCausalLM.from_pretrained(self.model_name , load_in_abit=A__ , device_map="""auto""" ) def __A ( self ): del self.model_fpaa del self.model_abit gc.collect() torch.cuda.empty_cache() def __A ( self ): A__ : int = self.model_abit.config self.assertTrue(hasattr(A__ , """quantization_config""" ) ) A__ : Optional[int] = config.to_dict() A__ : str = config.to_diff_dict() A__ : List[str] = config.to_json_string() def __A ( self ): from bitsandbytes.nn import Paramsabit A__ : Any = self.model_fpaa.get_memory_footprint() A__ : Union[str, Any] = self.model_abit.get_memory_footprint() self.assertAlmostEqual(mem_fpaa / mem_abit , self.EXPECTED_RELATIVE_DIFFERENCE ) A__ : Optional[int] = get_some_linear_layer(self.model_abit ) self.assertTrue(linear.weight.__class__ == Paramsabit ) def __A ( self ): from transformers import TaPreTrainedModel self.model_fpaa.get_memory_footprint() self.model_abit.get_memory_footprint() for name, module in self.model_abit.named_modules(): if isinstance(A__ , torch.nn.Linear ): if name not in ["lm_head"] + TaPreTrainedModel._keep_in_fpaa_modules: # 4-bit parameters are packed in uint8 variables self.assertTrue(module.weight.dtype == torch.uinta ) def __A ( self ): A__ : str = self.tokenizer(self.input_text , return_tensors="""pt""" ) A__ : Tuple = self.model_abit.generate(input_ids=encoded_input["""input_ids"""].to(0 ) , max_new_tokens=10 ) self.assertIn(self.tokenizer.decode(output_sequences[0] , skip_special_tokens=A__ ) , self.EXPECTED_OUTPUTS ) def __A ( self ): A__ : Tuple = BitsAndBytesConfig() A__ : Dict = True A__ : List[str] = AutoModelForCausalLM.from_pretrained( self.model_name , quantization_config=A__ , device_map="""auto""" ) A__ : str = self.tokenizer(self.input_text , return_tensors="""pt""" ) A__ : Dict = model_abit_from_config.generate( input_ids=encoded_input["""input_ids"""].to(0 ) , max_new_tokens=10 ) self.assertIn(self.tokenizer.decode(output_sequences[0] , skip_special_tokens=A__ ) , self.EXPECTED_OUTPUTS ) def __A ( self ): with self.assertRaises(A__ ), tempfile.TemporaryDirectory() as tmpdirname: self.model_abit.save_pretrained(A__ ) def __A ( self ): A__ : Dict = BitsAndBytesConfig() with self.assertRaises(A__ ): A__ : str = AutoModelForCausalLM.from_pretrained( self.model_name , quantization_config=A__ , load_in_abit=A__ , device_map="""auto""" , bnb_abit_quant_type="""nf4""" , ) def __A ( self ): with self.assertRaises(A__ ): # Tries with `str` self.model_abit.to("""cpu""" ) with self.assertRaises(A__ ): # Tries with a `dtype`` self.model_abit.to(torch.floataa ) with self.assertRaises(A__ ): # Tries with a `device` self.model_abit.to(torch.device("""cuda:0""" ) ) with self.assertRaises(A__ ): # Tries with a `device` self.model_abit.float() with self.assertRaises(A__ ): # Tries with a `device` self.model_abit.half() # Test if we did not break anything A__ : Optional[int] = self.tokenizer(self.input_text , return_tensors="""pt""" ) A__ : Tuple = self.model_fpaa.to(torch.floataa ) A__ : List[str] = self.model_fpaa.generate(input_ids=encoded_input["""input_ids"""].to(0 ) , max_new_tokens=10 ) # Check this does not throw an error A__ : int = self.model_fpaa.to("""cpu""" ) # Check this does not throw an error A__ : Tuple = self.model_fpaa.half() # Check this does not throw an error A__ : List[str] = self.model_fpaa.float() def __A ( self ): A__ : Optional[int] = AutoModelForSeqaSeqLM.from_pretrained("""t5-small""" , load_in_abit=A__ , device_map="""auto""" ) self.assertTrue(model.decoder.block[0].layer[2].DenseReluDense.wo.weight.dtype == torch.floataa ) @require_bitsandbytes @require_accelerate @require_torch @require_torch_gpu @slow class _a (unittest.TestCase ): '''simple docstring''' @classmethod def __A ( cls ): A__ : int = """t5-small""" A__ : Optional[Any] = """google/flan-t5-small""" # flan-t5 uses dense-act instead of dense-relu-dense A__ : List[Any] = AutoTokenizer.from_pretrained(cls.model_name ) A__ : str = """Translate in German: Hello, my dog is cute""" def __A ( self ): gc.collect() torch.cuda.empty_cache() def __A ( self ): from transformers import TaForConditionalGeneration A__ : Tuple = TaForConditionalGeneration._keep_in_fpaa_modules A__ : Optional[Any] = None # test with `t5-small` A__ : Any = TaForConditionalGeneration.from_pretrained(self.model_name , load_in_abit=A__ , device_map="""auto""" ) A__ : Optional[int] = self.tokenizer(self.input_text , return_tensors="""pt""" ).to(0 ) A__ : Any = model.generate(**A__ ) # test with `flan-t5-small` A__ : Optional[Any] = TaForConditionalGeneration.from_pretrained( self.dense_act_model_name , load_in_abit=A__ , device_map="""auto""" ) A__ : List[str] = self.tokenizer(self.input_text , return_tensors="""pt""" ).to(0 ) A__ : Union[str, Any] = model.generate(**A__ ) A__ : Dict = modules def __A ( self ): import bitsandbytes as bnb from transformers import TaForConditionalGeneration # test with `t5-small` A__ : Dict = TaForConditionalGeneration.from_pretrained(self.model_name , load_in_abit=A__ , device_map="""auto""" ) # there was a bug with decoders - this test checks that it is fixed self.assertTrue(isinstance(model.decoder.block[0].layer[0].SelfAttention.q , bnb.nn.Linearabit ) ) A__ : Tuple = self.tokenizer(self.input_text , return_tensors="""pt""" ).to(0 ) A__ : List[Any] = model.generate(**A__ ) # test with `flan-t5-small` A__ : str = TaForConditionalGeneration.from_pretrained( self.dense_act_model_name , load_in_abit=A__ , device_map="""auto""" ) A__ : Union[str, Any] = self.tokenizer(self.input_text , return_tensors="""pt""" ).to(0 ) A__ : Tuple = model.generate(**A__ ) class _a (__magic_name__ ): '''simple docstring''' def __A ( self ): super().setUp() # model_name A__ : Any = """bigscience/bloom-560m""" A__ : int = """t5-small""" # Different types of model A__ : Union[str, Any] = AutoModel.from_pretrained(self.model_name , load_in_abit=A__ , device_map="""auto""" ) # Sequence classification model A__ : Dict = AutoModelForSequenceClassification.from_pretrained( self.model_name , load_in_abit=A__ , device_map="""auto""" ) # CausalLM model A__ : str = AutoModelForCausalLM.from_pretrained(self.model_name , load_in_abit=A__ , device_map="""auto""" ) # Seq2seq model A__ : List[Any] = AutoModelForSeqaSeqLM.from_pretrained( self.seq_to_seq_name , load_in_abit=A__ , device_map="""auto""" ) def __A ( self ): del self.base_model del self.sequence_model del self.model_abit del self.seq_to_seq_model gc.collect() torch.cuda.empty_cache() def __A ( self ): from bitsandbytes.nn import Paramsabit self.assertTrue(self.base_model.h[-1].mlp.dense_ah_to_h.weight.__class__ == Paramsabit ) # Other heads should be nn.Parameter self.assertTrue(self.model_abit.lm_head.weight.__class__ == torch.nn.Parameter ) self.assertTrue(self.sequence_model.score.weight.__class__ == torch.nn.Parameter ) self.assertTrue(self.seq_to_seq_model.lm_head.weight.__class__ == torch.nn.Parameter ) class _a (__magic_name__ ): '''simple docstring''' def __A ( self ): super().setUp() def __A ( self ): del self.pipe gc.collect() torch.cuda.empty_cache() def __A ( self ): A__ : str = pipeline( """text-generation""" , model=self.model_name , model_kwargs={"""device_map""": """auto""", """load_in_4bit""": True, """torch_dtype""": torch.floataa} , max_new_tokens=self.MAX_NEW_TOKENS , ) # Real second forward pass A__ : str = self.pipe(self.input_text ) self.assertIn(pipeline_output[0]["""generated_text"""] , self.EXPECTED_OUTPUTS ) @require_torch_multi_gpu class _a (__magic_name__ ): '''simple docstring''' def __A ( self ): super().setUp() def __A ( self ): A__ : Optional[int] = AutoModelForCausalLM.from_pretrained( self.model_name , load_in_abit=A__ , device_map="""balanced""" ) # Check correct device map self.assertEqual(set(model_parallel.hf_device_map.values() ) , {0, 1} ) # Check that inference pass works on the model A__ : Dict = self.tokenizer(self.input_text , return_tensors="""pt""" ) # Second real batch A__ : Optional[int] = model_parallel.generate(input_ids=encoded_input["""input_ids"""].to(0 ) , max_new_tokens=10 ) self.assertIn(self.tokenizer.decode(output_parallel[0] , skip_special_tokens=A__ ) , self.EXPECTED_OUTPUTS ) class _a (__magic_name__ ): '''simple docstring''' def __A ( self ): A__ : str = """facebook/opt-350m""" super().setUp() def __A ( self ): if version.parse(importlib.metadata.version("""bitsandbytes""" ) ) < version.parse("""0.37.0""" ): return # Step 1: freeze all parameters A__ : str = AutoModelForCausalLM.from_pretrained(self.model_name , load_in_abit=A__ ) self.assertEqual(set(model.hf_device_map.values() ) , {torch.cuda.current_device()} ) for param in model.parameters(): A__ : Optional[Any] = False # freeze the model - train adapters later if param.ndim == 1: # cast the small parameters (e.g. layernorm) to fp32 for stability A__ : List[str] = param.data.to(torch.floataa ) # Step 2: add adapters for _, module in model.named_modules(): if "OPTAttention" in repr(type(A__ ) ): A__ : Any = LoRALayer(module.q_proj , rank=16 ) A__ : str = LoRALayer(module.k_proj , rank=16 ) A__ : List[str] = LoRALayer(module.v_proj , rank=16 ) # Step 3: dummy batch A__ : List[Any] = self.tokenizer("""Test batch """ , return_tensors="""pt""" ).to(0 ) # Step 4: Check if the gradient is not None with torch.cuda.amp.autocast(): A__ : List[str] = model.forward(**A__ ) out.logits.norm().backward() for module in model.modules(): if isinstance(A__ , A__ ): self.assertTrue(module.adapter[1].weight.grad is not None ) self.assertTrue(module.adapter[1].weight.grad.norm().item() > 0 ) elif isinstance(A__ , nn.Embedding ): self.assertTrue(module.weight.grad is None ) class _a (__magic_name__ ): '''simple docstring''' UpperCAmelCase__: Optional[int] = '''gpt2-xl''' UpperCAmelCase__: str = 3.31_91_85_48_54_15_21_87
64
from typing import Optional import numpy as np import torch from torch import nn from transformers import GPTaConfig, GPTaLMHeadModel from transformers.modeling_utils import ModuleUtilsMixin from ...configuration_utils import ConfigMixin, register_to_config from ...models import ModelMixin class _a (__magic_name__ , __magic_name__ , __magic_name__ ): '''simple docstring''' UpperCAmelCase__: str = [r'''h\.\d+\.attn\.bias''', r'''h\.\d+\.attn\.masked_bias'''] @register_to_config def __init__( self , A__ , A__ , A__ = None , A__ = 5_0257 , A__ = 1024 , A__ = 768 , A__ = 12 , A__ = 12 , A__ = None , A__ = "gelu_new" , A__ = 0.1 , A__ = 0.1 , A__ = 0.1 , A__ = 1e-5 , A__ = 0.0_2 , A__ = True , A__ = True , A__ = False , A__ = False , ): super().__init__() A__ : Union[str, Any] = prefix_length if prefix_inner_dim != n_embd and prefix_hidden_dim is None: raise ValueError( F"""`prefix_hidden_dim` cannot be `None` when `prefix_inner_dim`: {prefix_hidden_dim} and""" F""" `n_embd`: {n_embd} are not equal.""" ) A__ : str = prefix_inner_dim A__ : Optional[Any] = prefix_hidden_dim A__ : Tuple = ( nn.Linear(self.prefix_inner_dim , self.prefix_hidden_dim ) if self.prefix_hidden_dim is not None else nn.Identity() ) A__ : int = ( nn.Linear(self.prefix_hidden_dim , A__ ) if self.prefix_hidden_dim is not None else nn.Identity() ) A__ : Tuple = GPTaConfig( vocab_size=A__ , n_positions=A__ , n_embd=A__ , n_layer=A__ , n_head=A__ , n_inner=A__ , activation_function=A__ , resid_pdrop=A__ , embd_pdrop=A__ , attn_pdrop=A__ , layer_norm_epsilon=A__ , initializer_range=A__ , scale_attn_weights=A__ , use_cache=A__ , scale_attn_by_inverse_layer_idx=A__ , reorder_and_upcast_attn=A__ , ) A__ : int = GPTaLMHeadModel(A__ ) def __A ( self , A__ , A__ , A__ = None , A__ = None , ): A__ : List[str] = self.transformer.transformer.wte(A__ ) A__ : int = self.encode_prefix(A__ ) A__ : int = self.decode_prefix(A__ ) A__ : Optional[Any] = torch.cat((prefix_embeds, embedding_text) , dim=1 ) if labels is not None: A__ : Any = self.get_dummy_token(input_ids.shape[0] , input_ids.device ) A__ : List[Any] = torch.cat((dummy_token, input_ids) , dim=1 ) A__ : List[str] = self.transformer(inputs_embeds=A__ , labels=A__ , attention_mask=A__ ) if self.prefix_hidden_dim is not None: return out, hidden else: return out def __A ( self , A__ , A__ ): return torch.zeros(A__ , self.prefix_length , dtype=torch.intaa , device=A__ ) def __A ( self , A__ ): return self.encode_prefix(A__ ) @torch.no_grad() def __A ( self , A__ , A__ , A__ ): A__ : List[Any] = torch.split(A__ , 1 , dim=0 ) A__ : Optional[int] = [] A__ : str = [] for feature in features: A__ : Dict = self.decode_prefix(feature.to(A__ ) ) # back to the clip feature # Only support beam search for now A__ , A__ : Union[str, Any] = self.generate_beam( input_embeds=A__ , device=A__ , eos_token_id=A__ ) generated_tokens.append(output_tokens[0] ) generated_seq_lengths.append(seq_lengths[0] ) A__ : int = torch.stack(A__ ) A__ : List[Any] = torch.stack(A__ ) return generated_tokens, generated_seq_lengths @torch.no_grad() def __A ( self , A__=None , A__=None , A__=None , A__ = 5 , A__ = 67 , A__ = 1.0 , A__ = None , ): A__ : Any = eos_token_id A__ : Any = None A__ : Optional[int] = None A__ : Optional[Any] = torch.ones(A__ , device=A__ , dtype=torch.int ) A__ : Any = torch.zeros(A__ , device=A__ , dtype=torch.bool ) if input_embeds is not None: A__ : Dict = input_embeds else: A__ : str = self.transformer.transformer.wte(A__ ) for i in range(A__ ): A__ : Dict = self.transformer(inputs_embeds=A__ ) A__ : str = outputs.logits A__ : Union[str, Any] = logits[:, -1, :] / (temperature if temperature > 0 else 1.0) A__ : Any = logits.softmax(-1 ).log() if scores is None: A__ , A__ : Optional[int] = logits.topk(A__ , -1 ) A__ : List[Any] = generated.expand(A__ , *generated.shape[1:] ) A__ , A__ : List[Any] = next_tokens.permute(1 , 0 ), scores.squeeze(0 ) if tokens is None: A__ : Optional[Any] = next_tokens else: A__ : List[Any] = tokens.expand(A__ , *tokens.shape[1:] ) A__ : int = torch.cat((tokens, next_tokens) , dim=1 ) else: A__ : Optional[int] = -float(np.inf ) A__ : List[Any] = 0 A__ : str = scores[:, None] + logits seq_lengths[~is_stopped] += 1 A__ : Dict = scores_sum / seq_lengths[:, None] A__ , A__ : List[Any] = scores_sum_average.view(-1 ).topk(A__ , -1 ) A__ : Tuple = next_tokens // scores_sum.shape[1] A__ : Optional[Any] = seq_lengths[next_tokens_source] A__ : List[str] = next_tokens % scores_sum.shape[1] A__ : Optional[int] = next_tokens.unsqueeze(1 ) A__ : int = tokens[next_tokens_source] A__ : List[Any] = torch.cat((tokens, next_tokens) , dim=1 ) A__ : str = generated[next_tokens_source] A__ : Optional[Any] = scores_sum_average * seq_lengths A__ : Union[str, Any] = is_stopped[next_tokens_source] A__ : str = self.transformer.transformer.wte(next_tokens.squeeze() ).view(generated.shape[0] , 1 , -1 ) A__ : Optional[int] = torch.cat((generated, next_token_embed) , dim=1 ) A__ : List[str] = is_stopped + next_tokens.eq(A__ ).squeeze() if is_stopped.all(): break A__ : Dict = scores / seq_lengths A__ : Dict = scores.argsort(descending=A__ ) # tokens tensors are already padded to max_seq_length A__ : Union[str, Any] = [tokens[i] for i in order] A__ : Any = torch.stack(A__ , dim=0 ) A__ : Dict = torch.tensor([seq_lengths[i] for i in order] , dtype=seq_lengths.dtype ) return output_texts, seq_lengths
64
1
import shutil import tempfile import unittest from transformers import ( SPIECE_UNDERLINE, AddedToken, BatchEncoding, NllbTokenizer, NllbTokenizerFast, is_torch_available, ) from transformers.testing_utils import ( get_tests_dir, nested_simplify, require_sentencepiece, require_tokenizers, require_torch, ) from ...test_tokenization_common import TokenizerTesterMixin A_ : Union[str, Any] = get_tests_dir('fixtures/test_sentencepiece.model') if is_torch_available(): from transformers.models.mam_aaa.modeling_mam_aaa import shift_tokens_right A_ : List[str] = 25_6047 A_ : Union[str, Any] = 25_6145 @require_sentencepiece @require_tokenizers class _a (__magic_name__ , unittest.TestCase ): '''simple docstring''' UpperCAmelCase__: Any = NllbTokenizer UpperCAmelCase__: int = NllbTokenizerFast UpperCAmelCase__: Union[str, Any] = True UpperCAmelCase__: Union[str, Any] = True UpperCAmelCase__: List[str] = {} def __A ( self ): super().setUp() # We have a SentencePiece fixture for testing A__ : Tuple = NllbTokenizer(A__ , keep_accents=A__ ) tokenizer.save_pretrained(self.tmpdirname ) def __A ( self ): A__ : Optional[int] = NllbTokenizer(A__ , keep_accents=A__ ) A__ : Dict = tokenizer.tokenize("""This is a test""" ) self.assertListEqual(A__ , ["""▁This""", """▁is""", """▁a""", """▁t""", """est"""] ) self.assertListEqual( tokenizer.convert_tokens_to_ids(A__ ) , [value + tokenizer.fairseq_offset for value in [285, 46, 10, 170, 382]] , ) A__ : Dict = tokenizer.tokenize("""I was born in 92000, and this is falsé.""" ) self.assertListEqual( A__ , [ SPIECE_UNDERLINE + """I""", SPIECE_UNDERLINE + """was""", SPIECE_UNDERLINE + """b""", """or""", """n""", SPIECE_UNDERLINE + """in""", SPIECE_UNDERLINE + """""", """9""", """2""", """0""", """0""", """0""", """,""", SPIECE_UNDERLINE + """and""", SPIECE_UNDERLINE + """this""", SPIECE_UNDERLINE + """is""", SPIECE_UNDERLINE + """f""", """al""", """s""", """é""", """.""", ] , ) A__ : Dict = tokenizer.convert_tokens_to_ids(A__ ) self.assertListEqual( A__ , [ value + tokenizer.fairseq_offset for value in [8, 21, 84, 55, 24, 19, 7, 2, 602, 347, 347, 347, 3, 12, 66, 46, 72, 80, 6, 2, 4] ] , ) A__ : str = tokenizer.convert_ids_to_tokens(A__ ) self.assertListEqual( A__ , [ SPIECE_UNDERLINE + """I""", SPIECE_UNDERLINE + """was""", SPIECE_UNDERLINE + """b""", """or""", """n""", SPIECE_UNDERLINE + """in""", SPIECE_UNDERLINE + """""", """<unk>""", """2""", """0""", """0""", """0""", """,""", SPIECE_UNDERLINE + """and""", SPIECE_UNDERLINE + """this""", SPIECE_UNDERLINE + """is""", SPIECE_UNDERLINE + """f""", """al""", """s""", """<unk>""", """.""", ] , ) def __A ( self ): A__ : int = (self.rust_tokenizer_class, """hf-internal-testing/tiny-random-nllb""", {}) for tokenizer, pretrained_name, kwargs in self.tokenizers_list: with self.subTest(F"""{tokenizer.__class__.__name__} ({pretrained_name})""" ): A__ : Tuple = self.rust_tokenizer_class.from_pretrained(A__ , **A__ ) A__ : List[Any] = self.tokenizer_class.from_pretrained(A__ , **A__ ) A__ : Optional[int] = tempfile.mkdtemp() A__ : str = tokenizer_r.save_pretrained(A__ ) A__ : str = tokenizer_p.save_pretrained(A__ ) # Checks it save with the same files + the tokenizer.json file for the fast one self.assertTrue(any("""tokenizer.json""" in f for f in tokenizer_r_files ) ) A__ : List[str] = tuple(f for f in tokenizer_r_files if """tokenizer.json""" not in f ) self.assertSequenceEqual(A__ , A__ ) # Checks everything loads correctly in the same way A__ : str = tokenizer_r.from_pretrained(A__ ) A__ : Union[str, Any] = tokenizer_p.from_pretrained(A__ ) # Check special tokens are set accordingly on Rust and Python for key in tokenizer_pp.special_tokens_map: self.assertTrue(hasattr(A__ , A__ ) ) shutil.rmtree(A__ ) # Save tokenizer rust, legacy_format=True A__ : str = tempfile.mkdtemp() A__ : Optional[Any] = tokenizer_r.save_pretrained(A__ , legacy_format=A__ ) A__ : Union[str, Any] = tokenizer_p.save_pretrained(A__ ) # Checks it save with the same files self.assertSequenceEqual(A__ , A__ ) # Checks everything loads correctly in the same way A__ : Optional[int] = tokenizer_r.from_pretrained(A__ ) A__ : Union[str, Any] = tokenizer_p.from_pretrained(A__ ) # Check special tokens are set accordingly on Rust and Python for key in tokenizer_pp.special_tokens_map: self.assertTrue(hasattr(A__ , A__ ) ) shutil.rmtree(A__ ) # Save tokenizer rust, legacy_format=False A__ : Dict = tempfile.mkdtemp() A__ : int = tokenizer_r.save_pretrained(A__ , legacy_format=A__ ) A__ : Optional[int] = tokenizer_p.save_pretrained(A__ ) # Checks it saved the tokenizer.json file self.assertTrue(any("""tokenizer.json""" in f for f in tokenizer_r_files ) ) # Checks everything loads correctly in the same way A__ : int = tokenizer_r.from_pretrained(A__ ) A__ : str = tokenizer_p.from_pretrained(A__ ) # Check special tokens are set accordingly on Rust and Python for key in tokenizer_pp.special_tokens_map: self.assertTrue(hasattr(A__ , A__ ) ) shutil.rmtree(A__ ) @require_torch def __A ( self ): if not self.test_seqaseq: return A__ : List[Any] = self.get_tokenizers() for tokenizer in tokenizers: with self.subTest(F"""{tokenizer.__class__.__name__}""" ): # Longer text that will definitely require truncation. A__ : Tuple = [ """ UN Chief Says There Is No Military Solution in Syria""", """ Secretary-General Ban Ki-moon says his response to Russia's stepped up military support for""" """ Syria is that 'there is no military solution' to the nearly five-year conflict and more weapons""" """ will only worsen the violence and misery for millions of people.""", ] A__ : Tuple = [ """Şeful ONU declară că nu există o soluţie militară în Siria""", """Secretarul General Ban Ki-moon declară că răspunsul său la intensificarea sprijinului militar al""" """ Rusiei pentru Siria este că \"nu există o soluţie militară\" la conflictul de aproape cinci ani şi""" """ că noi arme nu vor face decât să înrăutăţească violenţele şi mizeria pentru milioane de oameni.""", ] try: A__ : Dict = tokenizer.prepare_seqaseq_batch( src_texts=A__ , tgt_texts=A__ , max_length=3 , max_target_length=10 , return_tensors="""pt""" , src_lang="""eng_Latn""" , tgt_lang="""ron_Latn""" , ) except NotImplementedError: return self.assertEqual(batch.input_ids.shape[1] , 3 ) self.assertEqual(batch.labels.shape[1] , 10 ) # max_target_length will default to max_length if not specified A__ : Optional[int] = tokenizer.prepare_seqaseq_batch( A__ , tgt_texts=A__ , max_length=3 , return_tensors="""pt""" ) self.assertEqual(batch.input_ids.shape[1] , 3 ) self.assertEqual(batch.labels.shape[1] , 3 ) A__ : List[Any] = tokenizer.prepare_seqaseq_batch( src_texts=A__ , max_length=3 , max_target_length=10 , return_tensors="""pt""" ) self.assertEqual(batch_encoder_only.input_ids.shape[1] , 3 ) self.assertEqual(batch_encoder_only.attention_mask.shape[1] , 3 ) self.assertNotIn("""decoder_input_ids""" , A__ ) @unittest.skip("""Unfortunately way too slow to build a BPE with SentencePiece.""" ) def __A ( self ): pass def __A ( self ): for tokenizer, pretrained_name, kwargs in self.tokenizers_list: with self.subTest(F"""{tokenizer.__class__.__name__} ({pretrained_name})""" ): A__ : int = [AddedToken("""<special>""" , lstrip=A__ )] A__ : List[str] = self.rust_tokenizer_class.from_pretrained( A__ , additional_special_tokens=A__ , **A__ ) A__ : Tuple = tokenizer_r.encode("""Hey this is a <special> token""" ) A__ : int = tokenizer_r.encode("""<special>""" , add_special_tokens=A__ )[0] self.assertTrue(special_token_id in r_output ) if self.test_slow_tokenizer: A__ : Dict = self.rust_tokenizer_class.from_pretrained( A__ , additional_special_tokens=A__ , **A__ , ) A__ : Union[str, Any] = self.tokenizer_class.from_pretrained( A__ , additional_special_tokens=A__ , **A__ ) A__ : Tuple = tokenizer_p.encode("""Hey this is a <special> token""" ) A__ : Any = tokenizer_cr.encode("""Hey this is a <special> token""" ) self.assertEqual(A__ , A__ ) self.assertEqual(A__ , A__ ) self.assertTrue(special_token_id in p_output ) self.assertTrue(special_token_id in cr_output ) @require_torch @require_sentencepiece @require_tokenizers class _a (unittest.TestCase ): '''simple docstring''' UpperCAmelCase__: Optional[Any] = '''facebook/nllb-200-distilled-600M''' UpperCAmelCase__: List[Any] = [ ''' UN Chief Says There Is No Military Solution in Syria''', ''' Secretary-General Ban Ki-moon says his response to Russia\'s stepped up military support for Syria is that "there is no military solution" to the nearly five-year conflict and more weapons will only worsen the violence and misery for millions of people.''', ] UpperCAmelCase__: Optional[Any] = [ '''Şeful ONU declară că nu există o soluţie militară în Siria''', '''Secretarul General Ban Ki-moon declară că răspunsul său la intensificarea sprijinului militar al Rusiei''' ''' pentru Siria este că "nu există o soluţie militară" la conflictul de aproape cinci ani şi că noi arme nu vor''' ''' face decât să înrăutăţească violenţele şi mizeria pentru milioane de oameni.''', ] UpperCAmelCase__: Optional[int] = [ 25_60_47, 1_62_97, 13_44_08, 81_65, 24_80_66, 1_47_34, 9_50, 11_35, 10_57_21, 35_73, 83, 2_73_52, 1_08, 4_94_86, 2, ] @classmethod def __A ( cls ): A__ : NllbTokenizer = NllbTokenizer.from_pretrained( cls.checkpoint_name , src_lang="""eng_Latn""" , tgt_lang="""ron_Latn""" ) A__ : Any = 1 return cls def __A ( self ): self.assertEqual(self.tokenizer.fairseq_tokens_to_ids["""ace_Arab"""] , 25_6001 ) self.assertEqual(self.tokenizer.fairseq_tokens_to_ids["""ace_Latn"""] , 25_6002 ) self.assertEqual(self.tokenizer.fairseq_tokens_to_ids["""fra_Latn"""] , 25_6057 ) def __A ( self ): A__ : List[Any] = self.tokenizer.batch_encode_plus(self.src_text ).input_ids[0] self.assertListEqual(self.expected_src_tokens , A__ ) def __A ( self ): self.assertIn(A__ , self.tokenizer.all_special_ids ) # fmt: off A__ : Union[str, Any] = [RO_CODE, 4254, 9_8068, 11_2923, 3_9072, 3909, 713, 10_2767, 26, 1_7314, 3_5642, 1_4683, 3_3118, 2022, 6_6987, 2, 25_6047] # fmt: on A__ : List[Any] = self.tokenizer.decode(A__ , skip_special_tokens=A__ ) A__ : Optional[int] = self.tokenizer.decode(generated_ids[1:] , skip_special_tokens=A__ ) self.assertEqual(A__ , A__ ) self.assertNotIn(self.tokenizer.eos_token , A__ ) def __A ( self ): A__ : Optional[int] = ["""this is gunna be a long sentence """ * 20] assert isinstance(src_text[0] , A__ ) A__ : int = 10 A__ : List[Any] = self.tokenizer(A__ , max_length=A__ , truncation=A__ ).input_ids[0] self.assertEqual(ids[-1] , 2 ) self.assertEqual(ids[0] , A__ ) self.assertEqual(len(A__ ) , A__ ) def __A ( self ): self.assertListEqual(self.tokenizer.convert_tokens_to_ids(["""<mask>""", """ar_AR"""] ) , [25_6203, 3] ) def __A ( self ): A__ : int = tempfile.mkdtemp() A__ : str = self.tokenizer.fairseq_tokens_to_ids self.tokenizer.save_pretrained(A__ ) A__ : List[str] = NllbTokenizer.from_pretrained(A__ ) self.assertDictEqual(new_tok.fairseq_tokens_to_ids , A__ ) @require_torch def __A ( self ): A__ : List[str] = self.tokenizer( self.src_text , text_target=self.tgt_text , padding=A__ , truncation=A__ , max_length=len(self.expected_src_tokens ) , return_tensors="""pt""" , ) A__ : Optional[Any] = shift_tokens_right( batch["""labels"""] , self.tokenizer.pad_token_id , self.tokenizer.lang_code_to_id["""ron_Latn"""] ) self.assertIsInstance(A__ , A__ ) self.assertEqual((2, 15) , batch.input_ids.shape ) self.assertEqual((2, 15) , batch.attention_mask.shape ) A__ : int = batch.input_ids.tolist()[0] self.assertListEqual(self.expected_src_tokens , A__ ) self.assertEqual(A__ , batch.decoder_input_ids[0, 0] ) # EOS # Test that special tokens are reset self.assertEqual(self.tokenizer.prefix_tokens , [EN_CODE] ) self.assertEqual(self.tokenizer.suffix_tokens , [self.tokenizer.eos_token_id] ) def __A ( self ): A__ : List[str] = self.tokenizer(self.src_text , padding=A__ , truncation=A__ , max_length=3 , return_tensors="""pt""" ) A__ : str = self.tokenizer( text_target=self.tgt_text , padding=A__ , truncation=A__ , max_length=10 , return_tensors="""pt""" ) A__ : Optional[int] = targets["""input_ids"""] A__ : Dict = shift_tokens_right( A__ , self.tokenizer.pad_token_id , decoder_start_token_id=self.tokenizer.lang_code_to_id[self.tokenizer.tgt_lang] , ) self.assertEqual(batch.input_ids.shape[1] , 3 ) self.assertEqual(batch.decoder_input_ids.shape[1] , 10 ) @require_torch def __A ( self ): A__ : List[Any] = self.tokenizer._build_translation_inputs( """A test""" , return_tensors="""pt""" , src_lang="""eng_Latn""" , tgt_lang="""fra_Latn""" ) self.assertEqual( nested_simplify(A__ ) , { # A, test, EOS, en_XX """input_ids""": [[25_6047, 70, 7356, 2]], """attention_mask""": [[1, 1, 1, 1]], # ar_AR """forced_bos_token_id""": 25_6057, } , ) @require_torch def __A ( self ): A__ : str = True A__ : int = self.tokenizer( """UN Chief says there is no military solution in Syria""" , src_lang="""eng_Latn""" , tgt_lang="""fra_Latn""" ) self.assertEqual( inputs.input_ids , [1_6297, 13_4408, 2_5653, 6370, 248, 254, 10_3929, 9_4995, 108, 4_9486, 2, 25_6047] ) A__ : Any = False A__ : Any = self.tokenizer( """UN Chief says there is no military solution in Syria""" , src_lang="""eng_Latn""" , tgt_lang="""fra_Latn""" ) self.assertEqual( inputs.input_ids , [25_6047, 1_6297, 13_4408, 2_5653, 6370, 248, 254, 10_3929, 9_4995, 108, 4_9486, 2] )
64
import io import itertools import json from dataclasses import dataclass from typing import Optional import pyarrow as pa import pyarrow.json as paj import datasets from datasets.table import table_cast from datasets.utils.file_utils import readline A_ : Tuple = datasets.utils.logging.get_logger(__name__) @dataclass class _a (datasets.BuilderConfig ): '''simple docstring''' UpperCAmelCase__: Optional[datasets.Features] = None UpperCAmelCase__: str = "utf-8" UpperCAmelCase__: Optional[str] = None UpperCAmelCase__: Optional[str] = None UpperCAmelCase__: bool = True # deprecated UpperCAmelCase__: Optional[int] = None # deprecated UpperCAmelCase__: int = 10 << 20 # 10MB UpperCAmelCase__: Optional[bool] = None class _a (datasets.ArrowBasedBuilder ): '''simple docstring''' UpperCAmelCase__: List[str] = JsonConfig def __A ( self ): if self.config.block_size is not None: logger.warning("""The JSON loader parameter `block_size` is deprecated. Please use `chunksize` instead""" ) A__ : Union[str, Any] = self.config.block_size if self.config.use_threads is not True: logger.warning( """The JSON loader parameter `use_threads` is deprecated and doesn't have any effect anymore.""" ) if self.config.newlines_in_values is not None: raise ValueError("""The JSON loader parameter `newlines_in_values` is no longer supported""" ) return datasets.DatasetInfo(features=self.config.features ) def __A ( self , A__ ): if not self.config.data_files: raise ValueError(F"""At least one data file must be specified, but got data_files={self.config.data_files}""" ) A__ : int = dl_manager.download_and_extract(self.config.data_files ) if isinstance(A__ , (str, list, tuple) ): A__ : Optional[Any] = data_files if isinstance(A__ , A__ ): A__ : List[str] = [files] A__ : int = [dl_manager.iter_files(A__ ) for file in files] return [datasets.SplitGenerator(name=datasets.Split.TRAIN , gen_kwargs={"""files""": files} )] A__ : List[str] = [] for split_name, files in data_files.items(): if isinstance(A__ , A__ ): A__ : Optional[int] = [files] A__ : Optional[int] = [dl_manager.iter_files(A__ ) for file in files] splits.append(datasets.SplitGenerator(name=A__ , gen_kwargs={"""files""": files} ) ) return splits def __A ( self , A__ ): if self.config.features is not None: # adding missing columns for column_name in set(self.config.features ) - set(pa_table.column_names ): A__ : Optional[Any] = self.config.features.arrow_schema.field(A__ ).type A__ : str = pa_table.append_column(A__ , pa.array([None] * len(A__ ) , type=A__ ) ) # more expensive cast to support nested structures with keys in a different order # allows str <-> int/float or str to Audio for example A__ : Optional[int] = table_cast(A__ , self.config.features.arrow_schema ) return pa_table def __A ( self , A__ ): for file_idx, file in enumerate(itertools.chain.from_iterable(A__ ) ): # If the file is one json object and if we need to look at the list of items in one specific field if self.config.field is not None: with open(A__ , encoding=self.config.encoding , errors=self.config.encoding_errors ) as f: A__ : Optional[Any] = json.load(A__ ) # We keep only the field we are interested in A__ : Optional[int] = dataset[self.config.field] # We accept two format: a list of dicts or a dict of lists if isinstance(A__ , (list, tuple) ): A__ : Union[str, Any] = set().union(*[row.keys() for row in dataset] ) A__ : Any = {col: [row.get(A__ ) for row in dataset] for col in keys} else: A__ : Any = dataset A__ : Any = pa.Table.from_pydict(A__ ) yield file_idx, self._cast_table(A__ ) # If the file has one json object per line else: with open(A__ , """rb""" ) as f: A__ : List[str] = 0 # Use block_size equal to the chunk size divided by 32 to leverage multithreading # Set a default minimum value of 16kB if the chunk size is really small A__ : List[str] = max(self.config.chunksize // 32 , 16 << 10 ) A__ : Any = ( self.config.encoding_errors if self.config.encoding_errors is not None else """strict""" ) while True: A__ : Dict = f.read(self.config.chunksize ) if not batch: break # Finish current line try: batch += f.readline() except (AttributeError, io.UnsupportedOperation): batch += readline(A__ ) # PyArrow only accepts utf-8 encoded bytes if self.config.encoding != "utf-8": A__ : List[Any] = batch.decode(self.config.encoding , errors=A__ ).encode("""utf-8""" ) try: while True: try: A__ : str = paj.read_json( io.BytesIO(A__ ) , read_options=paj.ReadOptions(block_size=A__ ) ) break except (pa.ArrowInvalid, pa.ArrowNotImplementedError) as e: if ( isinstance(A__ , pa.ArrowInvalid ) and "straddling" not in str(A__ ) or block_size > len(A__ ) ): raise else: # Increase the block size in case it was too small. # The block size will be reset for the next file. logger.debug( F"""Batch of {len(A__ )} bytes couldn't be parsed with block_size={block_size}. Retrying with block_size={block_size * 2}.""" ) block_size *= 2 except pa.ArrowInvalid as e: try: with open( A__ , encoding=self.config.encoding , errors=self.config.encoding_errors ) as f: A__ : Optional[Any] = json.load(A__ ) except json.JSONDecodeError: logger.error(F"""Failed to read file '{file}' with error {type(A__ )}: {e}""" ) raise e # If possible, parse the file as a list of json objects and exit the loop if isinstance(A__ , A__ ): # list is the only sequence type supported in JSON try: A__ : str = set().union(*[row.keys() for row in dataset] ) A__ : List[str] = {col: [row.get(A__ ) for row in dataset] for col in keys} A__ : int = pa.Table.from_pydict(A__ ) except (pa.ArrowInvalid, AttributeError) as e: logger.error(F"""Failed to read file '{file}' with error {type(A__ )}: {e}""" ) raise ValueError(F"""Not able to read records in the JSON file at {file}.""" ) from None yield file_idx, self._cast_table(A__ ) break else: logger.error(F"""Failed to read file '{file}' with error {type(A__ )}: {e}""" ) raise ValueError( F"""Not able to read records in the JSON file at {file}. """ F"""You should probably indicate the field of the JSON file containing your records. """ F"""This JSON file contain the following fields: {str(list(dataset.keys() ) )}. """ F"""Select the correct one and provide it as `field='XXX'` to the dataset loading method. """ ) from None # Uncomment for debugging (will print the Arrow table size and elements) # logger.warning(f"pa_table: {pa_table} num rows: {pa_table.num_rows}") # logger.warning('\n'.join(str(pa_table.slice(i, 1).to_pydict()) for i in range(pa_table.num_rows))) yield (file_idx, batch_idx), self._cast_table(A__ ) batch_idx += 1
64
1
import unittest from transformers import is_torch_available, is_vision_available from transformers.testing_utils import require_torch, require_vision, slow, torch_device if is_torch_available(): import torch from transformers import AutoModelForImageClassification if is_vision_available(): from transformers import AutoImageProcessor @require_torch @require_vision class _a (unittest.TestCase ): '''simple docstring''' @slow def __A ( self ): A__ : int = AutoImageProcessor.from_pretrained("""microsoft/dit-base-finetuned-rvlcdip""" ) A__ : Dict = AutoModelForImageClassification.from_pretrained("""microsoft/dit-base-finetuned-rvlcdip""" ) model.to(A__ ) from datasets import load_dataset A__ : Union[str, Any] = load_dataset("""nielsr/rvlcdip-demo""" ) A__ : List[str] = dataset["""train"""][0]["""image"""].convert("""RGB""" ) A__ : int = image_processor(A__ , return_tensors="""pt""" ).to(A__ ) # forward pass with torch.no_grad(): A__ : Optional[int] = model(**A__ ) A__ : int = outputs.logits A__ : List[str] = torch.Size((1, 16) ) self.assertEqual(logits.shape , A__ ) A__ : Optional[int] = torch.tensor( [-0.4_1_5_8, -0.4_0_9_2, -0.4_3_4_7] , device=A__ , dtype=torch.float , ) self.assertTrue(torch.allclose(logits[0, :3] , A__ , atol=1e-4 ) )
64
import os try: from .build_directory_md import good_file_paths except ImportError: from build_directory_md import good_file_paths # type: ignore A_ : Dict = list(good_file_paths()) assert filepaths, "good_file_paths() failed!" A_ : Optional[Any] = [file for file in filepaths if file != file.lower()] if upper_files: print(f'''{len(upper_files)} files contain uppercase characters:''') print('\n'.join(upper_files) + '\n') A_ : Tuple = [file for file in filepaths if ' ' in file] if space_files: print(f'''{len(space_files)} files contain space characters:''') print('\n'.join(space_files) + '\n') A_ : Any = [file for file in filepaths if '-' in file] if hyphen_files: print(f'''{len(hyphen_files)} files contain hyphen characters:''') print('\n'.join(hyphen_files) + '\n') A_ : List[str] = [file for file in filepaths if os.sep not in file] if nodir_files: print(f'''{len(nodir_files)} files are not in a directory:''') print('\n'.join(nodir_files) + '\n') A_ : Any = len(upper_files + space_files + hyphen_files + nodir_files) if bad_files: import sys sys.exit(bad_files)
64
1
from __future__ import annotations class _a : '''simple docstring''' def __init__( self , A__ = 0 ): A__ : Union[str, Any] = key def __A ( self , A__ , A__ ): assert isinstance(A__ , A__ ) and isinstance(A__ , A__ ) A__ : Union[str, Any] = key or self.__key or 1 # make sure key is an appropriate size key %= 255 return [chr(ord(A__ ) ^ key ) for ch in content] def __A ( self , A__ , A__ ): assert isinstance(A__ , A__ ) and isinstance(A__ , A__ ) A__ : str = key or self.__key or 1 # make sure key is an appropriate size key %= 255 return [chr(ord(A__ ) ^ key ) for ch in content] def __A ( self , A__ , A__ = 0 ): assert isinstance(A__ , A__ ) and isinstance(A__ , A__ ) A__ : Optional[Any] = key or self.__key or 1 # make sure key can be any size while key > 255: key -= 255 # This will be returned A__ : List[str] = """""" for ch in content: ans += chr(ord(A__ ) ^ key ) return ans def __A ( self , A__ , A__ = 0 ): assert isinstance(A__ , A__ ) and isinstance(A__ , A__ ) A__ : Optional[Any] = key or self.__key or 1 # make sure key can be any size while key > 255: key -= 255 # This will be returned A__ : List[Any] = """""" for ch in content: ans += chr(ord(A__ ) ^ key ) return ans def __A ( self , A__ , A__ = 0 ): assert isinstance(A__ , A__ ) and isinstance(A__ , A__ ) try: with open(A__ ) as fin, open("""encrypt.out""" , """w+""" ) as fout: # actual encrypt-process for line in fin: fout.write(self.encrypt_string(A__ , A__ ) ) except OSError: return False return True def __A ( self , A__ , A__ ): assert isinstance(A__ , A__ ) and isinstance(A__ , A__ ) try: with open(A__ ) as fin, open("""decrypt.out""" , """w+""" ) as fout: # actual encrypt-process for line in fin: fout.write(self.decrypt_string(A__ , A__ ) ) except OSError: return False return True # Tests # crypt = XORCipher() # key = 67 # # test encrypt # print(crypt.encrypt("hallo welt",key)) # # test decrypt # print(crypt.decrypt(crypt.encrypt("hallo welt",key), key)) # # test encrypt_string # print(crypt.encrypt_string("hallo welt",key)) # # test decrypt_string # print(crypt.decrypt_string(crypt.encrypt_string("hallo welt",key),key)) # if (crypt.encrypt_file("test.txt",key)): # print("encrypt successful") # else: # print("encrypt unsuccessful") # if (crypt.decrypt_file("encrypt.out",key)): # print("decrypt successful") # else: # print("decrypt unsuccessful")
64
import inspect import warnings from typing import Any, Dict, Optional, Union from packaging import version def UpperCamelCase (*lowercase_: Optional[int] , lowercase_: Optional[Union[Dict, Any]] = None , lowercase_: Dict=True , lowercase_: Tuple=2 ) -> Dict: from .. import __version__ A__ : Dict = take_from A__ : str = () if not isinstance(args[0] , lowercase_ ): A__ : int = (args,) for attribute, version_name, message in args: if version.parse(version.parse(lowercase_ ).base_version ) >= version.parse(lowercase_ ): raise ValueError( f"""The deprecation tuple {(attribute, version_name, message)} should be removed since diffusers'""" f""" version {__version__} is >= {version_name}""" ) A__ : Any = None if isinstance(lowercase_ , lowercase_ ) and attribute in deprecated_kwargs: values += (deprecated_kwargs.pop(lowercase_ ),) A__ : List[str] = f"""The `{attribute}` argument is deprecated and will be removed in version {version_name}.""" elif hasattr(lowercase_ , lowercase_ ): values += (getattr(lowercase_ , lowercase_ ),) A__ : Optional[Any] = f"""The `{attribute}` attribute is deprecated and will be removed in version {version_name}.""" elif deprecated_kwargs is None: A__ : int = f"""`{attribute}` is deprecated and will be removed in version {version_name}.""" if warning is not None: A__ : int = warning + """ """ if standard_warn else """""" warnings.warn(warning + message , lowercase_ , stacklevel=lowercase_ ) if isinstance(lowercase_ , lowercase_ ) and len(lowercase_ ) > 0: A__ : Union[str, Any] = inspect.getouterframes(inspect.currentframe() )[1] A__ : Optional[Any] = call_frame.filename A__ : Optional[int] = call_frame.lineno A__ : Any = call_frame.function A__ , A__ : List[str] = next(iter(deprecated_kwargs.items() ) ) raise TypeError(f"""{function} in {filename} line {line_number-1} got an unexpected keyword argument `{key}`""" ) if len(lowercase_ ) == 0: return elif len(lowercase_ ) == 1: return values[0] return values
64
1
import argparse import json from collections import OrderedDict from pathlib import Path import requests import torch from huggingface_hub import hf_hub_download from PIL import Image from transformers import ( ConditionalDetrConfig, ConditionalDetrForObjectDetection, ConditionalDetrForSegmentation, ConditionalDetrImageProcessor, ) from transformers.utils import logging logging.set_verbosity_info() A_ : str = logging.get_logger(__name__) # here we list all keys to be renamed (original name on the left, our name on the right) A_ : Tuple = [] for i in range(6): # encoder layers: output projection, 2 feedforward neural networks and 2 layernorms rename_keys.append( (f'''transformer.encoder.layers.{i}.self_attn.out_proj.weight''', f'''encoder.layers.{i}.self_attn.out_proj.weight''') ) rename_keys.append( (f'''transformer.encoder.layers.{i}.self_attn.out_proj.bias''', f'''encoder.layers.{i}.self_attn.out_proj.bias''') ) rename_keys.append((f'''transformer.encoder.layers.{i}.linear1.weight''', f'''encoder.layers.{i}.fc1.weight''')) rename_keys.append((f'''transformer.encoder.layers.{i}.linear1.bias''', f'''encoder.layers.{i}.fc1.bias''')) rename_keys.append((f'''transformer.encoder.layers.{i}.linear2.weight''', f'''encoder.layers.{i}.fc2.weight''')) rename_keys.append((f'''transformer.encoder.layers.{i}.linear2.bias''', f'''encoder.layers.{i}.fc2.bias''')) rename_keys.append( (f'''transformer.encoder.layers.{i}.norm1.weight''', f'''encoder.layers.{i}.self_attn_layer_norm.weight''') ) rename_keys.append((f'''transformer.encoder.layers.{i}.norm1.bias''', f'''encoder.layers.{i}.self_attn_layer_norm.bias''')) rename_keys.append((f'''transformer.encoder.layers.{i}.norm2.weight''', f'''encoder.layers.{i}.final_layer_norm.weight''')) rename_keys.append((f'''transformer.encoder.layers.{i}.norm2.bias''', f'''encoder.layers.{i}.final_layer_norm.bias''')) # decoder layers: 2 times output projection, 2 feedforward neural networks and 3 layernorms rename_keys.append( (f'''transformer.decoder.layers.{i}.self_attn.out_proj.weight''', f'''decoder.layers.{i}.self_attn.out_proj.weight''') ) rename_keys.append( (f'''transformer.decoder.layers.{i}.self_attn.out_proj.bias''', f'''decoder.layers.{i}.self_attn.out_proj.bias''') ) rename_keys.append( ( f'''transformer.decoder.layers.{i}.cross_attn.out_proj.weight''', f'''decoder.layers.{i}.encoder_attn.out_proj.weight''', ) ) rename_keys.append( ( f'''transformer.decoder.layers.{i}.cross_attn.out_proj.bias''', f'''decoder.layers.{i}.encoder_attn.out_proj.bias''', ) ) rename_keys.append((f'''transformer.decoder.layers.{i}.linear1.weight''', f'''decoder.layers.{i}.fc1.weight''')) rename_keys.append((f'''transformer.decoder.layers.{i}.linear1.bias''', f'''decoder.layers.{i}.fc1.bias''')) rename_keys.append((f'''transformer.decoder.layers.{i}.linear2.weight''', f'''decoder.layers.{i}.fc2.weight''')) rename_keys.append((f'''transformer.decoder.layers.{i}.linear2.bias''', f'''decoder.layers.{i}.fc2.bias''')) rename_keys.append( (f'''transformer.decoder.layers.{i}.norm1.weight''', f'''decoder.layers.{i}.self_attn_layer_norm.weight''') ) rename_keys.append((f'''transformer.decoder.layers.{i}.norm1.bias''', f'''decoder.layers.{i}.self_attn_layer_norm.bias''')) rename_keys.append( (f'''transformer.decoder.layers.{i}.norm2.weight''', f'''decoder.layers.{i}.encoder_attn_layer_norm.weight''') ) rename_keys.append( (f'''transformer.decoder.layers.{i}.norm2.bias''', f'''decoder.layers.{i}.encoder_attn_layer_norm.bias''') ) rename_keys.append((f'''transformer.decoder.layers.{i}.norm3.weight''', f'''decoder.layers.{i}.final_layer_norm.weight''')) rename_keys.append((f'''transformer.decoder.layers.{i}.norm3.bias''', f'''decoder.layers.{i}.final_layer_norm.bias''')) # q, k, v projections in self/cross-attention in decoder for conditional DETR rename_keys.append( (f'''transformer.decoder.layers.{i}.sa_qcontent_proj.weight''', f'''decoder.layers.{i}.sa_qcontent_proj.weight''') ) rename_keys.append( (f'''transformer.decoder.layers.{i}.sa_kcontent_proj.weight''', f'''decoder.layers.{i}.sa_kcontent_proj.weight''') ) rename_keys.append( (f'''transformer.decoder.layers.{i}.sa_qpos_proj.weight''', f'''decoder.layers.{i}.sa_qpos_proj.weight''') ) rename_keys.append( (f'''transformer.decoder.layers.{i}.sa_kpos_proj.weight''', f'''decoder.layers.{i}.sa_kpos_proj.weight''') ) rename_keys.append((f'''transformer.decoder.layers.{i}.sa_v_proj.weight''', f'''decoder.layers.{i}.sa_v_proj.weight''')) rename_keys.append( (f'''transformer.decoder.layers.{i}.ca_qcontent_proj.weight''', f'''decoder.layers.{i}.ca_qcontent_proj.weight''') ) # rename_keys.append((f"transformer.decoder.layers.{i}.ca_qpos_proj.weight", f"decoder.layers.{i}.ca_qpos_proj.weight")) rename_keys.append( (f'''transformer.decoder.layers.{i}.ca_kcontent_proj.weight''', f'''decoder.layers.{i}.ca_kcontent_proj.weight''') ) rename_keys.append( (f'''transformer.decoder.layers.{i}.ca_kpos_proj.weight''', f'''decoder.layers.{i}.ca_kpos_proj.weight''') ) rename_keys.append((f'''transformer.decoder.layers.{i}.ca_v_proj.weight''', f'''decoder.layers.{i}.ca_v_proj.weight''')) rename_keys.append( (f'''transformer.decoder.layers.{i}.ca_qpos_sine_proj.weight''', f'''decoder.layers.{i}.ca_qpos_sine_proj.weight''') ) rename_keys.append( (f'''transformer.decoder.layers.{i}.sa_qcontent_proj.bias''', f'''decoder.layers.{i}.sa_qcontent_proj.bias''') ) rename_keys.append( (f'''transformer.decoder.layers.{i}.sa_kcontent_proj.bias''', f'''decoder.layers.{i}.sa_kcontent_proj.bias''') ) rename_keys.append((f'''transformer.decoder.layers.{i}.sa_qpos_proj.bias''', f'''decoder.layers.{i}.sa_qpos_proj.bias''')) rename_keys.append((f'''transformer.decoder.layers.{i}.sa_kpos_proj.bias''', f'''decoder.layers.{i}.sa_kpos_proj.bias''')) rename_keys.append((f'''transformer.decoder.layers.{i}.sa_v_proj.bias''', f'''decoder.layers.{i}.sa_v_proj.bias''')) rename_keys.append( (f'''transformer.decoder.layers.{i}.ca_qcontent_proj.bias''', f'''decoder.layers.{i}.ca_qcontent_proj.bias''') ) # rename_keys.append((f"transformer.decoder.layers.{i}.ca_qpos_proj.bias", f"decoder.layers.{i}.ca_qpos_proj.bias")) rename_keys.append( (f'''transformer.decoder.layers.{i}.ca_kcontent_proj.bias''', f'''decoder.layers.{i}.ca_kcontent_proj.bias''') ) rename_keys.append((f'''transformer.decoder.layers.{i}.ca_kpos_proj.bias''', f'''decoder.layers.{i}.ca_kpos_proj.bias''')) rename_keys.append((f'''transformer.decoder.layers.{i}.ca_v_proj.bias''', f'''decoder.layers.{i}.ca_v_proj.bias''')) rename_keys.append( (f'''transformer.decoder.layers.{i}.ca_qpos_sine_proj.bias''', f'''decoder.layers.{i}.ca_qpos_sine_proj.bias''') ) # convolutional projection + query embeddings + layernorm of decoder + class and bounding box heads # for conditional DETR, also convert reference point head and query scale MLP rename_keys.extend( [ ('input_proj.weight', 'input_projection.weight'), ('input_proj.bias', 'input_projection.bias'), ('query_embed.weight', 'query_position_embeddings.weight'), ('transformer.decoder.norm.weight', 'decoder.layernorm.weight'), ('transformer.decoder.norm.bias', 'decoder.layernorm.bias'), ('class_embed.weight', 'class_labels_classifier.weight'), ('class_embed.bias', 'class_labels_classifier.bias'), ('bbox_embed.layers.0.weight', 'bbox_predictor.layers.0.weight'), ('bbox_embed.layers.0.bias', 'bbox_predictor.layers.0.bias'), ('bbox_embed.layers.1.weight', 'bbox_predictor.layers.1.weight'), ('bbox_embed.layers.1.bias', 'bbox_predictor.layers.1.bias'), ('bbox_embed.layers.2.weight', 'bbox_predictor.layers.2.weight'), ('bbox_embed.layers.2.bias', 'bbox_predictor.layers.2.bias'), ('transformer.decoder.ref_point_head.layers.0.weight', 'decoder.ref_point_head.layers.0.weight'), ('transformer.decoder.ref_point_head.layers.0.bias', 'decoder.ref_point_head.layers.0.bias'), ('transformer.decoder.ref_point_head.layers.1.weight', 'decoder.ref_point_head.layers.1.weight'), ('transformer.decoder.ref_point_head.layers.1.bias', 'decoder.ref_point_head.layers.1.bias'), ('transformer.decoder.query_scale.layers.0.weight', 'decoder.query_scale.layers.0.weight'), ('transformer.decoder.query_scale.layers.0.bias', 'decoder.query_scale.layers.0.bias'), ('transformer.decoder.query_scale.layers.1.weight', 'decoder.query_scale.layers.1.weight'), ('transformer.decoder.query_scale.layers.1.bias', 'decoder.query_scale.layers.1.bias'), ('transformer.decoder.layers.0.ca_qpos_proj.weight', 'decoder.layers.0.ca_qpos_proj.weight'), ('transformer.decoder.layers.0.ca_qpos_proj.bias', 'decoder.layers.0.ca_qpos_proj.bias'), ] ) def UpperCamelCase (lowercase_: Dict , lowercase_: Dict , lowercase_: Optional[int] ) -> Union[str, Any]: A__ : List[Any] = state_dict.pop(lowercase_ ) A__ : Tuple = val def UpperCamelCase (lowercase_: Dict ) -> List[str]: A__ : int = OrderedDict() for key, value in state_dict.items(): if "backbone.0.body" in key: A__ : int = key.replace("""backbone.0.body""" , """backbone.conv_encoder.model""" ) A__ : Any = value else: A__ : Optional[int] = value return new_state_dict def UpperCamelCase (lowercase_: Optional[int] , lowercase_: Dict=False ) -> Optional[int]: A__ : List[str] = """""" if is_panoptic: A__ : List[Any] = """conditional_detr.""" # first: transformer encoder for i in range(6 ): # read in weights + bias of input projection layer (in PyTorch's MultiHeadAttention, this is a single matrix + bias) A__ : int = state_dict.pop(f"""{prefix}transformer.encoder.layers.{i}.self_attn.in_proj_weight""" ) A__ : List[Any] = state_dict.pop(f"""{prefix}transformer.encoder.layers.{i}.self_attn.in_proj_bias""" ) # next, add query, keys and values (in that order) to the state dict A__ : Dict = in_proj_weight[:256, :] A__ : List[Any] = in_proj_bias[:256] A__ : Optional[Any] = in_proj_weight[256:512, :] A__ : Tuple = in_proj_bias[256:512] A__ : Dict = in_proj_weight[-256:, :] A__ : Dict = in_proj_bias[-256:] def UpperCamelCase () -> List[Any]: A__ : List[str] = """http://images.cocodataset.org/val2017/000000039769.jpg""" A__ : Optional[int] = Image.open(requests.get(lowercase_ , stream=lowercase_ ).raw ) return im @torch.no_grad() def UpperCamelCase (lowercase_: str , lowercase_: int ) -> int: A__ : Optional[Any] = ConditionalDetrConfig() # set backbone and dilation attributes if "resnet101" in model_name: A__ : Optional[int] = """resnet101""" if "dc5" in model_name: A__ : str = True A__ : Union[str, Any] = """panoptic""" in model_name if is_panoptic: A__ : Tuple = 250 else: A__ : Tuple = 91 A__ : Union[str, Any] = """huggingface/label-files""" A__ : str = """coco-detection-id2label.json""" A__ : Tuple = json.load(open(hf_hub_download(lowercase_ , lowercase_ , repo_type="""dataset""" ) , """r""" ) ) A__ : List[Any] = {int(lowercase_ ): v for k, v in idalabel.items()} A__ : Optional[Any] = idalabel A__ : Union[str, Any] = {v: k for k, v in idalabel.items()} # load image processor A__ : str = """coco_panoptic""" if is_panoptic else """coco_detection""" A__ : List[Any] = ConditionalDetrImageProcessor(format=lowercase_ ) # prepare image A__ : Union[str, Any] = prepare_img() A__ : Any = image_processor(images=lowercase_ , return_tensors="""pt""" ) A__ : int = encoding["""pixel_values"""] logger.info(f"""Converting model {model_name}...""" ) # load original model from torch hub A__ : Any = torch.hub.load("""DeppMeng/ConditionalDETR""" , lowercase_ , pretrained=lowercase_ ).eval() A__ : int = conditional_detr.state_dict() # rename keys for src, dest in rename_keys: if is_panoptic: A__ : List[str] = """conditional_detr.""" + src rename_key(lowercase_ , lowercase_ , lowercase_ ) A__ : int = rename_backbone_keys(lowercase_ ) # query, key and value matrices need special treatment read_in_q_k_v(lowercase_ , is_panoptic=lowercase_ ) # important: we need to prepend a prefix to each of the base model keys as the head models use different attributes for them A__ : Union[str, Any] = """conditional_detr.model.""" if is_panoptic else """model.""" for key in state_dict.copy().keys(): if is_panoptic: if ( key.startswith("""conditional_detr""" ) and not key.startswith("""class_labels_classifier""" ) and not key.startswith("""bbox_predictor""" ) ): A__ : List[Any] = state_dict.pop(lowercase_ ) A__ : int = val elif "class_labels_classifier" in key or "bbox_predictor" in key: A__ : Dict = state_dict.pop(lowercase_ ) A__ : int = val elif key.startswith("""bbox_attention""" ) or key.startswith("""mask_head""" ): continue else: A__ : Optional[int] = state_dict.pop(lowercase_ ) A__ : List[str] = val else: if not key.startswith("""class_labels_classifier""" ) and not key.startswith("""bbox_predictor""" ): A__ : Dict = state_dict.pop(lowercase_ ) A__ : str = val # finally, create HuggingFace model and load state dict A__ : Optional[int] = ConditionalDetrForSegmentation(lowercase_ ) if is_panoptic else ConditionalDetrForObjectDetection(lowercase_ ) model.load_state_dict(lowercase_ ) model.eval() model.push_to_hub(repo_id=lowercase_ , organization="""DepuMeng""" , commit_message="""Add model""" ) # verify our conversion A__ : Tuple = conditional_detr(lowercase_ ) A__ : Tuple = model(lowercase_ ) assert torch.allclose(outputs.logits , original_outputs["""pred_logits"""] , atol=1E-4 ) assert torch.allclose(outputs.pred_boxes , original_outputs["""pred_boxes"""] , atol=1E-4 ) if is_panoptic: assert torch.allclose(outputs.pred_masks , original_outputs["""pred_masks"""] , atol=1E-4 ) # Save model and image processor logger.info(f"""Saving PyTorch model and image processor to {pytorch_dump_folder_path}...""" ) Path(lowercase_ ).mkdir(exist_ok=lowercase_ ) model.save_pretrained(lowercase_ ) image_processor.save_pretrained(lowercase_ ) if __name__ == "__main__": A_ : Union[str, Any] = argparse.ArgumentParser() parser.add_argument( '--model_name', default='conditional_detr_resnet50', type=str, help='Name of the CONDITIONAL_DETR model you\'d like to convert.', ) parser.add_argument( '--pytorch_dump_folder_path', default=None, type=str, help='Path to the folder to output PyTorch model.' ) A_ : Union[str, Any] = parser.parse_args() convert_conditional_detr_checkpoint(args.model_name, args.pytorch_dump_folder_path)
64
import argparse import json import os import tensorstore as ts import torch from flax import serialization from flax.traverse_util import flatten_dict, unflatten_dict from tensorflow.io import gfile from transformers.modeling_utils import dtype_byte_size from transformers.models.switch_transformers.convert_switch_transformers_original_flax_checkpoint_to_pytorch import ( rename_keys, ) from transformers.utils import WEIGHTS_INDEX_NAME, WEIGHTS_NAME from transformers.utils.hub import convert_file_size_to_int def UpperCamelCase (lowercase_: List[str] , lowercase_: str ) -> Optional[Any]: if flax_key_tuple[-1] == "kernel" and flax_tensor.ndim == 3: # expert layer A__ : Union[str, Any] = flax_key_tuple[:-1] + ("""weight""",) A__ : Optional[int] = torch.permute(lowercase_ , (0, 2, 1) ) elif flax_key_tuple[-1] == "kernel" and ".".join(lowercase_ ): # linear layer A__ : Optional[Any] = flax_key_tuple[:-1] + ("""weight""",) A__ : int = flax_tensor.T elif flax_key_tuple[-1] in ["scale", "embedding"]: A__ : Optional[int] = flax_key_tuple[:-1] + ("""weight""",) return flax_key_tuple, flax_tensor def UpperCamelCase (lowercase_: Tuple , lowercase_: Optional[int] , lowercase_: str ) -> Union[str, Any]: if "metadata" in layer: A__ : Tuple = layer.split("""metadata""" ) A__ : Optional[Any] = """""".join(split_layer[0] )[:-1] A__ : Optional[Any] = [tuple(("""metadata""" + split_layer[1]).split("""/""" ) )] elif "kvstore" in layer: A__ : str = layer.split("""kvstore""" ) A__ : int = """""".join(split_layer[0] )[:-1] A__ : Optional[int] = [tuple(("""kvstore""" + split_layer[1]).split("""/""" ) )] else: A__ : Any = layer.split("""/""" ) A__ : int = """/""".join(split_layer[:-1] ) A__ : str = (split_layer[-1],) if "kvstore/path" in layer: A__ : Dict = f"""{switch_checkpoint_path}/{checkpoint_info[layer]}""" elif "kvstore/driver" in layer: A__ : Optional[int] = """file""" else: A__ : str = checkpoint_info[layer] return curr_real_layer_name, split_layer, content def UpperCamelCase (lowercase_: str , lowercase_: List[Any] ) -> int: A__ : int = rename_keys(lowercase_ ) A__ : Any = {} for k, v in current_block.items(): A__ : Dict = v A__ : str = new_current_block torch.save(lowercase_ , lowercase_ ) def UpperCamelCase (lowercase_: Dict , lowercase_: Optional[Any] , lowercase_: Optional[Any] , lowercase_: Optional[int] , lowercase_: str = WEIGHTS_NAME ) -> Tuple: A__ : Optional[int] = convert_file_size_to_int(lowercase_ ) A__ : List[Any] = [] A__ : int = {} A__ : List[str] = 0 A__ : Any = 0 os.makedirs(lowercase_ , exist_ok=lowercase_ ) with gfile.GFile(switch_checkpoint_path + """/checkpoint""" , """rb""" ) as fp: A__ : Optional[Any] = serialization.msgpack_restore(fp.read() )["""optimizer"""]["""target"""] A__ : Dict = flatten_dict(lowercase_ , sep="""/""" ) A__ : Any = {} for layer in checkpoint_info.keys(): A__ , A__ , A__ : Union[str, Any] = get_key_and_tensorstore_dict( lowercase_ , lowercase_ , lowercase_ ) if curr_real_layer_name in all_layers: A__ : Optional[int] = content else: A__ : List[Any] = {split_layer[-1]: content} for key in all_layers.keys(): # open tensorstore file A__ : Optional[Any] = ts.open(unflatten_dict(all_layers[key] ) ).result().read().result() A__ : List[Any] = torch.tensor(lowercase_ ) A__ : List[Any] = raw_weights.numel() * dtype_byte_size(raw_weights.dtype ) # use the renaming pattern from the small conversion scripts A__ , A__ : Any = rename_base_flax_keys(tuple(key.split("""/""" ) ) , lowercase_ ) A__ : Any = """/""".join(lowercase_ ) # If this weight is going to tip up over the maximal size, we split. if current_block_size + weight_size > max_shard_size: A__ : List[Any] = os.path.join( lowercase_ , weights_name.replace(""".bin""" , f"""-{len(lowercase_ )+1:05d}-of-???.bin""" ) ) rename_and_save_block(lowercase_ , lowercase_ ) sharded_state_dicts.append(current_block.keys() ) del current_block A__ : Any = {} A__ : str = 0 A__ : List[str] = raw_weights.to(getattr(lowercase_ , lowercase_ ) ) current_block_size += weight_size total_size += weight_size # Add the last block A__ : Union[str, Any] = os.path.join(lowercase_ , weights_name.replace(""".bin""" , f"""-{len(lowercase_ )+1:05d}-of-???.bin""" ) ) rename_and_save_block(lowercase_ , lowercase_ ) sharded_state_dicts.append(current_block.keys() ) # If we only have one shard, we return it if len(lowercase_ ) == 1: return {weights_name: sharded_state_dicts[0]}, None # Otherwise, let's build the index A__ : str = {} A__ : Any = {} for idx, shard in enumerate(lowercase_ ): A__ : Any = weights_name.replace( """.bin""" , f"""-{idx+1:05d}-of-{len(lowercase_ ):05d}.bin""" ) # len(sharded_state_dicts):05d} A__ : Dict = os.path.join(lowercase_ , weights_name.replace(""".bin""" , f"""-{idx+1:05d}-of-???.bin""" ) ) os.rename(lowercase_ , os.path.join(lowercase_ , lowercase_ ) ) A__ : str = shard for key in shard: A__ : Any = shard_file # Add the metadata A__ : Tuple = {"""total_size""": total_size} A__ : Union[str, Any] = {"""metadata""": metadata, """weight_map""": weight_map} with open(os.path.join(lowercase_ , lowercase_ ) , """w""" , encoding="""utf-8""" ) as f: A__ : Dict = json.dumps(lowercase_ , indent=2 , sort_keys=lowercase_ ) + """\n""" f.write(lowercase_ ) return metadata, index if __name__ == "__main__": A_ : Optional[int] = argparse.ArgumentParser() # Required parameters parser.add_argument( '--switch_t5x_checkpoint_path', default='/mnt/disks/disk_switch/original_checkpoints/switch-xxl-128/checkpoint_634600', type=str, required=False, help='Path to a directory containing a folder per layer. Follows the original Google format.', ) parser.add_argument('--max_shard_size', default='10GB', required=False, help='Max shard size') parser.add_argument('--dtype', default='bfloat16', type=str, required=False, help='dtype of the saved model') parser.add_argument( '--pytorch_dump_folder_path', default='/mnt/disks/disk_switch/original_checkpoints/switch-xxl-128-converted', type=str, required=False, help='Path to the output pytorch model.', ) A_ : Dict = parser.parse_args() shard_on_the_fly( args.switch_tax_checkpoint_path, args.pytorch_dump_folder_path, args.max_shard_size, args.dtype, ) def UpperCamelCase () -> int: from transformers import SwitchTransformersConfig, SwitchTransformersForConditionalGeneration, TaTokenizer A__ : str = SwitchTransformersConfig.from_pretrained("""google/switch-base-8""" ) config.save_pretrained("""/home/arthur_huggingface_co/transformers/switch_converted""" ) A__ : str = SwitchTransformersForConditionalGeneration.from_pretrained( """/home/arthur_huggingface_co/transformers/switch_converted""" , device_map="""auto""" ) A__ : Tuple = TaTokenizer.from_pretrained("""t5-small""" ) A__ : Dict = """A <extra_id_0> walks into a bar a orders a <extra_id_1> with <extra_id_2> pinch of <extra_id_3>.""" A__ : Union[str, Any] = tokenizer(lowercase_ , return_tensors="""pt""" ).input_ids A__ : Tuple = model.generate(lowercase_ , decoder_start_token_id=0 ) print(tokenizer.decode(out[0] ) )
64
1
import math def UpperCamelCase (lowercase_: Dict , lowercase_: List[Any] ) -> Optional[int]: if 0 not in (x, y): # We use the relation x^y = y*log10(x), where 10 is the base. return y * math.logaa(lowercase_ ) else: if x == 0: # 0 raised to any number is 0 return 0 elif y == 0: return 1 # any number raised to 0 is 1 raise AssertionError("""This should never happen""" ) if __name__ == "__main__": # Main function # Read two numbers from input and typecast them to int using map function. # Here x is the base and y is the power. A_ : Optional[int] = 'Enter the base and the power separated by a comma: ' A_ , A_ : Dict = map(int, input(prompt).split(',')) A_ , A_ : Optional[Any] = map(int, input(prompt).split(',')) # We find the log of each number, using the function res(), which takes two # arguments. A_ : Any = res(xa, ya) A_ : str = res(xa, ya) # We check for the largest number if resa > resa: print('Largest number is', xa, '^', ya) elif resa > resa: print('Largest number is', xa, '^', ya) else: print('Both are equal')
64
from typing import TYPE_CHECKING from ...utils import OptionalDependencyNotAvailable, _LazyModule, is_sentencepiece_available A_ : Optional[Any] = {} try: if not is_sentencepiece_available(): raise OptionalDependencyNotAvailable() except OptionalDependencyNotAvailable: pass else: A_ : str = ['BartphoTokenizer'] if TYPE_CHECKING: try: if not is_sentencepiece_available(): raise OptionalDependencyNotAvailable() except OptionalDependencyNotAvailable: pass else: from .tokenization_bartpho import BartphoTokenizer else: import sys A_ : Optional[int] = _LazyModule(__name__, globals()['__file__'], _import_structure, module_spec=__spec__)
64
1
import importlib import torch import yaml from omegaconf import OmegaConf from taming.models.vqgan import VQModel def UpperCamelCase (lowercase_: Any , lowercase_: Tuple=False ) -> Dict: A__ : List[str] = OmegaConf.load(lowercase_ ) if display: print(yaml.dump(OmegaConf.to_container(lowercase_ ) ) ) return config def UpperCamelCase (lowercase_: Tuple , lowercase_: Dict=None , lowercase_: Optional[int]=None ) -> List[Any]: if conf_path is None: A__ : Any = """./model_checkpoints/vqgan_only.yaml""" A__ : List[str] = load_config(lowercase_ , display=lowercase_ ) A__ : List[str] = VQModel(**config.model.params ) if ckpt_path is None: A__ : Optional[int] = """./model_checkpoints/vqgan_only.pt""" A__ : Union[str, Any] = torch.load(lowercase_ , map_location=lowercase_ ) if ".ckpt" in ckpt_path: A__ : Dict = sd["""state_dict"""] model.load_state_dict(lowercase_ , strict=lowercase_ ) model.to(lowercase_ ) del sd return model def UpperCamelCase (lowercase_: Tuple , lowercase_: Tuple ) -> Optional[int]: A__ , A__ , A__ : str = model.encode(lowercase_ ) print(f"""VQGAN --- {model.__class__.__name__}: latent shape: {z.shape[2:]}""" ) A__ : Tuple = model.decode(lowercase_ ) return xrec def UpperCamelCase (lowercase_: List[Any] , lowercase_: Tuple=False ) -> str: A__ , A__ : int = string.rsplit(""".""" , 1 ) if reload: A__ : List[Any] = importlib.import_module(lowercase_ ) importlib.reload(lowercase_ ) return getattr(importlib.import_module(lowercase_ , package=lowercase_ ) , cls ) def UpperCamelCase (lowercase_: List[str] ) -> List[Any]: if "target" not in config: raise KeyError("""Expected key `target` to instantiate.""" ) return get_obj_from_str(config["""target"""] )(**config.get("""params""" , {} ) ) def UpperCamelCase (lowercase_: int , lowercase_: Optional[Any] , lowercase_: Optional[Any]=True , lowercase_: Tuple=True ) -> str: A__ : Union[str, Any] = instantiate_from_config(lowercase_ ) if sd is not None: model.load_state_dict(lowercase_ ) if gpu: model.cuda() if eval_mode: model.eval() return {"model": model} def UpperCamelCase (lowercase_: Any , lowercase_: Tuple , lowercase_: Union[str, Any] , lowercase_: Union[str, Any] ) -> int: # load the specified checkpoint if ckpt: A__ : Any = torch.load(lowercase_ , map_location="""cpu""" ) A__ : str = pl_sd["""global_step"""] print(f"""loaded model from global step {global_step}.""" ) else: A__ : Any = {"""state_dict""": None} A__ : Dict = None A__ : Optional[int] = load_model_from_config(config.model , pl_sd["""state_dict"""] , gpu=lowercase_ , eval_mode=lowercase_ )["""model"""] return model, global_step
64
import argparse import hashlib import os import urllib import warnings import torch from torch import nn from tqdm import tqdm from transformers import WhisperConfig, WhisperForConditionalGeneration A_ : Dict = { 'tiny.en': 'https://openaipublic.azureedge.net/main/whisper/models/d3dd57d32accea0b295c96e26691aa14d8822fac7d9d27d5dc00b4ca2826dd03/tiny.en.pt', 'tiny': 'https://openaipublic.azureedge.net/main/whisper/models/65147644a518d12f04e32d6f3b26facc3f8dd46e5390956a9424a650c0ce22b9/tiny.pt', 'base.en': 'https://openaipublic.azureedge.net/main/whisper/models/25a8566e1d0c1e2231d1c762132cd20e0f96a85d16145c3a00adf5d1ac670ead/base.en.pt', 'base': 'https://openaipublic.azureedge.net/main/whisper/models/ed3a0b6b1c0edf879ad9b11b1af5a0e6ab5db9205f891f668f8b0e6c6326e34e/base.pt', 'small.en': 'https://openaipublic.azureedge.net/main/whisper/models/f953ad0fd29cacd07d5a9eda5624af0f6bcf2258be67c92b79389873d91e0872/small.en.pt', 'small': 'https://openaipublic.azureedge.net/main/whisper/models/9ecf779972d90ba49c06d968637d720dd632c55bbf19d441fb42bf17a411e794/small.pt', 'medium.en': 'https://openaipublic.azureedge.net/main/whisper/models/d7440d1dc186f76616474e0ff0b3b6b879abc9d1a4926b7adfa41db2d497ab4f/medium.en.pt', 'medium': 'https://openaipublic.azureedge.net/main/whisper/models/345ae4da62f9b3d59415adc60127b97c714f32e89e936602e85993674d08dcb1/medium.pt', 'large': 'https://openaipublic.azureedge.net/main/whisper/models/e4b87e7e0bf463eb8e6956e646f1e277e901512310def2c24bf0e11bd3c28e9a/large.pt', 'large-v2': 'https://openaipublic.azureedge.net/main/whisper/models/81f7c96c852ee8fc832187b0132e569d6c3065a3252ed18e56effd0b6a73e524/large-v2.pt', } def UpperCamelCase (lowercase_: Optional[Any] ) -> Optional[int]: A__ : List[Any] = ["""layers""", """blocks"""] for k in ignore_keys: state_dict.pop(lowercase_ , lowercase_ ) A_ : Any = { 'blocks': 'layers', 'mlp.0': 'fc1', 'mlp.2': 'fc2', 'mlp_ln': 'final_layer_norm', '.attn.query': '.self_attn.q_proj', '.attn.key': '.self_attn.k_proj', '.attn.value': '.self_attn.v_proj', '.attn_ln': '.self_attn_layer_norm', '.attn.out': '.self_attn.out_proj', '.cross_attn.query': '.encoder_attn.q_proj', '.cross_attn.key': '.encoder_attn.k_proj', '.cross_attn.value': '.encoder_attn.v_proj', '.cross_attn_ln': '.encoder_attn_layer_norm', '.cross_attn.out': '.encoder_attn.out_proj', 'decoder.ln.': 'decoder.layer_norm.', 'encoder.ln.': 'encoder.layer_norm.', 'token_embedding': 'embed_tokens', 'encoder.positional_embedding': 'encoder.embed_positions.weight', 'decoder.positional_embedding': 'decoder.embed_positions.weight', 'ln_post': 'layer_norm', } def UpperCamelCase (lowercase_: str ) -> Any: A__ : Dict = list(s_dict.keys() ) for key in keys: A__ : List[str] = key for k, v in WHISPER_MAPPING.items(): if k in key: A__ : List[Any] = new_key.replace(lowercase_ , lowercase_ ) print(f"""{key} -> {new_key}""" ) A__ : Tuple = s_dict.pop(lowercase_ ) return s_dict def UpperCamelCase (lowercase_: Tuple ) -> Optional[int]: A__ , A__ : Any = emb.weight.shape A__ : str = nn.Linear(lowercase_ , lowercase_ , bias=lowercase_ ) A__ : Union[str, Any] = emb.weight.data return lin_layer def UpperCamelCase (lowercase_: str , lowercase_: str ) -> bytes: os.makedirs(lowercase_ , exist_ok=lowercase_ ) A__ : Tuple = os.path.basename(lowercase_ ) A__ : int = url.split("""/""" )[-2] A__ : Dict = os.path.join(lowercase_ , lowercase_ ) if os.path.exists(lowercase_ ) and not os.path.isfile(lowercase_ ): raise RuntimeError(f"""{download_target} exists and is not a regular file""" ) if os.path.isfile(lowercase_ ): A__ : Optional[Any] = open(lowercase_ , """rb""" ).read() if hashlib.shaaaa(lowercase_ ).hexdigest() == expected_shaaaa: return model_bytes else: warnings.warn(f"""{download_target} exists, but the SHA256 checksum does not match; re-downloading the file""" ) with urllib.request.urlopen(lowercase_ ) as source, open(lowercase_ , """wb""" ) as output: with tqdm( total=int(source.info().get("""Content-Length""" ) ) , ncols=80 , unit="""iB""" , unit_scale=lowercase_ , unit_divisor=1024 ) as loop: while True: A__ : Any = source.read(8192 ) if not buffer: break output.write(lowercase_ ) loop.update(len(lowercase_ ) ) A__ : Dict = open(lowercase_ , """rb""" ).read() if hashlib.shaaaa(lowercase_ ).hexdigest() != expected_shaaaa: raise RuntimeError( """Model has been downloaded but the SHA256 checksum does not not match. Please retry loading the model.""" ) return model_bytes def UpperCamelCase (lowercase_: List[Any] , lowercase_: Tuple ) -> Optional[Any]: if ".pt" not in checkpoint_path: A__ : Tuple = _download(_MODELS[checkpoint_path] ) else: A__ : Optional[int] = torch.load(lowercase_ , map_location="""cpu""" ) A__ : str = original_checkpoint["""dims"""] A__ : List[Any] = original_checkpoint["""model_state_dict"""] A__ : Optional[Any] = state_dict["""decoder.token_embedding.weight"""] remove_ignore_keys_(lowercase_ ) rename_keys(lowercase_ ) A__ : List[str] = True A__ : Optional[Any] = state_dict["""decoder.layers.0.fc1.weight"""].shape[0] A__ : List[Any] = WhisperConfig( vocab_size=dimensions["""n_vocab"""] , encoder_ffn_dim=lowercase_ , decoder_ffn_dim=lowercase_ , num_mel_bins=dimensions["""n_mels"""] , d_model=dimensions["""n_audio_state"""] , max_target_positions=dimensions["""n_text_ctx"""] , encoder_layers=dimensions["""n_audio_layer"""] , encoder_attention_heads=dimensions["""n_audio_head"""] , decoder_layers=dimensions["""n_text_layer"""] , decoder_attention_heads=dimensions["""n_text_state"""] , max_source_positions=dimensions["""n_audio_ctx"""] , ) A__ : Optional[Any] = WhisperForConditionalGeneration(lowercase_ ) A__ , A__ : List[Any] = model.model.load_state_dict(lowercase_ , strict=lowercase_ ) if len(lowercase_ ) > 0 and not set(lowercase_ ) <= { "encoder.embed_positions.weights", "decoder.embed_positions.weights", }: raise ValueError( """Only `encoder.embed_positions.weights` and `decoder.embed_positions.weights` are allowed to be missing,""" f""" but all the following weights are missing {missing}""" ) if tie_embeds: A__ : Any = make_linear_from_emb(model.model.decoder.embed_tokens ) else: A__ : str = proj_out_weights model.save_pretrained(lowercase_ ) if __name__ == "__main__": A_ : Any = argparse.ArgumentParser() # # Required parameters parser.add_argument('--checkpoint_path', type=str, help='Patht to the downloaded checkpoints') parser.add_argument('--pytorch_dump_folder_path', default=None, type=str, help='Path to the output PyTorch model.') A_ : Tuple = parser.parse_args() convert_openai_whisper_to_tfms(args.checkpoint_path, args.pytorch_dump_folder_path)
64
1
import argparse from torch import nn # transformers_old should correspond to branch `save_old_prophetnet_model_structure` here # original prophetnet_checkpoints are saved under `patrickvonplaten/..._old` respectively from transformers_old.modeling_prophetnet import ( ProphetNetForConditionalGeneration as ProphetNetForConditionalGenerationOld, ) from transformers_old.modeling_xlm_prophetnet import ( XLMProphetNetForConditionalGeneration as XLMProphetNetForConditionalGenerationOld, ) from transformers import ProphetNetForConditionalGeneration, XLMProphetNetForConditionalGeneration, logging A_ : Dict = logging.get_logger(__name__) logging.set_verbosity_info() def UpperCamelCase (lowercase_: str , lowercase_: str ) -> Any: if "xprophetnet" in prophetnet_checkpoint_path: A__ : Union[str, Any] = XLMProphetNetForConditionalGenerationOld.from_pretrained(lowercase_ ) A__ , A__ : int = XLMProphetNetForConditionalGeneration.from_pretrained( lowercase_ , output_loading_info=lowercase_ ) else: A__ : int = ProphetNetForConditionalGenerationOld.from_pretrained(lowercase_ ) A__ , A__ : Optional[int] = ProphetNetForConditionalGeneration.from_pretrained( lowercase_ , output_loading_info=lowercase_ ) A__ : Any = ["""key_proj""", """value_proj""", """query_proj"""] A__ : Optional[int] = { """self_attn""": """ngram_self_attn""", """cross_attn""": """encoder_attn""", """cross_attn_layer_norm""": """encoder_attn_layer_norm""", """feed_forward_layer_norm""": """final_layer_norm""", """feed_forward""": """""", """intermediate""": """fc1""", """output""": """fc2""", """key_proj""": """k_proj""", """query_proj""": """q_proj""", """value_proj""": """v_proj""", """word_embeddings""": """embed_tokens""", """embeddings_layer_norm""": """emb_layer_norm""", """relative_pos_embeddings""": """relative_linear""", """ngram_embeddings""": """ngram_input_embed""", """position_embeddings""": """embed_positions""", } for key in loading_info["missing_keys"]: A__ : Optional[int] = key.split(""".""" ) if attributes[0] == "lm_head": A__ : int = prophet A__ : Tuple = prophet_old else: A__ : Optional[int] = prophet.prophetnet A__ : Optional[int] = prophet_old.model A__ : List[str] = False for attribute in attributes: if attribute in mapping: A__ : str = mapping[attribute] if not hasattr(lowercase_ , lowercase_ ) and len(lowercase_ ) > 0: A__ : Union[str, Any] = attribute elif hasattr(lowercase_ , lowercase_ ): A__ : List[Any] = attribute if attribute == "weight": assert old_model.weight.shape == model.weight.shape, "Shapes have to match!" A__ : Union[str, Any] = old_model.weight logger.info(f"""{attribute} is initialized.""" ) A__ : List[str] = True break elif attribute == "bias": assert old_model.bias.shape == model.bias.shape, "Shapes have to match!" A__ : Tuple = old_model.bias logger.info(f"""{attribute} is initialized""" ) A__ : Optional[Any] = True break elif attribute in special_keys and hasattr(lowercase_ , """in_proj_weight""" ): A__ : Union[str, Any] = old_model.in_proj_weight.shape[0] // 3 A__ : Dict = getattr(lowercase_ , lowercase_ ) param.weight.shape == old_model.in_proj_weight[:embed_dim, :].shape, "Shapes have to match" param.bias.shape == old_model.in_proj_bias[:embed_dim].shape, "Shapes have to match" if attribute == "query_proj": A__ : Tuple = nn.Parameter(old_model.in_proj_weight[:embed_dim, :] ) A__ : Optional[Any] = nn.Parameter(old_model.in_proj_bias[:embed_dim] ) elif attribute == "key_proj": A__ : int = nn.Parameter(old_model.in_proj_weight[embed_dim : 2 * embed_dim, :] ) A__ : List[Any] = nn.Parameter(old_model.in_proj_bias[embed_dim : 2 * embed_dim] ) elif attribute == "value_proj": A__ : Union[str, Any] = nn.Parameter(old_model.in_proj_weight[2 * embed_dim :, :] ) A__ : Dict = nn.Parameter(old_model.in_proj_bias[2 * embed_dim :] ) A__ : Union[str, Any] = True break elif attribute == "position_embeddings": assert ( model.position_embeddings.weight.shape[-1] == old_model.embed_positions.weight.shape[-1] ), "Hidden size has to match" assert model.position_embeddings.weight.shape[0] == 512, "We want 512 position_embeddings." A__ : Union[str, Any] = nn.Parameter(old_model.embed_positions.weight[:512, :] ) A__ : str = True break if attribute.isdigit(): A__ : Optional[Any] = model[int(lowercase_ )] A__ : Optional[int] = old_model[int(lowercase_ )] else: A__ : List[Any] = getattr(lowercase_ , lowercase_ ) if old_attribute == "": A__ : Any = old_model else: if not hasattr(lowercase_ , lowercase_ ): raise ValueError(f"""{old_model} does not have {old_attribute}""" ) A__ : str = getattr(lowercase_ , lowercase_ ) if not is_key_init: raise ValueError(f"""{key} was not correctly initialized!""" ) print(f"""Saving model to {pytorch_dump_folder_path}""" ) prophet.save_pretrained(lowercase_ ) if __name__ == "__main__": A_ : str = argparse.ArgumentParser() # Required parameters parser.add_argument( '--prophetnet_checkpoint_path', default=None, type=str, required=True, help='Path the official PyTorch dump.' ) parser.add_argument( '--pytorch_dump_folder_path', default=None, type=str, required=True, help='Path to the output PyTorch model.' ) A_ : Union[str, Any] = parser.parse_args() convert_prophetnet_checkpoint_to_pytorch(args.prophetnet_checkpoint_path, args.pytorch_dump_folder_path)
64
import unittest import numpy as np import torch from transformers import CLIPTextConfig, CLIPTextModel, CLIPTokenizer from diffusers import ( AutoencoderKL, DDIMScheduler, DPMSolverMultistepScheduler, TextToVideoSDPipeline, UNetaDConditionModel, ) from diffusers.utils import is_xformers_available, load_numpy, skip_mps, slow, torch_device from diffusers.utils.testing_utils import enable_full_determinism from ..pipeline_params import TEXT_TO_IMAGE_BATCH_PARAMS, TEXT_TO_IMAGE_PARAMS from ..test_pipelines_common import PipelineTesterMixin enable_full_determinism() @skip_mps class _a (__magic_name__ , unittest.TestCase ): '''simple docstring''' UpperCAmelCase__: Any = TextToVideoSDPipeline UpperCAmelCase__: Any = TEXT_TO_IMAGE_PARAMS UpperCAmelCase__: Optional[Any] = TEXT_TO_IMAGE_BATCH_PARAMS # No `output_type`. UpperCAmelCase__: Optional[int] = frozenset( [ '''num_inference_steps''', '''generator''', '''latents''', '''return_dict''', '''callback''', '''callback_steps''', ] ) def __A ( self ): torch.manual_seed(0 ) A__ : Optional[int] = UNetaDConditionModel( block_out_channels=(32, 64, 64, 64) , layers_per_block=2 , sample_size=32 , in_channels=4 , out_channels=4 , down_block_types=("""CrossAttnDownBlock3D""", """CrossAttnDownBlock3D""", """CrossAttnDownBlock3D""", """DownBlock3D""") , up_block_types=("""UpBlock3D""", """CrossAttnUpBlock3D""", """CrossAttnUpBlock3D""", """CrossAttnUpBlock3D""") , cross_attention_dim=32 , attention_head_dim=4 , ) A__ : Optional[int] = DDIMScheduler( beta_start=0.0_0_0_8_5 , beta_end=0.0_1_2 , beta_schedule="""scaled_linear""" , clip_sample=A__ , set_alpha_to_one=A__ , ) torch.manual_seed(0 ) A__ : Union[str, Any] = AutoencoderKL( block_out_channels=[32, 64] , in_channels=3 , out_channels=3 , down_block_types=["""DownEncoderBlock2D""", """DownEncoderBlock2D"""] , up_block_types=["""UpDecoderBlock2D""", """UpDecoderBlock2D"""] , latent_channels=4 , sample_size=128 , ) torch.manual_seed(0 ) A__ : Union[str, Any] = CLIPTextConfig( bos_token_id=0 , eos_token_id=2 , hidden_size=32 , intermediate_size=37 , layer_norm_eps=1e-05 , num_attention_heads=4 , num_hidden_layers=5 , pad_token_id=1 , vocab_size=1000 , hidden_act="""gelu""" , projection_dim=512 , ) A__ : Union[str, Any] = CLIPTextModel(A__ ) A__ : Tuple = CLIPTokenizer.from_pretrained("""hf-internal-testing/tiny-random-clip""" ) A__ : Dict = { """unet""": unet, """scheduler""": scheduler, """vae""": vae, """text_encoder""": text_encoder, """tokenizer""": tokenizer, } return components def __A ( self , A__ , A__=0 ): if str(A__ ).startswith("""mps""" ): A__ : Tuple = torch.manual_seed(A__ ) else: A__ : List[str] = torch.Generator(device=A__ ).manual_seed(A__ ) A__ : List[str] = { """prompt""": """A painting of a squirrel eating a burger""", """generator""": generator, """num_inference_steps""": 2, """guidance_scale""": 6.0, """output_type""": """pt""", } return inputs def __A ( self ): A__ : List[str] = """cpu""" # ensure determinism for the device-dependent torch.Generator A__ : Union[str, Any] = self.get_dummy_components() A__ : Union[str, Any] = TextToVideoSDPipeline(**A__ ) A__ : int = sd_pipe.to(A__ ) sd_pipe.set_progress_bar_config(disable=A__ ) A__ : int = self.get_dummy_inputs(A__ ) A__ : int = """np""" A__ : Any = sd_pipe(**A__ ).frames A__ : Dict = frames[0][-3:, -3:, -1] assert frames[0].shape == (64, 64, 3) A__ : Optional[Any] = np.array([1_5_8.0, 1_6_0.0, 1_5_3.0, 1_2_5.0, 1_0_0.0, 1_2_1.0, 1_1_1.0, 9_3.0, 1_1_3.0] ) assert np.abs(image_slice.flatten() - expected_slice ).max() < 1e-2 def __A ( self ): self._test_attention_slicing_forward_pass(test_mean_pixel_difference=A__ , expected_max_diff=3e-3 ) @unittest.skipIf( torch_device != """cuda""" or not is_xformers_available() , reason="""XFormers attention is only available with CUDA and `xformers` installed""" , ) def __A ( self ): self._test_xformers_attention_forwardGenerator_pass(test_mean_pixel_difference=A__ , expected_max_diff=1e-2 ) @unittest.skip(reason="""Batching needs to be properly figured out first for this pipeline.""" ) def __A ( self ): pass @unittest.skip(reason="""Batching needs to be properly figured out first for this pipeline.""" ) def __A ( self ): pass @unittest.skip(reason="""`num_images_per_prompt` argument is not supported for this pipeline.""" ) def __A ( self ): pass def __A ( self ): return super().test_progress_bar() @slow @skip_mps class _a (unittest.TestCase ): '''simple docstring''' def __A ( self ): A__ : Union[str, Any] = load_numpy( """https://huggingface.co/datasets/hf-internal-testing/diffusers-images/resolve/main/text_to_video/video.npy""" ) A__ : Tuple = TextToVideoSDPipeline.from_pretrained("""damo-vilab/text-to-video-ms-1.7b""" ) A__ : Any = DPMSolverMultistepScheduler.from_config(pipe.scheduler.config ) A__ : int = pipe.to("""cuda""" ) A__ : Optional[Any] = """Spiderman is surfing""" A__ : List[str] = torch.Generator(device="""cpu""" ).manual_seed(0 ) A__ : Optional[Any] = pipe(A__ , generator=A__ , num_inference_steps=25 , output_type="""pt""" ).frames A__ : Dict = video_frames.cpu().numpy() assert np.abs(expected_video - video ).mean() < 5e-2 def __A ( self ): A__ : List[Any] = load_numpy( """https://huggingface.co/datasets/hf-internal-testing/diffusers-images/resolve/main/text_to_video/video_2step.npy""" ) A__ : Optional[int] = TextToVideoSDPipeline.from_pretrained("""damo-vilab/text-to-video-ms-1.7b""" ) A__ : List[str] = pipe.to("""cuda""" ) A__ : Dict = """Spiderman is surfing""" A__ : Union[str, Any] = torch.Generator(device="""cpu""" ).manual_seed(0 ) A__ : Optional[int] = pipe(A__ , generator=A__ , num_inference_steps=2 , output_type="""pt""" ).frames A__ : Optional[int] = video_frames.cpu().numpy() assert np.abs(expected_video - video ).mean() < 5e-2
64
1
from typing import List, Optional, Union import numpy as np import PIL.Image from ...image_processing_utils import BaseImageProcessor, BatchFeature from ...image_transforms import rescale, resize, to_channel_dimension_format from ...image_utils import ( ChannelDimension, PILImageResampling, get_image_size, make_list_of_images, to_numpy_array, valid_images, ) from ...utils import TensorType, logging A_ : Union[str, Any] = logging.get_logger(__name__) class _a (__magic_name__ ): '''simple docstring''' UpperCAmelCase__: Union[str, Any] = ['''pixel_values'''] def __init__( self , A__ = True , A__ = 32 , A__=PILImageResampling.BILINEAR , A__ = True , **A__ , ): A__ : int = do_resize A__ : Tuple = do_rescale A__ : Tuple = size_divisor A__ : List[Any] = resample super().__init__(**A__ ) def __A ( self , A__ , A__ , A__ , A__ = None , **A__ ): A__ , A__ : str = get_image_size(A__ ) # Rounds the height and width down to the closest multiple of size_divisor A__ : List[str] = height // size_divisor * size_divisor A__ : List[Any] = width // size_divisor * size_divisor A__ : List[Any] = resize(A__ , (new_h, new_w) , resample=A__ , data_format=A__ , **A__ ) return image def __A ( self , A__ , A__ , A__ = None , **A__ ): return rescale(image=A__ , scale=A__ , data_format=A__ , **A__ ) def __A ( self , A__ , A__ = None , A__ = None , A__=None , A__ = None , A__ = None , A__ = ChannelDimension.FIRST , **A__ , ): A__ : Any = do_resize if do_resize is not None else self.do_resize A__ : Union[str, Any] = do_rescale if do_rescale is not None else self.do_rescale A__ : int = size_divisor if size_divisor is not None else self.size_divisor A__ : Optional[Any] = resample if resample is not None else self.resample if do_resize and size_divisor is None: raise ValueError("""size_divisor is required for resizing""" ) A__ : str = make_list_of_images(A__ ) if not valid_images(A__ ): raise ValueError("""Invalid image(s)""" ) # All transformations expect numpy arrays. A__ : int = [to_numpy_array(A__ ) for img in images] if do_resize: A__ : Optional[Any] = [self.resize(A__ , size_divisor=A__ , resample=A__ ) for image in images] if do_rescale: A__ : List[Any] = [self.rescale(A__ , scale=1 / 255 ) for image in images] A__ : Union[str, Any] = [to_channel_dimension_format(A__ , A__ ) for image in images] A__ : int = {"""pixel_values""": images} return BatchFeature(data=A__ , tensor_type=A__ )
64
def UpperCamelCase (lowercase_: int ) -> int: if not isinstance(lowercase_ , lowercase_ ): raise TypeError("""Input value must be an 'int' type""" ) A__ : int = 0 while number: position += 1 number >>= 1 return position if __name__ == "__main__": import doctest doctest.testmod()
64
1
import timeit import numpy as np import datasets from datasets.arrow_writer import ArrowWriter from datasets.features.features import _ArrayXD def UpperCamelCase (lowercase_: Dict ) -> List[str]: def wrapper(*lowercase_: List[Any] , **lowercase_: str ): A__ : List[Any] = timeit.default_timer() A__ : List[str] = func(*lowercase_ , **lowercase_ ) A__ : Any = timeit.default_timer() - starttime return delta A__ : List[str] = func.__name__ return wrapper def UpperCamelCase (lowercase_: dict , lowercase_: Any=100 , lowercase_: List[Any]=None ) -> str: A__ : Dict = [] A__ : List[Any] = seq_shapes or {} for i in range(lowercase_ ): A__ : Any = {} for col_id, (k, v) in enumerate(features.items() ): if isinstance(lowercase_ , _ArrayXD ): A__ : Union[str, Any] = np.random.rand(*v.shape ).astype(v.dtype ) elif isinstance(lowercase_ , datasets.Value ): if v.dtype == "string": A__ : Optional[Any] = """The small grey turtle was surprisingly fast when challenged.""" else: A__ : List[str] = np.random.randint(10 , size=1 ).astype(v.dtype ).item() elif isinstance(lowercase_ , datasets.Sequence ): while isinstance(lowercase_ , datasets.Sequence ): A__ : Optional[Any] = v.feature A__ : str = seq_shapes[k] A__ : Dict = np.random.rand(*lowercase_ ).astype(v.dtype ) A__ : List[Any] = data dummy_data.append((i, example) ) return dummy_data def UpperCamelCase (lowercase_: Any , lowercase_: Tuple , lowercase_: Union[str, Any]=100 , lowercase_: Optional[int]=None ) -> str: A__ : Union[str, Any] = generate_examples(lowercase_ , num_examples=lowercase_ , seq_shapes=lowercase_ ) with ArrowWriter(features=lowercase_ , path=lowercase_ ) as writer: for key, record in dummy_data: A__ : List[Any] = features.encode_example(lowercase_ ) writer.write(lowercase_ ) A__ , A__ : List[Any] = writer.finalize() if not num_final_examples == num_examples: raise ValueError( f"""Error writing the dataset, wrote {num_final_examples} examples but should have written {num_examples}.""" ) A__ : Optional[Any] = datasets.Dataset.from_file(filename=lowercase_ , info=datasets.DatasetInfo(features=lowercase_ ) ) return dataset
64
from pathlib import Path import cva import numpy as np from matplotlib import pyplot as plt def UpperCamelCase (lowercase_: np.ndarray , lowercase_: np.ndarray , lowercase_: np.ndarray , lowercase_: int , lowercase_: int ) -> np.ndarray: A__ : Any = cva.getAffineTransform(lowercase_ , lowercase_ ) return cva.warpAffine(lowercase_ , lowercase_ , (rows, cols) ) if __name__ == "__main__": # read original image A_ : List[Any] = cva.imread( str(Path(__file__).resolve().parent.parent / 'image_data' / 'lena.jpg') ) # turn image in gray scale value A_ : List[Any] = cva.cvtColor(image, cva.COLOR_BGR2GRAY) # get image shape A_ , A_ : Optional[Any] = gray_img.shape # set different points to rotate image A_ : str = np.array([[50, 50], [200, 50], [50, 200]], np.floataa) A_ : Dict = np.array([[10, 100], [200, 50], [100, 250]], np.floataa) A_ : Optional[int] = np.array([[50, 50], [150, 50], [120, 200]], np.floataa) A_ : Optional[int] = np.array([[10, 100], [80, 50], [180, 250]], np.floataa) # add all rotated images in a list A_ : Dict = [ gray_img, get_rotation(gray_img, ptsa, ptsa, img_rows, img_cols), get_rotation(gray_img, ptsa, ptsa, img_rows, img_cols), get_rotation(gray_img, ptsa, ptsa, img_rows, img_cols), ] # plot different image rotations A_ : Union[str, Any] = plt.figure(1) A_ : Union[str, Any] = ['Original', 'Rotation 1', 'Rotation 2', 'Rotation 3'] for i, image in enumerate(images): plt.subplot(2, 2, i + 1), plt.imshow(image, 'gray') plt.title(titles[i]) plt.axis('off') plt.subplots_adjust(left=0.0, bottom=0.05, right=1.0, top=0.95) plt.show()
64
1
def UpperCamelCase (lowercase_: str ) -> int: A__ : Optional[int] = hex_num.strip() if not hex_num: raise ValueError("""No value was passed to the function""" ) A__ : Optional[Any] = hex_num[0] == """-""" if is_negative: A__ : int = hex_num[1:] try: A__ : Optional[Any] = int(lowercase_ , 16 ) except ValueError: raise ValueError("""Invalid value was passed to the function""" ) A__ : Tuple = """""" while int_num > 0: A__ : List[Any] = str(int_num % 2 ) + bin_str int_num >>= 1 return int(("""-""" + bin_str) if is_negative else bin_str ) if __name__ == "__main__": import doctest doctest.testmod()
64
import os import tempfile import unittest from pathlib import Path from transformers import AutoConfig, is_torch_available from transformers.testing_utils import require_torch, torch_device if is_torch_available(): from transformers import PyTorchBenchmark, PyTorchBenchmarkArguments @require_torch class _a (unittest.TestCase ): '''simple docstring''' def __A ( self , A__ ): for model_result in results.values(): for batch_size, sequence_length in zip(model_result["""bs"""] , model_result["""ss"""] ): A__ : str = model_result["""result"""][batch_size][sequence_length] self.assertIsNotNone(A__ ) def __A ( self ): A__ : Dict = """sshleifer/tiny-gpt2""" A__ : Tuple = PyTorchBenchmarkArguments( models=[MODEL_ID] , training=A__ , inference=A__ , sequence_lengths=[8] , batch_sizes=[1] , multi_process=A__ , ) A__ : int = PyTorchBenchmark(A__ ) A__ : List[Any] = benchmark.run() self.check_results_dict_not_empty(results.time_inference_result ) self.check_results_dict_not_empty(results.memory_inference_result ) def __A ( self ): A__ : Dict = """sgugger/tiny-distilbert-classification""" A__ : Dict = PyTorchBenchmarkArguments( models=[MODEL_ID] , training=A__ , inference=A__ , sequence_lengths=[8] , batch_sizes=[1] , multi_process=A__ , only_pretrain_model=A__ , ) A__ : str = PyTorchBenchmark(A__ ) A__ : List[str] = benchmark.run() self.check_results_dict_not_empty(results.time_inference_result ) self.check_results_dict_not_empty(results.memory_inference_result ) def __A ( self ): A__ : Any = """sshleifer/tiny-gpt2""" A__ : List[Any] = PyTorchBenchmarkArguments( models=[MODEL_ID] , training=A__ , inference=A__ , torchscript=A__ , sequence_lengths=[8] , batch_sizes=[1] , multi_process=A__ , ) A__ : Tuple = PyTorchBenchmark(A__ ) A__ : str = benchmark.run() self.check_results_dict_not_empty(results.time_inference_result ) self.check_results_dict_not_empty(results.memory_inference_result ) @unittest.skipIf(torch_device == """cpu""" , """Cant do half precision""" ) def __A ( self ): A__ : Optional[Any] = """sshleifer/tiny-gpt2""" A__ : Optional[Any] = PyTorchBenchmarkArguments( models=[MODEL_ID] , training=A__ , inference=A__ , fpaa=A__ , sequence_lengths=[8] , batch_sizes=[1] , multi_process=A__ , ) A__ : str = PyTorchBenchmark(A__ ) A__ : Any = benchmark.run() self.check_results_dict_not_empty(results.time_inference_result ) self.check_results_dict_not_empty(results.memory_inference_result ) def __A ( self ): A__ : Optional[Any] = """sshleifer/tiny-gpt2""" A__ : Tuple = AutoConfig.from_pretrained(A__ ) # set architectures equal to `None` A__ : List[Any] = None A__ : str = PyTorchBenchmarkArguments( models=[MODEL_ID] , training=A__ , inference=A__ , sequence_lengths=[8] , batch_sizes=[1] , multi_process=A__ , ) A__ : List[str] = PyTorchBenchmark(A__ , configs=[config] ) A__ : Any = benchmark.run() self.check_results_dict_not_empty(results.time_inference_result ) self.check_results_dict_not_empty(results.memory_inference_result ) def __A ( self ): A__ : Optional[int] = """sshleifer/tiny-gpt2""" A__ : Optional[int] = PyTorchBenchmarkArguments( models=[MODEL_ID] , training=A__ , inference=A__ , sequence_lengths=[8] , batch_sizes=[1] , multi_process=A__ , ) A__ : Any = PyTorchBenchmark(A__ ) A__ : Dict = benchmark.run() self.check_results_dict_not_empty(results.time_train_result ) self.check_results_dict_not_empty(results.memory_train_result ) @unittest.skipIf(torch_device == """cpu""" , """Can't do half precision""" ) def __A ( self ): A__ : Optional[int] = """sshleifer/tiny-gpt2""" A__ : List[str] = PyTorchBenchmarkArguments( models=[MODEL_ID] , training=A__ , inference=A__ , sequence_lengths=[8] , batch_sizes=[1] , fpaa=A__ , multi_process=A__ , ) A__ : Dict = PyTorchBenchmark(A__ ) A__ : Optional[int] = benchmark.run() self.check_results_dict_not_empty(results.time_train_result ) self.check_results_dict_not_empty(results.memory_train_result ) def __A ( self ): A__ : int = """sshleifer/tiny-gpt2""" A__ : Optional[int] = AutoConfig.from_pretrained(A__ ) A__ : str = PyTorchBenchmarkArguments( models=[MODEL_ID] , training=A__ , inference=A__ , sequence_lengths=[8] , batch_sizes=[1] , multi_process=A__ , ) A__ : int = PyTorchBenchmark(A__ , configs=[config] ) A__ : Tuple = benchmark.run() self.check_results_dict_not_empty(results.time_inference_result ) self.check_results_dict_not_empty(results.memory_inference_result ) def __A ( self ): A__ : List[str] = """sshleifer/tinier_bart""" A__ : List[str] = AutoConfig.from_pretrained(A__ ) A__ : List[str] = PyTorchBenchmarkArguments( models=[MODEL_ID] , training=A__ , inference=A__ , sequence_lengths=[8] , batch_sizes=[1] , multi_process=A__ , ) A__ : Union[str, Any] = PyTorchBenchmark(A__ , configs=[config] ) A__ : str = benchmark.run() self.check_results_dict_not_empty(results.time_inference_result ) self.check_results_dict_not_empty(results.memory_inference_result ) def __A ( self ): A__ : Optional[int] = """sshleifer/tiny-gpt2""" A__ : Union[str, Any] = AutoConfig.from_pretrained(A__ ) A__ : Tuple = PyTorchBenchmarkArguments( models=[MODEL_ID] , training=A__ , inference=A__ , sequence_lengths=[8] , batch_sizes=[1] , multi_process=A__ , ) A__ : int = PyTorchBenchmark(A__ , configs=[config] ) A__ : Tuple = benchmark.run() self.check_results_dict_not_empty(results.time_train_result ) self.check_results_dict_not_empty(results.memory_train_result ) def __A ( self ): A__ : Dict = """sshleifer/tinier_bart""" A__ : int = AutoConfig.from_pretrained(A__ ) A__ : Union[str, Any] = PyTorchBenchmarkArguments( models=[MODEL_ID] , training=A__ , inference=A__ , sequence_lengths=[8] , batch_sizes=[1] , multi_process=A__ , ) A__ : List[Any] = PyTorchBenchmark(A__ , configs=[config] ) A__ : Optional[Any] = benchmark.run() self.check_results_dict_not_empty(results.time_train_result ) self.check_results_dict_not_empty(results.memory_train_result ) def __A ( self ): A__ : int = """sshleifer/tiny-gpt2""" with tempfile.TemporaryDirectory() as tmp_dir: A__ : Dict = PyTorchBenchmarkArguments( models=[MODEL_ID] , training=A__ , inference=A__ , save_to_csv=A__ , sequence_lengths=[8] , batch_sizes=[1] , inference_time_csv_file=os.path.join(A__ , """inf_time.csv""" ) , train_memory_csv_file=os.path.join(A__ , """train_mem.csv""" ) , inference_memory_csv_file=os.path.join(A__ , """inf_mem.csv""" ) , train_time_csv_file=os.path.join(A__ , """train_time.csv""" ) , env_info_csv_file=os.path.join(A__ , """env.csv""" ) , multi_process=A__ , ) A__ : Optional[Any] = PyTorchBenchmark(A__ ) benchmark.run() self.assertTrue(Path(os.path.join(A__ , """inf_time.csv""" ) ).exists() ) self.assertTrue(Path(os.path.join(A__ , """train_time.csv""" ) ).exists() ) self.assertTrue(Path(os.path.join(A__ , """inf_mem.csv""" ) ).exists() ) self.assertTrue(Path(os.path.join(A__ , """train_mem.csv""" ) ).exists() ) self.assertTrue(Path(os.path.join(A__ , """env.csv""" ) ).exists() ) def __A ( self ): A__ : Optional[int] = """sshleifer/tiny-gpt2""" def _check_summary_is_not_empty(A__ ): self.assertTrue(hasattr(A__ , """sequential""" ) ) self.assertTrue(hasattr(A__ , """cumulative""" ) ) self.assertTrue(hasattr(A__ , """current""" ) ) self.assertTrue(hasattr(A__ , """total""" ) ) with tempfile.TemporaryDirectory() as tmp_dir: A__ : Dict = PyTorchBenchmarkArguments( models=[MODEL_ID] , training=A__ , inference=A__ , sequence_lengths=[8] , batch_sizes=[1] , log_filename=os.path.join(A__ , """log.txt""" ) , log_print=A__ , trace_memory_line_by_line=A__ , multi_process=A__ , ) A__ : Dict = PyTorchBenchmark(A__ ) A__ : str = benchmark.run() _check_summary_is_not_empty(result.inference_summary ) _check_summary_is_not_empty(result.train_summary ) self.assertTrue(Path(os.path.join(A__ , """log.txt""" ) ).exists() )
64
1