code
stringlengths
86
54.5k
code_codestyle
int64
0
371
style_context
stringlengths
87
49.2k
style_context_codestyle
int64
0
349
label
int64
0
1
'''simple docstring''' import argparse import logging import os import sys import numpy as np import onnxruntime import torch from bart_onnx.generation_onnx import BARTBeamSearchGenerator from bart_onnx.reduce_onnx_size import remove_dup_initializers import transformers from transformers import BartForConditionalGeneration, BartTokenizer logging.basicConfig( format='%(asctime)s | %(levelname)s | %(name)s | [%(filename)s:%(lineno)d] %(message)s', datefmt='%Y-%m-%d %H:%M:%S', level=os.environ.get('LOGLEVEL', 'INFO').upper(), stream=sys.stdout, ) __snake_case : List[str] = logging.getLogger(__name__) __snake_case : Optional[int] = {'facebook/bart-base': BartForConditionalGeneration} __snake_case : Tuple = {'facebook/bart-base': BartTokenizer} def __lowerCamelCase ( ) -> str: """simple docstring""" A__ : List[Any] =argparse.ArgumentParser(description="""Export Bart model + Beam Search to ONNX graph.""" ) parser.add_argument( """--validation_file""", type=__snake_case, default=__snake_case, help="""A csv or a json file containing the validation data.""" ) parser.add_argument( """--max_length""", type=__snake_case, default=5, help="""The maximum total input sequence length after tokenization.""", ) parser.add_argument( """--num_beams""", type=__snake_case, default=__snake_case, help=( """Number of beams to use for evaluation. This argument will be """ """passed to ``model.generate``, which is used during ``evaluate`` and ``predict``.""" ), ) parser.add_argument( """--model_name_or_path""", type=__snake_case, help="""Path to pretrained model or model identifier from huggingface.co/models.""", required=__snake_case, ) parser.add_argument( """--config_name""", type=__snake_case, default=__snake_case, help="""Pretrained config name or path if not the same as model_name""", ) parser.add_argument( """--device""", type=__snake_case, default="""cpu""", help="""Device where the model will be run""", ) parser.add_argument("""--output_file_path""", type=__snake_case, default=__snake_case, help="""Where to store the final ONNX file.""" ) A__ : Union[str, Any] =parser.parse_args() return args def __lowerCamelCase ( __snake_case : Optional[int], __snake_case : Any="cpu" ) -> List[str]: """simple docstring""" A__ : Dict =model_dict[model_name].from_pretrained(__snake_case ).to(__snake_case ) A__ : Any =tokenizer_dict[model_name].from_pretrained(__snake_case ) if model_name in ["facebook/bart-base"]: A__ : List[Any] =0 A__ : Any =None A__ : Dict =0 return huggingface_model, tokenizer def __lowerCamelCase ( __snake_case : Optional[Any], __snake_case : Tuple, __snake_case : int, __snake_case : int, __snake_case : List[Any] ) -> List[Any]: """simple docstring""" model.eval() A__ : str =None A__ : List[Any] =torch.jit.script(BARTBeamSearchGenerator(__snake_case ) ) with torch.no_grad(): A__ : Tuple ="""My friends are cool but they eat too many carbs.""" A__ : Tuple =tokenizer([ARTICLE_TO_SUMMARIZE], max_length=1_024, return_tensors="""pt""" ).to(model.device ) A__ : Optional[Any] =model.generate( inputs["""input_ids"""], attention_mask=inputs["""attention_mask"""], num_beams=__snake_case, max_length=__snake_case, early_stopping=__snake_case, decoder_start_token_id=model.config.decoder_start_token_id, ) torch.onnx.export( __snake_case, ( inputs["""input_ids"""], inputs["""attention_mask"""], num_beams, max_length, model.config.decoder_start_token_id, ), __snake_case, opset_version=14, input_names=["""input_ids""", """attention_mask""", """num_beams""", """max_length""", """decoder_start_token_id"""], output_names=["""output_ids"""], dynamic_axes={ """input_ids""": {0: """batch""", 1: """seq"""}, """output_ids""": {0: """batch""", 1: """seq_out"""}, }, example_outputs=__snake_case, ) logger.info("""Model exported to {}""".format(__snake_case ) ) A__ : Union[str, Any] =remove_dup_initializers(os.path.abspath(__snake_case ) ) logger.info("""Deduplicated and optimized model written to {}""".format(__snake_case ) ) A__ : str =onnxruntime.InferenceSession(__snake_case ) A__ : List[Any] =ort_sess.run( __snake_case, { """input_ids""": inputs["""input_ids"""].cpu().numpy(), """attention_mask""": inputs["""attention_mask"""].cpu().numpy(), """num_beams""": np.array(__snake_case ), """max_length""": np.array(__snake_case ), """decoder_start_token_id""": np.array(model.config.decoder_start_token_id ), }, ) np.testing.assert_allclose(summary_ids.cpu().numpy(), ort_out[0], rtol=1E-3, atol=1E-3 ) logger.info("""Model outputs from torch and ONNX Runtime are similar.""" ) logger.info("""Success.""" ) def __lowerCamelCase ( ) -> Union[str, Any]: """simple docstring""" A__ : List[str] =parse_args() A__ : Optional[int] =5 A__ : Optional[int] =4 # Make one log on every process with the configuration for debugging. logging.basicConfig( format="""%(asctime)s - %(levelname)s - %(name)s - %(message)s""", datefmt="""%m/%d/%Y %H:%M:%S""", level=logging.INFO, ) logger.setLevel(logging.INFO ) transformers.utils.logging.set_verbosity_error() A__ : List[Any] =torch.device(args.device ) A__ , A__ : List[Any] =load_model_tokenizer(args.model_name_or_path, __snake_case ) if model.config.decoder_start_token_id is None: raise ValueError("""Make sure that `config.decoder_start_token_id` is correctly defined""" ) model.to(__snake_case ) if args.max_length: A__ : Union[str, Any] =args.max_length if args.num_beams: A__ : Tuple =args.num_beams if args.output_file_path: A__ : str =args.output_file_path else: A__ : Optional[Any] ="""BART.onnx""" logger.info("""Exporting model to ONNX""" ) export_and_validate_model(__snake_case, __snake_case, __snake_case, __snake_case, __snake_case ) if __name__ == "__main__": main()
134
'''simple docstring''' from __future__ import annotations from collections.abc import MutableSequence class lowerCamelCase : '''simple docstring''' def __init__( self : List[str] , lowerCAmelCase_ : int , lowerCAmelCase_ : MutableSequence[float] ) -> None: '''simple docstring''' if len(lowerCAmelCase_ ) != degree + 1: raise ValueError( """The number of coefficients should be equal to the degree + 1.""" ) A__ : list[float] =list(lowerCAmelCase_ ) A__ : Optional[int] =degree def __add__( self : Union[str, Any] , lowerCAmelCase_ : Polynomial ) -> Polynomial: '''simple docstring''' if self.degree > polynomial_a.degree: A__ : int =self.coefficients[:] for i in range(polynomial_a.degree + 1 ): coefficients[i] += polynomial_a.coefficients[i] return Polynomial(self.degree , lowerCAmelCase_ ) else: A__ : Any =polynomial_a.coefficients[:] for i in range(self.degree + 1 ): coefficients[i] += self.coefficients[i] return Polynomial(polynomial_a.degree , lowerCAmelCase_ ) def __sub__( self : str , lowerCAmelCase_ : Polynomial ) -> Polynomial: '''simple docstring''' return self + polynomial_a * Polynomial(0 , [-1] ) def __neg__( self : List[Any] ) -> Polynomial: '''simple docstring''' return Polynomial(self.degree , [-c for c in self.coefficients] ) def __mul__( self : str , lowerCAmelCase_ : Polynomial ) -> Polynomial: '''simple docstring''' A__ : list[float] =[0] * (self.degree + polynomial_a.degree + 1) for i in range(self.degree + 1 ): for j in range(polynomial_a.degree + 1 ): coefficients[i + j] += ( self.coefficients[i] * polynomial_a.coefficients[j] ) return Polynomial(self.degree + polynomial_a.degree , lowerCAmelCase_ ) def lowercase__ ( self : List[Any] , lowerCAmelCase_ : int | float ) -> int | float: '''simple docstring''' A__ : int | float =0 for i in range(self.degree + 1 ): result += self.coefficients[i] * (substitution**i) return result def __str__( self : List[str] ) -> str: '''simple docstring''' A__ : Optional[int] ="""""" for i in range(self.degree , -1 , -1 ): if self.coefficients[i] == 0: continue elif self.coefficients[i] > 0: if polynomial: polynomial += " + " else: polynomial += " - " if i == 0: polynomial += str(abs(self.coefficients[i] ) ) elif i == 1: polynomial += str(abs(self.coefficients[i] ) ) + "x" else: polynomial += str(abs(self.coefficients[i] ) ) + "x^" + str(lowerCAmelCase_ ) return polynomial def __repr__( self : Optional[Any] ) -> str: '''simple docstring''' return self.__str__() def lowercase__ ( self : str ) -> Polynomial: '''simple docstring''' A__ : list[float] =[0] * self.degree for i in range(self.degree ): A__ : Union[str, Any] =self.coefficients[i + 1] * (i + 1) return Polynomial(self.degree - 1 , lowerCAmelCase_ ) def lowercase__ ( self : Any , lowerCAmelCase_ : int | float = 0 ) -> Polynomial: '''simple docstring''' A__ : list[float] =[0] * (self.degree + 2) A__ : Any =constant for i in range(self.degree + 1 ): A__ : str =self.coefficients[i] / (i + 1) return Polynomial(self.degree + 1 , lowerCAmelCase_ ) def __eq__( self : Optional[int] , lowerCAmelCase_ : object ) -> bool: '''simple docstring''' if not isinstance(lowerCAmelCase_ , lowerCAmelCase_ ): return False if self.degree != polynomial_a.degree: return False for i in range(self.degree + 1 ): if self.coefficients[i] != polynomial_a.coefficients[i]: return False return True def __ne__( self : Optional[Any] , lowerCAmelCase_ : object ) -> bool: '''simple docstring''' return not self.__eq__(lowerCAmelCase_ )
134
1
'''simple docstring''' from ...configuration_utils import PretrainedConfig from ...utils import logging A__ : Any = logging.get_logger(__name__) A__ : Optional[int] = { '''edbeeching/decision-transformer-gym-hopper-medium''': ( '''https://huggingface.co/edbeeching/decision-transformer-gym-hopper-medium/resolve/main/config.json''' ), # See all DecisionTransformer models at https://huggingface.co/models?filter=decision_transformer } class snake_case__ ( SCREAMING_SNAKE_CASE_ ): A__ = '''decision_transformer''' A__ = ['''past_key_values'''] A__ = { '''max_position_embeddings''': '''n_positions''', '''num_attention_heads''': '''n_head''', '''num_hidden_layers''': '''n_layer''', } def __init__( self : str , __a : Union[str, Any]=17 , __a : Dict=4 , __a : str=128 , __a : Tuple=4096 , __a : str=True , __a : List[str]=1 , __a : Optional[Any]=1024 , __a : Any=3 , __a : List[str]=1 , __a : str=None , __a : Union[str, Any]="relu" , __a : Optional[Any]=0.1 , __a : str=0.1 , __a : List[str]=0.1 , __a : Any=1e-5 , __a : Dict=0.0_2 , __a : str=True , __a : str=True , __a : List[str]=50256 , __a : Any=50256 , __a : str=False , __a : List[str]=False , **__a : List[str] , ) -> Union[str, Any]: '''simple docstring''' __snake_case : Optional[Any] = state_dim __snake_case : Dict = act_dim __snake_case : Optional[int] = hidden_size __snake_case : int = max_ep_len __snake_case : Tuple = action_tanh __snake_case : str = vocab_size __snake_case : Tuple = n_positions __snake_case : Optional[Any] = n_layer __snake_case : int = n_head __snake_case : List[str] = n_inner __snake_case : List[Any] = activation_function __snake_case : Optional[Any] = resid_pdrop __snake_case : List[str] = embd_pdrop __snake_case : List[Any] = attn_pdrop __snake_case : Any = layer_norm_epsilon __snake_case : Union[str, Any] = initializer_range __snake_case : List[str] = scale_attn_weights __snake_case : str = use_cache __snake_case : List[Any] = scale_attn_by_inverse_layer_idx __snake_case : Optional[int] = reorder_and_upcast_attn __snake_case : Dict = bos_token_id __snake_case : Tuple = eos_token_id super().__init__(bos_token_id=__a , eos_token_id=__a , **__a )
364
'''simple docstring''' from typing import TYPE_CHECKING from ...utils import OptionalDependencyNotAvailable, _LazyModule, is_tf_available, is_torch_available A__ : int = { '''configuration_groupvit''': [ '''GROUPVIT_PRETRAINED_CONFIG_ARCHIVE_MAP''', '''GroupViTConfig''', '''GroupViTOnnxConfig''', '''GroupViTTextConfig''', '''GroupViTVisionConfig''', ], } try: if not is_torch_available(): raise OptionalDependencyNotAvailable() except OptionalDependencyNotAvailable: pass else: A__ : Tuple = [ '''GROUPVIT_PRETRAINED_MODEL_ARCHIVE_LIST''', '''GroupViTModel''', '''GroupViTPreTrainedModel''', '''GroupViTTextModel''', '''GroupViTVisionModel''', ] try: if not is_tf_available(): raise OptionalDependencyNotAvailable() except OptionalDependencyNotAvailable: pass else: A__ : Optional[int] = [ '''TF_GROUPVIT_PRETRAINED_MODEL_ARCHIVE_LIST''', '''TFGroupViTModel''', '''TFGroupViTPreTrainedModel''', '''TFGroupViTTextModel''', '''TFGroupViTVisionModel''', ] if TYPE_CHECKING: from .configuration_groupvit import ( GROUPVIT_PRETRAINED_CONFIG_ARCHIVE_MAP, GroupViTConfig, GroupViTOnnxConfig, GroupViTTextConfig, GroupViTVisionConfig, ) try: if not is_torch_available(): raise OptionalDependencyNotAvailable() except OptionalDependencyNotAvailable: pass else: from .modeling_groupvit import ( GROUPVIT_PRETRAINED_MODEL_ARCHIVE_LIST, GroupViTModel, GroupViTPreTrainedModel, GroupViTTextModel, GroupViTVisionModel, ) try: if not is_tf_available(): raise OptionalDependencyNotAvailable() except OptionalDependencyNotAvailable: pass else: from .modeling_tf_groupvit import ( TF_GROUPVIT_PRETRAINED_MODEL_ARCHIVE_LIST, TFGroupViTModel, TFGroupViTPreTrainedModel, TFGroupViTTextModel, TFGroupViTVisionModel, ) else: import sys A__ : List[str] = _LazyModule(__name__, globals()['''__file__'''], _import_structure, module_spec=__spec__)
0
0
def UpperCamelCase_( lowerCamelCase_ , lowerCamelCase_ ) -> str: if not isinstance(lowerCamelCase_ , lowerCamelCase_ ): raise ValueError('iterations must be defined as integers' ) if not isinstance(lowerCamelCase_ , lowerCamelCase_ ) or not number >= 1: raise ValueError( 'starting number must be\n and integer and be more than 0' ) if not iterations >= 1: raise ValueError('Iterations must be done more than 0 times to play FizzBuzz' ) _lowercase : Tuple = '' while number <= iterations: if number % 3 == 0: out += "Fizz" if number % 5 == 0: out += "Buzz" if 0 not in (number % 3, number % 5): out += str(lowerCamelCase_ ) # print(out) number += 1 out += " " return out if __name__ == "__main__": import doctest doctest.testmod()
21
import numpy # List of input, output pairs UpperCAmelCase : str = ( ((5, 2, 3), 15), ((6, 5, 9), 25), ((11, 12, 13), 41), ((1, 1, 1), 8), ((11, 12, 13), 41), ) UpperCAmelCase : Optional[int] = (((515, 22, 13), 555), ((61, 35, 49), 150)) UpperCAmelCase : str = [2, 4, 1, 5] UpperCAmelCase : List[str] = len(train_data) UpperCAmelCase : Dict = 0.0_0_9 def _A ( SCREAMING_SNAKE_CASE : Tuple , SCREAMING_SNAKE_CASE : Tuple="train" ): """simple docstring""" return calculate_hypothesis_value(SCREAMING_SNAKE_CASE , SCREAMING_SNAKE_CASE ) - output( SCREAMING_SNAKE_CASE , SCREAMING_SNAKE_CASE ) def _A ( SCREAMING_SNAKE_CASE : Tuple ): """simple docstring""" a__ : Tuple =0 for i in range(len(SCREAMING_SNAKE_CASE ) - 1 ): hyp_val += data_input_tuple[i] * parameter_vector[i + 1] hyp_val += parameter_vector[0] return hyp_val def _A ( SCREAMING_SNAKE_CASE : Dict , SCREAMING_SNAKE_CASE : Dict ): """simple docstring""" if data_set == "train": return train_data[example_no][1] elif data_set == "test": return test_data[example_no][1] return None def _A ( SCREAMING_SNAKE_CASE : Optional[Any] , SCREAMING_SNAKE_CASE : Tuple ): """simple docstring""" if data_set == "train": return _hypothesis_value(train_data[example_no][0] ) elif data_set == "test": return _hypothesis_value(test_data[example_no][0] ) return None def _A ( SCREAMING_SNAKE_CASE : Union[str, Any] , SCREAMING_SNAKE_CASE : int=m ): """simple docstring""" a__ : Any =0 for i in range(SCREAMING_SNAKE_CASE ): if index == -1: summation_value += _error(SCREAMING_SNAKE_CASE ) else: summation_value += _error(SCREAMING_SNAKE_CASE ) * train_data[i][0][index] return summation_value def _A ( SCREAMING_SNAKE_CASE : int ): """simple docstring""" a__ : Any =summation_of_cost_derivative(SCREAMING_SNAKE_CASE , SCREAMING_SNAKE_CASE ) / m return cost_derivative_value def _A ( ): """simple docstring""" global parameter_vector # Tune these values to set a tolerance value for predicted output a__ : Dict =0.0_0_0_0_0_2 a__ : Union[str, Any] =0 a__ : Any =0 while True: j += 1 a__ : Any =[0, 0, 0, 0] for i in range(0 , len(SCREAMING_SNAKE_CASE ) ): a__ : Tuple =get_cost_derivative(i - 1 ) a__ : List[Any] =( parameter_vector[i] - LEARNING_RATE * cost_derivative ) if numpy.allclose( SCREAMING_SNAKE_CASE , SCREAMING_SNAKE_CASE , atol=SCREAMING_SNAKE_CASE , rtol=SCREAMING_SNAKE_CASE , ): break a__ : Optional[Any] =temp_parameter_vector print(("Number of iterations:", j) ) def _A ( ): """simple docstring""" for i in range(len(SCREAMING_SNAKE_CASE ) ): print(("Actual output value:", output(SCREAMING_SNAKE_CASE , "test" )) ) print(("Hypothesis output:", calculate_hypothesis_value(SCREAMING_SNAKE_CASE , "test" )) ) if __name__ == "__main__": run_gradient_descent() print("""\nTesting gradient descent for a linear hypothesis function.\n""") test_gradient_descent()
95
0
"""simple docstring""" import math import time from transformers import Trainer, is_torch_tpu_available from transformers.trainer_utils import PredictionOutput, speed_metrics if is_torch_tpu_available(check_device=False): import torch_xla.core.xla_model as xm import torch_xla.debug.metrics as met class _SCREAMING_SNAKE_CASE ( lowercase__ ): def __init__( self , *__A , __A=None , __A=None , **__A ) -> Dict: super().__init__(*_a , **_a ) lowerCAmelCase_ :int = eval_examples lowerCAmelCase_ :Optional[Any] = post_process_function def __lowerCAmelCase ( self , __A=None , __A=None , __A=None , __A = "eval" ) -> Any: lowerCAmelCase_ :Dict = self.eval_dataset if eval_dataset is None else eval_dataset lowerCAmelCase_ :Union[str, Any] = self.get_eval_dataloader(_a ) lowerCAmelCase_ :Tuple = self.eval_examples if eval_examples is None else eval_examples # Temporarily disable metric computation, we will do it in the loop here. lowerCAmelCase_ :int = self.compute_metrics lowerCAmelCase_ :List[str] = None lowerCAmelCase_ :Tuple = self.prediction_loop if self.args.use_legacy_prediction_loop else self.evaluation_loop lowerCAmelCase_ :List[Any] = time.time() try: lowerCAmelCase_ :Optional[Any] = eval_loop( _a , description="""Evaluation""" , prediction_loss_only=True if compute_metrics is None else None , ignore_keys=_a , metric_key_prefix=_a , ) finally: lowerCAmelCase_ :List[Any] = compute_metrics lowerCAmelCase_ :Optional[Any] = self.args.eval_batch_size * self.args.world_size if f"""{metric_key_prefix}_jit_compilation_time""" in output.metrics: start_time += output.metrics[f"""{metric_key_prefix}_jit_compilation_time"""] output.metrics.update( speed_metrics( _a , _a , num_samples=output.num_samples , num_steps=math.ceil(output.num_samples / total_batch_size ) , ) ) if self.post_process_function is not None and self.compute_metrics is not None and self.args.should_save: # Only the main node write the results by default lowerCAmelCase_ :List[Any] = self.post_process_function(_a , _a , output.predictions ) lowerCAmelCase_ :str = self.compute_metrics(_a ) # Prefix all keys with metric_key_prefix + '_' for key in list(metrics.keys() ): if not key.startswith(f"""{metric_key_prefix}_""" ): lowerCAmelCase_ :Optional[int] = metrics.pop(_a ) metrics.update(output.metrics ) else: lowerCAmelCase_ :Dict = output.metrics if self.args.should_log: # Only the main node log the results by default self.log(_a ) if self.args.tpu_metrics_debug or self.args.debug: # tpu-comment: Logging debug metrics for PyTorch/XLA (compile, execute times, ops, etc.) xm.master_print(met.metrics_report() ) lowerCAmelCase_ :Optional[int] = self.callback_handler.on_evaluate(self.args , self.state , self.control , _a ) return metrics def __lowerCAmelCase ( self , __A , __A , __A=None , __A = "test" ) -> List[Any]: lowerCAmelCase_ :Optional[int] = self.get_test_dataloader(_a ) # Temporarily disable metric computation, we will do it in the loop here. lowerCAmelCase_ :int = self.compute_metrics lowerCAmelCase_ :Tuple = None lowerCAmelCase_ :Dict = self.prediction_loop if self.args.use_legacy_prediction_loop else self.evaluation_loop lowerCAmelCase_ :List[Any] = time.time() try: lowerCAmelCase_ :str = eval_loop( _a , description="""Prediction""" , prediction_loss_only=True if compute_metrics is None else None , ignore_keys=_a , metric_key_prefix=_a , ) finally: lowerCAmelCase_ :Optional[int] = compute_metrics lowerCAmelCase_ :Any = self.args.eval_batch_size * self.args.world_size if f"""{metric_key_prefix}_jit_compilation_time""" in output.metrics: start_time += output.metrics[f"""{metric_key_prefix}_jit_compilation_time"""] output.metrics.update( speed_metrics( _a , _a , num_samples=output.num_samples , num_steps=math.ceil(output.num_samples / total_batch_size ) , ) ) if self.post_process_function is None or self.compute_metrics is None: return output lowerCAmelCase_ :Dict = self.post_process_function(_a , _a , output.predictions , """predict""" ) lowerCAmelCase_ :Optional[Any] = self.compute_metrics(_a ) # Prefix all keys with metric_key_prefix + '_' for key in list(metrics.keys() ): if not key.startswith(f"""{metric_key_prefix}_""" ): lowerCAmelCase_ :Tuple = metrics.pop(_a ) metrics.update(output.metrics ) return PredictionOutput(predictions=predictions.predictions , label_ids=predictions.label_ids , metrics=_a )
354
"""simple docstring""" from .imports import is_tqdm_available if is_tqdm_available(): from tqdm.auto import tqdm as _tqdm from ..state import PartialState def _snake_case ( lowercase__ : bool = True , *lowercase__ : Optional[int] , **lowercase__ : str ) -> Optional[Any]: '''simple docstring''' if not is_tqdm_available(): raise ImportError("""Accelerate's `tqdm` module requires `tqdm` to be installed. Please run `pip install tqdm`.""" ) lowerCAmelCase_ :Tuple = False if main_process_only: lowerCAmelCase_ :Dict = PartialState().local_process_index == 0 return _tqdm(*lowercase__ , **lowercase__ , disable=lowercase__ )
1
0
"""simple docstring""" # Copyright 2021 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. from argparse import ArgumentParser from accelerate.commands.config import get_config_parser from accelerate.commands.env import env_command_parser from accelerate.commands.launch import launch_command_parser from accelerate.commands.test import test_command_parser from accelerate.commands.tpu import tpu_command_parser def UpperCAmelCase__ (): '''simple docstring''' __SCREAMING_SNAKE_CASE = ArgumentParser("Accelerate CLI tool" , usage="accelerate <command> [<args>]" , allow_abbrev=lowerCAmelCase_ ) __SCREAMING_SNAKE_CASE = parser.add_subparsers(help="accelerate command helpers" ) # Register commands get_config_parser(subparsers=lowerCAmelCase_ ) env_command_parser(subparsers=lowerCAmelCase_ ) launch_command_parser(subparsers=lowerCAmelCase_ ) tpu_command_parser(subparsers=lowerCAmelCase_ ) test_command_parser(subparsers=lowerCAmelCase_ ) # Let's go __SCREAMING_SNAKE_CASE = parser.parse_args() if not hasattr(lowerCAmelCase_ , "func" ): parser.print_help() exit(1 ) # Run args.func(lowerCAmelCase_ ) if __name__ == "__main__": main()
54
from ..utils import DummyObject, requires_backends class _snake_case ( metaclass=__snake_case ): '''simple docstring''' A__ : Tuple = ["flax"] def __init__( self: str ,*lowerCamelCase_: int ,**lowerCamelCase_: List[str] ) -> str: requires_backends(self ,["""flax"""] ) @classmethod def A__ ( cls: Optional[Any] ,*lowerCamelCase_: Dict ,**lowerCamelCase_: List[str] ) -> Any: requires_backends(cls ,["""flax"""] ) @classmethod def A__ ( cls: Optional[int] ,*lowerCamelCase_: Optional[int] ,**lowerCamelCase_: int ) -> Optional[int]: requires_backends(cls ,["""flax"""] ) class _snake_case ( metaclass=__snake_case ): '''simple docstring''' A__ : Any = ["flax"] def __init__( self: int ,*lowerCamelCase_: List[Any] ,**lowerCamelCase_: Tuple ) -> Union[str, Any]: requires_backends(self ,["""flax"""] ) @classmethod def A__ ( cls: Optional[int] ,*lowerCamelCase_: Optional[int] ,**lowerCamelCase_: List[str] ) -> Union[str, Any]: requires_backends(cls ,["""flax"""] ) @classmethod def A__ ( cls: Tuple ,*lowerCamelCase_: Tuple ,**lowerCamelCase_: Any ) -> int: requires_backends(cls ,["""flax"""] ) class _snake_case ( metaclass=__snake_case ): '''simple docstring''' A__ : Dict = ["flax"] def __init__( self: Dict ,*lowerCamelCase_: Optional[int] ,**lowerCamelCase_: List[Any] ) -> Any: requires_backends(self ,["""flax"""] ) @classmethod def A__ ( cls: Tuple ,*lowerCamelCase_: Optional[Any] ,**lowerCamelCase_: List[Any] ) -> str: requires_backends(cls ,["""flax"""] ) @classmethod def A__ ( cls: int ,*lowerCamelCase_: Optional[Any] ,**lowerCamelCase_: Optional[Any] ) -> int: requires_backends(cls ,["""flax"""] ) class _snake_case ( metaclass=__snake_case ): '''simple docstring''' A__ : List[str] = ["flax"] def __init__( self: str ,*lowerCamelCase_: List[str] ,**lowerCamelCase_: Optional[int] ) -> Union[str, Any]: requires_backends(self ,["""flax"""] ) @classmethod def A__ ( cls: Union[str, Any] ,*lowerCamelCase_: Any ,**lowerCamelCase_: Any ) -> Any: requires_backends(cls ,["""flax"""] ) @classmethod def A__ ( cls: Dict ,*lowerCamelCase_: int ,**lowerCamelCase_: Optional[Any] ) -> int: requires_backends(cls ,["""flax"""] ) class _snake_case ( metaclass=__snake_case ): '''simple docstring''' A__ : int = ["flax"] def __init__( self: Dict ,*lowerCamelCase_: Tuple ,**lowerCamelCase_: List[str] ) -> Optional[Any]: requires_backends(self ,["""flax"""] ) @classmethod def A__ ( cls: Optional[Any] ,*lowerCamelCase_: List[Any] ,**lowerCamelCase_: str ) -> Any: requires_backends(cls ,["""flax"""] ) @classmethod def A__ ( cls: Union[str, Any] ,*lowerCamelCase_: Dict ,**lowerCamelCase_: Optional[Any] ) -> str: requires_backends(cls ,["""flax"""] ) class _snake_case ( metaclass=__snake_case ): '''simple docstring''' A__ : Optional[int] = ["flax"] def __init__( self: str ,*lowerCamelCase_: Dict ,**lowerCamelCase_: Optional[int] ) -> Tuple: requires_backends(self ,["""flax"""] ) @classmethod def A__ ( cls: int ,*lowerCamelCase_: int ,**lowerCamelCase_: Tuple ) -> List[str]: requires_backends(cls ,["""flax"""] ) @classmethod def A__ ( cls: str ,*lowerCamelCase_: Union[str, Any] ,**lowerCamelCase_: Optional[Any] ) -> Any: requires_backends(cls ,["""flax"""] ) class _snake_case ( metaclass=__snake_case ): '''simple docstring''' A__ : List[Any] = ["flax"] def __init__( self: Union[str, Any] ,*lowerCamelCase_: Tuple ,**lowerCamelCase_: int ) -> List[Any]: requires_backends(self ,["""flax"""] ) @classmethod def A__ ( cls: Tuple ,*lowerCamelCase_: List[Any] ,**lowerCamelCase_: Dict ) -> Dict: requires_backends(cls ,["""flax"""] ) @classmethod def A__ ( cls: Dict ,*lowerCamelCase_: List[Any] ,**lowerCamelCase_: str ) -> Any: requires_backends(cls ,["""flax"""] ) class _snake_case ( metaclass=__snake_case ): '''simple docstring''' A__ : Tuple = ["flax"] def __init__( self: str ,*lowerCamelCase_: Any ,**lowerCamelCase_: int ) -> Tuple: requires_backends(self ,["""flax"""] ) @classmethod def A__ ( cls: Dict ,*lowerCamelCase_: Optional[int] ,**lowerCamelCase_: Union[str, Any] ) -> List[str]: requires_backends(cls ,["""flax"""] ) @classmethod def A__ ( cls: str ,*lowerCamelCase_: Union[str, Any] ,**lowerCamelCase_: Dict ) -> Optional[int]: requires_backends(cls ,["""flax"""] ) class _snake_case ( metaclass=__snake_case ): '''simple docstring''' A__ : str = ["flax"] def __init__( self: Optional[Any] ,*lowerCamelCase_: str ,**lowerCamelCase_: List[str] ) -> Optional[Any]: requires_backends(self ,["""flax"""] ) @classmethod def A__ ( cls: List[str] ,*lowerCamelCase_: Dict ,**lowerCamelCase_: int ) -> List[str]: requires_backends(cls ,["""flax"""] ) @classmethod def A__ ( cls: str ,*lowerCamelCase_: Optional[Any] ,**lowerCamelCase_: int ) -> Union[str, Any]: requires_backends(cls ,["""flax"""] ) class _snake_case ( metaclass=__snake_case ): '''simple docstring''' A__ : Union[str, Any] = ["flax"] def __init__( self: Any ,*lowerCamelCase_: Tuple ,**lowerCamelCase_: Optional[int] ) -> List[str]: requires_backends(self ,["""flax"""] ) @classmethod def A__ ( cls: Optional[int] ,*lowerCamelCase_: List[Any] ,**lowerCamelCase_: str ) -> Union[str, Any]: requires_backends(cls ,["""flax"""] ) @classmethod def A__ ( cls: List[Any] ,*lowerCamelCase_: Any ,**lowerCamelCase_: Any ) -> int: requires_backends(cls ,["""flax"""] ) class _snake_case ( metaclass=__snake_case ): '''simple docstring''' A__ : Tuple = ["flax"] def __init__( self: Any ,*lowerCamelCase_: Optional[Any] ,**lowerCamelCase_: Dict ) -> str: requires_backends(self ,["""flax"""] ) @classmethod def A__ ( cls: Tuple ,*lowerCamelCase_: Union[str, Any] ,**lowerCamelCase_: List[str] ) -> int: requires_backends(cls ,["""flax"""] ) @classmethod def A__ ( cls: List[Any] ,*lowerCamelCase_: str ,**lowerCamelCase_: str ) -> Any: requires_backends(cls ,["""flax"""] ) class _snake_case ( metaclass=__snake_case ): '''simple docstring''' A__ : Optional[Any] = ["flax"] def __init__( self: Dict ,*lowerCamelCase_: int ,**lowerCamelCase_: Optional[Any] ) -> Union[str, Any]: requires_backends(self ,["""flax"""] ) @classmethod def A__ ( cls: int ,*lowerCamelCase_: int ,**lowerCamelCase_: Tuple ) -> Union[str, Any]: requires_backends(cls ,["""flax"""] ) @classmethod def A__ ( cls: Optional[Any] ,*lowerCamelCase_: List[Any] ,**lowerCamelCase_: Optional[int] ) -> int: requires_backends(cls ,["""flax"""] ) class _snake_case ( metaclass=__snake_case ): '''simple docstring''' A__ : Optional[int] = ["flax"] def __init__( self: List[str] ,*lowerCamelCase_: Dict ,**lowerCamelCase_: Dict ) -> int: requires_backends(self ,["""flax"""] ) @classmethod def A__ ( cls: Dict ,*lowerCamelCase_: List[Any] ,**lowerCamelCase_: Dict ) -> Union[str, Any]: requires_backends(cls ,["""flax"""] ) @classmethod def A__ ( cls: int ,*lowerCamelCase_: Any ,**lowerCamelCase_: Any ) -> Optional[Any]: requires_backends(cls ,["""flax"""] )
345
0
from __future__ import annotations import pandas as pd def A__ ( __lowerCamelCase, __lowerCamelCase, __lowerCamelCase ): SCREAMING_SNAKE_CASE_ = [0] * no_of_processes SCREAMING_SNAKE_CASE_ = [0] * no_of_processes # Copy the burst time into remaining_time[] for i in range(_lowercase ): SCREAMING_SNAKE_CASE_ = burst_time[i] SCREAMING_SNAKE_CASE_ = 0 SCREAMING_SNAKE_CASE_ = 0 SCREAMING_SNAKE_CASE_ = 9_99_99_99_99 SCREAMING_SNAKE_CASE_ = 0 SCREAMING_SNAKE_CASE_ = False # Process until all processes are completed while complete != no_of_processes: for j in range(_lowercase ): if arrival_time[j] <= increment_time and remaining_time[j] > 0: if remaining_time[j] < minm: SCREAMING_SNAKE_CASE_ = remaining_time[j] SCREAMING_SNAKE_CASE_ = j SCREAMING_SNAKE_CASE_ = True if not check: increment_time += 1 continue remaining_time[short] -= 1 SCREAMING_SNAKE_CASE_ = remaining_time[short] if minm == 0: SCREAMING_SNAKE_CASE_ = 9_99_99_99_99 if remaining_time[short] == 0: complete += 1 SCREAMING_SNAKE_CASE_ = False # Find finish time of current process SCREAMING_SNAKE_CASE_ = increment_time + 1 # Calculate waiting time SCREAMING_SNAKE_CASE_ = finish_time - arrival_time[short] SCREAMING_SNAKE_CASE_ = finar - burst_time[short] if waiting_time[short] < 0: SCREAMING_SNAKE_CASE_ = 0 # Increment time increment_time += 1 return waiting_time def A__ ( __lowerCamelCase, __lowerCamelCase, __lowerCamelCase ): SCREAMING_SNAKE_CASE_ = [0] * no_of_processes for i in range(_lowercase ): SCREAMING_SNAKE_CASE_ = burst_time[i] + waiting_time[i] return turn_around_time def A__ ( __lowerCamelCase, __lowerCamelCase, __lowerCamelCase ): SCREAMING_SNAKE_CASE_ = 0 SCREAMING_SNAKE_CASE_ = 0 for i in range(_lowercase ): SCREAMING_SNAKE_CASE_ = total_waiting_time + waiting_time[i] SCREAMING_SNAKE_CASE_ = total_turn_around_time + turn_around_time[i] print(F'''Average waiting time = {total_waiting_time / no_of_processes:.5f}''' ) print('''Average turn around time =''', total_turn_around_time / no_of_processes ) if __name__ == "__main__": print("Enter how many process you want to analyze") __UpperCAmelCase = int(input()) __UpperCAmelCase = [0] * no_of_processes __UpperCAmelCase = [0] * no_of_processes __UpperCAmelCase = list(range(1, no_of_processes + 1)) for i in range(no_of_processes): print("Enter the arrival time and burst time for process:--" + str(i + 1)) __UpperCAmelCase = map(int, input().split()) __UpperCAmelCase = calculate_waitingtime(arrival_time, burst_time, no_of_processes) __UpperCAmelCase = burst_time __UpperCAmelCase = no_of_processes __UpperCAmelCase = waiting_time __UpperCAmelCase = calculate_turnaroundtime(bt, n, wt) calculate_average_times(waiting_time, turn_around_time, no_of_processes) __UpperCAmelCase = pd.DataFrame( list(zip(processes, burst_time, arrival_time, waiting_time, turn_around_time)), columns=[ "Process", "BurstTime", "ArrivalTime", "WaitingTime", "TurnAroundTime", ], ) # Printing the dataFrame pd.set_option("display.max_rows", fcfs.shape[0] + 1) print(fcfs)
363
import argparse import json from pathlib import Path import requests import torch from huggingface_hub import hf_hub_download from PIL import Image from timm import create_model from timm.data import resolve_data_config from timm.data.transforms_factory import create_transform from transformers import BitConfig, BitForImageClassification, BitImageProcessor from transformers.image_utils import PILImageResampling from transformers.utils import logging logging.set_verbosity_info() __UpperCAmelCase = logging.get_logger(__name__) def A__ ( __lowerCamelCase ): SCREAMING_SNAKE_CASE_ = '''huggingface/label-files''' SCREAMING_SNAKE_CASE_ = '''imagenet-1k-id2label.json''' SCREAMING_SNAKE_CASE_ = json.load(open(hf_hub_download(__lowerCamelCase, __lowerCamelCase, repo_type='''dataset''' ), '''r''' ) ) SCREAMING_SNAKE_CASE_ = {int(__lowerCamelCase ): v for k, v in idalabel.items()} SCREAMING_SNAKE_CASE_ = {v: k for k, v in idalabel.items()} SCREAMING_SNAKE_CASE_ = '''std_conv''' if '''bit''' in model_name else False # note that when using BiT as backbone for ViT-hybrid checkpoints, # one needs to additionally set config.layer_type = "bottleneck", config.stem_type = "same", # config.conv_layer = "std_conv_same" SCREAMING_SNAKE_CASE_ = BitConfig( conv_layer=__lowerCamelCase, num_labels=10_00, idalabel=__lowerCamelCase, labelaid=__lowerCamelCase, ) return config def A__ ( __lowerCamelCase ): if "stem.conv" in name: SCREAMING_SNAKE_CASE_ = name.replace('''stem.conv''', '''bit.embedder.convolution''' ) if "blocks" in name: SCREAMING_SNAKE_CASE_ = name.replace('''blocks''', '''layers''' ) if "head.fc" in name: SCREAMING_SNAKE_CASE_ = name.replace('''head.fc''', '''classifier.1''' ) if name.startswith('''norm''' ): SCREAMING_SNAKE_CASE_ = '''bit.''' + name if "bit" not in name and "classifier" not in name: SCREAMING_SNAKE_CASE_ = '''bit.encoder.''' + name return name def A__ ( ): SCREAMING_SNAKE_CASE_ = '''http://images.cocodataset.org/val2017/000000039769.jpg''' SCREAMING_SNAKE_CASE_ = Image.open(requests.get(__lowerCamelCase, stream=__lowerCamelCase ).raw ) return im @torch.no_grad() def A__ ( __lowerCamelCase, __lowerCamelCase, __lowerCamelCase=False ): SCREAMING_SNAKE_CASE_ = get_config(__lowerCamelCase ) # load original model from timm SCREAMING_SNAKE_CASE_ = create_model(__lowerCamelCase, pretrained=__lowerCamelCase ) timm_model.eval() # load state_dict of original model SCREAMING_SNAKE_CASE_ = timm_model.state_dict() for key in state_dict.copy().keys(): SCREAMING_SNAKE_CASE_ = state_dict.pop(__lowerCamelCase ) SCREAMING_SNAKE_CASE_ = val.squeeze() if '''head''' in key else val # load HuggingFace model SCREAMING_SNAKE_CASE_ = BitForImageClassification(__lowerCamelCase ) model.eval() model.load_state_dict(__lowerCamelCase ) # create image processor SCREAMING_SNAKE_CASE_ = create_transform(**resolve_data_config({}, model=__lowerCamelCase ) ) SCREAMING_SNAKE_CASE_ = transform.transforms SCREAMING_SNAKE_CASE_ = { '''bilinear''': PILImageResampling.BILINEAR, '''bicubic''': PILImageResampling.BICUBIC, '''nearest''': PILImageResampling.NEAREST, } SCREAMING_SNAKE_CASE_ = BitImageProcessor( do_resize=__lowerCamelCase, size={'''shortest_edge''': timm_transforms[0].size}, resample=pillow_resamplings[timm_transforms[0].interpolation.value], do_center_crop=__lowerCamelCase, crop_size={'''height''': timm_transforms[1].size[0], '''width''': timm_transforms[1].size[1]}, do_normalize=__lowerCamelCase, image_mean=timm_transforms[-1].mean.tolist(), image_std=timm_transforms[-1].std.tolist(), ) SCREAMING_SNAKE_CASE_ = prepare_img() SCREAMING_SNAKE_CASE_ = transform(__lowerCamelCase ).unsqueeze(0 ) SCREAMING_SNAKE_CASE_ = processor(__lowerCamelCase, return_tensors='''pt''' ).pixel_values # verify pixel values assert torch.allclose(__lowerCamelCase, __lowerCamelCase ) # verify logits with torch.no_grad(): SCREAMING_SNAKE_CASE_ = model(__lowerCamelCase ) SCREAMING_SNAKE_CASE_ = outputs.logits print('''Logits:''', logits[0, :3] ) print('''Predicted class:''', model.config.idalabel[logits.argmax(-1 ).item()] ) SCREAMING_SNAKE_CASE_ = timm_model(__lowerCamelCase ) assert timm_logits.shape == outputs.logits.shape assert torch.allclose(__lowerCamelCase, outputs.logits, atol=1E-3 ) print('''Looks ok!''' ) if pytorch_dump_folder_path is not None: Path(__lowerCamelCase ).mkdir(exist_ok=__lowerCamelCase ) print(F'''Saving model {model_name} and processor to {pytorch_dump_folder_path}''' ) model.save_pretrained(__lowerCamelCase ) processor.save_pretrained(__lowerCamelCase ) if push_to_hub: print(F'''Pushing model {model_name} and processor to the hub''' ) model.push_to_hub(F'''ybelkada/{model_name}''' ) processor.push_to_hub(F'''ybelkada/{model_name}''' ) if __name__ == "__main__": __UpperCAmelCase = argparse.ArgumentParser() # Required parameters parser.add_argument( "--model_name", default="resnetv2_50x1_bitm", type=str, help="Name of the BiT 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 push the model to the hub.", ) __UpperCAmelCase = parser.parse_args() convert_bit_checkpoint(args.model_name, args.pytorch_dump_folder_path, args.push_to_hub)
257
0
from typing import Dict, List, Optional, Union import numpy as np from transformers.utils import is_vision_available from transformers.utils.generic import TensorType from ...image_processing_utils import BaseImageProcessor, BatchFeature, get_size_dict from ...image_transforms import ( center_crop, get_resize_output_image_size, normalize, rescale, resize, to_channel_dimension_format, ) from ...image_utils import ( IMAGENET_STANDARD_MEAN, IMAGENET_STANDARD_STD, ChannelDimension, ImageInput, PILImageResampling, is_valid_image, to_numpy_array, valid_images, ) from ...utils import logging if is_vision_available(): import PIL _snake_case : int = logging.get_logger(__name__) def a_ ( lowerCAmelCase_ : List[str] ): if isinstance(lowerCAmelCase_, (list, tuple) ) and isinstance(videos[0], (list, tuple) ) and is_valid_image(videos[0][0] ): return videos elif isinstance(lowerCAmelCase_, (list, tuple) ) and is_valid_image(videos[0] ): return [videos] elif is_valid_image(lowerCAmelCase_ ): return [[videos]] raise ValueError(F"""Could not make batched video from {videos}""" ) class _UpperCAmelCase ( _UpperCamelCase ): """simple docstring""" a_ = ["""pixel_values"""] def __init__( self : int , lowerCAmelCase_ : bool = True , lowerCAmelCase_ : Dict[str, int] = None , lowerCAmelCase_ : PILImageResampling = PILImageResampling.BILINEAR , lowerCAmelCase_ : bool = True , lowerCAmelCase_ : Dict[str, int] = None , lowerCAmelCase_ : bool = True , lowerCAmelCase_ : Union[int, float] = 1 / 2_5_5 , lowerCAmelCase_ : bool = True , lowerCAmelCase_ : bool = True , lowerCAmelCase_ : Optional[Union[float, List[float]]] = None , lowerCAmelCase_ : Optional[Union[float, List[float]]] = None , **lowerCAmelCase_ : int , ) -> None: super().__init__(**lowerCAmelCase_ ) __lowerCAmelCase = size if size is not None else {'shortest_edge': 2_5_6} __lowerCAmelCase = get_size_dict(lowerCAmelCase_ , default_to_square=lowerCAmelCase_ ) __lowerCAmelCase = crop_size if crop_size is not None else {'height': 2_2_4, 'width': 2_2_4} __lowerCAmelCase = get_size_dict(lowerCAmelCase_ , param_name='crop_size' ) __lowerCAmelCase = do_resize __lowerCAmelCase = size __lowerCAmelCase = do_center_crop __lowerCAmelCase = crop_size __lowerCAmelCase = resample __lowerCAmelCase = do_rescale __lowerCAmelCase = rescale_factor __lowerCAmelCase = offset __lowerCAmelCase = do_normalize __lowerCAmelCase = image_mean if image_mean is not None else IMAGENET_STANDARD_MEAN __lowerCAmelCase = image_std if image_std is not None else IMAGENET_STANDARD_STD def lowercase ( self : str , lowerCAmelCase_ : np.ndarray , lowerCAmelCase_ : Dict[str, int] , lowerCAmelCase_ : PILImageResampling = PILImageResampling.BILINEAR , lowerCAmelCase_ : Optional[Union[str, ChannelDimension]] = None , **lowerCAmelCase_ : List[str] , ) -> np.ndarray: __lowerCAmelCase = get_size_dict(lowerCAmelCase_ , default_to_square=lowerCAmelCase_ ) if "shortest_edge" in size: __lowerCAmelCase = get_resize_output_image_size(lowerCAmelCase_ , size['shortest_edge'] , default_to_square=lowerCAmelCase_ ) elif "height" in size and "width" in size: __lowerCAmelCase = (size['height'], size['width']) else: raise ValueError(f"""Size must have 'height' and 'width' or 'shortest_edge' as keys. Got {size.keys()}""" ) return resize(lowerCAmelCase_ , size=lowerCAmelCase_ , resample=lowerCAmelCase_ , data_format=lowerCAmelCase_ , **lowerCAmelCase_ ) def lowercase ( self : Union[str, Any] , lowerCAmelCase_ : np.ndarray , lowerCAmelCase_ : Dict[str, int] , lowerCAmelCase_ : Optional[Union[str, ChannelDimension]] = None , **lowerCAmelCase_ : Dict , ) -> np.ndarray: __lowerCAmelCase = get_size_dict(lowerCAmelCase_ ) if "height" not in size or "width" not in size: raise ValueError(f"""Size must have 'height' and 'width' as keys. Got {size.keys()}""" ) return center_crop(lowerCAmelCase_ , size=(size['height'], size['width']) , data_format=lowerCAmelCase_ , **lowerCAmelCase_ ) def lowercase ( self : Optional[Any] , lowerCAmelCase_ : np.ndarray , lowerCAmelCase_ : Union[int, float] , lowerCAmelCase_ : bool = True , lowerCAmelCase_ : Optional[Union[str, ChannelDimension]] = None , **lowerCAmelCase_ : Tuple , ) -> Dict: __lowerCAmelCase = image.astype(np.floataa ) if offset: __lowerCAmelCase = image - (scale / 2) return rescale(lowerCAmelCase_ , scale=lowerCAmelCase_ , data_format=lowerCAmelCase_ , **lowerCAmelCase_ ) def lowercase ( self : Any , lowerCAmelCase_ : np.ndarray , lowerCAmelCase_ : Union[float, List[float]] , lowerCAmelCase_ : Union[float, List[float]] , lowerCAmelCase_ : Optional[Union[str, ChannelDimension]] = None , **lowerCAmelCase_ : List[Any] , ) -> np.ndarray: return normalize(lowerCAmelCase_ , mean=lowerCAmelCase_ , std=lowerCAmelCase_ , data_format=lowerCAmelCase_ , **lowerCAmelCase_ ) def lowercase ( self : int , lowerCAmelCase_ : ImageInput , lowerCAmelCase_ : bool = None , lowerCAmelCase_ : Dict[str, int] = None , lowerCAmelCase_ : PILImageResampling = None , lowerCAmelCase_ : bool = None , lowerCAmelCase_ : Dict[str, int] = None , lowerCAmelCase_ : bool = None , lowerCAmelCase_ : float = None , lowerCAmelCase_ : bool = None , lowerCAmelCase_ : bool = None , lowerCAmelCase_ : Optional[Union[float, List[float]]] = None , lowerCAmelCase_ : Optional[Union[float, List[float]]] = None , lowerCAmelCase_ : Optional[ChannelDimension] = ChannelDimension.FIRST , ) -> np.ndarray: if do_resize and size is None or resample is None: raise ValueError('Size and resample must be specified if do_resize is True.' ) if do_center_crop and crop_size is None: raise ValueError('Crop size must be specified if do_center_crop 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('Image mean and std must be specified if do_normalize is True.' ) if offset and not do_rescale: raise ValueError('For offset, do_rescale must also be set to True.' ) # All transformations expect numpy arrays. __lowerCAmelCase = to_numpy_array(lowerCAmelCase_ ) if do_resize: __lowerCAmelCase = self.resize(image=lowerCAmelCase_ , size=lowerCAmelCase_ , resample=lowerCAmelCase_ ) if do_center_crop: __lowerCAmelCase = self.center_crop(lowerCAmelCase_ , size=lowerCAmelCase_ ) if do_rescale: __lowerCAmelCase = self.rescale(image=lowerCAmelCase_ , scale=lowerCAmelCase_ , offset=lowerCAmelCase_ ) if do_normalize: __lowerCAmelCase = self.normalize(image=lowerCAmelCase_ , mean=lowerCAmelCase_ , std=lowerCAmelCase_ ) __lowerCAmelCase = to_channel_dimension_format(lowerCAmelCase_ , lowerCAmelCase_ ) return image def lowercase ( self : Optional[int] , lowerCAmelCase_ : ImageInput , lowerCAmelCase_ : bool = None , lowerCAmelCase_ : Dict[str, int] = None , lowerCAmelCase_ : PILImageResampling = None , lowerCAmelCase_ : bool = None , lowerCAmelCase_ : Dict[str, int] = None , lowerCAmelCase_ : bool = None , lowerCAmelCase_ : float = None , lowerCAmelCase_ : bool = None , lowerCAmelCase_ : bool = None , lowerCAmelCase_ : Optional[Union[float, List[float]]] = None , lowerCAmelCase_ : Optional[Union[float, List[float]]] = None , lowerCAmelCase_ : Optional[Union[str, TensorType]] = None , lowerCAmelCase_ : ChannelDimension = ChannelDimension.FIRST , **lowerCAmelCase_ : Any , ) -> PIL.Image.Image: __lowerCAmelCase = do_resize if do_resize is not None else self.do_resize __lowerCAmelCase = resample if resample is not None else self.resample __lowerCAmelCase = do_center_crop if do_center_crop is not None else self.do_center_crop __lowerCAmelCase = do_rescale if do_rescale is not None else self.do_rescale __lowerCAmelCase = rescale_factor if rescale_factor is not None else self.rescale_factor __lowerCAmelCase = offset if offset is not None else self.offset __lowerCAmelCase = do_normalize if do_normalize is not None else self.do_normalize __lowerCAmelCase = image_mean if image_mean is not None else self.image_mean __lowerCAmelCase = image_std if image_std is not None else self.image_std __lowerCAmelCase = size if size is not None else self.size __lowerCAmelCase = get_size_dict(lowerCAmelCase_ , default_to_square=lowerCAmelCase_ ) __lowerCAmelCase = crop_size if crop_size is not None else self.crop_size __lowerCAmelCase = get_size_dict(lowerCAmelCase_ , param_name='crop_size' ) if not valid_images(lowerCAmelCase_ ): raise ValueError( 'Invalid image type. Must be of type PIL.Image.Image, numpy.ndarray, ' 'torch.Tensor, tf.Tensor or jax.ndarray.' ) __lowerCAmelCase = make_batched(lowerCAmelCase_ ) __lowerCAmelCase = [ [ self._preprocess_image( image=lowerCAmelCase_ , do_resize=lowerCAmelCase_ , size=lowerCAmelCase_ , resample=lowerCAmelCase_ , do_center_crop=lowerCAmelCase_ , crop_size=lowerCAmelCase_ , do_rescale=lowerCAmelCase_ , rescale_factor=lowerCAmelCase_ , offset=lowerCAmelCase_ , do_normalize=lowerCAmelCase_ , image_mean=lowerCAmelCase_ , image_std=lowerCAmelCase_ , data_format=lowerCAmelCase_ , ) for img in video ] for video in videos ] __lowerCAmelCase = {'pixel_values': videos} return BatchFeature(data=lowerCAmelCase_ , tensor_type=lowerCAmelCase_ )
284
import os import tempfile from functools import partial from unittest import TestCase from unittest.mock import patch import numpy as np import pytest from datasets.arrow_dataset import Dataset from datasets.search import ElasticSearchIndex, FaissIndex, MissingIndex from .utils import require_elasticsearch, require_faiss _snake_case : Dict = pytest.mark.integration @require_faiss class _UpperCAmelCase ( _UpperCamelCase ): """simple docstring""" def lowercase ( self : List[Any] ) -> Optional[Any]: __lowerCAmelCase = Dataset.from_dict({'filename': ['my_name-train' + '_' + str(lowerCAmelCase_ ) for x in np.arange(3_0 ).tolist()]} ) return dset def lowercase ( self : List[str] ) -> Tuple: import faiss __lowerCAmelCase = self._create_dummy_dataset() __lowerCAmelCase = dset.map( lambda lowerCAmelCase_ , lowerCAmelCase_ : {"vecs": i * np.ones(5 , dtype=np.floataa )} , with_indices=lowerCAmelCase_ , keep_in_memory=lowerCAmelCase_ ) __lowerCAmelCase = dset.add_faiss_index('vecs' , batch_size=1_0_0 , metric_type=faiss.METRIC_INNER_PRODUCT ) __lowerCAmelCase , __lowerCAmelCase = dset.get_nearest_examples('vecs' , np.ones(5 , dtype=np.floataa ) ) self.assertEqual(examples['filename'][0] , 'my_name-train_29' ) dset.drop_index('vecs' ) def lowercase ( self : Optional[Any] ) -> str: import faiss __lowerCAmelCase = self._create_dummy_dataset() dset.add_faiss_index_from_external_arrays( external_arrays=np.ones((3_0, 5) ) * np.arange(3_0 ).reshape(-1 , 1 ) , index_name='vecs' , batch_size=1_0_0 , metric_type=faiss.METRIC_INNER_PRODUCT , ) __lowerCAmelCase , __lowerCAmelCase = dset.get_nearest_examples('vecs' , np.ones(5 , dtype=np.floataa ) ) self.assertEqual(examples['filename'][0] , 'my_name-train_29' ) def lowercase ( self : int ) -> Optional[Any]: import faiss __lowerCAmelCase = self._create_dummy_dataset() dset.add_faiss_index_from_external_arrays( external_arrays=np.ones((3_0, 5) ) * np.arange(3_0 ).reshape(-1 , 1 ) , index_name='vecs' , metric_type=faiss.METRIC_INNER_PRODUCT , ) # Setting delete=False and unlinking manually is not pretty... but it is required on Windows to # ensure somewhat stable behaviour. If we don't, we get PermissionErrors. This is an age-old issue. # see https://bugs.python.org/issue14243 and # https://stackoverflow.com/questions/23212435/permission-denied-to-write-to-my-temporary-file/23212515 with tempfile.NamedTemporaryFile(delete=lowerCAmelCase_ ) as tmp_file: dset.save_faiss_index('vecs' , tmp_file.name ) dset.load_faiss_index('vecs2' , tmp_file.name ) os.unlink(tmp_file.name ) __lowerCAmelCase , __lowerCAmelCase = dset.get_nearest_examples('vecs2' , np.ones(5 , dtype=np.floataa ) ) self.assertEqual(examples['filename'][0] , 'my_name-train_29' ) def lowercase ( self : Union[str, Any] ) -> List[Any]: __lowerCAmelCase = self._create_dummy_dataset() dset.add_faiss_index_from_external_arrays( external_arrays=np.ones((3_0, 5) ) * np.arange(3_0 ).reshape(-1 , 1 ) , index_name='vecs' ) dset.drop_index('vecs' ) self.assertRaises(lowerCAmelCase_ , partial(dset.get_nearest_examples , 'vecs2' , np.ones(5 , dtype=np.floataa ) ) ) def lowercase ( self : Union[str, Any] ) -> Tuple: from elasticsearch import Elasticsearch __lowerCAmelCase = self._create_dummy_dataset() with patch('elasticsearch.Elasticsearch.search' ) as mocked_search, patch( 'elasticsearch.client.IndicesClient.create' ) as mocked_index_create, patch('elasticsearch.helpers.streaming_bulk' ) as mocked_bulk: __lowerCAmelCase = {'acknowledged': True} mocked_bulk.return_value([(True, None)] * 3_0 ) __lowerCAmelCase = {'hits': {'hits': [{'_score': 1, '_id': 2_9}]}} __lowerCAmelCase = Elasticsearch() dset.add_elasticsearch_index('filename' , es_client=lowerCAmelCase_ ) __lowerCAmelCase , __lowerCAmelCase = dset.get_nearest_examples('filename' , 'my_name-train_29' ) self.assertEqual(examples['filename'][0] , 'my_name-train_29' ) @require_faiss class _UpperCAmelCase ( _UpperCamelCase ): """simple docstring""" def lowercase ( self : str ) -> int: import faiss __lowerCAmelCase = FaissIndex(metric_type=faiss.METRIC_INNER_PRODUCT ) # add vectors index.add_vectors(np.eye(5 , dtype=np.floataa ) ) self.assertIsNotNone(index.faiss_index ) self.assertEqual(index.faiss_index.ntotal , 5 ) index.add_vectors(np.zeros((5, 5) , dtype=np.floataa ) ) self.assertEqual(index.faiss_index.ntotal , 1_0 ) # single query __lowerCAmelCase = np.zeros(5 , dtype=np.floataa ) __lowerCAmelCase = 1 __lowerCAmelCase , __lowerCAmelCase = index.search(lowerCAmelCase_ ) self.assertRaises(lowerCAmelCase_ , index.search , query.reshape(-1 , 1 ) ) self.assertGreater(scores[0] , 0 ) self.assertEqual(indices[0] , 1 ) # batched queries __lowerCAmelCase = np.eye(5 , dtype=np.floataa )[::-1] __lowerCAmelCase , __lowerCAmelCase = index.search_batch(lowerCAmelCase_ ) self.assertRaises(lowerCAmelCase_ , index.search_batch , queries[0] ) __lowerCAmelCase = [scores[0] for scores in total_scores] __lowerCAmelCase = [indices[0] for indices in total_indices] self.assertGreater(np.min(lowerCAmelCase_ ) , 0 ) self.assertListEqual([4, 3, 2, 1, 0] , lowerCAmelCase_ ) def lowercase ( self : List[Any] ) -> List[str]: import faiss __lowerCAmelCase = FaissIndex(string_factory='Flat' ) index.add_vectors(np.eye(5 , dtype=np.floataa ) ) self.assertIsInstance(index.faiss_index , faiss.IndexFlat ) __lowerCAmelCase = FaissIndex(string_factory='LSH' ) index.add_vectors(np.eye(5 , dtype=np.floataa ) ) self.assertIsInstance(index.faiss_index , faiss.IndexLSH ) with self.assertRaises(lowerCAmelCase_ ): __lowerCAmelCase = FaissIndex(string_factory='Flat' , custom_index=faiss.IndexFlat(5 ) ) def lowercase ( self : Union[str, Any] ) -> Dict: import faiss __lowerCAmelCase = faiss.IndexFlat(5 ) __lowerCAmelCase = FaissIndex(custom_index=lowerCAmelCase_ ) index.add_vectors(np.eye(5 , dtype=np.floataa ) ) self.assertIsInstance(index.faiss_index , faiss.IndexFlat ) def lowercase ( self : str ) -> Any: import faiss __lowerCAmelCase = FaissIndex(metric_type=faiss.METRIC_INNER_PRODUCT ) index.add_vectors(np.eye(5 , dtype=np.floataa ) ) # Setting delete=False and unlinking manually is not pretty... but it is required on Windows to # ensure somewhat stable behaviour. If we don't, we get PermissionErrors. This is an age-old issue. # see https://bugs.python.org/issue14243 and # https://stackoverflow.com/questions/23212435/permission-denied-to-write-to-my-temporary-file/23212515 with tempfile.NamedTemporaryFile(delete=lowerCAmelCase_ ) as tmp_file: index.save(tmp_file.name ) __lowerCAmelCase = FaissIndex.load(tmp_file.name ) os.unlink(tmp_file.name ) __lowerCAmelCase = np.zeros(5 , dtype=np.floataa ) __lowerCAmelCase = 1 __lowerCAmelCase , __lowerCAmelCase = index.search(lowerCAmelCase_ ) self.assertGreater(scores[0] , 0 ) self.assertEqual(indices[0] , 1 ) @require_faiss def a_ ( lowerCAmelCase_ : Union[str, Any] ): import faiss __lowerCAmelCase = FaissIndex(metric_type=faiss.METRIC_INNER_PRODUCT ) index.add_vectors(np.eye(5, dtype=np.floataa ) ) __lowerCAmelCase = 'index.faiss' __lowerCAmelCase = F"""mock://{index_name}""" index.save(lowerCAmelCase_, storage_options=mockfs.storage_options ) __lowerCAmelCase = FaissIndex.load(lowerCAmelCase_, storage_options=mockfs.storage_options ) __lowerCAmelCase = np.zeros(5, dtype=np.floataa ) __lowerCAmelCase = 1 __lowerCAmelCase , __lowerCAmelCase = index.search(lowerCAmelCase_ ) assert scores[0] > 0 assert indices[0] == 1 @require_elasticsearch class _UpperCAmelCase ( _UpperCamelCase ): """simple docstring""" def lowercase ( self : Any ) -> int: from elasticsearch import Elasticsearch with patch('elasticsearch.Elasticsearch.search' ) as mocked_search, patch( 'elasticsearch.client.IndicesClient.create' ) as mocked_index_create, patch('elasticsearch.helpers.streaming_bulk' ) as mocked_bulk: __lowerCAmelCase = Elasticsearch() __lowerCAmelCase = {'acknowledged': True} __lowerCAmelCase = ElasticSearchIndex(es_client=lowerCAmelCase_ ) mocked_bulk.return_value([(True, None)] * 3 ) index.add_documents(['foo', 'bar', 'foobar'] ) # single query __lowerCAmelCase = 'foo' __lowerCAmelCase = {'hits': {'hits': [{'_score': 1, '_id': 0}]}} __lowerCAmelCase , __lowerCAmelCase = index.search(lowerCAmelCase_ ) self.assertEqual(scores[0] , 1 ) self.assertEqual(indices[0] , 0 ) # single query with timeout __lowerCAmelCase = 'foo' __lowerCAmelCase = {'hits': {'hits': [{'_score': 1, '_id': 0}]}} __lowerCAmelCase , __lowerCAmelCase = index.search(lowerCAmelCase_ , request_timeout=3_0 ) self.assertEqual(scores[0] , 1 ) self.assertEqual(indices[0] , 0 ) # batched queries __lowerCAmelCase = ['foo', 'bar', 'foobar'] __lowerCAmelCase = {'hits': {'hits': [{'_score': 1, '_id': 1}]}} __lowerCAmelCase , __lowerCAmelCase = index.search_batch(lowerCAmelCase_ ) __lowerCAmelCase = [scores[0] for scores in total_scores] __lowerCAmelCase = [indices[0] for indices in total_indices] self.assertGreater(np.min(lowerCAmelCase_ ) , 0 ) self.assertListEqual([1, 1, 1] , lowerCAmelCase_ ) # batched queries with timeout __lowerCAmelCase = ['foo', 'bar', 'foobar'] __lowerCAmelCase = {'hits': {'hits': [{'_score': 1, '_id': 1}]}} __lowerCAmelCase , __lowerCAmelCase = index.search_batch(lowerCAmelCase_ , request_timeout=3_0 ) __lowerCAmelCase = [scores[0] for scores in total_scores] __lowerCAmelCase = [indices[0] for indices in total_indices] self.assertGreater(np.min(lowerCAmelCase_ ) , 0 ) self.assertListEqual([1, 1, 1] , lowerCAmelCase_ )
284
1
"""simple docstring""" def _lowerCAmelCase ( UpperCamelCase_ ): if not isinstance(__lowerCAmelCase , __lowerCAmelCase ): raise ValueError("""Input series is not valid, valid series - [2, 4, 6]""" ) if len(__lowerCAmelCase ) == 0: raise ValueError("""Input list must be a non empty list""" ) if len(__lowerCAmelCase ) == 1: return True __SCREAMING_SNAKE_CASE = series[1] - series[0] for index in range(len(__lowerCAmelCase ) - 1 ): if series[index + 1] - series[index] != common_diff: return False return True def _lowerCAmelCase ( UpperCamelCase_ ): if not isinstance(__lowerCAmelCase , __lowerCAmelCase ): raise ValueError("""Input series is not valid, valid series - [2, 4, 6]""" ) if len(__lowerCAmelCase ) == 0: raise ValueError("""Input list must be a non empty list""" ) __SCREAMING_SNAKE_CASE = 0 for val in series: answer += val return answer / len(__lowerCAmelCase ) if __name__ == "__main__": import doctest doctest.testmod()
353
"""simple docstring""" from typing import Callable, Optional, Union from ...configuration_utils import PretrainedConfig from ...utils import logging __magic_name__ = logging.get_logger(__name__) __magic_name__ = { "microsoft/xprophetnet-large-wiki100-cased": ( "https://huggingface.co/microsoft/xprophetnet-large-wiki100-cased/resolve/main/config.json" ), } class SCREAMING_SNAKE_CASE_ ( __a ): """simple docstring""" __lowercase : List[str] = '''xlm-prophetnet''' __lowercase : Dict = ['''past_key_values'''] __lowercase : Any = { '''num_attention_heads''': '''num_encoder_attention_heads''', } def __init__( self , lowerCAmelCase__ = 0.1 , lowerCAmelCase__ = "gelu" , lowerCAmelCase__ = 3_0_5_2_2 , lowerCAmelCase__ = 1_0_2_4 , lowerCAmelCase__ = 4_0_9_6 , lowerCAmelCase__ = 1_2 , lowerCAmelCase__ = 1_6 , lowerCAmelCase__ = 4_0_9_6 , lowerCAmelCase__ = 1_2 , lowerCAmelCase__ = 1_6 , lowerCAmelCase__ = 0.1 , lowerCAmelCase__ = 0.1 , lowerCAmelCase__ = 5_1_2 , lowerCAmelCase__ = 0.02 , lowerCAmelCase__ = True , lowerCAmelCase__ = True , lowerCAmelCase__ = 0 , lowerCAmelCase__ = 2 , lowerCAmelCase__ = 3_2 , lowerCAmelCase__ = 1_2_8 , lowerCAmelCase__ = False , lowerCAmelCase__ = 0.0 , lowerCAmelCase__ = True , lowerCAmelCase__ = 0 , lowerCAmelCase__ = 1 , lowerCAmelCase__ = 2 , **lowerCAmelCase__ , ): __SCREAMING_SNAKE_CASE = vocab_size __SCREAMING_SNAKE_CASE = hidden_size __SCREAMING_SNAKE_CASE = encoder_ffn_dim __SCREAMING_SNAKE_CASE = num_encoder_layers __SCREAMING_SNAKE_CASE = num_encoder_attention_heads __SCREAMING_SNAKE_CASE = decoder_ffn_dim __SCREAMING_SNAKE_CASE = num_decoder_layers __SCREAMING_SNAKE_CASE = num_decoder_attention_heads __SCREAMING_SNAKE_CASE = max_position_embeddings __SCREAMING_SNAKE_CASE = init_std # Normal(0, this parameter) __SCREAMING_SNAKE_CASE = activation_function # parameters for xlmprophetnet __SCREAMING_SNAKE_CASE = ngram __SCREAMING_SNAKE_CASE = num_buckets __SCREAMING_SNAKE_CASE = relative_max_distance __SCREAMING_SNAKE_CASE = disable_ngram_loss __SCREAMING_SNAKE_CASE = eps # 3 Types of Dropout __SCREAMING_SNAKE_CASE = attention_dropout __SCREAMING_SNAKE_CASE = activation_dropout __SCREAMING_SNAKE_CASE = dropout __SCREAMING_SNAKE_CASE = use_cache super().__init__( pad_token_id=lowerCAmelCase__ , bos_token_id=lowerCAmelCase__ , eos_token_id=lowerCAmelCase__ , is_encoder_decoder=lowerCAmelCase__ , add_cross_attention=lowerCAmelCase__ , decoder_start_token_id=lowerCAmelCase__ , **lowerCAmelCase__ , ) @property def snake_case_ ( self): return self.num_encoder_layers + self.num_decoder_layers @num_hidden_layers.setter def snake_case_ ( self , lowerCAmelCase__): raise NotImplementedError( """This model does not support the setting of `num_hidden_layers`. Please set `num_encoder_layers` and""" """ `num_decoder_layers`.""")
255
0
import math from numpy import inf from scipy.integrate import quad def _a ( UpperCamelCase_ : float ) -> float: """simple docstring""" if num <= 0: raise ValueError("math domain error" ) return quad(UpperCamelCase_ , 0 , UpperCamelCase_ , args=(UpperCamelCase_) )[0] def _a ( UpperCamelCase_ : float , UpperCamelCase_ : float ) -> float: """simple docstring""" return math.pow(UpperCamelCase_ , z - 1 ) * math.exp(-x ) if __name__ == "__main__": from doctest import testmod testmod()
340
import json import os import shutil import tempfile import unittest import numpy as np import pytest from transformers import CLIPTokenizer, CLIPTokenizerFast from transformers.models.clip.tokenization_clip import VOCAB_FILES_NAMES from transformers.testing_utils import require_vision from transformers.utils import IMAGE_PROCESSOR_NAME, is_vision_available if is_vision_available(): from PIL import Image from transformers import OwlViTImageProcessor, OwlViTProcessor @require_vision class lowercase__ ( unittest.TestCase ): def UpperCAmelCase ( self )-> Dict: '''simple docstring''' lowerCAmelCase__ = tempfile.mkdtemp() # fmt: off lowerCAmelCase__ = ["", "l", "o", "w", "e", "r", "s", "t", "i", "d", "n", "lo", "l</w>", "w</w>", "r</w>", "t</w>", "low</w>", "er</w>", "lowest</w>", "newer</w>", "wider", "<unk>", "<|startoftext|>", "<|endoftext|>"] # fmt: on lowerCAmelCase__ = dict(zip(__UpperCAmelCase , range(len(__UpperCAmelCase ) ) ) ) lowerCAmelCase__ = ["#version: 0.2", "l o", "lo w</w>", "e r</w>", ""] lowerCAmelCase__ = {"unk_token": "<unk>"} lowerCAmelCase__ = os.path.join(self.tmpdirname , VOCAB_FILES_NAMES["vocab_file"] ) lowerCAmelCase__ = 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(__UpperCAmelCase ) + "\n" ) with open(self.merges_file , "w" , encoding="utf-8" ) as fp: fp.write("\n".join(__UpperCAmelCase ) ) lowerCAmelCase__ = { "do_resize": True, "size": 20, "do_center_crop": True, "crop_size": 18, "do_normalize": True, "image_mean": [0.48_145_466, 0.4_578_275, 0.40_821_073], "image_std": [0.26_862_954, 0.26_130_258, 0.27_577_711], } lowerCAmelCase__ = os.path.join(self.tmpdirname , __UpperCAmelCase ) with open(self.image_processor_file , "w" , encoding="utf-8" ) as fp: json.dump(__UpperCAmelCase , __UpperCAmelCase ) def UpperCAmelCase ( self , **__UpperCAmelCase )-> Union[str, Any]: '''simple docstring''' return CLIPTokenizer.from_pretrained(self.tmpdirname , pad_token="!" , **__UpperCAmelCase ) def UpperCAmelCase ( self , **__UpperCAmelCase )-> Any: '''simple docstring''' return CLIPTokenizerFast.from_pretrained(self.tmpdirname , pad_token="!" , **__UpperCAmelCase ) def UpperCAmelCase ( self , **__UpperCAmelCase )-> Optional[Any]: '''simple docstring''' return OwlViTImageProcessor.from_pretrained(self.tmpdirname , **__UpperCAmelCase ) def UpperCAmelCase ( self )-> Any: '''simple docstring''' shutil.rmtree(self.tmpdirname ) def UpperCAmelCase ( self )-> int: '''simple docstring''' lowerCAmelCase__ = [np.random.randint(255 , size=(3, 30, 400) , dtype=np.uinta )] lowerCAmelCase__ = [Image.fromarray(np.moveaxis(__UpperCAmelCase , 0 , -1 ) ) for x in image_inputs] return image_inputs def UpperCAmelCase ( self )-> Optional[int]: '''simple docstring''' lowerCAmelCase__ = self.get_tokenizer() lowerCAmelCase__ = self.get_rust_tokenizer() lowerCAmelCase__ = self.get_image_processor() lowerCAmelCase__ = OwlViTProcessor(tokenizer=__UpperCAmelCase , image_processor=__UpperCAmelCase ) processor_slow.save_pretrained(self.tmpdirname ) lowerCAmelCase__ = OwlViTProcessor.from_pretrained(self.tmpdirname , use_fast=__UpperCAmelCase ) lowerCAmelCase__ = OwlViTProcessor(tokenizer=__UpperCAmelCase , image_processor=__UpperCAmelCase ) processor_fast.save_pretrained(self.tmpdirname ) lowerCAmelCase__ = OwlViTProcessor.from_pretrained(self.tmpdirname ) self.assertEqual(processor_slow.tokenizer.get_vocab() , tokenizer_slow.get_vocab() ) self.assertEqual(processor_fast.tokenizer.get_vocab() , tokenizer_fast.get_vocab() ) self.assertEqual(tokenizer_slow.get_vocab() , tokenizer_fast.get_vocab() ) self.assertIsInstance(processor_slow.tokenizer , __UpperCAmelCase ) self.assertIsInstance(processor_fast.tokenizer , __UpperCAmelCase ) self.assertEqual(processor_slow.image_processor.to_json_string() , image_processor.to_json_string() ) self.assertEqual(processor_fast.image_processor.to_json_string() , image_processor.to_json_string() ) self.assertIsInstance(processor_slow.image_processor , __UpperCAmelCase ) self.assertIsInstance(processor_fast.image_processor , __UpperCAmelCase ) def UpperCAmelCase ( self )-> List[Any]: '''simple docstring''' lowerCAmelCase__ = OwlViTProcessor(tokenizer=self.get_tokenizer() , image_processor=self.get_image_processor() ) processor.save_pretrained(self.tmpdirname ) lowerCAmelCase__ = self.get_tokenizer(bos_token="(BOS)" , eos_token="(EOS)" ) lowerCAmelCase__ = self.get_image_processor(do_normalize=__UpperCAmelCase ) lowerCAmelCase__ = OwlViTProcessor.from_pretrained( self.tmpdirname , bos_token="(BOS)" , eos_token="(EOS)" , do_normalize=__UpperCAmelCase ) self.assertEqual(processor.tokenizer.get_vocab() , tokenizer_add_kwargs.get_vocab() ) self.assertIsInstance(processor.tokenizer , __UpperCAmelCase ) self.assertEqual(processor.image_processor.to_json_string() , image_processor_add_kwargs.to_json_string() ) self.assertIsInstance(processor.image_processor , __UpperCAmelCase ) def UpperCAmelCase ( self )-> List[str]: '''simple docstring''' lowerCAmelCase__ = self.get_image_processor() lowerCAmelCase__ = self.get_tokenizer() lowerCAmelCase__ = OwlViTProcessor(tokenizer=__UpperCAmelCase , image_processor=__UpperCAmelCase ) lowerCAmelCase__ = self.prepare_image_inputs() lowerCAmelCase__ = image_processor(__UpperCAmelCase , return_tensors="np" ) lowerCAmelCase__ = processor(images=__UpperCAmelCase , return_tensors="np" ) for key in input_image_proc.keys(): self.assertAlmostEqual(input_image_proc[key].sum() , input_processor[key].sum() , delta=1E-2 ) def UpperCAmelCase ( self )-> Dict: '''simple docstring''' lowerCAmelCase__ = self.get_image_processor() lowerCAmelCase__ = self.get_tokenizer() lowerCAmelCase__ = OwlViTProcessor(tokenizer=__UpperCAmelCase , image_processor=__UpperCAmelCase ) lowerCAmelCase__ = "lower newer" lowerCAmelCase__ = processor(text=__UpperCAmelCase , return_tensors="np" ) lowerCAmelCase__ = tokenizer(__UpperCAmelCase , return_tensors="np" ) for key in encoded_tok.keys(): self.assertListEqual(encoded_tok[key][0].tolist() , encoded_processor[key][0].tolist() ) def UpperCAmelCase ( self )-> int: '''simple docstring''' lowerCAmelCase__ = self.get_image_processor() lowerCAmelCase__ = self.get_tokenizer() lowerCAmelCase__ = OwlViTProcessor(tokenizer=__UpperCAmelCase , image_processor=__UpperCAmelCase ) lowerCAmelCase__ = "lower newer" lowerCAmelCase__ = self.prepare_image_inputs() lowerCAmelCase__ = processor(text=__UpperCAmelCase , images=__UpperCAmelCase ) self.assertListEqual(list(inputs.keys() ) , ["input_ids", "attention_mask", "pixel_values"] ) # test if it raises when no input is passed with pytest.raises(__UpperCAmelCase ): processor() def UpperCAmelCase ( self )-> Any: '''simple docstring''' lowerCAmelCase__ = "google/owlvit-base-patch32" lowerCAmelCase__ = OwlViTProcessor.from_pretrained(__UpperCAmelCase ) lowerCAmelCase__ = ["cat", "nasa badge"] lowerCAmelCase__ = processor(text=__UpperCAmelCase ) lowerCAmelCase__ = 16 self.assertListEqual(list(inputs.keys() ) , ["input_ids", "attention_mask"] ) self.assertEqual(inputs["input_ids"].shape , (2, seq_length) ) # test if it raises when no input is passed with pytest.raises(__UpperCAmelCase ): processor() def UpperCAmelCase ( self )-> Optional[int]: '''simple docstring''' lowerCAmelCase__ = "google/owlvit-base-patch32" lowerCAmelCase__ = OwlViTProcessor.from_pretrained(__UpperCAmelCase ) lowerCAmelCase__ = [["cat", "nasa badge"], ["person"]] lowerCAmelCase__ = processor(text=__UpperCAmelCase ) lowerCAmelCase__ = 16 lowerCAmelCase__ = len(__UpperCAmelCase ) lowerCAmelCase__ = max([len(__UpperCAmelCase ) for texts in input_texts] ) self.assertListEqual(list(inputs.keys() ) , ["input_ids", "attention_mask"] ) self.assertEqual(inputs["input_ids"].shape , (batch_size * num_max_text_queries, seq_length) ) # test if it raises when no input is passed with pytest.raises(__UpperCAmelCase ): processor() def UpperCAmelCase ( self )-> str: '''simple docstring''' lowerCAmelCase__ = "google/owlvit-base-patch32" lowerCAmelCase__ = OwlViTProcessor.from_pretrained(__UpperCAmelCase ) lowerCAmelCase__ = ["cat", "nasa badge"] lowerCAmelCase__ = processor(text=__UpperCAmelCase ) lowerCAmelCase__ = 16 lowerCAmelCase__ = inputs["input_ids"] lowerCAmelCase__ = [ [49406, 2368, 49407, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0], [49406, 6841, 11301, 49407, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0], ] self.assertListEqual(list(inputs.keys() ) , ["input_ids", "attention_mask"] ) self.assertEqual(inputs["input_ids"].shape , (2, seq_length) ) self.assertListEqual(list(input_ids[0] ) , predicted_ids[0] ) self.assertListEqual(list(input_ids[1] ) , predicted_ids[1] ) def UpperCAmelCase ( self )-> List[str]: '''simple docstring''' lowerCAmelCase__ = self.get_image_processor() lowerCAmelCase__ = self.get_tokenizer() lowerCAmelCase__ = OwlViTProcessor(tokenizer=__UpperCAmelCase , image_processor=__UpperCAmelCase ) lowerCAmelCase__ = self.prepare_image_inputs() lowerCAmelCase__ = self.prepare_image_inputs() lowerCAmelCase__ = processor(images=__UpperCAmelCase , query_images=__UpperCAmelCase ) self.assertListEqual(list(inputs.keys() ) , ["query_pixel_values", "pixel_values"] ) # test if it raises when no input is passed with pytest.raises(__UpperCAmelCase ): processor() def UpperCAmelCase ( self )-> Tuple: '''simple docstring''' lowerCAmelCase__ = self.get_image_processor() lowerCAmelCase__ = self.get_tokenizer() lowerCAmelCase__ = OwlViTProcessor(tokenizer=__UpperCAmelCase , image_processor=__UpperCAmelCase ) lowerCAmelCase__ = [[1, 4, 5, 8, 1, 0, 8], [3, 4, 3, 1, 1, 8, 9]] lowerCAmelCase__ = processor.batch_decode(__UpperCAmelCase ) lowerCAmelCase__ = tokenizer.batch_decode(__UpperCAmelCase ) self.assertListEqual(__UpperCAmelCase , __UpperCAmelCase )
340
1
'''simple docstring''' import inspect import unittest from transformers import ConvNextConfig 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_backbone_common import BackboneTesterMixin 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 transformers import ConvNextBackbone, ConvNextForImageClassification, ConvNextModel from transformers.models.convnext.modeling_convnext import CONVNEXT_PRETRAINED_MODEL_ARCHIVE_LIST if is_vision_available(): from PIL import Image from transformers import AutoImageProcessor class _a : '''simple docstring''' def __init__( self, A, A=13, A=32, A=3, A=4, A=[10, 20, 30, 40], A=[2, 2, 3, 2], A=True, A=True, A=37, A="gelu", A=10, A=0.02, A=["stage2", "stage3", "stage4"], A=[2, 3, 4], A=None, ): '''simple docstring''' SCREAMING_SNAKE_CASE : int = parent SCREAMING_SNAKE_CASE : Optional[int] = batch_size SCREAMING_SNAKE_CASE : str = image_size SCREAMING_SNAKE_CASE : Any = num_channels SCREAMING_SNAKE_CASE : Tuple = num_stages SCREAMING_SNAKE_CASE : Optional[Any] = hidden_sizes SCREAMING_SNAKE_CASE : Optional[Any] = depths SCREAMING_SNAKE_CASE : Any = is_training SCREAMING_SNAKE_CASE : Union[str, Any] = use_labels SCREAMING_SNAKE_CASE : Optional[Any] = intermediate_size SCREAMING_SNAKE_CASE : int = hidden_act SCREAMING_SNAKE_CASE : Dict = num_labels SCREAMING_SNAKE_CASE : int = initializer_range SCREAMING_SNAKE_CASE : List[str] = out_features SCREAMING_SNAKE_CASE : Optional[Any] = out_indices SCREAMING_SNAKE_CASE : Any = scope def UpperCamelCase_ ( self ): '''simple docstring''' SCREAMING_SNAKE_CASE : Optional[int] = floats_tensor([self.batch_size, self.num_channels, self.image_size, self.image_size] ) SCREAMING_SNAKE_CASE : Optional[int] = None if self.use_labels: SCREAMING_SNAKE_CASE : Tuple = ids_tensor([self.batch_size], self.num_labels ) SCREAMING_SNAKE_CASE : str = self.get_config() return config, pixel_values, labels def UpperCamelCase_ ( self ): '''simple docstring''' return ConvNextConfig( num_channels=self.num_channels, hidden_sizes=self.hidden_sizes, depths=self.depths, num_stages=self.num_stages, hidden_act=self.hidden_act, is_decoder=A, initializer_range=self.initializer_range, out_features=self.out_features, out_indices=self.out_indices, num_labels=self.num_labels, ) def UpperCamelCase_ ( self, A, A, A ): '''simple docstring''' SCREAMING_SNAKE_CASE : List[Any] = ConvNextModel(config=A ) model.to(A ) model.eval() SCREAMING_SNAKE_CASE : Any = model(A ) # expected last hidden states: B, C, H // 32, W // 32 self.parent.assertEqual( result.last_hidden_state.shape, (self.batch_size, self.hidden_sizes[-1], self.image_size // 32, self.image_size // 32), ) def UpperCamelCase_ ( self, A, A, A ): '''simple docstring''' SCREAMING_SNAKE_CASE : str = ConvNextForImageClassification(A ) model.to(A ) model.eval() SCREAMING_SNAKE_CASE : str = model(A, labels=A ) self.parent.assertEqual(result.logits.shape, (self.batch_size, self.num_labels) ) def UpperCamelCase_ ( self, A, A, A ): '''simple docstring''' SCREAMING_SNAKE_CASE : int = ConvNextBackbone(config=A ) model.to(A ) model.eval() SCREAMING_SNAKE_CASE : List[Any] = model(A ) # verify hidden states self.parent.assertEqual(len(result.feature_maps ), len(config.out_features ) ) self.parent.assertListEqual(list(result.feature_maps[0].shape ), [self.batch_size, self.hidden_sizes[1], 4, 4] ) # verify channels self.parent.assertEqual(len(model.channels ), len(config.out_features ) ) self.parent.assertListEqual(model.channels, config.hidden_sizes[1:] ) # verify backbone works with out_features=None SCREAMING_SNAKE_CASE : List[str] = None SCREAMING_SNAKE_CASE : Dict = ConvNextBackbone(config=A ) model.to(A ) model.eval() SCREAMING_SNAKE_CASE : Any = model(A ) # verify feature maps self.parent.assertEqual(len(result.feature_maps ), 1 ) self.parent.assertListEqual(list(result.feature_maps[0].shape ), [self.batch_size, self.hidden_sizes[-1], 1, 1] ) # verify channels self.parent.assertEqual(len(model.channels ), 1 ) self.parent.assertListEqual(model.channels, [config.hidden_sizes[-1]] ) def UpperCamelCase_ ( self ): '''simple docstring''' SCREAMING_SNAKE_CASE : Tuple = self.prepare_config_and_inputs() SCREAMING_SNAKE_CASE , SCREAMING_SNAKE_CASE , SCREAMING_SNAKE_CASE : Optional[Any] = config_and_inputs SCREAMING_SNAKE_CASE : List[str] = {'pixel_values': pixel_values} return config, inputs_dict @require_torch class _a ( SCREAMING_SNAKE_CASE , SCREAMING_SNAKE_CASE , unittest.TestCase ): '''simple docstring''' A : Optional[Any] = ( ( ConvNextModel, ConvNextForImageClassification, ConvNextBackbone, ) if is_torch_available() else () ) A : Dict = ( {'''feature-extraction''': ConvNextModel, '''image-classification''': ConvNextForImageClassification} if is_torch_available() else {} ) A : str = True A : Dict = False A : str = False A : Tuple = False A : Any = False def UpperCamelCase_ ( self ): '''simple docstring''' SCREAMING_SNAKE_CASE : List[Any] = ConvNextModelTester(self ) SCREAMING_SNAKE_CASE : Optional[int] = ConfigTester(self, config_class=A, has_text_modality=A, hidden_size=37 ) def UpperCamelCase_ ( self ): '''simple docstring''' self.create_and_test_config_common_properties() self.config_tester.create_and_test_config_to_json_string() self.config_tester.create_and_test_config_to_json_file() self.config_tester.create_and_test_config_from_and_save_pretrained() self.config_tester.create_and_test_config_with_num_labels() self.config_tester.check_config_can_be_init_without_params() self.config_tester.check_config_arguments_init() def UpperCamelCase_ ( self ): '''simple docstring''' return @unittest.skip(reason='ConvNext does not use inputs_embeds' ) def UpperCamelCase_ ( self ): '''simple docstring''' pass @unittest.skip(reason='ConvNext does not support input and output embeddings' ) def UpperCamelCase_ ( self ): '''simple docstring''' pass @unittest.skip(reason='ConvNext does not use feedforward chunking' ) def UpperCamelCase_ ( self ): '''simple docstring''' pass def UpperCamelCase_ ( self ): '''simple docstring''' SCREAMING_SNAKE_CASE , SCREAMING_SNAKE_CASE : str = self.model_tester.prepare_config_and_inputs_for_common() for model_class in self.all_model_classes: SCREAMING_SNAKE_CASE : int = model_class(A ) SCREAMING_SNAKE_CASE : Any = inspect.signature(model.forward ) # signature.parameters is an OrderedDict => so arg_names order is deterministic SCREAMING_SNAKE_CASE : Tuple = [*signature.parameters.keys()] SCREAMING_SNAKE_CASE : Optional[Any] = ['pixel_values'] self.assertListEqual(arg_names[:1], A ) def UpperCamelCase_ ( self ): '''simple docstring''' SCREAMING_SNAKE_CASE : Tuple = self.model_tester.prepare_config_and_inputs() self.model_tester.create_and_check_model(*A ) def UpperCamelCase_ ( self ): '''simple docstring''' SCREAMING_SNAKE_CASE : Tuple = self.model_tester.prepare_config_and_inputs() self.model_tester.create_and_check_backbone(*A ) def UpperCamelCase_ ( self ): '''simple docstring''' def check_hidden_states_output(A, A, A ): SCREAMING_SNAKE_CASE : str = model_class(A ) model.to(A ) model.eval() with torch.no_grad(): SCREAMING_SNAKE_CASE : Union[str, Any] = model(**self._prepare_for_class(A, A ) ) SCREAMING_SNAKE_CASE : Optional[int] = outputs.encoder_hidden_states if config.is_encoder_decoder else outputs.hidden_states SCREAMING_SNAKE_CASE : Optional[int] = self.model_tester.num_stages self.assertEqual(len(A ), expected_num_stages + 1 ) # ConvNext's feature maps are of shape (batch_size, num_channels, height, width) self.assertListEqual( list(hidden_states[0].shape[-2:] ), [self.model_tester.image_size // 4, self.model_tester.image_size // 4], ) SCREAMING_SNAKE_CASE , SCREAMING_SNAKE_CASE : List[Any] = self.model_tester.prepare_config_and_inputs_for_common() for model_class in self.all_model_classes: SCREAMING_SNAKE_CASE : Tuple = True check_hidden_states_output(A, A, A ) # check that output_hidden_states also work using config del inputs_dict["output_hidden_states"] SCREAMING_SNAKE_CASE : Dict = True check_hidden_states_output(A, A, A ) def UpperCamelCase_ ( self ): '''simple docstring''' SCREAMING_SNAKE_CASE : List[Any] = self.model_tester.prepare_config_and_inputs() self.model_tester.create_and_check_for_image_classification(*A ) @slow def UpperCamelCase_ ( self ): '''simple docstring''' for model_name in CONVNEXT_PRETRAINED_MODEL_ARCHIVE_LIST[:1]: SCREAMING_SNAKE_CASE : Dict = ConvNextModel.from_pretrained(A ) self.assertIsNotNone(A ) def lowercase__( ): """simple docstring""" SCREAMING_SNAKE_CASE : Optional[int] = Image.open('./tests/fixtures/tests_samples/COCO/000000039769.png' ) return image @require_torch @require_vision class _a ( unittest.TestCase ): '''simple docstring''' @cached_property def UpperCamelCase_ ( self ): '''simple docstring''' return AutoImageProcessor.from_pretrained('facebook/convnext-tiny-224' ) if is_vision_available() else None @slow def UpperCamelCase_ ( self ): '''simple docstring''' SCREAMING_SNAKE_CASE : str = ConvNextForImageClassification.from_pretrained('facebook/convnext-tiny-224' ).to(A ) SCREAMING_SNAKE_CASE : Tuple = self.default_image_processor SCREAMING_SNAKE_CASE : str = prepare_img() SCREAMING_SNAKE_CASE : int = image_processor(images=A, return_tensors='pt' ).to(A ) # forward pass with torch.no_grad(): SCREAMING_SNAKE_CASE : List[str] = model(**A ) # verify the logits SCREAMING_SNAKE_CASE : List[Any] = torch.Size((1, 1_000) ) self.assertEqual(outputs.logits.shape, A ) SCREAMING_SNAKE_CASE : Dict = torch.tensor([-0.02_60, -0.47_39, 0.19_11] ).to(A ) self.assertTrue(torch.allclose(outputs.logits[0, :3], A, atol=1E-4 ) ) @require_torch class _a ( unittest.TestCase , SCREAMING_SNAKE_CASE ): '''simple docstring''' A : int = (ConvNextBackbone,) if is_torch_available() else () A : int = ConvNextConfig A : Union[str, Any] = False def UpperCamelCase_ ( self ): '''simple docstring''' SCREAMING_SNAKE_CASE : List[str] = ConvNextModelTester(self )
246
'''simple docstring''' from typing import TYPE_CHECKING from ...utils import ( OptionalDependencyNotAvailable, _LazyModule, is_tf_available, is_tokenizers_available, is_torch_available, ) UpperCamelCase_ = { "configuration_mobilebert": [ "MOBILEBERT_PRETRAINED_CONFIG_ARCHIVE_MAP", "MobileBertConfig", "MobileBertOnnxConfig", ], "tokenization_mobilebert": ["MobileBertTokenizer"], } try: if not is_tokenizers_available(): raise OptionalDependencyNotAvailable() except OptionalDependencyNotAvailable: pass else: UpperCamelCase_ = ["MobileBertTokenizerFast"] try: if not is_torch_available(): raise OptionalDependencyNotAvailable() except OptionalDependencyNotAvailable: pass else: UpperCamelCase_ = [ "MOBILEBERT_PRETRAINED_MODEL_ARCHIVE_LIST", "MobileBertForMaskedLM", "MobileBertForMultipleChoice", "MobileBertForNextSentencePrediction", "MobileBertForPreTraining", "MobileBertForQuestionAnswering", "MobileBertForSequenceClassification", "MobileBertForTokenClassification", "MobileBertLayer", "MobileBertModel", "MobileBertPreTrainedModel", "load_tf_weights_in_mobilebert", ] try: if not is_tf_available(): raise OptionalDependencyNotAvailable() except OptionalDependencyNotAvailable: pass else: UpperCamelCase_ = [ "TF_MOBILEBERT_PRETRAINED_MODEL_ARCHIVE_LIST", "TFMobileBertForMaskedLM", "TFMobileBertForMultipleChoice", "TFMobileBertForNextSentencePrediction", "TFMobileBertForPreTraining", "TFMobileBertForQuestionAnswering", "TFMobileBertForSequenceClassification", "TFMobileBertForTokenClassification", "TFMobileBertMainLayer", "TFMobileBertModel", "TFMobileBertPreTrainedModel", ] if TYPE_CHECKING: from .configuration_mobilebert import ( MOBILEBERT_PRETRAINED_CONFIG_ARCHIVE_MAP, MobileBertConfig, MobileBertOnnxConfig, ) from .tokenization_mobilebert import MobileBertTokenizer try: if not is_tokenizers_available(): raise OptionalDependencyNotAvailable() except OptionalDependencyNotAvailable: pass else: from .tokenization_mobilebert_fast import MobileBertTokenizerFast try: if not is_torch_available(): raise OptionalDependencyNotAvailable() except OptionalDependencyNotAvailable: pass else: from .modeling_mobilebert import ( MOBILEBERT_PRETRAINED_MODEL_ARCHIVE_LIST, MobileBertForMaskedLM, MobileBertForMultipleChoice, MobileBertForNextSentencePrediction, MobileBertForPreTraining, MobileBertForQuestionAnswering, MobileBertForSequenceClassification, MobileBertForTokenClassification, MobileBertLayer, MobileBertModel, MobileBertPreTrainedModel, load_tf_weights_in_mobilebert, ) try: if not is_tf_available(): raise OptionalDependencyNotAvailable() except OptionalDependencyNotAvailable: pass else: from .modeling_tf_mobilebert import ( TF_MOBILEBERT_PRETRAINED_MODEL_ARCHIVE_LIST, TFMobileBertForMaskedLM, TFMobileBertForMultipleChoice, TFMobileBertForNextSentencePrediction, TFMobileBertForPreTraining, TFMobileBertForQuestionAnswering, TFMobileBertForSequenceClassification, TFMobileBertForTokenClassification, TFMobileBertMainLayer, TFMobileBertModel, TFMobileBertPreTrainedModel, ) else: import sys UpperCamelCase_ = _LazyModule(__name__, globals()["__file__"], _import_structure, module_spec=__spec__)
246
1
import argparse from pathlib import Path import torch from packaging import version from torch.onnx import export from diffusers import AutoencoderKL lowerCAmelCase__ : List[Any] = version.parse(version.parse(torch.__version__).base_version) < version.parse('''1.11''') def UpperCamelCase__ ( A__ , A__ , A__ , A__ , A__ , A__ , A__ , A__=False , ) -> Optional[int]: output_path.parent.mkdir(parents=A__ , exist_ok=A__ ) # PyTorch deprecated the `enable_onnx_checker` and `use_external_data_format` arguments in v1.11, # so we check the torch version for backwards compatibility if is_torch_less_than_1_11: export( A__ , A__ , f=output_path.as_posix() , input_names=A__ , output_names=A__ , dynamic_axes=A__ , do_constant_folding=A__ , use_external_data_format=A__ , enable_onnx_checker=A__ , opset_version=A__ , ) else: export( A__ , A__ , f=output_path.as_posix() , input_names=A__ , output_names=A__ , dynamic_axes=A__ , do_constant_folding=A__ , opset_version=A__ , ) @torch.no_grad() def UpperCamelCase__ ( A__ , A__ , A__ , A__ = False ) -> Union[str, Any]: snake_case__ : Dict = torch.floataa if fpaa else torch.floataa if fpaa and torch.cuda.is_available(): snake_case__ : str = 'cuda' elif fpaa and not torch.cuda.is_available(): raise ValueError('`float16` model export is only supported on GPUs with CUDA' ) else: snake_case__ : str = 'cpu' snake_case__ : Optional[Any] = Path(A__ ) # VAE DECODER snake_case__ : Tuple = AutoencoderKL.from_pretrained(model_path + '/vae' ) snake_case__ : Any = vae_decoder.config.latent_channels # forward only through the decoder part snake_case__ : str = vae_decoder.decode onnx_export( A__ , model_args=( torch.randn(1 , A__ , 25 , 25 ).to(device=A__ , dtype=A__ ), False, ) , output_path=output_path / 'vae_decoder' / 'model.onnx' , ordered_input_names=['latent_sample', 'return_dict'] , output_names=['sample'] , dynamic_axes={ 'latent_sample': {0: 'batch', 1: 'channels', 2: 'height', 3: 'width'}, } , opset=A__ , ) del vae_decoder if __name__ == "__main__": lowerCAmelCase__ : Union[str, Any] = argparse.ArgumentParser() parser.add_argument( '''--model_path''', type=str, required=True, help='''Path to the `diffusers` checkpoint to convert (either a local directory or on the Hub).''', ) parser.add_argument('''--output_path''', type=str, required=True, help='''Path to the output model.''') parser.add_argument( '''--opset''', default=14, type=int, help='''The version of the ONNX operator set to use.''', ) parser.add_argument('''--fp16''', action='''store_true''', default=False, help='''Export the models in `float16` mode''') lowerCAmelCase__ : Tuple = parser.parse_args() print(args.output_path) convert_models(args.model_path, args.output_path, args.opset, args.fpaa) print('''SD: Done: ONNX''')
143
import os from itertools import chain from random import randrange, shuffle import pytest from .sola import PokerHand lowerCAmelCase__ : str = ( '''4S 3H 2C 7S 5H''', '''9D 8H 2C 6S 7H''', '''2D 6D 9D TH 7D''', '''TC 8C 2S JH 6C''', '''JH 8S TH AH QH''', '''TS KS 5S 9S AC''', '''KD 6S 9D TH AD''', '''KS 8D 4D 9S 4S''', # pair '''8C 4S KH JS 4D''', # pair '''QH 8H KD JH 8S''', # pair '''KC 4H KS 2H 8D''', # pair '''KD 4S KC 3H 8S''', # pair '''AH 8S AS KC JH''', # pair '''3H 4C 4H 3S 2H''', # 2 pairs '''5S 5D 2C KH KH''', # 2 pairs '''3C KH 5D 5S KH''', # 2 pairs '''AS 3C KH AD KH''', # 2 pairs '''7C 7S 3S 7H 5S''', # 3 of a kind '''7C 7S KH 2H 7H''', # 3 of a kind '''AC KH QH AH AS''', # 3 of a kind '''2H 4D 3C AS 5S''', # straight (low ace) '''3C 5C 4C 2C 6H''', # straight '''6S 8S 7S 5H 9H''', # straight '''JS QS 9H TS KH''', # straight '''QC KH TS JS AH''', # straight (high ace) '''8C 9C 5C 3C TC''', # flush '''3S 8S 9S 5S KS''', # flush '''4C 5C 9C 8C KC''', # flush '''JH 8H AH KH QH''', # flush '''3D 2H 3H 2C 2D''', # full house '''2H 2C 3S 3H 3D''', # full house '''KH KC 3S 3H 3D''', # full house '''JC 6H JS JD JH''', # 4 of a kind '''JC 7H JS JD JH''', # 4 of a kind '''JC KH JS JD JH''', # 4 of a kind '''2S AS 4S 5S 3S''', # straight flush (low ace) '''2D 6D 3D 4D 5D''', # straight flush '''5C 6C 3C 7C 4C''', # straight flush '''JH 9H TH KH QH''', # straight flush '''JH AH TH KH QH''', # royal flush (high ace straight flush) ) lowerCAmelCase__ : Optional[int] = ( ('''2H 3H 4H 5H 6H''', '''KS AS TS QS JS''', '''Loss'''), ('''2H 3H 4H 5H 6H''', '''AS AD AC AH JD''', '''Win'''), ('''AS AH 2H AD AC''', '''JS JD JC JH 3D''', '''Win'''), ('''2S AH 2H AS AC''', '''JS JD JC JH AD''', '''Loss'''), ('''2S AH 2H AS AC''', '''2H 3H 5H 6H 7H''', '''Win'''), ('''AS 3S 4S 8S 2S''', '''2H 3H 5H 6H 7H''', '''Win'''), ('''2H 3H 5H 6H 7H''', '''2S 3H 4H 5S 6C''', '''Win'''), ('''2S 3H 4H 5S 6C''', '''3D 4C 5H 6H 2S''', '''Tie'''), ('''2S 3H 4H 5S 6C''', '''AH AC 5H 6H AS''', '''Win'''), ('''2S 2H 4H 5S 4C''', '''AH AC 5H 6H AS''', '''Loss'''), ('''2S 2H 4H 5S 4C''', '''AH AC 5H 6H 7S''', '''Win'''), ('''6S AD 7H 4S AS''', '''AH AC 5H 6H 7S''', '''Loss'''), ('''2S AH 4H 5S KC''', '''AH AC 5H 6H 7S''', '''Loss'''), ('''2S 3H 6H 7S 9C''', '''7H 3C TH 6H 9S''', '''Loss'''), ('''4S 5H 6H TS AC''', '''3S 5H 6H TS AC''', '''Win'''), ('''2S AH 4H 5S 6C''', '''AD 4C 5H 6H 2C''', '''Tie'''), ('''AS AH 3H AD AC''', '''AS AH 2H AD AC''', '''Win'''), ('''AH AC 5H 5C QS''', '''AH AC 5H 5C KS''', '''Loss'''), ('''AH AC 5H 5C QS''', '''KH KC 5H 5C QS''', '''Win'''), ('''7C 7S KH 2H 7H''', '''3C 3S AH 2H 3H''', '''Win'''), ('''3C 3S AH 2H 3H''', '''7C 7S KH 2H 7H''', '''Loss'''), ('''6H 5H 4H 3H 2H''', '''5H 4H 3H 2H AH''', '''Win'''), ('''5H 4H 3H 2H AH''', '''5H 4H 3H 2H AH''', '''Tie'''), ('''5H 4H 3H 2H AH''', '''6H 5H 4H 3H 2H''', '''Loss'''), ('''AH AD KS KC AC''', '''AH KD KH AC KC''', '''Win'''), ('''2H 4D 3C AS 5S''', '''2H 4D 3C 6S 5S''', '''Loss'''), ('''2H 3S 3C 3H 2S''', '''3S 3C 2S 2H 2D''', '''Win'''), ('''4D 6D 5D 2D JH''', '''3S 8S 3H TC KH''', '''Loss'''), ('''4S 6C 8S 3S 7S''', '''AD KS 2D 7D 7C''', '''Loss'''), ('''6S 4C 7H 8C 3H''', '''5H JC AH 9D 9C''', '''Loss'''), ('''9D 9H JH TC QH''', '''3C 2S JS 5C 7H''', '''Win'''), ('''2H TC 8S AD 9S''', '''4H TS 7H 2C 5C''', '''Win'''), ('''9D 3S 2C 7S 7C''', '''JC TD 3C TC 9H''', '''Loss'''), ) lowerCAmelCase__ : Optional[Any] = ( ('''2H 3H 4H 5H 6H''', True), ('''AS AH 2H AD AC''', False), ('''2H 3H 5H 6H 7H''', True), ('''KS AS TS QS JS''', True), ('''8H 9H QS JS TH''', False), ('''AS 3S 4S 8S 2S''', True), ) lowerCAmelCase__ : List[Any] = ( ('''2H 3H 4H 5H 6H''', True), ('''AS AH 2H AD AC''', False), ('''2H 3H 5H 6H 7H''', False), ('''KS AS TS QS JS''', True), ('''8H 9H QS JS TH''', True), ) lowerCAmelCase__ : Any = ( ('''2H 4D 3C AS 5S''', True, [5, 4, 3, 2, 14]), ('''2H 5D 3C AS 5S''', False, [14, 5, 5, 3, 2]), ('''JH QD KC AS TS''', False, [14, 13, 12, 11, 10]), ('''9D 3S 2C 7S 7C''', False, [9, 7, 7, 3, 2]), ) lowerCAmelCase__ : Dict = ( ('''JH AH TH KH QH''', 0), ('''JH 9H TH KH QH''', 0), ('''JC KH JS JD JH''', 7), ('''KH KC 3S 3H 3D''', 6), ('''8C 9C 5C 3C TC''', 0), ('''JS QS 9H TS KH''', 0), ('''7C 7S KH 2H 7H''', 3), ('''3C KH 5D 5S KH''', 2), ('''QH 8H KD JH 8S''', 1), ('''2D 6D 9D TH 7D''', 0), ) lowerCAmelCase__ : Optional[int] = ( ('''JH AH TH KH QH''', 23), ('''JH 9H TH KH QH''', 22), ('''JC KH JS JD JH''', 21), ('''KH KC 3S 3H 3D''', 20), ('''8C 9C 5C 3C TC''', 19), ('''JS QS 9H TS KH''', 18), ('''7C 7S KH 2H 7H''', 17), ('''3C KH 5D 5S KH''', 16), ('''QH 8H KD JH 8S''', 15), ('''2D 6D 9D TH 7D''', 14), ) def UpperCamelCase__ ( ) -> Any: snake_case__ , snake_case__ : List[str] = randrange(len(A__ ) ), randrange(len(A__ ) ) snake_case__ : str = ['Loss', 'Tie', 'Win'][(play >= oppo) + (play > oppo)] snake_case__ , snake_case__ : Any = SORTED_HANDS[play], SORTED_HANDS[oppo] return hand, other, expected def UpperCamelCase__ ( A__ = 100 ) -> Optional[int]: return (generate_random_hand() for _ in range(A__ )) @pytest.mark.parametrize('hand, expected' , A__ ) def UpperCamelCase__ ( A__ , A__ ) -> Union[str, Any]: assert PokerHand(A__ )._is_flush() == expected @pytest.mark.parametrize('hand, expected' , A__ ) def UpperCamelCase__ ( A__ , A__ ) -> Any: assert PokerHand(A__ )._is_straight() == expected @pytest.mark.parametrize('hand, expected, card_values' , A__ ) def UpperCamelCase__ ( A__ , A__ , A__ ) -> Dict: snake_case__ : Optional[int] = PokerHand(A__ ) assert player._is_five_high_straight() == expected assert player._card_values == card_values @pytest.mark.parametrize('hand, expected' , A__ ) def UpperCamelCase__ ( A__ , A__ ) -> str: assert PokerHand(A__ )._is_same_kind() == expected @pytest.mark.parametrize('hand, expected' , A__ ) def UpperCamelCase__ ( A__ , A__ ) -> Optional[Any]: assert PokerHand(A__ )._hand_type == expected @pytest.mark.parametrize('hand, other, expected' , A__ ) def UpperCamelCase__ ( A__ , A__ , A__ ) -> int: assert PokerHand(A__ ).compare_with(PokerHand(A__ ) ) == expected @pytest.mark.parametrize('hand, other, expected' , generate_random_hands() ) def UpperCamelCase__ ( A__ , A__ , A__ ) -> Union[str, Any]: assert PokerHand(A__ ).compare_with(PokerHand(A__ ) ) == expected def UpperCamelCase__ ( ) -> Union[str, Any]: snake_case__ : Union[str, Any] = [PokerHand(A__ ) for hand in SORTED_HANDS] snake_case__ : Optional[Any] = poker_hands.copy() shuffle(A__ ) snake_case__ : Tuple = chain(sorted(A__ ) ) for index, hand in enumerate(A__ ): assert hand == poker_hands[index] def UpperCamelCase__ ( ) -> str: # Test that five high straights are compared correctly. snake_case__ : int = [PokerHand('2D AC 3H 4H 5S' ), PokerHand('2S 3H 4H 5S 6C' )] pokerhands.sort(reverse=A__ ) assert pokerhands[0].__str__() == "2S 3H 4H 5S 6C" def UpperCamelCase__ ( ) -> Union[str, Any]: # Multiple calls to five_high_straight function should still return True # and shouldn't mutate the list in every call other than the first. snake_case__ : Optional[int] = PokerHand('2C 4S AS 3D 5C' ) snake_case__ : Optional[int] = True snake_case__ : Tuple = [5, 4, 3, 2, 14] for _ in range(10 ): assert pokerhand._is_five_high_straight() == expected assert pokerhand._card_values == expected_card_values def UpperCamelCase__ ( ) -> List[str]: # Problem number 54 from Project Euler # Testing from poker_hands.txt file snake_case__ : Any = 0 snake_case__ : Optional[Any] = os.path.abspath(os.path.dirname(A__ ) ) snake_case__ : List[str] = os.path.join(A__ , 'poker_hands.txt' ) with open(A__ ) as file_hand: for line in file_hand: snake_case__ : Tuple = line[:14].strip() snake_case__ : List[str] = line[15:].strip() snake_case__ , snake_case__ : Any = PokerHand(A__ ), PokerHand(A__ ) snake_case__ : Tuple = player.compare_with(A__ ) if output == "Win": answer += 1 assert answer == 376
143
1
"""simple docstring""" import argparse from transformers import TaConfig, TaForConditionalGeneration, load_tf_weights_in_ta from transformers.utils import logging logging.set_verbosity_info() def a__ ( SCREAMING_SNAKE_CASE : Tuple , SCREAMING_SNAKE_CASE : Optional[int] , SCREAMING_SNAKE_CASE : List[Any] ): '''simple docstring''' lowerCAmelCase : List[str] = TaConfig.from_json_file(SCREAMING_SNAKE_CASE ) print(f"""Building PyTorch model from configuration: {config}""" ) lowerCAmelCase : List[str] = TaForConditionalGeneration(SCREAMING_SNAKE_CASE ) # Load weights from tf checkpoint load_tf_weights_in_ta(SCREAMING_SNAKE_CASE , SCREAMING_SNAKE_CASE , SCREAMING_SNAKE_CASE ) # Save pytorch-model print(f"""Save PyTorch model to {pytorch_dump_path}""" ) model.save_pretrained(SCREAMING_SNAKE_CASE ) if __name__ == "__main__": lowerCAmelCase__ = 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 T5 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.''' ) lowerCAmelCase__ = parser.parse_args() convert_tf_checkpoint_to_pytorch(args.tf_checkpoint_path, args.config_file, args.pytorch_dump_path)
133
"""simple docstring""" def a__ ( SCREAMING_SNAKE_CASE : list[int] ): '''simple docstring''' lowerCAmelCase : str = len(SCREAMING_SNAKE_CASE ) for i in range(SCREAMING_SNAKE_CASE ): for j in range(i + 1 , SCREAMING_SNAKE_CASE ): if numbers[j] < numbers[i]: lowerCAmelCase , lowerCAmelCase : Any = numbers[j], numbers[i] return numbers if __name__ == "__main__": lowerCAmelCase__ = input('''Enter numbers separated by a comma:\n''').strip() lowerCAmelCase__ = [int(item) for item in user_input.split(''',''')] print(exchange_sort(unsorted))
133
1
"""simple docstring""" from collections import OrderedDict from ...utils import logging from .auto_factory import _BaseAutoModelClass, _LazyAutoMapping, auto_class_update from .configuration_auto import CONFIG_MAPPING_NAMES SCREAMING_SNAKE_CASE__ = logging.get_logger(__name__) SCREAMING_SNAKE_CASE__ = OrderedDict( [ # Base model mapping ("albert", "FlaxAlbertModel"), ("bart", "FlaxBartModel"), ("beit", "FlaxBeitModel"), ("bert", "FlaxBertModel"), ("big_bird", "FlaxBigBirdModel"), ("blenderbot", "FlaxBlenderbotModel"), ("blenderbot-small", "FlaxBlenderbotSmallModel"), ("clip", "FlaxCLIPModel"), ("distilbert", "FlaxDistilBertModel"), ("electra", "FlaxElectraModel"), ("gpt-sw3", "FlaxGPT2Model"), ("gpt2", "FlaxGPT2Model"), ("gpt_neo", "FlaxGPTNeoModel"), ("gptj", "FlaxGPTJModel"), ("longt5", "FlaxLongT5Model"), ("marian", "FlaxMarianModel"), ("mbart", "FlaxMBartModel"), ("mt5", "FlaxMT5Model"), ("opt", "FlaxOPTModel"), ("pegasus", "FlaxPegasusModel"), ("regnet", "FlaxRegNetModel"), ("resnet", "FlaxResNetModel"), ("roberta", "FlaxRobertaModel"), ("roberta-prelayernorm", "FlaxRobertaPreLayerNormModel"), ("roformer", "FlaxRoFormerModel"), ("t5", "FlaxT5Model"), ("vision-text-dual-encoder", "FlaxVisionTextDualEncoderModel"), ("vit", "FlaxViTModel"), ("wav2vec2", "FlaxWav2Vec2Model"), ("whisper", "FlaxWhisperModel"), ("xglm", "FlaxXGLMModel"), ("xlm-roberta", "FlaxXLMRobertaModel"), ] ) SCREAMING_SNAKE_CASE__ = OrderedDict( [ # Model for pre-training mapping ("albert", "FlaxAlbertForPreTraining"), ("bart", "FlaxBartForConditionalGeneration"), ("bert", "FlaxBertForPreTraining"), ("big_bird", "FlaxBigBirdForPreTraining"), ("electra", "FlaxElectraForPreTraining"), ("longt5", "FlaxLongT5ForConditionalGeneration"), ("mbart", "FlaxMBartForConditionalGeneration"), ("mt5", "FlaxMT5ForConditionalGeneration"), ("roberta", "FlaxRobertaForMaskedLM"), ("roberta-prelayernorm", "FlaxRobertaPreLayerNormForMaskedLM"), ("roformer", "FlaxRoFormerForMaskedLM"), ("t5", "FlaxT5ForConditionalGeneration"), ("wav2vec2", "FlaxWav2Vec2ForPreTraining"), ("whisper", "FlaxWhisperForConditionalGeneration"), ("xlm-roberta", "FlaxXLMRobertaForMaskedLM"), ] ) SCREAMING_SNAKE_CASE__ = OrderedDict( [ # Model for Masked LM mapping ("albert", "FlaxAlbertForMaskedLM"), ("bart", "FlaxBartForConditionalGeneration"), ("bert", "FlaxBertForMaskedLM"), ("big_bird", "FlaxBigBirdForMaskedLM"), ("distilbert", "FlaxDistilBertForMaskedLM"), ("electra", "FlaxElectraForMaskedLM"), ("mbart", "FlaxMBartForConditionalGeneration"), ("roberta", "FlaxRobertaForMaskedLM"), ("roberta-prelayernorm", "FlaxRobertaPreLayerNormForMaskedLM"), ("roformer", "FlaxRoFormerForMaskedLM"), ("xlm-roberta", "FlaxXLMRobertaForMaskedLM"), ] ) SCREAMING_SNAKE_CASE__ = OrderedDict( [ # Model for Seq2Seq Causal LM mapping ("bart", "FlaxBartForConditionalGeneration"), ("blenderbot", "FlaxBlenderbotForConditionalGeneration"), ("blenderbot-small", "FlaxBlenderbotSmallForConditionalGeneration"), ("encoder-decoder", "FlaxEncoderDecoderModel"), ("longt5", "FlaxLongT5ForConditionalGeneration"), ("marian", "FlaxMarianMTModel"), ("mbart", "FlaxMBartForConditionalGeneration"), ("mt5", "FlaxMT5ForConditionalGeneration"), ("pegasus", "FlaxPegasusForConditionalGeneration"), ("t5", "FlaxT5ForConditionalGeneration"), ] ) SCREAMING_SNAKE_CASE__ = OrderedDict( [ # Model for Image-classsification ("beit", "FlaxBeitForImageClassification"), ("regnet", "FlaxRegNetForImageClassification"), ("resnet", "FlaxResNetForImageClassification"), ("vit", "FlaxViTForImageClassification"), ] ) SCREAMING_SNAKE_CASE__ = OrderedDict( [ ("vision-encoder-decoder", "FlaxVisionEncoderDecoderModel"), ] ) SCREAMING_SNAKE_CASE__ = OrderedDict( [ # Model for Causal LM mapping ("bart", "FlaxBartForCausalLM"), ("bert", "FlaxBertForCausalLM"), ("big_bird", "FlaxBigBirdForCausalLM"), ("electra", "FlaxElectraForCausalLM"), ("gpt-sw3", "FlaxGPT2LMHeadModel"), ("gpt2", "FlaxGPT2LMHeadModel"), ("gpt_neo", "FlaxGPTNeoForCausalLM"), ("gptj", "FlaxGPTJForCausalLM"), ("opt", "FlaxOPTForCausalLM"), ("roberta", "FlaxRobertaForCausalLM"), ("roberta-prelayernorm", "FlaxRobertaPreLayerNormForCausalLM"), ("xglm", "FlaxXGLMForCausalLM"), ("xlm-roberta", "FlaxXLMRobertaForCausalLM"), ] ) SCREAMING_SNAKE_CASE__ = OrderedDict( [ # Model for Sequence Classification mapping ("albert", "FlaxAlbertForSequenceClassification"), ("bart", "FlaxBartForSequenceClassification"), ("bert", "FlaxBertForSequenceClassification"), ("big_bird", "FlaxBigBirdForSequenceClassification"), ("distilbert", "FlaxDistilBertForSequenceClassification"), ("electra", "FlaxElectraForSequenceClassification"), ("mbart", "FlaxMBartForSequenceClassification"), ("roberta", "FlaxRobertaForSequenceClassification"), ("roberta-prelayernorm", "FlaxRobertaPreLayerNormForSequenceClassification"), ("roformer", "FlaxRoFormerForSequenceClassification"), ("xlm-roberta", "FlaxXLMRobertaForSequenceClassification"), ] ) SCREAMING_SNAKE_CASE__ = OrderedDict( [ # Model for Question Answering mapping ("albert", "FlaxAlbertForQuestionAnswering"), ("bart", "FlaxBartForQuestionAnswering"), ("bert", "FlaxBertForQuestionAnswering"), ("big_bird", "FlaxBigBirdForQuestionAnswering"), ("distilbert", "FlaxDistilBertForQuestionAnswering"), ("electra", "FlaxElectraForQuestionAnswering"), ("mbart", "FlaxMBartForQuestionAnswering"), ("roberta", "FlaxRobertaForQuestionAnswering"), ("roberta-prelayernorm", "FlaxRobertaPreLayerNormForQuestionAnswering"), ("roformer", "FlaxRoFormerForQuestionAnswering"), ("xlm-roberta", "FlaxXLMRobertaForQuestionAnswering"), ] ) SCREAMING_SNAKE_CASE__ = OrderedDict( [ # Model for Token Classification mapping ("albert", "FlaxAlbertForTokenClassification"), ("bert", "FlaxBertForTokenClassification"), ("big_bird", "FlaxBigBirdForTokenClassification"), ("distilbert", "FlaxDistilBertForTokenClassification"), ("electra", "FlaxElectraForTokenClassification"), ("roberta", "FlaxRobertaForTokenClassification"), ("roberta-prelayernorm", "FlaxRobertaPreLayerNormForTokenClassification"), ("roformer", "FlaxRoFormerForTokenClassification"), ("xlm-roberta", "FlaxXLMRobertaForTokenClassification"), ] ) SCREAMING_SNAKE_CASE__ = OrderedDict( [ # Model for Multiple Choice mapping ("albert", "FlaxAlbertForMultipleChoice"), ("bert", "FlaxBertForMultipleChoice"), ("big_bird", "FlaxBigBirdForMultipleChoice"), ("distilbert", "FlaxDistilBertForMultipleChoice"), ("electra", "FlaxElectraForMultipleChoice"), ("roberta", "FlaxRobertaForMultipleChoice"), ("roberta-prelayernorm", "FlaxRobertaPreLayerNormForMultipleChoice"), ("roformer", "FlaxRoFormerForMultipleChoice"), ("xlm-roberta", "FlaxXLMRobertaForMultipleChoice"), ] ) SCREAMING_SNAKE_CASE__ = OrderedDict( [ ("bert", "FlaxBertForNextSentencePrediction"), ] ) SCREAMING_SNAKE_CASE__ = OrderedDict( [ ("speech-encoder-decoder", "FlaxSpeechEncoderDecoderModel"), ("whisper", "FlaxWhisperForConditionalGeneration"), ] ) SCREAMING_SNAKE_CASE__ = OrderedDict( [ ("whisper", "FlaxWhisperForAudioClassification"), ] ) SCREAMING_SNAKE_CASE__ = _LazyAutoMapping(CONFIG_MAPPING_NAMES, FLAX_MODEL_MAPPING_NAMES) SCREAMING_SNAKE_CASE__ = _LazyAutoMapping(CONFIG_MAPPING_NAMES, FLAX_MODEL_FOR_PRETRAINING_MAPPING_NAMES) SCREAMING_SNAKE_CASE__ = _LazyAutoMapping(CONFIG_MAPPING_NAMES, FLAX_MODEL_FOR_MASKED_LM_MAPPING_NAMES) SCREAMING_SNAKE_CASE__ = _LazyAutoMapping( CONFIG_MAPPING_NAMES, FLAX_MODEL_FOR_SEQ_TO_SEQ_CAUSAL_LM_MAPPING_NAMES ) SCREAMING_SNAKE_CASE__ = _LazyAutoMapping( CONFIG_MAPPING_NAMES, FLAX_MODEL_FOR_IMAGE_CLASSIFICATION_MAPPING_NAMES ) SCREAMING_SNAKE_CASE__ = _LazyAutoMapping(CONFIG_MAPPING_NAMES, FLAX_MODEL_FOR_VISION_2_SEQ_MAPPING_NAMES) SCREAMING_SNAKE_CASE__ = _LazyAutoMapping(CONFIG_MAPPING_NAMES, FLAX_MODEL_FOR_CAUSAL_LM_MAPPING_NAMES) SCREAMING_SNAKE_CASE__ = _LazyAutoMapping( CONFIG_MAPPING_NAMES, FLAX_MODEL_FOR_SEQUENCE_CLASSIFICATION_MAPPING_NAMES ) SCREAMING_SNAKE_CASE__ = _LazyAutoMapping( CONFIG_MAPPING_NAMES, FLAX_MODEL_FOR_QUESTION_ANSWERING_MAPPING_NAMES ) SCREAMING_SNAKE_CASE__ = _LazyAutoMapping( CONFIG_MAPPING_NAMES, FLAX_MODEL_FOR_TOKEN_CLASSIFICATION_MAPPING_NAMES ) SCREAMING_SNAKE_CASE__ = _LazyAutoMapping( CONFIG_MAPPING_NAMES, FLAX_MODEL_FOR_MULTIPLE_CHOICE_MAPPING_NAMES ) SCREAMING_SNAKE_CASE__ = _LazyAutoMapping( CONFIG_MAPPING_NAMES, FLAX_MODEL_FOR_NEXT_SENTENCE_PREDICTION_MAPPING_NAMES ) SCREAMING_SNAKE_CASE__ = _LazyAutoMapping( CONFIG_MAPPING_NAMES, FLAX_MODEL_FOR_SPEECH_SEQ_2_SEQ_MAPPING_NAMES ) SCREAMING_SNAKE_CASE__ = _LazyAutoMapping( CONFIG_MAPPING_NAMES, FLAX_MODEL_FOR_AUDIO_CLASSIFICATION_MAPPING_NAMES ) class lowercase ( _BaseAutoModelClass ): _SCREAMING_SNAKE_CASE = FLAX_MODEL_MAPPING SCREAMING_SNAKE_CASE__ = auto_class_update(FlaxAutoModel) class lowercase ( _BaseAutoModelClass ): _SCREAMING_SNAKE_CASE = FLAX_MODEL_FOR_PRETRAINING_MAPPING SCREAMING_SNAKE_CASE__ = auto_class_update(FlaxAutoModelForPreTraining, head_doc="pretraining") class lowercase ( _BaseAutoModelClass ): _SCREAMING_SNAKE_CASE = FLAX_MODEL_FOR_CAUSAL_LM_MAPPING SCREAMING_SNAKE_CASE__ = auto_class_update(FlaxAutoModelForCausalLM, head_doc="causal language modeling") class lowercase ( _BaseAutoModelClass ): _SCREAMING_SNAKE_CASE = FLAX_MODEL_FOR_MASKED_LM_MAPPING SCREAMING_SNAKE_CASE__ = auto_class_update(FlaxAutoModelForMaskedLM, head_doc="masked language modeling") class lowercase ( _BaseAutoModelClass ): _SCREAMING_SNAKE_CASE = FLAX_MODEL_FOR_SEQ_TO_SEQ_CAUSAL_LM_MAPPING SCREAMING_SNAKE_CASE__ = auto_class_update( FlaxAutoModelForSeqaSeqLM, head_doc="sequence-to-sequence language modeling", checkpoint_for_example="t5-base" ) class lowercase ( _BaseAutoModelClass ): _SCREAMING_SNAKE_CASE = FLAX_MODEL_FOR_SEQUENCE_CLASSIFICATION_MAPPING SCREAMING_SNAKE_CASE__ = auto_class_update( FlaxAutoModelForSequenceClassification, head_doc="sequence classification" ) class lowercase ( _BaseAutoModelClass ): _SCREAMING_SNAKE_CASE = FLAX_MODEL_FOR_QUESTION_ANSWERING_MAPPING SCREAMING_SNAKE_CASE__ = auto_class_update(FlaxAutoModelForQuestionAnswering, head_doc="question answering") class lowercase ( _BaseAutoModelClass ): _SCREAMING_SNAKE_CASE = FLAX_MODEL_FOR_TOKEN_CLASSIFICATION_MAPPING SCREAMING_SNAKE_CASE__ = auto_class_update( FlaxAutoModelForTokenClassification, head_doc="token classification" ) class lowercase ( _BaseAutoModelClass ): _SCREAMING_SNAKE_CASE = FLAX_MODEL_FOR_MULTIPLE_CHOICE_MAPPING SCREAMING_SNAKE_CASE__ = auto_class_update(FlaxAutoModelForMultipleChoice, head_doc="multiple choice") class lowercase ( _BaseAutoModelClass ): _SCREAMING_SNAKE_CASE = FLAX_MODEL_FOR_NEXT_SENTENCE_PREDICTION_MAPPING SCREAMING_SNAKE_CASE__ = auto_class_update( FlaxAutoModelForNextSentencePrediction, head_doc="next sentence prediction" ) class lowercase ( _BaseAutoModelClass ): _SCREAMING_SNAKE_CASE = FLAX_MODEL_FOR_IMAGE_CLASSIFICATION_MAPPING SCREAMING_SNAKE_CASE__ = auto_class_update( FlaxAutoModelForImageClassification, head_doc="image classification" ) class lowercase ( _BaseAutoModelClass ): _SCREAMING_SNAKE_CASE = FLAX_MODEL_FOR_VISION_2_SEQ_MAPPING SCREAMING_SNAKE_CASE__ = auto_class_update(FlaxAutoModelForVisionaSeq, head_doc="vision-to-text modeling") class lowercase ( _BaseAutoModelClass ): _SCREAMING_SNAKE_CASE = FLAX_MODEL_FOR_SPEECH_SEQ_2_SEQ_MAPPING SCREAMING_SNAKE_CASE__ = auto_class_update( FlaxAutoModelForSpeechSeqaSeq, head_doc="sequence-to-sequence speech-to-text modeling" )
46
"""simple docstring""" import argparse import os import shutil import torch from emmental.modules import MagnitudeBinarizer, ThresholdBinarizer, TopKBinarizer def lowercase ( _SCREAMING_SNAKE_CASE : Any ): '''simple docstring''' _UpperCAmelCase = args.pruning_method _UpperCAmelCase = args.threshold _UpperCAmelCase = args.model_name_or_path.rstrip('''/''' ) _UpperCAmelCase = args.target_model_path print(f'Load fine-pruned model from {model_name_or_path}' ) _UpperCAmelCase = torch.load(os.path.join(_SCREAMING_SNAKE_CASE , '''pytorch_model.bin''' ) ) _UpperCAmelCase = {} for name, tensor in model.items(): if "embeddings" in name or "LayerNorm" in name or "pooler" in name: _UpperCAmelCase = tensor print(f'Copied layer {name}' ) elif "classifier" in name or "qa_output" in name: _UpperCAmelCase = tensor print(f'Copied layer {name}' ) elif "bias" in name: _UpperCAmelCase = tensor print(f'Copied layer {name}' ) else: if pruning_method == "magnitude": _UpperCAmelCase = MagnitudeBinarizer.apply(inputs=_SCREAMING_SNAKE_CASE , threshold=_SCREAMING_SNAKE_CASE ) _UpperCAmelCase = tensor * mask print(f'Pruned layer {name}' ) elif pruning_method == "topK": if "mask_scores" in name: continue _UpperCAmelCase = name[:-6] _UpperCAmelCase = model[f'{prefix_}mask_scores'] _UpperCAmelCase = TopKBinarizer.apply(_SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE ) _UpperCAmelCase = tensor * mask print(f'Pruned layer {name}' ) elif pruning_method == "sigmoied_threshold": if "mask_scores" in name: continue _UpperCAmelCase = name[:-6] _UpperCAmelCase = model[f'{prefix_}mask_scores'] _UpperCAmelCase = ThresholdBinarizer.apply(_SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE ) _UpperCAmelCase = tensor * mask print(f'Pruned layer {name}' ) elif pruning_method == "l0": if "mask_scores" in name: continue _UpperCAmelCase = name[:-6] _UpperCAmelCase = model[f'{prefix_}mask_scores'] _UpperCAmelCase , _UpperCAmelCase = -0.1, 1.1 _UpperCAmelCase = torch.sigmoid(_SCREAMING_SNAKE_CASE ) _UpperCAmelCase = s * (r - l) + l _UpperCAmelCase = s_bar.clamp(min=0.0 , max=1.0 ) _UpperCAmelCase = tensor * mask print(f'Pruned layer {name}' ) else: raise ValueError('''Unknown pruning method''' ) if target_model_path is None: _UpperCAmelCase = os.path.join( os.path.dirname(_SCREAMING_SNAKE_CASE ) , f'bertarized_{os.path.basename(_SCREAMING_SNAKE_CASE )}' ) if not os.path.isdir(_SCREAMING_SNAKE_CASE ): shutil.copytree(_SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE ) print(f'\nCreated folder {target_model_path}' ) torch.save(_SCREAMING_SNAKE_CASE , os.path.join(_SCREAMING_SNAKE_CASE , '''pytorch_model.bin''' ) ) print('''\nPruned model saved! See you later!''' ) if __name__ == "__main__": __A : Tuple = argparse.ArgumentParser() parser.add_argument( "--pruning_method", choices=["l0", "magnitude", "topK", "sigmoied_threshold"], type=str, required=True, help=( "Pruning Method (l0 = L0 regularization, magnitude = Magnitude pruning, topK = Movement pruning," " sigmoied_threshold = Soft movement pruning)" ), ) parser.add_argument( "--threshold", type=float, required=False, help=( "For `magnitude` and `topK`, it is the level of remaining weights (in %) in the fine-pruned model." "For `sigmoied_threshold`, it is the threshold \tau against which the (sigmoied) scores are compared." "Not needed for `l0`" ), ) parser.add_argument( "--model_name_or_path", type=str, required=True, help="Folder containing the model that was previously fine-pruned", ) parser.add_argument( "--target_model_path", default=None, type=str, required=False, help="Folder containing the model that was previously fine-pruned", ) __A : Optional[int] = parser.parse_args() main(args)
260
0
"""simple docstring""" import torch def _snake_case ( ) -> Union[str, Any]: if torch.cuda.is_available(): lowerCamelCase_ : int =torch.cuda.device_count() else: lowerCamelCase_ : List[str] =0 print(F"""Successfully ran on {num_gpus} GPUs""" ) if __name__ == "__main__": main()
209
"""simple docstring""" import inspect from typing import Optional, Union import numpy as np import PIL import torch from torch.nn import functional as F from torchvision import transforms from transformers import CLIPFeatureExtractor, CLIPModel, CLIPTextModel, CLIPTokenizer from diffusers import ( AutoencoderKL, DDIMScheduler, DiffusionPipeline, DPMSolverMultistepScheduler, LMSDiscreteScheduler, PNDMScheduler, UNetaDConditionModel, ) from diffusers.pipelines.stable_diffusion.pipeline_stable_diffusion import StableDiffusionPipelineOutput from diffusers.utils import ( PIL_INTERPOLATION, randn_tensor, ) def _snake_case ( lowerCamelCase__ : int , lowerCamelCase__ : Union[str, Any] , lowerCamelCase__ : int ) -> List[Any]: if isinstance(lowerCamelCase__ , torch.Tensor ): return image elif isinstance(lowerCamelCase__ , PIL.Image.Image ): lowerCamelCase_ : int =[image] if isinstance(image[0] , PIL.Image.Image ): lowerCamelCase_ : Tuple =[np.array(i.resize((w, h) , resample=PIL_INTERPOLATION["lanczos"] ) )[None, :] for i in image] lowerCamelCase_ : Any =np.concatenate(lowerCamelCase__ , axis=0 ) lowerCamelCase_ : Any =np.array(lowerCamelCase__ ).astype(np.floataa ) / 255.0 lowerCamelCase_ : Any =image.transpose(0 , 3 , 1 , 2 ) lowerCamelCase_ : Dict =2.0 * image - 1.0 lowerCamelCase_ : List[Any] =torch.from_numpy(lowerCamelCase__ ) elif isinstance(image[0] , torch.Tensor ): lowerCamelCase_ : Tuple =torch.cat(lowerCamelCase__ , dim=0 ) return image def _snake_case ( lowerCamelCase__ : Union[str, Any] , lowerCamelCase__ : Dict , lowerCamelCase__ : Dict , lowerCamelCase__ : Optional[int]=0.9995 ) -> List[str]: if not isinstance(lowerCamelCase__ , np.ndarray ): lowerCamelCase_ : List[Any] =True lowerCamelCase_ : Tuple =va.device lowerCamelCase_ : Optional[int] =va.cpu().numpy() lowerCamelCase_ : List[Any] =va.cpu().numpy() lowerCamelCase_ : Any =np.sum(va * va / (np.linalg.norm(lowerCamelCase__ ) * np.linalg.norm(lowerCamelCase__ )) ) if np.abs(lowerCamelCase__ ) > DOT_THRESHOLD: lowerCamelCase_ : str =(1 - t) * va + t * va else: lowerCamelCase_ : Tuple =np.arccos(lowerCamelCase__ ) lowerCamelCase_ : Any =np.sin(lowerCamelCase__ ) lowerCamelCase_ : Optional[int] =theta_a * t lowerCamelCase_ : Tuple =np.sin(lowerCamelCase__ ) lowerCamelCase_ : List[str] =np.sin(theta_a - theta_t ) / sin_theta_a lowerCamelCase_ : Union[str, Any] =sin_theta_t / sin_theta_a lowerCamelCase_ : str =sa * va + sa * va if inputs_are_torch: lowerCamelCase_ : Any =torch.from_numpy(lowerCamelCase__ ).to(lowerCamelCase__ ) return va def _snake_case ( lowerCamelCase__ : Tuple , lowerCamelCase__ : Dict ) -> List[str]: lowerCamelCase_ : List[Any] =F.normalize(lowerCamelCase__ , dim=-1 ) lowerCamelCase_ : Any =F.normalize(lowerCamelCase__ , dim=-1 ) return (x - y).norm(dim=-1 ).div(2 ).arcsin().pow(2 ).mul(2 ) def _snake_case ( lowerCamelCase__ : str , lowerCamelCase__ : List[Any] ) -> Any: for param in model.parameters(): lowerCamelCase_ : Any =value class lowercase__ ( snake_case__ ): def __init__( self : Union[str, Any] , snake_case__ : AutoencoderKL , snake_case__ : CLIPTextModel , snake_case__ : CLIPModel , snake_case__ : CLIPTokenizer , snake_case__ : UNetaDConditionModel , snake_case__ : Union[PNDMScheduler, LMSDiscreteScheduler, DDIMScheduler, DPMSolverMultistepScheduler] , snake_case__ : CLIPFeatureExtractor , snake_case__ : Optional[Any]=None , snake_case__ : int=None , snake_case__ : str=None , ): super().__init__() self.register_modules( vae=snake_case__ , text_encoder=snake_case__ , clip_model=snake_case__ , tokenizer=snake_case__ , unet=snake_case__ , scheduler=snake_case__ , feature_extractor=snake_case__ , coca_model=snake_case__ , coca_tokenizer=snake_case__ , coca_transform=snake_case__ , ) lowerCamelCase_ : Optional[Any] =( feature_extractor.size if isinstance(feature_extractor.size , snake_case__ ) else feature_extractor.size["shortest_edge"] ) lowerCamelCase_ : Union[str, Any] =transforms.Normalize(mean=feature_extractor.image_mean , std=feature_extractor.image_std ) set_requires_grad(self.text_encoder , snake_case__ ) set_requires_grad(self.clip_model , snake_case__ ) def UpperCAmelCase__ ( self : Optional[Any] , snake_case__ : Optional[Union[str, int]] = "auto" ): if slice_size == "auto": # half the attention head size is usually a good trade-off between # speed and memory lowerCamelCase_ : List[Any] =self.unet.config.attention_head_dim // 2 self.unet.set_attention_slice(snake_case__ ) def UpperCAmelCase__ ( self : int ): self.enable_attention_slicing(snake_case__ ) def UpperCAmelCase__ ( self : str ): set_requires_grad(self.vae , snake_case__ ) def UpperCAmelCase__ ( self : List[str] ): set_requires_grad(self.vae , snake_case__ ) def UpperCAmelCase__ ( self : List[Any] ): set_requires_grad(self.unet , snake_case__ ) def UpperCAmelCase__ ( self : Tuple ): set_requires_grad(self.unet , snake_case__ ) def UpperCAmelCase__ ( self : str , snake_case__ : Optional[int] , snake_case__ : List[str] , snake_case__ : Tuple ): # get the original timestep using init_timestep lowerCamelCase_ : Optional[int] =min(int(num_inference_steps * strength ) , snake_case__ ) lowerCamelCase_ : Dict =max(num_inference_steps - init_timestep , 0 ) lowerCamelCase_ : Union[str, Any] =self.scheduler.timesteps[t_start:] return timesteps, num_inference_steps - t_start def UpperCAmelCase__ ( self : List[Any] , snake_case__ : List[str] , snake_case__ : Union[str, Any] , snake_case__ : Optional[int] , snake_case__ : Optional[Any] , snake_case__ : List[Any] , snake_case__ : List[Any]=None ): if not isinstance(snake_case__ , torch.Tensor ): raise ValueError(F"""`image` has to be of type `torch.Tensor` but is {type(snake_case__ )}""" ) lowerCamelCase_ : List[str] =image.to(device=snake_case__ , dtype=snake_case__ ) if isinstance(snake_case__ , snake_case__ ): lowerCamelCase_ : Optional[int] =[ self.vae.encode(image[i : i + 1] ).latent_dist.sample(generator[i] ) for i in range(snake_case__ ) ] lowerCamelCase_ : Tuple =torch.cat(snake_case__ , dim=0 ) else: lowerCamelCase_ : Any =self.vae.encode(snake_case__ ).latent_dist.sample(snake_case__ ) # Hardcode 0.18215 because stable-diffusion-2-base has not self.vae.config.scaling_factor lowerCamelCase_ : Optional[Any] =0.18_215 * init_latents lowerCamelCase_ : Optional[Any] =init_latents.repeat_interleave(snake_case__ , dim=0 ) lowerCamelCase_ : List[Any] =randn_tensor(init_latents.shape , generator=snake_case__ , device=snake_case__ , dtype=snake_case__ ) # get latents lowerCamelCase_ : Optional[int] =self.scheduler.add_noise(snake_case__ , snake_case__ , snake_case__ ) lowerCamelCase_ : int =init_latents return latents def UpperCAmelCase__ ( self : Dict , snake_case__ : List[Any] ): lowerCamelCase_ : Optional[int] =self.coca_transform(snake_case__ ).unsqueeze(0 ) with torch.no_grad(), torch.cuda.amp.autocast(): lowerCamelCase_ : Tuple =self.coca_model.generate(transformed_image.to(device=self.device , dtype=self.coca_model.dtype ) ) lowerCamelCase_ : str =self.coca_tokenizer.decode(generated[0].cpu().numpy() ) return generated.split("<end_of_text>" )[0].replace("<start_of_text>" , "" ).rstrip(" .," ) def UpperCAmelCase__ ( self : str , snake_case__ : Union[str, Any] , snake_case__ : Optional[int] ): lowerCamelCase_ : int =self.feature_extractor.preprocess(snake_case__ ) lowerCamelCase_ : Tuple =torch.from_numpy(clip_image_input["pixel_values"][0] ).unsqueeze(0 ).to(self.device ).half() lowerCamelCase_ : Optional[int] =self.clip_model.get_image_features(snake_case__ ) lowerCamelCase_ : List[str] =image_embeddings_clip / image_embeddings_clip.norm(p=2 , dim=-1 , keepdim=snake_case__ ) lowerCamelCase_ : str =image_embeddings_clip.repeat_interleave(snake_case__ , dim=0 ) return image_embeddings_clip @torch.enable_grad() def UpperCAmelCase__ ( self : Union[str, Any] , snake_case__ : Dict , snake_case__ : Any , snake_case__ : str , snake_case__ : Optional[int] , snake_case__ : Union[str, Any] , snake_case__ : List[Any] , snake_case__ : str , ): lowerCamelCase_ : Any =latents.detach().requires_grad_() lowerCamelCase_ : Union[str, Any] =self.scheduler.scale_model_input(snake_case__ , snake_case__ ) # predict the noise residual lowerCamelCase_ : Any =self.unet(snake_case__ , snake_case__ , encoder_hidden_states=snake_case__ ).sample if isinstance(self.scheduler , (PNDMScheduler, DDIMScheduler, DPMSolverMultistepScheduler) ): lowerCamelCase_ : Any =self.scheduler.alphas_cumprod[timestep] lowerCamelCase_ : str =1 - alpha_prod_t # compute predicted original sample from predicted noise also called # "predicted x_0" of formula (12) from https://arxiv.org/pdf/2010.02502.pdf lowerCamelCase_ : Dict =(latents - beta_prod_t ** 0.5 * noise_pred) / alpha_prod_t ** 0.5 lowerCamelCase_ : Tuple =torch.sqrt(snake_case__ ) lowerCamelCase_ : Optional[int] =pred_original_sample * (fac) + latents * (1 - fac) elif isinstance(self.scheduler , snake_case__ ): lowerCamelCase_ : Optional[Any] =self.scheduler.sigmas[index] lowerCamelCase_ : Optional[int] =latents - sigma * noise_pred else: raise ValueError(F"""scheduler type {type(self.scheduler )} not supported""" ) # Hardcode 0.18215 because stable-diffusion-2-base has not self.vae.config.scaling_factor lowerCamelCase_ : str =1 / 0.18_215 * sample lowerCamelCase_ : List[Any] =self.vae.decode(snake_case__ ).sample lowerCamelCase_ : List[Any] =(image / 2 + 0.5).clamp(0 , 1 ) lowerCamelCase_ : Any =transforms.Resize(self.feature_extractor_size )(snake_case__ ) lowerCamelCase_ : Any =self.normalize(snake_case__ ).to(latents.dtype ) lowerCamelCase_ : Any =self.clip_model.get_image_features(snake_case__ ) lowerCamelCase_ : Optional[int] =image_embeddings_clip / image_embeddings_clip.norm(p=2 , dim=-1 , keepdim=snake_case__ ) lowerCamelCase_ : Union[str, Any] =spherical_dist_loss(snake_case__ , snake_case__ ).mean() * clip_guidance_scale lowerCamelCase_ : Union[str, Any] =-torch.autograd.grad(snake_case__ , snake_case__ )[0] if isinstance(self.scheduler , snake_case__ ): lowerCamelCase_ : int =latents.detach() + grads * (sigma**2) lowerCamelCase_ : Tuple =noise_pred_original else: lowerCamelCase_ : Union[str, Any] =noise_pred_original - torch.sqrt(snake_case__ ) * grads return noise_pred, latents @torch.no_grad() def __call__( self : Any , snake_case__ : Union[torch.FloatTensor, PIL.Image.Image] , snake_case__ : Union[torch.FloatTensor, PIL.Image.Image] , snake_case__ : Optional[str] = None , snake_case__ : Optional[str] = None , snake_case__ : Optional[int] = 512 , snake_case__ : Optional[int] = 512 , snake_case__ : float = 0.6 , snake_case__ : Optional[int] = 50 , snake_case__ : Optional[float] = 7.5 , snake_case__ : Optional[int] = 1 , snake_case__ : float = 0.0 , snake_case__ : Optional[float] = 100 , snake_case__ : Optional[torch.Generator] = None , snake_case__ : Optional[str] = "pil" , snake_case__ : bool = True , snake_case__ : float = 0.8 , snake_case__ : float = 0.1 , snake_case__ : float = 0.1 , ): if isinstance(snake_case__ , snake_case__ ) and len(snake_case__ ) != batch_size: raise ValueError(F"""You have passed {batch_size} batch_size, but only {len(snake_case__ )} generators.""" ) if height % 8 != 0 or width % 8 != 0: raise ValueError(F"""`height` and `width` have to be divisible by 8 but are {height} and {width}.""" ) if isinstance(snake_case__ , torch.Generator ) and batch_size > 1: lowerCamelCase_ : List[str] =[generator] + [None] * (batch_size - 1) lowerCamelCase_ : Any =[ ("model", self.coca_model is None), ("tokenizer", self.coca_tokenizer is None), ("transform", self.coca_transform is None), ] lowerCamelCase_ : Optional[Any] =[x[0] for x in coca_is_none if x[1]] lowerCamelCase_ : Any =", ".join(snake_case__ ) # generate prompts with coca model if prompt is None if content_prompt is None: if len(snake_case__ ): raise ValueError( F"""Content prompt is None and CoCa [{coca_is_none_str}] is None.""" F"""Set prompt or pass Coca [{coca_is_none_str}] to DiffusionPipeline.""" ) lowerCamelCase_ : Union[str, Any] =self.get_image_description(snake_case__ ) if style_prompt is None: if len(snake_case__ ): raise ValueError( F"""Style prompt is None and CoCa [{coca_is_none_str}] is None.""" F""" Set prompt or pass Coca [{coca_is_none_str}] to DiffusionPipeline.""" ) lowerCamelCase_ : str =self.get_image_description(snake_case__ ) # get prompt text embeddings for content and style lowerCamelCase_ : str =self.tokenizer( snake_case__ , padding="max_length" , max_length=self.tokenizer.model_max_length , truncation=snake_case__ , return_tensors="pt" , ) lowerCamelCase_ : List[str] =self.text_encoder(content_text_input.input_ids.to(self.device ) )[0] lowerCamelCase_ : int =self.tokenizer( snake_case__ , padding="max_length" , max_length=self.tokenizer.model_max_length , truncation=snake_case__ , return_tensors="pt" , ) lowerCamelCase_ : Optional[int] =self.text_encoder(style_text_input.input_ids.to(self.device ) )[0] lowerCamelCase_ : Dict =slerp(snake_case__ , snake_case__ , snake_case__ ) # duplicate text embeddings for each generation per prompt lowerCamelCase_ : str =text_embeddings.repeat_interleave(snake_case__ , dim=0 ) # set timesteps lowerCamelCase_ : List[Any] ="offset" in set(inspect.signature(self.scheduler.set_timesteps ).parameters.keys() ) lowerCamelCase_ : List[Any] ={} if accepts_offset: lowerCamelCase_ : Optional[int] =1 self.scheduler.set_timesteps(snake_case__ , **snake_case__ ) # Some schedulers like PNDM have timesteps as arrays # It's more optimized to move all timesteps to correct device beforehand self.scheduler.timesteps.to(self.device ) lowerCamelCase_ , lowerCamelCase_ : Optional[int] =self.get_timesteps(snake_case__ , snake_case__ , self.device ) lowerCamelCase_ : Union[str, Any] =timesteps[:1].repeat(snake_case__ ) # Preprocess image lowerCamelCase_ : str =preprocess(snake_case__ , snake_case__ , snake_case__ ) lowerCamelCase_ : int =self.prepare_latents( snake_case__ , snake_case__ , snake_case__ , text_embeddings.dtype , self.device , snake_case__ ) lowerCamelCase_ : Dict =preprocess(snake_case__ , snake_case__ , snake_case__ ) lowerCamelCase_ : Tuple =self.prepare_latents( snake_case__ , snake_case__ , snake_case__ , text_embeddings.dtype , self.device , snake_case__ ) lowerCamelCase_ : int =slerp(snake_case__ , snake_case__ , snake_case__ ) if clip_guidance_scale > 0: lowerCamelCase_ : List[Any] =self.get_clip_image_embeddings(snake_case__ , snake_case__ ) lowerCamelCase_ : Optional[Any] =self.get_clip_image_embeddings(snake_case__ , snake_case__ ) lowerCamelCase_ : List[Any] =slerp( snake_case__ , snake_case__ , snake_case__ ) # here `guidance_scale` is defined analog to the guidance weight `w` of equation (2) # of the Imagen paper: https://arxiv.org/pdf/2205.11487.pdf . `guidance_scale = 1` # corresponds to doing no classifier free guidance. lowerCamelCase_ : str =guidance_scale > 1.0 # get unconditional embeddings for classifier free guidance if do_classifier_free_guidance: lowerCamelCase_ : List[Any] =content_text_input.input_ids.shape[-1] lowerCamelCase_ : Union[str, Any] =self.tokenizer([""] , padding="max_length" , max_length=snake_case__ , return_tensors="pt" ) lowerCamelCase_ : List[Any] =self.text_encoder(uncond_input.input_ids.to(self.device ) )[0] # duplicate unconditional embeddings for each generation per prompt lowerCamelCase_ : List[str] =uncond_embeddings.repeat_interleave(snake_case__ , dim=0 ) # For classifier free guidance, we need to do two forward passes. # Here we concatenate the unconditional and text embeddings into a single batch # to avoid doing two forward passes lowerCamelCase_ : List[str] =torch.cat([uncond_embeddings, text_embeddings] ) # get the initial random noise unless the user supplied it # Unlike in other pipelines, latents need to be generated in the target device # for 1-to-1 results reproducibility with the CompVis implementation. # However this currently doesn't work in `mps`. lowerCamelCase_ : Optional[Any] =(batch_size, self.unet.config.in_channels, height // 8, width // 8) lowerCamelCase_ : List[str] =text_embeddings.dtype if latents is None: if self.device.type == "mps": # randn does not work reproducibly on mps lowerCamelCase_ : Union[str, Any] =torch.randn(snake_case__ , generator=snake_case__ , device="cpu" , dtype=snake_case__ ).to( self.device ) else: lowerCamelCase_ : Tuple =torch.randn(snake_case__ , generator=snake_case__ , device=self.device , dtype=snake_case__ ) else: if latents.shape != latents_shape: raise ValueError(F"""Unexpected latents shape, got {latents.shape}, expected {latents_shape}""" ) lowerCamelCase_ : Any =latents.to(self.device ) # scale the initial noise by the standard deviation required by the scheduler lowerCamelCase_ : Optional[int] =latents * self.scheduler.init_noise_sigma # prepare extra kwargs for the scheduler step, since not all schedulers have the same signature # eta (η) is only used with the DDIMScheduler, it will be ignored for other schedulers. # eta corresponds to η in DDIM paper: https://arxiv.org/abs/2010.02502 # and should be between [0, 1] lowerCamelCase_ : List[Any] ="eta" in set(inspect.signature(self.scheduler.step ).parameters.keys() ) lowerCamelCase_ : Dict ={} if accepts_eta: lowerCamelCase_ : Tuple =eta # check if the scheduler accepts generator lowerCamelCase_ : Tuple ="generator" in set(inspect.signature(self.scheduler.step ).parameters.keys() ) if accepts_generator: lowerCamelCase_ : Optional[int] =generator with self.progress_bar(total=snake_case__ ): for i, t in enumerate(snake_case__ ): # expand the latents if we are doing classifier free guidance lowerCamelCase_ : List[str] =torch.cat([latents] * 2 ) if do_classifier_free_guidance else latents lowerCamelCase_ : Optional[Any] =self.scheduler.scale_model_input(snake_case__ , snake_case__ ) # predict the noise residual lowerCamelCase_ : Tuple =self.unet(snake_case__ , snake_case__ , encoder_hidden_states=snake_case__ ).sample # perform classifier free guidance if do_classifier_free_guidance: lowerCamelCase_ , lowerCamelCase_ : Dict =noise_pred.chunk(2 ) lowerCamelCase_ : Optional[Any] =noise_pred_uncond + guidance_scale * (noise_pred_text - noise_pred_uncond) # perform clip guidance if clip_guidance_scale > 0: lowerCamelCase_ : Optional[int] =( text_embeddings.chunk(2 )[1] if do_classifier_free_guidance else text_embeddings ) lowerCamelCase_ , lowerCamelCase_ : Union[str, Any] =self.cond_fn( snake_case__ , snake_case__ , snake_case__ , snake_case__ , snake_case__ , snake_case__ , snake_case__ , ) # compute the previous noisy sample x_t -> x_t-1 lowerCamelCase_ : Dict =self.scheduler.step(snake_case__ , snake_case__ , snake_case__ , **snake_case__ ).prev_sample # Hardcode 0.18215 because stable-diffusion-2-base has not self.vae.config.scaling_factor lowerCamelCase_ : str =1 / 0.18_215 * latents lowerCamelCase_ : List[str] =self.vae.decode(snake_case__ ).sample lowerCamelCase_ : str =(image / 2 + 0.5).clamp(0 , 1 ) lowerCamelCase_ : List[Any] =image.cpu().permute(0 , 2 , 3 , 1 ).numpy() if output_type == "pil": lowerCamelCase_ : List[str] =self.numpy_to_pil(snake_case__ ) if not return_dict: return (image, None) return StableDiffusionPipelineOutput(images=snake_case__ , nsfw_content_detected=snake_case__ )
209
1
class lowerCamelCase__ : def __init__(self , UpperCAmelCase , UpperCAmelCase=None , UpperCAmelCase=None ) -> Dict: _lowercase =data _lowercase =previous _lowercase =next_node def __str__(self ) -> str: return f"{self.data}" def __A (self ) -> int: return self.data def __A (self ) -> Tuple: return self.next def __A (self ) -> str: return self.previous class lowerCamelCase__ : def __init__(self , UpperCAmelCase ) -> int: _lowercase =head def __iter__(self ) -> Optional[Any]: return self def __A (self ) -> List[str]: if not self.current: raise StopIteration else: _lowercase =self.current.get_data() _lowercase =self.current.get_next() return value class lowerCamelCase__ : def __init__(self ) -> List[Any]: _lowercase =None # First node in list _lowercase =None # Last node in list def __str__(self ) -> Any: _lowercase =self.head _lowercase =[] while current is not None: nodes.append(current.get_data() ) _lowercase =current.get_next() return " ".join(str(lowerCamelCase__ ) for node in nodes ) def __contains__(self , UpperCAmelCase ) -> Tuple: _lowercase =self.head while current: if current.get_data() == value: return True _lowercase =current.get_next() return False def __iter__(self ) -> Any: return LinkedListIterator(self.head ) def __A (self ) -> str: if self.head: return self.head.get_data() return None def __A (self ) -> Any: if self.tail: return self.tail.get_data() return None def __A (self , UpperCAmelCase ) -> None: if self.head is None: _lowercase =node _lowercase =node else: self.insert_before_node(self.head , lowerCamelCase__ ) def __A (self , UpperCAmelCase ) -> None: if self.head is None: self.set_head(lowerCamelCase__ ) else: self.insert_after_node(self.tail , lowerCamelCase__ ) def __A (self , UpperCAmelCase ) -> None: _lowercase =Node(lowerCamelCase__ ) if self.head is None: self.set_head(lowerCamelCase__ ) else: self.set_tail(lowerCamelCase__ ) def __A (self , UpperCAmelCase , UpperCAmelCase ) -> None: _lowercase =node _lowercase =node.previous if node.get_previous() is None: _lowercase =node_to_insert else: _lowercase =node_to_insert _lowercase =node_to_insert def __A (self , UpperCAmelCase , UpperCAmelCase ) -> None: _lowercase =node _lowercase =node.next if node.get_next() is None: _lowercase =node_to_insert else: _lowercase =node_to_insert _lowercase =node_to_insert def __A (self , UpperCAmelCase , UpperCAmelCase ) -> None: _lowercase =1 _lowercase =Node(lowerCamelCase__ ) _lowercase =self.head while node: if current_position == position: self.insert_before_node(lowerCamelCase__ , lowerCamelCase__ ) return current_position += 1 _lowercase =node.next self.insert_after_node(self.tail , lowerCamelCase__ ) def __A (self , UpperCAmelCase ) -> Node: _lowercase =self.head while node: if node.get_data() == item: return node _lowercase =node.get_next() raise Exception('''Node not found''' ) def __A (self , UpperCAmelCase ) -> Tuple: if (node := self.get_node(lowerCamelCase__ )) is not None: if node == self.head: _lowercase =self.head.get_next() if node == self.tail: _lowercase =self.tail.get_previous() self.remove_node_pointers(lowerCamelCase__ ) @staticmethod def __A (UpperCAmelCase ) -> None: if node.get_next(): _lowercase =node.previous if node.get_previous(): _lowercase =node.next _lowercase =None _lowercase =None def __A (self ) -> Any: return self.head is None def UpperCAmelCase_ ( ) -> Dict: """simple docstring""" if __name__ == "__main__": import doctest doctest.testmod()
5
import json import os import re import sys import urllib.request import requests from bsa import BeautifulSoup __UpperCamelCase : Dict = { "User-Agent": "Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36" " (KHTML, like Gecko) Chrome/70.0.3538.102 Safari/537.36 Edge/18.19582" } def _a ( SCREAMING_SNAKE_CASE : str = "dhaka" , SCREAMING_SNAKE_CASE : int = 5 ): """simple docstring""" UpperCamelCase__ : Optional[int] = min(SCREAMING_SNAKE_CASE , 50 ) # Prevent abuse! UpperCamelCase__ : str = { '''q''': query, '''tbm''': '''isch''', '''hl''': '''en''', '''ijn''': '''0''', } UpperCamelCase__ : List[str] = requests.get('''https://www.google.com/search''' , params=SCREAMING_SNAKE_CASE , headers=SCREAMING_SNAKE_CASE ) UpperCamelCase__ : Any = BeautifulSoup(html.text , '''html.parser''' ) UpperCamelCase__ : Union[str, Any] = ''''''.join( re.findall(r'''AF_initDataCallback\(([^<]+)\);''' , str(soup.select('''script''' ) ) ) ) UpperCamelCase__ : Optional[Any] = json.dumps(SCREAMING_SNAKE_CASE ) UpperCamelCase__ : List[Any] = json.loads(SCREAMING_SNAKE_CASE ) UpperCamelCase__ : List[str] = re.findall( r'''\[\"GRID_STATE0\",null,\[\[1,\[0,\".*?\",(.*),\"All\",''' , SCREAMING_SNAKE_CASE , ) if not matched_google_image_data: return 0 UpperCamelCase__ : Optional[Any] = re.sub( r'''\[\"(https\:\/\/encrypted-tbn0\.gstatic\.com\/images\?.*?)\",\d+,\d+\]''' , '''''' , str(SCREAMING_SNAKE_CASE ) , ) UpperCamelCase__ : List[Any] = re.findall( r'''(?:\'|,),\[\"(https:|http.*?)\",\d+,\d+\]''' , SCREAMING_SNAKE_CASE , ) for index, fixed_full_res_image in enumerate(SCREAMING_SNAKE_CASE ): if index >= max_images: return index UpperCamelCase__ : Optional[int] = bytes(SCREAMING_SNAKE_CASE , '''ascii''' ).decode( '''unicode-escape''' ) UpperCamelCase__ : List[Any] = bytes(SCREAMING_SNAKE_CASE , '''ascii''' ).decode( '''unicode-escape''' ) UpperCamelCase__ : List[Any] = urllib.request.build_opener() UpperCamelCase__ : Optional[Any] = [ ( '''User-Agent''', '''Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36''' ''' (KHTML, like Gecko) Chrome/70.0.3538.102 Safari/537.36 Edge/18.19582''', ) ] urllib.request.install_opener(SCREAMING_SNAKE_CASE ) UpperCamelCase__ : Dict = F"query_{query.replace(' ' , '_' )}" if not os.path.exists(SCREAMING_SNAKE_CASE ): os.makedirs(SCREAMING_SNAKE_CASE ) urllib.request.urlretrieve( # noqa: S310 SCREAMING_SNAKE_CASE , F"{path_name}/original_size_img_{index}.jpg" ) return index if __name__ == "__main__": try: __UpperCamelCase : List[Any] = download_images_from_google_query(sys.argv[1]) print(f"{image_count} images were downloaded to disk.") except IndexError: print("Please provide a search term.") raise
146
0
"""simple docstring""" import unittest from diffusers import FlaxAutoencoderKL from diffusers.utils import is_flax_available from diffusers.utils.testing_utils import require_flax from .test_modeling_common_flax import FlaxModelTesterMixin if is_flax_available(): import jax @require_flax class _UpperCamelCase ( lowerCAmelCase__ ,unittest.TestCase ): '''simple docstring''' __UpperCAmelCase : List[Any] =FlaxAutoencoderKL @property def snake_case ( self ): __lowerCAmelCase = 4 __lowerCAmelCase = 3 __lowerCAmelCase = (32, 32) __lowerCAmelCase = jax.random.PRNGKey(0 ) __lowerCAmelCase = jax.random.uniform(__a , ((batch_size, num_channels) + sizes) ) return {"sample": image, "prng_key": prng_key} def snake_case ( self ): __lowerCAmelCase = { "block_out_channels": [32, 64], "in_channels": 3, "out_channels": 3, "down_block_types": ["DownEncoderBlock2D", "DownEncoderBlock2D"], "up_block_types": ["UpDecoderBlock2D", "UpDecoderBlock2D"], "latent_channels": 4, } __lowerCAmelCase = self.dummy_input return init_dict, inputs_dict
259
"""simple docstring""" import math from collections.abc import Iterator from itertools import takewhile def _lowerCamelCase ( _UpperCamelCase ): '''simple docstring''' if 1 < number < 4: # 2 and 3 are primes return True elif number < 2 or number % 2 == 0 or number % 3 == 0: # Negatives, 0, 1, all even numbers, all multiples of 3 are not primes return False # All primes number are in format of 6k +/- 1 for i in range(5 , int(math.sqrt(_UpperCamelCase ) + 1 ) , 6 ): if number % i == 0 or number % (i + 2) == 0: return False return True def _lowerCamelCase ( ): '''simple docstring''' __lowerCAmelCase = 2 while True: if is_prime(_UpperCamelCase ): yield num num += 1 def _lowerCamelCase ( _UpperCamelCase = 200_0000 ): '''simple docstring''' return sum(takewhile(lambda _UpperCamelCase : x < n , prime_generator() ) ) if __name__ == "__main__": print(f'''{solution() = }''')
259
1
from dataclasses import asdict, dataclass from typing import Optional from ...configuration_utils import PretrainedConfig from ...utils import logging __A : Optional[int] = logging.get_logger(__name__) # TODO Update this __A : Optional[Any] = { """facebook/esm-1b""": """https://huggingface.co/facebook/esm-1b/resolve/main/config.json""", # See all ESM models at https://huggingface.co/models?filter=esm } class __A ( UpperCamelCase__ ): lowerCAmelCase_ : List[Any] = """esm""" def __init__( self : Optional[Any] , UpperCAmelCase_ : str=None , UpperCAmelCase_ : Union[str, Any]=None , UpperCAmelCase_ : Union[str, Any]=None , UpperCAmelCase_ : Union[str, Any]=768 , UpperCAmelCase_ : Dict=12 , UpperCAmelCase_ : Tuple=12 , UpperCAmelCase_ : Union[str, Any]=3072 , UpperCAmelCase_ : Dict=0.1 , UpperCAmelCase_ : Optional[int]=0.1 , UpperCAmelCase_ : int=1026 , UpperCAmelCase_ : Dict=0.02 , UpperCAmelCase_ : Optional[int]=1E-12 , UpperCAmelCase_ : List[str]="absolute" , UpperCAmelCase_ : int=True , UpperCAmelCase_ : List[Any]=None , UpperCAmelCase_ : Optional[Any]=False , UpperCAmelCase_ : Dict=False , UpperCAmelCase_ : List[str]=None , UpperCAmelCase_ : List[Any]=None , **UpperCAmelCase_ : List[str] , ): super().__init__(pad_token_id=lowerCAmelCase__ , mask_token_id=lowerCAmelCase__ , **lowerCAmelCase__ ) lowerCAmelCase : Union[str, Any] = vocab_size lowerCAmelCase : List[Any] = hidden_size lowerCAmelCase : Optional[Any] = num_hidden_layers lowerCAmelCase : str = num_attention_heads lowerCAmelCase : Tuple = intermediate_size lowerCAmelCase : List[Any] = hidden_dropout_prob lowerCAmelCase : Optional[int] = attention_probs_dropout_prob lowerCAmelCase : int = max_position_embeddings lowerCAmelCase : List[str] = initializer_range lowerCAmelCase : Optional[int] = layer_norm_eps lowerCAmelCase : Dict = position_embedding_type lowerCAmelCase : int = use_cache lowerCAmelCase : Tuple = emb_layer_norm_before lowerCAmelCase : Union[str, Any] = token_dropout lowerCAmelCase : List[Any] = is_folding_model if is_folding_model: if esmfold_config is None: logger.info('No esmfold_config supplied for folding model, using default values.' ) lowerCAmelCase : List[Any] = EsmFoldConfig() elif isinstance(lowerCAmelCase__ , lowerCAmelCase__ ): lowerCAmelCase : Any = EsmFoldConfig(**lowerCAmelCase__ ) lowerCAmelCase : Any = esmfold_config if vocab_list is None: logger.warning('No vocab_list supplied for folding model, assuming the ESM-2 vocabulary!' ) lowerCAmelCase : List[Any] = get_default_vocab_list() else: lowerCAmelCase : List[str] = vocab_list else: lowerCAmelCase : Any = None lowerCAmelCase : str = None if self.esmfold_config is not None and getattr(self.esmfold_config , 'use_esm_attn_map' , lowerCAmelCase__ ): raise ValueError('The HuggingFace port of ESMFold does not support use_esm_attn_map at this time!' ) def lowercase__ ( self : List[Any] ): lowerCAmelCase : Optional[Any] = super().to_dict() if isinstance(self.esmfold_config , lowerCAmelCase__ ): lowerCAmelCase : List[str] = self.esmfold_config.to_dict() return output @dataclass class __A : lowerCAmelCase_ : str = None lowerCAmelCase_ : bool = True lowerCAmelCase_ : bool = False lowerCAmelCase_ : bool = False lowerCAmelCase_ : bool = False lowerCAmelCase_ : float = 0 lowerCAmelCase_ : bool = True lowerCAmelCase_ : bool = False lowerCAmelCase_ : int = 128 lowerCAmelCase_ : "TrunkConfig" = None def lowercase__ ( self : Union[str, Any] ): if self.trunk is None: lowerCAmelCase : List[Any] = TrunkConfig() elif isinstance(self.trunk , lowerCAmelCase__ ): lowerCAmelCase : Dict = TrunkConfig(**self.trunk ) def lowercase__ ( self : int ): lowerCAmelCase : Any = asdict(self ) lowerCAmelCase : List[str] = self.trunk.to_dict() return output @dataclass class __A : lowerCAmelCase_ : int = 48 lowerCAmelCase_ : int = 1024 lowerCAmelCase_ : int = 128 lowerCAmelCase_ : int = 32 lowerCAmelCase_ : int = 32 lowerCAmelCase_ : int = 32 lowerCAmelCase_ : float = 0 lowerCAmelCase_ : float = 0 lowerCAmelCase_ : bool = False lowerCAmelCase_ : int = 4 lowerCAmelCase_ : Optional[int] = 128 lowerCAmelCase_ : "StructureModuleConfig" = None def lowercase__ ( self : List[Any] ): if self.structure_module is None: lowerCAmelCase : List[Any] = StructureModuleConfig() elif isinstance(self.structure_module , lowerCAmelCase__ ): lowerCAmelCase : Dict = StructureModuleConfig(**self.structure_module ) if self.max_recycles <= 0: raise ValueError(f"`max_recycles` should be positive, got {self.max_recycles}." ) if self.sequence_state_dim % self.sequence_state_dim != 0: raise ValueError( '`sequence_state_dim` should be a round multiple of `sequence_state_dim`, got' f" {self.sequence_state_dim} and {self.sequence_state_dim}." ) if self.pairwise_state_dim % self.pairwise_state_dim != 0: raise ValueError( '`pairwise_state_dim` should be a round multiple of `pairwise_state_dim`, got' f" {self.pairwise_state_dim} and {self.pairwise_state_dim}." ) lowerCAmelCase : int = self.sequence_state_dim // self.sequence_head_width lowerCAmelCase : Dict = self.pairwise_state_dim // self.pairwise_head_width if self.sequence_state_dim != sequence_num_heads * self.sequence_head_width: raise ValueError( '`sequence_state_dim` should be equal to `sequence_num_heads * sequence_head_width, got' f" {self.sequence_state_dim} != {sequence_num_heads} * {self.sequence_head_width}." ) if self.pairwise_state_dim != pairwise_num_heads * self.pairwise_head_width: raise ValueError( '`pairwise_state_dim` should be equal to `pairwise_num_heads * pairwise_head_width, got' f" {self.pairwise_state_dim} != {pairwise_num_heads} * {self.pairwise_head_width}." ) if self.pairwise_state_dim % 2 != 0: raise ValueError(f"`pairwise_state_dim` should be even, got {self.pairwise_state_dim}." ) if self.dropout >= 0.4: raise ValueError(f"`dropout` should not be greater than 0.4, got {self.dropout}." ) def lowercase__ ( self : Optional[Any] ): lowerCAmelCase : str = asdict(self ) lowerCAmelCase : List[str] = self.structure_module.to_dict() return output @dataclass class __A : lowerCAmelCase_ : int = 384 lowerCAmelCase_ : int = 128 lowerCAmelCase_ : int = 16 lowerCAmelCase_ : int = 128 lowerCAmelCase_ : int = 12 lowerCAmelCase_ : int = 4 lowerCAmelCase_ : int = 8 lowerCAmelCase_ : float = 0.1 lowerCAmelCase_ : int = 8 lowerCAmelCase_ : int = 1 lowerCAmelCase_ : int = 2 lowerCAmelCase_ : int = 7 lowerCAmelCase_ : int = 10 lowerCAmelCase_ : float = 1E-8 lowerCAmelCase_ : float = 1E5 def lowercase__ ( self : Any ): return asdict(self ) def SCREAMING_SNAKE_CASE__ ( ) -> Dict: '''simple docstring''' return ( "<cls>", "<pad>", "<eos>", "<unk>", "L", "A", "G", "V", "S", "E", "R", "T", "I", "D", "P", "K", "Q", "N", "F", "Y", "M", "H", "W", "C", "X", "B", "U", "Z", "O", ".", "-", "<null_1>", "<mask>", )
138
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 MobileViTImageProcessor class __lowerCAmelCase ( unittest.TestCase): def __init__( self , lowerCAmelCase__ , lowerCAmelCase__=7 , lowerCAmelCase__=3 , lowerCAmelCase__=1_8 , lowerCAmelCase__=3_0 , lowerCAmelCase__=4_0_0 , lowerCAmelCase__=True , lowerCAmelCase__=None , lowerCAmelCase__=True , lowerCAmelCase__=None , lowerCAmelCase__=True , ) -> Optional[Any]: '''simple docstring''' a__ : Union[str, Any] =size if size is not None else {"shortest_edge": 2_0} a__ : List[str] =crop_size if crop_size is not None else {"height": 1_8, "width": 1_8} a__ : Tuple =parent a__ : Union[str, Any] =batch_size a__ : List[str] =num_channels a__ : List[Any] =image_size a__ : str =min_resolution a__ : Optional[int] =max_resolution a__ : Tuple =do_resize a__ : Union[str, Any] =size a__ : List[Any] =do_center_crop a__ : List[str] =crop_size a__ : Optional[int] =do_flip_channel_order def _lowercase ( self ) -> Optional[int]: '''simple docstring''' return { "do_resize": self.do_resize, "size": self.size, "do_center_crop": self.do_center_crop, "crop_size": self.crop_size, "do_flip_channel_order": self.do_flip_channel_order, } @require_torch @require_vision class __lowerCAmelCase ( UpperCamelCase__ , unittest.TestCase): _lowercase : int = MobileViTImageProcessor if is_vision_available() else None def _lowercase ( self ) -> Optional[Any]: '''simple docstring''' a__ : Tuple =MobileViTImageProcessingTester(self ) @property def _lowercase ( self ) -> List[str]: '''simple docstring''' return self.image_processor_tester.prepare_image_processor_dict() def _lowercase ( self ) -> List[str]: '''simple docstring''' a__ : str =self.image_processing_class(**self.image_processor_dict ) self.assertTrue(hasattr(lowerCAmelCase__ , "do_resize" ) ) self.assertTrue(hasattr(lowerCAmelCase__ , "size" ) ) self.assertTrue(hasattr(lowerCAmelCase__ , "do_center_crop" ) ) self.assertTrue(hasattr(lowerCAmelCase__ , "center_crop" ) ) self.assertTrue(hasattr(lowerCAmelCase__ , "do_flip_channel_order" ) ) def _lowercase ( self ) -> Optional[int]: '''simple docstring''' a__ : List[Any] =self.image_processing_class.from_dict(self.image_processor_dict ) self.assertEqual(image_processor.size , {"shortest_edge": 2_0} ) self.assertEqual(image_processor.crop_size , {"height": 1_8, "width": 1_8} ) a__ : Union[str, Any] =self.image_processing_class.from_dict(self.image_processor_dict , size=4_2 , crop_size=8_4 ) self.assertEqual(image_processor.size , {"shortest_edge": 4_2} ) self.assertEqual(image_processor.crop_size , {"height": 8_4, "width": 8_4} ) def _lowercase ( self ) -> Optional[Any]: '''simple docstring''' pass def _lowercase ( self ) -> Tuple: '''simple docstring''' a__ : List[str] =self.image_processing_class(**self.image_processor_dict ) # create random PIL images a__ : Any =prepare_image_inputs(self.image_processor_tester , equal_resolution=lowerCAmelCase__ ) for image in image_inputs: self.assertIsInstance(lowerCAmelCase__ , Image.Image ) # Test not batched input a__ : Tuple =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.crop_size["height"], self.image_processor_tester.crop_size["width"], ) , ) # Test batched a__ : List[Any] =image_processing(lowerCAmelCase__ , 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.crop_size["height"], self.image_processor_tester.crop_size["width"], ) , ) def _lowercase ( self ) -> Any: '''simple docstring''' a__ : Optional[Any] =self.image_processing_class(**self.image_processor_dict ) # create random numpy tensors a__ : List[Any] =prepare_image_inputs(self.image_processor_tester , equal_resolution=lowerCAmelCase__ , numpify=lowerCAmelCase__ ) for image in image_inputs: self.assertIsInstance(lowerCAmelCase__ , np.ndarray ) # Test not batched input a__ : Tuple =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.crop_size["height"], self.image_processor_tester.crop_size["width"], ) , ) # Test batched a__ : int =image_processing(lowerCAmelCase__ , 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.crop_size["height"], self.image_processor_tester.crop_size["width"], ) , ) def _lowercase ( self ) -> Optional[Any]: '''simple docstring''' a__ : int =self.image_processing_class(**self.image_processor_dict ) # create random PyTorch tensors a__ : Tuple =prepare_image_inputs(self.image_processor_tester , equal_resolution=lowerCAmelCase__ , torchify=lowerCAmelCase__ ) for image in image_inputs: self.assertIsInstance(lowerCAmelCase__ , torch.Tensor ) # Test not batched input a__ : List[str] =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.crop_size["height"], self.image_processor_tester.crop_size["width"], ) , ) # Test batched a__ : List[str] =image_processing(lowerCAmelCase__ , 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.crop_size["height"], self.image_processor_tester.crop_size["width"], ) , )
95
0
from typing import TYPE_CHECKING from ...utils import OptionalDependencyNotAvailable, _LazyModule, is_tokenizers_available, is_torch_available SCREAMING_SNAKE_CASE : Optional[int] = { "configuration_altclip": [ "ALTCLIP_PRETRAINED_CONFIG_ARCHIVE_MAP", "AltCLIPConfig", "AltCLIPTextConfig", "AltCLIPVisionConfig", ], "processing_altclip": ["AltCLIPProcessor"], } try: if not is_torch_available(): raise OptionalDependencyNotAvailable() except OptionalDependencyNotAvailable: pass else: SCREAMING_SNAKE_CASE : List[Any] = [ "ALTCLIP_PRETRAINED_MODEL_ARCHIVE_LIST", "AltCLIPPreTrainedModel", "AltCLIPModel", "AltCLIPTextModel", "AltCLIPVisionModel", ] if TYPE_CHECKING: from .configuration_altclip import ( ALTCLIP_PRETRAINED_CONFIG_ARCHIVE_MAP, AltCLIPConfig, AltCLIPTextConfig, AltCLIPVisionConfig, ) from .processing_altclip import AltCLIPProcessor try: if not is_torch_available(): raise OptionalDependencyNotAvailable() except OptionalDependencyNotAvailable: pass else: from .modeling_altclip import ( ALTCLIP_PRETRAINED_MODEL_ARCHIVE_LIST, AltCLIPModel, AltCLIPPreTrainedModel, AltCLIPTextModel, AltCLIPVisionModel, ) else: import sys SCREAMING_SNAKE_CASE : int = _LazyModule(__name__, globals()["__file__"], _import_structure, module_spec=__spec__)
360
import argparse import logging import sys from unittest.mock import patch import run_glue_deebert from transformers.testing_utils import TestCasePlus, get_gpu_count, require_torch_non_multi_gpu, slow logging.basicConfig(level=logging.DEBUG) SCREAMING_SNAKE_CASE : str = logging.getLogger() def UpperCamelCase_( ) -> Any: _lowercase : int = argparse.ArgumentParser() parser.add_argument('-f' ) _lowercase : Optional[Any] = parser.parse_args() return args.f class _lowerCamelCase( _a ): def UpperCamelCase ( self) -> None: """simple docstring""" _lowercase : List[Any] = logging.StreamHandler(sys.stdout) logger.addHandler(lowerCamelCase) def UpperCamelCase ( self, lowerCamelCase) -> List[str]: """simple docstring""" _lowercase : str = get_gpu_count() if n_gpu > 1: pass # XXX: doesn't quite work with n_gpu > 1 https://github.com/huggingface/transformers/issues/10560 # script = f"{self.examples_dir_str}/research_projects/deebert/run_glue_deebert.py" # distributed_args = f"-m torch.distributed.launch --nproc_per_node={n_gpu} {script}".split() # cmd = [sys.executable] + distributed_args + args # execute_subprocess_async(cmd, env=self.get_env()) # XXX: test the results - need to save them first into .json file else: args.insert(0, 'run_glue_deebert.py') with patch.object(lowerCamelCase, 'argv', lowerCamelCase): _lowercase : Optional[Any] = run_glue_deebert.main() for value in result.values(): self.assertGreaterEqual(lowerCamelCase, 0.6_6_6) @slow @require_torch_non_multi_gpu def UpperCamelCase ( self) -> Tuple: """simple docstring""" _lowercase : Union[str, Any] = '\n --model_type roberta\n --model_name_or_path roberta-base\n --task_name MRPC\n --do_train\n --do_eval\n --do_lower_case\n --data_dir ./tests/fixtures/tests_samples/MRPC/\n --max_seq_length 128\n --per_gpu_eval_batch_size=1\n --per_gpu_train_batch_size=8\n --learning_rate 2e-4\n --num_train_epochs 3\n --overwrite_output_dir\n --seed 42\n --output_dir ./examples/deebert/saved_models/roberta-base/MRPC/two_stage\n --plot_data_dir ./examples/deebert/results/\n --save_steps 0\n --overwrite_cache\n --eval_after_first_stage\n '.split() self.run_and_check(lowerCamelCase) _lowercase : Union[str, Any] = '\n --model_type roberta\n --model_name_or_path ./examples/deebert/saved_models/roberta-base/MRPC/two_stage\n --task_name MRPC\n --do_eval\n --do_lower_case\n --data_dir ./tests/fixtures/tests_samples/MRPC/\n --output_dir ./examples/deebert/saved_models/roberta-base/MRPC/two_stage\n --plot_data_dir ./examples/deebert/results/\n --max_seq_length 128\n --eval_each_highway\n --eval_highway\n --overwrite_cache\n --per_gpu_eval_batch_size=1\n '.split() self.run_and_check(lowerCamelCase) _lowercase : Union[str, Any] = '\n --model_type roberta\n --model_name_or_path ./examples/deebert/saved_models/roberta-base/MRPC/two_stage\n --task_name MRPC\n --do_eval\n --do_lower_case\n --data_dir ./tests/fixtures/tests_samples/MRPC/\n --output_dir ./examples/deebert/saved_models/roberta-base/MRPC/two_stage\n --plot_data_dir ./examples/deebert/results/\n --max_seq_length 128\n --early_exit_entropy 0.1\n --eval_highway\n --overwrite_cache\n --per_gpu_eval_batch_size=1\n '.split() self.run_and_check(lowerCamelCase)
84
0
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 transformers import DeiTConfig, DeiTForImageClassificationWithTeacher, DeiTImageProcessor from transformers.utils import logging logging.set_verbosity_info() _UpperCAmelCase : List[str] = logging.get_logger(__name__) def A ( lowercase , lowercase=False ) -> List[Any]: '''simple docstring''' UpperCamelCase = [] 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'''deit.encoder.layer.{i}.layernorm_before.weight''') ) rename_keys.append((f'''blocks.{i}.norm1.bias''', f'''deit.encoder.layer.{i}.layernorm_before.bias''') ) rename_keys.append((f'''blocks.{i}.attn.proj.weight''', f'''deit.encoder.layer.{i}.attention.output.dense.weight''') ) rename_keys.append((f'''blocks.{i}.attn.proj.bias''', f'''deit.encoder.layer.{i}.attention.output.dense.bias''') ) rename_keys.append((f'''blocks.{i}.norm2.weight''', f'''deit.encoder.layer.{i}.layernorm_after.weight''') ) rename_keys.append((f'''blocks.{i}.norm2.bias''', f'''deit.encoder.layer.{i}.layernorm_after.bias''') ) rename_keys.append((f'''blocks.{i}.mlp.fc1.weight''', f'''deit.encoder.layer.{i}.intermediate.dense.weight''') ) rename_keys.append((f'''blocks.{i}.mlp.fc1.bias''', f'''deit.encoder.layer.{i}.intermediate.dense.bias''') ) rename_keys.append((f'''blocks.{i}.mlp.fc2.weight''', f'''deit.encoder.layer.{i}.output.dense.weight''') ) rename_keys.append((f'''blocks.{i}.mlp.fc2.bias''', f'''deit.encoder.layer.{i}.output.dense.bias''') ) # projection layer + position embeddings rename_keys.extend( [ ('cls_token', 'deit.embeddings.cls_token'), ('dist_token', 'deit.embeddings.distillation_token'), ('patch_embed.proj.weight', 'deit.embeddings.patch_embeddings.projection.weight'), ('patch_embed.proj.bias', 'deit.embeddings.patch_embeddings.projection.bias'), ('pos_embed', 'deit.embeddings.position_embeddings'), ] ) 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 "deit" from all keys that start with "deit" UpperCamelCase = [(pair[0], pair[1][4:]) if pair[1].startswith('deit' ) else pair for pair in rename_keys] else: # layernorm + classification heads rename_keys.extend( [ ('norm.weight', 'deit.layernorm.weight'), ('norm.bias', 'deit.layernorm.bias'), ('head.weight', 'cls_classifier.weight'), ('head.bias', 'cls_classifier.bias'), ('head_dist.weight', 'distillation_classifier.weight'), ('head_dist.bias', 'distillation_classifier.bias'), ] ) return rename_keys def A ( lowercase , lowercase , lowercase=False ) -> Dict: '''simple docstring''' for i in range(config.num_hidden_layers ): if base_model: UpperCamelCase = '' else: UpperCamelCase = 'deit.' # read in weights + bias of input projection layer (in timm, this is a single matrix + bias) UpperCamelCase = state_dict.pop(f'''blocks.{i}.attn.qkv.weight''' ) UpperCamelCase = state_dict.pop(f'''blocks.{i}.attn.qkv.bias''' ) # next, add query, keys and values (in that order) to the state dict UpperCamelCase = in_proj_weight[ : config.hidden_size, : ] UpperCamelCase = in_proj_bias[: config.hidden_size] UpperCamelCase = in_proj_weight[ config.hidden_size : config.hidden_size * 2, : ] UpperCamelCase = in_proj_bias[ config.hidden_size : config.hidden_size * 2 ] UpperCamelCase = in_proj_weight[ -config.hidden_size :, : ] UpperCamelCase = in_proj_bias[-config.hidden_size :] def A ( lowercase , lowercase , lowercase ) -> Dict: '''simple docstring''' UpperCamelCase = dct.pop(lowercase ) UpperCamelCase = val def A ( ) -> int: '''simple docstring''' UpperCamelCase = 'http://images.cocodataset.org/val2017/000000039769.jpg' UpperCamelCase = Image.open(requests.get(lowercase , stream=lowercase ).raw ) return im @torch.no_grad() def A ( lowercase , lowercase ) -> Dict: '''simple docstring''' UpperCamelCase = DeiTConfig() # all deit models have fine-tuned heads UpperCamelCase = False # dataset (fine-tuned on ImageNet 2012), patch_size and image_size UpperCamelCase = 1_000 UpperCamelCase = 'huggingface/label-files' UpperCamelCase = 'imagenet-1k-id2label.json' UpperCamelCase = json.load(open(hf_hub_download(lowercase , lowercase , repo_type='dataset' ) , 'r' ) ) UpperCamelCase = {int(lowercase ): v for k, v in idalabel.items()} UpperCamelCase = idalabel UpperCamelCase = {v: k for k, v in idalabel.items()} UpperCamelCase = int(deit_name[-6:-4] ) UpperCamelCase = int(deit_name[-3:] ) # size of the architecture if deit_name[9:].startswith('tiny' ): UpperCamelCase = 192 UpperCamelCase = 768 UpperCamelCase = 12 UpperCamelCase = 3 elif deit_name[9:].startswith('small' ): UpperCamelCase = 384 UpperCamelCase = 1_536 UpperCamelCase = 12 UpperCamelCase = 6 if deit_name[9:].startswith('base' ): pass elif deit_name[4:].startswith('large' ): UpperCamelCase = 1_024 UpperCamelCase = 4_096 UpperCamelCase = 24 UpperCamelCase = 16 # load original model from timm UpperCamelCase = timm.create_model(lowercase , pretrained=lowercase ) timm_model.eval() # load state_dict of original model, remove and rename some keys UpperCamelCase = timm_model.state_dict() UpperCamelCase = create_rename_keys(lowercase , lowercase ) for src, dest in rename_keys: rename_key(lowercase , lowercase , lowercase ) read_in_q_k_v(lowercase , lowercase , lowercase ) # load HuggingFace model UpperCamelCase = DeiTForImageClassificationWithTeacher(lowercase ).eval() model.load_state_dict(lowercase ) # Check outputs on an image, prepared by DeiTImageProcessor UpperCamelCase = int( (256 / 224) * config.image_size ) # to maintain same ratio w.r.t. 224 images, see https://github.com/facebookresearch/deit/blob/ab5715372db8c6cad5740714b2216d55aeae052e/datasets.py#L103 UpperCamelCase = DeiTImageProcessor(size=lowercase , crop_size=config.image_size ) UpperCamelCase = image_processor(images=prepare_img() , return_tensors='pt' ) UpperCamelCase = encoding['pixel_values'] UpperCamelCase = model(lowercase ) UpperCamelCase = timm_model(lowercase ) assert timm_logits.shape == outputs.logits.shape assert torch.allclose(lowercase , outputs.logits , atol=1e-3 ) Path(lowercase ).mkdir(exist_ok=lowercase ) print(f'''Saving model {deit_name} to {pytorch_dump_folder_path}''' ) model.save_pretrained(lowercase ) print(f'''Saving image processor to {pytorch_dump_folder_path}''' ) image_processor.save_pretrained(lowercase ) if __name__ == "__main__": _UpperCAmelCase : Union[str, Any] = argparse.ArgumentParser() # Required parameters parser.add_argument( "--deit_name", default="vit_deit_base_distilled_patch16_224", type=str, help="Name of the DeiT 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." ) _UpperCAmelCase : str = parser.parse_args() convert_deit_checkpoint(args.deit_name, args.pytorch_dump_folder_path)
222
import os import re from shutil import copyfile from typing import Any, Dict, List, Optional, Tuple import sentencepiece as spm from ...tokenization_utils import AddedToken, PreTrainedTokenizer from ...utils import logging _UpperCAmelCase : List[Any] = logging.get_logger(__name__) _UpperCAmelCase : Dict = {"vocab_file": "spiece.model"} _UpperCAmelCase : Dict = { "vocab_file": { "google/bigbird-roberta-base": "https://huggingface.co/google/bigbird-roberta-base/resolve/main/spiece.model", "google/bigbird-roberta-large": ( "https://huggingface.co/google/bigbird-roberta-large/resolve/main/spiece.model" ), "google/bigbird-base-trivia-itc": ( "https://huggingface.co/google/bigbird-base-trivia-itc/resolve/main/spiece.model" ), } } _UpperCAmelCase : int = { "google/bigbird-roberta-base": 4_096, "google/bigbird-roberta-large": 4_096, "google/bigbird-base-trivia-itc": 4_096, } class lowercase ( _SCREAMING_SNAKE_CASE ): __lowercase : List[str] = VOCAB_FILES_NAMES __lowercase : int = PRETRAINED_VOCAB_FILES_MAP __lowercase : List[Any] = PRETRAINED_POSITIONAL_EMBEDDINGS_SIZES __lowercase : List[Any] = ["input_ids", "attention_mask"] __lowercase : List[int] = [] def __init__( self , A_ , A_="<unk>" , A_="<s>" , A_="</s>" , A_="<pad>" , A_="[SEP]" , A_="[MASK]" , A_="[CLS]" , A_ = None , **A_ , ) -> None: """simple docstring""" UpperCamelCase = AddedToken(A_ , lstrip=A_ , rstrip=A_ ) if isinstance(A_ , A_ ) else bos_token UpperCamelCase = AddedToken(A_ , lstrip=A_ , rstrip=A_ ) if isinstance(A_ , A_ ) else eos_token UpperCamelCase = AddedToken(A_ , lstrip=A_ , rstrip=A_ ) if isinstance(A_ , A_ ) else unk_token UpperCamelCase = AddedToken(A_ , lstrip=A_ , rstrip=A_ ) if isinstance(A_ , A_ ) else pad_token UpperCamelCase = AddedToken(A_ , lstrip=A_ , rstrip=A_ ) if isinstance(A_ , A_ ) else cls_token UpperCamelCase = AddedToken(A_ , lstrip=A_ , rstrip=A_ ) if isinstance(A_ , A_ ) else sep_token # Mask token behave like a normal word, i.e. include the space before it UpperCamelCase = AddedToken(A_ , lstrip=A_ , rstrip=A_ ) if isinstance(A_ , A_ ) else mask_token UpperCamelCase = {} if sp_model_kwargs is None else sp_model_kwargs super().__init__( bos_token=A_ , eos_token=A_ , unk_token=A_ , pad_token=A_ , sep_token=A_ , mask_token=A_ , cls_token=A_ , sp_model_kwargs=self.sp_model_kwargs , **A_ , ) UpperCamelCase = vocab_file UpperCamelCase = spm.SentencePieceProcessor(**self.sp_model_kwargs ) self.sp_model.Load(A_ ) @property def __UpperCamelCase ( self ) -> Dict: """simple docstring""" return self.sp_model.get_piece_size() def __UpperCamelCase ( self ) -> List[Any]: """simple docstring""" UpperCamelCase = {self.convert_ids_to_tokens(A_ ): i for i in range(self.vocab_size )} vocab.update(self.added_tokens_encoder ) return vocab def __getstate__( self ) -> Optional[int]: """simple docstring""" UpperCamelCase = self.__dict__.copy() UpperCamelCase = None return state def __setstate__( self , A_ ) -> Optional[int]: """simple docstring""" UpperCamelCase = d # for backward compatibility if not hasattr(self , 'sp_model_kwargs' ): UpperCamelCase = {} UpperCamelCase = spm.SentencePieceProcessor(**self.sp_model_kwargs ) self.sp_model.Load(self.vocab_file ) def __UpperCamelCase ( self , A_ ) -> List[str]: """simple docstring""" return self.sp_model.encode(A_ , out_type=A_ ) def __UpperCamelCase ( self , A_ ) -> Optional[Any]: """simple docstring""" return self.sp_model.piece_to_id(A_ ) def __UpperCamelCase ( self , A_ ) -> Tuple: """simple docstring""" UpperCamelCase = self.sp_model.IdToPiece(A_ ) return token def __UpperCamelCase ( self , A_ ) -> List[Any]: """simple docstring""" UpperCamelCase = [] UpperCamelCase = '' UpperCamelCase = False for token in tokens: # make sure that special tokens are not decoded using sentencepiece model if token in self.all_special_tokens: if not prev_is_special: out_string += " " out_string += self.sp_model.decode(A_ ) + token UpperCamelCase = True UpperCamelCase = [] else: current_sub_tokens.append(A_ ) UpperCamelCase = False out_string += self.sp_model.decode(A_ ) return out_string.strip() def __UpperCamelCase ( self , A_ , A_ = False , A_ = None , A_ = True , **A_ , ) -> str: """simple docstring""" UpperCamelCase = kwargs.pop('use_source_tokenizer' , A_ ) UpperCamelCase = self.convert_ids_to_tokens(A_ , skip_special_tokens=A_ ) # To avoid mixing byte-level and unicode for byte-level BPT # we need to build string separately for added tokens and byte-level tokens # cf. https://github.com/huggingface/transformers/issues/1133 UpperCamelCase = [] UpperCamelCase = [] for token in filtered_tokens: if skip_special_tokens and token in self.all_special_ids: continue if token in self.added_tokens_encoder: if current_sub_text: sub_texts.append(self.convert_tokens_to_string(A_ ) ) UpperCamelCase = [] sub_texts.append(A_ ) else: current_sub_text.append(A_ ) if current_sub_text: sub_texts.append(self.convert_tokens_to_string(A_ ) ) # Mimic the behavior of the Rust tokenizer: # No space before [MASK] and [SEP] if spaces_between_special_tokens: UpperCamelCase = re.sub(r' (\[(MASK|SEP)\])' , r'\1' , ' '.join(A_ ) ) else: UpperCamelCase = ''.join(A_ ) UpperCamelCase = ( clean_up_tokenization_spaces if clean_up_tokenization_spaces is not None else self.clean_up_tokenization_spaces ) if clean_up_tokenization_spaces: UpperCamelCase = self.clean_up_tokenization(A_ ) return clean_text else: return text def __UpperCamelCase ( self , A_ , A_ = None ) -> Tuple[str]: """simple docstring""" if not os.path.isdir(A_ ): logger.error(F'''Vocabulary path ({save_directory}) should be a directory''' ) return UpperCamelCase = os.path.join( A_ , (filename_prefix + '-' if filename_prefix else '') + VOCAB_FILES_NAMES['vocab_file'] ) if os.path.abspath(self.vocab_file ) != os.path.abspath(A_ ) and os.path.isfile(self.vocab_file ): copyfile(self.vocab_file , A_ ) elif not os.path.isfile(self.vocab_file ): with open(A_ , 'wb' ) as fi: UpperCamelCase = self.sp_model.serialized_model_proto() fi.write(A_ ) return (out_vocab_file,) def __UpperCamelCase ( self , A_ , A_ = None ) -> List[int]: """simple docstring""" if token_ids_a is None: return [self.cls_token_id] + token_ids_a + [self.sep_token_id] UpperCamelCase = [self.cls_token_id] UpperCamelCase = [self.sep_token_id] return cls + token_ids_a + sep + token_ids_a + sep def __UpperCamelCase ( self , A_ , A_ = None , A_ = False ) -> List[int]: """simple docstring""" 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 __UpperCamelCase ( self , A_ , A_ = None ) -> List[int]: """simple docstring""" UpperCamelCase = [self.sep_token_id] UpperCamelCase = [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]
222
1
from ...configuration_utils import PretrainedConfig from ...utils import logging _lowerCamelCase : Optional[Any] = logging.get_logger(__name__) _lowerCamelCase : Optional[Any] = {} class lowerCamelCase (__lowerCamelCase ): """simple docstring""" UpperCAmelCase_ = "llama" UpperCAmelCase_ = ["past_key_values"] def __init__( self : List[Any], _UpperCAmelCase : Dict=3_2_0_0_0, _UpperCAmelCase : List[str]=4_0_9_6, _UpperCAmelCase : Any=1_1_0_0_8, _UpperCAmelCase : List[Any]=3_2, _UpperCAmelCase : Dict=3_2, _UpperCAmelCase : List[str]=None, _UpperCAmelCase : Dict="silu", _UpperCAmelCase : Optional[int]=2_0_4_8, _UpperCAmelCase : Union[str, Any]=0.02, _UpperCAmelCase : Tuple=1E-6, _UpperCAmelCase : List[Any]=True, _UpperCAmelCase : List[str]=0, _UpperCAmelCase : Optional[int]=1, _UpperCAmelCase : Tuple=2, _UpperCAmelCase : Union[str, Any]=1, _UpperCAmelCase : List[Any]=False, _UpperCAmelCase : List[Any]=None, **_UpperCAmelCase : Dict, ) -> List[str]: """simple docstring""" SCREAMING_SNAKE_CASE__ : Optional[int] = vocab_size SCREAMING_SNAKE_CASE__ : str = max_position_embeddings SCREAMING_SNAKE_CASE__ : List[str] = hidden_size SCREAMING_SNAKE_CASE__ : Tuple = intermediate_size SCREAMING_SNAKE_CASE__ : Optional[int] = num_hidden_layers SCREAMING_SNAKE_CASE__ : List[Any] = num_attention_heads # for backward compatibility if num_key_value_heads is None: SCREAMING_SNAKE_CASE__ : List[Any] = num_attention_heads SCREAMING_SNAKE_CASE__ : List[Any] = num_key_value_heads SCREAMING_SNAKE_CASE__ : List[Any] = hidden_act SCREAMING_SNAKE_CASE__ : Optional[Any] = initializer_range SCREAMING_SNAKE_CASE__ : List[str] = rms_norm_eps SCREAMING_SNAKE_CASE__ : Tuple = pretraining_tp SCREAMING_SNAKE_CASE__ : str = use_cache SCREAMING_SNAKE_CASE__ : Dict = rope_scaling self._rope_scaling_validation() super().__init__( pad_token_id=_UpperCAmelCase, bos_token_id=_UpperCAmelCase, eos_token_id=_UpperCAmelCase, tie_word_embeddings=_UpperCAmelCase, **_UpperCAmelCase, ) def A_ ( self : Optional[int] ) -> List[Any]: """simple docstring""" if self.rope_scaling is None: return if not isinstance(self.rope_scaling, _UpperCAmelCase ) or len(self.rope_scaling ) != 2: raise ValueError( "`rope_scaling` must be a dictionary with with two fields, `name` and `factor`, " F'''got {self.rope_scaling}''' ) SCREAMING_SNAKE_CASE__ : Optional[int] = self.rope_scaling.get("type", _UpperCAmelCase ) SCREAMING_SNAKE_CASE__ : Any = self.rope_scaling.get("factor", _UpperCAmelCase ) if rope_scaling_type is None or rope_scaling_type not in ["linear", "dynamic"]: raise ValueError( F'''`rope_scaling`\'s name field must be one of [\'linear\', \'dynamic\'], got {rope_scaling_type}''' ) if rope_scaling_factor is None or not isinstance(_UpperCAmelCase, _UpperCAmelCase ) or rope_scaling_factor <= 1.0: raise ValueError(F'''`rope_scaling`\'s factor field must be an float > 1, got {rope_scaling_factor}''' )
191
import sys from typing import Tuple import numpy as np import torch from PIL import Image from torch import nn from transformers.image_utils import PILImageResampling from utils import img_tensorize class lowerCamelCase : """simple docstring""" def __init__( self : Dict, _UpperCAmelCase : Any, _UpperCAmelCase : List[Any]=sys.maxsize ) -> Optional[int]: """simple docstring""" SCREAMING_SNAKE_CASE__ : Optional[int] = "bilinear" SCREAMING_SNAKE_CASE__ : Optional[int] = max_size SCREAMING_SNAKE_CASE__ : Optional[int] = short_edge_length def __call__( self : Optional[int], _UpperCAmelCase : Union[str, Any] ) -> str: """simple docstring""" SCREAMING_SNAKE_CASE__ : List[Any] = [] for img in imgs: SCREAMING_SNAKE_CASE__ ,SCREAMING_SNAKE_CASE__ : Tuple = img.shape[:2] # later: provide list and randomly choose index for resize SCREAMING_SNAKE_CASE__ : List[str] = np.random.randint(self.short_edge_length[0], self.short_edge_length[1] + 1 ) if size == 0: return img SCREAMING_SNAKE_CASE__ : int = size * 1.0 / min(_UpperCAmelCase, _UpperCAmelCase ) if h < w: SCREAMING_SNAKE_CASE__ ,SCREAMING_SNAKE_CASE__ : Optional[int] = size, scale * w else: SCREAMING_SNAKE_CASE__ ,SCREAMING_SNAKE_CASE__ : List[str] = scale * h, size if max(_UpperCAmelCase, _UpperCAmelCase ) > self.max_size: SCREAMING_SNAKE_CASE__ : str = self.max_size * 1.0 / max(_UpperCAmelCase, _UpperCAmelCase ) SCREAMING_SNAKE_CASE__ : Any = newh * scale SCREAMING_SNAKE_CASE__ : List[str] = neww * scale SCREAMING_SNAKE_CASE__ : Any = int(neww + 0.5 ) SCREAMING_SNAKE_CASE__ : List[Any] = int(newh + 0.5 ) if img.dtype == np.uinta: SCREAMING_SNAKE_CASE__ : List[Any] = Image.fromarray(_UpperCAmelCase ) SCREAMING_SNAKE_CASE__ : Union[str, Any] = pil_image.resize((neww, newh), PILImageResampling.BILINEAR ) SCREAMING_SNAKE_CASE__ : List[str] = np.asarray(_UpperCAmelCase ) else: SCREAMING_SNAKE_CASE__ : str = img.permute(2, 0, 1 ).unsqueeze(0 ) # 3, 0, 1) # hw(c) -> nchw SCREAMING_SNAKE_CASE__ : Tuple = nn.functional.interpolate( _UpperCAmelCase, (newh, neww), mode=self.interp_method, align_corners=_UpperCAmelCase ).squeeze(0 ) img_augs.append(_UpperCAmelCase ) return img_augs class lowerCamelCase : """simple docstring""" def __init__( self : Dict, _UpperCAmelCase : Optional[Any] ) -> Optional[int]: """simple docstring""" SCREAMING_SNAKE_CASE__ : Any = ResizeShortestEdge([cfg.INPUT.MIN_SIZE_TEST, cfg.INPUT.MIN_SIZE_TEST], cfg.INPUT.MAX_SIZE_TEST ) SCREAMING_SNAKE_CASE__ : Any = cfg.INPUT.FORMAT SCREAMING_SNAKE_CASE__ : List[str] = cfg.SIZE_DIVISIBILITY SCREAMING_SNAKE_CASE__ : List[Any] = cfg.PAD_VALUE SCREAMING_SNAKE_CASE__ : Dict = cfg.INPUT.MAX_SIZE_TEST SCREAMING_SNAKE_CASE__ : Optional[int] = cfg.MODEL.DEVICE SCREAMING_SNAKE_CASE__ : int = torch.tensor(cfg.MODEL.PIXEL_STD ).to(self.device ).view(len(cfg.MODEL.PIXEL_STD ), 1, 1 ) SCREAMING_SNAKE_CASE__ : str = torch.tensor(cfg.MODEL.PIXEL_MEAN ).to(self.device ).view(len(cfg.MODEL.PIXEL_STD ), 1, 1 ) SCREAMING_SNAKE_CASE__ : Optional[Any] = lambda _UpperCAmelCase : (x - self.pixel_mean) / self.pixel_std def A_ ( self : str, _UpperCAmelCase : Union[str, Any] ) -> Union[str, Any]: """simple docstring""" SCREAMING_SNAKE_CASE__ : Tuple = tuple(max(_UpperCAmelCase ) for s in zip(*[img.shape for img in images] ) ) SCREAMING_SNAKE_CASE__ : Optional[Any] = [im.shape[-2:] for im in images] SCREAMING_SNAKE_CASE__ : Union[str, Any] = [ nn.functional.pad( _UpperCAmelCase, [0, max_size[-1] - size[1], 0, max_size[-2] - size[0]], value=self.pad_value, ) for size, im in zip(_UpperCAmelCase, _UpperCAmelCase ) ] return torch.stack(_UpperCAmelCase ), torch.tensor(_UpperCAmelCase ) def __call__( self : Any, _UpperCAmelCase : Dict, _UpperCAmelCase : List[str]=False ) -> Optional[Any]: """simple docstring""" with torch.no_grad(): if not isinstance(_UpperCAmelCase, _UpperCAmelCase ): SCREAMING_SNAKE_CASE__ : str = [images] if single_image: assert len(_UpperCAmelCase ) == 1 for i in range(len(_UpperCAmelCase ) ): if isinstance(images[i], torch.Tensor ): images.insert(_UpperCAmelCase, images.pop(_UpperCAmelCase ).to(self.device ).float() ) elif not isinstance(images[i], torch.Tensor ): images.insert( _UpperCAmelCase, torch.as_tensor(img_tensorize(images.pop(_UpperCAmelCase ), input_format=self.input_format ) ) .to(self.device ) .float(), ) # resize smallest edge SCREAMING_SNAKE_CASE__ : List[str] = torch.tensor([im.shape[:2] for im in images] ) SCREAMING_SNAKE_CASE__ : Tuple = self.aug(_UpperCAmelCase ) # transpose images and convert to torch tensors # images = [torch.as_tensor(i.astype("float32")).permute(2, 0, 1).to(self.device) for i in images] # now normalize before pad to avoid useless arithmetic SCREAMING_SNAKE_CASE__ : List[Any] = [self.normalizer(_UpperCAmelCase ) for x in images] # now pad them to do the following operations SCREAMING_SNAKE_CASE__ ,SCREAMING_SNAKE_CASE__ : List[str] = self.pad(_UpperCAmelCase ) # Normalize if self.size_divisibility > 0: raise NotImplementedError() # pad SCREAMING_SNAKE_CASE__ : Union[str, Any] = torch.true_divide(_UpperCAmelCase, _UpperCAmelCase ) if single_image: return images[0], sizes[0], scales_yx[0] else: return images, sizes, scales_yx def _a ( SCREAMING_SNAKE_CASE__ : int , SCREAMING_SNAKE_CASE__ : Dict ) -> Union[str, Any]: '''simple docstring''' boxes[:, 0::2] *= scale_yx[:, 1] boxes[:, 1::2] *= scale_yx[:, 0] return boxes def _a ( SCREAMING_SNAKE_CASE__ : List[Any] , SCREAMING_SNAKE_CASE__ : Tuple[int, int] ) -> List[Any]: '''simple docstring''' assert torch.isfinite(SCREAMING_SNAKE_CASE__ ).all(), "Box tensor contains infinite or NaN!" SCREAMING_SNAKE_CASE__ ,SCREAMING_SNAKE_CASE__ : int = box_size tensor[:, 0].clamp_(min=0 , max=SCREAMING_SNAKE_CASE__ ) tensor[:, 1].clamp_(min=0 , max=SCREAMING_SNAKE_CASE__ ) tensor[:, 2].clamp_(min=0 , max=SCREAMING_SNAKE_CASE__ ) tensor[:, 3].clamp_(min=0 , max=SCREAMING_SNAKE_CASE__ )
191
1
'''simple docstring''' import os import sys import unittest __snake_case = os.path.abspath(os.path.dirname(os.path.dirname(os.path.dirname(__file__)))) sys.path.append(os.path.join(git_repo_path, '''utils''')) import get_test_info # noqa: E402 from get_test_info import ( # noqa: E402 get_model_to_test_mapping, get_model_to_tester_mapping, get_test_to_tester_mapping, ) __snake_case = os.path.join('''tests''', '''models''', '''bert''', '''test_modeling_bert.py''') __snake_case = os.path.join('''tests''', '''models''', '''blip''', '''test_modeling_blip.py''') class lowercase ( unittest.TestCase ): """simple docstring""" def lowerCAmelCase__ ( self ): '''simple docstring''' UpperCamelCase__ :Optional[Any] = get_test_to_tester_mapping(_SCREAMING_SNAKE_CASE ) UpperCamelCase__ :Union[str, Any] = get_test_to_tester_mapping(_SCREAMING_SNAKE_CASE ) UpperCamelCase__ :Tuple = {'''BertModelTest''': '''BertModelTester'''} UpperCamelCase__ :Tuple = { '''BlipModelTest''': '''BlipModelTester''', '''BlipTextImageModelTest''': '''BlipTextImageModelsModelTester''', '''BlipTextModelTest''': '''BlipTextModelTester''', '''BlipTextRetrievalModelTest''': '''BlipTextRetrievalModelTester''', '''BlipVQAModelTest''': '''BlipVQAModelTester''', '''BlipVisionModelTest''': '''BlipVisionModelTester''', } self.assertEqual(get_test_info.to_json(_SCREAMING_SNAKE_CASE ) , _SCREAMING_SNAKE_CASE ) self.assertEqual(get_test_info.to_json(_SCREAMING_SNAKE_CASE ) , _SCREAMING_SNAKE_CASE ) def lowerCAmelCase__ ( self ): '''simple docstring''' UpperCamelCase__ :Any = get_model_to_test_mapping(_SCREAMING_SNAKE_CASE ) UpperCamelCase__ :Tuple = get_model_to_test_mapping(_SCREAMING_SNAKE_CASE ) UpperCamelCase__ :int = { '''BertForMaskedLM''': ['''BertModelTest'''], '''BertForMultipleChoice''': ['''BertModelTest'''], '''BertForNextSentencePrediction''': ['''BertModelTest'''], '''BertForPreTraining''': ['''BertModelTest'''], '''BertForQuestionAnswering''': ['''BertModelTest'''], '''BertForSequenceClassification''': ['''BertModelTest'''], '''BertForTokenClassification''': ['''BertModelTest'''], '''BertLMHeadModel''': ['''BertModelTest'''], '''BertModel''': ['''BertModelTest'''], } UpperCamelCase__ :Optional[int] = { '''BlipForConditionalGeneration''': ['''BlipTextImageModelTest'''], '''BlipForImageTextRetrieval''': ['''BlipTextRetrievalModelTest'''], '''BlipForQuestionAnswering''': ['''BlipVQAModelTest'''], '''BlipModel''': ['''BlipModelTest'''], '''BlipTextModel''': ['''BlipTextModelTest'''], '''BlipVisionModel''': ['''BlipVisionModelTest'''], } self.assertEqual(get_test_info.to_json(_SCREAMING_SNAKE_CASE ) , _SCREAMING_SNAKE_CASE ) self.assertEqual(get_test_info.to_json(_SCREAMING_SNAKE_CASE ) , _SCREAMING_SNAKE_CASE ) def lowerCAmelCase__ ( self ): '''simple docstring''' UpperCamelCase__ :List[str] = get_model_to_tester_mapping(_SCREAMING_SNAKE_CASE ) UpperCamelCase__ :int = get_model_to_tester_mapping(_SCREAMING_SNAKE_CASE ) UpperCamelCase__ :str = { '''BertForMaskedLM''': ['''BertModelTester'''], '''BertForMultipleChoice''': ['''BertModelTester'''], '''BertForNextSentencePrediction''': ['''BertModelTester'''], '''BertForPreTraining''': ['''BertModelTester'''], '''BertForQuestionAnswering''': ['''BertModelTester'''], '''BertForSequenceClassification''': ['''BertModelTester'''], '''BertForTokenClassification''': ['''BertModelTester'''], '''BertLMHeadModel''': ['''BertModelTester'''], '''BertModel''': ['''BertModelTester'''], } UpperCamelCase__ :Tuple = { '''BlipForConditionalGeneration''': ['''BlipTextImageModelsModelTester'''], '''BlipForImageTextRetrieval''': ['''BlipTextRetrievalModelTester'''], '''BlipForQuestionAnswering''': ['''BlipVQAModelTester'''], '''BlipModel''': ['''BlipModelTester'''], '''BlipTextModel''': ['''BlipTextModelTester'''], '''BlipVisionModel''': ['''BlipVisionModelTester'''], } self.assertEqual(get_test_info.to_json(_SCREAMING_SNAKE_CASE ) , _SCREAMING_SNAKE_CASE ) self.assertEqual(get_test_info.to_json(_SCREAMING_SNAKE_CASE ) , _SCREAMING_SNAKE_CASE )
97
from typing import TYPE_CHECKING from ...file_utils import _LazyModule, is_torch_available from ...utils import OptionalDependencyNotAvailable __A : List[Any] = { 'configuration_gpt_neox_japanese': ['GPT_NEOX_JAPANESE_PRETRAINED_CONFIG_ARCHIVE_MAP', 'GPTNeoXJapaneseConfig'], 'tokenization_gpt_neox_japanese': ['GPTNeoXJapaneseTokenizer'], } try: if not is_torch_available(): raise OptionalDependencyNotAvailable() except OptionalDependencyNotAvailable: pass else: __A : Union[str, Any] = [ 'GPT_NEOX_JAPANESE_PRETRAINED_MODEL_ARCHIVE_LIST', 'GPTNeoXJapaneseForCausalLM', 'GPTNeoXJapaneseLayer', 'GPTNeoXJapaneseModel', 'GPTNeoXJapanesePreTrainedModel', ] if TYPE_CHECKING: from .configuration_gpt_neox_japanese import GPT_NEOX_JAPANESE_PRETRAINED_CONFIG_ARCHIVE_MAP, GPTNeoXJapaneseConfig from .tokenization_gpt_neox_japanese import GPTNeoXJapaneseTokenizer try: if not is_torch_available(): raise OptionalDependencyNotAvailable() except OptionalDependencyNotAvailable: pass else: from .modeling_gpt_neox_japanese import ( GPT_NEOX_JAPANESE_PRETRAINED_MODEL_ARCHIVE_LIST, GPTNeoXJapaneseForCausalLM, GPTNeoXJapaneseLayer, GPTNeoXJapaneseModel, GPTNeoXJapanesePreTrainedModel, ) else: import sys __A : str = _LazyModule(__name__, globals()['__file__'], _import_structure, module_spec=__spec__)
154
0
'''simple docstring''' 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 __UpperCamelCase : 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] , ): '''simple docstring''' __a : Optional[int] = parent __a : Dict = batch_size __a : Dict = num_channels __a : str = is_training __a : Optional[int] = use_labels __a : Union[str, Any] = hidden_dropout_prob __a : Dict = attention_probs_dropout_prob __a : Optional[Any] = num_labels __a : Tuple = image_size __a : Optional[Any] = layer_depths __a : Dict = embed_dims def __UpperCAmelCase ( self ): '''simple docstring''' __a : Tuple = floats_tensor([self.batch_size, self.num_channels, self.image_size, self.image_size] ) __a : int = None if self.use_labels: __a : Optional[Any] = ids_tensor([self.batch_size] , self.num_labels ) __a : List[str] = self.get_config() return config, pixel_values, labels def __UpperCAmelCase ( self ): '''simple docstring''' 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 __UpperCAmelCase ( self , __a , __a , __a ): '''simple docstring''' __a : int = SwiftFormerModel(config=__a ) model.to(__a ) model.eval() __a : Optional[Any] = model(__a ) self.parent.assertEqual(result.last_hidden_state.shape , (self.batch_size, self.embed_dims[-1], 7, 7) ) def __UpperCAmelCase ( self , __a , __a , __a ): '''simple docstring''' __a : Tuple = self.num_labels __a : Optional[int] = 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 : int = SwiftFormerForImageClassification(__a ) model.to(__a ) model.eval() __a : Union[str, Any] = floats_tensor([self.batch_size, self.num_channels, self.image_size, self.image_size] ) __a : Optional[Any] = model(__a ) self.parent.assertEqual(result.logits.shape , (self.batch_size, self.num_labels) ) def __UpperCAmelCase ( self ): '''simple docstring''' (__a) : Tuple = self.prepare_config_and_inputs() __a : Dict = {'pixel_values': pixel_values} return config, inputs_dict @require_torch class __UpperCamelCase ( lowerCAmelCase_ , lowerCAmelCase_ , unittest.TestCase ): A_ = (SwiftFormerModel, SwiftFormerForImageClassification) if is_torch_available() else () A_ = ( {"feature-extraction": SwiftFormerModel, "image-classification": SwiftFormerForImageClassification} if is_torch_available() else {} ) A_ = False A_ = False A_ = False A_ = False A_ = False def __UpperCAmelCase ( self ): '''simple docstring''' __a : Optional[Any] = SwiftFormerModelTester(self ) __a : Dict = ConfigTester( self , config_class=__a , has_text_modality=__a , hidden_size=37 , num_attention_heads=12 , num_hidden_layers=12 , ) def __UpperCAmelCase ( self ): '''simple docstring''' self.config_tester.run_common_tests() @unittest.skip(reason='SwiftFormer does not use inputs_embeds' ) def __UpperCAmelCase ( self ): '''simple docstring''' pass def __UpperCAmelCase ( self ): '''simple docstring''' __a : Optional[int] = self.model_tester.prepare_config_and_inputs_for_common() for model_class in self.all_model_classes: __a : Optional[Any] = model_class(__a ) __a : int = model.get_output_embeddings() self.assertTrue(x is None or isinstance(__a , nn.Linear ) ) def __UpperCAmelCase ( self ): '''simple docstring''' __a : Dict = self.model_tester.prepare_config_and_inputs_for_common() for model_class in self.all_model_classes: __a : List[Any] = model_class(__a ) __a : List[str] = inspect.signature(model.forward ) # signature.parameters is an OrderedDict => so arg_names order is deterministic __a : Optional[int] = [*signature.parameters.keys()] __a : Optional[Any] = ['pixel_values'] self.assertListEqual(arg_names[:1] , __a ) def __UpperCAmelCase ( self ): '''simple docstring''' __a : Tuple = self.model_tester.prepare_config_and_inputs() self.model_tester.create_and_check_model(*__a ) def __UpperCAmelCase ( self ): '''simple docstring''' __a : int = self.model_tester.prepare_config_and_inputs() self.model_tester.create_and_check_for_image_classification(*__a ) @slow def __UpperCAmelCase ( self ): '''simple docstring''' for model_name in SWIFTFORMER_PRETRAINED_MODEL_ARCHIVE_LIST[:1]: __a : int = SwiftFormerModel.from_pretrained(__a ) self.assertIsNotNone(__a ) @unittest.skip(reason='SwiftFormer does not output attentions' ) def __UpperCAmelCase ( self ): '''simple docstring''' pass def __UpperCAmelCase ( self ): '''simple docstring''' def check_hidden_states_output(__a , __a , __a ): __a : Optional[Any] = model_class(__a ) model.to(__a ) model.eval() with torch.no_grad(): __a : Union[str, Any] = model(**self._prepare_for_class(__a , __a ) ) __a : Any = 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 : List[Any] = 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 : Union[str, Any] = True check_hidden_states_output(__a , __a , __a ) def __UpperCAmelCase ( self ): '''simple docstring''' def _config_zero_init(__a ): __a : List[Any] = 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-1_0 ) if isinstance(getattr(__a , __a , __a ) , __a ): __a : List[Any] = _config_zero_init(getattr(__a , __a ) ) setattr(__a , __a , __a ) return configs_no_init __a : Dict = self.model_tester.prepare_config_and_inputs_for_common() __a : Optional[Any] = _config_zero_init(__a ) for model_class in self.all_model_classes: __a : Tuple = 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 __UpperCAmelCase ( self ): '''simple docstring''' pass def lowerCamelCase (): __a : Dict = Image.open('./tests/fixtures/tests_samples/COCO/000000039769.png' ) return image @require_torch @require_vision class __UpperCamelCase ( unittest.TestCase ): @cached_property def __UpperCAmelCase ( self ): '''simple docstring''' return ViTImageProcessor.from_pretrained('MBZUAI/swiftformer-xs' ) if is_vision_available() else None @slow def __UpperCAmelCase ( self ): '''simple docstring''' __a : Tuple = SwiftFormerForImageClassification.from_pretrained('MBZUAI/swiftformer-xs' ).to(__a ) __a : Optional[Any] = self.default_image_processor __a : Tuple = prepare_img() __a : int = image_processor(images=__a , return_tensors='pt' ).to(__a ) # forward pass with torch.no_grad(): __a : Optional[Any] = model(**__a ) # verify the logits __a : Tuple = torch.Size((1, 1000) ) self.assertEqual(outputs.logits.shape , __a ) __a : int = torch.tensor([[-2.1_7_0_3E0_0, 2.1_1_0_7E0_0, -2.0_8_1_1E0_0]] ).to(__a ) self.assertTrue(torch.allclose(outputs.logits[0, :3] , __a , atol=1E-4 ) )
359
'''simple docstring''' from typing import TYPE_CHECKING from ...utils import ( OptionalDependencyNotAvailable, _LazyModule, is_flax_available, is_tf_available, is_tokenizers_available, is_torch_available, ) __lowercase : Tuple = { 'configuration_distilbert': [ 'DISTILBERT_PRETRAINED_CONFIG_ARCHIVE_MAP', 'DistilBertConfig', 'DistilBertOnnxConfig', ], 'tokenization_distilbert': ['DistilBertTokenizer'], } try: if not is_tokenizers_available(): raise OptionalDependencyNotAvailable() except OptionalDependencyNotAvailable: pass else: __lowercase : str = ['DistilBertTokenizerFast'] try: if not is_torch_available(): raise OptionalDependencyNotAvailable() except OptionalDependencyNotAvailable: pass else: __lowercase : Any = [ 'DISTILBERT_PRETRAINED_MODEL_ARCHIVE_LIST', 'DistilBertForMaskedLM', 'DistilBertForMultipleChoice', 'DistilBertForQuestionAnswering', 'DistilBertForSequenceClassification', 'DistilBertForTokenClassification', 'DistilBertModel', 'DistilBertPreTrainedModel', ] try: if not is_tf_available(): raise OptionalDependencyNotAvailable() except OptionalDependencyNotAvailable: pass else: __lowercase : List[str] = [ 'TF_DISTILBERT_PRETRAINED_MODEL_ARCHIVE_LIST', 'TFDistilBertForMaskedLM', 'TFDistilBertForMultipleChoice', 'TFDistilBertForQuestionAnswering', 'TFDistilBertForSequenceClassification', 'TFDistilBertForTokenClassification', 'TFDistilBertMainLayer', 'TFDistilBertModel', 'TFDistilBertPreTrainedModel', ] try: if not is_flax_available(): raise OptionalDependencyNotAvailable() except OptionalDependencyNotAvailable: pass else: __lowercase : List[str] = [ 'FlaxDistilBertForMaskedLM', 'FlaxDistilBertForMultipleChoice', 'FlaxDistilBertForQuestionAnswering', 'FlaxDistilBertForSequenceClassification', 'FlaxDistilBertForTokenClassification', 'FlaxDistilBertModel', 'FlaxDistilBertPreTrainedModel', ] if TYPE_CHECKING: from .configuration_distilbert import ( DISTILBERT_PRETRAINED_CONFIG_ARCHIVE_MAP, DistilBertConfig, DistilBertOnnxConfig, ) from .tokenization_distilbert import DistilBertTokenizer try: if not is_tokenizers_available(): raise OptionalDependencyNotAvailable() except OptionalDependencyNotAvailable: pass else: from .tokenization_distilbert_fast import DistilBertTokenizerFast try: if not is_torch_available(): raise OptionalDependencyNotAvailable() except OptionalDependencyNotAvailable: pass else: from .modeling_distilbert import ( DISTILBERT_PRETRAINED_MODEL_ARCHIVE_LIST, DistilBertForMaskedLM, DistilBertForMultipleChoice, DistilBertForQuestionAnswering, DistilBertForSequenceClassification, DistilBertForTokenClassification, DistilBertModel, DistilBertPreTrainedModel, ) try: if not is_tf_available(): raise OptionalDependencyNotAvailable() except OptionalDependencyNotAvailable: pass else: from .modeling_tf_distilbert import ( TF_DISTILBERT_PRETRAINED_MODEL_ARCHIVE_LIST, TFDistilBertForMaskedLM, TFDistilBertForMultipleChoice, TFDistilBertForQuestionAnswering, TFDistilBertForSequenceClassification, TFDistilBertForTokenClassification, TFDistilBertMainLayer, TFDistilBertModel, TFDistilBertPreTrainedModel, ) try: if not is_flax_available(): raise OptionalDependencyNotAvailable() except OptionalDependencyNotAvailable: pass else: from .modeling_flax_distilbert import ( FlaxDistilBertForMaskedLM, FlaxDistilBertForMultipleChoice, FlaxDistilBertForQuestionAnswering, FlaxDistilBertForSequenceClassification, FlaxDistilBertForTokenClassification, FlaxDistilBertModel, FlaxDistilBertPreTrainedModel, ) else: import sys __lowercase : int = _LazyModule(__name__, globals()['__file__'], _import_structure, module_spec=__spec__)
294
0
from maths.is_square_free import is_square_free from maths.prime_factors import prime_factors def SCREAMING_SNAKE_CASE__ ( __a ): snake_case_ : int = prime_factors(__a ) if is_square_free(__a ): return -1 if len(__a ) % 2 else 1 return 0 if __name__ == "__main__": import doctest doctest.testmod()
327
import re import string import numpy as np import datasets _SCREAMING_SNAKE_CASE = """ Returns the rate at which the input predicted strings exactly match their references, ignoring any strings input as part of the regexes_to_ignore list. """ _SCREAMING_SNAKE_CASE = """ Args: predictions: List of predicted texts. references: List of reference texts. regexes_to_ignore: List, defaults to None. Regex expressions of characters to ignore when calculating the exact matches. Note: these regexes are removed from the input data before the changes based on the options below (e.g. ignore_case, ignore_punctuation, ignore_numbers) are applied. ignore_case: Boolean, defaults to False. If true, turns everything to lowercase so that capitalization differences are ignored. ignore_punctuation: Boolean, defaults to False. If true, removes all punctuation before comparing predictions and references. ignore_numbers: Boolean, defaults to False. If true, removes all punctuation before comparing predictions and references. Returns: exact_match: Dictionary containing exact_match rate. Possible values are between 0.0 and 100.0, inclusive. Examples: >>> exact_match = datasets.load_metric(\"exact_match\") >>> refs = [\"the cat\", \"theater\", \"YELLING\", \"agent007\"] >>> preds = [\"cat?\", \"theater\", \"yelling\", \"agent\"] >>> results = exact_match.compute(references=refs, predictions=preds) >>> print(round(results[\"exact_match\"], 1)) 25.0 >>> exact_match = datasets.load_metric(\"exact_match\") >>> refs = [\"the cat\", \"theater\", \"YELLING\", \"agent007\"] >>> preds = [\"cat?\", \"theater\", \"yelling\", \"agent\"] >>> results = exact_match.compute(references=refs, predictions=preds, regexes_to_ignore=[\"the \", \"yell\"], ignore_case=True, ignore_punctuation=True) >>> print(round(results[\"exact_match\"], 1)) 50.0 >>> exact_match = datasets.load_metric(\"exact_match\") >>> refs = [\"the cat\", \"theater\", \"YELLING\", \"agent007\"] >>> preds = [\"cat?\", \"theater\", \"yelling\", \"agent\"] >>> results = exact_match.compute(references=refs, predictions=preds, regexes_to_ignore=[\"the \", \"yell\", \"YELL\"], ignore_case=True, ignore_punctuation=True) >>> print(round(results[\"exact_match\"], 1)) 75.0 >>> exact_match = datasets.load_metric(\"exact_match\") >>> refs = [\"the cat\", \"theater\", \"YELLING\", \"agent007\"] >>> preds = [\"cat?\", \"theater\", \"yelling\", \"agent\"] >>> results = exact_match.compute(references=refs, predictions=preds, regexes_to_ignore=[\"the \", \"yell\", \"YELL\"], ignore_case=True, ignore_punctuation=True, ignore_numbers=True) >>> print(round(results[\"exact_match\"], 1)) 100.0 >>> exact_match = datasets.load_metric(\"exact_match\") >>> refs = [\"The cat sat on the mat.\", \"Theaters are great.\", \"It's like comparing oranges and apples.\"] >>> preds = [\"The cat sat on the mat?\", \"Theaters are great.\", \"It's like comparing apples and oranges.\"] >>> results = exact_match.compute(references=refs, predictions=preds) >>> print(round(results[\"exact_match\"], 1)) 33.3 """ _SCREAMING_SNAKE_CASE = """ """ @datasets.utils.file_utils.add_start_docstrings(_DESCRIPTION , _KWARGS_DESCRIPTION ) class SCREAMING_SNAKE_CASE_ ( datasets.Metric ): def UpperCAmelCase_ ( self : Union[str, Any] ) -> Optional[int]: """simple docstring""" return datasets.MetricInfo( description=_DESCRIPTION , citation=_CITATION , inputs_description=_KWARGS_DESCRIPTION , features=datasets.Features( { 'predictions': datasets.Value('string' , id='sequence' ), 'references': datasets.Value('string' , id='sequence' ), } ) , reference_urls=[] , ) def UpperCAmelCase_ ( self : int , _A : Tuple , _A : Tuple , _A : str=None , _A : Dict=False , _A : Tuple=False , _A : str=False , ) -> Tuple: """simple docstring""" if regexes_to_ignore is not None: for s in regexes_to_ignore: snake_case_ : List[Any] = np.array([re.sub(_A , '' , _A ) for x in predictions] ) snake_case_ : Optional[Any] = np.array([re.sub(_A , '' , _A ) for x in references] ) else: snake_case_ : Dict = np.asarray(_A ) snake_case_ : Tuple = np.asarray(_A ) if ignore_case: snake_case_ : List[str] = np.char.lower(_A ) snake_case_ : Any = np.char.lower(_A ) if ignore_punctuation: snake_case_ : int = string.punctuation.maketrans('' , '' , string.punctuation ) snake_case_ : Tuple = np.char.translate(_A , table=_A ) snake_case_ : str = np.char.translate(_A , table=_A ) if ignore_numbers: snake_case_ : Optional[int] = string.digits.maketrans('' , '' , string.digits ) snake_case_ : str = np.char.translate(_A , table=_A ) snake_case_ : Union[str, Any] = np.char.translate(_A , table=_A ) snake_case_ : int = predictions == references return {"exact_match": np.mean(_A ) * 100}
327
1
"""simple docstring""" import gc import unittest import torch from parameterized import parameterized from diffusers import AutoencoderKL from diffusers.utils import floats_tensor, load_hf_numpy, require_torch_gpu, slow, torch_all_close, torch_device from diffusers.utils.import_utils import is_xformers_available from diffusers.utils.testing_utils import enable_full_determinism from .test_modeling_common import ModelTesterMixin, UNetTesterMixin enable_full_determinism() class __a (UpperCamelCase_ , UpperCamelCase_ , unittest.TestCase): '''simple docstring''' _SCREAMING_SNAKE_CASE :Optional[Any] = AutoencoderKL _SCREAMING_SNAKE_CASE :int = """sample""" _SCREAMING_SNAKE_CASE :Optional[Any] = 1E-2 @property def _a ( self ) -> List[Any]: """simple docstring""" SCREAMING_SNAKE_CASE__ : Union[str, Any] = 4 SCREAMING_SNAKE_CASE__ : List[Any] = 3 SCREAMING_SNAKE_CASE__ : Union[str, Any] = (32, 32) SCREAMING_SNAKE_CASE__ : List[Any] = floats_tensor((batch_size, num_channels) + sizes ).to(_a ) return {"sample": image} @property def _a ( self ) -> Tuple: """simple docstring""" return (3, 32, 32) @property def _a ( self ) -> int: """simple docstring""" return (3, 32, 32) def _a ( self ) -> Tuple: """simple docstring""" SCREAMING_SNAKE_CASE__ : List[str] = { """block_out_channels""": [32, 64], """in_channels""": 3, """out_channels""": 3, """down_block_types""": ["""DownEncoderBlock2D""", """DownEncoderBlock2D"""], """up_block_types""": ["""UpDecoderBlock2D""", """UpDecoderBlock2D"""], """latent_channels""": 4, } SCREAMING_SNAKE_CASE__ : Optional[Any] = self.dummy_input return init_dict, inputs_dict def _a ( self ) -> Optional[Any]: """simple docstring""" pass def _a ( self ) -> Dict: """simple docstring""" pass @unittest.skipIf(torch_device == """mps""" , """Gradient checkpointing skipped on MPS""" ) def _a ( self ) -> Tuple: """simple docstring""" SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ : str = self.prepare_init_args_and_inputs_for_common() SCREAMING_SNAKE_CASE__ : Any = self.model_class(**_a ) model.to(_a ) assert not model.is_gradient_checkpointing and model.training SCREAMING_SNAKE_CASE__ : Dict = model(**_a ).sample # run the backwards pass on the model. For backwards pass, for simplicity purpose, # we won't calculate the loss and rather backprop on out.sum() model.zero_grad() SCREAMING_SNAKE_CASE__ : List[Any] = torch.randn_like(_a ) SCREAMING_SNAKE_CASE__ : List[Any] = (out - labels).mean() loss.backward() # re-instantiate the model now enabling gradient checkpointing SCREAMING_SNAKE_CASE__ : Optional[Any] = self.model_class(**_a ) # clone model model_a.load_state_dict(model.state_dict() ) model_a.to(_a ) model_a.enable_gradient_checkpointing() assert model_a.is_gradient_checkpointing and model_a.training SCREAMING_SNAKE_CASE__ : Tuple = model_a(**_a ).sample # run the backwards pass on the model. For backwards pass, for simplicity purpose, # we won't calculate the loss and rather backprop on out.sum() model_a.zero_grad() SCREAMING_SNAKE_CASE__ : Tuple = (out_a - labels).mean() loss_a.backward() # compare the output and parameters gradients self.assertTrue((loss - loss_a).abs() < 1E-5 ) SCREAMING_SNAKE_CASE__ : Union[str, Any] = dict(model.named_parameters() ) SCREAMING_SNAKE_CASE__ : List[Any] = dict(model_a.named_parameters() ) for name, param in named_params.items(): self.assertTrue(torch_all_close(param.grad.data , named_params_a[name].grad.data , atol=5E-5 ) ) def _a ( self ) -> List[Any]: """simple docstring""" SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ : Tuple = AutoencoderKL.from_pretrained("""fusing/autoencoder-kl-dummy""" , output_loading_info=_a ) self.assertIsNotNone(_a ) self.assertEqual(len(loading_info["""missing_keys"""] ) , 0 ) model.to(_a ) SCREAMING_SNAKE_CASE__ : Any = model(**self.dummy_input ) assert image is not None, "Make sure output is not None" def _a ( self ) -> List[Any]: """simple docstring""" SCREAMING_SNAKE_CASE__ : List[str] = AutoencoderKL.from_pretrained("""fusing/autoencoder-kl-dummy""" ) SCREAMING_SNAKE_CASE__ : List[str] = model.to(_a ) model.eval() if torch_device == "mps": SCREAMING_SNAKE_CASE__ : List[Any] = torch.manual_seed(0 ) else: SCREAMING_SNAKE_CASE__ : Dict = torch.Generator(device=_a ).manual_seed(0 ) SCREAMING_SNAKE_CASE__ : List[Any] = torch.randn( 1 , model.config.in_channels , model.config.sample_size , model.config.sample_size , generator=torch.manual_seed(0 ) , ) SCREAMING_SNAKE_CASE__ : List[Any] = image.to(_a ) with torch.no_grad(): SCREAMING_SNAKE_CASE__ : Tuple = model(_a , sample_posterior=_a , generator=_a ).sample SCREAMING_SNAKE_CASE__ : Any = output[0, -1, -3:, -3:].flatten().cpu() # Since the VAE Gaussian prior's generator is seeded on the appropriate device, # the expected output slices are not the same for CPU and GPU. if torch_device == "mps": SCREAMING_SNAKE_CASE__ : Optional[Any] = torch.tensor( [ -4.0_0_7_8E-0_1, -3.8_3_2_3E-0_4, -1.2_6_8_1E-0_1, -1.1_4_6_2E-0_1, 2.0_0_9_5E-0_1, 1.0_8_9_3E-0_1, -8.8_2_4_7E-0_2, -3.0_3_6_1E-0_1, -9.8_6_4_4E-0_3, ] ) elif torch_device == "cpu": SCREAMING_SNAKE_CASE__ : Optional[Any] = torch.tensor( [-0.1_352, 0.0_878, 0.0_419, -0.0_818, -0.1_069, 0.0_688, -0.1_458, -0.4_446, -0.0_026] ) else: SCREAMING_SNAKE_CASE__ : Any = torch.tensor( [-0.2_421, 0.4_642, 0.2_507, -0.0_438, 0.0_682, 0.3_160, -0.2_018, -0.0_727, 0.2_485] ) self.assertTrue(torch_all_close(_a , _a , rtol=1E-2 ) ) @slow class __a (unittest.TestCase): '''simple docstring''' def _a ( self , _a , _a ) -> List[Any]: """simple docstring""" return f'''gaussian_noise_s={seed}_shape={'_'.join([str(_a ) for s in shape] )}.npy''' def _a ( self ) -> List[str]: """simple docstring""" super().tearDown() gc.collect() torch.cuda.empty_cache() def _a ( self , _a=0 , _a=(4, 3, 512, 512) , _a=False ) -> List[Any]: """simple docstring""" SCREAMING_SNAKE_CASE__ : Union[str, Any] = torch.floataa if fpaa else torch.floataa SCREAMING_SNAKE_CASE__ : str = torch.from_numpy(load_hf_numpy(self.get_file_format(_a , _a ) ) ).to(_a ).to(_a ) return image def _a ( self , _a="CompVis/stable-diffusion-v1-4" , _a=False ) -> Optional[int]: """simple docstring""" SCREAMING_SNAKE_CASE__ : int = """fp16""" if fpaa else None SCREAMING_SNAKE_CASE__ : Union[str, Any] = torch.floataa if fpaa else torch.floataa SCREAMING_SNAKE_CASE__ : List[str] = AutoencoderKL.from_pretrained( _a , subfolder="""vae""" , torch_dtype=_a , revision=_a , ) model.to(_a ).eval() return model def _a ( self , _a=0 ) -> List[str]: """simple docstring""" if torch_device == "mps": return torch.manual_seed(_a ) return torch.Generator(device=_a ).manual_seed(_a ) @parameterized.expand( [ # fmt: off [33, [-0.1_603, 0.9_878, -0.0_495, -0.0_790, -0.2_709, 0.8_375, -0.2_060, -0.0_824], [-0.2_395, 0.0_098, 0.0_102, -0.0_709, -0.2_840, -0.0_274, -0.0_718, -0.1_824]], [47, [-0.2_376, 0.1_168, 0.1_332, -0.4_840, -0.2_508, -0.0_791, -0.0_493, -0.4_089], [0.0_350, 0.0_847, 0.0_467, 0.0_344, -0.0_842, -0.0_547, -0.0_633, -0.1_131]], # fmt: on ] ) def _a ( self , _a , _a , _a ) -> List[str]: """simple docstring""" SCREAMING_SNAKE_CASE__ : List[Any] = self.get_sd_vae_model() SCREAMING_SNAKE_CASE__ : List[Any] = self.get_sd_image(_a ) SCREAMING_SNAKE_CASE__ : Optional[int] = self.get_generator(_a ) with torch.no_grad(): SCREAMING_SNAKE_CASE__ : Optional[Any] = model(_a , generator=_a , sample_posterior=_a ).sample assert sample.shape == image.shape SCREAMING_SNAKE_CASE__ : int = sample[-1, -2:, -2:, :2].flatten().float().cpu() SCREAMING_SNAKE_CASE__ : Dict = torch.tensor(expected_slice_mps if torch_device == """mps""" else expected_slice ) assert torch_all_close(_a , _a , atol=3E-3 ) @parameterized.expand( [ # fmt: off [33, [-0.0_513, 0.0_289, 1.3_799, 0.2_166, -0.2_573, -0.0_871, 0.5_103, -0.0_999]], [47, [-0.4_128, -0.1_320, -0.3_704, 0.1_965, -0.4_116, -0.2_332, -0.3_340, 0.2_247]], # fmt: on ] ) @require_torch_gpu def _a ( self , _a , _a ) -> Optional[Any]: """simple docstring""" SCREAMING_SNAKE_CASE__ : Tuple = self.get_sd_vae_model(fpaa=_a ) SCREAMING_SNAKE_CASE__ : Union[str, Any] = self.get_sd_image(_a , fpaa=_a ) SCREAMING_SNAKE_CASE__ : Any = self.get_generator(_a ) with torch.no_grad(): SCREAMING_SNAKE_CASE__ : List[str] = model(_a , generator=_a , sample_posterior=_a ).sample assert sample.shape == image.shape SCREAMING_SNAKE_CASE__ : Union[str, Any] = sample[-1, -2:, :2, -2:].flatten().float().cpu() SCREAMING_SNAKE_CASE__ : Tuple = torch.tensor(_a ) assert torch_all_close(_a , _a , atol=1E-2 ) @parameterized.expand( [ # fmt: off [33, [-0.1_609, 0.9_866, -0.0_487, -0.0_777, -0.2_716, 0.8_368, -0.2_055, -0.0_814], [-0.2_395, 0.0_098, 0.0_102, -0.0_709, -0.2_840, -0.0_274, -0.0_718, -0.1_824]], [47, [-0.2_377, 0.1_147, 0.1_333, -0.4_841, -0.2_506, -0.0_805, -0.0_491, -0.4_085], [0.0_350, 0.0_847, 0.0_467, 0.0_344, -0.0_842, -0.0_547, -0.0_633, -0.1_131]], # fmt: on ] ) def _a ( self , _a , _a , _a ) -> Dict: """simple docstring""" SCREAMING_SNAKE_CASE__ : str = self.get_sd_vae_model() SCREAMING_SNAKE_CASE__ : Any = self.get_sd_image(_a ) with torch.no_grad(): SCREAMING_SNAKE_CASE__ : Tuple = model(_a ).sample assert sample.shape == image.shape SCREAMING_SNAKE_CASE__ : Any = sample[-1, -2:, -2:, :2].flatten().float().cpu() SCREAMING_SNAKE_CASE__ : List[Any] = torch.tensor(expected_slice_mps if torch_device == """mps""" else expected_slice ) assert torch_all_close(_a , _a , atol=3E-3 ) @parameterized.expand( [ # fmt: off [13, [-0.2_051, -0.1_803, -0.2_311, -0.2_114, -0.3_292, -0.3_574, -0.2_953, -0.3_323]], [37, [-0.2_632, -0.2_625, -0.2_199, -0.2_741, -0.4_539, -0.4_990, -0.3_720, -0.4_925]], # fmt: on ] ) @require_torch_gpu def _a ( self , _a , _a ) -> Optional[int]: """simple docstring""" SCREAMING_SNAKE_CASE__ : Tuple = self.get_sd_vae_model() SCREAMING_SNAKE_CASE__ : int = self.get_sd_image(_a , shape=(3, 4, 64, 64) ) with torch.no_grad(): SCREAMING_SNAKE_CASE__ : Union[str, Any] = model.decode(_a ).sample assert list(sample.shape ) == [3, 3, 512, 512] SCREAMING_SNAKE_CASE__ : Optional[int] = sample[-1, -2:, :2, -2:].flatten().cpu() SCREAMING_SNAKE_CASE__ : Union[str, Any] = torch.tensor(_a ) assert torch_all_close(_a , _a , atol=1E-3 ) @parameterized.expand( [ # fmt: off [27, [-0.0_369, 0.0_207, -0.0_776, -0.0_682, -0.1_747, -0.1_930, -0.1_465, -0.2_039]], [16, [-0.1_628, -0.2_134, -0.2_747, -0.2_642, -0.3_774, -0.4_404, -0.3_687, -0.4_277]], # fmt: on ] ) @require_torch_gpu def _a ( self , _a , _a ) -> Optional[Any]: """simple docstring""" SCREAMING_SNAKE_CASE__ : List[Any] = self.get_sd_vae_model(fpaa=_a ) SCREAMING_SNAKE_CASE__ : List[Any] = self.get_sd_image(_a , shape=(3, 4, 64, 64) , fpaa=_a ) with torch.no_grad(): SCREAMING_SNAKE_CASE__ : Optional[Any] = model.decode(_a ).sample assert list(sample.shape ) == [3, 3, 512, 512] SCREAMING_SNAKE_CASE__ : Any = sample[-1, -2:, :2, -2:].flatten().float().cpu() SCREAMING_SNAKE_CASE__ : Optional[Any] = torch.tensor(_a ) assert torch_all_close(_a , _a , atol=5E-3 ) @parameterized.expand([(13,), (16,), (27,)] ) @require_torch_gpu @unittest.skipIf(not is_xformers_available() , reason="""xformers is not required when using PyTorch 2.0.""" ) def _a ( self , _a ) -> Any: """simple docstring""" SCREAMING_SNAKE_CASE__ : List[str] = self.get_sd_vae_model(fpaa=_a ) SCREAMING_SNAKE_CASE__ : Dict = self.get_sd_image(_a , shape=(3, 4, 64, 64) , fpaa=_a ) with torch.no_grad(): SCREAMING_SNAKE_CASE__ : List[Any] = model.decode(_a ).sample model.enable_xformers_memory_efficient_attention() with torch.no_grad(): SCREAMING_SNAKE_CASE__ : str = model.decode(_a ).sample assert list(sample.shape ) == [3, 3, 512, 512] assert torch_all_close(_a , _a , atol=1E-1 ) @parameterized.expand([(13,), (16,), (37,)] ) @require_torch_gpu @unittest.skipIf(not is_xformers_available() , reason="""xformers is not required when using PyTorch 2.0.""" ) def _a ( self , _a ) -> Optional[Any]: """simple docstring""" SCREAMING_SNAKE_CASE__ : Optional[Any] = self.get_sd_vae_model() SCREAMING_SNAKE_CASE__ : Dict = self.get_sd_image(_a , shape=(3, 4, 64, 64) ) with torch.no_grad(): SCREAMING_SNAKE_CASE__ : str = model.decode(_a ).sample model.enable_xformers_memory_efficient_attention() with torch.no_grad(): SCREAMING_SNAKE_CASE__ : Union[str, Any] = model.decode(_a ).sample assert list(sample.shape ) == [3, 3, 512, 512] assert torch_all_close(_a , _a , atol=1E-2 ) @parameterized.expand( [ # fmt: off [33, [-0.3_001, 0.0_918, -2.6_984, -3.9_720, -3.2_099, -5.0_353, 1.7_338, -0.2_065, 3.4_267]], [47, [-1.5_030, -4.3_871, -6.0_355, -9.1_157, -1.6_661, -2.7_853, 2.1_607, -5.0_823, 2.5_633]], # fmt: on ] ) def _a ( self , _a , _a ) -> str: """simple docstring""" SCREAMING_SNAKE_CASE__ : str = self.get_sd_vae_model() SCREAMING_SNAKE_CASE__ : int = self.get_sd_image(_a ) SCREAMING_SNAKE_CASE__ : Optional[int] = self.get_generator(_a ) with torch.no_grad(): SCREAMING_SNAKE_CASE__ : Optional[Any] = model.encode(_a ).latent_dist SCREAMING_SNAKE_CASE__ : List[Any] = dist.sample(generator=_a ) assert list(sample.shape ) == [image.shape[0], 4] + [i // 8 for i in image.shape[2:]] SCREAMING_SNAKE_CASE__ : int = sample[0, -1, -3:, -3:].flatten().cpu() SCREAMING_SNAKE_CASE__ : Optional[int] = torch.tensor(_a ) SCREAMING_SNAKE_CASE__ : str = 3E-3 if torch_device != """mps""" else 1E-2 assert torch_all_close(_a , _a , atol=_a )
56
"""simple docstring""" def _lowercase ( __lowerCAmelCase ) -> Tuple: SCREAMING_SNAKE_CASE__ : Union[str, Any] = [0] * len(__lowerCAmelCase ) SCREAMING_SNAKE_CASE__ : List[Any] = [] SCREAMING_SNAKE_CASE__ : Any = [1] * len(__lowerCAmelCase ) for values in graph.values(): for i in values: indegree[i] += 1 for i in range(len(__lowerCAmelCase ) ): if indegree[i] == 0: queue.append(__lowerCAmelCase ) while queue: SCREAMING_SNAKE_CASE__ : str = queue.pop(0 ) for x in graph[vertex]: indegree[x] -= 1 if long_dist[vertex] + 1 > long_dist[x]: SCREAMING_SNAKE_CASE__ : str = long_dist[vertex] + 1 if indegree[x] == 0: queue.append(__lowerCAmelCase ) print(max(__lowerCAmelCase ) ) # Adjacency list of Graph a :int = {0: [2, 3, 4], 1: [2, 7], 2: [5], 3: [5, 7], 4: [7], 5: [6], 6: [7], 7: []} longest_distance(graph)
56
1
import inspect from typing import Optional, Union import numpy as np import PIL import torch from torch.nn import functional as F from torchvision import transforms from transformers import CLIPFeatureExtractor, CLIPModel, CLIPTextModel, CLIPTokenizer from diffusers import ( AutoencoderKL, DDIMScheduler, DiffusionPipeline, DPMSolverMultistepScheduler, LMSDiscreteScheduler, PNDMScheduler, UNetaDConditionModel, ) from diffusers.pipelines.stable_diffusion.pipeline_stable_diffusion import StableDiffusionPipelineOutput from diffusers.utils import ( PIL_INTERPOLATION, randn_tensor, ) def lowercase_ (A : Tuple , A : List[Any] , A : List[str] ): if isinstance(SCREAMING_SNAKE_CASE__ , torch.Tensor ): return image elif isinstance(SCREAMING_SNAKE_CASE__ , PIL.Image.Image ): snake_case__ : Tuple = [image] if isinstance(image[0] , PIL.Image.Image ): snake_case__ : List[str] = [np.array(i.resize((w, h) , resample=PIL_INTERPOLATION['lanczos'] ) )[None, :] for i in image] snake_case__ : Optional[Any] = np.concatenate(SCREAMING_SNAKE_CASE__ , axis=0 ) snake_case__ : Optional[int] = np.array(SCREAMING_SNAKE_CASE__ ).astype(np.floataa ) / 255.0 snake_case__ : Tuple = image.transpose(0 , 3 , 1 , 2 ) snake_case__ : Dict = 2.0 * image - 1.0 snake_case__ : Optional[Any] = torch.from_numpy(SCREAMING_SNAKE_CASE__ ) elif isinstance(image[0] , torch.Tensor ): snake_case__ : Any = torch.cat(SCREAMING_SNAKE_CASE__ , dim=0 ) return image def lowercase_ (A : List[Any] , A : Optional[int] , A : Any , A : Optional[Any]=0.9995 ): if not isinstance(SCREAMING_SNAKE_CASE__ , np.ndarray ): snake_case__ : Optional[Any] = True snake_case__ : Any = va.device snake_case__ : Tuple = va.cpu().numpy() snake_case__ : Dict = va.cpu().numpy() snake_case__ : Any = np.sum(va * va / (np.linalg.norm(SCREAMING_SNAKE_CASE__ ) * np.linalg.norm(SCREAMING_SNAKE_CASE__ )) ) if np.abs(SCREAMING_SNAKE_CASE__ ) > DOT_THRESHOLD: snake_case__ : Any = (1 - t) * va + t * va else: snake_case__ : int = np.arccos(SCREAMING_SNAKE_CASE__ ) snake_case__ : List[str] = np.sin(SCREAMING_SNAKE_CASE__ ) snake_case__ : str = theta_a * t snake_case__ : List[Any] = np.sin(SCREAMING_SNAKE_CASE__ ) snake_case__ : int = np.sin(theta_a - theta_t ) / sin_theta_a snake_case__ : int = sin_theta_t / sin_theta_a snake_case__ : Dict = sa * va + sa * va if inputs_are_torch: snake_case__ : Optional[int] = torch.from_numpy(SCREAMING_SNAKE_CASE__ ).to(SCREAMING_SNAKE_CASE__ ) return va def lowercase_ (A : Union[str, Any] , A : Union[str, Any] ): snake_case__ : Union[str, Any] = F.normalize(SCREAMING_SNAKE_CASE__ , dim=-1 ) snake_case__ : List[Any] = F.normalize(SCREAMING_SNAKE_CASE__ , dim=-1 ) return (x - y).norm(dim=-1 ).div(2 ).arcsin().pow(2 ).mul(2 ) def lowercase_ (A : str , A : Tuple ): for param in model.parameters(): snake_case__ : List[str] = value class snake_case__ ( lowerCAmelCase_ ): """simple docstring""" def __init__( self : List[Any], _snake_case : Union[str, Any], _snake_case : Any, _snake_case : List[str], _snake_case : int, _snake_case : Any, _snake_case : Optional[int], _snake_case : int, _snake_case : str=None, _snake_case : Optional[Any]=None, _snake_case : Tuple=None, ) ->Optional[Any]: super().__init__() self.register_modules( vae=_snake_case, text_encoder=_snake_case, clip_model=_snake_case, tokenizer=_snake_case, unet=_snake_case, scheduler=_snake_case, feature_extractor=_snake_case, coca_model=_snake_case, coca_tokenizer=_snake_case, coca_transform=_snake_case, ) snake_case__ : Optional[int] = ( feature_extractor.size if isinstance(feature_extractor.size, _snake_case ) else feature_extractor.size["""shortest_edge"""] ) snake_case__ : Dict = transforms.Normalize(mean=feature_extractor.image_mean, std=feature_extractor.image_std ) set_requires_grad(self.text_encoder, _snake_case ) set_requires_grad(self.clip_model, _snake_case ) def lowercase_ ( self : List[str], _snake_case : Optional[int] = "auto" ) ->Union[str, Any]: if slice_size == "auto": # half the attention head size is usually a good trade-off between # speed and memory snake_case__ : List[Any] = self.unet.config.attention_head_dim // 2 self.unet.set_attention_slice(_snake_case ) def lowercase_ ( self : Optional[Any] ) ->Any: self.enable_attention_slicing(_snake_case ) def lowercase_ ( self : Optional[Any] ) ->Union[str, Any]: set_requires_grad(self.vae, _snake_case ) def lowercase_ ( self : Optional[Any] ) ->Union[str, Any]: set_requires_grad(self.vae, _snake_case ) def lowercase_ ( self : List[str] ) ->List[str]: set_requires_grad(self.unet, _snake_case ) def lowercase_ ( self : str ) ->Any: set_requires_grad(self.unet, _snake_case ) def lowercase_ ( self : Optional[Any], _snake_case : Optional[Any], _snake_case : Optional[int], _snake_case : Optional[int] ) ->str: snake_case__ : Optional[int] = min(int(num_inference_steps * strength ), _snake_case ) snake_case__ : List[Any] = max(num_inference_steps - init_timestep, 0 ) snake_case__ : int = self.scheduler.timesteps[t_start:] return timesteps, num_inference_steps - t_start def lowercase_ ( self : Union[str, Any], _snake_case : List[Any], _snake_case : List[str], _snake_case : List[Any], _snake_case : Any, _snake_case : Optional[Any], _snake_case : Union[str, Any]=None ) ->Dict: if not isinstance(_snake_case, torch.Tensor ): raise ValueError(F'''`image` has to be of type `torch.Tensor` but is {type(_snake_case )}''' ) snake_case__ : List[str] = image.to(device=_snake_case, dtype=_snake_case ) if isinstance(_snake_case, _snake_case ): snake_case__ : Optional[int] = [ self.vae.encode(image[i : i + 1] ).latent_dist.sample(generator[i] ) for i in range(_snake_case ) ] snake_case__ : Tuple = torch.cat(_snake_case, dim=0 ) else: snake_case__ : List[str] = self.vae.encode(_snake_case ).latent_dist.sample(_snake_case ) # Hardcode 0.18215 because stable-diffusion-2-base has not self.vae.config.scaling_factor snake_case__ : Any = 0.1_8_2_1_5 * init_latents snake_case__ : Dict = init_latents.repeat_interleave(_snake_case, dim=0 ) snake_case__ : List[str] = randn_tensor(init_latents.shape, generator=_snake_case, device=_snake_case, dtype=_snake_case ) # get latents snake_case__ : Optional[int] = self.scheduler.add_noise(_snake_case, _snake_case, _snake_case ) snake_case__ : Optional[Any] = init_latents return latents def lowercase_ ( self : Optional[Any], _snake_case : List[str] ) ->Any: snake_case__ : Dict = self.coca_transform(_snake_case ).unsqueeze(0 ) with torch.no_grad(), torch.cuda.amp.autocast(): snake_case__ : List[Any] = self.coca_model.generate(transformed_image.to(device=self.device, dtype=self.coca_model.dtype ) ) snake_case__ : int = self.coca_tokenizer.decode(generated[0].cpu().numpy() ) return generated.split('<end_of_text>' )[0].replace('<start_of_text>', '' ).rstrip(' .,' ) def lowercase_ ( self : Optional[Any], _snake_case : List[str], _snake_case : Optional[int] ) ->int: snake_case__ : Optional[int] = self.feature_extractor.preprocess(_snake_case ) snake_case__ : Optional[int] = torch.from_numpy(clip_image_input['pixel_values'][0] ).unsqueeze(0 ).to(self.device ).half() snake_case__ : List[Any] = self.clip_model.get_image_features(_snake_case ) snake_case__ : List[str] = image_embeddings_clip / image_embeddings_clip.norm(p=2, dim=-1, keepdim=_snake_case ) snake_case__ : Tuple = image_embeddings_clip.repeat_interleave(_snake_case, dim=0 ) return image_embeddings_clip @torch.enable_grad() def lowercase_ ( self : List[Any], _snake_case : str, _snake_case : Tuple, _snake_case : Optional[Any], _snake_case : int, _snake_case : Any, _snake_case : List[Any], _snake_case : Dict, ) ->int: snake_case__ : Optional[int] = latents.detach().requires_grad_() snake_case__ : Optional[int] = self.scheduler.scale_model_input(_snake_case, _snake_case ) # predict the noise residual snake_case__ : Optional[int] = self.unet(_snake_case, _snake_case, encoder_hidden_states=_snake_case ).sample if isinstance(self.scheduler, (PNDMScheduler, DDIMScheduler, DPMSolverMultistepScheduler) ): snake_case__ : Optional[int] = self.scheduler.alphas_cumprod[timestep] snake_case__ : Union[str, Any] = 1 - alpha_prod_t # compute predicted original sample from predicted noise also called # "predicted x_0" of formula (12) from https://arxiv.org/pdf/2010.02502.pdf snake_case__ : Tuple = (latents - beta_prod_t ** 0.5 * noise_pred) / alpha_prod_t ** 0.5 snake_case__ : int = torch.sqrt(_snake_case ) snake_case__ : Any = pred_original_sample * (fac) + latents * (1 - fac) elif isinstance(self.scheduler, _snake_case ): snake_case__ : Dict = self.scheduler.sigmas[index] snake_case__ : Tuple = latents - sigma * noise_pred else: raise ValueError(F'''scheduler type {type(self.scheduler )} not supported''' ) # Hardcode 0.18215 because stable-diffusion-2-base has not self.vae.config.scaling_factor snake_case__ : Optional[Any] = 1 / 0.1_8_2_1_5 * sample snake_case__ : Union[str, Any] = self.vae.decode(_snake_case ).sample snake_case__ : str = (image / 2 + 0.5).clamp(0, 1 ) snake_case__ : int = transforms.Resize(self.feature_extractor_size )(_snake_case ) snake_case__ : Tuple = self.normalize(_snake_case ).to(latents.dtype ) snake_case__ : Tuple = self.clip_model.get_image_features(_snake_case ) snake_case__ : List[str] = image_embeddings_clip / image_embeddings_clip.norm(p=2, dim=-1, keepdim=_snake_case ) snake_case__ : str = spherical_dist_loss(_snake_case, _snake_case ).mean() * clip_guidance_scale snake_case__ : List[Any] = -torch.autograd.grad(_snake_case, _snake_case )[0] if isinstance(self.scheduler, _snake_case ): snake_case__ : Any = latents.detach() + grads * (sigma**2) snake_case__ : Optional[Any] = noise_pred_original else: snake_case__ : int = noise_pred_original - torch.sqrt(_snake_case ) * grads return noise_pred, latents @torch.no_grad() def __call__( self : Any, _snake_case : List[Any], _snake_case : Dict, _snake_case : List[str] = None, _snake_case : Dict = None, _snake_case : Tuple = 5_1_2, _snake_case : Any = 5_1_2, _snake_case : Optional[Any] = 0.6, _snake_case : str = 5_0, _snake_case : Optional[Any] = 7.5, _snake_case : Any = 1, _snake_case : Union[str, Any] = 0.0, _snake_case : Any = 1_0_0, _snake_case : Optional[Any] = None, _snake_case : List[Any] = "pil", _snake_case : Union[str, Any] = True, _snake_case : Union[str, Any] = 0.8, _snake_case : Dict = 0.1, _snake_case : Tuple = 0.1, ) ->Optional[Any]: if isinstance(_snake_case, _snake_case ) and len(_snake_case ) != batch_size: raise ValueError(F'''You have passed {batch_size} batch_size, but only {len(_snake_case )} generators.''' ) if height % 8 != 0 or width % 8 != 0: raise ValueError(F'''`height` and `width` have to be divisible by 8 but are {height} and {width}.''' ) if isinstance(_snake_case, torch.Generator ) and batch_size > 1: snake_case__ : str = [generator] + [None] * (batch_size - 1) snake_case__ : Union[str, Any] = [ ("""model""", self.coca_model is None), ("""tokenizer""", self.coca_tokenizer is None), ("""transform""", self.coca_transform is None), ] snake_case__ : Optional[int] = [x[0] for x in coca_is_none if x[1]] snake_case__ : int = """, """.join(_snake_case ) # generate prompts with coca model if prompt is None if content_prompt is None: if len(_snake_case ): raise ValueError( F'''Content prompt is None and CoCa [{coca_is_none_str}] is None.''' F'''Set prompt or pass Coca [{coca_is_none_str}] to DiffusionPipeline.''' ) snake_case__ : Optional[int] = self.get_image_description(_snake_case ) if style_prompt is None: if len(_snake_case ): raise ValueError( F'''Style prompt is None and CoCa [{coca_is_none_str}] is None.''' F''' Set prompt or pass Coca [{coca_is_none_str}] to DiffusionPipeline.''' ) snake_case__ : str = self.get_image_description(_snake_case ) # get prompt text embeddings for content and style snake_case__ : List[Any] = self.tokenizer( _snake_case, padding='max_length', max_length=self.tokenizer.model_max_length, truncation=_snake_case, return_tensors='pt', ) snake_case__ : List[Any] = self.text_encoder(content_text_input.input_ids.to(self.device ) )[0] snake_case__ : Optional[int] = self.tokenizer( _snake_case, padding='max_length', max_length=self.tokenizer.model_max_length, truncation=_snake_case, return_tensors='pt', ) snake_case__ : Tuple = self.text_encoder(style_text_input.input_ids.to(self.device ) )[0] snake_case__ : Optional[Any] = slerp(_snake_case, _snake_case, _snake_case ) # duplicate text embeddings for each generation per prompt snake_case__ : str = text_embeddings.repeat_interleave(_snake_case, dim=0 ) # set timesteps snake_case__ : str = """offset""" in set(inspect.signature(self.scheduler.set_timesteps ).parameters.keys() ) snake_case__ : Tuple = {} if accepts_offset: snake_case__ : int = 1 self.scheduler.set_timesteps(_snake_case, **_snake_case ) # Some schedulers like PNDM have timesteps as arrays # It's more optimized to move all timesteps to correct device beforehand self.scheduler.timesteps.to(self.device ) snake_case__ : Optional[int] = self.get_timesteps(_snake_case, _snake_case, self.device ) snake_case__ : Tuple = timesteps[:1].repeat(_snake_case ) # Preprocess image snake_case__ : str = preprocess(_snake_case, _snake_case, _snake_case ) snake_case__ : int = self.prepare_latents( _snake_case, _snake_case, _snake_case, text_embeddings.dtype, self.device, _snake_case ) snake_case__ : List[Any] = preprocess(_snake_case, _snake_case, _snake_case ) snake_case__ : Tuple = self.prepare_latents( _snake_case, _snake_case, _snake_case, text_embeddings.dtype, self.device, _snake_case ) snake_case__ : List[str] = slerp(_snake_case, _snake_case, _snake_case ) if clip_guidance_scale > 0: snake_case__ : Union[str, Any] = self.get_clip_image_embeddings(_snake_case, _snake_case ) snake_case__ : Optional[int] = self.get_clip_image_embeddings(_snake_case, _snake_case ) snake_case__ : Optional[int] = slerp( _snake_case, _snake_case, _snake_case ) # here `guidance_scale` is defined analog to the guidance weight `w` of equation (2) # of the Imagen paper: https://arxiv.org/pdf/2205.11487.pdf . `guidance_scale = 1` # corresponds to doing no classifier free guidance. snake_case__ : str = guidance_scale > 1.0 # get unconditional embeddings for classifier free guidance if do_classifier_free_guidance: snake_case__ : List[str] = content_text_input.input_ids.shape[-1] snake_case__ : Optional[Any] = self.tokenizer([''], padding='max_length', max_length=_snake_case, return_tensors='pt' ) snake_case__ : Any = self.text_encoder(uncond_input.input_ids.to(self.device ) )[0] # duplicate unconditional embeddings for each generation per prompt snake_case__ : Any = uncond_embeddings.repeat_interleave(_snake_case, dim=0 ) # For classifier free guidance, we need to do two forward passes. # Here we concatenate the unconditional and text embeddings into a single batch # to avoid doing two forward passes snake_case__ : List[str] = torch.cat([uncond_embeddings, text_embeddings] ) # get the initial random noise unless the user supplied it # Unlike in other pipelines, latents need to be generated in the target device # for 1-to-1 results reproducibility with the CompVis implementation. # However this currently doesn't work in `mps`. snake_case__ : Tuple = (batch_size, self.unet.config.in_channels, height // 8, width // 8) snake_case__ : Tuple = text_embeddings.dtype if latents is None: if self.device.type == "mps": # randn does not work reproducibly on mps snake_case__ : str = torch.randn(_snake_case, generator=_snake_case, device='cpu', dtype=_snake_case ).to( self.device ) else: snake_case__ : Optional[int] = torch.randn(_snake_case, generator=_snake_case, device=self.device, dtype=_snake_case ) else: if latents.shape != latents_shape: raise ValueError(F'''Unexpected latents shape, got {latents.shape}, expected {latents_shape}''' ) snake_case__ : Any = latents.to(self.device ) # scale the initial noise by the standard deviation required by the scheduler snake_case__ : Dict = latents * self.scheduler.init_noise_sigma # prepare extra kwargs for the scheduler step, since not all schedulers have the same signature # eta (η) is only used with the DDIMScheduler, it will be ignored for other schedulers. # eta corresponds to η in DDIM paper: https://arxiv.org/abs/2010.02502 # and should be between [0, 1] snake_case__ : Union[str, Any] = """eta""" in set(inspect.signature(self.scheduler.step ).parameters.keys() ) snake_case__ : str = {} if accepts_eta: snake_case__ : str = eta # check if the scheduler accepts generator snake_case__ : Optional[Any] = """generator""" in set(inspect.signature(self.scheduler.step ).parameters.keys() ) if accepts_generator: snake_case__ : Tuple = generator with self.progress_bar(total=_snake_case ): for i, t in enumerate(_snake_case ): # expand the latents if we are doing classifier free guidance snake_case__ : Optional[Any] = torch.cat([latents] * 2 ) if do_classifier_free_guidance else latents snake_case__ : Optional[int] = self.scheduler.scale_model_input(_snake_case, _snake_case ) # predict the noise residual snake_case__ : Dict = self.unet(_snake_case, _snake_case, encoder_hidden_states=_snake_case ).sample # perform classifier free guidance if do_classifier_free_guidance: snake_case__ : str = noise_pred.chunk(2 ) snake_case__ : Optional[int] = noise_pred_uncond + guidance_scale * (noise_pred_text - noise_pred_uncond) # perform clip guidance if clip_guidance_scale > 0: snake_case__ : int = ( text_embeddings.chunk(2 )[1] if do_classifier_free_guidance else text_embeddings ) snake_case__ : Union[str, Any] = self.cond_fn( _snake_case, _snake_case, _snake_case, _snake_case, _snake_case, _snake_case, _snake_case, ) # compute the previous noisy sample x_t -> x_t-1 snake_case__ : Any = self.scheduler.step(_snake_case, _snake_case, _snake_case, **_snake_case ).prev_sample # Hardcode 0.18215 because stable-diffusion-2-base has not self.vae.config.scaling_factor snake_case__ : Optional[Any] = 1 / 0.1_8_2_1_5 * latents snake_case__ : Any = self.vae.decode(_snake_case ).sample snake_case__ : Optional[Any] = (image / 2 + 0.5).clamp(0, 1 ) snake_case__ : Optional[Any] = image.cpu().permute(0, 2, 3, 1 ).numpy() if output_type == "pil": snake_case__ : List[str] = self.numpy_to_pil(_snake_case ) if not return_dict: return (image, None) return StableDiffusionPipelineOutput(images=_snake_case, nsfw_content_detected=_snake_case )
277
from scipy.stats import pearsonr, spearmanr from sklearn.metrics import fa_score, matthews_corrcoef import datasets lowercase : str = """\ @inproceedings{wang2019glue, title={{GLUE}: A Multi-Task Benchmark and Analysis Platform for Natural Language Understanding}, author={Wang, Alex and Singh, Amanpreet and Michael, Julian and Hill, Felix and Levy, Omer and Bowman, Samuel R.}, note={In the Proceedings of ICLR.}, year={2019} } """ lowercase : Dict = """\ GLUE, the General Language Understanding Evaluation benchmark (https://gluebenchmark.com/) is a collection of resources for training, evaluating, and analyzing natural language understanding systems. """ lowercase : int = """ Compute GLUE evaluation metric associated to each GLUE dataset. Args: predictions: list of predictions to score. Each translation should be tokenized into a list of tokens. references: list of lists of references for each translation. Each reference should be tokenized into a list of tokens. Returns: depending on the GLUE subset, one or several of: \"accuracy\": Accuracy \"f1\": F1 score \"pearson\": Pearson Correlation \"spearmanr\": Spearman Correlation \"matthews_correlation\": Matthew Correlation Examples: >>> glue_metric = datasets.load_metric('glue', 'sst2') # 'sst2' or any of [\"mnli\", \"mnli_mismatched\", \"mnli_matched\", \"qnli\", \"rte\", \"wnli\", \"hans\"] >>> references = [0, 1] >>> predictions = [0, 1] >>> results = glue_metric.compute(predictions=predictions, references=references) >>> print(results) {'accuracy': 1.0} >>> glue_metric = datasets.load_metric('glue', 'mrpc') # 'mrpc' or 'qqp' >>> references = [0, 1] >>> predictions = [0, 1] >>> results = glue_metric.compute(predictions=predictions, references=references) >>> print(results) {'accuracy': 1.0, 'f1': 1.0} >>> glue_metric = datasets.load_metric('glue', 'stsb') >>> references = [0., 1., 2., 3., 4., 5.] >>> predictions = [0., 1., 2., 3., 4., 5.] >>> results = glue_metric.compute(predictions=predictions, references=references) >>> print({\"pearson\": round(results[\"pearson\"], 2), \"spearmanr\": round(results[\"spearmanr\"], 2)}) {'pearson': 1.0, 'spearmanr': 1.0} >>> glue_metric = datasets.load_metric('glue', 'cola') >>> references = [0, 1] >>> predictions = [0, 1] >>> results = glue_metric.compute(predictions=predictions, references=references) >>> print(results) {'matthews_correlation': 1.0} """ def _snake_case( SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ ) -> Optional[int]: return float((preds == labels).mean() ) def _snake_case( SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ ) -> Optional[Any]: lowercase : Any = simple_accuracy(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ ) lowercase : Union[str, Any] = float(fa_score(y_true=SCREAMING_SNAKE_CASE__ , y_pred=SCREAMING_SNAKE_CASE__ ) ) return { "accuracy": acc, "f1": fa, } def _snake_case( SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ ) -> List[Any]: lowercase : Union[str, Any] = float(pearsonr(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ )[0] ) lowercase : Dict = float(spearmanr(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ )[0] ) return { "pearson": pearson_corr, "spearmanr": spearman_corr, } @datasets.utils.file_utils.add_start_docstrings(_DESCRIPTION , _KWARGS_DESCRIPTION ) class __snake_case ( datasets.Metric ): def _SCREAMING_SNAKE_CASE ( self ): '''simple docstring''' if self.config_name not in [ "sst2", "mnli", "mnli_mismatched", "mnli_matched", "cola", "stsb", "mrpc", "qqp", "qnli", "rte", "wnli", "hans", ]: raise KeyError( """You should supply a configuration name selected in """ """[\"sst2\", \"mnli\", \"mnli_mismatched\", \"mnli_matched\", """ """\"cola\", \"stsb\", \"mrpc\", \"qqp\", \"qnli\", \"rte\", \"wnli\", \"hans\"]""" ) return datasets.MetricInfo( description=_DESCRIPTION ,citation=_CITATION ,inputs_description=_KWARGS_DESCRIPTION ,features=datasets.Features( { """predictions""": datasets.Value("""int64""" if self.config_name != """stsb""" else """float32""" ), """references""": datasets.Value("""int64""" if self.config_name != """stsb""" else """float32""" ), } ) ,codebase_urls=[] ,reference_urls=[] ,format="""numpy""" ,) def _SCREAMING_SNAKE_CASE ( self ,snake_case ,snake_case ): '''simple docstring''' if self.config_name == "cola": return {"matthews_correlation": matthews_corrcoef(snake_case ,snake_case )} elif self.config_name == "stsb": return pearson_and_spearman(snake_case ,snake_case ) elif self.config_name in ["mrpc", "qqp"]: return acc_and_fa(snake_case ,snake_case ) elif self.config_name in ["sst2", "mnli", "mnli_mismatched", "mnli_matched", "qnli", "rte", "wnli", "hans"]: return {"accuracy": simple_accuracy(snake_case ,snake_case )} else: raise KeyError( """You should supply a configuration name selected in """ """[\"sst2\", \"mnli\", \"mnli_mismatched\", \"mnli_matched\", """ """\"cola\", \"stsb\", \"mrpc\", \"qqp\", \"qnli\", \"rte\", \"wnli\", \"hans\"]""" )
20
0
from typing import TYPE_CHECKING from ...utils import OptionalDependencyNotAvailable, _LazyModule, is_torch_available lowercase : Union[str, Any] = { "configuration_bigbird_pegasus": [ "BIGBIRD_PEGASUS_PRETRAINED_CONFIG_ARCHIVE_MAP", "BigBirdPegasusConfig", "BigBirdPegasusOnnxConfig", ], } try: if not is_torch_available(): raise OptionalDependencyNotAvailable() except OptionalDependencyNotAvailable: pass else: lowercase : Optional[int] = [ "BIGBIRD_PEGASUS_PRETRAINED_MODEL_ARCHIVE_LIST", "BigBirdPegasusForCausalLM", "BigBirdPegasusForConditionalGeneration", "BigBirdPegasusForQuestionAnswering", "BigBirdPegasusForSequenceClassification", "BigBirdPegasusModel", "BigBirdPegasusPreTrainedModel", ] if TYPE_CHECKING: from .configuration_bigbird_pegasus import ( BIGBIRD_PEGASUS_PRETRAINED_CONFIG_ARCHIVE_MAP, BigBirdPegasusConfig, BigBirdPegasusOnnxConfig, ) try: if not is_torch_available(): raise OptionalDependencyNotAvailable() except OptionalDependencyNotAvailable: pass else: from .modeling_bigbird_pegasus import ( BIGBIRD_PEGASUS_PRETRAINED_MODEL_ARCHIVE_LIST, BigBirdPegasusForCausalLM, BigBirdPegasusForConditionalGeneration, BigBirdPegasusForQuestionAnswering, BigBirdPegasusForSequenceClassification, BigBirdPegasusModel, BigBirdPegasusPreTrainedModel, ) else: import sys lowercase : Any = _LazyModule(__name__, globals()["""__file__"""], _import_structure, module_spec=__spec__)
366
def A_ ( A__ , A__ , A__ ) -> float: if principal <= 0: raise Exception('Principal borrowed must be > 0' ) if rate_per_annum < 0: raise Exception('Rate of interest must be >= 0' ) if years_to_repay <= 0 or not isinstance(A__ , A__ ): raise Exception('Years to repay must be an integer > 0' ) # Yearly rate is divided by 12 to get monthly rate a__ : str = rate_per_annum / 12 # Years to repay is multiplied by 12 to get number of payments as payment is monthly a__ : List[Any] = years_to_repay * 12 return ( principal * rate_per_month * (1 + rate_per_month) ** number_of_payments / ((1 + rate_per_month) ** number_of_payments - 1) ) if __name__ == "__main__": import doctest doctest.testmod()
225
0
"""simple docstring""" 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 _a ( _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE ) -> List[Any]: snake_case_ = XCLIPTextConfig() # derive patch size from model name snake_case_ = model_name.find("""patch""" ) snake_case_ = int(model_name[start_idx + len("""patch""" ) : start_idx + len("""patch""" ) + 2] ) snake_case_ = XCLIPVisionConfig(patch_size=_SCREAMING_SNAKE_CASE , num_frames=_SCREAMING_SNAKE_CASE ) if "large" in model_name: snake_case_ = 768 snake_case_ = 3_072 snake_case_ = 12 snake_case_ = 1_024 snake_case_ = 4_096 snake_case_ = 16 snake_case_ = 24 snake_case_ = 768 snake_case_ = 3_072 if model_name == "xclip-large-patch14-16-frames": snake_case_ = 336 snake_case_ = XCLIPConfig.from_text_vision_configs(_SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE ) if "large" in model_name: snake_case_ = 768 return config def _a ( _SCREAMING_SNAKE_CASE ) -> Tuple: # text encoder if name == "token_embedding.weight": snake_case_ = name.replace("""token_embedding.weight""" , """text_model.embeddings.token_embedding.weight""" ) if name == "positional_embedding": snake_case_ = name.replace("""positional_embedding""" , """text_model.embeddings.position_embedding.weight""" ) if "ln_1" in name: snake_case_ = name.replace("""ln_1""" , """layer_norm1""" ) if "ln_2" in name: snake_case_ = name.replace("""ln_2""" , """layer_norm2""" ) if "c_fc" in name: snake_case_ = name.replace("""c_fc""" , """fc1""" ) if "c_proj" in name: snake_case_ = name.replace("""c_proj""" , """fc2""" ) if name.startswith("""transformer.resblocks""" ): snake_case_ = name.replace("""transformer.resblocks""" , """text_model.encoder.layers""" ) if "attn.out_proj" in name and "message" not in name: snake_case_ = name.replace("""attn.out_proj""" , """self_attn.out_proj""" ) if "ln_final" in name: snake_case_ = name.replace("""ln_final""" , """text_model.final_layer_norm""" ) # visual encoder if name == "visual.class_embedding": snake_case_ = name.replace("""visual.class_embedding""" , """vision_model.embeddings.class_embedding""" ) if name == "visual.positional_embedding": snake_case_ = name.replace("""visual.positional_embedding""" , """vision_model.embeddings.position_embedding.weight""" ) if name.startswith("""visual.transformer.resblocks""" ): snake_case_ = name.replace("""visual.transformer.resblocks""" , """vision_model.encoder.layers""" ) if "visual.conv1" in name: snake_case_ = name.replace("""visual.conv1""" , """vision_model.embeddings.patch_embedding""" ) if "visual.ln_pre" in name: snake_case_ = name.replace("""visual.ln_pre""" , """vision_model.pre_layernorm""" ) if "visual.ln_post" in name: snake_case_ = name.replace("""visual.ln_post""" , """vision_model.post_layernorm""" ) if "visual.proj" in name: snake_case_ = name.replace("""visual.proj""" , """visual_projection.weight""" ) if "text_projection" in name: snake_case_ = name.replace("""text_projection""" , """text_projection.weight""" ) # things on top if "prompts_visual_proj" in name: snake_case_ = name.replace("""prompts_visual_proj""" , """prompts_visual_projection""" ) if "prompts_visual_ln" in name: snake_case_ = name.replace("""prompts_visual_ln""" , """prompts_visual_layernorm""" ) # mit if name == "mit.positional_embedding": snake_case_ = name.replace("""positional""" , """position""" ) if name.startswith("""mit.resblocks""" ): snake_case_ = name.replace("""mit.resblocks""" , """mit.encoder.layers""" ) # prompts generator if name.startswith("""prompts_generator.norm""" ): snake_case_ = name.replace("""prompts_generator.norm""" , """prompts_generator.layernorm""" ) return name def _a ( _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE ) -> Optional[Any]: for key in orig_state_dict.copy().keys(): snake_case_ = orig_state_dict.pop(_SCREAMING_SNAKE_CASE ) if "attn.in_proj" in key: snake_case_ = key.split(""".""" ) if key.startswith("""visual""" ): snake_case_ = key_split[3] snake_case_ = config.vision_config.hidden_size if "message_attn" in key: if "weight" in key: snake_case_ = val[ :dim, : ] snake_case_ = val[ dim : dim * 2, : ] snake_case_ = val[ -dim:, : ] else: snake_case_ = val[ :dim ] snake_case_ = val[ dim : dim * 2 ] snake_case_ = val[ -dim: ] else: if "weight" in key: snake_case_ = val[ :dim, : ] snake_case_ = val[ dim : dim * 2, : ] snake_case_ = val[ -dim:, : ] else: snake_case_ = val[:dim] snake_case_ = val[ dim : dim * 2 ] snake_case_ = val[-dim:] elif key.startswith("""mit""" ): snake_case_ = key_split[2] snake_case_ = config.vision_config.mit_hidden_size if "weight" in key: snake_case_ = val[:dim, :] snake_case_ = val[dim : dim * 2, :] snake_case_ = val[-dim:, :] else: snake_case_ = val[:dim] snake_case_ = val[dim : dim * 2] snake_case_ = val[-dim:] else: snake_case_ = key_split[2] snake_case_ = config.text_config.hidden_size if "weight" in key: snake_case_ = val[:dim, :] snake_case_ = val[ dim : dim * 2, : ] snake_case_ = val[-dim:, :] else: snake_case_ = val[:dim] snake_case_ = val[ dim : dim * 2 ] snake_case_ = val[-dim:] else: snake_case_ = rename_key(_SCREAMING_SNAKE_CASE ) if new_key_name in ["visual_projection.weight", "text_projection.weight"]: snake_case_ = val.T snake_case_ = val return orig_state_dict def _a ( _SCREAMING_SNAKE_CASE ) -> Optional[int]: if num_frames == 8: snake_case_ = """eating_spaghetti_8_frames.npy""" elif num_frames == 16: snake_case_ = """eating_spaghetti.npy""" elif num_frames == 32: snake_case_ = """eating_spaghetti_32_frames.npy""" snake_case_ = hf_hub_download( repo_id="""hf-internal-testing/spaghetti-video""" , filename=_SCREAMING_SNAKE_CASE , repo_type="""dataset""" , ) snake_case_ = np.load(_SCREAMING_SNAKE_CASE ) return list(_SCREAMING_SNAKE_CASE ) def _a ( _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE=None , _SCREAMING_SNAKE_CASE=False ) -> Tuple: snake_case_ = { # 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""", } snake_case_ = model_to_url[model_name] snake_case_ = 8 if "16-frames" in model_name: snake_case_ = 16 elif "shot" in model_name: snake_case_ = 32 snake_case_ = get_xclip_config(_SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE ) snake_case_ = XCLIPModel(_SCREAMING_SNAKE_CASE ) model.eval() if "drive" in checkpoint_url: snake_case_ = """pytorch_model.bin""" gdown.cached_download(_SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , quiet=_SCREAMING_SNAKE_CASE ) snake_case_ = torch.load(_SCREAMING_SNAKE_CASE , map_location="""cpu""" )["""model"""] else: snake_case_ = torch.hub.load_state_dict_from_url(_SCREAMING_SNAKE_CASE )["""model"""] snake_case_ = convert_state_dict(_SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE ) snake_case_ = XCLIPModel(_SCREAMING_SNAKE_CASE ) snake_case_ , snake_case_ = model.load_state_dict(_SCREAMING_SNAKE_CASE , strict=_SCREAMING_SNAKE_CASE ) assert missing_keys == ["text_model.embeddings.position_ids", "vision_model.embeddings.position_ids"] model.eval() snake_case_ = 336 if model_name == """xclip-large-patch14-16-frames""" else 224 snake_case_ = VideoMAEImageProcessor(size=_SCREAMING_SNAKE_CASE ) snake_case_ = CLIPTokenizer.from_pretrained("""openai/clip-vit-base-patch32""" ) snake_case_ = CLIPTokenizerFast.from_pretrained("""openai/clip-vit-base-patch32""" ) snake_case_ = XCLIPProcessor(image_processor=_SCREAMING_SNAKE_CASE , tokenizer=_SCREAMING_SNAKE_CASE ) snake_case_ = prepare_video(_SCREAMING_SNAKE_CASE ) snake_case_ = processor( text=["""playing sports""", """eating spaghetti""", """go shopping"""] , videos=_SCREAMING_SNAKE_CASE , return_tensors="""pt""" , padding=_SCREAMING_SNAKE_CASE ) print("""Shape of pixel values:""" , inputs.pixel_values.shape ) with torch.no_grad(): snake_case_ = model(**_SCREAMING_SNAKE_CASE ) # Verify outputs snake_case_ = outputs.logits_per_video snake_case_ = logits_per_video.softmax(dim=1 ) print("""Probs:""" , _SCREAMING_SNAKE_CASE ) # kinetics-400 if model_name == "xclip-base-patch32": snake_case_ = torch.tensor([[0.0019, 0.9951, 0.0030]] ) elif model_name == "xclip-base-patch32-16-frames": snake_case_ = torch.tensor([[7.0_9_9_9E-0_4, 9.9_8_8_3E-0_1, 4.5_5_8_0E-0_4]] ) elif model_name == "xclip-base-patch16": snake_case_ = torch.tensor([[0.0083, 0.9681, 0.0236]] ) elif model_name == "xclip-base-patch16-16-frames": snake_case_ = torch.tensor([[7.6_9_3_7E-0_4, 9.9_7_2_8E-0_1, 1.9_4_7_3E-0_3]] ) elif model_name == "xclip-large-patch14": snake_case_ = torch.tensor([[0.0062, 0.9864, 0.0075]] ) elif model_name == "xclip-large-patch14-16-frames": snake_case_ = torch.tensor([[3.3_8_7_7E-0_4, 9.9_9_3_7E-0_1, 2.8_8_8_8E-0_4]] ) # kinetics-600 elif model_name == "xclip-base-patch16-kinetics-600": snake_case_ = torch.tensor([[0.0555, 0.8914, 0.0531]] ) elif model_name == "xclip-base-patch16-kinetics-600-16-frames": snake_case_ = torch.tensor([[3.8_5_5_4E-0_4, 9.9_9_2_9E-0_1, 3.2_7_5_4E-0_4]] ) elif model_name == "xclip-large-patch14-kinetics-600": snake_case_ = torch.tensor([[0.0036, 0.9920, 0.0045]] ) # few shot elif model_name == "xclip-base-patch16-hmdb-2-shot": snake_case_ = torch.tensor([[7.1_8_9_0E-0_6, 9.9_9_9_4E-0_1, 5.6_5_5_9E-0_5]] ) elif model_name == "xclip-base-patch16-hmdb-4-shot": snake_case_ = torch.tensor([[1.0_3_2_0E-0_5, 9.9_9_9_3E-0_1, 6.2_4_3_5E-0_5]] ) elif model_name == "xclip-base-patch16-hmdb-8-shot": snake_case_ = torch.tensor([[4.1_3_7_7E-0_6, 9.9_9_9_0E-0_1, 9.8_3_8_6E-0_5]] ) elif model_name == "xclip-base-patch16-hmdb-16-shot": snake_case_ = torch.tensor([[4.1_3_4_7E-0_5, 9.9_9_6_2E-0_1, 3.3_4_1_1E-0_4]] ) elif model_name == "xclip-base-patch16-ucf-2-shot": snake_case_ = torch.tensor([[8.5_8_5_7E-0_5, 9.9_9_2_8E-0_1, 6.3_2_9_1E-0_4]] ) elif model_name == "xclip-base-patch16-ucf-4-shot": snake_case_ = torch.tensor([[8.5_8_5_7E-0_5, 9.9_9_2_8E-0_1, 6.3_2_9_1E-0_4]] ) elif model_name == "xclip-base-patch16-ucf-8-shot": snake_case_ = torch.tensor([[0.0027, 0.9904, 0.0070]] ) elif model_name == "xclip-base-patch16-ucf-16-shot": snake_case_ = torch.tensor([[9.8_2_1_9E-0_4, 9.9_5_9_3E-0_1, 3.0_8_6_3E-0_3]] ) # zero shot elif model_name == "xclip-base-patch16-zero-shot": snake_case_ = torch.tensor([[3.5_0_8_2E-0_4, 9.9_7_8_5E-0_1, 1.7_9_6_6E-0_3]] ) else: raise ValueError(f"""Model name {model_name} not supported""" ) assert torch.allclose(_SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , 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(_SCREAMING_SNAKE_CASE ) if push_to_hub: print("""Pushing model, processor and slow tokenizer files to the hub...""" ) model.push_to_hub(_SCREAMING_SNAKE_CASE , organization="""nielsr""" ) processor.push_to_hub(_SCREAMING_SNAKE_CASE , organization="""nielsr""" ) slow_tokenizer.push_to_hub(_SCREAMING_SNAKE_CASE , organization="""nielsr""" ) if __name__ == "__main__": __SCREAMING_SNAKE_CASE : Union[str, Any] = 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.' ) __SCREAMING_SNAKE_CASE : Any = parser.parse_args() convert_xclip_checkpoint(args.model_name, args.pytorch_dump_folder_path, args.push_to_hub)
347
"""simple docstring""" # this script reports modified .py files under the desired list of top-level sub-dirs passed as a list of arguments, e.g.: # python ./utils/get_modified_files.py utils src tests examples # # it uses git to find the forking point and which files were modified - i.e. files not under git won't be considered # since the output of this script is fed into Makefile commands it doesn't print a newline after the results import re import subprocess import sys __SCREAMING_SNAKE_CASE : Tuple = subprocess.check_output('git merge-base main HEAD'.split()).decode('utf-8') __SCREAMING_SNAKE_CASE : Tuple = subprocess.check_output(f"""git diff --name-only {fork_point_sha}""".split()).decode('utf-8').split() __SCREAMING_SNAKE_CASE : Any = '|'.join(sys.argv[1:]) __SCREAMING_SNAKE_CASE : Optional[Any] = re.compile(Rf"""^({joined_dirs}).*?\.py$""") __SCREAMING_SNAKE_CASE : List[str] = [x for x in modified_files if regex.match(x)] print(' '.join(relevant_modified_files), end='')
347
1
"""simple docstring""" import random import timeit from functools import wraps from typing import Callable, Optional from ..configuration_utils import PretrainedConfig from ..models.auto.modeling_tf_auto import TF_MODEL_MAPPING, TF_MODEL_WITH_LM_HEAD_MAPPING from ..utils import is_pyanvml_available, is_tf_available, logging from .benchmark_utils import ( Benchmark, Memory, MemorySummary, measure_peak_memory_cpu, start_memory_tracing, stop_memory_tracing, ) if is_tf_available(): import tensorflow as tf from tensorflow.python.framework.errors_impl import ResourceExhaustedError from .benchmark_args_tf import TensorFlowBenchmarkArguments if is_pyanvml_available(): import pyanvml.pyanvml as nvml A_ = logging.get_logger(__name__) def UpperCAmelCase__ (snake_case__ : bool , snake_case__ : bool ): """simple docstring""" def run_func(snake_case__ : List[Any] ): @wraps(_SCREAMING_SNAKE_CASE ) def run_in_eager_mode(*snake_case__ : List[Any] , **snake_case__ : Union[str, Any] ): return func(*_SCREAMING_SNAKE_CASE , **_SCREAMING_SNAKE_CASE ) @wraps(_SCREAMING_SNAKE_CASE ) @tf.function(experimental_compile=_SCREAMING_SNAKE_CASE ) def run_in_graph_mode(*snake_case__ : int , **snake_case__ : List[Any] ): return func(*_SCREAMING_SNAKE_CASE , **_SCREAMING_SNAKE_CASE ) if do_eager_mode is True: if use_xla is not False: raise ValueError( """Cannot run model in XLA, if `args.eager_mode` is set to `True`. Please set `args.eager_mode=False`.""" ) return run_in_eager_mode else: return run_in_graph_mode return run_func def UpperCAmelCase__ (snake_case__ : int , snake_case__ : int , snake_case__ : int ): """simple docstring""" _snake_case : str = random.Random() _snake_case : List[Any] = [rng.randint(0 , vocab_size - 1 ) for i in range(batch_size * sequence_length )] return tf.constant(_SCREAMING_SNAKE_CASE , shape=(batch_size, sequence_length) , dtype=tf.intaa ) class lowercase( __SCREAMING_SNAKE_CASE ): '''simple docstring''' lowercase__ = 42 lowercase__ = 42 lowercase__ = "TensorFlow" @property def UpperCamelCase_ ( self: Optional[Any] ): '''simple docstring''' return tf.__version__ def UpperCamelCase_ ( self: str, a_: Any, a_: str, a_: Tuple ): '''simple docstring''' _snake_case : Optional[int] = self.args.strategy if strategy is None: raise ValueError("""A device strategy has to be initialized before using TensorFlow.""" ) _snake_case : str = self._prepare_inference_func(_snake_case, _snake_case, _snake_case ) return self._measure_speed(_inference ) def UpperCamelCase_ ( self: Tuple, a_: Any, a_: str, a_: List[Any] ): '''simple docstring''' _snake_case : List[str] = self.args.strategy if strategy is None: raise ValueError("""A device strategy has to be initialized before using TensorFlow.""" ) _snake_case : List[Any] = self._prepare_train_func(_snake_case, _snake_case, _snake_case ) return self._measure_speed(_train ) def UpperCamelCase_ ( self: Dict, a_: Dict, a_: List[Any], a_: str ): '''simple docstring''' if self.args.is_gpu: tf.config.experimental.set_memory_growth(self.args.gpu_list[self.args.device_idx], _snake_case ) _snake_case : Dict = self.args.strategy if strategy is None: raise ValueError("""A device strategy has to be initialized before using TensorFlow.""" ) _snake_case : Tuple = self._prepare_inference_func(_snake_case, _snake_case, _snake_case ) return self._measure_memory(_inference ) def UpperCamelCase_ ( self: Any, a_: Any, a_: Tuple, a_: Tuple ): '''simple docstring''' if self.args.is_gpu: tf.config.experimental.set_memory_growth(self.args.gpu_list[self.args.device_idx], _snake_case ) _snake_case : int = self.args.strategy if strategy is None: raise ValueError("""A device strategy has to be initialized before using TensorFlow.""" ) _snake_case : Dict = self._prepare_train_func(_snake_case, _snake_case, _snake_case ) return self._measure_memory(_train ) def UpperCamelCase_ ( self: str, a_: Optional[Any], a_: Dict, a_: Any ): '''simple docstring''' _snake_case : List[Any] = self.config_dict[model_name] if self.args.fpaa: raise NotImplementedError("""Mixed precision is currently not supported.""" ) _snake_case : int = ( hasattr(_snake_case, """architectures""" ) and isinstance(config.architectures, _snake_case ) and len(config.architectures ) > 0 ) if not self.args.only_pretrain_model and has_model_class_in_config: try: _snake_case : Tuple = "TF" + config.architectures[0] # prepend 'TF' for tensorflow model _snake_case : Any = __import__("""transformers""", fromlist=[model_class] ) _snake_case : Any = getattr(_snake_case, _snake_case ) _snake_case : List[Any] = model_cls(_snake_case ) except ImportError: raise ImportError( f"{model_class} does not exist. If you just want to test the pretrained model, you might want to" """ set `--only_pretrain_model` or `args.only_pretrain_model=True`.""" ) else: _snake_case : int = TF_MODEL_MAPPING[config.__class__](_snake_case ) # encoder-decoder has vocab size saved differently _snake_case : List[Any] = config.vocab_size if hasattr(_snake_case, """vocab_size""" ) else config.encoder.vocab_size _snake_case : List[str] = random_input_ids(_snake_case, _snake_case, _snake_case ) @run_with_tf_optimizations(self.args.eager_mode, self.args.use_xla ) def encoder_decoder_forward(): return model(_snake_case, decoder_input_ids=_snake_case, training=_snake_case ) @run_with_tf_optimizations(self.args.eager_mode, self.args.use_xla ) def encoder_forward(): return model(_snake_case, training=_snake_case ) _snake_case : Optional[int] = encoder_decoder_forward if config.is_encoder_decoder else encoder_forward return _inference def UpperCamelCase_ ( self: Dict, a_: Tuple, a_: Union[str, Any], a_: Dict ): '''simple docstring''' _snake_case : Dict = self.config_dict[model_name] if self.args.eager_mode is not False: raise ValueError("""Training cannot be done in eager mode. Please make sure that `args.eager_mode = False`.""" ) if self.args.fpaa: raise NotImplementedError("""Mixed precision is currently not supported.""" ) _snake_case : str = ( hasattr(_snake_case, """architectures""" ) and isinstance(config.architectures, _snake_case ) and len(config.architectures ) > 0 ) if not self.args.only_pretrain_model and has_model_class_in_config: try: _snake_case : List[str] = "TF" + config.architectures[0] # prepend 'TF' for tensorflow model _snake_case : Optional[int] = __import__("""transformers""", fromlist=[model_class] ) _snake_case : List[str] = getattr(_snake_case, _snake_case ) _snake_case : Dict = model_cls(_snake_case ) except ImportError: raise ImportError( f"{model_class} does not exist. If you just want to test the pretrained model, you might want to" """ set `--only_pretrain_model` or `args.only_pretrain_model=True`.""" ) else: _snake_case : Optional[int] = TF_MODEL_WITH_LM_HEAD_MAPPING[config.__class__](_snake_case ) # encoder-decoder has vocab size saved differently _snake_case : Optional[Any] = config.vocab_size if hasattr(_snake_case, """vocab_size""" ) else config.encoder.vocab_size _snake_case : Tuple = random_input_ids(_snake_case, _snake_case, _snake_case ) @run_with_tf_optimizations(self.args.eager_mode, self.args.use_xla ) def encoder_decoder_train(): _snake_case : List[str] = model(_snake_case, decoder_input_ids=_snake_case, labels=_snake_case, training=_snake_case )[0] _snake_case : int = tf.gradients(_snake_case, model.trainable_variables ) return gradients @run_with_tf_optimizations(self.args.eager_mode, self.args.use_xla ) def encoder_train(): _snake_case : Dict = model(_snake_case, labels=_snake_case, training=_snake_case )[0] _snake_case : Any = tf.gradients(_snake_case, model.trainable_variables ) return gradients _snake_case : Dict = encoder_decoder_train if config.is_encoder_decoder else encoder_train return _train def UpperCamelCase_ ( self: Any, a_: List[str] ): '''simple docstring''' with self.args.strategy.scope(): try: if self.args.is_tpu or self.args.use_xla: # run additional 10 times to stabilize compilation for tpu logger.info("""Do inference on TPU. Running model 5 times to stabilize compilation""" ) timeit.repeat(_snake_case, repeat=1, number=5 ) # as written in https://docs.python.org/2/library/timeit.html#timeit.Timer.repeat, min should be taken rather than the average _snake_case : Dict = timeit.repeat( _snake_case, repeat=self.args.repeat, number=10, ) return min(_snake_case ) / 10.0 except ResourceExhaustedError as e: self.print_fn(f"Doesn\'t fit on GPU. {e}" ) def UpperCamelCase_ ( self: Tuple, a_: Union[str, Any] ): '''simple docstring''' logger.info( """Note that TensorFlow allocates more memory than """ """it might need to speed up computation. """ """The memory reported here corresponds to the memory """ """reported by `nvidia-smi`, which can vary depending """ """on total available memory on the GPU that is used.""" ) with self.args.strategy.scope(): try: if self.args.trace_memory_line_by_line: if not self.args.eager_mode: raise ValueError( """`args.eager_mode` is set to `False`. Make sure to run model in eager mode to measure memory""" """ consumption line by line.""" ) _snake_case : List[Any] = start_memory_tracing("""transformers""" ) if self.args.is_tpu: # tpu raise NotImplementedError( """Memory Benchmarking is currently not implemented for TPU. Please disable memory benchmarking""" """ with `args.memory=False`""" ) elif self.args.is_gpu: # gpu if not is_pyanvml_available(): logger.warning( """py3nvml not installed, we won't log GPU memory usage. """ """Install py3nvml (pip install py3nvml) to log information about GPU.""" ) _snake_case : Any = "N/A" else: logger.info( """Measuring total GPU usage on GPU device. Make sure to not have additional processes""" """ running on the same GPU.""" ) # init nvml nvml.nvmlInit() func() _snake_case : Optional[Any] = nvml.nvmlDeviceGetHandleByIndex(self.args.device_idx ) _snake_case : Optional[int] = nvml.nvmlDeviceGetMemoryInfo(_snake_case ) _snake_case : Union[str, Any] = meminfo.used _snake_case : List[Any] = Memory(_snake_case ) # shutdown nvml nvml.nvmlShutdown() else: # cpu if self.args.trace_memory_line_by_line: logger.info( """When enabling line by line tracing, the max peak memory for CPU is inaccurate in""" """ TensorFlow.""" ) _snake_case : Any = None else: _snake_case : Any = measure_peak_memory_cpu(_snake_case ) _snake_case : Dict = Memory(_snake_case ) if isinstance(_snake_case, _snake_case ) else memory_bytes if self.args.trace_memory_line_by_line: _snake_case : int = stop_memory_tracing(_snake_case ) if memory is None: _snake_case : int = summary.total else: _snake_case : Optional[int] = None return memory, summary except ResourceExhaustedError as e: self.print_fn(f"Doesn\'t fit on GPU. {e}" ) return "N/A", None
351
"""simple docstring""" from collections.abc import Generator from math import sin def UpperCAmelCase__ (snake_case__ : bytes ): """simple docstring""" if len(snake_case__ ) != 32: raise ValueError("""Input must be of length 32""" ) _snake_case : Optional[int] = B"""""" for i in [3, 2, 1, 0]: little_endian += string_aa[8 * i : 8 * i + 8] return little_endian def UpperCAmelCase__ (snake_case__ : int ): """simple docstring""" if i < 0: raise ValueError("""Input must be non-negative""" ) _snake_case : Optional[Any] = format(snake_case__ , """08x""" )[-8:] _snake_case : Optional[Any] = B"""""" for i in [3, 2, 1, 0]: little_endian_hex += hex_rep[2 * i : 2 * i + 2].encode("""utf-8""" ) return little_endian_hex def UpperCAmelCase__ (snake_case__ : bytes ): """simple docstring""" _snake_case : Union[str, Any] = B"""""" for char in message: bit_string += format(snake_case__ , """08b""" ).encode("""utf-8""" ) _snake_case : List[Any] = format(len(snake_case__ ) , """064b""" ).encode("""utf-8""" ) # Pad bit_string to a multiple of 512 chars bit_string += b"1" while len(snake_case__ ) % 5_12 != 4_48: bit_string += b"0" bit_string += to_little_endian(start_len[32:] ) + to_little_endian(start_len[:32] ) return bit_string def UpperCAmelCase__ (snake_case__ : bytes ): """simple docstring""" if len(snake_case__ ) % 5_12 != 0: raise ValueError("""Input must have length that's a multiple of 512""" ) for pos in range(0 , len(snake_case__ ) , 5_12 ): _snake_case : List[str] = bit_string[pos : pos + 5_12] _snake_case : List[str] = [] for i in range(0 , 5_12 , 32 ): block_words.append(int(to_little_endian(block[i : i + 32] ) , 2 ) ) yield block_words def UpperCAmelCase__ (snake_case__ : int ): """simple docstring""" if i < 0: raise ValueError("""Input must be non-negative""" ) _snake_case : Optional[int] = format(snake_case__ , """032b""" ) _snake_case : str = """""" for c in i_str: new_str += "1" if c == "0" else "0" return int(snake_case__ , 2 ) def UpperCAmelCase__ (snake_case__ : int , snake_case__ : int ): """simple docstring""" return (a + b) % 2**32 def UpperCAmelCase__ (snake_case__ : int , snake_case__ : int ): """simple docstring""" if i < 0: raise ValueError("""Input must be non-negative""" ) if shift < 0: raise ValueError("""Shift must be non-negative""" ) return ((i << shift) ^ (i >> (32 - shift))) % 2**32 def UpperCAmelCase__ (snake_case__ : bytes ): """simple docstring""" _snake_case : Any = preprocess(snake_case__ ) _snake_case : Optional[Any] = [int(2**32 * abs(sin(i + 1 ) ) ) for i in range(64 )] # Starting states _snake_case : Union[str, Any] = 0x6745_2301 _snake_case : List[Any] = 0xEFCD_AB89 _snake_case : Optional[Any] = 0x98BA_DCFE _snake_case : Optional[int] = 0x1032_5476 _snake_case : Tuple = [ 7, 12, 17, 22, 7, 12, 17, 22, 7, 12, 17, 22, 7, 12, 17, 22, 5, 9, 14, 20, 5, 9, 14, 20, 5, 9, 14, 20, 5, 9, 14, 20, 4, 11, 16, 23, 4, 11, 16, 23, 4, 11, 16, 23, 4, 11, 16, 23, 6, 10, 15, 21, 6, 10, 15, 21, 6, 10, 15, 21, 6, 10, 15, 21, ] # Process bit string in chunks, each with 16 32-char words for block_words in get_block_words(snake_case__ ): _snake_case : Tuple = aa _snake_case : str = ba _snake_case : int = ca _snake_case : Dict = da # Hash current chunk for i in range(64 ): if i <= 15: # f = (b & c) | (not_32(b) & d) # Alternate definition for f _snake_case : int = d ^ (b & (c ^ d)) _snake_case : Dict = i elif i <= 31: # f = (d & b) | (not_32(d) & c) # Alternate definition for f _snake_case : Tuple = c ^ (d & (b ^ c)) _snake_case : int = (5 * i + 1) % 16 elif i <= 47: _snake_case : List[Any] = b ^ c ^ d _snake_case : Union[str, Any] = (3 * i + 5) % 16 else: _snake_case : Tuple = c ^ (b | not_aa(snake_case__ )) _snake_case : Optional[int] = (7 * i) % 16 _snake_case : Dict = (f + a + added_consts[i] + block_words[g]) % 2**32 _snake_case : List[str] = d _snake_case : List[Any] = c _snake_case : str = b _snake_case : List[str] = sum_aa(snake_case__ , left_rotate_aa(snake_case__ , shift_amounts[i] ) ) # Add hashed chunk to running total _snake_case : Union[str, Any] = sum_aa(snake_case__ , snake_case__ ) _snake_case : str = sum_aa(snake_case__ , snake_case__ ) _snake_case : Any = sum_aa(snake_case__ , snake_case__ ) _snake_case : List[str] = sum_aa(snake_case__ , snake_case__ ) _snake_case : Any = reformat_hex(snake_case__ ) + reformat_hex(snake_case__ ) + reformat_hex(snake_case__ ) + reformat_hex(snake_case__ ) return digest if __name__ == "__main__": import doctest doctest.testmod()
132
0
from ...configuration_utils import PretrainedConfig from ...utils import logging _a = logging.get_logger(__name__) _a = { '''funnel-transformer/small''': '''https://huggingface.co/funnel-transformer/small/resolve/main/config.json''', '''funnel-transformer/small-base''': '''https://huggingface.co/funnel-transformer/small-base/resolve/main/config.json''', '''funnel-transformer/medium''': '''https://huggingface.co/funnel-transformer/medium/resolve/main/config.json''', '''funnel-transformer/medium-base''': '''https://huggingface.co/funnel-transformer/medium-base/resolve/main/config.json''', '''funnel-transformer/intermediate''': ( '''https://huggingface.co/funnel-transformer/intermediate/resolve/main/config.json''' ), '''funnel-transformer/intermediate-base''': ( '''https://huggingface.co/funnel-transformer/intermediate-base/resolve/main/config.json''' ), '''funnel-transformer/large''': '''https://huggingface.co/funnel-transformer/large/resolve/main/config.json''', '''funnel-transformer/large-base''': '''https://huggingface.co/funnel-transformer/large-base/resolve/main/config.json''', '''funnel-transformer/xlarge''': '''https://huggingface.co/funnel-transformer/xlarge/resolve/main/config.json''', '''funnel-transformer/xlarge-base''': '''https://huggingface.co/funnel-transformer/xlarge-base/resolve/main/config.json''', } class __lowerCamelCase ( snake_case__): """simple docstring""" UpperCamelCase__ = "funnel" UpperCamelCase__ = { "hidden_size": "d_model", "num_attention_heads": "n_head", } def __init__( self , UpperCAmelCase=3_0522 , UpperCAmelCase=[4, 4, 4] , UpperCAmelCase=None , UpperCAmelCase=2 , UpperCAmelCase=768 , UpperCAmelCase=12 , UpperCAmelCase=64 , UpperCAmelCase=3072 , UpperCAmelCase="gelu_new" , UpperCAmelCase=0.1 , UpperCAmelCase=0.1 , UpperCAmelCase=0.0 , UpperCAmelCase=0.1 , UpperCAmelCase=None , UpperCAmelCase=1e-9 , UpperCAmelCase="mean" , UpperCAmelCase="relative_shift" , UpperCAmelCase=True , UpperCAmelCase=True , UpperCAmelCase=True , **UpperCAmelCase , ): """simple docstring""" _UpperCAmelCase = vocab_size _UpperCAmelCase = block_sizes _UpperCAmelCase = [1] * len(UpperCAmelCase ) if block_repeats is None else block_repeats assert len(UpperCAmelCase ) == len( self.block_repeats ), "`block_sizes` and `block_repeats` should have the same length." _UpperCAmelCase = num_decoder_layers _UpperCAmelCase = d_model _UpperCAmelCase = n_head _UpperCAmelCase = d_head _UpperCAmelCase = d_inner _UpperCAmelCase = hidden_act _UpperCAmelCase = hidden_dropout _UpperCAmelCase = attention_dropout _UpperCAmelCase = activation_dropout _UpperCAmelCase = initializer_range _UpperCAmelCase = initializer_std _UpperCAmelCase = layer_norm_eps assert pooling_type in [ "mean", "max", ], F"""Got {pooling_type} for `pooling_type` but only 'mean' and 'max' are supported.""" _UpperCAmelCase = pooling_type assert attention_type in [ "relative_shift", "factorized", ], F"""Got {attention_type} for `attention_type` but only 'relative_shift' and 'factorized' are supported.""" _UpperCAmelCase = attention_type _UpperCAmelCase = separate_cls _UpperCAmelCase = truncate_seq _UpperCAmelCase = pool_q_only super().__init__(**UpperCAmelCase ) @property def UpperCamelCase ( self ): """simple docstring""" return sum(self.block_sizes ) @num_hidden_layers.setter def UpperCamelCase ( self , UpperCAmelCase ): """simple docstring""" raise NotImplementedError( 'This model does not support the setting of `num_hidden_layers`. Please set `block_sizes`.' ) @property def UpperCamelCase ( self ): """simple docstring""" return len(self.block_sizes ) @num_blocks.setter def UpperCamelCase ( self , UpperCAmelCase ): """simple docstring""" raise NotImplementedError('This model does not support the setting of `num_blocks`. Please set `block_sizes`.' )
39
class __lowerCamelCase : """simple docstring""" def __init__( self , UpperCAmelCase = "" , UpperCAmelCase = False ): """simple docstring""" _UpperCAmelCase = {} # A node will be a leaf if the tree contains its word _UpperCAmelCase = is_leaf _UpperCAmelCase = prefix def UpperCamelCase ( self , UpperCAmelCase ): """simple docstring""" _UpperCAmelCase = 0 for q, w in zip(self.prefix , UpperCAmelCase ): if q != w: break x += 1 return self.prefix[:x], self.prefix[x:], word[x:] def UpperCamelCase ( self , UpperCAmelCase ): """simple docstring""" for word in words: self.insert(UpperCAmelCase ) def UpperCamelCase ( self , UpperCAmelCase ): """simple docstring""" if self.prefix == word: _UpperCAmelCase = True # Case 2: The node has no edges that have a prefix to the word # Solution: We create an edge from the current node to a new one # containing the word elif word[0] not in self.nodes: _UpperCAmelCase = RadixNode(prefix=UpperCAmelCase , is_leaf=UpperCAmelCase ) else: _UpperCAmelCase = self.nodes[word[0]] _UpperCAmelCase , _UpperCAmelCase , _UpperCAmelCase = incoming_node.match( UpperCAmelCase ) # Case 3: The node prefix is equal to the matching # Solution: We insert remaining word on the next node if remaining_prefix == "": self.nodes[matching_string[0]].insert(UpperCAmelCase ) # Case 4: The word is greater equal to the matching # Solution: Create a node in between both nodes, change # prefixes and add the new node for the remaining word else: _UpperCAmelCase = remaining_prefix _UpperCAmelCase = self.nodes[matching_string[0]] _UpperCAmelCase = RadixNode(UpperCAmelCase , UpperCAmelCase ) _UpperCAmelCase = aux_node if remaining_word == "": _UpperCAmelCase = True else: self.nodes[matching_string[0]].insert(UpperCAmelCase ) def UpperCamelCase ( self , UpperCAmelCase ): """simple docstring""" _UpperCAmelCase = self.nodes.get(word[0] , UpperCAmelCase ) if not incoming_node: return False else: _UpperCAmelCase , _UpperCAmelCase , _UpperCAmelCase = incoming_node.match( UpperCAmelCase ) # If there is remaining prefix, the word can't be on the tree if remaining_prefix != "": return False # This applies when the word and the prefix are equal elif remaining_word == "": return incoming_node.is_leaf # We have word remaining so we check the next node else: return incoming_node.find(UpperCAmelCase ) def UpperCamelCase ( self , UpperCAmelCase ): """simple docstring""" _UpperCAmelCase = self.nodes.get(word[0] , UpperCAmelCase ) if not incoming_node: return False else: _UpperCAmelCase , _UpperCAmelCase , _UpperCAmelCase = incoming_node.match( UpperCAmelCase ) # If there is remaining prefix, the word can't be on the tree if remaining_prefix != "": return False # We have word remaining so we check the next node elif remaining_word != "": return incoming_node.delete(UpperCAmelCase ) else: # If it is not a leaf, we don't have to delete if not incoming_node.is_leaf: return False else: # We delete the nodes if no edges go from it if len(incoming_node.nodes ) == 0: del self.nodes[word[0]] # We merge the current node with its only child if len(self.nodes ) == 1 and not self.is_leaf: _UpperCAmelCase = list(self.nodes.values() )[0] _UpperCAmelCase = merging_node.is_leaf self.prefix += merging_node.prefix _UpperCAmelCase = merging_node.nodes # If there is more than 1 edge, we just mark it as non-leaf elif len(incoming_node.nodes ) > 1: _UpperCAmelCase = False # If there is 1 edge, we merge it with its child else: _UpperCAmelCase = list(incoming_node.nodes.values() )[0] _UpperCAmelCase = merging_node.is_leaf incoming_node.prefix += merging_node.prefix _UpperCAmelCase = merging_node.nodes return True def UpperCamelCase ( self , UpperCAmelCase = 0 ): """simple docstring""" if self.prefix != "": print('-' * height , self.prefix , ' (leaf)' if self.is_leaf else '' ) for value in self.nodes.values(): value.print_tree(height + 1 ) def __A ( )-> bool: """simple docstring""" _UpperCAmelCase = 'banana bananas bandana band apple all beast'.split() _UpperCAmelCase = RadixNode() root.insert_many(__lowerCAmelCase ) assert all(root.find(__lowerCAmelCase ) for word in words ) assert not root.find('bandanas' ) assert not root.find('apps' ) root.delete('all' ) assert not root.find('all' ) root.delete('banana' ) assert not root.find('banana' ) assert root.find('bananas' ) return True def __A ( )-> None: """simple docstring""" assert test_trie() def __A ( )-> None: """simple docstring""" _UpperCAmelCase = RadixNode() _UpperCAmelCase = 'banana bananas bandanas bandana band apple all beast'.split() root.insert_many(__lowerCAmelCase ) print('Words:' , __lowerCAmelCase ) print('Tree:' ) root.print_tree() if __name__ == "__main__": main()
39
1
import argparse import json import os import fairseq import torch from fairseq.data import Dictionary from transformers import ( UniSpeechConfig, UniSpeechForCTC, UniSpeechForPreTraining, WavaVecaFeatureExtractor, WavaVecaPhonemeCTCTokenizer, WavaVecaProcessor, logging, ) logging.set_verbosity_info() __a = logging.get_logger(__name__) __a = { '''post_extract_proj''': '''feature_projection.projection''', '''encoder.pos_conv.0''': '''encoder.pos_conv_embed.conv''', '''self_attn.k_proj''': '''encoder.layers.*.attention.k_proj''', '''self_attn.v_proj''': '''encoder.layers.*.attention.v_proj''', '''self_attn.q_proj''': '''encoder.layers.*.attention.q_proj''', '''self_attn.out_proj''': '''encoder.layers.*.attention.out_proj''', '''self_attn_layer_norm''': '''encoder.layers.*.layer_norm''', '''fc1''': '''encoder.layers.*.feed_forward.intermediate_dense''', '''fc2''': '''encoder.layers.*.feed_forward.output_dense''', '''final_layer_norm''': '''encoder.layers.*.final_layer_norm''', '''encoder.layer_norm''': '''encoder.layer_norm''', '''w2v_model.layer_norm''': '''feature_projection.layer_norm''', '''quantizer.weight_proj''': '''quantizer.weight_proj''', '''quantizer.vars''': '''quantizer.codevectors''', '''project_q''': '''project_q''', '''final_proj''': '''project_hid''', '''w2v_encoder.proj''': '''ctc_proj''', '''mask_emb''': '''masked_spec_embed''', } __a = [ '''ctc_proj''', '''quantizer.weight_proj''', '''quantizer.codevectors''', '''project_q''', '''project_hid''', ] def __lowercase ( _UpperCamelCase, _UpperCamelCase, _UpperCamelCase, _UpperCamelCase, _UpperCamelCase, _UpperCamelCase ) ->Optional[int]: """simple docstring""" for attribute in key.split('''.''' ): if is_finetuned: if attribute in ["quantizer", "project_q", "project_hid"]: # those layers are only relevant for pretraining and should be dropped return if attribute == "ctc_proj": # we should rename `ctc_proj` to `lm_head` for fine-tuned phoneme models lowercase : Union[str, Any] = '''lm_head''' lowercase : Optional[Any] = getattr(_UpperCamelCase, _UpperCamelCase ) if weight_type is not None: lowercase : Optional[Any] = getattr(_UpperCamelCase, _UpperCamelCase ).shape else: lowercase : Optional[Any] = hf_pointer.shape assert hf_shape == value.shape, ( 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": lowercase : Any = value elif weight_type == "weight_g": lowercase : Any = value elif weight_type == "weight_v": lowercase : List[Any] = value elif weight_type == "bias": lowercase : List[Any] = value else: lowercase : Tuple = value logger.info(f"""{key + '.' + weight_type if weight_type is not None else ''} was initialized from {full_name}.""" ) def __lowercase ( _UpperCamelCase, _UpperCamelCase, _UpperCamelCase ) ->Union[str, Any]: """simple docstring""" lowercase : Union[str, Any] = [] lowercase : Any = fairseq_model.state_dict() lowercase : Optional[int] = hf_model.unispeech.feature_extractor for name, value in fairseq_dict.items(): lowercase : int = False if "conv_layers" in name: load_conv_layer( _UpperCamelCase, _UpperCamelCase, _UpperCamelCase, _UpperCamelCase, hf_model.config.feat_extract_norm == '''group''', ) lowercase : List[str] = True else: for key, mapped_key in MAPPING.items(): lowercase : Tuple = '''unispeech.''' + mapped_key if mapped_key not in TOP_LEVEL_KEYS else mapped_key if key in name or key.split('''w2v_model.''' )[-1] == name.split('''.''' )[0]: lowercase : List[Any] = True if "*" in mapped_key: lowercase : Dict = name.split(_UpperCamelCase )[0].split('''.''' )[-2] lowercase : Any = mapped_key.replace('''*''', _UpperCamelCase ) if "weight_g" in name: lowercase : Dict = '''weight_g''' elif "weight_v" in name: lowercase : Union[str, Any] = '''weight_v''' elif "bias" in name: lowercase : List[Any] = '''bias''' elif "weight" in name: # TODO: don't match quantizer.weight_proj lowercase : Union[str, Any] = '''weight''' else: lowercase : List[Any] = None set_recursively(_UpperCamelCase, _UpperCamelCase, _UpperCamelCase, _UpperCamelCase, _UpperCamelCase, _UpperCamelCase ) continue if not is_used: unused_weights.append(_UpperCamelCase ) logger.warning(f"""Unused weights: {unused_weights}""" ) def __lowercase ( _UpperCamelCase, _UpperCamelCase, _UpperCamelCase, _UpperCamelCase, _UpperCamelCase ) ->Union[str, Any]: """simple docstring""" lowercase : Any = full_name.split('''conv_layers.''' )[-1] lowercase : Tuple = name.split('''.''' ) lowercase : Optional[int] = int(items[0] ) lowercase : Optional[int] = int(items[1] ) if type_id == 0: if "bias" in name: assert value.shape == feature_extractor.conv_layers[layer_id].conv.bias.data.shape, ( f"""{full_name} has size {value.shape}, but""" f""" {feature_extractor.conv_layers[layer_id].conv.bias.data.shape} was found.""" ) lowercase : Optional[int] = value logger.info(f"""Feat extract conv layer {layer_id} was initialized from {full_name}.""" ) elif "weight" in name: assert value.shape == feature_extractor.conv_layers[layer_id].conv.weight.data.shape, ( f"""{full_name} has size {value.shape}, but""" f""" {feature_extractor.conv_layers[layer_id].conv.weight.data.shape} was found.""" ) lowercase : List[Any] = 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: assert value.shape == feature_extractor.conv_layers[layer_id].layer_norm.bias.data.shape, ( f"""{full_name} has size {value.shape}, but {feature_extractor[layer_id].layer_norm.bias.data.shape} was""" " found." ) lowercase : int = value logger.info(f"""Feat extract layer norm weight of layer {layer_id} was initialized from {full_name}.""" ) elif "weight" in name: assert value.shape == feature_extractor.conv_layers[layer_id].layer_norm.weight.data.shape, ( f"""{full_name} has size {value.shape}, but""" f""" {feature_extractor[layer_id].layer_norm.weight.data.shape} was found.""" ) lowercase : str = value logger.info(f"""Feat extract layer norm weight of layer {layer_id} was initialized from {full_name}.""" ) else: unused_weights.append(_UpperCamelCase ) @torch.no_grad() def __lowercase ( _UpperCamelCase, _UpperCamelCase, _UpperCamelCase=None, _UpperCamelCase=None, _UpperCamelCase=True ) ->Tuple: """simple docstring""" if config_path is not None: lowercase : Any = UniSpeechConfig.from_pretrained(_UpperCamelCase ) else: lowercase : Dict = UniSpeechConfig() if is_finetuned: if dict_path: lowercase : Optional[Any] = Dictionary.load_from_json(_UpperCamelCase ) # important change bos & pad token id since CTC symbol is <pad> and # not <s> as in fairseq lowercase : Optional[int] = target_dict.pad_index lowercase : Tuple = target_dict.bos_index lowercase : List[Any] = target_dict.eos_index lowercase : Dict = len(target_dict.symbols ) lowercase : Optional[Any] = os.path.join(_UpperCamelCase, '''vocab.json''' ) if not os.path.isdir(_UpperCamelCase ): logger.error('''--pytorch_dump_folder_path ({}) should be a directory'''.format(_UpperCamelCase ) ) return os.makedirs(_UpperCamelCase, exist_ok=_UpperCamelCase ) lowercase : Union[str, Any] = target_dict.indices # fairseq has the <pad> and <s> switched lowercase : List[str] = 42 lowercase : str = 43 with open(_UpperCamelCase, '''w''', encoding='''utf-8''' ) as vocab_handle: json.dump(_UpperCamelCase, _UpperCamelCase ) lowercase : Any = WavaVecaPhonemeCTCTokenizer( _UpperCamelCase, unk_token=target_dict.unk_word, pad_token=target_dict.pad_word, bos_token=target_dict.bos_word, eos_token=target_dict.eos_word, word_delimiter_token='''|''', do_lower_case=_UpperCamelCase, ) lowercase : Optional[Any] = True if config.feat_extract_norm == '''layer''' else False lowercase : Union[str, Any] = WavaVecaFeatureExtractor( feature_size=1, sampling_rate=16000, padding_value=0, do_normalize=_UpperCamelCase, return_attention_mask=_UpperCamelCase, ) lowercase : Union[str, Any] = WavaVecaProcessor(feature_extractor=_UpperCamelCase, tokenizer=_UpperCamelCase ) processor.save_pretrained(_UpperCamelCase ) lowercase : List[str] = UniSpeechForCTC(_UpperCamelCase ) else: lowercase : Any = UniSpeechForPreTraining(_UpperCamelCase ) if is_finetuned: lowercase , lowercase , lowercase : str = fairseq.checkpoint_utils.load_model_ensemble_and_task( [checkpoint_path], arg_overrides={'''data''': '''/'''.join(dict_path.split('''/''' )[:-1] ), '''w2v_path''': checkpoint_path} ) else: lowercase , lowercase , lowercase : Optional[Any] = fairseq.checkpoint_utils.load_model_ensemble_and_task([checkpoint_path] ) lowercase : Optional[int] = model[0].eval() recursively_load_weights(_UpperCamelCase, _UpperCamelCase, _UpperCamelCase ) hf_unispeech.save_pretrained(_UpperCamelCase ) if __name__ == "__main__": __a = argparse.ArgumentParser() parser.add_argument('''--pytorch_dump_folder_path''', default=None, type=str, help='''Path to the output PyTorch model.''') parser.add_argument('''--checkpoint_path''', default=None, type=str, help='''Path to fairseq checkpoint''') parser.add_argument('''--dict_path''', default=None, type=str, help='''Path to dict of fine-tuned model''') parser.add_argument('''--config_path''', default=None, type=str, help='''Path to hf config.json of model to convert''') parser.add_argument( '''--not_finetuned''', action='''store_true''', help='''Whether the model to convert is a fine-tuned model or not''' ) __a = parser.parse_args() convert_unispeech_checkpoint( args.checkpoint_path, args.pytorch_dump_folder_path, args.config_path, args.dict_path, not args.not_finetuned )
173
import argparse import os import re import tensorflow as tf import torch from transformers import BertConfig, BertModel from transformers.utils import logging logging.set_verbosity_info() __a = logging.get_logger(__name__) def __lowercase ( _UpperCamelCase, _UpperCamelCase, _UpperCamelCase ) ->Union[str, Any]: """simple docstring""" lowercase : Any = os.path.abspath(_UpperCamelCase ) logger.info(f"""Converting TensorFlow checkpoint from {tf_path}""" ) # Load weights from TF model lowercase : Optional[int] = tf.train.list_variables(_UpperCamelCase ) lowercase : Optional[int] = [] lowercase : Optional[int] = [] lowercase : Optional[Any] = [] for full_name, shape in init_vars: # logger.info(f"Loading TF weight {name} with shape {shape}") lowercase : int = full_name.split('''/''' ) if full_name == "_CHECKPOINTABLE_OBJECT_GRAPH" or name[0] in ["global_step", "save_counter"]: logger.info(f"""Skipping non-model layer {full_name}""" ) continue if "optimizer" in full_name: logger.info(f"""Skipping optimization layer {full_name}""" ) continue if name[0] == "model": # ignore initial 'model' lowercase : List[Any] = name[1:] # figure out how many levels deep the name is lowercase : Optional[int] = 0 for _name in name: if _name.startswith('''layer_with_weights''' ): depth += 1 else: break layer_depth.append(_UpperCamelCase ) # read data lowercase : Any = tf.train.load_variable(_UpperCamelCase, _UpperCamelCase ) names.append('''/'''.join(_UpperCamelCase ) ) arrays.append(_UpperCamelCase ) logger.info(f"""Read a total of {len(_UpperCamelCase ):,} layers""" ) # Sanity check if len(set(_UpperCamelCase ) ) != 1: raise ValueError(f"""Found layer names with different depths (layer depth {list(set(_UpperCamelCase ) )})""" ) lowercase : List[str] = list(set(_UpperCamelCase ) )[0] if layer_depth != 1: raise ValueError( '''The model contains more than just the embedding/encoder layers. This script does not handle MLM/NSP''' ''' heads.''' ) # convert layers logger.info('''Converting weights...''' ) for full_name, array in zip(_UpperCamelCase, _UpperCamelCase ): lowercase : Optional[int] = full_name.split('''/''' ) lowercase : Tuple = model lowercase : Any = [] for i, m_name in enumerate(_UpperCamelCase ): if m_name == ".ATTRIBUTES": # variable names end with .ATTRIBUTES/VARIABLE_VALUE break if m_name.startswith('''layer_with_weights''' ): lowercase : Tuple = int(m_name.split('''-''' )[-1] ) if layer_num <= 2: # embedding layers # layer_num 0: word_embeddings # layer_num 1: position_embeddings # layer_num 2: token_type_embeddings continue elif layer_num == 3: # embedding LayerNorm trace.extend(['''embeddings''', '''LayerNorm'''] ) lowercase : int = getattr(_UpperCamelCase, '''embeddings''' ) lowercase : Optional[Any] = getattr(_UpperCamelCase, '''LayerNorm''' ) elif layer_num > 3 and layer_num < config.num_hidden_layers + 4: # encoder layers trace.extend(['''encoder''', '''layer''', str(layer_num - 4 )] ) lowercase : int = getattr(_UpperCamelCase, '''encoder''' ) lowercase : Tuple = getattr(_UpperCamelCase, '''layer''' ) lowercase : List[Any] = pointer[layer_num - 4] elif layer_num == config.num_hidden_layers + 4: # pooler layer trace.extend(['''pooler''', '''dense'''] ) lowercase : str = getattr(_UpperCamelCase, '''pooler''' ) lowercase : str = getattr(_UpperCamelCase, '''dense''' ) elif m_name == "embeddings": trace.append('''embeddings''' ) lowercase : Optional[int] = getattr(_UpperCamelCase, '''embeddings''' ) if layer_num == 0: trace.append('''word_embeddings''' ) lowercase : str = getattr(_UpperCamelCase, '''word_embeddings''' ) elif layer_num == 1: trace.append('''position_embeddings''' ) lowercase : List[Any] = getattr(_UpperCamelCase, '''position_embeddings''' ) elif layer_num == 2: trace.append('''token_type_embeddings''' ) lowercase : int = getattr(_UpperCamelCase, '''token_type_embeddings''' ) else: raise ValueError(f"""Unknown embedding layer with name {full_name}""" ) trace.append('''weight''' ) lowercase : Union[str, Any] = getattr(_UpperCamelCase, '''weight''' ) elif m_name == "_attention_layer": # self-attention layer trace.extend(['''attention''', '''self'''] ) lowercase : Tuple = getattr(_UpperCamelCase, '''attention''' ) lowercase : str = getattr(_UpperCamelCase, '''self''' ) elif m_name == "_attention_layer_norm": # output attention norm trace.extend(['''attention''', '''output''', '''LayerNorm'''] ) lowercase : Dict = getattr(_UpperCamelCase, '''attention''' ) lowercase : Any = getattr(_UpperCamelCase, '''output''' ) lowercase : Union[str, Any] = getattr(_UpperCamelCase, '''LayerNorm''' ) elif m_name == "_attention_output_dense": # output attention dense trace.extend(['''attention''', '''output''', '''dense'''] ) lowercase : Union[str, Any] = getattr(_UpperCamelCase, '''attention''' ) lowercase : str = getattr(_UpperCamelCase, '''output''' ) lowercase : Optional[int] = getattr(_UpperCamelCase, '''dense''' ) elif m_name == "_output_dense": # output dense trace.extend(['''output''', '''dense'''] ) lowercase : List[str] = getattr(_UpperCamelCase, '''output''' ) lowercase : str = getattr(_UpperCamelCase, '''dense''' ) elif m_name == "_output_layer_norm": # output dense trace.extend(['''output''', '''LayerNorm'''] ) lowercase : Dict = getattr(_UpperCamelCase, '''output''' ) lowercase : int = getattr(_UpperCamelCase, '''LayerNorm''' ) elif m_name == "_key_dense": # attention key trace.append('''key''' ) lowercase : Optional[Any] = getattr(_UpperCamelCase, '''key''' ) elif m_name == "_query_dense": # attention query trace.append('''query''' ) lowercase : Dict = getattr(_UpperCamelCase, '''query''' ) elif m_name == "_value_dense": # attention value trace.append('''value''' ) lowercase : Optional[Any] = getattr(_UpperCamelCase, '''value''' ) elif m_name == "_intermediate_dense": # attention intermediate dense trace.extend(['''intermediate''', '''dense'''] ) lowercase : List[str] = getattr(_UpperCamelCase, '''intermediate''' ) lowercase : Optional[int] = getattr(_UpperCamelCase, '''dense''' ) elif m_name == "_output_layer_norm": # output layer norm trace.append('''output''' ) lowercase : Tuple = getattr(_UpperCamelCase, '''output''' ) # weights & biases elif m_name in ["bias", "beta"]: trace.append('''bias''' ) lowercase : str = getattr(_UpperCamelCase, '''bias''' ) elif m_name in ["kernel", "gamma"]: trace.append('''weight''' ) lowercase : Dict = getattr(_UpperCamelCase, '''weight''' ) else: logger.warning(f"""Ignored {m_name}""" ) # for certain layers reshape is necessary lowercase : Any = '''.'''.join(_UpperCamelCase ) if re.match(R'''(\S+)\.attention\.self\.(key|value|query)\.(bias|weight)''', _UpperCamelCase ) or re.match( R'''(\S+)\.attention\.output\.dense\.weight''', _UpperCamelCase ): lowercase : Any = array.reshape(pointer.data.shape ) if "kernel" in full_name: lowercase : List[str] = array.transpose() if pointer.shape == array.shape: lowercase : Optional[Any] = torch.from_numpy(_UpperCamelCase ) else: raise ValueError( f"""Shape mismatch in layer {full_name}: Model expects shape {pointer.shape} but layer contains shape:""" f""" {array.shape}""" ) logger.info(f"""Successfully set variable {full_name} to PyTorch layer {trace}""" ) return model def __lowercase ( _UpperCamelCase, _UpperCamelCase, _UpperCamelCase ) ->Tuple: """simple docstring""" logger.info(f"""Loading model based on config from {config_path}...""" ) lowercase : List[Any] = BertConfig.from_json_file(_UpperCamelCase ) lowercase : Dict = BertModel(_UpperCamelCase ) # Load weights from checkpoint logger.info(f"""Loading weights from checkpoint {tf_checkpoint_path}...""" ) load_tfa_weights_in_bert(_UpperCamelCase, _UpperCamelCase, _UpperCamelCase ) # Save pytorch-model logger.info(f"""Saving PyTorch model to {pytorch_dump_path}...""" ) torch.save(model.state_dict(), _UpperCamelCase ) if __name__ == "__main__": __a = argparse.ArgumentParser() parser.add_argument( '''--tf_checkpoint_path''', type=str, required=True, help='''Path to the TensorFlow 2.x checkpoint path.''' ) parser.add_argument( '''--bert_config_file''', type=str, required=True, help='''The config json file corresponding to the BERT model. This specifies the model architecture.''', ) parser.add_argument( '''--pytorch_dump_path''', type=str, required=True, help='''Path to the output PyTorch model (must include filename).''', ) __a = parser.parse_args() convert_tfa_checkpoint_to_pytorch(args.tf_checkpoint_path, args.bert_config_file, args.pytorch_dump_path)
173
1
'''simple docstring''' import unicodedata from dataclasses import dataclass from typing import Optional, Union import numpy as np from transformers.data.data_collator import DataCollatorMixin from transformers.file_utils import PaddingStrategy from transformers.tokenization_utils_base import PreTrainedTokenizerBase def __magic_name__ ( __UpperCAmelCase, __UpperCAmelCase, __UpperCAmelCase, __UpperCAmelCase ) -> List[Any]: '''simple docstring''' if isinstance(__UpperCAmelCase, __UpperCAmelCase ): snake_case_ = np.full((len(__UpperCAmelCase ), sequence_length, 2), __UpperCAmelCase ) else: snake_case_ = np.full((len(__UpperCAmelCase ), sequence_length), __UpperCAmelCase ) for i, tensor in enumerate(__UpperCAmelCase ): if padding_side == "right": if isinstance(__UpperCAmelCase, __UpperCAmelCase ): snake_case_ = tensor[:sequence_length] else: snake_case_ = tensor[:sequence_length] else: if isinstance(__UpperCAmelCase, __UpperCAmelCase ): snake_case_ = tensor[:sequence_length] else: snake_case_ = tensor[:sequence_length] return out_tensor.tolist() def __magic_name__ ( __UpperCAmelCase ) -> Dict: '''simple docstring''' snake_case_ = ord(__UpperCAmelCase ) if (cp >= 33 and cp <= 47) or (cp >= 58 and cp <= 64) or (cp >= 91 and cp <= 96) or (cp >= 123 and cp <= 126): return True snake_case_ = unicodedata.category(__UpperCAmelCase ) if cat.startswith('''P''' ): return True return False @dataclass class a ( _lowerCamelCase ): snake_case_ = 42 snake_case_ = True snake_case_ = None snake_case_ = None snake_case_ = -100 snake_case_ = "pt" def A_ ( self : int , lowercase_ : List[str] ): import torch snake_case_ = '''label''' if '''label''' in features[0].keys() else '''labels''' snake_case_ = [feature[label_name] for feature in features] if label_name in features[0].keys() else None snake_case_ = self.tokenizer.pad( lowercase_ , padding=self.padding , max_length=self.max_length , pad_to_multiple_of=self.pad_to_multiple_of , return_tensors='''pt''' if labels is None else None , ) if labels is None: return batch snake_case_ = torch.tensor(batch['''entity_ids'''] ).shape[1] snake_case_ = self.tokenizer.padding_side if padding_side == "right": snake_case_ = [ list(lowercase_ ) + [self.label_pad_token_id] * (sequence_length - len(lowercase_ )) for label in labels ] else: snake_case_ = [ [self.label_pad_token_id] * (sequence_length - len(lowercase_ )) + list(lowercase_ ) for label in labels ] snake_case_ = [feature['''ner_tags'''] for feature in features] snake_case_ = padding_tensor(lowercase_ , -1 , lowercase_ , lowercase_ ) snake_case_ = [feature['''original_entity_spans'''] for feature in features] snake_case_ = padding_tensor(lowercase_ , (-1, -1) , lowercase_ , lowercase_ ) snake_case_ = {k: torch.tensor(lowercase_ , dtype=torch.intaa ) for k, v in batch.items()} return batch
56
'''simple docstring''' import unittest from diffusers import FlaxAutoencoderKL from diffusers.utils import is_flax_available from diffusers.utils.testing_utils import require_flax from .test_modeling_common_flax import FlaxModelTesterMixin if is_flax_available(): import jax @require_flax class lowerCAmelCase__ ( a , unittest.TestCase ): """simple docstring""" lowerCAmelCase__ = FlaxAutoencoderKL @property def UpperCAmelCase__ ( self : Tuple ) -> List[str]: """simple docstring""" __SCREAMING_SNAKE_CASE = 4 __SCREAMING_SNAKE_CASE = 3 __SCREAMING_SNAKE_CASE = (32, 32) __SCREAMING_SNAKE_CASE = jax.random.PRNGKey(0 ) __SCREAMING_SNAKE_CASE = jax.random.uniform(__SCREAMING_SNAKE_CASE , ((batch_size, num_channels) + sizes) ) return {"sample": image, "prng_key": prng_key} def UpperCAmelCase__ ( self : Union[str, Any] ) -> Any: """simple docstring""" __SCREAMING_SNAKE_CASE = { """block_out_channels""": [32, 64], """in_channels""": 3, """out_channels""": 3, """down_block_types""": ["""DownEncoderBlock2D""", """DownEncoderBlock2D"""], """up_block_types""": ["""UpDecoderBlock2D""", """UpDecoderBlock2D"""], """latent_channels""": 4, } __SCREAMING_SNAKE_CASE = self.dummy_input return init_dict, inputs_dict
267
0
'''simple docstring''' from importlib import import_module from .logging import get_logger __a = get_logger(__name__) class A__ : """simple docstring""" def __init__( self : List[str] , lowerCAmelCase__ : List[Any] , lowerCAmelCase__ : Optional[Any]=None ) -> Optional[int]: """simple docstring""" _UpperCAmelCase : Any = attrs or [] if module is not None: for key in module.__dict__: if key in attrs or not key.startswith("__" ): setattr(self , lowerCAmelCase__ , getattr(lowerCAmelCase__ , lowerCAmelCase__ ) ) _UpperCAmelCase : int = module._original_module if isinstance(lowerCAmelCase__ , _PatchedModuleObj ) else module class A__ : """simple docstring""" UpperCamelCase_ : Union[str, Any] = [] def __init__( self : int , lowerCAmelCase__ : List[str] , lowerCAmelCase__ : str , lowerCAmelCase__ : List[str] , lowerCAmelCase__ : Optional[int]=None ) -> Optional[Any]: """simple docstring""" _UpperCAmelCase : List[Any] = obj _UpperCAmelCase : int = target _UpperCAmelCase : Optional[int] = new _UpperCAmelCase : Any = target.split("." )[0] _UpperCAmelCase : Optional[int] = {} _UpperCAmelCase : Dict = attrs or [] def __enter__( self : List[str] ) -> int: """simple docstring""" *_UpperCAmelCase , _UpperCAmelCase : List[str] = self.target.split("." ) # Patch modules: # it's used to patch attributes of submodules like "os.path.join"; # in this case we need to patch "os" and "os.path" for i in range(len(lowerCAmelCase__ ) ): try: _UpperCAmelCase : int = import_module(".".join(submodules[: i + 1] ) ) except ModuleNotFoundError: continue # We iterate over all the globals in self.obj in case we find "os" or "os.path" for attr in self.obj.__dir__(): _UpperCAmelCase : List[Any] = getattr(self.obj , lowerCAmelCase__ ) # We don't check for the name of the global, but rather if its value *is* "os" or "os.path". # This allows to patch renamed modules like "from os import path as ospath". if obj_attr is submodule or ( (isinstance(lowerCAmelCase__ , _PatchedModuleObj ) and obj_attr._original_module is submodule) ): _UpperCAmelCase : Tuple = obj_attr # patch at top level setattr(self.obj , lowerCAmelCase__ , _PatchedModuleObj(lowerCAmelCase__ , attrs=self.attrs ) ) _UpperCAmelCase : List[Any] = getattr(self.obj , lowerCAmelCase__ ) # construct lower levels patches for key in submodules[i + 1 :]: setattr(lowerCAmelCase__ , lowerCAmelCase__ , _PatchedModuleObj(getattr(lowerCAmelCase__ , lowerCAmelCase__ , lowerCAmelCase__ ) , attrs=self.attrs ) ) _UpperCAmelCase : Any = getattr(lowerCAmelCase__ , lowerCAmelCase__ ) # finally set the target attribute setattr(lowerCAmelCase__ , lowerCAmelCase__ , self.new ) # Patch attribute itself: # it's used for builtins like "open", # and also to patch "os.path.join" we may also need to patch "join" # itself if it was imported as "from os.path import join". if submodules: # if it's an attribute of a submodule like "os.path.join" try: _UpperCAmelCase : Dict = getattr(import_module(".".join(lowerCAmelCase__ ) ) , lowerCAmelCase__ ) except (AttributeError, ModuleNotFoundError): return # We iterate over all the globals in self.obj in case we find "os.path.join" for attr in self.obj.__dir__(): # We don't check for the name of the global, but rather if its value *is* "os.path.join". # This allows to patch renamed attributes like "from os.path import join as pjoin". if getattr(self.obj , lowerCAmelCase__ ) is attr_value: _UpperCAmelCase : Optional[Any] = getattr(self.obj , lowerCAmelCase__ ) setattr(self.obj , lowerCAmelCase__ , self.new ) elif target_attr in globals()["__builtins__"]: # if it'a s builtin like "open" _UpperCAmelCase : Dict = globals()["__builtins__"][target_attr] setattr(self.obj , lowerCAmelCase__ , self.new ) else: raise RuntimeError(F"""Tried to patch attribute {target_attr} instead of a submodule.""" ) def __exit__( self : Optional[int] , *lowerCAmelCase__ : List[str] ) -> Union[str, Any]: """simple docstring""" for attr in list(self.original ): setattr(self.obj , lowerCAmelCase__ , self.original.pop(lowerCAmelCase__ ) ) def _lowerCAmelCase ( self : Union[str, Any] ) -> Optional[Any]: """simple docstring""" self.__enter__() self._active_patches.append(self ) def _lowerCAmelCase ( self : Optional[int] ) -> Tuple: """simple docstring""" try: self._active_patches.remove(self ) except ValueError: # If the patch hasn't been started this will fail return None return self.__exit__()
17
'''simple docstring''' import baseaa def __UpperCAmelCase ( a_: str ): return baseaa.baaencode(string.encode("utf-8" ) ) def __UpperCAmelCase ( a_: bytes ): return baseaa.baadecode(a_ ).decode("utf-8" ) if __name__ == "__main__": __a = 'Hello World!' __a = baseaa_encode(test) print(encoded) __a = baseaa_decode(encoded) print(decoded)
17
1
"""simple docstring""" import unittest from transformers import ( MODEL_FOR_SEQ_TO_SEQ_CAUSAL_LM_MAPPING, TF_MODEL_FOR_SEQ_TO_SEQ_CAUSAL_LM_MAPPING, TextaTextGenerationPipeline, pipeline, ) from transformers.testing_utils import is_pipeline_test, require_tf, require_torch from transformers.utils import is_torch_available from .test_pipelines_common import ANY if is_torch_available(): import torch @is_pipeline_test class UpperCAmelCase_ ( unittest.TestCase): snake_case__ = MODEL_FOR_SEQ_TO_SEQ_CAUSAL_LM_MAPPING snake_case__ = TF_MODEL_FOR_SEQ_TO_SEQ_CAUSAL_LM_MAPPING def _UpperCamelCase ( self : int , __UpperCamelCase : Tuple , __UpperCamelCase : Union[str, Any] , __UpperCamelCase : int ) -> Tuple: _UpperCamelCase = TextaTextGenerationPipeline(model=UpperCamelCase__ , tokenizer=UpperCamelCase__ ) return generator, ["Something to write", "Something else"] def _UpperCamelCase ( self : Tuple , __UpperCamelCase : Dict , __UpperCamelCase : Optional[int] ) -> Any: _UpperCamelCase = generator('''Something there''' ) self.assertEqual(UpperCamelCase__ , [{'''generated_text''': ANY(UpperCamelCase__ )}] ) # These are encoder decoder, they don't just append to incoming string self.assertFalse(outputs[0]['''generated_text'''].startswith('''Something there''' ) ) _UpperCamelCase = generator(['''This is great !''', '''Something else'''] , num_return_sequences=2 , do_sample=UpperCamelCase__ ) self.assertEqual( UpperCamelCase__ , [ [{'''generated_text''': ANY(UpperCamelCase__ )}, {'''generated_text''': ANY(UpperCamelCase__ )}], [{'''generated_text''': ANY(UpperCamelCase__ )}, {'''generated_text''': ANY(UpperCamelCase__ )}], ] , ) _UpperCamelCase = generator( ['''This is great !''', '''Something else'''] , num_return_sequences=2 , batch_size=2 , do_sample=UpperCamelCase__ ) self.assertEqual( UpperCamelCase__ , [ [{'''generated_text''': ANY(UpperCamelCase__ )}, {'''generated_text''': ANY(UpperCamelCase__ )}], [{'''generated_text''': ANY(UpperCamelCase__ )}, {'''generated_text''': ANY(UpperCamelCase__ )}], ] , ) with self.assertRaises(UpperCamelCase__ ): generator(4 ) @require_torch def _UpperCamelCase ( self : Optional[Any] ) -> Dict: _UpperCamelCase = pipeline('''text2text-generation''' , model='''patrickvonplaten/t5-tiny-random''' , framework='''pt''' ) # do_sample=False necessary for reproducibility _UpperCamelCase = generator('''Something there''' , do_sample=UpperCamelCase__ ) self.assertEqual(UpperCamelCase__ , [{'''generated_text''': ''''''}] ) _UpperCamelCase = 3 _UpperCamelCase = generator( '''Something there''' , num_return_sequences=UpperCamelCase__ , num_beams=UpperCamelCase__ , ) _UpperCamelCase = [ {'''generated_text''': '''Beide Beide Beide Beide Beide Beide Beide Beide Beide'''}, {'''generated_text''': '''Beide Beide Beide Beide Beide Beide Beide Beide'''}, {'''generated_text''': ''''''}, ] self.assertEqual(UpperCamelCase__ , UpperCamelCase__ ) _UpperCamelCase = generator('''This is a test''' , do_sample=UpperCamelCase__ , num_return_sequences=2 , return_tensors=UpperCamelCase__ ) self.assertEqual( UpperCamelCase__ , [ {'''generated_token_ids''': ANY(torch.Tensor )}, {'''generated_token_ids''': ANY(torch.Tensor )}, ] , ) _UpperCamelCase = generator.model.config.eos_token_id _UpperCamelCase = '''<pad>''' _UpperCamelCase = generator( ['''This is a test''', '''This is a second test'''] , do_sample=UpperCamelCase__ , num_return_sequences=2 , batch_size=2 , return_tensors=UpperCamelCase__ , ) self.assertEqual( UpperCamelCase__ , [ [ {'''generated_token_ids''': ANY(torch.Tensor )}, {'''generated_token_ids''': ANY(torch.Tensor )}, ], [ {'''generated_token_ids''': ANY(torch.Tensor )}, {'''generated_token_ids''': ANY(torch.Tensor )}, ], ] , ) @require_tf def _UpperCamelCase ( self : Tuple ) -> Any: _UpperCamelCase = pipeline('''text2text-generation''' , model='''patrickvonplaten/t5-tiny-random''' , framework='''tf''' ) # do_sample=False necessary for reproducibility _UpperCamelCase = generator('''Something there''' , do_sample=UpperCamelCase__ ) self.assertEqual(UpperCamelCase__ , [{'''generated_text''': ''''''}] )
256
'''simple docstring''' from __future__ import annotations class A__ : def __init__( self , UpperCamelCase__=None ) -> Any: '''simple docstring''' A_ = data A_ = None def __repr__( self ) -> List[str]: '''simple docstring''' A_ = [] A_ = self while temp: string_rep.append(f'''{temp.data}''' ) A_ = temp.next return "->".join(UpperCamelCase__ ) def UpperCAmelCase__ ( UpperCAmelCase__ ) -> Tuple: if not elements_list: raise Exception("""The Elements List is empty""" ) A_ = A_ = Node(elements_list[0] ) for i in range(1, len(UpperCAmelCase__ ) ): A_ = Node(elements_list[i] ) A_ = current.next return head def UpperCAmelCase__ ( UpperCAmelCase__ ) -> None: if head_node is not None and isinstance(UpperCAmelCase__, UpperCAmelCase__ ): print_reverse(head_node.next ) print(head_node.data ) def UpperCAmelCase__ ( ) -> Optional[Any]: from doctest import testmod testmod() A_ = make_linked_list([14, 52, 14, 12, 43] ) print("""Linked List:""" ) print(UpperCAmelCase__ ) print("""Elements in Reverse:""" ) print_reverse(UpperCAmelCase__ ) if __name__ == "__main__": main()
162
0
'''simple docstring''' from argparse import ArgumentParser from .env import EnvironmentCommand def _A ( ): """simple docstring""" __lowercase = ArgumentParser('''Diffusers CLI tool''' , usage='''diffusers-cli <command> [<args>]''' ) __lowercase = parser.add_subparsers(help='''diffusers-cli command helpers''' ) # Register commands EnvironmentCommand.register_subcommand(A__ ) # Let's go __lowercase = parser.parse_args() if not hasattr(A__ , '''func''' ): parser.print_help() exit(1 ) # Run __lowercase = args.func(A__ ) service.run() if __name__ == "__main__": main()
52
'''simple docstring''' def _A ( A__ = 1000 ): """simple docstring""" __lowercase , __lowercase = 1, 1 __lowercase = 2 while True: __lowercase = 0 __lowercase = fa + fa __lowercase , __lowercase = fa, f index += 1 for _ in str(A__ ): i += 1 if i == n: break return index if __name__ == "__main__": print(solution(int(str(input()).strip())))
52
1
from __future__ import annotations import random # Maximum size of the population. Bigger could be faster but is more memory expensive. __snake_case = 2_0_0 # Number of elements selected in every generation of evolution. The selection takes # place from best to worst of that generation and must be smaller than N_POPULATION. __snake_case = 5_0 # Probability that an element of a generation can mutate, changing one of its genes. # This will guarantee that all genes will be used during evolution. __snake_case = 0.4 # Just a seed to improve randomness required by the algorithm. random.seed(random.randint(0, 1_0_0_0)) def _A ( _lowercase , _lowercase ) -> tuple[str, float]: """simple docstring""" __UpperCamelCase = len([g for position, g in enumerate(_lowercase ) if g == main_target[position]] ) return (item, float(_lowercase )) def _A ( _lowercase , _lowercase ) -> tuple[str, str]: """simple docstring""" __UpperCamelCase = random.randint(0 , len(_lowercase ) - 1 ) __UpperCamelCase = parent_a[:random_slice] + parent_a[random_slice:] __UpperCamelCase = parent_a[:random_slice] + parent_a[random_slice:] return (child_a, child_a) def _A ( _lowercase , _lowercase ) -> str: """simple docstring""" __UpperCamelCase = list(_lowercase ) if random.uniform(0 , 1 ) < MUTATION_PROBABILITY: __UpperCamelCase = random.choice(_lowercase ) return "".join(_lowercase ) def _A ( _lowercase , _lowercase , _lowercase , ) -> list[str]: """simple docstring""" __UpperCamelCase = [] # Generate more children proportionally to the fitness score. __UpperCamelCase = int(parent_a[1] * 1_00 ) + 1 __UpperCamelCase = 10 if child_n >= 10 else child_n for _ in range(_lowercase ): __UpperCamelCase = population_score[random.randint(0 , _lowercase )][0] __UpperCamelCase, __UpperCamelCase = crossover(parent_a[0] , _lowercase ) # Append new string to the population list. pop.append(mutate(_lowercase , _lowercase ) ) pop.append(mutate(_lowercase , _lowercase ) ) return pop def _A ( _lowercase , _lowercase , _lowercase = True ) -> tuple[int, int, str]: """simple docstring""" if N_POPULATION < N_SELECTED: __UpperCamelCase = f'''{N_POPULATION} must be bigger than {N_SELECTED}''' raise ValueError(_lowercase ) # Verify that the target contains no genes besides the ones inside genes variable. __UpperCamelCase = sorted({c for c in target if c not in genes} ) if not_in_genes_list: __UpperCamelCase = f'''{not_in_genes_list} is not in genes list, evolution cannot converge''' raise ValueError(_lowercase ) # Generate random starting population. __UpperCamelCase = [] for _ in range(_lowercase ): population.append(''.join([random.choice(_lowercase ) for i in range(len(_lowercase ) )] ) ) # Just some logs to know what the algorithms is doing. __UpperCamelCase, __UpperCamelCase = 0, 0 # This loop will end when we find a perfect match for our target. while True: generation += 1 total_population += len(_lowercase ) # Random population created. Now it's time to evaluate. # Adding a bit of concurrency can make everything faster, # # import concurrent.futures # population_score: list[tuple[str, float]] = [] # with concurrent.futures.ThreadPoolExecutor( # max_workers=NUM_WORKERS) as executor: # futures = {executor.submit(evaluate, item) for item in population} # concurrent.futures.wait(futures) # population_score = [item.result() for item in futures] # # but with a simple algorithm like this, it will probably be slower. # We just need to call evaluate for every item inside the population. __UpperCamelCase = [evaluate(_lowercase , _lowercase ) for item in population] # Check if there is a matching evolution. __UpperCamelCase = sorted(_lowercase , key=lambda _lowercase : x[1] , reverse=_lowercase ) if population_score[0][0] == target: return (generation, total_population, population_score[0][0]) # Print the best result every 10 generation. # Just to know that the algorithm is working. if debug and generation % 10 == 0: print( f'''\nGeneration: {generation}''' f'''\nTotal Population:{total_population}''' f'''\nBest score: {population_score[0][1]}''' f'''\nBest string: {population_score[0][0]}''' ) # Flush the old population, keeping some of the best evolutions. # Keeping this avoid regression of evolution. __UpperCamelCase = population[: int(N_POPULATION / 3 )] population.clear() population.extend(_lowercase ) # Normalize population score to be between 0 and 1. __UpperCamelCase = [ (item, score / len(_lowercase )) for item, score in population_score ] # This is selection for i in range(_lowercase ): population.extend(select(population_score[int(_lowercase )] , _lowercase , _lowercase ) ) # Check if the population has already reached the maximum value and if so, # break the cycle. If this check is disabled, the algorithm will take # forever to compute large strings, but will also calculate small strings in # a far fewer generations. if len(_lowercase ) > N_POPULATION: break if __name__ == "__main__": __snake_case = ( '''This is a genetic algorithm to evaluate, combine, evolve, and mutate a string!''' ) __snake_case = list( ''' ABCDEFGHIJKLMNOPQRSTUVWXYZabcdefghijklm''' '''nopqrstuvwxyz.,;!?+-*#@^\'èéòà€ù=)(&%$£/\\''' ) __snake_case , __snake_case , __snake_case = basic(target_str, genes_list) print( f"""\nGeneration: {generation}\nTotal Population: {population}\nTarget: {target}""" )
310
import json import os import unittest from transformers.models.biogpt.tokenization_biogpt import VOCAB_FILES_NAMES, BioGptTokenizer from transformers.testing_utils import slow from ...test_tokenization_common import TokenizerTesterMixin class __lowerCamelCase (_a , unittest.TestCase ): _lowercase = BioGptTokenizer _lowercase = False def snake_case_ ( self: Any ): '''simple docstring''' super().setUp() # Adapted from Sennrich et al. 2015 and https://github.com/rsennrich/subword-nmt __UpperCamelCase = [ 'l', 'o', 'w', 'e', 'r', 's', 't', 'i', 'd', 'n', 'w</w>', 'r</w>', 't</w>', 'lo', 'low', 'er</w>', 'low</w>', 'lowest</w>', 'newer</w>', 'wider</w>', '<unk>', ] __UpperCamelCase = dict(zip(A_,range(len(A_ ) ) ) ) __UpperCamelCase = ['l o 123', 'lo w 1456', 'e r</w> 1789', ''] __UpperCamelCase = os.path.join(self.tmpdirname,VOCAB_FILES_NAMES['vocab_file'] ) __UpperCamelCase = os.path.join(self.tmpdirname,VOCAB_FILES_NAMES['merges_file'] ) with open(self.vocab_file,'w' ) as fp: fp.write(json.dumps(A_ ) ) with open(self.merges_file,'w' ) as fp: fp.write('\n'.join(A_ ) ) def snake_case_ ( self: Optional[int],A_: List[Any] ): '''simple docstring''' __UpperCamelCase = 'lower newer' __UpperCamelCase = 'lower newer' return input_text, output_text def snake_case_ ( self: Any ): '''simple docstring''' __UpperCamelCase = BioGptTokenizer(self.vocab_file,self.merges_file ) __UpperCamelCase = 'lower' __UpperCamelCase = ['low', 'er</w>'] __UpperCamelCase = tokenizer.tokenize(A_ ) self.assertListEqual(A_,A_ ) __UpperCamelCase = tokens + ['<unk>'] __UpperCamelCase = [14, 15, 20] self.assertListEqual(tokenizer.convert_tokens_to_ids(A_ ),A_ ) @slow def snake_case_ ( self: Dict ): '''simple docstring''' __UpperCamelCase = BioGptTokenizer.from_pretrained('microsoft/biogpt' ) __UpperCamelCase = tokenizer.encode('sequence builders',add_special_tokens=A_ ) __UpperCamelCase = tokenizer.encode('multi-sequence build',add_special_tokens=A_ ) __UpperCamelCase = tokenizer.build_inputs_with_special_tokens(A_ ) __UpperCamelCase = tokenizer.build_inputs_with_special_tokens(A_,A_ ) self.assertTrue(encoded_sentence == [2] + text ) self.assertTrue(encoded_pair == [2] + text + [2] + text_a )
310
1
'''simple docstring''' import random import unittest import torch from diffusers import IFInpaintingSuperResolutionPipeline from diffusers.utils import floats_tensor from diffusers.utils.import_utils import is_xformers_available from diffusers.utils.testing_utils import skip_mps, torch_device from ..pipeline_params import ( TEXT_GUIDED_IMAGE_INPAINTING_BATCH_PARAMS, TEXT_GUIDED_IMAGE_INPAINTING_PARAMS, ) from ..test_pipelines_common import PipelineTesterMixin from . import IFPipelineTesterMixin @skip_mps class lowercase_ (lowerCamelCase__ , lowerCamelCase__ , unittest.TestCase ): """simple docstring""" SCREAMING_SNAKE_CASE : str = IFInpaintingSuperResolutionPipeline SCREAMING_SNAKE_CASE : List[str] = TEXT_GUIDED_IMAGE_INPAINTING_PARAMS - {'width', 'height'} SCREAMING_SNAKE_CASE : int = TEXT_GUIDED_IMAGE_INPAINTING_BATCH_PARAMS.union({'original_image'} ) SCREAMING_SNAKE_CASE : Dict = PipelineTesterMixin.required_optional_params - {'latents'} def SCREAMING_SNAKE_CASE ( self : Union[str, Any] ): return self._get_superresolution_dummy_components() def SCREAMING_SNAKE_CASE ( self : List[Any] ,lowercase__ : List[str] ,lowercase__ : Dict=0 ): if str(lowercase__ ).startswith('''mps''' ): __lowercase = torch.manual_seed(lowercase__ ) else: __lowercase = torch.Generator(device=lowercase__ ).manual_seed(lowercase__ ) __lowercase = floats_tensor((1, 3, 1_6, 1_6) ,rng=random.Random(lowercase__ ) ).to(lowercase__ ) __lowercase = floats_tensor((1, 3, 3_2, 3_2) ,rng=random.Random(lowercase__ ) ).to(lowercase__ ) __lowercase = floats_tensor((1, 3, 3_2, 3_2) ,rng=random.Random(lowercase__ ) ).to(lowercase__ ) __lowercase = { '''prompt''': '''A painting of a squirrel eating a burger''', '''image''': image, '''original_image''': original_image, '''mask_image''': mask_image, '''generator''': generator, '''num_inference_steps''': 2, '''output_type''': '''numpy''', } return inputs @unittest.skipIf( torch_device != '''cuda''' or not is_xformers_available() ,reason='''XFormers attention is only available with CUDA and `xformers` installed''' ,) def SCREAMING_SNAKE_CASE ( self : List[str] ): self._test_xformers_attention_forwardGenerator_pass(expected_max_diff=1e-3 ) def SCREAMING_SNAKE_CASE ( self : Tuple ): self._test_save_load_optional_components() @unittest.skipIf(torch_device != '''cuda''' ,reason='''float16 requires CUDA''' ) def SCREAMING_SNAKE_CASE ( self : Optional[int] ): # Due to non-determinism in save load of the hf-internal-testing/tiny-random-t5 text encoder super().test_save_load_floataa(expected_max_diff=1e-1 ) def SCREAMING_SNAKE_CASE ( self : int ): self._test_attention_slicing_forward_pass(expected_max_diff=1e-2 ) def SCREAMING_SNAKE_CASE ( self : str ): self._test_save_load_local() def SCREAMING_SNAKE_CASE ( self : List[Any] ): self._test_inference_batch_single_identical( expected_max_diff=1e-2 ,)
52
'''simple docstring''' import warnings from ...utils import logging from .image_processing_glpn import GLPNImageProcessor lowerCAmelCase__ = logging.get_logger(__name__) class lowercase_ (lowerCamelCase__ ): """simple docstring""" def __init__( self : List[Any] ,*lowercase__ : List[str] ,**lowercase__ : Optional[Any] ): warnings.warn( '''The class GLPNFeatureExtractor is deprecated and will be removed in version 5 of Transformers. Please''' ''' use GLPNImageProcessor instead.''' ,lowercase__ ,) super().__init__(*lowercase__ ,**lowercase__ )
52
1
'''simple docstring''' class a_ : '''simple docstring''' def __init__( self , A , A ) -> int: _SCREAMING_SNAKE_CASE = name _SCREAMING_SNAKE_CASE = val def __str__( self ) -> Dict: return f'{self.__class__.__name__}({self.name}, {self.val})' def __lt__( self , A ) -> Any: return self.val < other.val class a_ : '''simple docstring''' def __init__( self , A ) -> Optional[Any]: _SCREAMING_SNAKE_CASE = {} _SCREAMING_SNAKE_CASE = {} _SCREAMING_SNAKE_CASE = self.build_heap(A ) def __getitem__( self , A ) -> Any: return self.get_value(A ) def snake_case_( self , A ) -> Optional[int]: return (idx - 1) // 2 def snake_case_( self , A ) -> List[Any]: return idx * 2 + 1 def snake_case_( self , A ) -> Dict: return idx * 2 + 2 def snake_case_( self , A ) -> Tuple: return self.heap_dict[key] def snake_case_( self , A ) -> Optional[int]: _SCREAMING_SNAKE_CASE = len(A ) - 1 _SCREAMING_SNAKE_CASE = self.get_parent_idx(A ) for idx, i in enumerate(A ): _SCREAMING_SNAKE_CASE = idx _SCREAMING_SNAKE_CASE = i.val for i in range(A , -1 , -1 ): self.sift_down(A , A ) return array def snake_case_( self , A , A ) -> List[str]: while True: _SCREAMING_SNAKE_CASE = self.get_left_child_idx(A ) # noqa: E741 _SCREAMING_SNAKE_CASE = self.get_right_child_idx(A ) _SCREAMING_SNAKE_CASE = idx if l < len(A ) and array[l] < array[idx]: _SCREAMING_SNAKE_CASE = l if r < len(A ) and array[r] < array[smallest]: _SCREAMING_SNAKE_CASE = r if smallest != idx: _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE = array[smallest], array[idx] ( ( _SCREAMING_SNAKE_CASE ) , ( _SCREAMING_SNAKE_CASE ) , ) = ( self.idx_of_element[array[smallest]], self.idx_of_element[array[idx]], ) _SCREAMING_SNAKE_CASE = smallest else: break def snake_case_( self , A ) -> int: _SCREAMING_SNAKE_CASE = self.get_parent_idx(A ) while p >= 0 and self.heap[p] > self.heap[idx]: _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE = self.heap[idx], self.heap[p] _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE = ( self.idx_of_element[self.heap[idx]], self.idx_of_element[self.heap[p]], ) _SCREAMING_SNAKE_CASE = p _SCREAMING_SNAKE_CASE = self.get_parent_idx(A ) def snake_case_( self ) -> Optional[Any]: return self.heap[0] def snake_case_( self ) -> Dict: _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE = self.heap[-1], self.heap[0] _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE = ( self.idx_of_element[self.heap[-1]], self.idx_of_element[self.heap[0]], ) _SCREAMING_SNAKE_CASE = self.heap.pop() del self.idx_of_element[x] self.sift_down(0 , self.heap ) return x def snake_case_( self , A ) -> Optional[int]: self.heap.append(A ) _SCREAMING_SNAKE_CASE = len(self.heap ) - 1 _SCREAMING_SNAKE_CASE = node.val self.sift_up(len(self.heap ) - 1 ) def snake_case_( self ) -> Optional[int]: return len(self.heap ) == 0 def snake_case_( self , A , A ) -> Any: assert ( self.heap[self.idx_of_element[node]].val > new_value ), "newValue must be less that current value" _SCREAMING_SNAKE_CASE = new_value _SCREAMING_SNAKE_CASE = new_value self.sift_up(self.idx_of_element[node] ) lowercase_ = Node("""R""", -1) lowercase_ = Node("""B""", 6) lowercase_ = Node("""A""", 3) lowercase_ = Node("""X""", 1) lowercase_ = Node("""E""", 4) # Use one of these two ways to generate Min-Heap # Generating Min-Heap from array lowercase_ = MinHeap([r, b, a, x, e]) # Generating Min-Heap by Insert method # myMinHeap.insert(a) # myMinHeap.insert(b) # myMinHeap.insert(x) # myMinHeap.insert(r) # myMinHeap.insert(e) # Before print("""Min Heap - before decrease key""") for i in my_min_heap.heap: print(i) print("""Min Heap - After decrease key of node [B -> -17]""") my_min_heap.decrease_key(b, -17) # After for i in my_min_heap.heap: print(i) if __name__ == "__main__": import doctest doctest.testmod()
58
import random import timeit from functools import wraps from typing import Callable, Optional from ..configuration_utils import PretrainedConfig from ..models.auto.modeling_tf_auto import TF_MODEL_MAPPING, TF_MODEL_WITH_LM_HEAD_MAPPING from ..utils import is_pyanvml_available, is_tf_available, logging from .benchmark_utils import ( Benchmark, Memory, MemorySummary, measure_peak_memory_cpu, start_memory_tracing, stop_memory_tracing, ) if is_tf_available(): import tensorflow as tf from tensorflow.python.framework.errors_impl import ResourceExhaustedError from .benchmark_args_tf import TensorFlowBenchmarkArguments if is_pyanvml_available(): import pyanvml.pyanvml as nvml snake_case_ : Tuple = logging.get_logger(__name__) def A (__A : bool , __A : bool ) -> Optional[Any]: """simple docstring""" def run_func(__A : Optional[Any] ): @wraps(__A ) def run_in_eager_mode(*__A : Dict , **__A : List[Any] ): return func(*__A , **__A ) @wraps(__A ) @tf.function(experimental_compile=__A ) def run_in_graph_mode(*__A : Optional[Any] , **__A : Any ): return func(*__A , **__A ) if do_eager_mode is True: if use_xla is not False: raise ValueError( '''Cannot run model in XLA, if `args.eager_mode` is set to `True`. Please set `args.eager_mode=False`.''' ) return run_in_eager_mode else: return run_in_graph_mode return run_func def A (__A : int , __A : int , __A : int ) -> ["tf.Tensor"]: """simple docstring""" UpperCAmelCase_ = random.Random() UpperCAmelCase_ = [rng.randint(0 , vocab_size - 1 ) for i in range(batch_size * sequence_length )] return tf.constant(__A , shape=(batch_size, sequence_length) , dtype=tf.intaa ) class __snake_case ( a ): UpperCAmelCase__ : TensorFlowBenchmarkArguments UpperCAmelCase__ : PretrainedConfig UpperCAmelCase__ : str = "TensorFlow" @property def lowerCamelCase ( self : List[str]): """simple docstring""" return tf.__version__ def lowerCamelCase ( self : Dict , _snake_case : str , _snake_case : int , _snake_case : int): """simple docstring""" UpperCAmelCase_ = self.args.strategy if strategy is None: raise ValueError('''A device strategy has to be initialized before using TensorFlow.''') UpperCAmelCase_ = self._prepare_inference_func(_snake_case , _snake_case , _snake_case) return self._measure_speed(_inference) def lowerCamelCase ( self : Any , _snake_case : str , _snake_case : int , _snake_case : int): """simple docstring""" UpperCAmelCase_ = self.args.strategy if strategy is None: raise ValueError('''A device strategy has to be initialized before using TensorFlow.''') UpperCAmelCase_ = self._prepare_train_func(_snake_case , _snake_case , _snake_case) return self._measure_speed(_train) def lowerCamelCase ( self : Any , _snake_case : str , _snake_case : int , _snake_case : int): """simple docstring""" if self.args.is_gpu: tf.config.experimental.set_memory_growth(self.args.gpu_list[self.args.device_idx] , _snake_case) UpperCAmelCase_ = self.args.strategy if strategy is None: raise ValueError('''A device strategy has to be initialized before using TensorFlow.''') UpperCAmelCase_ = self._prepare_inference_func(_snake_case , _snake_case , _snake_case) return self._measure_memory(_inference) def lowerCamelCase ( self : Optional[Any] , _snake_case : str , _snake_case : int , _snake_case : int): """simple docstring""" if self.args.is_gpu: tf.config.experimental.set_memory_growth(self.args.gpu_list[self.args.device_idx] , _snake_case) UpperCAmelCase_ = self.args.strategy if strategy is None: raise ValueError('''A device strategy has to be initialized before using TensorFlow.''') UpperCAmelCase_ = self._prepare_train_func(_snake_case , _snake_case , _snake_case) return self._measure_memory(_train) def lowerCamelCase ( self : Optional[int] , _snake_case : str , _snake_case : int , _snake_case : int): """simple docstring""" UpperCAmelCase_ = self.config_dict[model_name] if self.args.fpaa: raise NotImplementedError('''Mixed precision is currently not supported.''') UpperCAmelCase_ = ( hasattr(_snake_case , '''architectures''') and isinstance(config.architectures , _snake_case) and len(config.architectures) > 0 ) if not self.args.only_pretrain_model and has_model_class_in_config: try: UpperCAmelCase_ = '''TF''' + config.architectures[0] # prepend 'TF' for tensorflow model UpperCAmelCase_ = __import__('''transformers''' , fromlist=[model_class]) UpperCAmelCase_ = getattr(_snake_case , _snake_case) UpperCAmelCase_ = model_cls(_snake_case) except ImportError: raise ImportError( F"""{model_class} does not exist. If you just want to test the pretrained model, you might want to""" ''' set `--only_pretrain_model` or `args.only_pretrain_model=True`.''') else: UpperCAmelCase_ = TF_MODEL_MAPPING[config.__class__](_snake_case) # encoder-decoder has vocab size saved differently UpperCAmelCase_ = config.vocab_size if hasattr(_snake_case , '''vocab_size''') else config.encoder.vocab_size UpperCAmelCase_ = random_input_ids(_snake_case , _snake_case , _snake_case) @run_with_tf_optimizations(self.args.eager_mode , self.args.use_xla) def encoder_decoder_forward(): return model(_snake_case , decoder_input_ids=_snake_case , training=_snake_case) @run_with_tf_optimizations(self.args.eager_mode , self.args.use_xla) def encoder_forward(): return model(_snake_case , training=_snake_case) UpperCAmelCase_ = encoder_decoder_forward if config.is_encoder_decoder else encoder_forward return _inference def lowerCamelCase ( self : Optional[Any] , _snake_case : str , _snake_case : int , _snake_case : int): """simple docstring""" UpperCAmelCase_ = self.config_dict[model_name] if self.args.eager_mode is not False: raise ValueError('''Training cannot be done in eager mode. Please make sure that `args.eager_mode = False`.''') if self.args.fpaa: raise NotImplementedError('''Mixed precision is currently not supported.''') UpperCAmelCase_ = ( hasattr(_snake_case , '''architectures''') and isinstance(config.architectures , _snake_case) and len(config.architectures) > 0 ) if not self.args.only_pretrain_model and has_model_class_in_config: try: UpperCAmelCase_ = '''TF''' + config.architectures[0] # prepend 'TF' for tensorflow model UpperCAmelCase_ = __import__('''transformers''' , fromlist=[model_class]) UpperCAmelCase_ = getattr(_snake_case , _snake_case) UpperCAmelCase_ = model_cls(_snake_case) except ImportError: raise ImportError( F"""{model_class} does not exist. If you just want to test the pretrained model, you might want to""" ''' set `--only_pretrain_model` or `args.only_pretrain_model=True`.''') else: UpperCAmelCase_ = TF_MODEL_WITH_LM_HEAD_MAPPING[config.__class__](_snake_case) # encoder-decoder has vocab size saved differently UpperCAmelCase_ = config.vocab_size if hasattr(_snake_case , '''vocab_size''') else config.encoder.vocab_size UpperCAmelCase_ = random_input_ids(_snake_case , _snake_case , _snake_case) @run_with_tf_optimizations(self.args.eager_mode , self.args.use_xla) def encoder_decoder_train(): UpperCAmelCase_ = model(_snake_case , decoder_input_ids=_snake_case , labels=_snake_case , training=_snake_case)[0] UpperCAmelCase_ = tf.gradients(_snake_case , model.trainable_variables) return gradients @run_with_tf_optimizations(self.args.eager_mode , self.args.use_xla) def encoder_train(): UpperCAmelCase_ = model(_snake_case , labels=_snake_case , training=_snake_case)[0] UpperCAmelCase_ = tf.gradients(_snake_case , model.trainable_variables) return gradients UpperCAmelCase_ = encoder_decoder_train if config.is_encoder_decoder else encoder_train return _train def lowerCamelCase ( self : Any , _snake_case : Optional[Any]): """simple docstring""" with self.args.strategy.scope(): try: if self.args.is_tpu or self.args.use_xla: # run additional 10 times to stabilize compilation for tpu logger.info('''Do inference on TPU. Running model 5 times to stabilize compilation''') timeit.repeat(_snake_case , repeat=1 , number=5) # as written in https://docs.python.org/2/library/timeit.html#timeit.Timer.repeat, min should be taken rather than the average UpperCAmelCase_ = timeit.repeat( _snake_case , repeat=self.args.repeat , number=10 , ) return min(_snake_case) / 1_0.0 except ResourceExhaustedError as e: self.print_fn(F"""Doesn't fit on GPU. {e}""") def lowerCamelCase ( self : Dict , _snake_case : Callable[[], None]): """simple docstring""" logger.info( '''Note that TensorFlow allocates more memory than ''' '''it might need to speed up computation. ''' '''The memory reported here corresponds to the memory ''' '''reported by `nvidia-smi`, which can vary depending ''' '''on total available memory on the GPU that is used.''') with self.args.strategy.scope(): try: if self.args.trace_memory_line_by_line: if not self.args.eager_mode: raise ValueError( '''`args.eager_mode` is set to `False`. Make sure to run model in eager mode to measure memory''' ''' consumption line by line.''') UpperCAmelCase_ = start_memory_tracing('''transformers''') if self.args.is_tpu: # tpu raise NotImplementedError( '''Memory Benchmarking is currently not implemented for TPU. Please disable memory benchmarking''' ''' with `args.memory=False`''') elif self.args.is_gpu: # gpu if not is_pyanvml_available(): logger.warning( '''py3nvml not installed, we won\'t log GPU memory usage. ''' '''Install py3nvml (pip install py3nvml) to log information about GPU.''') UpperCAmelCase_ = '''N/A''' else: logger.info( '''Measuring total GPU usage on GPU device. Make sure to not have additional processes''' ''' running on the same GPU.''') # init nvml nvml.nvmlInit() func() UpperCAmelCase_ = nvml.nvmlDeviceGetHandleByIndex(self.args.device_idx) UpperCAmelCase_ = nvml.nvmlDeviceGetMemoryInfo(_snake_case) UpperCAmelCase_ = meminfo.used UpperCAmelCase_ = Memory(_snake_case) # shutdown nvml nvml.nvmlShutdown() else: # cpu if self.args.trace_memory_line_by_line: logger.info( '''When enabling line by line tracing, the max peak memory for CPU is inaccurate in''' ''' TensorFlow.''') UpperCAmelCase_ = None else: UpperCAmelCase_ = measure_peak_memory_cpu(_snake_case) UpperCAmelCase_ = Memory(_snake_case) if isinstance(_snake_case , _snake_case) else memory_bytes if self.args.trace_memory_line_by_line: UpperCAmelCase_ = stop_memory_tracing(_snake_case) if memory is None: UpperCAmelCase_ = summary.total else: UpperCAmelCase_ = None return memory, summary except ResourceExhaustedError as e: self.print_fn(F"""Doesn't fit on GPU. {e}""") return "N/A", None
51
0
'''simple docstring''' import os from tempfile import TemporaryDirectory from unittest import TestCase import pytest from absl.testing import parameterized from datasets import config from datasets.arrow_reader import HF_GCP_BASE_URL from datasets.builder import DatasetBuilder from datasets.dataset_dict import IterableDatasetDict from datasets.iterable_dataset import IterableDataset from datasets.load import dataset_module_factory, import_main_class from datasets.utils.file_utils import cached_path __lowercase: Union[str, Any] = [ {"dataset": "wikipedia", "config_name": "20220301.de"}, {"dataset": "wikipedia", "config_name": "20220301.en"}, {"dataset": "wikipedia", "config_name": "20220301.fr"}, {"dataset": "wikipedia", "config_name": "20220301.frr"}, {"dataset": "wikipedia", "config_name": "20220301.it"}, {"dataset": "wikipedia", "config_name": "20220301.simple"}, {"dataset": "snli", "config_name": "plain_text"}, {"dataset": "eli5", "config_name": "LFQA_reddit"}, {"dataset": "wiki40b", "config_name": "en"}, {"dataset": "wiki_dpr", "config_name": "psgs_w100.nq.compressed"}, {"dataset": "wiki_dpr", "config_name": "psgs_w100.nq.no_index"}, {"dataset": "wiki_dpr", "config_name": "psgs_w100.multiset.no_index"}, {"dataset": "natural_questions", "config_name": "default"}, ] def SCREAMING_SNAKE_CASE__( _UpperCamelCase : Dict=True ) -> Tuple: '''simple docstring''' if with_config: return [ { "testcase_name": d["dataset"] + "/" + d["config_name"], "dataset": d["dataset"], "config_name": d["config_name"], } for d in DATASETS_ON_HF_GCP ] else: return [ {"testcase_name": dataset, "dataset": dataset} for dataset in {d["dataset"] for d in DATASETS_ON_HF_GCP} ] @parameterized.named_parameters(list_datasets_on_hf_gcp_parameters(with_config=SCREAMING_SNAKE_CASE__)) class UpperCAmelCase ( SCREAMING_SNAKE_CASE__): _lowerCamelCase : List[str] = None _lowerCamelCase : Tuple = None def lowercase_ ( self : Optional[int], a_ : Any, a_ : Optional[int] ): """simple docstring""" with TemporaryDirectory() as tmp_dir: UpperCamelCase__ = dataset_module_factory(a_, cache_dir=a_ ) UpperCamelCase__ = import_main_class(dataset_module.module_path, dataset=a_ ) UpperCamelCase__ = builder_cls( cache_dir=a_, config_name=a_, hash=dataset_module.hash, ) UpperCamelCase__ = "/".join( [ HF_GCP_BASE_URL, builder_instance._relative_data_dir(with_hash=a_ ).replace(os.sep, "/" ), config.DATASET_INFO_FILENAME, ] ) UpperCamelCase__ = cached_path(a_, cache_dir=a_ ) self.assertTrue(os.path.exists(a_ ) ) @pytest.mark.integration def SCREAMING_SNAKE_CASE__( _UpperCamelCase : List[str] ) -> Dict: '''simple docstring''' UpperCamelCase__ = tmp_path_factory.mktemp("test_hf_gcp" ) / "test_wikipedia_simple" UpperCamelCase__ = dataset_module_factory("wikipedia" , cache_dir=_UpperCamelCase ) UpperCamelCase__ = import_main_class(dataset_module.module_path ) UpperCamelCase__ = builder_cls( cache_dir=_UpperCamelCase , config_name="20220301.frr" , hash=dataset_module.hash , ) # use the HF cloud storage, not the original download_and_prepare that uses apache-beam UpperCamelCase__ = None builder_instance.download_and_prepare() UpperCamelCase__ = builder_instance.as_dataset() assert ds @pytest.mark.integration def SCREAMING_SNAKE_CASE__( _UpperCamelCase : Dict ) -> Dict: '''simple docstring''' UpperCamelCase__ = dataset_module_factory("wikipedia" , cache_dir=_UpperCamelCase ) UpperCamelCase__ = import_main_class(dataset_module.module_path , dataset=_UpperCamelCase ) UpperCamelCase__ = builder_cls( cache_dir=_UpperCamelCase , config_name="20220301.frr" , hash=dataset_module.hash , ) UpperCamelCase__ = builder_instance.as_streaming_dataset() assert ds assert isinstance(_UpperCamelCase , _UpperCamelCase ) assert "train" in ds assert isinstance(ds["train"] , _UpperCamelCase ) assert next(iter(ds["train"] ) )
31
'''simple docstring''' import itertools import random import unittest import numpy as np from transformers import is_speech_available from transformers.testing_utils import require_torch, require_torchaudio from ...test_sequence_feature_extraction_common import SequenceFeatureExtractionTestMixin if is_speech_available(): from transformers import SpeechaTextFeatureExtractor __lowercase: str = random.Random() def SCREAMING_SNAKE_CASE__( _UpperCamelCase : List[str] , _UpperCamelCase : Optional[int]=1.0 , _UpperCamelCase : Dict=None , _UpperCamelCase : List[str]=None ) -> Union[str, Any]: '''simple docstring''' if rng is None: UpperCamelCase__ = global_rng UpperCamelCase__ = [] for batch_idx in range(shape[0] ): values.append([] ) for _ in range(shape[1] ): values[-1].append(rng.random() * scale ) return values @require_torch @require_torchaudio class UpperCAmelCase ( unittest.TestCase): def __init__( self : List[Any], a_ : List[str], a_ : Any=7, a_ : Dict=400, a_ : str=2000, a_ : List[Any]=24, a_ : int=24, a_ : int=0.0, a_ : Union[str, Any]=1_6000, a_ : Union[str, Any]=True, a_ : Optional[Any]=True, ): """simple docstring""" UpperCamelCase__ = parent UpperCamelCase__ = batch_size UpperCamelCase__ = min_seq_length UpperCamelCase__ = max_seq_length UpperCamelCase__ = (self.max_seq_length - self.min_seq_length) // (self.batch_size - 1) UpperCamelCase__ = feature_size UpperCamelCase__ = num_mel_bins UpperCamelCase__ = padding_value UpperCamelCase__ = sampling_rate UpperCamelCase__ = return_attention_mask UpperCamelCase__ = do_normalize def lowercase_ ( self : Tuple ): """simple docstring""" return { "feature_size": self.feature_size, "num_mel_bins": self.num_mel_bins, "padding_value": self.padding_value, "sampling_rate": self.sampling_rate, "return_attention_mask": self.return_attention_mask, "do_normalize": self.do_normalize, } def lowercase_ ( self : Optional[Any], a_ : Union[str, Any]=False, a_ : Optional[int]=False ): """simple docstring""" def _flatten(a_ : Dict ): return list(itertools.chain(*a_ ) ) if equal_length: UpperCamelCase__ = [floats_list((self.max_seq_length, self.feature_size) ) for _ in range(self.batch_size )] else: # make sure that inputs increase in size UpperCamelCase__ = [ floats_list((x, self.feature_size) ) for x in range(self.min_seq_length, self.max_seq_length, self.seq_length_diff ) ] if numpify: UpperCamelCase__ = [np.asarray(a_ ) for x in speech_inputs] return speech_inputs @require_torch @require_torchaudio class UpperCAmelCase ( SCREAMING_SNAKE_CASE__ , unittest.TestCase): _lowerCamelCase : Dict = SpeechaTextFeatureExtractor if is_speech_available() else None def lowercase_ ( self : Any ): """simple docstring""" UpperCamelCase__ = SpeechaTextFeatureExtractionTester(self ) def lowercase_ ( self : Optional[int], a_ : Tuple ): """simple docstring""" self.assertTrue(np.all(np.mean(a_, axis=0 ) < 1e-3 ) ) self.assertTrue(np.all(np.abs(np.var(a_, axis=0 ) - 1 ) < 1e-3 ) ) def lowercase_ ( self : Any ): """simple docstring""" UpperCamelCase__ = self.feature_extraction_class(**self.feat_extract_tester.prepare_feat_extract_dict() ) # create three inputs of length 800, 1000, and 1200 UpperCamelCase__ = [floats_list((1, x) )[0] for x in range(800, 1400, 200 )] UpperCamelCase__ = [np.asarray(a_ ) for speech_input in speech_inputs] # Test feature size UpperCamelCase__ = feature_extractor(a_, padding=a_, return_tensors="np" ).input_features self.assertTrue(input_features.ndim == 3 ) self.assertTrue(input_features.shape[-1] == feature_extractor.feature_size ) # Test not batched input UpperCamelCase__ = feature_extractor(speech_inputs[0], return_tensors="np" ).input_features UpperCamelCase__ = feature_extractor(np_speech_inputs[0], return_tensors="np" ).input_features self.assertTrue(np.allclose(a_, a_, atol=1e-3 ) ) # Test batched UpperCamelCase__ = feature_extractor(a_, return_tensors="np" ).input_features UpperCamelCase__ = feature_extractor(a_, return_tensors="np" ).input_features 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. UpperCamelCase__ = [floats_list((1, x) )[0] for x in (800, 800, 800)] UpperCamelCase__ = np.asarray(a_ ) UpperCamelCase__ = feature_extractor(a_, return_tensors="np" ).input_features UpperCamelCase__ = feature_extractor(a_, return_tensors="np" ).input_features for enc_seq_a, enc_seq_a in zip(a_, a_ ): self.assertTrue(np.allclose(a_, a_, atol=1e-3 ) ) def lowercase_ ( self : List[str] ): """simple docstring""" UpperCamelCase__ = self.feature_extraction_class(**self.feat_extract_tester.prepare_feat_extract_dict() ) UpperCamelCase__ = [floats_list((1, x) )[0] for x in range(800, 1400, 200 )] UpperCamelCase__ = ["longest", "max_length", "do_not_pad"] UpperCamelCase__ = [None, 16, None] for max_length, padding in zip(a_, a_ ): UpperCamelCase__ = feature_extractor( a_, padding=a_, max_length=a_, return_attention_mask=a_ ) UpperCamelCase__ = inputs.input_features UpperCamelCase__ = inputs.attention_mask UpperCamelCase__ = [np.sum(a_ ) for x in attention_mask] self._check_zero_mean_unit_variance(input_features[0][: fbank_feat_lengths[0]] ) self._check_zero_mean_unit_variance(input_features[1][: fbank_feat_lengths[1]] ) self._check_zero_mean_unit_variance(input_features[2][: fbank_feat_lengths[2]] ) def lowercase_ ( self : Any ): """simple docstring""" UpperCamelCase__ = self.feature_extraction_class(**self.feat_extract_tester.prepare_feat_extract_dict() ) UpperCamelCase__ = [floats_list((1, x) )[0] for x in range(800, 1400, 200 )] UpperCamelCase__ = ["longest", "max_length", "do_not_pad"] UpperCamelCase__ = [None, 16, None] for max_length, padding in zip(a_, a_ ): UpperCamelCase__ = feature_extractor( a_, max_length=a_, padding=a_, return_tensors="np", return_attention_mask=a_ ) UpperCamelCase__ = inputs.input_features UpperCamelCase__ = inputs.attention_mask UpperCamelCase__ = [np.sum(a_ ) for x in attention_mask] self._check_zero_mean_unit_variance(input_features[0][: fbank_feat_lengths[0]] ) self.assertTrue(input_features[0][fbank_feat_lengths[0] :].sum() < 1e-6 ) self._check_zero_mean_unit_variance(input_features[1][: fbank_feat_lengths[1]] ) self.assertTrue(input_features[0][fbank_feat_lengths[1] :].sum() < 1e-6 ) self._check_zero_mean_unit_variance(input_features[2][: fbank_feat_lengths[2]] ) def lowercase_ ( self : str ): """simple docstring""" UpperCamelCase__ = self.feature_extraction_class(**self.feat_extract_tester.prepare_feat_extract_dict() ) UpperCamelCase__ = [floats_list((1, x) )[0] for x in range(800, 1400, 200 )] UpperCamelCase__ = feature_extractor( a_, padding="max_length", max_length=4, truncation=a_, return_tensors="np", return_attention_mask=a_, ) UpperCamelCase__ = inputs.input_features UpperCamelCase__ = inputs.attention_mask UpperCamelCase__ = np.sum(attention_mask == 1, axis=1 ) self._check_zero_mean_unit_variance(input_features[0, : fbank_feat_lengths[0]] ) self._check_zero_mean_unit_variance(input_features[1] ) self._check_zero_mean_unit_variance(input_features[2] ) def lowercase_ ( self : Any ): """simple docstring""" UpperCamelCase__ = self.feature_extraction_class(**self.feat_extract_tester.prepare_feat_extract_dict() ) UpperCamelCase__ = [floats_list((1, x) )[0] for x in range(800, 1400, 200 )] UpperCamelCase__ = feature_extractor( a_, padding="longest", max_length=4, truncation=a_, return_tensors="np", return_attention_mask=a_, ) UpperCamelCase__ = inputs.input_features UpperCamelCase__ = inputs.attention_mask UpperCamelCase__ = np.sum(attention_mask == 1, axis=1 ) self._check_zero_mean_unit_variance(input_features[0, : fbank_feat_lengths[0]] ) self._check_zero_mean_unit_variance(input_features[1, : fbank_feat_lengths[1]] ) self._check_zero_mean_unit_variance(input_features[2] ) # make sure that if max_length < longest -> then pad to max_length self.assertEqual(input_features.shape, (3, 4, 24) ) UpperCamelCase__ = [floats_list((1, x) )[0] for x in range(800, 1400, 200 )] UpperCamelCase__ = feature_extractor( a_, padding="longest", max_length=16, truncation=a_, return_tensors="np", return_attention_mask=a_, ) UpperCamelCase__ = inputs.input_features UpperCamelCase__ = inputs.attention_mask UpperCamelCase__ = np.sum(attention_mask == 1, axis=1 ) self._check_zero_mean_unit_variance(input_features[0, : fbank_feat_lengths[0]] ) self._check_zero_mean_unit_variance(input_features[1, : fbank_feat_lengths[1]] ) self._check_zero_mean_unit_variance(input_features[2] ) # make sure that if max_length < longest -> then pad to max_length self.assertEqual(input_features.shape, (3, 6, 24) ) def lowercase_ ( self : Optional[Any] ): """simple docstring""" import torch UpperCamelCase__ = self.feature_extraction_class(**self.feat_extract_tester.prepare_feat_extract_dict() ) UpperCamelCase__ = np.random.rand(100, 32 ).astype(np.floataa ) UpperCamelCase__ = np_speech_inputs.tolist() for inputs in [py_speech_inputs, np_speech_inputs]: UpperCamelCase__ = feature_extractor.pad([{"input_features": inputs}], return_tensors="np" ) self.assertTrue(np_processed.input_features.dtype == np.floataa ) UpperCamelCase__ = feature_extractor.pad([{"input_features": inputs}], return_tensors="pt" ) self.assertTrue(pt_processed.input_features.dtype == torch.floataa ) def lowercase_ ( self : List[str], a_ : int ): """simple docstring""" from datasets import load_dataset UpperCamelCase__ = load_dataset("hf-internal-testing/librispeech_asr_dummy", "clean", split="validation" ) # automatic decoding with librispeech UpperCamelCase__ = ds.sort("id" ).select(range(a_ ) )[:num_samples]["audio"] return [x["array"] for x in speech_samples] def lowercase_ ( self : int ): """simple docstring""" UpperCamelCase__ = np.array([ -1.5_745, -1.7_713, -1.7_020, -1.6_069, -1.2_250, -1.1_105, -0.9_072, -0.8_241, -1.2_310, -0.8_098, -0.3_320, -0.4_101, -0.7_985, -0.4_996, -0.8_213, -0.9_128, -1.0_420, -1.1_286, -1.0_440, -0.7_999, -0.8_405, -1.2_275, -1.5_443, -1.4_625, ] ) # fmt: on UpperCamelCase__ = self._load_datasamples(1 ) UpperCamelCase__ = self.feature_extraction_class(**self.feat_extract_tester.prepare_feat_extract_dict() ) UpperCamelCase__ = feature_extractor(a_, return_tensors="pt" ).input_features self.assertEquals(input_features.shape, (1, 584, 24) ) self.assertTrue(np.allclose(input_features[0, 0, :30], a_, atol=1e-4 ) )
31
1
'''simple docstring''' from cva import destroyAllWindows, imread, imshow, waitKey def __snake_case ( UpperCAmelCase_ : Any ): # getting number of pixels in the image lowerCamelCase_ ,lowerCamelCase_ = img.shape[0], img.shape[1] # converting each pixel's color to its negative for i in range(UpperCAmelCase_ ): for j in range(UpperCAmelCase_ ): lowerCamelCase_ = [255, 255, 255] - img[i][j] return img if __name__ == "__main__": # read original image a_ : int = imread("""image_data/lena.jpg""", 1) # convert to its negative a_ : Any = convert_to_negative(img) # show result image imshow("""negative of original image""", img) waitKey(0) destroyAllWindows()
55
'''simple docstring''' from __future__ import annotations def __snake_case ( UpperCAmelCase_ : int ): lowerCamelCase_ = 2 lowerCamelCase_ = [] while i * i <= n: if n % i: i += 1 else: n //= i factors.append(UpperCAmelCase_ ) if n > 1: factors.append(UpperCAmelCase_ ) return factors if __name__ == "__main__": import doctest doctest.testmod()
55
1
"""simple docstring""" from ...configuration_utils import PretrainedConfig from ...utils import logging A = logging.get_logger(__name__) A = { '''uw-madison/mra-base-512-4''': '''https://huggingface.co/uw-madison/mra-base-512-4/resolve/main/config.json''', } class __lowercase ( _UpperCamelCase ): '''simple docstring''' __lowerCAmelCase = '''mra''' def __init__( self , _UpperCAmelCase=50265 , _UpperCAmelCase=768 , _UpperCAmelCase=12 , _UpperCAmelCase=12 , _UpperCAmelCase=3072 , _UpperCAmelCase="gelu" , _UpperCAmelCase=0.1 , _UpperCAmelCase=0.1 , _UpperCAmelCase=512 , _UpperCAmelCase=1 , _UpperCAmelCase=0.0_2 , _UpperCAmelCase=1e-5 , _UpperCAmelCase="absolute" , _UpperCAmelCase=4 , _UpperCAmelCase="full" , _UpperCAmelCase=0 , _UpperCAmelCase=0 , _UpperCAmelCase=1 , _UpperCAmelCase=0 , _UpperCAmelCase=2 , **_UpperCAmelCase , ): super().__init__(pad_token_id=_UpperCAmelCase , bos_token_id=_UpperCAmelCase , eos_token_id=_UpperCAmelCase , **_UpperCAmelCase ) __a : Any = vocab_size __a : Tuple = max_position_embeddings __a : Union[str, Any] = hidden_size __a : Dict = num_hidden_layers __a : Tuple = num_attention_heads __a : Optional[Any] = intermediate_size __a : Dict = hidden_act __a : List[str] = hidden_dropout_prob __a : List[Any] = attention_probs_dropout_prob __a : List[Any] = initializer_range __a : Optional[int] = type_vocab_size __a : Union[str, Any] = layer_norm_eps __a : int = position_embedding_type __a : Optional[Any] = block_per_row __a : Any = approx_mode __a : Optional[Any] = initial_prior_first_n_blocks __a : List[Any] = initial_prior_diagonal_n_blocks
358
"""simple docstring""" 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 ViTImageProcessor class __lowercase ( unittest.TestCase ): '''simple docstring''' def __init__( self , _UpperCAmelCase , _UpperCAmelCase=13 , _UpperCAmelCase=3 , _UpperCAmelCase=224 , _UpperCAmelCase=30 , _UpperCAmelCase=400 , _UpperCAmelCase=True , _UpperCAmelCase=None , _UpperCAmelCase=True , _UpperCAmelCase=[0.5, 0.5, 0.5] , _UpperCAmelCase=[0.5, 0.5, 0.5] , ): __a : int = size if size is not None else {'''height''': 18, '''width''': 18} __a : List[Any] = parent __a : Dict = batch_size __a : Dict = num_channels __a : int = image_size __a : Optional[Any] = min_resolution __a : Optional[int] = max_resolution __a : Dict = do_resize __a : List[Any] = size __a : int = do_normalize __a : Optional[Any] = image_mean __a : int = image_std def _lowerCamelCase ( self ): return { "image_mean": self.image_mean, "image_std": self.image_std, "do_normalize": self.do_normalize, "do_resize": self.do_resize, "size": self.size, } @require_torch @require_vision class __lowercase ( _UpperCamelCase , unittest.TestCase ): '''simple docstring''' __lowerCAmelCase = ViTImageProcessor if is_vision_available() else None def _lowerCamelCase ( self ): __a : int = EfficientFormerImageProcessorTester(self ) @property def _lowerCamelCase ( self ): return self.image_proc_tester.prepare_image_processor_dict() def _lowerCamelCase ( self ): __a : List[str] = self.image_processing_class(**self.image_processor_dict ) self.assertTrue(hasattr(_UpperCAmelCase , '''image_mean''' ) ) self.assertTrue(hasattr(_UpperCAmelCase , '''image_std''' ) ) self.assertTrue(hasattr(_UpperCAmelCase , '''do_normalize''' ) ) self.assertTrue(hasattr(_UpperCAmelCase , '''do_resize''' ) ) self.assertTrue(hasattr(_UpperCAmelCase , '''size''' ) ) def _lowerCamelCase ( self ): pass def _lowerCamelCase ( self ): # Initialize image_processor __a : Any = self.image_processing_class(**self.image_processor_dict ) # create random PIL images __a : int = prepare_image_inputs(self.image_proc_tester , equal_resolution=_UpperCAmelCase ) for image in image_inputs: self.assertIsInstance(_UpperCAmelCase , Image.Image ) # Test not batched input __a : Dict = image_processor(image_inputs[0] , return_tensors='''pt''' ).pixel_values self.assertEqual( encoded_images.shape , ( 1, self.image_proc_tester.num_channels, self.image_proc_tester.size['''height'''], self.image_proc_tester.size['''width'''], ) , ) # Test batched __a : Any = image_processor(_UpperCAmelCase , return_tensors='''pt''' ).pixel_values self.assertEqual( encoded_images.shape , ( self.image_proc_tester.batch_size, self.image_proc_tester.num_channels, self.image_proc_tester.size['''height'''], self.image_proc_tester.size['''width'''], ) , ) def _lowerCamelCase ( self ): # Initialize image_processor __a : Tuple = self.image_processing_class(**self.image_processor_dict ) # create random numpy tensors __a : int = prepare_image_inputs(self.image_proc_tester , equal_resolution=_UpperCAmelCase , numpify=_UpperCAmelCase ) for image in image_inputs: self.assertIsInstance(_UpperCAmelCase , np.ndarray ) # Test not batched input __a : Optional[int] = image_processor(image_inputs[0] , return_tensors='''pt''' ).pixel_values self.assertEqual( encoded_images.shape , ( 1, self.image_proc_tester.num_channels, self.image_proc_tester.size['''height'''], self.image_proc_tester.size['''width'''], ) , ) # Test batched __a : str = image_processor(_UpperCAmelCase , return_tensors='''pt''' ).pixel_values self.assertEqual( encoded_images.shape , ( self.image_proc_tester.batch_size, self.image_proc_tester.num_channels, self.image_proc_tester.size['''height'''], self.image_proc_tester.size['''width'''], ) , ) def _lowerCamelCase ( self ): # Initialize image_processor __a : int = self.image_processing_class(**self.image_processor_dict ) # create random PyTorch tensors __a : List[Any] = prepare_image_inputs(self.image_proc_tester , equal_resolution=_UpperCAmelCase , torchify=_UpperCAmelCase ) for image in image_inputs: self.assertIsInstance(_UpperCAmelCase , torch.Tensor ) # Test not batched input __a : str = image_processor(image_inputs[0] , return_tensors='''pt''' ).pixel_values self.assertEqual( encoded_images.shape , ( 1, self.image_proc_tester.num_channels, self.image_proc_tester.size['''height'''], self.image_proc_tester.size['''width'''], ) , ) # Test batched __a : Tuple = image_processor(_UpperCAmelCase , return_tensors='''pt''' ).pixel_values self.assertEqual( encoded_images.shape , ( self.image_proc_tester.batch_size, self.image_proc_tester.num_channels, self.image_proc_tester.size['''height'''], self.image_proc_tester.size['''width'''], ) , )
188
0
"""simple docstring""" from __future__ import annotations def lowercase_ ( _snake_case ): SCREAMING_SNAKE_CASE__ : Tuple = [True] * limit SCREAMING_SNAKE_CASE__ : List[Any] = False SCREAMING_SNAKE_CASE__ : int = False SCREAMING_SNAKE_CASE__ : List[Any] = True for i in range(3 ,int(limit**0.5 + 1 ) ,2 ): SCREAMING_SNAKE_CASE__ : Tuple = i * 2 while index < limit: SCREAMING_SNAKE_CASE__ : Union[str, Any] = False SCREAMING_SNAKE_CASE__ : int = index + i SCREAMING_SNAKE_CASE__ : Tuple = [2] for i in range(3 ,_snake_case ,2 ): if is_prime[i]: primes.append(_snake_case ) return primes def lowercase_ ( _snake_case = 1_000_000 ): SCREAMING_SNAKE_CASE__ : Union[str, Any] = prime_sieve(_snake_case ) SCREAMING_SNAKE_CASE__ : int = 0 SCREAMING_SNAKE_CASE__ : Union[str, Any] = 0 for i in range(len(_snake_case ) ): for j in range(i + length ,len(_snake_case ) ): SCREAMING_SNAKE_CASE__ : int = sum(primes[i:j] ) if sol >= ceiling: break if sol in primes: SCREAMING_SNAKE_CASE__ : str = j - i SCREAMING_SNAKE_CASE__ : str = sol return largest if __name__ == "__main__": print(f"""{solution() = }""")
25
import math import os import re import sys import unittest from pathlib import Path from typing import Tuple from unittest.mock import patch from parameterized import parameterized from transformers.testing_utils import ( CaptureStderr, ExtendSysPath, TestCasePlus, execute_subprocess_async, get_gpu_count, get_torch_dist_unique_port, require_apex, require_bitsandbytes, require_fairscale, require_torch, require_torch_gpu, require_torch_multi_gpu, require_torch_non_multi_gpu, slow, ) from transformers.trainer_callback import TrainerState from transformers.trainer_utils import set_seed _SCREAMING_SNAKE_CASE = os.path.abspath(os.path.dirname(__file__)) with ExtendSysPath(F'''{bindir}/../../examples/pytorch/translation'''): from run_translation import main # noqa set_seed(42) _SCREAMING_SNAKE_CASE = 'sshleifer/student_marian_en_ro_6_1' _SCREAMING_SNAKE_CASE = 'sshleifer/tiny-mbart' @require_torch class a ( __lowerCAmelCase ): """simple docstring""" def UpperCAmelCase ( self , lowerCAmelCase_=False , lowerCAmelCase_=None , lowerCAmelCase_=True , lowerCAmelCase_=True , lowerCAmelCase_=True , lowerCAmelCase_=True , ) -> int: _A = self.run_trainer( eval_steps=1 , max_len=12 , model_name=lowerCAmelCase_ , num_train_epochs=1 , distributed=lowerCAmelCase_ , extra_args_str=lowerCAmelCase_ , predict_with_generate=lowerCAmelCase_ , do_train=lowerCAmelCase_ , do_eval=lowerCAmelCase_ , do_predict=lowerCAmelCase_ , ) _A = TrainerState.load_from_json(os.path.join(lowerCAmelCase_ , """trainer_state.json""" ) ).log_history if not do_eval: return _A = [log for log in logs if """eval_loss""" in log.keys()] _A = eval_metrics[0] if predict_with_generate: assert "eval_bleu" in first_step_stats _A = eval_metrics[-1] assert isinstance(last_step_stats["""eval_bleu"""] , lowerCAmelCase_ ) assert not math.isnan(float(last_step_stats["""eval_loss"""] ) ), "eval_loss must not be `nan`" @require_torch_non_multi_gpu def UpperCAmelCase ( self ) -> Optional[int]: self.run_seqaseq_quick() @require_torch_multi_gpu def UpperCAmelCase ( self ) -> Dict: self.run_seqaseq_quick(distributed=lowerCAmelCase_ ) @require_torch_multi_gpu def UpperCAmelCase ( self ) -> Dict: self.run_seqaseq_quick(distributed=lowerCAmelCase_ ) @unittest.skip("""Requires an update of the env running those tests""" ) @require_torch_multi_gpu @require_fairscale def UpperCAmelCase ( self ) -> str: self.run_seqaseq_quick(distributed=lowerCAmelCase_ , extra_args_str="""--sharded_ddp simple""" ) @unittest.skip("""Requires an update of the env running those tests""" ) @require_torch_multi_gpu @require_fairscale def UpperCAmelCase ( self ) -> Dict: self.run_seqaseq_quick(distributed=lowerCAmelCase_ , extra_args_str="""--sharded_ddp simple --fp16""" ) @unittest.skip("""Requires an update of the env running those tests""" ) @require_torch_multi_gpu @require_fairscale def UpperCAmelCase ( self ) -> Optional[Any]: self.run_seqaseq_quick(distributed=lowerCAmelCase_ , extra_args_str="""--sharded_ddp zero_dp_2""" , predict_with_generate=lowerCAmelCase_ ) @unittest.skip("""Requires an update of the env running those tests""" ) @require_torch_multi_gpu @require_fairscale def UpperCAmelCase ( self ) -> Tuple: self.run_seqaseq_quick( distributed=lowerCAmelCase_ , extra_args_str="""--sharded_ddp zero_dp_2 --fp16""" , predict_with_generate=lowerCAmelCase_ ) @require_apex @require_torch_gpu def UpperCAmelCase ( self ) -> int: # XXX: apex breaks the trainer if it's run twice e.g. run_seq2seq.main() from the same # program and it breaks other tests that run from the same pytest worker, therefore until this is # sorted out it must be run only in an external program, that is distributed=True in this # test and only under one or more gpus - if we want cpu will need to make a special test # # specifically to the problem traced it to self.optimizer.step() - if it's run 2nd time via # 2nd main() call it botches the future eval. # self.run_seqaseq_quick(distributed=lowerCAmelCase_ , extra_args_str="""--fp16 --fp16_backend=apex""" ) # test 2nd time - was getting eval_loss': nan' # to reproduce the problem set distributed=False self.run_seqaseq_quick(distributed=lowerCAmelCase_ , extra_args_str="""--fp16 --fp16_backend=apex""" ) @parameterized.expand(["""base""", """low""", """high""", """mixed"""] ) @require_torch_multi_gpu def UpperCAmelCase ( self , lowerCAmelCase_ ) -> int: # as each sub-test is slow-ish split into multiple sub-tests to avoid CI timeout _A = { # test with the default log_level - should be info and thus log info once """base""": {"""extra_args_str""": """""", """n_matches""": 1}, # test with low log_level and log_level_replica - should be noisy on all processes # now the info string should appear twice on 2 processes """low""": {"""extra_args_str""": """--log_level debug --log_level_replica debug""", """n_matches""": 2}, # test with high log_level and low log_level_replica # now the info string should appear once only on the replica """high""": {"""extra_args_str""": """--log_level error --log_level_replica debug""", """n_matches""": 1}, # test with high log_level and log_level_replica - should be quiet on all processes """mixed""": {"""extra_args_str""": """--log_level error --log_level_replica error""", """n_matches""": 0}, } _A = experiments[experiment_id] _A = {"""distributed""": True, """predict_with_generate""": False, """do_eval""": False, """do_predict""": False} _A = """Running training""" with CaptureStderr() as cl: self.run_seqaseq_quick(**lowerCAmelCase_ , extra_args_str=data["""extra_args_str"""] ) _A = len(re.findall(lowerCAmelCase_ , cl.err ) ) self.assertEqual(lowerCAmelCase_ , data["""n_matches"""] ) @slow def UpperCAmelCase ( self ) -> Dict: _A = self.run_trainer( eval_steps=2 , max_len=1_28 , model_name=lowerCAmelCase_ , learning_rate=3E-4 , num_train_epochs=10 , distributed=lowerCAmelCase_ , ) # Check metrics _A = TrainerState.load_from_json(os.path.join(lowerCAmelCase_ , """trainer_state.json""" ) ).log_history _A = [log for log in logs if """eval_loss""" in log.keys()] _A = eval_metrics[0] _A = eval_metrics[-1] assert first_step_stats["eval_loss"] > last_step_stats["eval_loss"], "model learned nothing" assert isinstance(last_step_stats["""eval_bleu"""] , lowerCAmelCase_ ) # test if do_predict saves generations and metrics _A = os.listdir(lowerCAmelCase_ ) _A = {os.path.basename(lowerCAmelCase_ ) for p in contents} assert "generated_predictions.txt" in contents assert "predict_results.json" in contents @slow @require_bitsandbytes def UpperCAmelCase ( self ) -> Optional[Any]: from transformers.training_args import OptimizerNames def train_and_return_metrics(lowerCAmelCase_ ) -> Tuple[int, float]: _A = """--skip_memory_metrics 0""" _A = self.run_trainer( max_len=1_28 , model_name=lowerCAmelCase_ , learning_rate=3E-4 , num_train_epochs=1 , optim=lowerCAmelCase_ , distributed=lowerCAmelCase_ , extra_args_str=lowerCAmelCase_ , do_eval=lowerCAmelCase_ , do_predict=lowerCAmelCase_ , n_gpus_to_use=1 , ) # Check metrics _A = TrainerState.load_from_json(Path(lowerCAmelCase_ , """trainer_state.json""" ) ).log_history _A = int(logs[0]["""train_mem_gpu_peaked_delta"""] / 2**20 ) _A = int(logs[0]["""train_mem_gpu_alloc_delta"""] / 2**20 ) _A = logs[0]["""train_loss"""] return gpu_peak_mem_mb, gpu_alloc_mem_mb, loss _A , _A , _A = train_and_return_metrics(OptimizerNames.ADAMW_TORCH.value ) _A , _A , _A = train_and_return_metrics(OptimizerNames.ADAMW_BNB.value ) _A = gpu_alloc_mem_orig - gpu_alloc_mem_bnb _A = gpu_peak_mem_orig + gpu_alloc_mem_orig _A = gpu_peak_mem_bnb + gpu_alloc_mem_bnb _A = gpu_total_mem_orig - gpu_total_mem_bnb # sshleifer/student_marian_en_ro_6_1 has 54M parameter, 29M of which is `nn.Embedding` which # doesn't get quantized and remains in fp32. Therefore we only have 25M parameters quantized # in 2 bytes and the diff in optim memory usage is derived as so: # # - normal 25*8=~200MB (8 bytes per param) # - bnb 25*2= ~50MB (2 bytes per param) # # Thus we should expect ~150MB total memory saved. # # Peak memory should be the same - the total should be different by about that same margin # # After leaving a small margin to accommodate for differences between gpus let's check # that we have at least 120MB in savings _A = 1_20 # uncomment the following if this test starts failing - requires py38 for a new print feature # gpu_peak_mem_diff = gpu_peak_mem_orig - gpu_peak_mem_bnb # print(f"{gpu_alloc_mem_orig=}MB {gpu_peak_mem_orig=}MB {gpu_alloc_mem_orig+gpu_peak_mem_orig=}MB") # print(f" {gpu_alloc_mem_bnb=}MB {gpu_peak_mem_bnb=}MB {gpu_alloc_mem_bnb+gpu_peak_mem_bnb=}MB") # print(f"{gpu_alloc_mem_diff=}MB") # print(f"{gpu_peak_mem_diff=}MB") # print(f"{gpu_total_mem_orig=}MB, {gpu_total_mem_bnb=}MB") # print(f"{gpu_total_mem_diff=}MB, {gpu_total_mem_diff=}MB") self.assertGreater( lowerCAmelCase_ , lowerCAmelCase_ , """should use ~150MB less alloc gpu memory with BNB, compared to without it for this model but got""" F''' a difference of {gpu_alloc_mem_diff}MB, with gpu_alloc_mem_orig={gpu_alloc_mem_orig}MB and''' F''' gpu_alloc_mem_bnb={gpu_alloc_mem_bnb}MB''' , ) self.assertGreater( lowerCAmelCase_ , lowerCAmelCase_ , """should use ~150MB less total gpu memory with BNB, compared to without it for this model but got""" F''' a difference of {gpu_total_mem_diff}MB, with gpu_total_mem_orig={gpu_total_mem_orig}MB and''' F''' gpu_total_mem_bnb={gpu_total_mem_bnb}MB''' , ) self.assertEqual( lowerCAmelCase_ , lowerCAmelCase_ , F'''loss should be the same, but got loss_orig={loss_orig}, loss_bnb={loss_bnb}''' ) def UpperCAmelCase ( self , lowerCAmelCase_ , lowerCAmelCase_ , lowerCAmelCase_ , lowerCAmelCase_ = 3E-3 , lowerCAmelCase_ = "adafactor" , lowerCAmelCase_ = False , lowerCAmelCase_ = None , lowerCAmelCase_ = 0 , lowerCAmelCase_ = True , lowerCAmelCase_ = True , lowerCAmelCase_ = True , lowerCAmelCase_ = True , lowerCAmelCase_ = None , ) -> str: _A = self.test_file_dir / """../fixtures/tests_samples/wmt_en_ro""" _A = self.get_auto_remove_tmp_dir() _A = F''' --model_name_or_path {model_name} --train_file {data_dir}/train.json --validation_file {data_dir}/val.json --test_file {data_dir}/test.json --output_dir {output_dir} --overwrite_output_dir --max_train_samples 8 --max_source_length {max_len} --max_target_length {max_len} --do_train --num_train_epochs {str(lowerCAmelCase_ )} --per_device_train_batch_size 4 --learning_rate {learning_rate} --warmup_steps 8 --logging_steps 0 --logging_strategy no --save_steps {str(lowerCAmelCase_ )} --group_by_length --label_smoothing_factor 0.1 --target_lang ro_RO --source_lang en_XX '''.split() _A = F''' --do_eval --per_device_eval_batch_size 4 --max_eval_samples 8 --val_max_target_length {max_len} --evaluation_strategy steps --eval_steps {str(lowerCAmelCase_ )} '''.split() _A = """ --do_predict """.split() _A = [] if do_train: args += args_train if do_eval: args += args_eval if do_predict: args += args_predict if predict_with_generate: args += "--predict_with_generate".split() if do_train: if optim == "adafactor": args += "--adafactor".split() else: args += F'''--optim {optim}'''.split() if extra_args_str is not None: args += extra_args_str.split() if distributed: if n_gpus_to_use is None: _A = get_gpu_count() _A = get_torch_dist_unique_port() _A = F''' -m torch.distributed.run --nproc_per_node={n_gpus_to_use} --master_port={master_port} {self.examples_dir_str}/pytorch/translation/run_translation.py '''.split() _A = [sys.executable] + distributed_args + args # keep for quick debug # print(" ".join([f"\nPYTHONPATH={self.src_dir_str}"] +cmd)); die execute_subprocess_async(lowerCAmelCase_ , env=self.get_env() ) else: _A = ["""run_translation.py"""] + args with patch.object(lowerCAmelCase_ , """argv""" , lowerCAmelCase_ ): main() return output_dir
180
0
def lowerCAmelCase_ ( UpperCamelCase_ ) -> int: UpperCamelCase_ = hex_num.strip() if not hex_num: raise ValueError("No value was passed to the function" ) UpperCamelCase_ = hex_num[0] == "-" if is_negative: UpperCamelCase_ = hex_num[1:] try: UpperCamelCase_ = int(UpperCamelCase_ , 16 ) except ValueError: raise ValueError("Invalid value was passed to the function" ) UpperCamelCase_ = "" while int_num > 0: UpperCamelCase_ = 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()
359
from datetime import datetime import matplotlib.pyplot as plt import torch def lowerCAmelCase_ ( UpperCamelCase_ ) -> int: for param in module.parameters(): UpperCamelCase_ = False def lowerCAmelCase_ ( ) -> Dict: UpperCamelCase_ = "cuda" if torch.cuda.is_available() else "cpu" if torch.backends.mps.is_available() and torch.backends.mps.is_built(): UpperCamelCase_ = "mps" if device == "mps": print( "WARNING: MPS currently doesn't seem to work, and messes up backpropagation without any visible torch" " errors. I recommend using CUDA on a colab notebook or CPU instead if you're facing inexplicable issues" " with generations." ) return device def lowerCAmelCase_ ( UpperCamelCase_ ) -> Union[str, Any]: UpperCamelCase_ = plt.imshow(UpperCamelCase_ ) fig.axes.get_xaxis().set_visible(UpperCamelCase_ ) fig.axes.get_yaxis().set_visible(UpperCamelCase_ ) plt.show() def lowerCAmelCase_ ( ) -> List[str]: UpperCamelCase_ = datetime.now() UpperCamelCase_ = current_time.strftime("%H:%M:%S" ) return timestamp
328
0
from __future__ import annotations __a = 1.6021E-19 # units = C def a ( snake_case__: float , snake_case__: float , snake_case__: float , ): '''simple docstring''' if (conductivity, electron_conc, mobility).count(0 ) != 1: raise ValueError('''You cannot supply more or less than 2 values''' ) elif conductivity < 0: raise ValueError('''Conductivity cannot be negative''' ) elif electron_conc < 0: raise ValueError('''Electron concentration cannot be negative''' ) elif mobility < 0: raise ValueError('''mobility cannot be negative''' ) elif conductivity == 0: return ( "conductivity", mobility * electron_conc * ELECTRON_CHARGE, ) elif electron_conc == 0: return ( "electron_conc", conductivity / (mobility * ELECTRON_CHARGE), ) else: return ( "mobility", conductivity / (electron_conc * ELECTRON_CHARGE), ) if __name__ == "__main__": import doctest doctest.testmod()
30
'''simple docstring''' from typing import TYPE_CHECKING from ...utils import ( OptionalDependencyNotAvailable, _LazyModule, is_sentencepiece_available, is_tokenizers_available, is_torch_available, ) a : Tuple = { 'configuration_llama': ['LLAMA_PRETRAINED_CONFIG_ARCHIVE_MAP', 'LlamaConfig'], } try: if not is_sentencepiece_available(): raise OptionalDependencyNotAvailable() except OptionalDependencyNotAvailable: pass else: a : Optional[Any] = ['LlamaTokenizer'] try: if not is_tokenizers_available(): raise OptionalDependencyNotAvailable() except OptionalDependencyNotAvailable: pass else: a : str = ['LlamaTokenizerFast'] try: if not is_torch_available(): raise OptionalDependencyNotAvailable() except OptionalDependencyNotAvailable: pass else: a : Optional[Any] = [ 'LlamaForCausalLM', 'LlamaModel', 'LlamaPreTrainedModel', 'LlamaForSequenceClassification', ] if TYPE_CHECKING: from .configuration_llama import LLAMA_PRETRAINED_CONFIG_ARCHIVE_MAP, LlamaConfig try: if not is_sentencepiece_available(): raise OptionalDependencyNotAvailable() except OptionalDependencyNotAvailable: pass else: from .tokenization_llama import LlamaTokenizer try: if not is_tokenizers_available(): raise OptionalDependencyNotAvailable() except OptionalDependencyNotAvailable: pass else: from .tokenization_llama_fast import LlamaTokenizerFast try: if not is_torch_available(): raise OptionalDependencyNotAvailable() except OptionalDependencyNotAvailable: pass else: from .modeling_llama import LlamaForCausalLM, LlamaForSequenceClassification, LlamaModel, LlamaPreTrainedModel else: import sys a : str = _LazyModule(__name__, globals()['__file__'], _import_structure, module_spec=__spec__)
56
0
import os # All paths are set with the intent you should run this script from the root of the repo with the command # python utils/check_doctest_list.py UpperCamelCase : Any = "." if __name__ == "__main__": UpperCamelCase : Optional[int] = os.path.join(REPO_PATH, "utils/documentation_tests.txt") UpperCamelCase : Optional[int] = [] UpperCamelCase : Optional[int] = [] with open(doctest_file_path) as fp: for line in fp: UpperCamelCase : Optional[int] = line.strip() UpperCamelCase : Tuple = os.path.join(REPO_PATH, line) if not (os.path.isfile(path) or os.path.isdir(path)): non_existent_paths.append(line) all_paths.append(path) if len(non_existent_paths) > 0: UpperCamelCase : Dict = "\n".join(non_existent_paths) raise ValueError(f'''`utils/documentation_tests.txt` contains non-existent paths:\n{non_existent_paths}''') if all_paths != sorted(all_paths): raise ValueError("Files in `utils/documentation_tests.txt` are not in alphabetical order.")
366
"""simple docstring""" import json import re from typing import TYPE_CHECKING, List, Optional, Tuple, Union import numpy as np from ...utils import is_tf_available, is_torch_available, logging if TYPE_CHECKING: if is_torch_available(): import torch if is_tf_available(): import tensorflow as tf from tokenizers import pre_tokenizers from ...tokenization_utils_base import BatchEncoding from ...tokenization_utils_fast import PreTrainedTokenizerFast from .tokenization_codegen import CodeGenTokenizer UpperCamelCase : List[Any] = logging.get_logger(__name__) UpperCamelCase : List[str] = {"vocab_file": "vocab.json", "merges_file": "merges.txt", "tokenizer_file": "tokenizer.json"} UpperCamelCase : List[Any] = { "vocab_file": { "Salesforce/codegen-350M-mono": "https://huggingface.co/Salesforce/codegen-350M-mono/resolve/main/vocab.json", }, "merges_file": { "Salesforce/codegen-350M-mono": "https://huggingface.co/Salesforce/codegen-350M-mono/resolve/main/merges.txt", }, "tokenizer_file": { "Salesforce/codegen-350M-mono": ( "https://huggingface.co/Salesforce/codegen-350M-mono/resolve/main/tokenizer.json" ), }, } UpperCamelCase : List[Any] = { "Salesforce/codegen-350M-mono": 2_0_4_8, } class __lowerCAmelCase ( __SCREAMING_SNAKE_CASE ): lowercase = VOCAB_FILES_NAMES lowercase = PRETRAINED_VOCAB_FILES_MAP lowercase = PRETRAINED_POSITIONAL_EMBEDDINGS_SIZES lowercase = ["input_ids", "attention_mask"] lowercase = CodeGenTokenizer def __init__( self , __UpperCAmelCase=None , __UpperCAmelCase=None , __UpperCAmelCase=None , __UpperCAmelCase="<|endoftext|>" , __UpperCAmelCase="<|endoftext|>" , __UpperCAmelCase="<|endoftext|>" , __UpperCAmelCase=False , **__UpperCAmelCase , ): '''simple docstring''' super().__init__( __UpperCAmelCase , __UpperCAmelCase , tokenizer_file=__UpperCAmelCase , unk_token=__UpperCAmelCase , bos_token=__UpperCAmelCase , eos_token=__UpperCAmelCase , add_prefix_space=__UpperCAmelCase , **__UpperCAmelCase , ) if kwargs.pop('add_bos_token' , __UpperCAmelCase ): __UpperCamelCase = kwargs.pop('name_or_path' , '' ) raise ValueError( 'Currenty GPT2\'s fast tokenizer does NOT support adding a BOS token.' 'Instead you should use GPT2\'s slow tokenizer class `CodeGenTokenizer` as follows: \n' F'`CodeGenTokenizer.from_pretrained(\'{model_id}\')`\nor\n' F'`AutoTokenizer.from_pretrained(\'{model_id}\', use_fast=False)`\n' 'This issue will be fixed soon, see: https://github.com/huggingface/tokenizers/pull/1005.' ' so that the fast tokenizer works correctly.' ) __UpperCamelCase = json.loads(self.backend_tokenizer.pre_tokenizer.__getstate__() ) if pre_tok_state.get('add_prefix_space' , __UpperCAmelCase ) != add_prefix_space: __UpperCamelCase = getattr(__UpperCAmelCase , pre_tok_state.pop('type' ) ) __UpperCamelCase = add_prefix_space __UpperCamelCase = pre_tok_class(**__UpperCAmelCase ) __UpperCamelCase = add_prefix_space def UpperCAmelCase ( self , *__UpperCAmelCase , **__UpperCAmelCase ): '''simple docstring''' __UpperCamelCase = kwargs.get('is_split_into_words' , __UpperCAmelCase ) 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(*__UpperCAmelCase , **__UpperCAmelCase ) def UpperCAmelCase ( self , *__UpperCAmelCase , **__UpperCAmelCase ): '''simple docstring''' __UpperCamelCase = kwargs.get('is_split_into_words' , __UpperCAmelCase ) 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(*__UpperCAmelCase , **__UpperCAmelCase ) def UpperCAmelCase ( self , __UpperCAmelCase , __UpperCAmelCase = None ): '''simple docstring''' __UpperCamelCase = self._tokenizer.model.save(__UpperCAmelCase , name=__UpperCAmelCase ) return tuple(__UpperCAmelCase ) def UpperCAmelCase ( self , __UpperCAmelCase , __UpperCAmelCase = False , __UpperCAmelCase = None , __UpperCAmelCase = None , **__UpperCAmelCase , ): '''simple docstring''' __UpperCamelCase = super().decode( token_ids=__UpperCAmelCase , skip_special_tokens=__UpperCAmelCase , clean_up_tokenization_spaces=__UpperCAmelCase , **__UpperCAmelCase , ) if truncate_before_pattern is not None and len(__UpperCAmelCase ) > 0: __UpperCamelCase = self.truncate(__UpperCAmelCase , __UpperCAmelCase ) return decoded_text def UpperCAmelCase ( self , __UpperCAmelCase , __UpperCAmelCase ): '''simple docstring''' def find_re(__UpperCAmelCase , __UpperCAmelCase , __UpperCAmelCase ): __UpperCamelCase = pattern.search(__UpperCAmelCase , __UpperCAmelCase ) return m.start() if m else -1 __UpperCamelCase = [re.compile(__UpperCAmelCase , re.MULTILINE ) for pattern in truncate_before_pattern] __UpperCamelCase = list(re.finditer('^print' , __UpperCAmelCase , re.MULTILINE ) ) if len(__UpperCAmelCase ) > 1: __UpperCamelCase = completion[: prints[1].start()] __UpperCamelCase = list(re.finditer('^def' , __UpperCAmelCase , re.MULTILINE ) ) if len(__UpperCAmelCase ) > 1: __UpperCamelCase = completion[: defs[1].start()] __UpperCamelCase = 0 __UpperCamelCase = [ pos for pos in [find_re(__UpperCAmelCase , __UpperCAmelCase , __UpperCAmelCase ) for terminal in terminals] if pos != -1 ] if len(__UpperCAmelCase ) > 0: return completion[: min(__UpperCAmelCase )] else: return completion
263
0
# Copyright 2021 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 from accelerate.test_utils import execute_subprocess_async def SCREAMING_SNAKE_CASE ( _UpperCAmelCase=None ) -> Tuple: if subparsers is not None: lowerCamelCase__ : Any = subparsers.add_parser('test' ) else: lowerCamelCase__ : int = argparse.ArgumentParser('Accelerate test command' ) parser.add_argument( '--config_file' , default=_UpperCAmelCase , help=( 'The path to use to store the config file. Will default to a file named default_config.yaml in the cache ' 'location, which is the content of the environment `HF_HOME` suffixed with \'accelerate\', or if you don\'t have ' 'such an environment variable, your cache directory (\'~/.cache\' or the content of `XDG_CACHE_HOME`) suffixed ' 'with \'huggingface\'.' ) , ) if subparsers is not None: parser.set_defaults(func=_UpperCAmelCase ) return parser def SCREAMING_SNAKE_CASE ( _UpperCAmelCase ) -> Union[str, Any]: lowerCamelCase__ : Tuple = os.path.sep.join(__file__.split(os.path.sep )[:-2] + ['test_utils', 'scripts', 'test_script.py'] ) if args.config_file is None: lowerCamelCase__ : List[str] = script_name else: lowerCamelCase__ : List[Any] = F"""--config_file={args.config_file} {script_name}""" lowerCamelCase__ : str = ['accelerate-launch'] + test_args.split() lowerCamelCase__ : Dict = execute_subprocess_async(_UpperCAmelCase , env=os.environ.copy() ) if result.returncode == 0: print('Test is a success! You are ready for your distributed training!' ) def SCREAMING_SNAKE_CASE ( ) -> Any: lowerCamelCase__ : Any = test_command_parser() lowerCamelCase__ : List[Any] = parser.parse_args() test_command(_UpperCAmelCase ) if __name__ == "__main__": main()
50
import tempfile import unittest from transformers import TaConfig, is_torch_available from transformers.testing_utils import ( require_sentencepiece, require_tokenizers, require_torch, slow, torch_device, ) from ...generation.test_utils import GenerationTesterMixin from ...test_modeling_common import ModelTesterMixin, ids_tensor from ...test_pipeline_mixin import PipelineTesterMixin if is_torch_available(): import torch from transformers import AutoTokenizer, UMTaForConditionalGeneration, UMTaForQuestionAnswering, UMTaModel class lowerCAmelCase : def __init__( self : Tuple , UpperCAmelCase : Dict , UpperCAmelCase : Union[str, Any]=99 , UpperCAmelCase : str=13 , UpperCAmelCase : List[str]=7 , UpperCAmelCase : str=9 , UpperCAmelCase : Optional[int]=True , UpperCAmelCase : str=True , UpperCAmelCase : Any=False , UpperCAmelCase : Union[str, Any]=32 , UpperCAmelCase : List[str]=5 , UpperCAmelCase : Tuple=4 , UpperCAmelCase : Union[str, Any]=37 , UpperCAmelCase : int=8 , UpperCAmelCase : List[str]=0.1 , UpperCAmelCase : Any=0.0_0_2 , UpperCAmelCase : Optional[Any]=1 , UpperCAmelCase : List[Any]=0 , UpperCAmelCase : Union[str, Any]=0 , UpperCAmelCase : Tuple=None , UpperCAmelCase : Optional[Any]=None , ) -> Union[str, Any]: lowerCamelCase__ : int = parent lowerCamelCase__ : Any = batch_size lowerCamelCase__ : Optional[int] = encoder_seq_length lowerCamelCase__ : int = decoder_seq_length # For common tests lowerCamelCase__ : List[str] = self.decoder_seq_length lowerCamelCase__ : Optional[int] = is_training lowerCamelCase__ : List[Any] = use_attention_mask lowerCamelCase__ : Optional[Any] = use_labels lowerCamelCase__ : Union[str, Any] = vocab_size lowerCamelCase__ : Union[str, Any] = hidden_size lowerCamelCase__ : Optional[Any] = num_hidden_layers lowerCamelCase__ : Any = num_attention_heads lowerCamelCase__ : str = d_ff lowerCamelCase__ : Optional[Any] = relative_attention_num_buckets lowerCamelCase__ : Any = dropout_rate lowerCamelCase__ : Any = initializer_factor lowerCamelCase__ : Union[str, Any] = eos_token_id lowerCamelCase__ : List[str] = pad_token_id lowerCamelCase__ : List[str] = decoder_start_token_id lowerCamelCase__ : List[Any] = None lowerCamelCase__ : Optional[Any] = decoder_layers def A_ ( self : List[Any] ) -> int: return TaConfig.from_pretrained('google/umt5-base' ) def A_ ( self : List[Any] , UpperCAmelCase : str , UpperCAmelCase : Optional[int] , UpperCAmelCase : Union[str, Any] , UpperCAmelCase : Tuple=None , UpperCAmelCase : List[str]=None , UpperCAmelCase : str=None , UpperCAmelCase : Optional[Any]=None , UpperCAmelCase : Optional[Any]=None , ) -> List[str]: if attention_mask is None: lowerCamelCase__ : Optional[Any] = input_ids.ne(config.pad_token_id ) if decoder_attention_mask is None: lowerCamelCase__ : Optional[Any] = decoder_input_ids.ne(config.pad_token_id ) if head_mask is None: lowerCamelCase__ : int = torch.ones(config.num_hidden_layers , config.num_attention_heads , device=UpperCAmelCase ) if decoder_head_mask is None: lowerCamelCase__ : Dict = torch.ones(config.num_decoder_layers , config.num_attention_heads , device=UpperCAmelCase ) if cross_attn_head_mask is None: lowerCamelCase__ : Dict = torch.ones( config.num_decoder_layers , config.num_attention_heads , device=UpperCAmelCase ) 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, } def A_ ( self : str ) -> List[str]: lowerCamelCase__ : Any = ids_tensor([self.batch_size, self.encoder_seq_length] , self.vocab_size ) lowerCamelCase__ : Optional[Any] = ids_tensor([self.batch_size, self.decoder_seq_length] , self.vocab_size ) # we need to clamp the input ids here to avoid having pad token in between # this is because for NllbMoe the position_ids are prepared such that # all pad tokens have pos id = 2 and rest are between 2..seq_length # and the seq_length here is seq_length - num_pad_tokens # but when using past, there is no way of knowing if the past input ids had # pad tokens in them, which results in incorrect seq_lenth and which in turn results in # position_ids being off by num_pad_tokens in past input lowerCamelCase__ : List[str] = input_ids.clamp(self.pad_token_id + 1 ) lowerCamelCase__ : Union[str, Any] = decoder_input_ids.clamp(self.pad_token_id + 1 ) lowerCamelCase__ : Dict = self.get_config() lowerCamelCase__ : Tuple = config.num_attention_heads lowerCamelCase__ : Any = self.prepare_inputs_dict(UpperCAmelCase , UpperCAmelCase , UpperCAmelCase ) return config, input_dict def A_ ( self : Tuple ) -> Union[str, Any]: lowerCamelCase__ , lowerCamelCase__ : Dict = self.prepare_config_and_inputs() return config, inputs_dict def A_ ( self : Optional[int] ) -> List[str]: return TaConfig( vocab_size=166 , d_model=self.hidden_size , d_ff=self.d_ff , d_kv=self.hidden_size // self.num_attention_heads , num_layers=self.num_hidden_layers , num_decoder_layers=self.decoder_layers , num_heads=self.num_attention_heads , relative_attention_num_buckets=self.relative_attention_num_buckets , dropout_rate=self.dropout_rate , initializer_factor=self.initializer_factor , eos_token_id=self.eos_token_id , bos_token_id=self.pad_token_id , pad_token_id=self.pad_token_id , decoder_start_token_id=self.decoder_start_token_id , ) def A_ ( self : Union[str, Any] ) -> Dict: return TaConfig( vocab_size=self.vocab_size , d_model=self.hidden_size , d_ff=self.d_ff , d_kv=self.hidden_size // self.num_attention_heads , num_layers=self.num_hidden_layers , num_decoder_layers=self.decoder_layers , num_heads=self.num_attention_heads , relative_attention_num_buckets=self.relative_attention_num_buckets , dropout_rate=self.dropout_rate , initializer_factor=self.initializer_factor , eos_token_id=self.eos_token_id , bos_token_id=self.pad_token_id , pad_token_id=self.pad_token_id , decoder_start_token_id=self.decoder_start_token_id , ) def A_ ( self : Any , UpperCAmelCase : Union[str, Any] , UpperCAmelCase : List[str] , UpperCAmelCase : int , UpperCAmelCase : str , UpperCAmelCase : Any , UpperCAmelCase : Dict , ) -> str: lowerCamelCase__ : Dict = UMTaModel(config=UpperCAmelCase ) model.to(UpperCAmelCase ) model.eval() lowerCamelCase__ : Optional[int] = model( input_ids=UpperCAmelCase , decoder_input_ids=UpperCAmelCase , attention_mask=UpperCAmelCase , decoder_attention_mask=UpperCAmelCase , ) lowerCamelCase__ : Any = model(input_ids=UpperCAmelCase , decoder_input_ids=UpperCAmelCase ) lowerCamelCase__ : Dict = result.last_hidden_state lowerCamelCase__ : Any = result.past_key_values lowerCamelCase__ : List[Any] = result.encoder_last_hidden_state self.parent.assertEqual(encoder_output.size() , (self.batch_size, self.encoder_seq_length, self.hidden_size) ) self.parent.assertEqual(decoder_output.size() , (self.batch_size, self.decoder_seq_length, self.hidden_size) ) # There should be `num_layers` key value embeddings stored in decoder_past self.parent.assertEqual(len(UpperCAmelCase ) , config.num_layers ) # There should be a self attn key, a self attn value, a cross attn key and a cross attn value stored in each decoder_past tuple self.parent.assertEqual(len(decoder_past[0] ) , 4 ) def A_ ( self : Optional[int] , UpperCAmelCase : str , UpperCAmelCase : Union[str, Any] , UpperCAmelCase : List[Any] , UpperCAmelCase : int , UpperCAmelCase : List[str] , UpperCAmelCase : Optional[Any] , ) -> Optional[int]: lowerCamelCase__ : List[Any] = UMTaModel(config=UpperCAmelCase ).get_decoder().to(UpperCAmelCase ).eval() # first forward pass lowerCamelCase__ : Tuple = model(UpperCAmelCase , use_cache=UpperCAmelCase ) lowerCamelCase__ : List[Any] = model(UpperCAmelCase ) lowerCamelCase__ : int = model(UpperCAmelCase , use_cache=UpperCAmelCase ) self.parent.assertTrue(len(UpperCAmelCase ) == len(UpperCAmelCase ) ) self.parent.assertTrue(len(UpperCAmelCase ) == len(UpperCAmelCase ) + 1 ) lowerCamelCase__ , lowerCamelCase__ : Dict = outputs.to_tuple() # create hypothetical next token and extent to next_input_ids lowerCamelCase__ : Optional[int] = ids_tensor((self.batch_size, 1) , config.vocab_size ) # append to next input_ids and lowerCamelCase__ : List[Any] = torch.cat([input_ids, next_tokens] , dim=-1 ) lowerCamelCase__ : List[str] = model(UpperCAmelCase )['last_hidden_state'] lowerCamelCase__ : str = model(UpperCAmelCase , past_key_values=UpperCAmelCase )['last_hidden_state'] # select random slice lowerCamelCase__ : List[str] = ids_tensor((1,) , output_from_past.shape[-1] ).item() lowerCamelCase__ : Tuple = output_from_no_past[:, -1, random_slice_idx].detach() lowerCamelCase__ : List[str] = output_from_past[:, 0, random_slice_idx].detach() # test that outputs are equal for slice self.parent.assertTrue(torch.allclose(UpperCAmelCase , UpperCAmelCase , atol=1e-3 ) ) def A_ ( self : Optional[int] , UpperCAmelCase : Optional[Any] , UpperCAmelCase : List[str] , ) -> Tuple: lowerCamelCase__ : Union[str, Any] = UMTaModel(config=UpperCAmelCase ).to(UpperCAmelCase ).half().eval() lowerCamelCase__ : Optional[int] = model(**UpperCAmelCase )['last_hidden_state'] self.parent.assertFalse(torch.isnan(UpperCAmelCase ).any().item() ) @require_torch class lowerCAmelCase ( __UpperCamelCase, __UpperCamelCase, __UpperCamelCase, unittest.TestCase ): UpperCAmelCase__ = ( (UMTaModel, UMTaForConditionalGeneration, UMTaForQuestionAnswering) if is_torch_available() else () ) UpperCAmelCase__ = (UMTaForConditionalGeneration,) if is_torch_available() else () UpperCAmelCase__ = ( { """conversational""": UMTaForConditionalGeneration, """feature-extraction""": UMTaModel, """summarization""": UMTaForConditionalGeneration, """text2text-generation""": UMTaForConditionalGeneration, """translation""": UMTaForConditionalGeneration, """question-answering""": UMTaForQuestionAnswering, } if is_torch_available() else {} ) UpperCAmelCase__ = True UpperCAmelCase__ = False UpperCAmelCase__ = False UpperCAmelCase__ = True UpperCAmelCase__ = True # The small UMT5 model needs higher percentages for CPU/MP tests UpperCAmelCase__ = [0.8, 0.9] def A_ ( self : Union[str, Any] ) -> List[Any]: lowerCamelCase__ : Union[str, Any] = UMTaModelTester(self ) @unittest.skip('Test has a segmentation fault on torch 1.8.0' ) def A_ ( self : Tuple ) -> int: lowerCamelCase__ : str = self.model_tester.prepare_config_and_inputs() lowerCamelCase__ : Tuple = UMTaModel(config_and_inputs[0] ).to(UpperCAmelCase ) with tempfile.TemporaryDirectory() as tmpdirname: torch.onnx.export( UpperCAmelCase , (config_and_inputs[1], config_and_inputs[3], config_and_inputs[2]) , F"""{tmpdirname}/t5_test.onnx""" , export_params=UpperCAmelCase , opset_version=9 , input_names=['input_ids', 'decoder_input_ids'] , ) @unittest.skipIf(torch_device == 'cpu' , 'Cant do half precision' ) def A_ ( self : Tuple ) -> Optional[Any]: lowerCamelCase__ : Any = self.model_tester.prepare_config_and_inputs() self.model_tester.create_and_check_model_fpaa_forward(*UpperCAmelCase ) def A_ ( self : List[Any] ) -> str: lowerCamelCase__ : int = ['encoder_attentions', 'decoder_attentions', 'cross_attentions'] lowerCamelCase__ : str = self.model_tester.prepare_config_and_inputs() lowerCamelCase__ : Any = config_and_inputs[0] lowerCamelCase__ : Any = UMTaForConditionalGeneration(UpperCAmelCase ).eval() model.to(UpperCAmelCase ) lowerCamelCase__ : Tuple = { 'head_mask': torch.zeros(config.num_layers , config.num_heads , device=UpperCAmelCase ), 'decoder_head_mask': torch.zeros(config.num_decoder_layers , config.num_heads , device=UpperCAmelCase ), 'cross_attn_head_mask': torch.zeros(config.num_decoder_layers , config.num_heads , device=UpperCAmelCase ), } for attn_name, (name, mask) in zip(UpperCAmelCase , head_masking.items() ): lowerCamelCase__ : Union[str, Any] = {name: mask} # Explicitly pass decoder_head_mask as it is required from T5 model when head_mask specified if name == "head_mask": lowerCamelCase__ : Union[str, Any] = torch.ones( config.num_decoder_layers , config.num_heads , device=UpperCAmelCase ) lowerCamelCase__ : Tuple = model.generate( config_and_inputs[1]['input_ids'] , num_beams=1 , max_length=3 , output_attentions=UpperCAmelCase , return_dict_in_generate=UpperCAmelCase , **UpperCAmelCase , ) # We check the state of decoder_attentions and cross_attentions just from the last step lowerCamelCase__ : Union[str, Any] = out[attn_name] if attn_name == attention_names[0] else out[attn_name][-1] self.assertEqual(sum([w.sum().item() for w in attn_weights] ) , 0.0 ) @unittest.skip('Does not work on the tiny model as we keep hitting edge cases.' ) def A_ ( self : Optional[Any] ) -> Optional[Any]: pass @require_torch @require_sentencepiece @require_tokenizers class lowerCAmelCase ( unittest.TestCase ): @slow @unittest.skip( 'Unless we stop stripping left and right by default for all special tokens, the expected ids obtained here will not match the original ones. Wait for https://github.com/huggingface/transformers/pull/23909 to be merged' ) def A_ ( self : Any ) -> int: lowerCamelCase__ : Optional[Any] = UMTaForConditionalGeneration.from_pretrained('google/umt5-small' , return_dict=UpperCAmelCase ).to(UpperCAmelCase ) lowerCamelCase__ : List[str] = AutoTokenizer.from_pretrained('google/umt5-small' , use_fast=UpperCAmelCase , legacy=UpperCAmelCase ) lowerCamelCase__ : Dict = [ 'Bonjour monsieur <extra_id_0> bien <extra_id_1>.', 'No se como puedo <extra_id_0>.', 'This is the reason why we <extra_id_0> them.', 'The <extra_id_0> walks in <extra_id_1>, seats', 'A <extra_id_0> walks into a bar and orders a <extra_id_1> with <extra_id_2> pinch of <extra_id_3>.', ] lowerCamelCase__ : Tuple = tokenizer(UpperCAmelCase , return_tensors='pt' , padding=UpperCAmelCase ).input_ids # fmt: off lowerCamelCase__ : Any = torch.tensor( [ [ 38530, 210703, 256299, 1410, 256298, 274, 1, 0,0, 0, 0, 0, 0, 0, 0, 0,0, 0], [ 826, 321, 671, 25922, 256299, 274, 1, 0,0, 0, 0, 0, 0, 0, 0, 0,0, 0], [ 1460, 339, 312, 19014, 10620, 758, 256299, 2355,274, 1, 0, 0, 0, 0, 0, 0,0, 0], [ 517, 256299, 14869, 281, 301, 256298, 275, 119983,1, 0, 0, 0, 0, 0, 0, 0,0, 0], [ 320, 256299, 14869, 281, 2234, 289, 2275, 333,61391, 289, 256298, 543, 256297, 168714, 329, 256296,274, 1], ] ) # fmt: on torch.testing.assert_allclose(UpperCAmelCase , UpperCAmelCase ) lowerCamelCase__ : Optional[int] = model.generate(input_ids.to(UpperCAmelCase ) ) lowerCamelCase__ : List[Any] = [ '<pad><extra_id_0> et<extra_id_1> [eod] <extra_id_2><extra_id_55>.. [eod] 💐 💐 💐 💐 💐 💐 💐 💐 💐 💐 💐 <extra_id_56>ajšietosto<extra_id_56>lleux<extra_id_19><extra_id_6>ajšie</s>', '<pad><extra_id_0>.<extra_id_1>.,<0x0A>...spech <0x0A><extra_id_20> <extra_id_21></s><pad><pad><pad><pad><pad><pad><pad><pad><pad><pad><pad><pad><pad><pad><pad><pad><pad><pad><pad><pad><pad><pad><pad><pad><pad><pad><pad><pad><pad><pad><pad>', '<pad><extra_id_0> are not going to be a part of the world. We are not going to be a part of<extra_id_1> and<extra_id_2><0x0A><extra_id_48>.<extra_id_48></s><pad><pad><pad><pad><pad><pad><pad><pad><pad><pad><pad><pad><pad><pad><pad><pad>', '<pad><extra_id_0> door<extra_id_1>, the door<extra_id_2> 피해[/</s><pad><pad><pad><pad><pad><pad><pad><pad><pad><pad><pad><pad><pad><pad><pad><pad><pad><pad><pad><pad><pad><pad><pad><pad><pad><pad><pad><pad><pad><pad><pad><pad><pad><pad><pad>', '<pad><extra_id_0>nyone who<extra_id_1> drink<extra_id_2> a<extra_id_3> alcohol<extra_id_4> A<extra_id_5> A. This<extra_id_6> I<extra_id_7><extra_id_52><extra_id_53></s><pad><pad><pad><pad><pad><pad><pad><pad><pad><pad><pad><pad><pad><pad><pad><pad><pad><pad><pad><pad><pad><pad><pad><pad>', ] lowerCamelCase__ : Union[str, Any] = tokenizer.batch_decode(UpperCAmelCase ) self.assertEqual(UpperCAmelCase , UpperCAmelCase )
50
1
'''simple docstring''' def a_ ( _UpperCAmelCase : int ,_UpperCAmelCase : int ) -> int: return x if y == 0 else greatest_common_divisor(_UpperCAmelCase ,x % y ) def a_ ( _UpperCAmelCase : int ,_UpperCAmelCase : int ) -> int: return (x * y) // greatest_common_divisor(_UpperCAmelCase ,_UpperCAmelCase ) def a_ ( _UpperCAmelCase : int = 20 ) -> int: __snake_case : Optional[Any] = 1 for i in range(1 ,n + 1 ): __snake_case : Optional[int] = lcm(_UpperCAmelCase ,_UpperCAmelCase ) return g if __name__ == "__main__": print(F"""{solution() = }""")
0
'''simple docstring''' import math def a_ ( _UpperCAmelCase : int ) -> list: __snake_case : Optional[Any] = [True] * n __snake_case : Optional[int] = False __snake_case : Dict = False __snake_case : List[Any] = True for i in range(3 ,int(n**0.5 + 1 ) ,2 ): __snake_case : Optional[int] = i * 2 while index < n: __snake_case : Union[str, Any] = False __snake_case : int = index + i __snake_case : Dict = [2] for i in range(3 ,_UpperCAmelCase ,2 ): if is_prime[i]: primes.append(_UpperCAmelCase ) return primes def a_ ( _UpperCAmelCase : int = 99_99_66_66_33_33 ) -> int: __snake_case : List[Any] = math.floor(math.sqrt(_UpperCAmelCase ) ) + 1_00 __snake_case : Tuple = prime_sieve(_UpperCAmelCase ) __snake_case : List[Any] = 0 __snake_case : List[Any] = 0 __snake_case : Optional[int] = primes[prime_index] while (last_prime**2) <= limit: __snake_case : Optional[int] = primes[prime_index + 1] __snake_case : Union[str, Any] = last_prime**2 __snake_case : Dict = next_prime**2 # Get numbers divisible by lps(current) __snake_case : Optional[Any] = lower_bound + last_prime while upper_bound > current <= limit: matches_sum += current current += last_prime # Reset the upper_bound while (upper_bound - next_prime) > limit: upper_bound -= next_prime # Add the numbers divisible by ups(current) __snake_case : Optional[Any] = upper_bound - next_prime while current > lower_bound: matches_sum += current current -= next_prime # Remove the numbers divisible by both ups and lps __snake_case : List[str] = 0 while upper_bound > current <= limit: if current <= lower_bound: # Increment the current number current += last_prime * next_prime continue if current > limit: break # Remove twice since it was added by both ups and lps matches_sum -= current * 2 # Increment the current number current += last_prime * next_prime # Setup for next pair __snake_case : Dict = next_prime prime_index += 1 return matches_sum if __name__ == "__main__": print(solution())
0
1
import argparse import os import torch from diffusers import ( CMStochasticIterativeScheduler, ConsistencyModelPipeline, UNetaDModel, ) __A : int = { '''sample_size''': 32, '''in_channels''': 3, '''out_channels''': 3, '''layers_per_block''': 2, '''num_class_embeds''': 1000, '''block_out_channels''': [32, 64], '''attention_head_dim''': 8, '''down_block_types''': [ '''ResnetDownsampleBlock2D''', '''AttnDownBlock2D''', ], '''up_block_types''': [ '''AttnUpBlock2D''', '''ResnetUpsampleBlock2D''', ], '''resnet_time_scale_shift''': '''scale_shift''', '''upsample_type''': '''resnet''', '''downsample_type''': '''resnet''', } __A : Union[str, Any] = { '''sample_size''': 64, '''in_channels''': 3, '''out_channels''': 3, '''layers_per_block''': 3, '''num_class_embeds''': 1000, '''block_out_channels''': [192, 192 * 2, 192 * 3, 192 * 4], '''attention_head_dim''': 64, '''down_block_types''': [ '''ResnetDownsampleBlock2D''', '''AttnDownBlock2D''', '''AttnDownBlock2D''', '''AttnDownBlock2D''', ], '''up_block_types''': [ '''AttnUpBlock2D''', '''AttnUpBlock2D''', '''AttnUpBlock2D''', '''ResnetUpsampleBlock2D''', ], '''resnet_time_scale_shift''': '''scale_shift''', '''upsample_type''': '''resnet''', '''downsample_type''': '''resnet''', } __A : Any = { '''sample_size''': 256, '''in_channels''': 3, '''out_channels''': 3, '''layers_per_block''': 2, '''num_class_embeds''': None, '''block_out_channels''': [256, 256, 256 * 2, 256 * 2, 256 * 4, 256 * 4], '''attention_head_dim''': 64, '''down_block_types''': [ '''ResnetDownsampleBlock2D''', '''ResnetDownsampleBlock2D''', '''ResnetDownsampleBlock2D''', '''AttnDownBlock2D''', '''AttnDownBlock2D''', '''AttnDownBlock2D''', ], '''up_block_types''': [ '''AttnUpBlock2D''', '''AttnUpBlock2D''', '''AttnUpBlock2D''', '''ResnetUpsampleBlock2D''', '''ResnetUpsampleBlock2D''', '''ResnetUpsampleBlock2D''', ], '''resnet_time_scale_shift''': '''default''', '''upsample_type''': '''resnet''', '''downsample_type''': '''resnet''', } __A : List[Any] = { '''num_train_timesteps''': 40, '''sigma_min''': 0.0_02, '''sigma_max''': 80.0, } __A : Union[str, Any] = { '''num_train_timesteps''': 201, '''sigma_min''': 0.0_02, '''sigma_max''': 80.0, } __A : int = { '''num_train_timesteps''': 151, '''sigma_min''': 0.0_02, '''sigma_max''': 80.0, } def SCREAMING_SNAKE_CASE__ ( _UpperCAmelCase ) -> Dict: '''simple docstring''' if isinstance(_UpperCAmelCase, _UpperCAmelCase ): return v if v.lower() in ("yes", "true", "t", "y", "1"): return True elif v.lower() in ("no", "false", "f", "n", "0"): return False else: raise argparse.ArgumentTypeError('boolean value expected' ) def SCREAMING_SNAKE_CASE__ ( _UpperCAmelCase, _UpperCAmelCase, _UpperCAmelCase, _UpperCAmelCase, _UpperCAmelCase=False ) -> List[str]: '''simple docstring''' lowerCAmelCase : int = checkpoint[f"{old_prefix}.in_layers.0.weight"] lowerCAmelCase : List[str] = checkpoint[f"{old_prefix}.in_layers.0.bias"] lowerCAmelCase : Optional[Any] = checkpoint[f"{old_prefix}.in_layers.2.weight"] lowerCAmelCase : str = checkpoint[f"{old_prefix}.in_layers.2.bias"] lowerCAmelCase : int = checkpoint[f"{old_prefix}.emb_layers.1.weight"] lowerCAmelCase : Tuple = checkpoint[f"{old_prefix}.emb_layers.1.bias"] lowerCAmelCase : Any = checkpoint[f"{old_prefix}.out_layers.0.weight"] lowerCAmelCase : List[Any] = checkpoint[f"{old_prefix}.out_layers.0.bias"] lowerCAmelCase : List[str] = checkpoint[f"{old_prefix}.out_layers.3.weight"] lowerCAmelCase : Optional[int] = checkpoint[f"{old_prefix}.out_layers.3.bias"] if has_skip: lowerCAmelCase : str = checkpoint[f"{old_prefix}.skip_connection.weight"] lowerCAmelCase : Dict = checkpoint[f"{old_prefix}.skip_connection.bias"] return new_checkpoint def SCREAMING_SNAKE_CASE__ ( _UpperCAmelCase, _UpperCAmelCase, _UpperCAmelCase, _UpperCAmelCase, _UpperCAmelCase=None ) -> Dict: '''simple docstring''' lowerCAmelCase , lowerCAmelCase , lowerCAmelCase : List[Any] = checkpoint[f"{old_prefix}.qkv.weight"].chunk(3, dim=0 ) lowerCAmelCase , lowerCAmelCase , lowerCAmelCase : str = checkpoint[f"{old_prefix}.qkv.bias"].chunk(3, dim=0 ) lowerCAmelCase : int = checkpoint[f"{old_prefix}.norm.weight"] lowerCAmelCase : List[Any] = checkpoint[f"{old_prefix}.norm.bias"] lowerCAmelCase : List[Any] = weight_q.squeeze(-1 ).squeeze(-1 ) lowerCAmelCase : Tuple = bias_q.squeeze(-1 ).squeeze(-1 ) lowerCAmelCase : Union[str, Any] = weight_k.squeeze(-1 ).squeeze(-1 ) lowerCAmelCase : Dict = bias_k.squeeze(-1 ).squeeze(-1 ) lowerCAmelCase : List[Any] = weight_v.squeeze(-1 ).squeeze(-1 ) lowerCAmelCase : str = bias_v.squeeze(-1 ).squeeze(-1 ) lowerCAmelCase : Tuple = ( checkpoint[f"{old_prefix}.proj_out.weight"].squeeze(-1 ).squeeze(-1 ) ) lowerCAmelCase : Optional[Any] = checkpoint[f"{old_prefix}.proj_out.bias"].squeeze(-1 ).squeeze(-1 ) return new_checkpoint def SCREAMING_SNAKE_CASE__ ( _UpperCAmelCase, _UpperCAmelCase ) -> int: '''simple docstring''' lowerCAmelCase : Union[str, Any] = torch.load(_UpperCAmelCase, map_location='cpu' ) lowerCAmelCase : Tuple = {} lowerCAmelCase : Any = checkpoint['time_embed.0.weight'] lowerCAmelCase : Any = checkpoint['time_embed.0.bias'] lowerCAmelCase : List[Any] = checkpoint['time_embed.2.weight'] lowerCAmelCase : Tuple = checkpoint['time_embed.2.bias'] if unet_config["num_class_embeds"] is not None: lowerCAmelCase : str = checkpoint['label_emb.weight'] lowerCAmelCase : Optional[Any] = checkpoint['input_blocks.0.0.weight'] lowerCAmelCase : Optional[Any] = checkpoint['input_blocks.0.0.bias'] lowerCAmelCase : int = unet_config['down_block_types'] lowerCAmelCase : str = unet_config['layers_per_block'] lowerCAmelCase : Union[str, Any] = unet_config['attention_head_dim'] lowerCAmelCase : Any = unet_config['block_out_channels'] lowerCAmelCase : Dict = 1 lowerCAmelCase : List[str] = channels_list[0] for i, layer_type in enumerate(_UpperCAmelCase ): lowerCAmelCase : List[str] = channels_list[i] lowerCAmelCase : Optional[Any] = current_channels != prev_channels if layer_type == "ResnetDownsampleBlock2D": for j in range(_UpperCAmelCase ): lowerCAmelCase : Optional[Any] = f"down_blocks.{i}.resnets.{j}" lowerCAmelCase : List[Any] = f"input_blocks.{current_layer}.0" lowerCAmelCase : int = True if j == 0 and downsample_block_has_skip else False lowerCAmelCase : List[Any] = convert_resnet(_UpperCAmelCase, _UpperCAmelCase, _UpperCAmelCase, _UpperCAmelCase, has_skip=_UpperCAmelCase ) current_layer += 1 elif layer_type == "AttnDownBlock2D": for j in range(_UpperCAmelCase ): lowerCAmelCase : List[Any] = f"down_blocks.{i}.resnets.{j}" lowerCAmelCase : Union[str, Any] = f"input_blocks.{current_layer}.0" lowerCAmelCase : Dict = True if j == 0 and downsample_block_has_skip else False lowerCAmelCase : List[str] = convert_resnet(_UpperCAmelCase, _UpperCAmelCase, _UpperCAmelCase, _UpperCAmelCase, has_skip=_UpperCAmelCase ) lowerCAmelCase : Optional[Any] = f"down_blocks.{i}.attentions.{j}" lowerCAmelCase : List[str] = f"input_blocks.{current_layer}.1" lowerCAmelCase : Any = convert_attention( _UpperCAmelCase, _UpperCAmelCase, _UpperCAmelCase, _UpperCAmelCase, _UpperCAmelCase ) current_layer += 1 if i != len(_UpperCAmelCase ) - 1: lowerCAmelCase : List[Any] = f"down_blocks.{i}.downsamplers.0" lowerCAmelCase : List[str] = f"input_blocks.{current_layer}.0" lowerCAmelCase : Optional[Any] = convert_resnet(_UpperCAmelCase, _UpperCAmelCase, _UpperCAmelCase, _UpperCAmelCase ) current_layer += 1 lowerCAmelCase : List[str] = current_channels # hardcoded the mid-block for now lowerCAmelCase : Tuple = 'mid_block.resnets.0' lowerCAmelCase : Dict = 'middle_block.0' lowerCAmelCase : Any = convert_resnet(_UpperCAmelCase, _UpperCAmelCase, _UpperCAmelCase, _UpperCAmelCase ) lowerCAmelCase : Union[str, Any] = 'mid_block.attentions.0' lowerCAmelCase : Optional[int] = 'middle_block.1' lowerCAmelCase : List[Any] = convert_attention(_UpperCAmelCase, _UpperCAmelCase, _UpperCAmelCase, _UpperCAmelCase, _UpperCAmelCase ) lowerCAmelCase : List[Any] = 'mid_block.resnets.1' lowerCAmelCase : List[Any] = 'middle_block.2' lowerCAmelCase : Tuple = convert_resnet(_UpperCAmelCase, _UpperCAmelCase, _UpperCAmelCase, _UpperCAmelCase ) lowerCAmelCase : Optional[Any] = 0 lowerCAmelCase : Optional[int] = unet_config['up_block_types'] for i, layer_type in enumerate(_UpperCAmelCase ): if layer_type == "ResnetUpsampleBlock2D": for j in range(layers_per_block + 1 ): lowerCAmelCase : Optional[Any] = f"up_blocks.{i}.resnets.{j}" lowerCAmelCase : Optional[int] = f"output_blocks.{current_layer}.0" lowerCAmelCase : Any = convert_resnet(_UpperCAmelCase, _UpperCAmelCase, _UpperCAmelCase, _UpperCAmelCase, has_skip=_UpperCAmelCase ) current_layer += 1 if i != len(_UpperCAmelCase ) - 1: lowerCAmelCase : str = f"up_blocks.{i}.upsamplers.0" lowerCAmelCase : Optional[Any] = f"output_blocks.{current_layer-1}.1" lowerCAmelCase : Dict = convert_resnet(_UpperCAmelCase, _UpperCAmelCase, _UpperCAmelCase, _UpperCAmelCase ) elif layer_type == "AttnUpBlock2D": for j in range(layers_per_block + 1 ): lowerCAmelCase : Any = f"up_blocks.{i}.resnets.{j}" lowerCAmelCase : Dict = f"output_blocks.{current_layer}.0" lowerCAmelCase : Optional[int] = convert_resnet(_UpperCAmelCase, _UpperCAmelCase, _UpperCAmelCase, _UpperCAmelCase, has_skip=_UpperCAmelCase ) lowerCAmelCase : Optional[int] = f"up_blocks.{i}.attentions.{j}" lowerCAmelCase : List[str] = f"output_blocks.{current_layer}.1" lowerCAmelCase : Optional[Any] = convert_attention( _UpperCAmelCase, _UpperCAmelCase, _UpperCAmelCase, _UpperCAmelCase, _UpperCAmelCase ) current_layer += 1 if i != len(_UpperCAmelCase ) - 1: lowerCAmelCase : Optional[Any] = f"up_blocks.{i}.upsamplers.0" lowerCAmelCase : Union[str, Any] = f"output_blocks.{current_layer-1}.2" lowerCAmelCase : List[str] = convert_resnet(_UpperCAmelCase, _UpperCAmelCase, _UpperCAmelCase, _UpperCAmelCase ) lowerCAmelCase : List[str] = checkpoint['out.0.weight'] lowerCAmelCase : List[str] = checkpoint['out.0.bias'] lowerCAmelCase : List[Any] = checkpoint['out.2.weight'] lowerCAmelCase : str = checkpoint['out.2.bias'] return new_checkpoint if __name__ == "__main__": __A : Any = argparse.ArgumentParser() parser.add_argument('''--unet_path''', default=None, type=str, required=True, help='''Path to the unet.pt to convert.''') parser.add_argument( '''--dump_path''', default=None, type=str, required=True, help='''Path to output the converted UNet model.''' ) parser.add_argument('''--class_cond''', default=True, type=str, help='''Whether the model is class-conditional.''') __A : Optional[int] = parser.parse_args() __A : List[Any] = strabool(args.class_cond) __A : Dict = os.path.basename(args.unet_path) print(F'Checkpoint: {ckpt_name}') # Get U-Net config if "imagenet64" in ckpt_name: __A : Dict = IMAGENET_64_UNET_CONFIG elif "256" in ckpt_name and (("bedroom" in ckpt_name) or ("cat" in ckpt_name)): __A : Optional[int] = LSUN_256_UNET_CONFIG elif "test" in ckpt_name: __A : Any = TEST_UNET_CONFIG else: raise ValueError(F'Checkpoint type {ckpt_name} is not currently supported.') if not args.class_cond: __A : List[str] = None __A : Any = con_pt_to_diffuser(args.unet_path, unet_config) __A : List[str] = UNetaDModel(**unet_config) image_unet.load_state_dict(converted_unet_ckpt) # Get scheduler config if "cd" in ckpt_name or "test" in ckpt_name: __A : Optional[Any] = CD_SCHEDULER_CONFIG elif "ct" in ckpt_name and "imagenet64" in ckpt_name: __A : Optional[int] = CT_IMAGENET_64_SCHEDULER_CONFIG elif "ct" in ckpt_name and "256" in ckpt_name and (("bedroom" in ckpt_name) or ("cat" in ckpt_name)): __A : Union[str, Any] = CT_LSUN_256_SCHEDULER_CONFIG else: raise ValueError(F'Checkpoint type {ckpt_name} is not currently supported.') __A : Union[str, Any] = CMStochasticIterativeScheduler(**scheduler_config) __A : Any = ConsistencyModelPipeline(unet=image_unet, scheduler=cm_scheduler) consistency_model.save_pretrained(args.dump_path)
138
import itertools from dataclasses import dataclass from typing import List, Optional import pyarrow as pa import pyarrow.parquet as pq import datasets from datasets.table import table_cast __A : Any = datasets.utils.logging.get_logger(__name__) @dataclass class __A ( datasets.BuilderConfig ): lowerCAmelCase_ : int = 1_0000 lowerCAmelCase_ : Optional[List[str]] = None lowerCAmelCase_ : Optional[datasets.Features] = None class __A ( datasets.ArrowBasedBuilder ): lowerCAmelCase_ : List[Any] = ParquetConfig def lowercase__ ( self : Tuple ): return datasets.DatasetInfo(features=self.config.features ) def lowercase__ ( self : List[str] , UpperCAmelCase_ : Dict ): 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}" ) lowerCAmelCase : Any = dl_manager.download_and_extract(self.config.data_files ) if isinstance(UpperCAmelCase_ , (str, list, tuple) ): lowerCAmelCase : int = data_files if isinstance(UpperCAmelCase_ , UpperCAmelCase_ ): lowerCAmelCase : Optional[Any] = [files] # Use `dl_manager.iter_files` to skip hidden files in an extracted archive lowerCAmelCase : Dict = [dl_manager.iter_files(UpperCAmelCase_ ) for file in files] return [datasets.SplitGenerator(name=datasets.Split.TRAIN , gen_kwargs={'files': files} )] lowerCAmelCase : Any = [] for split_name, files in data_files.items(): if isinstance(UpperCAmelCase_ , UpperCAmelCase_ ): lowerCAmelCase : Tuple = [files] # Use `dl_manager.iter_files` to skip hidden files in an extracted archive lowerCAmelCase : Dict = [dl_manager.iter_files(UpperCAmelCase_ ) for file in files] # Infer features is they are stoed in the arrow schema if self.info.features is None: for file in itertools.chain.from_iterable(UpperCAmelCase_ ): with open(UpperCAmelCase_ , 'rb' ) as f: lowerCAmelCase : int = datasets.Features.from_arrow_schema(pq.read_schema(UpperCAmelCase_ ) ) break splits.append(datasets.SplitGenerator(name=UpperCAmelCase_ , gen_kwargs={'files': files} ) ) return splits def lowercase__ ( self : Any , UpperCAmelCase_ : pa.Table ): if self.info.features is not None: # more expensive cast to support nested features with keys in a different order # allows str <-> int/float or str to Audio for example lowerCAmelCase : Union[str, Any] = table_cast(UpperCAmelCase_ , self.info.features.arrow_schema ) return pa_table def lowercase__ ( self : Union[str, Any] , UpperCAmelCase_ : Optional[int] ): lowerCAmelCase : Optional[int] = self.info.features.arrow_schema if self.info.features is not None else None if self.info.features is not None and self.config.columns is not None: if sorted(field.name for field in schema ) != sorted(self.config.columns ): raise ValueError( f"Tried to load parquet data with columns '{self.config.columns}' with mismatching features '{self.info.features}'" ) for file_idx, file in enumerate(itertools.chain.from_iterable(UpperCAmelCase_ ) ): with open(UpperCAmelCase_ , 'rb' ) as f: lowerCAmelCase : Optional[Any] = pq.ParquetFile(UpperCAmelCase_ ) try: for batch_idx, record_batch in enumerate( parquet_file.iter_batches(batch_size=self.config.batch_size , columns=self.config.columns ) ): lowerCAmelCase : Dict = pa.Table.from_batches([record_batch] ) # 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 f"{file_idx}_{batch_idx}", self._cast_table(UpperCAmelCase_ ) except ValueError as e: logger.error(f"Failed to read file '{file}' with error {type(UpperCAmelCase_ )}: {e}" ) raise
138
1
import argparse import math import traceback import dateutil.parser as date_parser import requests def a__ ( A__ ): SCREAMING_SNAKE_CASE_ : Tuple = {} SCREAMING_SNAKE_CASE_ : Any = job['started_at'] SCREAMING_SNAKE_CASE_ : Any = job['completed_at'] SCREAMING_SNAKE_CASE_ : Optional[int] = date_parser.parse(A__ ) SCREAMING_SNAKE_CASE_ : Optional[int] = date_parser.parse(A__ ) SCREAMING_SNAKE_CASE_ : Union[str, Any] = round((end_datetime - start_datetime).total_seconds() / 60.0 ) SCREAMING_SNAKE_CASE_ : Dict = start SCREAMING_SNAKE_CASE_ : Optional[int] = end SCREAMING_SNAKE_CASE_ : Optional[Any] = duration_in_min return job_info def a__ ( A__, A__=None ): SCREAMING_SNAKE_CASE_ : List[str] = None if token is not None: SCREAMING_SNAKE_CASE_ : Union[str, Any] = {'Accept': 'application/vnd.github+json', 'Authorization': F'''Bearer {token}'''} SCREAMING_SNAKE_CASE_ : str = F'''https://api.github.com/repos/huggingface/transformers/actions/runs/{workflow_run_id}/jobs?per_page=100''' SCREAMING_SNAKE_CASE_ : Tuple = requests.get(A__, headers=A__ ).json() SCREAMING_SNAKE_CASE_ : int = {} try: job_time.update({job['name']: extract_time_from_single_job(A__ ) for job in result['jobs']} ) SCREAMING_SNAKE_CASE_ : int = math.ceil((result['total_count'] - 1_0_0) / 1_0_0 ) for i in range(A__ ): SCREAMING_SNAKE_CASE_ : Union[str, Any] = requests.get(url + F'''&page={i + 2}''', headers=A__ ).json() job_time.update({job['name']: extract_time_from_single_job(A__ ) for job in result['jobs']} ) return job_time except Exception: print(F'''Unknown error, could not fetch links:\n{traceback.format_exc()}''' ) return {} if __name__ == "__main__": lowerCAmelCase__ : Any =argparse.ArgumentParser() # Required parameters parser.add_argument('--workflow_run_id', type=str, required=True, help='A GitHub Actions workflow run id.') lowerCAmelCase__ : str =parser.parse_args() lowerCAmelCase__ : List[Any] =get_job_time(args.workflow_run_id) lowerCAmelCase__ : List[Any] =dict(sorted(job_time.items(), key=lambda item: item[1]["duration"], reverse=True)) for k, v in job_time.items(): print(F"""{k}: {v["duration"]}""")
358
from __future__ import annotations def a__ ( A__ ): return len(set(A__ ) ) == len(A__ ) if __name__ == "__main__": import doctest doctest.testmod()
162
0
import inspect import os import unittest import torch import accelerate from accelerate import debug_launcher from accelerate.test_utils import ( execute_subprocess_async, require_cpu, require_huggingface_suite, require_multi_gpu, require_single_gpu, ) from accelerate.utils import patch_environment @require_huggingface_suite class _UpperCamelCase ( unittest.TestCase ): """simple docstring""" def _SCREAMING_SNAKE_CASE ( self ) -> Union[str, Any]: '''simple docstring''' __lowercase = inspect.getfile(accelerate.test_utils ) __lowercase = os.path.sep.join( mod_file.split(os.path.sep )[:-1] + ['''scripts''', '''external_deps''', '''test_metrics.py'''] ) from accelerate.test_utils.scripts.external_deps import test_metrics # noqa: F401 __lowercase = test_metrics @require_cpu def _SCREAMING_SNAKE_CASE ( self ) -> Tuple: '''simple docstring''' debug_launcher(self.test_metrics.main , num_processes=1 ) @require_cpu def _SCREAMING_SNAKE_CASE ( self ) -> Tuple: '''simple docstring''' debug_launcher(self.test_metrics.main ) @require_single_gpu def _SCREAMING_SNAKE_CASE ( self ) -> List[str]: '''simple docstring''' self.test_metrics.main() @require_multi_gpu def _SCREAMING_SNAKE_CASE ( self ) -> Tuple: '''simple docstring''' print(F"Found {torch.cuda.device_count()} devices." ) __lowercase = ['''torchrun''', F"--nproc_per_node={torch.cuda.device_count()}", self.test_file_path] with patch_environment(omp_num_threads=1 ): execute_subprocess_async(lowerCAmelCase__ , env=os.environ.copy() )
210
import argparse import logging import os import datasets import tensorflow as tf from transformers import AutoTokenizer __a : List[Any] = logging.getLogger(__name__) def UpperCAmelCase ( ): """simple docstring""" __lowercase = argparse.ArgumentParser( description='''Prepare TFRecord shards from pre-tokenized samples of the wikitext dataset.''' ) parser.add_argument( '''--dataset_name''' , type=lowercase , default='''wikitext''' , help='''Name of the training. Explore datasets at: hf.co/datasets.''' , ) parser.add_argument( '''--dataset_config''' , type=lowercase , default='''wikitext-103-raw-v1''' , help='''Configuration name of the dataset.''' ) parser.add_argument( '''--tokenizer_name_or_path''' , type=lowercase , default='''sayakpaul/unigram-tokenizer-wikitext''' , help='''Tokenizer identifier. Can be a local filepath or a Hub identifier.''' , ) parser.add_argument( '''--shard_size''' , type=lowercase , default=1000 , help='''Number of entries to go in a single shard.''' , ) parser.add_argument('''--split''' , type=lowercase , default='''train''' , choices=['''train''', '''test''', '''validation'''] ) parser.add_argument( '''--limit''' , default=lowercase , type=lowercase , help='''Limit the number of shards (used for debugging).''' , ) parser.add_argument( '''--max_length''' , type=lowercase , default=512 , help='''Maximum sequence length. For training on TPUs, it helps to have a maximum''' ''' sequence length that is a multiple of 8.''' , ) parser.add_argument( '''--output_dir''' , default='''tf-tpu''' , type=lowercase , help='''Output directory where the TFRecord shards will be saved. If the''' ''' path is appended with `gs://` (\'gs://tf-tpu\', for example) then the TFRecord''' ''' shards will be directly saved to a Google Cloud Storage bucket.''' , ) __lowercase = parser.parse_args() return args def UpperCAmelCase ( lowercase ): """simple docstring""" def fn(lowercase ): return tokenizer(examples['''text'''] ) return fn def UpperCAmelCase ( lowercase ): """simple docstring""" __lowercase = [] for i in range(len(tokenized_data['''input_ids'''] ) ): __lowercase = { '''input_ids''': tf.train.Feature(intaa_list=tf.train.IntaaList(value=tokenized_data['''input_ids'''][i] ) ), '''attention_mask''': tf.train.Feature( intaa_list=tf.train.IntaaList(value=tokenized_data['''attention_mask'''][i] ) ), } __lowercase = tf.train.Features(feature=lowercase ) __lowercase = tf.train.Example(features=lowercase ) __lowercase = example.SerializeToString() records.append(lowercase ) return records def UpperCAmelCase ( lowercase ): """simple docstring""" __lowercase = datasets.load_dataset(args.dataset_name , args.dataset_config , split=args.split ) if args.limit is not None: __lowercase = min(len(lowercase ) , args.limit ) __lowercase = dataset.select(range(lowercase ) ) print(F"Limiting the dataset to {args.limit} entries." ) __lowercase = AutoTokenizer.from_pretrained(args.tokenizer_name_or_path ) # Handle output directory creation. # For serializing into a Google Cloud Storage Bucket, one needs to first # create a bucket. if "gs" not in args.output_dir: if not os.path.exists(args.output_dir ): os.makedirs(args.output_dir ) __lowercase = os.path.join(args.output_dir , args.split ) if not os.path.exists(lowercase ): os.makedirs(lowercase ) else: __lowercase = os.path.join(args.output_dir , args.split ) # Tokenize the whole dataset at once. __lowercase = tokenize_function(lowercase ) __lowercase = dataset.map(lowercase , batched=lowercase , num_proc=4 , remove_columns=['''text'''] ) # We need to concatenate all our texts together, and then split the result # into chunks of a fixed size, which we will call block_size. To do this, we # will use the map method again, with the option batched=True. When we use batched=True, # the function we pass to map() will be passed multiple inputs at once, allowing us # to group them into more or fewer examples than we had in the input. # This allows us to create our new fixed-length samples. The advantage of this # method is that we don't lose a whole lot of content from the dataset compared to the # case where we simply tokenize with a pre-defined max_length. def group_texts(lowercase ): # Concatenate all texts. __lowercase = {k: sum(examples[k] , [] ) for k in examples.keys()} __lowercase = len(concatenated_examples[list(examples.keys() )[0]] ) # We drop the small remainder, though you could add padding instead if the model supports it # In this, as in all things, we advise you to follow your heart 🫀 __lowercase = (total_length // args.max_length) * args.max_length # Split by chunks of max_len. __lowercase = { k: [t[i : i + args.max_length] for i in range(0 , lowercase , args.max_length )] for k, t in concatenated_examples.items() } return result __lowercase = dataset_tokenized.map(lowercase , batched=lowercase , batch_size=1000 , num_proc=4 ) __lowercase = 0 __lowercase = 0 for shard in range(0 , len(lowercase ) , args.shard_size ): __lowercase = grouped_dataset[shard : shard + args.shard_size] __lowercase = len(dataset_snapshot['''input_ids'''] ) __lowercase = os.path.join(lowercase , F"dataset-{shard_count}-{records_containing}.tfrecord" ) __lowercase = get_serialized_examples(lowercase ) with tf.io.TFRecordWriter(lowercase ) as out_file: for i in range(len(lowercase ) ): __lowercase = serialized_examples[i] out_file.write(lowercase ) print('''Wrote file {} containing {} records'''.format(lowercase , lowercase ) ) shard_count += 1 total_records += records_containing with open(F"split-{args.split}-records-count.txt" , '''w''' ) as f: print(F"Total {args.split} records: {total_records}" , file=lowercase ) if __name__ == "__main__": __a : Optional[Any] = parse_args() main(args)
210
1
def __lowerCamelCase ( UpperCAmelCase_ : str , UpperCAmelCase_ : str ): """simple docstring""" if not (isinstance(UpperCAmelCase_ , UpperCAmelCase_ ) and isinstance(UpperCAmelCase_ , UpperCAmelCase_ )): raise ValueError('''longest_common_substring() takes two strings for inputs''' ) a :Optional[int] = len(UpperCAmelCase_ ) a :str = len(UpperCAmelCase_ ) a :List[Any] = [[0] * (texta_length + 1) for _ in range(texta_length + 1 )] a :Dict = 0 a :Union[str, Any] = 0 for i in range(1 , texta_length + 1 ): for j in range(1 , texta_length + 1 ): if texta[i - 1] == texta[j - 1]: a :Union[str, Any] = 1 + dp[i - 1][j - 1] if dp[i][j] > ans_length: a :Dict = i a :Optional[Any] = dp[i][j] return texta[ans_index - ans_length : ans_index] if __name__ == "__main__": import doctest doctest.testmod()
352
import os import pytest from datasets import ( get_dataset_config_info, get_dataset_config_names, get_dataset_infos, get_dataset_split_names, inspect_dataset, inspect_metric, ) snake_case : Any = pytest.mark.integration @pytest.mark.parametrize('''path''' , ['''paws''', '''csv'''] ) def __lowerCamelCase ( UpperCAmelCase_ : Optional[int] , UpperCAmelCase_ : Optional[Any] ): """simple docstring""" inspect_dataset(UpperCAmelCase_ , UpperCAmelCase_ ) a :List[Any] = path + '''.py''' assert script_name in os.listdir(UpperCAmelCase_ ) assert "__pycache__" not in os.listdir(UpperCAmelCase_ ) @pytest.mark.filterwarnings('''ignore:inspect_metric is deprecated:FutureWarning''' ) @pytest.mark.filterwarnings('''ignore:metric_module_factory is deprecated:FutureWarning''' ) @pytest.mark.parametrize('''path''' , ['''accuracy'''] ) def __lowerCamelCase ( UpperCAmelCase_ : List[Any] , UpperCAmelCase_ : Any ): """simple docstring""" inspect_metric(UpperCAmelCase_ , UpperCAmelCase_ ) a :Dict = path + '''.py''' assert script_name in os.listdir(UpperCAmelCase_ ) assert "__pycache__" not in os.listdir(UpperCAmelCase_ ) @pytest.mark.parametrize( '''path, config_name, expected_splits''' , [ ('''squad''', '''plain_text''', ['''train''', '''validation''']), ('''dalle-mini/wit''', '''dalle-mini--wit''', ['''train''']), ('''paws''', '''labeled_final''', ['''train''', '''test''', '''validation''']), ] , ) def __lowerCamelCase ( UpperCAmelCase_ : Optional[int] , UpperCAmelCase_ : List[Any] , UpperCAmelCase_ : str ): """simple docstring""" a :List[str] = get_dataset_config_info(UpperCAmelCase_ , config_name=UpperCAmelCase_ ) assert info.config_name == config_name assert list(info.splits.keys() ) == expected_splits @pytest.mark.parametrize( '''path, config_name, expected_exception''' , [ ('''paws''', None, ValueError), ] , ) def __lowerCamelCase ( UpperCAmelCase_ : List[Any] , UpperCAmelCase_ : str , UpperCAmelCase_ : Tuple ): """simple docstring""" with pytest.raises(UpperCAmelCase_ ): get_dataset_config_info(UpperCAmelCase_ , config_name=UpperCAmelCase_ ) @pytest.mark.parametrize( '''path, expected''' , [ ('''squad''', '''plain_text'''), ('''acronym_identification''', '''default'''), ('''lhoestq/squad''', '''plain_text'''), ('''lhoestq/test''', '''default'''), ('''lhoestq/demo1''', '''lhoestq--demo1'''), ('''dalle-mini/wit''', '''dalle-mini--wit'''), ] , ) def __lowerCamelCase ( UpperCAmelCase_ : Tuple , UpperCAmelCase_ : int ): """simple docstring""" a :List[str] = get_dataset_config_names(UpperCAmelCase_ ) assert expected in config_names @pytest.mark.parametrize( '''path, expected_configs, expected_splits_in_first_config''' , [ ('''squad''', ['''plain_text'''], ['''train''', '''validation''']), ('''dalle-mini/wit''', ['''dalle-mini--wit'''], ['''train''']), ('''paws''', ['''labeled_final''', '''labeled_swap''', '''unlabeled_final'''], ['''train''', '''test''', '''validation''']), ] , ) def __lowerCamelCase ( UpperCAmelCase_ : Tuple , UpperCAmelCase_ : Tuple , UpperCAmelCase_ : Optional[int] ): """simple docstring""" a :Optional[int] = get_dataset_infos(UpperCAmelCase_ ) assert list(infos.keys() ) == expected_configs a :Union[str, Any] = expected_configs[0] assert expected_config in infos a :List[Any] = infos[expected_config] assert info.config_name == expected_config assert list(info.splits.keys() ) == expected_splits_in_first_config @pytest.mark.parametrize( '''path, expected_config, expected_splits''' , [ ('''squad''', '''plain_text''', ['''train''', '''validation''']), ('''dalle-mini/wit''', '''dalle-mini--wit''', ['''train''']), ('''paws''', '''labeled_final''', ['''train''', '''test''', '''validation''']), ] , ) def __lowerCamelCase ( UpperCAmelCase_ : Optional[int] , UpperCAmelCase_ : Any , UpperCAmelCase_ : Optional[int] ): """simple docstring""" a :Union[str, Any] = get_dataset_infos(UpperCAmelCase_ ) assert expected_config in infos a :int = infos[expected_config] assert info.config_name == expected_config assert list(info.splits.keys() ) == expected_splits @pytest.mark.parametrize( '''path, config_name, expected_exception''' , [ ('''paws''', None, ValueError), ] , ) def __lowerCamelCase ( UpperCAmelCase_ : Union[str, Any] , UpperCAmelCase_ : Optional[int] , UpperCAmelCase_ : str ): """simple docstring""" with pytest.raises(UpperCAmelCase_ ): get_dataset_split_names(UpperCAmelCase_ , config_name=UpperCAmelCase_ )
281
0
'''simple docstring''' import argparse import os import transformers from .convert_slow_tokenizer import SLOW_TO_FAST_CONVERTERS from .utils import logging logging.set_verbosity_info() lowerCamelCase : Optional[Any] = logging.get_logger(__name__) lowerCamelCase : str = {name: getattr(transformers, name + "Fast") for name in SLOW_TO_FAST_CONVERTERS} def _lowerCAmelCase ( _UpperCamelCase : Dict , _UpperCamelCase : str , _UpperCamelCase : Optional[Any] , _UpperCamelCase : str ) -> str: """simple docstring""" if tokenizer_name is not None and tokenizer_name not in TOKENIZER_CLASSES: raise ValueError(f"Unrecognized tokenizer name, should be one of {list(TOKENIZER_CLASSES.keys() )}." ) if tokenizer_name is None: _SCREAMING_SNAKE_CASE =TOKENIZER_CLASSES else: _SCREAMING_SNAKE_CASE ={tokenizer_name: getattr(__UpperCamelCase , tokenizer_name + 'Fast' )} logger.info(f"Loading tokenizer classes: {tokenizer_names}" ) for tokenizer_name in tokenizer_names: _SCREAMING_SNAKE_CASE =TOKENIZER_CLASSES[tokenizer_name] _SCREAMING_SNAKE_CASE =True if checkpoint_name is None: _SCREAMING_SNAKE_CASE =list(tokenizer_class.max_model_input_sizes.keys() ) else: _SCREAMING_SNAKE_CASE =[checkpoint_name] logger.info(f"For tokenizer {tokenizer_class.__class__.__name__} loading checkpoints: {checkpoint_names}" ) for checkpoint in checkpoint_names: logger.info(f"Loading {tokenizer_class.__class__.__name__} {checkpoint}" ) # Load tokenizer _SCREAMING_SNAKE_CASE =tokenizer_class.from_pretrained(__UpperCamelCase , force_download=__UpperCamelCase ) # Save fast tokenizer logger.info(f"Save fast tokenizer to {dump_path} with prefix {checkpoint} add_prefix {add_prefix}" ) # For organization names we create sub-directories if "/" in checkpoint: _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE =checkpoint.split('/' ) _SCREAMING_SNAKE_CASE =os.path.join(__UpperCamelCase , __UpperCamelCase ) elif add_prefix: _SCREAMING_SNAKE_CASE =checkpoint _SCREAMING_SNAKE_CASE =dump_path else: _SCREAMING_SNAKE_CASE =None _SCREAMING_SNAKE_CASE =dump_path logger.info(f"=> {dump_path_full} with prefix {checkpoint_prefix_name}, add_prefix {add_prefix}" ) if checkpoint in list(tokenizer.pretrained_vocab_files_map.values() )[0]: _SCREAMING_SNAKE_CASE =list(tokenizer.pretrained_vocab_files_map.values() )[0][checkpoint] _SCREAMING_SNAKE_CASE =file_path.split(__UpperCamelCase )[-1][0] if next_char == "/": _SCREAMING_SNAKE_CASE =os.path.join(__UpperCamelCase , __UpperCamelCase ) _SCREAMING_SNAKE_CASE =None logger.info(f"=> {dump_path_full} with prefix {checkpoint_prefix_name}, add_prefix {add_prefix}" ) _SCREAMING_SNAKE_CASE =tokenizer.save_pretrained( __UpperCamelCase , legacy_format=__UpperCamelCase , filename_prefix=__UpperCamelCase ) logger.info(f"=> File names {file_names}" ) for file_name in file_names: if not file_name.endswith('tokenizer.json' ): os.remove(__UpperCamelCase ) logger.info(f"=> removing {file_name}" ) if __name__ == "__main__": lowerCamelCase : Optional[Any] = argparse.ArgumentParser() # Required parameters parser.add_argument( "--dump_path", default=None, type=str, required=True, help="Path to output generated fast tokenizer files." ) parser.add_argument( "--tokenizer_name", default=None, type=str, help=( f'''Optional tokenizer type selected in the list of {list(TOKENIZER_CLASSES.keys())}. If not given, will ''' "download and convert all the checkpoints from AWS." ), ) parser.add_argument( "--checkpoint_name", default=None, type=str, help="Optional checkpoint name. If not given, will download and convert the canonical checkpoints from AWS.", ) parser.add_argument( "--force_download", action="store_true", help="Re-download checkpoints.", ) lowerCamelCase : Union[str, Any] = parser.parse_args() convert_slow_checkpoint_to_fast(args.tokenizer_name, args.checkpoint_name, args.dump_path, args.force_download)
47
'''simple docstring''' def lowercase__ ( __UpperCamelCase , __UpperCamelCase , __UpperCamelCase )-> bool: return not any( neighbour == 1 and colored_vertices[i] == color for i, neighbour in enumerate(__UpperCamelCase ) ) def lowercase__ ( __UpperCamelCase , __UpperCamelCase , __UpperCamelCase , __UpperCamelCase )-> bool: # Base Case if index == len(__UpperCamelCase ): return True # Recursive Step for i in range(__UpperCamelCase ): if valid_coloring(graph[index] , __UpperCamelCase , __UpperCamelCase ): # Color current vertex UpperCamelCase = i # Validate coloring if util_color(__UpperCamelCase , __UpperCamelCase , __UpperCamelCase , index + 1 ): return True # Backtrack UpperCamelCase = -1 return False def lowercase__ ( __UpperCamelCase , __UpperCamelCase )-> list[int]: UpperCamelCase = [-1] * len(__UpperCamelCase ) if util_color(__UpperCamelCase , __UpperCamelCase , __UpperCamelCase , 0 ): return colored_vertices return []
321
0
'''simple docstring''' import argparse import os import pickle import sys import torch from transformers import TransfoXLConfig, TransfoXLLMHeadModel, load_tf_weights_in_transfo_xl from transformers.models.transfo_xl import tokenization_transfo_xl as data_utils from transformers.models.transfo_xl.tokenization_transfo_xl import CORPUS_NAME, VOCAB_FILES_NAMES from transformers.utils import CONFIG_NAME, WEIGHTS_NAME, logging logging.set_verbosity_info() # We do this to be able to load python 2 datasets pickles # See e.g. https://stackoverflow.com/questions/2121874/python-pickling-after-changing-a-modules-directory/2121918#2121918 __lowercase: Optional[Any] = data_utils.TransfoXLTokenizer __lowercase: Union[str, Any] = data_utils.TransfoXLCorpus __lowercase: Any = data_utils __lowercase: Optional[Any] = data_utils def SCREAMING_SNAKE_CASE__( _UpperCamelCase : str , _UpperCamelCase : Optional[Any] , _UpperCamelCase : Optional[Any] , _UpperCamelCase : Optional[Any] ) -> Dict: '''simple docstring''' if transfo_xl_dataset_file: # Convert a pre-processed corpus (see original TensorFlow repo) with open(_UpperCamelCase , "rb" ) as fp: UpperCamelCase__ = pickle.load(_UpperCamelCase , encoding="latin1" ) # Save vocabulary and dataset cache as Dictionaries (should be better than pickles for the long-term) UpperCamelCase__ = pytorch_dump_folder_path + "/" + VOCAB_FILES_NAMES["pretrained_vocab_file"] print(F'Save vocabulary to {pytorch_vocab_dump_path}' ) UpperCamelCase__ = corpus.vocab.__dict__ torch.save(_UpperCamelCase , _UpperCamelCase ) UpperCamelCase__ = corpus.__dict__ corpus_dict_no_vocab.pop("vocab" , _UpperCamelCase ) UpperCamelCase__ = pytorch_dump_folder_path + "/" + CORPUS_NAME print(F'Save dataset to {pytorch_dataset_dump_path}' ) torch.save(_UpperCamelCase , _UpperCamelCase ) if tf_checkpoint_path: # Convert a pre-trained TensorFlow model UpperCamelCase__ = os.path.abspath(_UpperCamelCase ) UpperCamelCase__ = os.path.abspath(_UpperCamelCase ) print(F'Converting Transformer XL checkpoint from {tf_path} with config at {config_path}.' ) # Initialise PyTorch model if transfo_xl_config_file == "": UpperCamelCase__ = TransfoXLConfig() else: UpperCamelCase__ = TransfoXLConfig.from_json_file(_UpperCamelCase ) print(F'Building PyTorch model from configuration: {config}' ) UpperCamelCase__ = TransfoXLLMHeadModel(_UpperCamelCase ) UpperCamelCase__ = load_tf_weights_in_transfo_xl(_UpperCamelCase , _UpperCamelCase , _UpperCamelCase ) # Save pytorch-model UpperCamelCase__ = os.path.join(_UpperCamelCase , _UpperCamelCase ) UpperCamelCase__ = os.path.join(_UpperCamelCase , _UpperCamelCase ) print(F'Save PyTorch model to {os.path.abspath(_UpperCamelCase )}' ) torch.save(model.state_dict() , _UpperCamelCase ) print(F'Save configuration file to {os.path.abspath(_UpperCamelCase )}' ) with open(_UpperCamelCase , "w" , encoding="utf-8" ) as f: f.write(config.to_json_string() ) if __name__ == "__main__": __lowercase: Any = argparse.ArgumentParser() parser.add_argument( "--pytorch_dump_folder_path", default=None, type=str, required=True, help="Path to the folder to store the PyTorch model or dataset/vocab.", ) parser.add_argument( "--tf_checkpoint_path", default="", type=str, help="An optional path to a TensorFlow checkpoint path to be converted.", ) parser.add_argument( "--transfo_xl_config_file", default="", type=str, help=( "An optional config json file corresponding to the pre-trained BERT model. \n" "This specifies the model architecture." ), ) parser.add_argument( "--transfo_xl_dataset_file", default="", type=str, help="An optional dataset file to be converted in a vocabulary.", ) __lowercase: Tuple = parser.parse_args() convert_transfo_xl_checkpoint_to_pytorch( args.tf_checkpoint_path, args.transfo_xl_config_file, args.pytorch_dump_folder_path, args.transfo_xl_dataset_file, )
31
'''simple docstring''' def SCREAMING_SNAKE_CASE__( _UpperCamelCase : list ) -> float: '''simple docstring''' UpperCamelCase__ = 0 while len(_UpperCamelCase ) > 1: UpperCamelCase__ = 0 # Consider two files with minimum cost to be merged for _ in range(2 ): UpperCamelCase__ = files.index(min(_UpperCamelCase ) ) temp += files[min_index] files.pop(_UpperCamelCase ) files.append(_UpperCamelCase ) optimal_merge_cost += temp return optimal_merge_cost if __name__ == "__main__": import doctest doctest.testmod()
31
1
from collections import OrderedDict from typing import Mapping from packaging import version from ...configuration_utils import PretrainedConfig from ...onnx import OnnxConfig from ...utils import logging __snake_case :int = logging.get_logger(__name__) __snake_case :Optional[Any] = { '''google/mobilenet_v1_1.0_224''': '''https://huggingface.co/google/mobilenet_v1_1.0_224/resolve/main/config.json''', '''google/mobilenet_v1_0.75_192''': '''https://huggingface.co/google/mobilenet_v1_0.75_192/resolve/main/config.json''', # See all MobileNetV1 models at https://huggingface.co/models?filter=mobilenet_v1 } class _A ( __UpperCAmelCase ): UpperCamelCase__ : List[str] = '''mobilenet_v1''' def __init__( self : int , __SCREAMING_SNAKE_CASE : List[Any]=3 , __SCREAMING_SNAKE_CASE : Tuple=224 , __SCREAMING_SNAKE_CASE : Any=1.0 , __SCREAMING_SNAKE_CASE : List[Any]=8 , __SCREAMING_SNAKE_CASE : Dict="relu6" , __SCREAMING_SNAKE_CASE : Optional[int]=True , __SCREAMING_SNAKE_CASE : Any=0.9_99 , __SCREAMING_SNAKE_CASE : Optional[int]=0.02 , __SCREAMING_SNAKE_CASE : Dict=0.0_01 , **__SCREAMING_SNAKE_CASE : str , ): '''simple docstring''' super().__init__(**__SCREAMING_SNAKE_CASE) if depth_multiplier <= 0: raise ValueError('''depth_multiplier must be greater than zero.''') __a = num_channels __a = image_size __a = depth_multiplier __a = min_depth __a = hidden_act __a = tf_padding __a = classifier_dropout_prob __a = initializer_range __a = layer_norm_eps class _A ( __UpperCAmelCase ): UpperCamelCase__ : List[str] = version.parse('''1.11''' ) @property def _lowerCamelCase ( self : Union[str, Any]): '''simple docstring''' return OrderedDict([('''pixel_values''', {0: '''batch'''})]) @property def _lowerCamelCase ( self : List[str]): '''simple docstring''' if self.task == "image-classification": return OrderedDict([('''logits''', {0: '''batch'''})]) else: return OrderedDict([('''last_hidden_state''', {0: '''batch'''}), ('''pooler_output''', {0: '''batch'''})]) @property def _lowerCamelCase ( self : Dict): '''simple docstring''' return 1E-4
49
"""simple docstring""" from typing import TYPE_CHECKING from ...utils import OptionalDependencyNotAvailable, _LazyModule, is_tf_available, is_torch_available __A = { """configuration_data2vec_audio""": ["""DATA2VEC_AUDIO_PRETRAINED_CONFIG_ARCHIVE_MAP""", """Data2VecAudioConfig"""], """configuration_data2vec_text""": [ """DATA2VEC_TEXT_PRETRAINED_CONFIG_ARCHIVE_MAP""", """Data2VecTextConfig""", """Data2VecTextOnnxConfig""", ], """configuration_data2vec_vision""": [ """DATA2VEC_VISION_PRETRAINED_CONFIG_ARCHIVE_MAP""", """Data2VecVisionConfig""", """Data2VecVisionOnnxConfig""", ], } try: if not is_torch_available(): raise OptionalDependencyNotAvailable() except OptionalDependencyNotAvailable: pass else: __A = [ """DATA2VEC_AUDIO_PRETRAINED_MODEL_ARCHIVE_LIST""", """Data2VecAudioForAudioFrameClassification""", """Data2VecAudioForCTC""", """Data2VecAudioForSequenceClassification""", """Data2VecAudioForXVector""", """Data2VecAudioModel""", """Data2VecAudioPreTrainedModel""", ] __A = [ """DATA2VEC_TEXT_PRETRAINED_MODEL_ARCHIVE_LIST""", """Data2VecTextForCausalLM""", """Data2VecTextForMaskedLM""", """Data2VecTextForMultipleChoice""", """Data2VecTextForQuestionAnswering""", """Data2VecTextForSequenceClassification""", """Data2VecTextForTokenClassification""", """Data2VecTextModel""", """Data2VecTextPreTrainedModel""", ] __A = [ """DATA2VEC_VISION_PRETRAINED_MODEL_ARCHIVE_LIST""", """Data2VecVisionForImageClassification""", """Data2VecVisionForMaskedImageModeling""", """Data2VecVisionForSemanticSegmentation""", """Data2VecVisionModel""", """Data2VecVisionPreTrainedModel""", ] if is_tf_available(): __A = [ """TFData2VecVisionForImageClassification""", """TFData2VecVisionForSemanticSegmentation""", """TFData2VecVisionModel""", """TFData2VecVisionPreTrainedModel""", ] if TYPE_CHECKING: from .configuration_dataavec_audio import DATA2VEC_AUDIO_PRETRAINED_CONFIG_ARCHIVE_MAP, DataaVecAudioConfig from .configuration_dataavec_text import ( DATA2VEC_TEXT_PRETRAINED_CONFIG_ARCHIVE_MAP, DataaVecTextConfig, DataaVecTextOnnxConfig, ) from .configuration_dataavec_vision import ( DATA2VEC_VISION_PRETRAINED_CONFIG_ARCHIVE_MAP, DataaVecVisionConfig, DataaVecVisionOnnxConfig, ) try: if not is_torch_available(): raise OptionalDependencyNotAvailable() except OptionalDependencyNotAvailable: pass else: from .modeling_dataavec_audio import ( DATA2VEC_AUDIO_PRETRAINED_MODEL_ARCHIVE_LIST, DataaVecAudioForAudioFrameClassification, DataaVecAudioForCTC, DataaVecAudioForSequenceClassification, DataaVecAudioForXVector, DataaVecAudioModel, DataaVecAudioPreTrainedModel, ) from .modeling_dataavec_text import ( DATA2VEC_TEXT_PRETRAINED_MODEL_ARCHIVE_LIST, DataaVecTextForCausalLM, DataaVecTextForMaskedLM, DataaVecTextForMultipleChoice, DataaVecTextForQuestionAnswering, DataaVecTextForSequenceClassification, DataaVecTextForTokenClassification, DataaVecTextModel, DataaVecTextPreTrainedModel, ) from .modeling_dataavec_vision import ( DATA2VEC_VISION_PRETRAINED_MODEL_ARCHIVE_LIST, DataaVecVisionForImageClassification, DataaVecVisionForMaskedImageModeling, DataaVecVisionForSemanticSegmentation, DataaVecVisionModel, DataaVecVisionPreTrainedModel, ) if is_tf_available(): from .modeling_tf_dataavec_vision import ( TFDataaVecVisionForImageClassification, TFDataaVecVisionForSemanticSegmentation, TFDataaVecVisionModel, TFDataaVecVisionPreTrainedModel, ) else: import sys __A = _LazyModule(__name__, globals()["""__file__"""], _import_structure, module_spec=__spec__)
293
0
"""simple docstring""" import ast import os import re import shutil import tempfile import unittest from unittest import mock import torch from accelerate.test_utils.examples import compare_against_test from accelerate.test_utils.testing import TempDirTestCase, require_trackers, run_command, slow from accelerate.utils import write_basic_config # DataLoaders built from `test_samples/MRPC` for quick testing # Should mock `{script_name}.get_dataloaders` via: # @mock.patch("{script_name}.get_dataloaders", mocked_dataloaders) __SCREAMING_SNAKE_CASE : Tuple = [ 'cross_validation.py', 'gradient_accumulation.py', 'local_sgd.py', 'multi_process_metrics.py', 'memory.py', 'automatic_gradient_accumulation.py', 'fsdp_with_peak_mem_tracking.py', 'deepspeed_with_config_support.py', 'megatron_lm_gpt_pretraining.py', ] class __A (unittest.TestCase): '''simple docstring''' def lowerCAmelCase ( self : str , UpperCAmelCase_ : str , UpperCAmelCase_ : bool , UpperCAmelCase_ : str = None , UpperCAmelCase_ : list = None ) ->List[Any]: """simple docstring""" snake_case_ = None snake_case_ = os.path.abspath(os.path.join("""examples""" , """by_feature""" ) ) snake_case_ = os.path.abspath("""examples""" ) for item in os.listdir(UpperCAmelCase_ ): if item not in EXCLUDE_EXAMPLES: snake_case_ = os.path.join(UpperCAmelCase_ , UpperCAmelCase_ ) if os.path.isfile(UpperCAmelCase_ ) and ".py" in item_path: with self.subTest( tested_script=UpperCAmelCase_ , feature_script=UpperCAmelCase_ , tested_section="""main()""" if parser_only else """training_function()""" , ): snake_case_ = compare_against_test( os.path.join(UpperCAmelCase_ , UpperCAmelCase_ ) , UpperCAmelCase_ , UpperCAmelCase_ , UpperCAmelCase_ ) snake_case_ = """\n""".join(UpperCAmelCase_ ) if special_strings is not None: for string in special_strings: snake_case_ = diff.replace(UpperCAmelCase_ , """""" ) self.assertEqual(UpperCAmelCase_ , """""" ) def lowerCAmelCase ( self : Tuple ) ->List[Any]: """simple docstring""" self.one_complete_example("""complete_nlp_example.py""" , UpperCAmelCase_ ) self.one_complete_example("""complete_nlp_example.py""" , UpperCAmelCase_ ) def lowerCAmelCase ( self : Tuple ) ->Dict: """simple docstring""" snake_case_ = os.path.abspath(os.path.join("""examples""" , """cv_example.py""" ) ) snake_case_ = [ """ """ * 16 + """{\n\n""", """ """ * 20 + """\"accuracy\": eval_metric[\"accuracy\"],\n\n""", """ """ * 20 + """\"f1\": eval_metric[\"f1\"],\n\n""", """ """ * 20 + """\"train_loss\": total_loss.item() / len(train_dataloader),\n\n""", """ """ * 20 + """\"epoch\": epoch,\n\n""", """ """ * 16 + """},\n\n""", """ """ * 16 + """step=epoch,\n""", """ """ * 12, """ """ * 8 + """for step, batch in enumerate(active_dataloader):\n""", ] self.one_complete_example("""complete_cv_example.py""" , UpperCAmelCase_ , UpperCAmelCase_ , UpperCAmelCase_ ) self.one_complete_example("""complete_cv_example.py""" , UpperCAmelCase_ , UpperCAmelCase_ , UpperCAmelCase_ ) @mock.patch.dict(os.environ , {"""TESTING_MOCKED_DATALOADERS""": """1"""}) class __A (snake_case__): '''simple docstring''' __lowercase: str = False @classmethod def lowerCAmelCase ( cls : Any ) ->List[str]: """simple docstring""" super().setUpClass() snake_case_ = tempfile.mkdtemp() snake_case_ = os.path.join(cls._tmpdir , """default_config.yml""" ) write_basic_config(save_location=cls.configPath ) snake_case_ = ["""accelerate""", """launch""", """--config_file""", cls.configPath] @classmethod def lowerCAmelCase ( cls : List[str] ) ->int: """simple docstring""" super().tearDownClass() shutil.rmtree(cls._tmpdir ) def lowerCAmelCase ( self : List[str] ) ->Optional[Any]: """simple docstring""" snake_case_ = F""" examples/by_feature/checkpointing.py --checkpointing_steps epoch --output_dir {self.tmpdir} """.split() run_command(self._launch_args + testargs ) self.assertTrue(os.path.exists(os.path.join(self.tmpdir , """epoch_0""" ) ) ) def lowerCAmelCase ( self : Any ) ->Optional[Any]: """simple docstring""" snake_case_ = F""" examples/by_feature/checkpointing.py --checkpointing_steps 1 --output_dir {self.tmpdir} """.split() snake_case_ = run_command(self._launch_args + testargs ) self.assertTrue(os.path.exists(os.path.join(self.tmpdir , """step_2""" ) ) ) def lowerCAmelCase ( self : Union[str, Any] ) ->str: """simple docstring""" snake_case_ = F""" examples/by_feature/checkpointing.py --resume_from_checkpoint {os.path.join(self.tmpdir , "epoch_0" )} """.split() snake_case_ = run_command(self._launch_args + testargs , return_stdout=UpperCAmelCase_ ) self.assertNotIn("""epoch 0:""" , UpperCAmelCase_ ) self.assertIn("""epoch 1:""" , UpperCAmelCase_ ) def lowerCAmelCase ( self : Optional[int] ) ->Tuple: """simple docstring""" snake_case_ = F""" examples/by_feature/checkpointing.py --resume_from_checkpoint {os.path.join(self.tmpdir , "step_2" )} """.split() snake_case_ = run_command(self._launch_args + testargs , return_stdout=UpperCAmelCase_ ) if torch.cuda.is_available(): snake_case_ = torch.cuda.device_count() else: snake_case_ = 1 if num_processes > 1: self.assertNotIn("""epoch 0:""" , UpperCAmelCase_ ) self.assertIn("""epoch 1:""" , UpperCAmelCase_ ) else: self.assertIn("""epoch 0:""" , UpperCAmelCase_ ) self.assertIn("""epoch 1:""" , UpperCAmelCase_ ) @slow def lowerCAmelCase ( self : Dict ) ->Dict: """simple docstring""" snake_case_ = """ examples/by_feature/cross_validation.py --num_folds 2 """.split() with mock.patch.dict(os.environ , {"""TESTING_MOCKED_DATALOADERS""": """0"""} ): snake_case_ = run_command(self._launch_args + testargs , return_stdout=UpperCAmelCase_ ) snake_case_ = re.findall("""({.+})""" , UpperCAmelCase_ ) snake_case_ = [r for r in results if """accuracy""" in r][-1] snake_case_ = ast.literal_eval(UpperCAmelCase_ ) self.assertGreaterEqual(results["""accuracy"""] , 0.75 ) def lowerCAmelCase ( self : Optional[int] ) ->int: """simple docstring""" snake_case_ = ["""examples/by_feature/multi_process_metrics.py"""] run_command(self._launch_args + testargs ) @require_trackers @mock.patch.dict(os.environ , {"""WANDB_MODE""": """offline"""} ) def lowerCAmelCase ( self : Dict ) ->Optional[Any]: """simple docstring""" with tempfile.TemporaryDirectory() as tmpdir: snake_case_ = F""" examples/by_feature/tracking.py --with_tracking --project_dir {tmpdir} """.split() run_command(self._launch_args + testargs ) self.assertTrue(os.path.exists(os.path.join(UpperCAmelCase_ , """tracking""" ) ) ) def lowerCAmelCase ( self : List[str] ) ->Union[str, Any]: """simple docstring""" snake_case_ = ["""examples/by_feature/gradient_accumulation.py"""] run_command(self._launch_args + testargs ) def lowerCAmelCase ( self : str ) ->Dict: """simple docstring""" snake_case_ = ["""examples/by_feature/local_sgd.py"""] run_command(self._launch_args + testargs )
233
"""simple docstring""" import copy import tempfile import unittest from transformers import MaMaaaConfig, is_torch_available from transformers.testing_utils import require_sentencepiece, require_tokenizers, require_torch, slow, torch_device from transformers.utils import cached_property from ...generation.test_utils import GenerationTesterMixin from ...test_configuration_common import ConfigTester from ...test_modeling_common import ModelTesterMixin, ids_tensor from ...test_pipeline_mixin import PipelineTesterMixin if is_torch_available(): import torch from transformers import MaMaaaForConditionalGeneration, MaMaaaModel, MaMaaaTokenizer from transformers.models.mam_aaa.modeling_mam_aaa import MaMaaaDecoder, MaMaaaEncoder def _a ( _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE=None , _SCREAMING_SNAKE_CASE=None , _SCREAMING_SNAKE_CASE=None , _SCREAMING_SNAKE_CASE=None , _SCREAMING_SNAKE_CASE=None , ) -> Optional[Any]: if attention_mask is None: snake_case_ = input_ids.ne(config.pad_token_id ) if decoder_attention_mask is None: snake_case_ = decoder_input_ids.ne(config.pad_token_id ) if head_mask is None: snake_case_ = torch.ones(config.encoder_layers , config.encoder_attention_heads , device=_SCREAMING_SNAKE_CASE ) if decoder_head_mask is None: snake_case_ = torch.ones(config.decoder_layers , config.decoder_attention_heads , device=_SCREAMING_SNAKE_CASE ) if cross_attn_head_mask is None: snake_case_ = torch.ones(config.decoder_layers , config.decoder_attention_heads , device=_SCREAMING_SNAKE_CASE ) return { "input_ids": input_ids, "decoder_input_ids": decoder_input_ids, "attention_mask": attention_mask, "decoder_attention_mask": attention_mask, "head_mask": head_mask, "decoder_head_mask": decoder_head_mask, "cross_attn_head_mask": cross_attn_head_mask, } class __A : '''simple docstring''' def __init__( self : str , UpperCAmelCase_ : int , UpperCAmelCase_ : str=13 , UpperCAmelCase_ : Optional[int]=7 , UpperCAmelCase_ : Optional[int]=True , UpperCAmelCase_ : Optional[Any]=False , UpperCAmelCase_ : int=99 , UpperCAmelCase_ : Union[str, Any]=16 , UpperCAmelCase_ : str=2 , UpperCAmelCase_ : List[Any]=4 , UpperCAmelCase_ : int=4 , UpperCAmelCase_ : Optional[Any]="relu" , UpperCAmelCase_ : int=0.1 , UpperCAmelCase_ : List[Any]=0.1 , UpperCAmelCase_ : List[Any]=0.0 , UpperCAmelCase_ : List[str]=0.0 , UpperCAmelCase_ : List[str]=20 , UpperCAmelCase_ : Dict=2 , UpperCAmelCase_ : List[Any]=1 , UpperCAmelCase_ : Optional[Any]=0 , ) ->Dict: """simple docstring""" snake_case_ = parent snake_case_ = batch_size snake_case_ = seq_length snake_case_ = is_training snake_case_ = use_labels snake_case_ = vocab_size snake_case_ = hidden_size snake_case_ = num_hidden_layers snake_case_ = num_attention_heads snake_case_ = intermediate_size snake_case_ = hidden_act snake_case_ = hidden_dropout_prob snake_case_ = attention_probs_dropout_prob snake_case_ = encoder_layerdrop snake_case_ = decoder_layerdrop snake_case_ = max_position_embeddings snake_case_ = eos_token_id snake_case_ = pad_token_id snake_case_ = bos_token_id def lowerCAmelCase ( self : Tuple ) ->List[str]: """simple docstring""" snake_case_ = ids_tensor([self.batch_size, self.seq_length] , self.vocab_size ) snake_case_ = self.eos_token_id # Eos Token snake_case_ = ids_tensor([self.batch_size, self.seq_length] , self.vocab_size ) # we need to clamp the input ids here to avoid having pad token in between # this is because for M2M100 the position_ids are prepared such that # all pad tokens have pos id = 2 and rest are between 2..seq_length # and the seq_length here is seq_length - num_pad_tokens # but when using past, there is no way of knowing if the past input ids had # pad tokens in them, which results in incorrect seq_lenth and which in turn results in # position_ids being off by num_pad_tokens in past input snake_case_ = input_ids.clamp(self.pad_token_id + 1 ) snake_case_ = decoder_input_ids.clamp(self.pad_token_id + 1 ) snake_case_ = self.get_config() snake_case_ = prepare_mam_aaa_inputs_dict(UpperCAmelCase_ , UpperCAmelCase_ , UpperCAmelCase_ ) return config, inputs_dict def lowerCAmelCase ( self : Optional[int] ) ->Union[str, Any]: """simple docstring""" return MaMaaaConfig( 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 , encoder_layerdrop=self.encoder_layerdrop , decoder_layerdrop=self.decoder_layerdrop , max_position_embeddings=self.max_position_embeddings , eos_token_id=self.eos_token_id , bos_token_id=self.bos_token_id , pad_token_id=self.pad_token_id , ) def lowerCAmelCase ( self : int ) ->Optional[int]: """simple docstring""" snake_case_ , snake_case_ = self.prepare_config_and_inputs() return config, inputs_dict def lowerCAmelCase ( self : Dict , UpperCAmelCase_ : Optional[Any] , UpperCAmelCase_ : Dict ) ->Dict: """simple docstring""" snake_case_ = MaMaaaModel(config=UpperCAmelCase_ ).get_decoder().to(UpperCAmelCase_ ).eval() snake_case_ = inputs_dict["""input_ids"""] snake_case_ = inputs_dict["""attention_mask"""] snake_case_ = inputs_dict["""head_mask"""] # first forward pass snake_case_ = model(UpperCAmelCase_ , attention_mask=UpperCAmelCase_ , head_mask=UpperCAmelCase_ , use_cache=UpperCAmelCase_ ) snake_case_ , snake_case_ = outputs.to_tuple() # create hypothetical multiple next token and extent to next_input_ids snake_case_ = ids_tensor((self.batch_size, 3) , config.vocab_size ) snake_case_ = ids_tensor((self.batch_size, 3) , 2 ) # append to next input_ids and snake_case_ = torch.cat([input_ids, next_tokens] , dim=-1 ) snake_case_ = torch.cat([attention_mask, next_attn_mask] , dim=-1 ) snake_case_ = model(UpperCAmelCase_ , attention_mask=UpperCAmelCase_ )["""last_hidden_state"""] snake_case_ = model(UpperCAmelCase_ , attention_mask=UpperCAmelCase_ , past_key_values=UpperCAmelCase_ )[ """last_hidden_state""" ] # select random slice snake_case_ = ids_tensor((1,) , output_from_past.shape[-1] ).item() snake_case_ = output_from_no_past[:, -3:, random_slice_idx].detach() snake_case_ = 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(UpperCAmelCase_ , UpperCAmelCase_ , atol=1E-2 ) ) def lowerCAmelCase ( self : Dict , UpperCAmelCase_ : Union[str, Any] , UpperCAmelCase_ : Dict ) ->int: """simple docstring""" snake_case_ = MaMaaaModel(config=UpperCAmelCase_ ).to(UpperCAmelCase_ ).eval() snake_case_ = model(**UpperCAmelCase_ ) snake_case_ = outputs.encoder_last_hidden_state snake_case_ = outputs.last_hidden_state with tempfile.TemporaryDirectory() as tmpdirname: snake_case_ = model.get_encoder() encoder.save_pretrained(UpperCAmelCase_ ) snake_case_ = MaMaaaEncoder.from_pretrained(UpperCAmelCase_ ).to(UpperCAmelCase_ ) snake_case_ = encoder(inputs_dict["""input_ids"""] , attention_mask=inputs_dict["""attention_mask"""] )[ 0 ] self.parent.assertTrue((encoder_last_hidden_state_a - encoder_last_hidden_state).abs().max().item() < 1E-3 ) with tempfile.TemporaryDirectory() as tmpdirname: snake_case_ = model.get_decoder() decoder.save_pretrained(UpperCAmelCase_ ) snake_case_ = MaMaaaDecoder.from_pretrained(UpperCAmelCase_ ).to(UpperCAmelCase_ ) snake_case_ = decoder( input_ids=inputs_dict["""decoder_input_ids"""] , attention_mask=inputs_dict["""decoder_attention_mask"""] , encoder_hidden_states=UpperCAmelCase_ , encoder_attention_mask=inputs_dict["""attention_mask"""] , )[0] self.parent.assertTrue((last_hidden_state_a - last_hidden_state).abs().max().item() < 1E-3 ) @require_torch class __A (snake_case__ , snake_case__ , snake_case__ , unittest.TestCase): '''simple docstring''' __lowercase: Optional[Any] = ( ( MaMaaaModel, MaMaaaForConditionalGeneration, ) if is_torch_available() else () ) __lowercase: Union[str, Any] = (MaMaaaForConditionalGeneration,) if is_torch_available() else () __lowercase: Tuple = ( { """conversational""": MaMaaaForConditionalGeneration, """feature-extraction""": MaMaaaModel, """summarization""": MaMaaaForConditionalGeneration, """text2text-generation""": MaMaaaForConditionalGeneration, """translation""": MaMaaaForConditionalGeneration, } if is_torch_available() else {} ) __lowercase: Dict = True __lowercase: List[Any] = True __lowercase: Union[str, Any] = False __lowercase: Optional[int] = False def lowerCAmelCase ( self : Union[str, Any] , UpperCAmelCase_ : List[str] , UpperCAmelCase_ : int , UpperCAmelCase_ : Optional[int] , UpperCAmelCase_ : Optional[int] , UpperCAmelCase_ : int ) ->str: """simple docstring""" if pipeline_test_casse_name == "TranslationPipelineTests": # Get `ValueError: Translation requires a `src_lang` and a `tgt_lang` for this model`. # `M2M100Config` was never used in pipeline tests: cannot create a simple tokenizer. return True return False def lowerCAmelCase ( self : int ) ->Dict: """simple docstring""" snake_case_ = MaMaaaModelTester(self ) snake_case_ = ConfigTester(self , config_class=UpperCAmelCase_ ) def lowerCAmelCase ( self : Optional[Any] ) ->Tuple: """simple docstring""" self.config_tester.run_common_tests() def lowerCAmelCase ( self : Union[str, Any] ) ->Optional[Any]: """simple docstring""" snake_case_ , snake_case_ = self.model_tester.prepare_config_and_inputs() for model_class in self.all_model_classes: snake_case_ = model_class(UpperCAmelCase_ ) with tempfile.TemporaryDirectory() as tmpdirname: model.save_pretrained(UpperCAmelCase_ ) snake_case_ , snake_case_ = model_class.from_pretrained(UpperCAmelCase_ , output_loading_info=UpperCAmelCase_ ) self.assertEqual(info["""missing_keys"""] , [] ) def lowerCAmelCase ( self : Optional[int] ) ->List[Any]: """simple docstring""" snake_case_ = self.model_tester.prepare_config_and_inputs() self.model_tester.create_and_check_decoder_model_past_large_inputs(*UpperCAmelCase_ ) def lowerCAmelCase ( self : str ) ->List[str]: """simple docstring""" snake_case_ = self.model_tester.prepare_config_and_inputs_for_common() self.model_tester.check_encoder_decoder_model_standalone(*UpperCAmelCase_ ) def lowerCAmelCase ( self : str ) ->List[str]: """simple docstring""" snake_case_ , snake_case_ = self.model_tester.prepare_config_and_inputs_for_common() for model_class in (MaMaaaModel, MaMaaaForConditionalGeneration): snake_case_ = model_class(UpperCAmelCase_ ) model.to(UpperCAmelCase_ ) model.eval() snake_case_ = copy.deepcopy(self._prepare_for_class(UpperCAmelCase_ , UpperCAmelCase_ ) ) if not self.is_encoder_decoder: snake_case_ = inputs["""input_ids"""] del inputs["input_ids"] else: snake_case_ = inputs["""input_ids"""] snake_case_ = inputs.get("""decoder_input_ids""" , UpperCAmelCase_ ) del inputs["input_ids"] inputs.pop("""decoder_input_ids""" , UpperCAmelCase_ ) snake_case_ = model.get_input_embeddings() if not self.is_encoder_decoder: snake_case_ = wte(UpperCAmelCase_ ) else: snake_case_ = wte(UpperCAmelCase_ ) snake_case_ = wte(UpperCAmelCase_ ) with torch.no_grad(): model(**UpperCAmelCase_ )[0] def lowerCAmelCase ( self : Any ) ->Any: """simple docstring""" snake_case_ , snake_case_ = self.model_tester.prepare_config_and_inputs() snake_case_ = input_dict["""input_ids"""] snake_case_ = input_ids.ne(1 ).to(UpperCAmelCase_ ) snake_case_ = MaMaaaForConditionalGeneration(UpperCAmelCase_ ).eval().to(UpperCAmelCase_ ) if torch_device == "cuda": model.half() model.generate(UpperCAmelCase_ , attention_mask=UpperCAmelCase_ ) model.generate(num_beams=4 , do_sample=UpperCAmelCase_ , early_stopping=UpperCAmelCase_ , num_return_sequences=3 ) def _a ( _SCREAMING_SNAKE_CASE ) -> Optional[int]: return torch.tensor(_SCREAMING_SNAKE_CASE , dtype=torch.long , device=_SCREAMING_SNAKE_CASE ) __SCREAMING_SNAKE_CASE : Tuple = 1E-4 @require_torch @require_sentencepiece @require_tokenizers @slow class __A (unittest.TestCase): '''simple docstring''' @cached_property def lowerCAmelCase ( self : Dict ) ->str: """simple docstring""" return MaMaaaTokenizer.from_pretrained("""facebook/m2m100_418M""" ) def lowerCAmelCase ( self : str ) ->Any: """simple docstring""" snake_case_ = MaMaaaModel.from_pretrained("""facebook/m2m100_418M""" ).to(UpperCAmelCase_ ) snake_case_ = _long_tensor([[128_028, 98, 12, 30_527, 2_732, 159, 7_755, 61_904, 39_144, 38, 2]] ) snake_case_ = _long_tensor([[2, 128_028, 98, 12, 30_527, 2_732, 159, 7_755, 61_904, 39_144, 38]] ) snake_case_ = prepare_mam_aaa_inputs_dict(model.config , UpperCAmelCase_ , UpperCAmelCase_ ) with torch.no_grad(): snake_case_ = model(**UpperCAmelCase_ )[0] snake_case_ = torch.Size((1, 11, 1_024) ) self.assertEqual(output.shape , UpperCAmelCase_ ) # change to expected output here snake_case_ = torch.tensor( [[-0.7_780, -0.1_676, 0.1_038], [-6.7_556, -1.3_992, 0.0_567], [-7.5_383, -0.5_920, -0.2_779]] , device=UpperCAmelCase_ ) self.assertTrue(torch.allclose(output[:, :3, :3] , UpperCAmelCase_ , atol=UpperCAmelCase_ ) ) def lowerCAmelCase ( self : Optional[int] ) ->Any: """simple docstring""" snake_case_ = MaMaaaForConditionalGeneration.from_pretrained("""facebook/m2m100_418M""" ).to(UpperCAmelCase_ ) # change to intended input snake_case_ = _long_tensor([[128_028, 98, 12, 30_527, 2_732, 159, 7_755, 61_904, 39_144, 38, 2]] ) snake_case_ = _long_tensor([[2, 128_028, 98, 12, 30_527, 2_732, 159, 7_755, 61_904, 39_144, 38]] ) snake_case_ = prepare_mam_aaa_inputs_dict(model.config , UpperCAmelCase_ , UpperCAmelCase_ ) with torch.no_grad(): snake_case_ = model(**UpperCAmelCase_ )[0] snake_case_ = torch.Size((1, 11, model.config.vocab_size) ) self.assertEqual(output.shape , UpperCAmelCase_ ) # change to expected output here snake_case_ = torch.tensor( [[-1.0_448, -1.0_411, 3.7_992], [-3.2_191, -3.2_386, -1.3_451], [-3.6_210, -3.5_993, 0.4_925]] , device=UpperCAmelCase_ ) self.assertTrue(torch.allclose(output[:, :3, :3] , UpperCAmelCase_ , atol=UpperCAmelCase_ ) ) def lowerCAmelCase ( self : Dict ) ->str: """simple docstring""" snake_case_ = MaMaaaForConditionalGeneration.from_pretrained("""facebook/m2m100_418M""" ).to(UpperCAmelCase_ ) snake_case_ = MaMaaaTokenizer.from_pretrained("""facebook/m2m100_418M""" , src_lang="""fr""" , tgt_lang="""en""" ) snake_case_ = [ """L'affaire NSA souligne l'absence totale de débat sur le renseignement""", """Selon moi, il y a deux niveaux de réponse de la part du gouvernement français.""", """Lorsque François Hollande téléphone à Barack Obama ou quand le ministre des affaires étrangères Laurent""" """ Fabius convoque l'ambassadeur des Etats-Unis, ils réagissent à une vraie découverte, qui est celle de""" """ l'ampleur de la surveillance américaine sur l'ensemble des communications en France.""", ] # The below article tests that we don't add any hypotheses outside of the top n_beams snake_case_ = tokenizer(UpperCAmelCase_ , padding=UpperCAmelCase_ , return_tensors="""pt""" ) snake_case_ = model.generate( input_ids=dct["""input_ids"""].to(UpperCAmelCase_ ) , attention_mask=dct["""attention_mask"""].to(UpperCAmelCase_ ) , num_beams=5 , forced_bos_token_id=tokenizer.get_lang_id("""en""" ) , ) snake_case_ = [ """The NSA case highlights the total absence of intelligence debate""", """I think there are two levels of response from the French government.""", """When François Hollande calls Barack Obama or when Foreign Minister Laurent Fabius calls the U.S.""" """ Ambassador, they respond to a real discovery, which is that of the scale of U.S. surveillance on all""" """ communications in France.""", ] snake_case_ = tokenizer.batch_decode( hypotheses_batch.tolist() , clean_up_tokenization_spaces=UpperCAmelCase_ , skip_special_tokens=UpperCAmelCase_ ) assert generated == expected_en
233
1
"""simple docstring""" import gc import random import unittest import numpy as np import torch from PIL import Image from diffusers import ( DDIMScheduler, KandinskyVaaControlnetImgaImgPipeline, KandinskyVaaPriorEmbaEmbPipeline, UNetaDConditionModel, VQModel, ) from diffusers.utils import floats_tensor, load_image, load_numpy, slow, torch_device from diffusers.utils.testing_utils import enable_full_determinism, require_torch_gpu from ..test_pipelines_common import PipelineTesterMixin, assert_mean_pixel_difference enable_full_determinism() class _UpperCAmelCase ( _A , unittest.TestCase ): SCREAMING_SNAKE_CASE_ : Any = KandinskyVaaControlnetImgaImgPipeline SCREAMING_SNAKE_CASE_ : Optional[int] = ["image_embeds", "negative_image_embeds", "image", "hint"] SCREAMING_SNAKE_CASE_ : str = ["image_embeds", "negative_image_embeds", "image", "hint"] SCREAMING_SNAKE_CASE_ : Dict = [ "generator", "height", "width", "strength", "guidance_scale", "num_inference_steps", "return_dict", "guidance_scale", "num_images_per_prompt", "output_type", "return_dict", ] SCREAMING_SNAKE_CASE_ : Dict = False @property def A ( self : Any ) -> Any: return 32 @property def A ( self : Optional[int] ) -> Any: return 32 @property def A ( self : Dict ) -> int: return self.time_input_dim @property def A ( self : Tuple ) -> str: return self.time_input_dim * 4 @property def A ( self : Any ) -> str: return 1_00 @property def A ( self : str ) -> List[str]: torch.manual_seed(0 ) lowercase_ : List[Any] = { '''in_channels''': 8, # Out channels is double in channels because predicts mean and variance '''out_channels''': 8, '''addition_embed_type''': '''image_hint''', '''down_block_types''': ('''ResnetDownsampleBlock2D''', '''SimpleCrossAttnDownBlock2D'''), '''up_block_types''': ('''SimpleCrossAttnUpBlock2D''', '''ResnetUpsampleBlock2D'''), '''mid_block_type''': '''UNetMidBlock2DSimpleCrossAttn''', '''block_out_channels''': (self.block_out_channels_a, self.block_out_channels_a * 2), '''layers_per_block''': 1, '''encoder_hid_dim''': self.text_embedder_hidden_size, '''encoder_hid_dim_type''': '''image_proj''', '''cross_attention_dim''': self.cross_attention_dim, '''attention_head_dim''': 4, '''resnet_time_scale_shift''': '''scale_shift''', '''class_embed_type''': None, } lowercase_ : Dict = UNetaDConditionModel(**A ) return model @property def A ( self : Optional[Any] ) -> Union[str, Any]: return { "block_out_channels": [32, 32, 64, 64], "down_block_types": [ "DownEncoderBlock2D", "DownEncoderBlock2D", "DownEncoderBlock2D", "AttnDownEncoderBlock2D", ], "in_channels": 3, "latent_channels": 4, "layers_per_block": 1, "norm_num_groups": 8, "norm_type": "spatial", "num_vq_embeddings": 12, "out_channels": 3, "up_block_types": ["AttnUpDecoderBlock2D", "UpDecoderBlock2D", "UpDecoderBlock2D", "UpDecoderBlock2D"], "vq_embed_dim": 4, } @property def A ( self : List[Any] ) -> Dict: torch.manual_seed(0 ) lowercase_ : int = VQModel(**self.dummy_movq_kwargs ) return model def A ( self : Union[str, Any] ) -> Optional[int]: lowercase_ : Tuple = self.dummy_unet lowercase_ : int = self.dummy_movq lowercase_ : List[Any] = { '''num_train_timesteps''': 10_00, '''beta_schedule''': '''linear''', '''beta_start''': 0.00085, '''beta_end''': 0.012, '''clip_sample''': False, '''set_alpha_to_one''': False, '''steps_offset''': 0, '''prediction_type''': '''epsilon''', '''thresholding''': False, } lowercase_ : str = DDIMScheduler(**A ) lowercase_ : Tuple = { '''unet''': unet, '''scheduler''': scheduler, '''movq''': movq, } return components def A ( self : Optional[int] , A : int , A : List[str]=0 ) -> int: lowercase_ : str = floats_tensor((1, self.text_embedder_hidden_size) , rng=random.Random(A ) ).to(A ) lowercase_ : Tuple = floats_tensor((1, self.text_embedder_hidden_size) , rng=random.Random(seed + 1 ) ).to( A ) # create init_image lowercase_ : Union[str, Any] = floats_tensor((1, 3, 64, 64) , rng=random.Random(A ) ).to(A ) lowercase_ : Any = image.cpu().permute(0 , 2 , 3 , 1 )[0] lowercase_ : Optional[Any] = Image.fromarray(np.uinta(A ) ).convert('''RGB''' ).resize((2_56, 2_56) ) # create hint lowercase_ : Optional[int] = floats_tensor((1, 3, 64, 64) , rng=random.Random(A ) ).to(A ) if str(A ).startswith('''mps''' ): lowercase_ : Optional[Any] = torch.manual_seed(A ) else: lowercase_ : List[Any] = torch.Generator(device=A ).manual_seed(A ) lowercase_ : Dict = { '''image''': init_image, '''image_embeds''': image_embeds, '''negative_image_embeds''': negative_image_embeds, '''hint''': hint, '''generator''': generator, '''height''': 64, '''width''': 64, '''num_inference_steps''': 10, '''guidance_scale''': 7.0, '''strength''': 0.2, '''output_type''': '''np''', } return inputs def A ( self : Any ) -> List[Any]: lowercase_ : List[str] = '''cpu''' lowercase_ : Any = self.get_dummy_components() lowercase_ : Any = self.pipeline_class(**A ) lowercase_ : int = pipe.to(A ) pipe.set_progress_bar_config(disable=A ) lowercase_ : Dict = pipe(**self.get_dummy_inputs(A ) ) lowercase_ : str = output.images lowercase_ : int = pipe( **self.get_dummy_inputs(A ) , return_dict=A , )[0] lowercase_ : Dict = image[0, -3:, -3:, -1] lowercase_ : Union[str, Any] = image_from_tuple[0, -3:, -3:, -1] assert image.shape == (1, 64, 64, 3) lowercase_ : List[str] = np.array( [0.54985034, 0.55509365, 0.52561504, 0.5570494, 0.5593818, 0.5263979, 0.50285643, 0.5069846, 0.51196736] ) assert ( np.abs(image_slice.flatten() - expected_slice ).max() < 1e-2 ), F''' expected_slice {expected_slice}, but got {image_slice.flatten()}''' assert ( np.abs(image_from_tuple_slice.flatten() - expected_slice ).max() < 1e-2 ), F''' expected_slice {expected_slice}, but got {image_from_tuple_slice.flatten()}''' @slow @require_torch_gpu class _UpperCAmelCase ( unittest.TestCase ): def A ( self : Tuple ) -> str: # clean up the VRAM after each test super().tearDown() gc.collect() torch.cuda.empty_cache() def A ( self : Any ) -> Optional[int]: lowercase_ : Dict = load_numpy( '''https://huggingface.co/datasets/hf-internal-testing/diffusers-images/resolve/main''' '''/kandinskyv22/kandinskyv22_controlnet_img2img_robotcat_fp16.npy''' ) lowercase_ : Dict = load_image( '''https://huggingface.co/datasets/hf-internal-testing/diffusers-images/resolve/main''' '''/kandinsky/cat.png''' ) lowercase_ : Optional[int] = init_image.resize((5_12, 5_12) ) lowercase_ : Dict = load_image( '''https://huggingface.co/datasets/hf-internal-testing/diffusers-images/resolve/main''' '''/kandinskyv22/hint_image_cat.png''' ) lowercase_ : Optional[int] = torch.from_numpy(np.array(A ) ).float() / 255.0 lowercase_ : Tuple = hint.permute(2 , 0 , 1 ).unsqueeze(0 ) lowercase_ : Optional[Any] = '''A robot, 4k photo''' lowercase_ : Tuple = KandinskyVaaPriorEmbaEmbPipeline.from_pretrained( '''kandinsky-community/kandinsky-2-2-prior''' , torch_dtype=torch.floataa ) pipe_prior.to(A ) lowercase_ : Dict = KandinskyVaaControlnetImgaImgPipeline.from_pretrained( '''kandinsky-community/kandinsky-2-2-controlnet-depth''' , torch_dtype=torch.floataa ) lowercase_ : int = pipeline.to(A ) pipeline.set_progress_bar_config(disable=A ) lowercase_ : Tuple = torch.Generator(device='''cpu''' ).manual_seed(0 ) lowercase_ , lowercase_ : int = pipe_prior( A , image=A , strength=0.85 , generator=A , negative_prompt='''''' , ).to_tuple() lowercase_ : str = pipeline( image=A , image_embeds=A , negative_image_embeds=A , hint=A , generator=A , num_inference_steps=1_00 , height=5_12 , width=5_12 , strength=0.5 , output_type='''np''' , ) lowercase_ : Optional[Any] = output.images[0] assert image.shape == (5_12, 5_12, 3) assert_mean_pixel_difference(A , A )
33
"""simple docstring""" def lowercase ( __snake_case : int = 1_0_0_0 ): lowercase_ , lowercase_ : str = 1, 1 lowercase_ : List[str] = 2 while True: lowercase_ : Tuple = 0 lowercase_ : List[Any] = fa + fa lowercase_ , lowercase_ : Optional[int] = fa, f index += 1 for _ in str(__snake_case ): i += 1 if i == n: break return index if __name__ == "__main__": print(solution(int(str(input()).strip())))
33
1
'''simple docstring''' import torch from transformers import PreTrainedModel, XLMRobertaConfig, XLMRobertaModel class SCREAMING_SNAKE_CASE__ ( snake_case__ ): """simple docstring""" SCREAMING_SNAKE_CASE = '''M-CLIP''' def __init__( self : List[str] , UpperCAmelCase_ : str=1_024 , UpperCAmelCase_ : Optional[Any]=768 , **UpperCAmelCase_ : Tuple ): """simple docstring""" __UpperCAmelCase : Any = transformerDimSize __UpperCAmelCase : Union[str, Any] = imageDimSize super().__init__(**UpperCAmelCase_ ) class SCREAMING_SNAKE_CASE__ ( snake_case__ ): """simple docstring""" SCREAMING_SNAKE_CASE = MCLIPConfig def __init__( self : Optional[Any] , UpperCAmelCase_ : Dict , *UpperCAmelCase_ : int , **UpperCAmelCase_ : Optional[Any] ): """simple docstring""" super().__init__(UpperCAmelCase_ , *UpperCAmelCase_ , **UpperCAmelCase_ ) __UpperCAmelCase : int = XLMRobertaModel(UpperCAmelCase_ ) __UpperCAmelCase : str = torch.nn.Linear( in_features=config.transformerDimensions , out_features=config.numDims ) def lowerCamelCase_ ( self : str , UpperCAmelCase_ : Any , UpperCAmelCase_ : Any ): """simple docstring""" __UpperCAmelCase : Optional[Any] = self.transformer(input_ids=UpperCAmelCase_ , attention_mask=UpperCAmelCase_ )[0] __UpperCAmelCase : List[Any] = (embs * attention_mask.unsqueeze(2 )).sum(dim=1 ) / attention_mask.sum(dim=1 )[:, None] return self.LinearTransformation(UpperCAmelCase_ ), embs
368
'''simple docstring''' import argparse import re from flax.traverse_util import flatten_dict, unflatten_dict from tax import checkpoints from transformers import SwitchTransformersConfig, SwitchTransformersForConditionalGeneration from transformers.modeling_flax_pytorch_utils import load_flax_weights_in_pytorch_model from transformers.utils import logging logging.set_verbosity_info() # should not include what is already done by the `from_pt` argument lowerCAmelCase__ : Optional[Any] = { "/attention/": "/0/SelfAttention/", "/self_attention/": "/0/SelfAttention/", "/encoder_decoder_attention/": "/1/EncDecAttention/", "value": "v", "query": "q", "key": "k", "out": "o", "pre_self_attention_layer_norm": "0/layer_norm", "pre_cross_attention_layer_norm": "1/layer_norm", "pre_attention_layer_norm": "0/layer_norm", # previously 1, but seems wrong "token_embedder": "shared", "encoder_norm": "final_layer_norm", "decoder_norm": "final_layer_norm", "relpos_bias/rel_embedding": "block/0/layer/0/SelfAttention/relative_attention_bias/weight", "router/router_weights/w/": "router/classifier/", "roer/roer_weights/w/": "router/classifier/", "logits_dense": "lm_head", } def __UpperCamelCase ( _UpperCAmelCase ): # 1. in HF T5, we have block.{x}.layer.{y}. which corresponds to layer.{x} in # the original model __UpperCAmelCase : List[str] = list(s_dict.keys() ) for key in keys: __UpperCAmelCase : int = R".*/layers_(\d+)" __UpperCAmelCase : List[str] = key if re.match(_UpperCAmelCase, _UpperCAmelCase ): __UpperCAmelCase : Optional[int] = re.sub(R"layers_(\d+)", R"block/\1/layer", _UpperCAmelCase ) __UpperCAmelCase : Any = R"(encoder|decoder)\/" if re.match(_UpperCAmelCase, _UpperCAmelCase ): __UpperCAmelCase : List[Any] = re.match(_UpperCAmelCase, _UpperCAmelCase ).groups() if groups[0] == "encoder": __UpperCAmelCase : Optional[Any] = re.sub(R"/mlp/", R"/1/mlp/", _UpperCAmelCase ) __UpperCAmelCase : List[Any] = re.sub(R"/pre_mlp_layer_norm/", R"/1/layer_norm/", _UpperCAmelCase ) elif groups[0] == "decoder": __UpperCAmelCase : List[Any] = re.sub(R"/mlp/", R"/2/mlp/", _UpperCAmelCase ) __UpperCAmelCase : Any = re.sub(R"/pre_mlp_layer_norm/", R"/2/layer_norm/", _UpperCAmelCase ) # 2. Convert other classic mappings for old_key, temp_key in MOE_LAYER_NAME_MAPPING.items(): if old_key in new_key: __UpperCAmelCase : List[str] = new_key.replace(_UpperCAmelCase, _UpperCAmelCase ) print(F"{key} -> {new_key}" ) __UpperCAmelCase : Any = s_dict.pop(_UpperCAmelCase ) if "encoder/block/0/layer/0/SelfAttention/relative_attention_bias/weight" in s_dict: __UpperCAmelCase : Tuple = s_dict[ "encoder/block/0/layer/0/SelfAttention/relative_attention_bias/weight" ].T if "decoder/block/0/layer/0/SelfAttention/relative_attention_bias/weight" in s_dict: __UpperCAmelCase : Optional[Any] = s_dict[ "decoder/block/0/layer/0/SelfAttention/relative_attention_bias/weight" ].T # 3. Take extra care of the EXPERTS layer for key in list(s_dict.keys() ): if "expert" in key: __UpperCAmelCase : Any = s_dict[key].shape[0] __UpperCAmelCase : str = s_dict[key] for idx in range(_UpperCAmelCase ): __UpperCAmelCase : Optional[Any] = expert_weihts[idx] print(F"{key} -> {key.replace('expert/', 'nested fstring' )}" ) s_dict.pop(_UpperCAmelCase ) return s_dict lowerCAmelCase__ : Optional[Any] = { "NUM_ENCODER_LAYERS": "num_layers", "NUM_DECODER_LAYERS": "num_decoder_layers", "NUM_HEADS": "num_heads", "HEAD_DIM": "d_kv", "EMBED_DIM": "d_model", "MLP_DIM": "d_ff", "NUM_SELECTED_EXPERTS": "num_selected_experts", "NUM_ENCODER_SPARSE_LAYERS": "num_sparse_encoder_layers", "NUM_DECODER_SPARSE_LAYERS": "num_sparse_decoder_layers", "dense.MlpBlock.activations": "feed_forward_proj", } def __UpperCamelCase ( _UpperCAmelCase, _UpperCAmelCase ): # Convert a google style config to the hugging face fromat import regex as re with open(_UpperCAmelCase, "r" ) as f: __UpperCAmelCase : List[Any] = f.read() __UpperCAmelCase : Union[str, Any] = re.findall(R"(.*) = ([0-9.]*)", _UpperCAmelCase ) __UpperCAmelCase : Dict = {} for param, value in regex_match: if param in GIN_TO_CONFIG_MAPPING and value != "": __UpperCAmelCase : Tuple = float(_UpperCAmelCase ) if "." in value else int(_UpperCAmelCase ) __UpperCAmelCase : str = re.findall(R"(.*activations) = \(\'(.*)\',\)", _UpperCAmelCase )[0] __UpperCAmelCase : int = str(activation[1] ) __UpperCAmelCase : int = num_experts __UpperCAmelCase : List[str] = SwitchTransformersConfig(**_UpperCAmelCase ) return config def __UpperCamelCase ( _UpperCAmelCase, _UpperCAmelCase, _UpperCAmelCase=None, _UpperCAmelCase="./", _UpperCAmelCase=8 ): # Initialise PyTorch model print(F"Loading flax weights from : {flax_checkpoint_path}" ) __UpperCAmelCase : Dict = checkpoints.load_tax_checkpoint(_UpperCAmelCase ) if gin_file is not None: __UpperCAmelCase : int = convert_gin_to_config(_UpperCAmelCase, _UpperCAmelCase ) else: __UpperCAmelCase : int = SwitchTransformersConfig.from_pretrained(_UpperCAmelCase ) __UpperCAmelCase : Any = SwitchTransformersForConditionalGeneration(_UpperCAmelCase ) __UpperCAmelCase : str = flax_params["target"] __UpperCAmelCase : Any = flatten_dict(_UpperCAmelCase, sep="/" ) __UpperCAmelCase : Optional[Any] = rename_keys(_UpperCAmelCase ) __UpperCAmelCase : Any = unflatten_dict(_UpperCAmelCase, sep="/" ) # Load the flax params in the PT model load_flax_weights_in_pytorch_model(_UpperCAmelCase, _UpperCAmelCase ) print(F"Save PyTorch model to {pytorch_dump_path}" ) pt_model.save_pretrained(_UpperCAmelCase ) if __name__ == "__main__": lowerCAmelCase__ : int = argparse.ArgumentParser() # Required parameters parser.add_argument( "--switch_t5x_checkpoint_path", default=None, type=str, required=True, help=( "The config json file corresponding to the pre-trained SwitchTransformers model. \nThis specifies the" " model architecture. If not provided, a `gin_file` has to be provided." ), ) parser.add_argument( "--gin_file", default=None, type=str, required=False, help="Path to the gin config file. If not provided, a `config_file` has to be passed ", ) parser.add_argument( "--config_name", default=None, type=str, required=False, help="Config name of SwitchTransformers model." ) parser.add_argument( "--pytorch_dump_folder_path", default=None, type=str, required=True, help="Path to the output pytorch model." ) parser.add_argument("--num_experts", default=8, type=int, required=False, help="Number of experts") lowerCAmelCase__ : int = parser.parse_args() convert_flax_checkpoint_to_pytorch( args.switch_tax_checkpoint_path, args.config_name, args.gin_file, args.pytorch_dump_folder_path, args.num_experts, )
37
0
"""simple docstring""" from collections.abc import Iterator, MutableMapping from dataclasses import dataclass from typing import Generic, TypeVar _snake_case = TypeVar('KEY') _snake_case = TypeVar('VAL') @dataclass(frozen=snake_case_ , slots=snake_case_ ) class UpperCamelCase ( Generic[KEY, VAL] ): UpperCamelCase : KEY UpperCamelCase : VAL class UpperCamelCase ( _Item ): def __init__( self : Optional[int] ) -> None: super().__init__(UpperCAmelCase__ , UpperCAmelCase__ ) def __bool__( self : List[str] ) -> bool: return False _snake_case = _DeletedItem() class UpperCamelCase ( MutableMapping[KEY, VAL] ): def __init__( self : Optional[Any] , UpperCAmelCase__ : int = 8 , UpperCAmelCase__ : float = 0.7_5 ) -> None: _a : List[Any] = initial_block_size _a : list[_Item | None] = [None] * initial_block_size assert 0.0 < capacity_factor < 1.0 _a : Optional[int] = capacity_factor _a : int = 0 def _lowercase ( self : List[str] , UpperCAmelCase__ : KEY ) -> int: return hash(UpperCAmelCase__ ) % len(self._buckets ) def _lowercase ( self : Dict , UpperCAmelCase__ : int ) -> int: return (ind + 1) % len(self._buckets ) def _lowercase ( self : Any , UpperCAmelCase__ : int , UpperCAmelCase__ : KEY , UpperCAmelCase__ : VAL ) -> bool: _a : Tuple = self._buckets[ind] if not stored: _a : Any = _Item(UpperCAmelCase__ , UpperCAmelCase__ ) self._len += 1 return True elif stored.key == key: _a : List[Any] = _Item(UpperCAmelCase__ , UpperCAmelCase__ ) return True else: return False def _lowercase ( self : int ) -> bool: _a : Optional[int] = len(self._buckets ) * self._capacity_factor return len(self ) >= int(UpperCAmelCase__ ) def _lowercase ( self : Union[str, Any] ) -> bool: if len(self._buckets ) <= self._initial_block_size: return False _a : Union[str, Any] = len(self._buckets ) * self._capacity_factor / 2 return len(self ) < limit def _lowercase ( self : List[str] , UpperCAmelCase__ : int ) -> None: _a : Any = self._buckets _a : Union[str, Any] = [None] * new_size _a : str = 0 for item in old_buckets: if item: self._add_item(item.key , item.val ) def _lowercase ( self : int ) -> None: self._resize(len(self._buckets ) * 2 ) def _lowercase ( self : Dict ) -> None: self._resize(len(self._buckets ) // 2 ) def _lowercase ( self : Dict , UpperCAmelCase__ : KEY ) -> Iterator[int]: _a : str = self._get_bucket_index(UpperCAmelCase__ ) for _ in range(len(self._buckets ) ): yield ind _a : Union[str, Any] = self._get_next_ind(UpperCAmelCase__ ) def _lowercase ( self : Any , UpperCAmelCase__ : KEY , UpperCAmelCase__ : VAL ) -> None: for ind in self._iterate_buckets(UpperCAmelCase__ ): if self._try_set(UpperCAmelCase__ , UpperCAmelCase__ , UpperCAmelCase__ ): break def __setitem__( self : Union[str, Any] , UpperCAmelCase__ : KEY , UpperCAmelCase__ : VAL ) -> None: if self._is_full(): self._size_up() self._add_item(UpperCAmelCase__ , UpperCAmelCase__ ) def __delitem__( self : List[str] , UpperCAmelCase__ : KEY ) -> None: for ind in self._iterate_buckets(UpperCAmelCase__ ): _a : Union[str, Any] = self._buckets[ind] if item is None: raise KeyError(UpperCAmelCase__ ) if item is _deleted: continue if item.key == key: _a : Any = _deleted self._len -= 1 break if self._is_sparse(): self._size_down() def __getitem__( self : Any , UpperCAmelCase__ : KEY ) -> VAL: for ind in self._iterate_buckets(UpperCAmelCase__ ): _a : List[Any] = self._buckets[ind] if item is None: break if item is _deleted: continue if item.key == key: return item.val raise KeyError(UpperCAmelCase__ ) def __len__( self : int ) -> int: return self._len def __iter__( self : int ) -> Iterator[KEY]: yield from (item.key for item in self._buckets if item) def __repr__( self : Tuple ) -> str: _a : int = """ ,""".join( f"""{item.key}: {item.val}""" for item in self._buckets if item ) return f"""HashMap({val_string})"""
294
"""simple docstring""" import cva import numpy as np class UpperCamelCase : def __init__( self : Optional[int] , UpperCAmelCase__ : float , UpperCAmelCase__ : int ) -> Dict: if k in (0.0_4, 0.0_6): _a : List[str] = k _a : List[Any] = window_size else: raise ValueError("""invalid k value""" ) def __str__( self : Dict ) -> str: return str(self.k ) def _lowercase ( self : int , UpperCAmelCase__ : str ) -> tuple[cva.Mat, list[list[int]]]: _a : Dict = cva.imread(UpperCAmelCase__ , 0 ) _a , _a : List[Any] = img.shape _a : list[list[int]] = [] _a : List[Any] = img.copy() _a : int = cva.cvtColor(UpperCAmelCase__ , cva.COLOR_GRAY2RGB ) _a , _a : Any = np.gradient(UpperCAmelCase__ ) _a : Tuple = dx**2 _a : Union[str, Any] = dy**2 _a : Union[str, Any] = dx * dy _a : int = 0.0_4 _a : List[str] = self.window_size // 2 for y in range(UpperCAmelCase__ , h - offset ): for x in range(UpperCAmelCase__ , w - offset ): _a : str = ixx[ y - offset : y + offset + 1, x - offset : x + offset + 1 ].sum() _a : List[Any] = iyy[ y - offset : y + offset + 1, x - offset : x + offset + 1 ].sum() _a : Tuple = ixy[ y - offset : y + offset + 1, x - offset : x + offset + 1 ].sum() _a : Any = (wxx * wyy) - (wxy**2) _a : Tuple = wxx + wyy _a : Any = det - k * (trace**2) # Can change the value if r > 0.5: corner_list.append([x, y, r] ) color_img.itemset((y, x, 0) , 0 ) color_img.itemset((y, x, 1) , 0 ) color_img.itemset((y, x, 2) , 255 ) return color_img, corner_list if __name__ == "__main__": _snake_case = HarrisCorner(0.04, 3) _snake_case , _snake_case = edge_detect.detect('path_to_image') cva.imwrite('detect.png', color_img)
294
1
"""simple docstring""" 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`')
365
import baseaa def SCREAMING_SNAKE_CASE ( __UpperCamelCase : str ) -> bytes: return baseaa.baaencode(string.encode('''utf-8''' ) ) def SCREAMING_SNAKE_CASE ( __UpperCamelCase : bytes ) -> str: return baseaa.baadecode(__UpperCamelCase ).decode('''utf-8''' ) if __name__ == "__main__": _lowerCamelCase = 'Hello World!' _lowerCamelCase = baseaa_encode(test) print(encoded) _lowerCamelCase = baseaa_decode(encoded) print(decoded)
177
0
from __future__ import annotations def lowerCamelCase__ ( __lowerCamelCase : Optional[int] , __lowerCamelCase : str , __lowerCamelCase : Optional[int] ): 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 lowerCamelCase__ ( __lowerCamelCase : List[str] , __lowerCamelCase : int , __lowerCamelCase : Tuple , ): 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 lowerCamelCase__ ( __lowerCamelCase : Union[str, Any] , __lowerCamelCase : Any , __lowerCamelCase : Union[str, Any] , ): 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( a__ , nominal_annual_percentage_rate / 365 , number_of_years * 365 ) if __name__ == "__main__": import doctest doctest.testmod()
114
import gc import unittest import numpy as np import torch from transformers import CLIPTextConfig, CLIPTextModel, CLIPTokenizer from diffusers import ( AutoencoderKL, DDIMScheduler, PNDMScheduler, StableDiffusionLDMaDPipeline, UNetaDConditionModel, ) from diffusers.utils import nightly, slow, torch_device from diffusers.utils.testing_utils import enable_full_determinism, require_torch_gpu from ..pipeline_params import TEXT_TO_IMAGE_BATCH_PARAMS, TEXT_TO_IMAGE_IMAGE_PARAMS, TEXT_TO_IMAGE_PARAMS enable_full_determinism() class A__(unittest.TestCase ): """simple docstring""" _A : List[str] = StableDiffusionLDMaDPipeline _A : int = TEXT_TO_IMAGE_PARAMS _A : Dict = TEXT_TO_IMAGE_BATCH_PARAMS _A : str = TEXT_TO_IMAGE_IMAGE_PARAMS def UpperCamelCase__ ( self ) -> Union[str, Any]: torch.manual_seed(0 ) a_ : Tuple = 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 , ) a_ : List[Any] = DDIMScheduler( beta_start=0.0_0_0_8_5 , beta_end=0.0_1_2 , beta_schedule="""scaled_linear""" , clip_sample=_lowercase , set_alpha_to_one=_lowercase , ) torch.manual_seed(0 ) a_ : List[str] = AutoencoderKL( block_out_channels=[32, 64] , in_channels=6 , out_channels=6 , down_block_types=["""DownEncoderBlock2D""", """DownEncoderBlock2D"""] , up_block_types=["""UpDecoderBlock2D""", """UpDecoderBlock2D"""] , latent_channels=4 , ) torch.manual_seed(0 ) a_ : Dict = 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=1_000 , ) a_ : Tuple = CLIPTextModel(_lowercase ) a_ : Tuple = CLIPTokenizer.from_pretrained("""hf-internal-testing/tiny-random-clip""" ) a_ : Any = { """unet""": unet, """scheduler""": scheduler, """vae""": vae, """text_encoder""": text_encoder, """tokenizer""": tokenizer, """safety_checker""": None, """feature_extractor""": None, } return components def UpperCamelCase__ ( self , _lowercase , _lowercase=0 ) -> Any: if str(_lowercase ).startswith("""mps""" ): a_ : Optional[Any] = torch.manual_seed(_lowercase ) else: a_ : Optional[Any] = torch.Generator(device=_lowercase ).manual_seed(_lowercase ) a_ : Optional[Any] = { """prompt""": """A painting of a squirrel eating a burger""", """generator""": generator, """num_inference_steps""": 2, """guidance_scale""": 6.0, """output_type""": """numpy""", } return inputs def UpperCamelCase__ ( self ) -> List[Any]: a_ : str = """cpu""" # ensure determinism for the device-dependent torch.Generator a_ : Any = self.get_dummy_components() a_ : List[str] = StableDiffusionLDMaDPipeline(**_lowercase ) a_ : Union[str, Any] = ldmad_pipe.to(_lowercase ) ldmad_pipe.set_progress_bar_config(disable=_lowercase ) a_ : int = self.get_dummy_inputs(_lowercase ) a_ : List[Any] = ldmad_pipe(**_lowercase ) a_ , a_ : Tuple = output.rgb, output.depth a_ : Union[str, Any] = rgb[0, -3:, -3:, -1] a_ : Any = depth[0, -3:, -1] assert rgb.shape == (1, 64, 64, 3) assert depth.shape == (1, 64, 64) a_ : Optional[Any] = np.array( [0.3_7_3_3_8_1_7_6, 0.7_0_2_4_7, 0.7_4_2_0_3_1_9_3, 0.5_1_6_4_3_6_0_4, 0.5_8_2_5_6_7_9_3, 0.6_0_9_3_2_1_3_6, 0.4_1_8_1_0_9_5, 0.4_8_3_5_5_8_7_7, 0.4_6_5_3_5_2_6_2] ) a_ : int = np.array([1_0_3.4_6_7_2_7, 8_5.8_1_2_0_0_4, 8_7.8_4_9_2_3_6] ) assert np.abs(image_slice_rgb.flatten() - expected_slice_rgb ).max() < 1e-2 assert np.abs(image_slice_depth.flatten() - expected_slice_depth ).max() < 1e-2 def UpperCamelCase__ ( self ) -> Optional[Any]: a_ : Tuple = self.get_dummy_components() a_ : Optional[int] = StableDiffusionLDMaDPipeline(**_lowercase ) a_ : Optional[Any] = ldmad_pipe.to(_lowercase ) ldmad_pipe.set_progress_bar_config(disable=_lowercase ) a_ : Dict = self.get_dummy_inputs(_lowercase ) a_ : List[str] = 3 * [inputs["""prompt"""]] # forward a_ : Optional[int] = ldmad_pipe(**_lowercase ) a_ , a_ : Any = output.rgb, output.depth a_ : Union[str, Any] = rgb_slice_a[0, -3:, -3:, -1] a_ : Union[str, Any] = depth_slice_a[0, -3:, -1] a_ : Dict = self.get_dummy_inputs(_lowercase ) a_ : List[str] = 3 * [inputs.pop("""prompt""" )] a_ : List[Any] = ldmad_pipe.tokenizer( _lowercase , padding="""max_length""" , max_length=ldmad_pipe.tokenizer.model_max_length , truncation=_lowercase , return_tensors="""pt""" , ) a_ : int = text_inputs["""input_ids"""].to(_lowercase ) a_ : Any = ldmad_pipe.text_encoder(_lowercase )[0] a_ : Dict = prompt_embeds # forward a_ : int = ldmad_pipe(**_lowercase ) a_ , a_ : Optional[int] = output.rgb, output.depth a_ : List[str] = rgb_slice_a[0, -3:, -3:, -1] a_ : Tuple = depth_slice_a[0, -3:, -1] assert np.abs(rgb_slice_a.flatten() - rgb_slice_a.flatten() ).max() < 1e-4 assert np.abs(depth_slice_a.flatten() - depth_slice_a.flatten() ).max() < 1e-4 def UpperCamelCase__ ( self ) -> Dict: a_ : int = """cpu""" # ensure determinism for the device-dependent torch.Generator a_ : Dict = self.get_dummy_components() a_ : Any = PNDMScheduler(skip_prk_steps=_lowercase ) a_ : List[str] = StableDiffusionLDMaDPipeline(**_lowercase ) a_ : str = ldmad_pipe.to(_lowercase ) ldmad_pipe.set_progress_bar_config(disable=_lowercase ) a_ : List[Any] = self.get_dummy_inputs(_lowercase ) a_ : int = """french fries""" a_ : Any = ldmad_pipe(**_lowercase , negative_prompt=_lowercase ) a_ , a_ : Optional[Any] = output.rgb, output.depth a_ : Tuple = rgb[0, -3:, -3:, -1] a_ : Union[str, Any] = depth[0, -3:, -1] assert rgb.shape == (1, 64, 64, 3) assert depth.shape == (1, 64, 64) a_ : Optional[int] = np.array( [0.3_7_0_4_4, 0.7_1_8_1_1_5_0_3, 0.7_2_2_3_2_5_1, 0.4_8_6_0_3_6_7_5, 0.5_6_3_8_3_9_1, 0.6_3_6_4_9_4_8, 0.4_2_8_3_3_7_0_4, 0.4_9_0_1_3_1_5, 0.4_7_9_2_6_2_1_7] ) a_ : Union[str, Any] = np.array([1_0_7.8_4_7_3_8, 8_4.6_2_8_0_2, 8_9.9_6_2_1_3_5] ) assert np.abs(rgb_slice.flatten() - expected_slice_rgb ).max() < 1e-2 assert np.abs(depth_slice.flatten() - expected_slice_depth ).max() < 1e-2 @slow @require_torch_gpu class A__(unittest.TestCase ): """simple docstring""" def UpperCamelCase__ ( self ) -> str: super().tearDown() gc.collect() torch.cuda.empty_cache() def UpperCamelCase__ ( self , _lowercase , _lowercase="cpu" , _lowercase=torch.floataa , _lowercase=0 ) -> List[str]: a_ : Union[str, Any] = torch.Generator(device=_lowercase ).manual_seed(_lowercase ) a_ : Dict = np.random.RandomState(_lowercase ).standard_normal((1, 4, 64, 64) ) a_ : Tuple = torch.from_numpy(_lowercase ).to(device=_lowercase , dtype=_lowercase ) a_ : Any = { """prompt""": """a photograph of an astronaut riding a horse""", """latents""": latents, """generator""": generator, """num_inference_steps""": 3, """guidance_scale""": 7.5, """output_type""": """numpy""", } return inputs def UpperCamelCase__ ( self ) -> Any: a_ : Tuple = StableDiffusionLDMaDPipeline.from_pretrained("""Intel/ldm3d""" ) a_ : str = ldmad_pipe.to(_lowercase ) ldmad_pipe.set_progress_bar_config(disable=_lowercase ) a_ : Dict = self.get_inputs(_lowercase ) a_ : Optional[Any] = ldmad_pipe(**_lowercase ) a_ , a_ : int = output.rgb, output.depth a_ : str = rgb[0, -3:, -3:, -1].flatten() a_ : Tuple = rgb[0, -3:, -1].flatten() assert rgb.shape == (1, 512, 512, 3) assert depth.shape == (1, 512, 512) a_ : Optional[int] = np.array( [0.5_3_8_0_5_4_6_5, 0.5_6_7_0_7_3_0_5, 0.5_4_8_6_5_1_5, 0.5_7_0_1_2_2_3_6, 0.5_8_1_4_5_1_1, 0.5_6_2_5_3_4_8_7, 0.5_4_8_4_3_0_1_4, 0.5_5_0_9_2_2_6_3, 0.6_4_5_9_7_0_6] ) a_ : Optional[int] = np.array( [0.9_2_6_3_7_8_1, 0.6_6_7_8_6_7_2, 0.5_4_8_6_5_1_5, 0.9_2_2_0_2_1_4_5, 0.6_7_8_3_1_1_3_5, 0.5_6_2_5_3_4_8_7, 0.9_2_4_1_6_9_4, 0.7_5_5_1_4_7_8, 0.6_4_5_9_7_0_6] ) assert np.abs(rgb_slice - expected_slice_rgb ).max() < 3e-3 assert np.abs(depth_slice - expected_slice_depth ).max() < 3e-3 @nightly @require_torch_gpu class A__(unittest.TestCase ): """simple docstring""" def UpperCamelCase__ ( self ) -> Optional[int]: super().tearDown() gc.collect() torch.cuda.empty_cache() def UpperCamelCase__ ( self , _lowercase , _lowercase="cpu" , _lowercase=torch.floataa , _lowercase=0 ) -> str: a_ : List[Any] = torch.Generator(device=_lowercase ).manual_seed(_lowercase ) a_ : Tuple = np.random.RandomState(_lowercase ).standard_normal((1, 4, 64, 64) ) a_ : Any = torch.from_numpy(_lowercase ).to(device=_lowercase , dtype=_lowercase ) a_ : Dict = { """prompt""": """a photograph of an astronaut riding a horse""", """latents""": latents, """generator""": generator, """num_inference_steps""": 50, """guidance_scale""": 7.5, """output_type""": """numpy""", } return inputs def UpperCamelCase__ ( self ) -> Optional[Any]: a_ : Tuple = StableDiffusionLDMaDPipeline.from_pretrained("""Intel/ldm3d""" ).to(_lowercase ) ldmad_pipe.set_progress_bar_config(disable=_lowercase ) a_ : List[str] = self.get_inputs(_lowercase ) a_ : Union[str, Any] = ldmad_pipe(**_lowercase ) a_ , a_ : str = output.rgb, output.depth a_ : List[str] = 0.4_9_5_5_8_6 a_ : int = 0.3_3_7_9_5_5_1_5 a_ : int = 1_1_2.4_8_5_1_8 a_ : Optional[int] = 9_8.4_8_9_7_4_6 assert np.abs(expected_rgb_mean - rgb.mean() ) < 1e-3 assert np.abs(expected_rgb_std - rgb.std() ) < 1e-3 assert np.abs(expected_depth_mean - depth.mean() ) < 1e-3 assert np.abs(expected_depth_std - depth.std() ) < 1e-3 def UpperCamelCase__ ( self ) -> Optional[int]: a_ : Optional[Any] = StableDiffusionLDMaDPipeline.from_pretrained("""Intel/ldm3d-4c""" ).to(_lowercase ) ldmad_pipe.set_progress_bar_config(disable=_lowercase ) a_ : List[str] = self.get_inputs(_lowercase ) a_ : List[Any] = ldmad_pipe(**_lowercase ) a_ , a_ : List[Any] = output.rgb, output.depth a_ : int = 0.4_1_9_4_1_2_7 a_ : List[str] = 0.3_5_3_7_5_5_8_6 a_ : Optional[int] = 0.5_6_3_8_5_0_2 a_ : str = 0.3_4_6_8_6_1_0_3 assert rgb.shape == (1, 512, 512, 3) assert depth.shape == (1, 512, 512, 1) assert np.abs(expected_rgb_mean - rgb.mean() ) < 1e-3 assert np.abs(expected_rgb_std - rgb.std() ) < 1e-3 assert np.abs(expected_depth_mean - depth.mean() ) < 1e-3 assert np.abs(expected_depth_std - depth.std() ) < 1e-3
248
0
import os import numpy import onnx def SCREAMING_SNAKE_CASE__ ( lowerCamelCase__ , lowerCamelCase__ ) -> Optional[int]: __lowerCamelCase : List[Any] = a.name __lowerCamelCase : Union[str, Any] = b.name __lowerCamelCase : str = '' __lowerCamelCase : str = '' __lowerCamelCase : Any = a == b __lowerCamelCase : Tuple = name_a __lowerCamelCase : Tuple = name_b return res def SCREAMING_SNAKE_CASE__ ( lowerCamelCase__ , lowerCamelCase__ , lowerCamelCase__ ) -> Any: for i, input_name in enumerate(node_proto.input ): if input_name == name: node_proto.input.insert(lowerCamelCase__ , lowerCamelCase__ ) node_proto.input.pop(i + 1 ) if node_proto.op_type == "If": _graph_replace_input_with(node_proto.attribute[0].g , lowerCamelCase__ , lowerCamelCase__ ) _graph_replace_input_with(node_proto.attribute[1].g , lowerCamelCase__ , lowerCamelCase__ ) if node_proto.op_type == "Loop": _graph_replace_input_with(node_proto.attribute[0].g , lowerCamelCase__ , lowerCamelCase__ ) def SCREAMING_SNAKE_CASE__ ( lowerCamelCase__ , lowerCamelCase__ , lowerCamelCase__ ) -> str: for n in graph_proto.node: _node_replace_input_with(lowerCamelCase__ , lowerCamelCase__ , lowerCamelCase__ ) def SCREAMING_SNAKE_CASE__ ( lowerCamelCase__ , lowerCamelCase__ , lowerCamelCase__ ) -> int: __lowerCamelCase : Union[str, Any] = list(model.graph.initializer ) __lowerCamelCase : Union[str, Any] = list(model_without_ext.graph.initializer ) for i, ref_i in ind_to_replace: assert inits_with_data[i].name == inits[i].name assert inits_with_data[ref_i].name == inits[ref_i].name assert i > ref_i __lowerCamelCase : Optional[int] = inits[i].name __lowerCamelCase : List[Any] = inits[ref_i].name model_without_ext.graph.initializer.remove(inits[i] ) # for n in model.graph.node: _graph_replace_input_with(model_without_ext.graph , lowerCamelCase__ , lowerCamelCase__ ) def SCREAMING_SNAKE_CASE__ ( lowerCamelCase__ ) -> Optional[Any]: __lowerCamelCase : Any = os.path.dirname(lowerCamelCase__ ) __lowerCamelCase : List[Any] = os.path.basename(lowerCamelCase__ ) __lowerCamelCase : List[str] = onnx.load(os.path.join(lowerCamelCase__ , lowerCamelCase__ ) ) __lowerCamelCase : Any = list(model.graph.initializer ) __lowerCamelCase : int = set() __lowerCamelCase : Optional[int] = {} __lowerCamelCase : Dict = [] __lowerCamelCase : Optional[Any] = 0 for i in range(len(lowerCamelCase__ ) ): if i in dup_set: continue for j in range(i + 1 , len(lowerCamelCase__ ) ): if j in dup_set: continue if _is_equal_tensor_proto(inits[i] , inits[j] ): dup_set.add(lowerCamelCase__ ) dup_set.add(lowerCamelCase__ ) __lowerCamelCase : Optional[Any] = inits[j].data_type __lowerCamelCase : List[Any] = numpy.prod(inits[j].dims ) if dtype == 1: mem_size *= 4 elif dtype == 6: mem_size *= 4 elif dtype == 7 or dtype == 1_1: mem_size *= 8 else: print('unexpected data type: ' , lowerCamelCase__ ) total_reduced_size += mem_size __lowerCamelCase : Tuple = inits[i].name __lowerCamelCase : Optional[int] = inits[j].name if name_i in dup_map: dup_map[name_i].append(lowerCamelCase__ ) else: __lowerCamelCase : Tuple = [name_j] ind_to_replace.append((j, i) ) print('total reduced size: ' , total_reduced_size / 1_0_2_4 / 1_0_2_4 / 1_0_2_4 , 'GB' ) __lowerCamelCase : Tuple = sorted(lowerCamelCase__ ) _remove_dup_initializers_from_model(lowerCamelCase__ , lowerCamelCase__ , lowerCamelCase__ ) __lowerCamelCase : List[str] = 'optimized_' + model_file_name __lowerCamelCase : Dict = os.path.join(lowerCamelCase__ , lowerCamelCase__ ) onnx.save(lowerCamelCase__ , lowerCamelCase__ ) return new_model
113
from typing import TYPE_CHECKING from ...utils import ( OptionalDependencyNotAvailable, _LazyModule, is_flax_available, is_tf_available, is_tokenizers_available, is_torch_available, ) a ={ """configuration_roformer""": ["""ROFORMER_PRETRAINED_CONFIG_ARCHIVE_MAP""", """RoFormerConfig""", """RoFormerOnnxConfig"""], """tokenization_roformer""": ["""RoFormerTokenizer"""], } try: if not is_tokenizers_available(): raise OptionalDependencyNotAvailable() except OptionalDependencyNotAvailable: pass else: a =["""RoFormerTokenizerFast"""] try: if not is_torch_available(): raise OptionalDependencyNotAvailable() except OptionalDependencyNotAvailable: pass else: a =[ """ROFORMER_PRETRAINED_MODEL_ARCHIVE_LIST""", """RoFormerForCausalLM""", """RoFormerForMaskedLM""", """RoFormerForMultipleChoice""", """RoFormerForQuestionAnswering""", """RoFormerForSequenceClassification""", """RoFormerForTokenClassification""", """RoFormerLayer""", """RoFormerModel""", """RoFormerPreTrainedModel""", """load_tf_weights_in_roformer""", ] try: if not is_tf_available(): raise OptionalDependencyNotAvailable() except OptionalDependencyNotAvailable: pass else: a =[ """TF_ROFORMER_PRETRAINED_MODEL_ARCHIVE_LIST""", """TFRoFormerForCausalLM""", """TFRoFormerForMaskedLM""", """TFRoFormerForMultipleChoice""", """TFRoFormerForQuestionAnswering""", """TFRoFormerForSequenceClassification""", """TFRoFormerForTokenClassification""", """TFRoFormerLayer""", """TFRoFormerModel""", """TFRoFormerPreTrainedModel""", ] try: if not is_flax_available(): raise OptionalDependencyNotAvailable() except OptionalDependencyNotAvailable: pass else: a =[ """FLAX_ROFORMER_PRETRAINED_MODEL_ARCHIVE_LIST""", """FlaxRoFormerForMaskedLM""", """FlaxRoFormerForMultipleChoice""", """FlaxRoFormerForQuestionAnswering""", """FlaxRoFormerForSequenceClassification""", """FlaxRoFormerForTokenClassification""", """FlaxRoFormerModel""", """FlaxRoFormerPreTrainedModel""", ] if TYPE_CHECKING: from .configuration_roformer import ROFORMER_PRETRAINED_CONFIG_ARCHIVE_MAP, RoFormerConfig, RoFormerOnnxConfig from .tokenization_roformer import RoFormerTokenizer try: if not is_tokenizers_available(): raise OptionalDependencyNotAvailable() except OptionalDependencyNotAvailable: pass else: from .tokenization_roformer_fast import RoFormerTokenizerFast try: if not is_torch_available(): raise OptionalDependencyNotAvailable() except OptionalDependencyNotAvailable: pass else: from .modeling_roformer import ( ROFORMER_PRETRAINED_MODEL_ARCHIVE_LIST, RoFormerForCausalLM, RoFormerForMaskedLM, RoFormerForMultipleChoice, RoFormerForQuestionAnswering, RoFormerForSequenceClassification, RoFormerForTokenClassification, RoFormerLayer, RoFormerModel, RoFormerPreTrainedModel, load_tf_weights_in_roformer, ) try: if not is_tf_available(): raise OptionalDependencyNotAvailable() except OptionalDependencyNotAvailable: pass else: from .modeling_tf_roformer import ( TF_ROFORMER_PRETRAINED_MODEL_ARCHIVE_LIST, TFRoFormerForCausalLM, TFRoFormerForMaskedLM, TFRoFormerForMultipleChoice, TFRoFormerForQuestionAnswering, TFRoFormerForSequenceClassification, TFRoFormerForTokenClassification, TFRoFormerLayer, TFRoFormerModel, TFRoFormerPreTrainedModel, ) try: if not is_flax_available(): raise OptionalDependencyNotAvailable() except OptionalDependencyNotAvailable: pass else: from .modeling_flax_roformer import ( FLAX_ROFORMER_PRETRAINED_MODEL_ARCHIVE_LIST, FlaxRoFormerForMaskedLM, FlaxRoFormerForMultipleChoice, FlaxRoFormerForQuestionAnswering, FlaxRoFormerForSequenceClassification, FlaxRoFormerForTokenClassification, FlaxRoFormerModel, FlaxRoFormerPreTrainedModel, ) else: import sys a =_LazyModule(__name__, globals()["""__file__"""], _import_structure, module_spec=__spec__)
113
1
"""simple docstring""" from collections import defaultdict def lowerCAmelCase ( __UpperCamelCase ): """simple docstring""" __A = 1 __A = True for v in tree[start]: if v not in visited: ret += dfs(__UpperCamelCase ) if ret % 2 == 0: cuts.append(__UpperCamelCase ) return ret def lowerCAmelCase ( ): """simple docstring""" dfs(1 ) if __name__ == "__main__": lowercase_ , lowercase_ = 10, 9 lowercase_ = defaultdict(list) lowercase_ = {} lowercase_ = [] lowercase_ = 0 lowercase_ = [(2, 1), (3, 1), (4, 3), (5, 2), (6, 1), (7, 2), (8, 6), (9, 8), (10, 8)] for u, v in edges: tree[u].append(v) tree[v].append(u) even_tree() print(len(cuts) - 1)
266
"""simple docstring""" 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 lowercase_ = get_tests_dir('fixtures/test_sentencepiece_no_bos.model') @require_sentencepiece @require_tokenizers class snake_case ( _lowerCAmelCase , unittest.TestCase ): '''simple docstring''' A_ : Optional[Any] = PegasusTokenizer A_ : int = PegasusTokenizerFast A_ : Optional[Any] = True A_ : Union[str, Any] = True def _SCREAMING_SNAKE_CASE ( self : Union[str, Any] ): '''simple docstring''' super().setUp() # We have a SentencePiece fixture for testing __A = PegasusTokenizer(_lowerCamelCase ) tokenizer.save_pretrained(self.tmpdirname ) @cached_property def _SCREAMING_SNAKE_CASE ( self : Tuple ): '''simple docstring''' return PegasusTokenizer.from_pretrained('''google/pegasus-large''' ) def _SCREAMING_SNAKE_CASE ( self : int, **_lowerCamelCase : List[Any] ): '''simple docstring''' return PegasusTokenizer.from_pretrained(self.tmpdirname, **_lowerCamelCase ) def _SCREAMING_SNAKE_CASE ( self : Optional[Any], _lowerCamelCase : Dict ): '''simple docstring''' return ("This is a test", "This is a test") def _SCREAMING_SNAKE_CASE ( self : Tuple ): '''simple docstring''' __A = '''</s>''' __A = 1 self.assertEqual(self.get_tokenizer()._convert_token_to_id(_lowerCamelCase ), _lowerCamelCase ) self.assertEqual(self.get_tokenizer()._convert_id_to_token(_lowerCamelCase ), _lowerCamelCase ) def _SCREAMING_SNAKE_CASE ( self : int ): '''simple docstring''' __A = 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(_lowerCamelCase ), 11_03 ) def _SCREAMING_SNAKE_CASE ( self : int ): '''simple docstring''' self.assertEqual(self.get_tokenizer().vocab_size, 11_03 ) def _SCREAMING_SNAKE_CASE ( self : Dict ): '''simple docstring''' __A = self.rust_tokenizer_class.from_pretrained(self.tmpdirname ) __A = self.tokenizer_class.from_pretrained(self.tmpdirname ) __A = ( '''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 = rust_tokenizer([raw_input_str], return_tensors=_lowerCamelCase, add_special_tokens=_lowerCamelCase ).input_ids[0] __A = py_tokenizer([raw_input_str], return_tensors=_lowerCamelCase, add_special_tokens=_lowerCamelCase ).input_ids[0] self.assertListEqual(_lowerCamelCase, _lowerCamelCase ) def _SCREAMING_SNAKE_CASE ( self : List[Any] ): '''simple docstring''' __A = self._large_tokenizer # <mask_1> masks whole sentence while <mask_2> masks single word __A = '''<mask_1> To ensure a <mask_2> flow of bank resolutions.''' __A = [2, 4_13, 6_15, 1_14, 3, 19_71, 1_13, 16_79, 1_07_10, 1_07, 1] __A = tokenizer([raw_input_str], return_tensors=_lowerCamelCase ).input_ids[0] self.assertListEqual(_lowerCamelCase, _lowerCamelCase ) def _SCREAMING_SNAKE_CASE ( self : Optional[int] ): '''simple docstring''' __A = self._large_tokenizer # The tracebacks for the following asserts are **better** without messages or self.assertEqual assert tokenizer.vocab_size == 9_61_03 assert tokenizer.pad_token_id == 0 assert tokenizer.eos_token_id == 1 assert tokenizer.offset == 1_03 assert tokenizer.unk_token_id == tokenizer.offset + 2 == 1_05 assert tokenizer.unk_token == "<unk>" assert tokenizer.model_max_length == 10_24 __A = '''To ensure a smooth flow of bank resolutions.''' __A = [4_13, 6_15, 1_14, 22_91, 19_71, 1_13, 16_79, 1_07_10, 1_07, 1] __A = tokenizer([raw_input_str], return_tensors=_lowerCamelCase ).input_ids[0] self.assertListEqual(_lowerCamelCase, _lowerCamelCase ) assert tokenizer.convert_ids_to_tokens([0, 1, 2, 3] ) == ["<pad>", "</s>", "<mask_1>", "<mask_2>"] @require_torch def _SCREAMING_SNAKE_CASE ( self : Dict ): '''simple docstring''' __A = ['''This is going to be way too long.''' * 1_50, '''short example'''] __A = ['''not super long but more than 5 tokens''', '''tiny'''] __A = self._large_tokenizer(_lowerCamelCase, padding=_lowerCamelCase, truncation=_lowerCamelCase, return_tensors='''pt''' ) __A = self._large_tokenizer( text_target=_lowerCamelCase, max_length=5, padding=_lowerCamelCase, truncation=_lowerCamelCase, return_tensors='''pt''' ) assert batch.input_ids.shape == (2, 10_24) assert batch.attention_mask.shape == (2, 10_24) assert targets["input_ids"].shape == (2, 5) assert len(_lowerCamelCase ) == 2 # input_ids, attention_mask. @slow def _SCREAMING_SNAKE_CASE ( self : Tuple ): '''simple docstring''' # fmt: off __A = {'''input_ids''': [[3_89_79, 1_43, 1_84_85, 6_06, 1_30, 2_66_69, 8_76_86, 1_21, 5_41_89, 11_29, 1_11, 2_66_69, 8_76_86, 1_21, 91_14, 1_47_87, 1_21, 1_32_49, 1_58, 5_92, 9_56, 1_21, 1_46_21, 3_15_76, 1_43, 6_26_13, 1_08, 96_88, 9_30, 4_34_30, 1_15_62, 6_26_13, 3_04, 1_08, 1_14_43, 8_97, 1_08, 93_14, 1_74_15, 6_33_99, 1_08, 1_14_43, 76_14, 1_83_16, 1_18, 42_84, 71_48, 1_24_30, 1_43, 14_00, 2_57_03, 1_58, 1_11, 42_84, 71_48, 1_17_72, 1_43, 2_12_97, 10_64, 1_58, 1_22, 2_04, 35_06, 17_54, 11_33, 1_47_87, 15_81, 1_15, 3_32_24, 44_82, 1_11, 13_55, 1_10, 2_91_73, 3_17, 5_08_33, 1_08, 2_01_47, 9_46_65, 1_11, 7_71_98, 1_07, 1], [1_10, 6_26_13, 1_17, 6_38, 1_12, 11_33, 1_21, 2_00_98, 13_55, 7_90_50, 1_38_72, 1_35, 15_96, 5_35_41, 13_52, 1_41, 1_30_39, 55_42, 1_24, 3_02, 5_18, 1_11, 2_68, 29_56, 1_15, 1_49, 44_27, 1_07, 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_39, 12_35, 27_99, 1_82_89, 1_77_80, 2_04, 1_09, 94_74, 12_96, 1_07, 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=_lowerCamelCase, model_name='''google/bigbird-pegasus-large-arxiv''', revision='''ba85d0851d708441f91440d509690f1ab6353415''', ) @require_sentencepiece @require_tokenizers class snake_case ( _lowerCAmelCase , unittest.TestCase ): '''simple docstring''' A_ : str = PegasusTokenizer A_ : Union[str, Any] = PegasusTokenizerFast A_ : Any = True A_ : str = True def _SCREAMING_SNAKE_CASE ( self : List[Any] ): '''simple docstring''' super().setUp() # We have a SentencePiece fixture for testing __A = PegasusTokenizer(_lowerCamelCase, offset=0, mask_token_sent=_lowerCamelCase, mask_token='''[MASK]''' ) tokenizer.save_pretrained(self.tmpdirname ) @cached_property def _SCREAMING_SNAKE_CASE ( self : str ): '''simple docstring''' return PegasusTokenizer.from_pretrained('''google/bigbird-pegasus-large-arxiv''' ) def _SCREAMING_SNAKE_CASE ( self : Optional[int], **_lowerCamelCase : Dict ): '''simple docstring''' return PegasusTokenizer.from_pretrained(self.tmpdirname, **_lowerCamelCase ) def _SCREAMING_SNAKE_CASE ( self : List[str], _lowerCamelCase : List[str] ): '''simple docstring''' return ("This is a test", "This is a test") def _SCREAMING_SNAKE_CASE ( self : List[Any] ): '''simple docstring''' __A = self.rust_tokenizer_class.from_pretrained(self.tmpdirname ) __A = self.tokenizer_class.from_pretrained(self.tmpdirname ) __A = ( '''Let\'s see which <unk> is the better <unk_token> one [MASK] It seems like this [MASK] was important </s>''' ''' <pad> <pad> <pad>''' ) __A = rust_tokenizer([raw_input_str], return_tensors=_lowerCamelCase, add_special_tokens=_lowerCamelCase ).input_ids[0] __A = py_tokenizer([raw_input_str], return_tensors=_lowerCamelCase, add_special_tokens=_lowerCamelCase ).input_ids[0] self.assertListEqual(_lowerCamelCase, _lowerCamelCase ) @require_torch def _SCREAMING_SNAKE_CASE ( self : Tuple ): '''simple docstring''' __A = ['''This is going to be way too long.''' * 10_00, '''short example'''] __A = ['''not super long but more than 5 tokens''', '''tiny'''] __A = self._large_tokenizer(_lowerCamelCase, padding=_lowerCamelCase, truncation=_lowerCamelCase, return_tensors='''pt''' ) __A = self._large_tokenizer( text_target=_lowerCamelCase, max_length=5, padding=_lowerCamelCase, truncation=_lowerCamelCase, return_tensors='''pt''' ) assert batch.input_ids.shape == (2, 40_96) assert batch.attention_mask.shape == (2, 40_96) assert targets["input_ids"].shape == (2, 5) assert len(_lowerCamelCase ) == 2 # input_ids, attention_mask. def _SCREAMING_SNAKE_CASE ( self : Tuple ): '''simple docstring''' __A = ( '''This is an example string that is used to test the original TF implementation against the HF''' ''' implementation''' ) __A = self._large_tokenizer(_lowerCamelCase ).input_ids self.assertListEqual( _lowerCamelCase, [1_82, 1_17, 1_42, 5_87, 42_11, 1_20, 1_17, 2_63, 1_12, 8_04, 1_09, 8_56, 2_50_16, 31_37, 4_64, 1_09, 2_69_55, 31_37, 1], )
266
1
import inspect import unittest from transformers import ConvNextVaConfig from transformers.models.auto import get_values from transformers.models.auto.modeling_auto import MODEL_FOR_BACKBONE_MAPPING_NAMES, MODEL_MAPPING_NAMES 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 transformers import ConvNextVaBackbone, ConvNextVaForImageClassification, ConvNextVaModel from transformers.models.convnextva.modeling_convnextva import CONVNEXTV2_PRETRAINED_MODEL_ARCHIVE_LIST if is_vision_available(): from PIL import Image from transformers import AutoImageProcessor class lowerCamelCase_ : '''simple docstring''' def __init__( self : List[Any] , __lowerCamelCase : str , __lowerCamelCase : List[Any]=13 , __lowerCamelCase : Union[str, Any]=32 , __lowerCamelCase : Union[str, Any]=3 , __lowerCamelCase : List[str]=4 , __lowerCamelCase : Dict=[10, 20, 30, 40] , __lowerCamelCase : List[str]=[2, 2, 3, 2] , __lowerCamelCase : Union[str, Any]=True , __lowerCamelCase : List[str]=True , __lowerCamelCase : List[Any]=37 , __lowerCamelCase : Union[str, Any]="gelu" , __lowerCamelCase : List[Any]=10 , __lowerCamelCase : int=0.02 , __lowerCamelCase : Optional[Any]=["stage2", "stage3", "stage4"] , __lowerCamelCase : Dict=[2, 3, 4] , __lowerCamelCase : Any=None , ) -> int: A : Union[str, Any] = parent A : str = batch_size A : Any = image_size A : Optional[Any] = num_channels A : str = num_stages A : List[str] = hidden_sizes A : Tuple = depths A : Tuple = is_training A : int = use_labels A : List[str] = intermediate_size A : Union[str, Any] = hidden_act A : int = num_labels A : Dict = initializer_range A : List[str] = out_features A : Tuple = out_indices A : List[Any] = scope def SCREAMING_SNAKE_CASE__ ( self : int ) -> Optional[int]: A : str = floats_tensor([self.batch_size, self.num_channels, self.image_size, self.image_size] ) A : Any = None if self.use_labels: A : Any = ids_tensor([self.batch_size] , self.num_labels ) A : Dict = self.get_config() return config, pixel_values, labels def SCREAMING_SNAKE_CASE__ ( self : Union[str, Any] ) -> int: return ConvNextVaConfig( num_channels=self.num_channels , hidden_sizes=self.hidden_sizes , depths=self.depths , num_stages=self.num_stages , hidden_act=self.hidden_act , is_decoder=__lowerCamelCase , initializer_range=self.initializer_range , out_features=self.out_features , out_indices=self.out_indices , num_labels=self.num_labels , ) def SCREAMING_SNAKE_CASE__ ( self : Optional[Any] , __lowerCamelCase : List[str] , __lowerCamelCase : int , __lowerCamelCase : Optional[int] ) -> List[Any]: A : Tuple = ConvNextVaModel(config=__lowerCamelCase ) model.to(__lowerCamelCase ) model.eval() A : Optional[int] = model(__lowerCamelCase ) # expected last hidden states: B, C, H // 32, W // 32 self.parent.assertEqual( result.last_hidden_state.shape , (self.batch_size, self.hidden_sizes[-1], self.image_size // 32, self.image_size // 32) , ) def SCREAMING_SNAKE_CASE__ ( self : Optional[Any] , __lowerCamelCase : List[str] , __lowerCamelCase : Tuple , __lowerCamelCase : List[str] ) -> Optional[int]: A : Optional[int] = ConvNextVaForImageClassification(__lowerCamelCase ) model.to(__lowerCamelCase ) model.eval() A : Any = model(__lowerCamelCase , labels=__lowerCamelCase ) self.parent.assertEqual(result.logits.shape , (self.batch_size, self.num_labels) ) def SCREAMING_SNAKE_CASE__ ( self : List[str] , __lowerCamelCase : Union[str, Any] , __lowerCamelCase : str , __lowerCamelCase : Any ) -> List[str]: A : List[str] = ConvNextVaBackbone(config=__lowerCamelCase ) model.to(__lowerCamelCase ) model.eval() A : Optional[Any] = model(__lowerCamelCase ) # verify hidden states self.parent.assertEqual(len(result.feature_maps ) , len(config.out_features ) ) self.parent.assertListEqual(list(result.feature_maps[0].shape ) , [self.batch_size, self.hidden_sizes[1], 4, 4] ) # verify channels self.parent.assertEqual(len(model.channels ) , len(config.out_features ) ) self.parent.assertListEqual(model.channels , config.hidden_sizes[1:] ) # verify backbone works with out_features=None A : Optional[Any] = None A : List[str] = ConvNextVaBackbone(config=__lowerCamelCase ) model.to(__lowerCamelCase ) model.eval() A : Any = model(__lowerCamelCase ) # verify feature maps self.parent.assertEqual(len(result.feature_maps ) , 1 ) self.parent.assertListEqual(list(result.feature_maps[0].shape ) , [self.batch_size, self.hidden_sizes[-1], 1, 1] ) # verify channels self.parent.assertEqual(len(model.channels ) , 1 ) self.parent.assertListEqual(model.channels , [config.hidden_sizes[-1]] ) def SCREAMING_SNAKE_CASE__ ( self : Dict ) -> Union[str, Any]: A : str = self.prepare_config_and_inputs() A , A , A : Optional[int] = config_and_inputs A : Tuple = {"pixel_values": pixel_values} return config, inputs_dict def SCREAMING_SNAKE_CASE__ ( self : Optional[Any] ) -> int: A : Optional[Any] = self.prepare_config_and_inputs() A , A , A : Tuple = config_and_inputs A : Optional[int] = {"pixel_values": pixel_values, "labels": labels} return config, inputs_dict @require_torch class lowerCamelCase_ ( _A ,_A ,unittest.TestCase ): '''simple docstring''' a__ = ( ( ConvNextVaModel, ConvNextVaForImageClassification, ConvNextVaBackbone, ) if is_torch_available() else () ) a__ = ( {"feature-extraction": ConvNextVaModel, "image-classification": ConvNextVaForImageClassification} if is_torch_available() else {} ) a__ = False a__ = False a__ = False a__ = False a__ = False def SCREAMING_SNAKE_CASE__ ( self : Optional[Any] ) -> Any: A : str = ConvNextVaModelTester(self ) A : int = ConfigTester(self , config_class=__lowerCamelCase , has_text_modality=__lowerCamelCase , hidden_size=37 ) def SCREAMING_SNAKE_CASE__ ( self : Optional[int] ) -> List[Any]: self.create_and_test_config_common_properties() self.config_tester.create_and_test_config_to_json_string() self.config_tester.create_and_test_config_to_json_file() self.config_tester.create_and_test_config_from_and_save_pretrained() self.config_tester.create_and_test_config_with_num_labels() self.config_tester.check_config_can_be_init_without_params() self.config_tester.check_config_arguments_init() def SCREAMING_SNAKE_CASE__ ( self : Dict ) -> List[str]: return @unittest.skip(reason="ConvNextV2 does not use inputs_embeds" ) def SCREAMING_SNAKE_CASE__ ( self : Union[str, Any] ) -> Optional[Any]: pass @unittest.skip(reason="ConvNextV2 does not support input and output embeddings" ) def SCREAMING_SNAKE_CASE__ ( self : Dict ) -> Dict: pass @unittest.skip(reason="ConvNextV2 does not use feedforward chunking" ) def SCREAMING_SNAKE_CASE__ ( self : Union[str, Any] ) -> Dict: pass def SCREAMING_SNAKE_CASE__ ( self : str ) -> Any: if not self.model_tester.is_training: return for model_class in self.all_model_classes: A , A : str = self.model_tester.prepare_config_and_inputs_with_labels() A : List[str] = True if model_class.__name__ in [ *get_values(__lowerCamelCase ), *get_values(__lowerCamelCase ), ]: continue A : str = model_class(__lowerCamelCase ) model.to(__lowerCamelCase ) model.train() A : Tuple = self._prepare_for_class(__lowerCamelCase , __lowerCamelCase , return_labels=__lowerCamelCase ) A : Union[str, Any] = model(**__lowerCamelCase ).loss loss.backward() def SCREAMING_SNAKE_CASE__ ( self : List[Any] ) -> List[Any]: if not self.model_tester.is_training: return for model_class in self.all_model_classes: A , A : List[Any] = self.model_tester.prepare_config_and_inputs_with_labels() A : Any = False A : Optional[Any] = True if ( model_class.__name__ in [*get_values(__lowerCamelCase ), *get_values(__lowerCamelCase )] or not model_class.supports_gradient_checkpointing ): continue A : Union[str, Any] = model_class(__lowerCamelCase ) model.to(__lowerCamelCase ) model.gradient_checkpointing_enable() model.train() A : int = self._prepare_for_class(__lowerCamelCase , __lowerCamelCase , return_labels=__lowerCamelCase ) A : Tuple = model(**__lowerCamelCase ).loss loss.backward() def SCREAMING_SNAKE_CASE__ ( self : int ) -> Union[str, Any]: A , A : Any = self.model_tester.prepare_config_and_inputs_for_common() for model_class in self.all_model_classes: A : Optional[int] = model_class(__lowerCamelCase ) A : Union[str, Any] = inspect.signature(model.forward ) # signature.parameters is an OrderedDict => so arg_names order is deterministic A : int = [*signature.parameters.keys()] A : Optional[int] = ["pixel_values"] self.assertListEqual(arg_names[:1] , __lowerCamelCase ) def SCREAMING_SNAKE_CASE__ ( self : int ) -> Dict: A : str = self.model_tester.prepare_config_and_inputs() self.model_tester.create_and_check_model(*__lowerCamelCase ) def SCREAMING_SNAKE_CASE__ ( self : Optional[Any] ) -> Union[str, Any]: def check_hidden_states_output(__lowerCamelCase : int , __lowerCamelCase : Union[str, Any] , __lowerCamelCase : str ): A : int = model_class(__lowerCamelCase ) model.to(__lowerCamelCase ) model.eval() with torch.no_grad(): A : List[str] = model(**self._prepare_for_class(__lowerCamelCase , __lowerCamelCase ) ) A : List[str] = outputs.encoder_hidden_states if config.is_encoder_decoder else outputs.hidden_states A : str = self.model_tester.num_stages self.assertEqual(len(__lowerCamelCase ) , expected_num_stages + 1 ) # ConvNextV2's feature maps are of shape (batch_size, num_channels, height, width) self.assertListEqual( list(hidden_states[0].shape[-2:] ) , [self.model_tester.image_size // 4, self.model_tester.image_size // 4] , ) A , A : Dict = self.model_tester.prepare_config_and_inputs_for_common() for model_class in self.all_model_classes: A : Optional[int] = True check_hidden_states_output(__lowerCamelCase , __lowerCamelCase , __lowerCamelCase ) # check that output_hidden_states also work using config del inputs_dict["output_hidden_states"] A : List[Any] = True check_hidden_states_output(__lowerCamelCase , __lowerCamelCase , __lowerCamelCase ) def SCREAMING_SNAKE_CASE__ ( self : Dict ) -> List[str]: A : Tuple = self.model_tester.prepare_config_and_inputs() self.model_tester.create_and_check_for_image_classification(*__lowerCamelCase ) @slow def SCREAMING_SNAKE_CASE__ ( self : int ) -> Dict: for model_name in CONVNEXTV2_PRETRAINED_MODEL_ARCHIVE_LIST[:1]: A : Dict = ConvNextVaModel.from_pretrained(__lowerCamelCase ) self.assertIsNotNone(__lowerCamelCase ) def UpperCAmelCase ( ): A : List[str] = Image.open("./tests/fixtures/tests_samples/COCO/000000039769.png" ) return image @require_torch @require_vision class lowerCamelCase_ ( unittest.TestCase ): '''simple docstring''' @cached_property def SCREAMING_SNAKE_CASE__ ( self : Dict ) -> List[str]: return AutoImageProcessor.from_pretrained("facebook/convnextv2-tiny-1k-224" ) if is_vision_available() else None @slow def SCREAMING_SNAKE_CASE__ ( self : int ) -> int: A : Optional[int] = ConvNextVaForImageClassification.from_pretrained("facebook/convnextv2-tiny-1k-224" ).to(__lowerCamelCase ) A : Union[str, Any] = self.default_image_processor A : Dict = prepare_img() A : int = preprocessor(images=__lowerCamelCase , return_tensors="pt" ).to(__lowerCamelCase ) # forward pass with torch.no_grad(): A : Optional[Any] = model(**__lowerCamelCase ) # verify the logits A : List[str] = torch.Size((1, 10_00) ) self.assertEqual(outputs.logits.shape , __lowerCamelCase ) A : Any = torch.tensor([0.9996, 0.1966, -0.4386] ).to(__lowerCamelCase ) self.assertTrue(torch.allclose(outputs.logits[0, :3] , __lowerCamelCase , atol=1e-4 ) )
256
import argparse import csv import logging import os import random import numpy as np import torch from torch.utils.data import DataLoader, RandomSampler, SequentialSampler, TensorDataset from tqdm import tqdm, trange from transformers import ( CONFIG_NAME, WEIGHTS_NAME, AdamW, OpenAIGPTDoubleHeadsModel, OpenAIGPTTokenizer, get_linear_schedule_with_warmup, ) logging.basicConfig( format="""%(asctime)s - %(levelname)s - %(name)s - %(message)s""", datefmt="""%m/%d/%Y %H:%M:%S""", level=logging.INFO ) __SCREAMING_SNAKE_CASE = logging.getLogger(__name__) def UpperCAmelCase ( _lowerCamelCase , _lowerCamelCase ): A : Dict = np.argmax(_lowerCamelCase , axis=1 ) return np.sum(outputs == labels ) def UpperCAmelCase ( _lowerCamelCase ): with open(_lowerCamelCase , encoding="utf_8" ) as f: A : Dict = csv.reader(_lowerCamelCase ) A : Optional[int] = [] next(_lowerCamelCase ) # skip the first line for line in tqdm(_lowerCamelCase ): output.append((" ".join(line[1:5] ), line[5], line[6], int(line[-1] ) - 1) ) return output def UpperCAmelCase ( _lowerCamelCase , _lowerCamelCase , _lowerCamelCase , _lowerCamelCase , _lowerCamelCase , _lowerCamelCase ): A : str = [] for dataset in encoded_datasets: A : List[str] = len(_lowerCamelCase ) A : Dict = np.zeros((n_batch, 2, input_len) , dtype=np.intaa ) A : List[str] = np.zeros((n_batch, 2) , dtype=np.intaa ) A : Any = np.full((n_batch, 2, input_len) , fill_value=-100 , dtype=np.intaa ) A : str = np.zeros((n_batch,) , dtype=np.intaa ) for ( i, (story, conta, conta, mc_label), ) in enumerate(_lowerCamelCase ): A : Optional[int] = [start_token] + story[:cap_length] + [delimiter_token] + conta[:cap_length] + [clf_token] A : str = [start_token] + story[:cap_length] + [delimiter_token] + conta[:cap_length] + [clf_token] A : str = with_conta A : Any = with_conta A : Union[str, Any] = len(_lowerCamelCase ) - 1 A : Dict = len(_lowerCamelCase ) - 1 A : int = with_conta A : int = with_conta A : int = mc_label A : Optional[Any] = (input_ids, mc_token_ids, lm_labels, mc_labels) tensor_datasets.append(tuple(torch.tensor(_lowerCamelCase ) for t in all_inputs ) ) return tensor_datasets def UpperCAmelCase ( ): A : Any = argparse.ArgumentParser() parser.add_argument("--model_name" , type=_lowerCamelCase , default="openai-gpt" , help="pretrained model name" ) parser.add_argument("--do_train" , action="store_true" , help="Whether to run training." ) parser.add_argument("--do_eval" , action="store_true" , help="Whether to run eval on the dev set." ) parser.add_argument( "--output_dir" , default=_lowerCamelCase , type=_lowerCamelCase , required=_lowerCamelCase , help="The output directory where the model predictions and checkpoints will be written." , ) parser.add_argument("--train_dataset" , type=_lowerCamelCase , default="" ) parser.add_argument("--eval_dataset" , type=_lowerCamelCase , default="" ) parser.add_argument("--seed" , type=_lowerCamelCase , default=42 ) parser.add_argument("--num_train_epochs" , type=_lowerCamelCase , default=3 ) parser.add_argument("--train_batch_size" , type=_lowerCamelCase , default=8 ) parser.add_argument("--eval_batch_size" , type=_lowerCamelCase , default=16 ) parser.add_argument("--adam_epsilon" , default=1e-8 , type=_lowerCamelCase , help="Epsilon for Adam optimizer." ) parser.add_argument("--max_grad_norm" , type=_lowerCamelCase , default=1 ) parser.add_argument( "--max_steps" , default=-1 , type=_lowerCamelCase , help=( "If > 0: set total number of training steps to perform. Override num_train_epochs." ) , ) parser.add_argument( "--gradient_accumulation_steps" , type=_lowerCamelCase , default=1 , help="Number of updates steps to accumulate before performing a backward/update pass." , ) parser.add_argument("--learning_rate" , type=_lowerCamelCase , default=6.25e-5 ) parser.add_argument("--warmup_steps" , default=0 , type=_lowerCamelCase , help="Linear warmup over warmup_steps." ) parser.add_argument("--lr_schedule" , type=_lowerCamelCase , default="warmup_linear" ) parser.add_argument("--weight_decay" , type=_lowerCamelCase , default=0.01 ) parser.add_argument("--lm_coef" , type=_lowerCamelCase , default=0.9 ) parser.add_argument("--n_valid" , type=_lowerCamelCase , default=374 ) parser.add_argument("--server_ip" , type=_lowerCamelCase , default="" , help="Can be used for distant debugging." ) parser.add_argument("--server_port" , type=_lowerCamelCase , default="" , help="Can be used for distant debugging." ) A : List[Any] = parser.parse_args() print(_lowerCamelCase ) if args.server_ip and args.server_port: # Distant debugging - see https://code.visualstudio.com/docs/python/debugging#_attach-to-a-local-script import ptvsd print("Waiting for debugger attach" ) ptvsd.enable_attach(address=(args.server_ip, args.server_port) , redirect_output=_lowerCamelCase ) ptvsd.wait_for_attach() random.seed(args.seed ) np.random.seed(args.seed ) torch.manual_seed(args.seed ) torch.cuda.manual_seed_all(args.seed ) A : Union[str, Any] = torch.device("cuda" if torch.cuda.is_available() else "cpu" ) A : List[Any] = torch.cuda.device_count() logger.info("device: {}, n_gpu {}".format(_lowerCamelCase , _lowerCamelCase ) ) if not args.do_train and not args.do_eval: raise ValueError("At least one of `do_train` or `do_eval` must be True." ) if not os.path.exists(args.output_dir ): os.makedirs(args.output_dir ) # Load tokenizer and model # This loading functions also add new tokens and embeddings called `special tokens` # These new embeddings will be fine-tuned on the RocStories dataset A : List[Any] = ["_start_", "_delimiter_", "_classify_"] A : Dict = OpenAIGPTTokenizer.from_pretrained(args.model_name ) tokenizer.add_tokens(_lowerCamelCase ) A : str = tokenizer.convert_tokens_to_ids(_lowerCamelCase ) A : str = OpenAIGPTDoubleHeadsModel.from_pretrained(args.model_name ) model.resize_token_embeddings(len(_lowerCamelCase ) ) model.to(_lowerCamelCase ) # Load and encode the datasets def tokenize_and_encode(_lowerCamelCase ): if isinstance(_lowerCamelCase , _lowerCamelCase ): return tokenizer.convert_tokens_to_ids(tokenizer.tokenize(_lowerCamelCase ) ) elif isinstance(_lowerCamelCase , _lowerCamelCase ): return obj return [tokenize_and_encode(_lowerCamelCase ) for o in obj] logger.info("Encoding dataset..." ) A : List[str] = load_rocstories_dataset(args.train_dataset ) A : List[str] = load_rocstories_dataset(args.eval_dataset ) A : str = (train_dataset, eval_dataset) A : Any = tokenize_and_encode(_lowerCamelCase ) # Compute the max input length for the Transformer A : Any = model.config.n_positions // 2 - 2 A : List[str] = max( len(story[:max_length] ) + max(len(conta[:max_length] ) , len(conta[:max_length] ) ) + 3 for dataset in encoded_datasets for story, conta, conta, _ in dataset ) A : Dict = min(_lowerCamelCase , model.config.n_positions ) # Max size of input for the pre-trained model # Prepare inputs tensors and dataloaders A : List[Any] = pre_process_datasets(_lowerCamelCase , _lowerCamelCase , _lowerCamelCase , *_lowerCamelCase ) A , A : Dict = tensor_datasets[0], tensor_datasets[1] A : Union[str, Any] = TensorDataset(*_lowerCamelCase ) A : Dict = RandomSampler(_lowerCamelCase ) A : Dict = DataLoader(_lowerCamelCase , sampler=_lowerCamelCase , batch_size=args.train_batch_size ) A : int = TensorDataset(*_lowerCamelCase ) A : List[str] = SequentialSampler(_lowerCamelCase ) A : Dict = DataLoader(_lowerCamelCase , sampler=_lowerCamelCase , batch_size=args.eval_batch_size ) # Prepare optimizer if args.do_train: if args.max_steps > 0: A : Any = args.max_steps A : str = args.max_steps // (len(_lowerCamelCase ) // args.gradient_accumulation_steps) + 1 else: A : Optional[int] = len(_lowerCamelCase ) // args.gradient_accumulation_steps * args.num_train_epochs A : str = list(model.named_parameters() ) A : Tuple = ["bias", "LayerNorm.bias", "LayerNorm.weight"] A : List[Any] = [ { "params": [p for n, p in param_optimizer if not any(nd in n for nd in no_decay )], "weight_decay": args.weight_decay, }, {"params": [p for n, p in param_optimizer if any(nd in n for nd in no_decay )], "weight_decay": 0.0}, ] A : Optional[Any] = AdamW(_lowerCamelCase , lr=args.learning_rate , eps=args.adam_epsilon ) A : Any = get_linear_schedule_with_warmup( _lowerCamelCase , num_warmup_steps=args.warmup_steps , num_training_steps=_lowerCamelCase ) if args.do_train: A , A , A : Tuple = 0, 0, None model.train() for _ in trange(int(args.num_train_epochs ) , desc="Epoch" ): A : List[Any] = 0 A : Any = 0 A : Dict = tqdm(_lowerCamelCase , desc="Training" ) for step, batch in enumerate(_lowerCamelCase ): A : List[str] = tuple(t.to(_lowerCamelCase ) for t in batch ) A , A , A , A : Any = batch A : Any = model(_lowerCamelCase , mc_token_ids=_lowerCamelCase , lm_labels=_lowerCamelCase , mc_labels=_lowerCamelCase ) A : Any = args.lm_coef * losses[0] + losses[1] loss.backward() optimizer.step() scheduler.step() optimizer.zero_grad() tr_loss += loss.item() A : Dict = ( loss.item() if exp_average_loss is None else 0.7 * exp_average_loss + 0.3 * loss.item() ) nb_tr_steps += 1 A : Optional[int] = "Training loss: {:.2e} lr: {:.2e}".format(_lowerCamelCase , scheduler.get_lr()[0] ) # Save a trained model if args.do_train: # Save a trained model, configuration and tokenizer A : Tuple = model.module if hasattr(_lowerCamelCase , "module" ) else model # Only save the model itself # If we save using the predefined names, we can load using `from_pretrained` A : Dict = os.path.join(args.output_dir , _lowerCamelCase ) A : Any = os.path.join(args.output_dir , _lowerCamelCase ) torch.save(model_to_save.state_dict() , _lowerCamelCase ) model_to_save.config.to_json_file(_lowerCamelCase ) tokenizer.save_vocabulary(args.output_dir ) # Load a trained model and vocabulary that you have fine-tuned A : Optional[Any] = OpenAIGPTDoubleHeadsModel.from_pretrained(args.output_dir ) A : List[str] = OpenAIGPTTokenizer.from_pretrained(args.output_dir ) model.to(_lowerCamelCase ) if args.do_eval: model.eval() A , A : Union[str, Any] = 0, 0 A , A : List[Any] = 0, 0 for batch in tqdm(_lowerCamelCase , desc="Evaluating" ): A : List[str] = tuple(t.to(_lowerCamelCase ) for t in batch ) A , A , A , A : Optional[int] = batch with torch.no_grad(): A , A , A , A : Any = model( _lowerCamelCase , mc_token_ids=_lowerCamelCase , lm_labels=_lowerCamelCase , mc_labels=_lowerCamelCase ) A : Dict = mc_logits.detach().cpu().numpy() A : Union[str, Any] = mc_labels.to("cpu" ).numpy() A : Union[str, Any] = accuracy(_lowerCamelCase , _lowerCamelCase ) eval_loss += mc_loss.mean().item() eval_accuracy += tmp_eval_accuracy nb_eval_examples += input_ids.size(0 ) nb_eval_steps += 1 A : Union[str, Any] = eval_loss / nb_eval_steps A : Tuple = eval_accuracy / nb_eval_examples A : Optional[int] = tr_loss / nb_tr_steps if args.do_train else None A : Dict = {"eval_loss": eval_loss, "eval_accuracy": eval_accuracy, "train_loss": train_loss} A : Optional[Any] = os.path.join(args.output_dir , "eval_results.txt" ) with open(_lowerCamelCase , "w" ) as writer: logger.info("***** Eval results *****" ) for key in sorted(result.keys() ): logger.info(" %s = %s" , _lowerCamelCase , str(result[key] ) ) writer.write("%s = %s\n" % (key, str(result[key] )) ) if __name__ == "__main__": main()
256
1
"""simple docstring""" def __lowercase ( snake_case_ : Optional[Any] ) ->Union[str, Any]: '''simple docstring''' __A : int = 0 __A : List[Any] = len(snake_case_ ) for i in range(n - 1 ): for j in range(i + 1 ,snake_case_ ): if arr[i] > arr[j]: num_inversions += 1 return num_inversions def __lowercase ( snake_case_ : Any ) ->Optional[int]: '''simple docstring''' if len(snake_case_ ) <= 1: return arr, 0 __A : Optional[int] = len(snake_case_ ) // 2 __A : Dict = arr[0:mid] __A : List[str] = arr[mid:] __A , __A : Optional[Any] = count_inversions_recursive(snake_case_ ) __A , __A : Any = count_inversions_recursive(snake_case_ ) __A , __A : List[Any] = _count_cross_inversions(snake_case_ ,snake_case_ ) __A : Optional[int] = inversion_p + inversions_q + cross_inversions return c, num_inversions def __lowercase ( snake_case_ : List[str] ,snake_case_ : Optional[int] ) ->List[Any]: '''simple docstring''' __A : Optional[int] = [] __A : Tuple = 0 while i < len(snake_case_ ) and j < len(snake_case_ ): if p[i] > q[j]: # if P[1] > Q[j], then P[k] > Q[k] for all i < k <= len(P) # These are all inversions. The claim emerges from the # property that P is sorted. num_inversion += len(snake_case_ ) - i r.append(q[j] ) j += 1 else: r.append(p[i] ) i += 1 if i < len(snake_case_ ): r.extend(p[i:] ) else: r.extend(q[j:] ) return r, num_inversion def __lowercase ( ) ->Optional[int]: '''simple docstring''' __A : int = [10, 2, 1, 5, 5, 2, 11] # this arr has 8 inversions: # (10, 2), (10, 1), (10, 5), (10, 5), (10, 2), (2, 1), (5, 2), (5, 2) __A : Dict = count_inversions_bf(snake_case_ ) __A , __A : Optional[Any] = count_inversions_recursive(snake_case_ ) assert num_inversions_bf == num_inversions_recursive == 8 print('''number of inversions = ''' ,snake_case_ ) # testing an array with zero inversion (a sorted arr_1) arr_a.sort() __A : List[str] = count_inversions_bf(snake_case_ ) __A , __A : int = count_inversions_recursive(snake_case_ ) assert num_inversions_bf == num_inversions_recursive == 0 print('''number of inversions = ''' ,snake_case_ ) # an empty list should also have zero inversions __A : Union[str, Any] = [] __A : Any = count_inversions_bf(snake_case_ ) __A , __A : str = count_inversions_recursive(snake_case_ ) assert num_inversions_bf == num_inversions_recursive == 0 print('''number of inversions = ''' ,snake_case_ ) if __name__ == "__main__": main()
179
"""simple docstring""" import logging import os from dataclasses import dataclass, field from functools import partial from pathlib import Path from tempfile import TemporaryDirectory from typing import List, Optional import faiss import torch from datasets import Features, Sequence, Value, load_dataset from transformers import DPRContextEncoder, DPRContextEncoderTokenizerFast, HfArgumentParser a_ = logging.getLogger(__name__) torch.set_grad_enabled(False) a_ = """cuda""" if torch.cuda.is_available() else """cpu""" def __lowercase ( snake_case_ : str ,snake_case_ : Tuple=100 ,snake_case_ : int=" " ) ->List[str]: '''simple docstring''' __A : Dict = text.split(snake_case_ ) return [character.join(text[i : i + n] ).strip() for i in range(0 ,len(snake_case_ ) ,snake_case_ )] def __lowercase ( snake_case_ : dict ) ->dict: '''simple docstring''' __A , __A : Optional[int] = [], [] for title, text in zip(documents['''title'''] ,documents['''text'''] ): if text is not None: for passage in split_text(snake_case_ ): titles.append(title if title is not None else '''''' ) texts.append(snake_case_ ) return {"title": titles, "text": texts} def __lowercase ( snake_case_ : dict ,snake_case_ : DPRContextEncoder ,snake_case_ : DPRContextEncoderTokenizerFast ) ->dict: '''simple docstring''' __A : Tuple = ctx_tokenizer( documents['''title'''] ,documents['''text'''] ,truncation=snake_case_ ,padding='''longest''' ,return_tensors='''pt''' )['''input_ids'''] __A : int = ctx_encoder(input_ids.to(device=snake_case_ ) ,return_dict=snake_case_ ).pooler_output return {"embeddings": embeddings.detach().cpu().numpy()} def __lowercase ( snake_case_ : "RagExampleArguments" ,snake_case_ : "ProcessingArguments" ,snake_case_ : "IndexHnswArguments" ,) ->Tuple: '''simple docstring''' logger.info('''Step 1 - Create the dataset''' ) ###################################### # The dataset needed for RAG must have three columns: # - title (string): title of the document # - text (string): text of a passage of the document # - embeddings (array of dimension d): DPR representation of the passage # Let's say you have documents in tab-separated csv files with columns "title" and "text" assert os.path.isfile(rag_example_args.csv_path ), "Please provide a valid path to a csv file" # You can load a Dataset object this way __A : Optional[Any] = load_dataset( '''csv''' ,data_files=[rag_example_args.csv_path] ,split='''train''' ,delimiter='''\t''' ,column_names=['''title''', '''text'''] ) # More info about loading csv files in the documentation: https://huggingface.co/docs/datasets/loading_datasets.html?highlight=csv#csv-files # Then split the documents into passages of 100 words __A : str = dataset.map(snake_case_ ,batched=snake_case_ ,num_proc=processing_args.num_proc ) # And compute the embeddings __A : Union[str, Any] = DPRContextEncoder.from_pretrained(rag_example_args.dpr_ctx_encoder_model_name ).to(device=snake_case_ ) __A : str = DPRContextEncoderTokenizerFast.from_pretrained(rag_example_args.dpr_ctx_encoder_model_name ) __A : Optional[int] = Features( {'''text''': Value('''string''' ), '''title''': Value('''string''' ), '''embeddings''': Sequence(Value('''float32''' ) )} ) # optional, save as float32 instead of float64 to save space __A : List[Any] = dataset.map( partial(snake_case_ ,ctx_encoder=snake_case_ ,ctx_tokenizer=snake_case_ ) ,batched=snake_case_ ,batch_size=processing_args.batch_size ,features=snake_case_ ,) # And finally save your dataset __A : int = os.path.join(rag_example_args.output_dir ,'''my_knowledge_dataset''' ) dataset.save_to_disk(snake_case_ ) # from datasets import load_from_disk # dataset = load_from_disk(passages_path) # to reload the dataset ###################################### logger.info('''Step 2 - Index the dataset''' ) ###################################### # Let's use the Faiss implementation of HNSW for fast approximate nearest neighbor search __A : Tuple = faiss.IndexHNSWFlat(index_hnsw_args.d ,index_hnsw_args.m ,faiss.METRIC_INNER_PRODUCT ) dataset.add_faiss_index('''embeddings''' ,custom_index=snake_case_ ) # And save the index __A : List[str] = os.path.join(rag_example_args.output_dir ,'''my_knowledge_dataset_hnsw_index.faiss''' ) dataset.get_index('''embeddings''' ).save(snake_case_ ) # dataset.load_faiss_index("embeddings", index_path) # to reload the index @dataclass class __snake_case : """simple docstring""" _lowerCamelCase = field( default=str(Path(SCREAMING_SNAKE_CASE__ ).parent / """test_run""" / """dummy-kb""" / """my_knowledge_dataset.csv""" ) , metadata={"""help""": """Path to a tab-separated csv file with columns 'title' and 'text'"""} , ) _lowerCamelCase = field( default=SCREAMING_SNAKE_CASE__ , metadata={"""help""": """Question that is passed as input to RAG. Default is 'What does Moses' rod turn into ?'."""} , ) _lowerCamelCase = field( default="""facebook/rag-sequence-nq""" , metadata={"""help""": """The RAG model to use. Either 'facebook/rag-sequence-nq' or 'facebook/rag-token-nq'"""} , ) _lowerCamelCase = field( default="""facebook/dpr-ctx_encoder-multiset-base""" , metadata={ """help""": ( """The DPR context encoder model to use. Either 'facebook/dpr-ctx_encoder-single-nq-base' or""" """ 'facebook/dpr-ctx_encoder-multiset-base'""" ) } , ) _lowerCamelCase = field( default=str(Path(SCREAMING_SNAKE_CASE__ ).parent / """test_run""" / """dummy-kb""" ) , metadata={"""help""": """Path to a directory where the dataset passages and the index will be saved"""} , ) @dataclass class __snake_case : """simple docstring""" _lowerCamelCase = field( default=SCREAMING_SNAKE_CASE__ , metadata={ """help""": """The number of processes to use to split the documents into passages. Default is single process.""" } , ) _lowerCamelCase = field( default=16 , metadata={ """help""": """The batch size to use when computing the passages embeddings using the DPR context encoder.""" } , ) @dataclass class __snake_case : """simple docstring""" _lowerCamelCase = field( default=7_68 , metadata={"""help""": """The dimension of the embeddings to pass to the HNSW Faiss index."""} , ) _lowerCamelCase = field( default=1_28 , metadata={ """help""": ( """The number of bi-directional links created for every new element during the HNSW index construction.""" ) } , ) if __name__ == "__main__": logging.basicConfig(level=logging.WARNING) logger.setLevel(logging.INFO) a_ = HfArgumentParser((RagExampleArguments, ProcessingArguments, IndexHnswArguments)) a_ , a_ , a_ = parser.parse_args_into_dataclasses() with TemporaryDirectory() as tmp_dir: a_ = rag_example_args.output_dir or tmp_dir main(rag_example_args, processing_args, index_hnsw_args)
179
1
'''simple docstring''' import sys def a ( lowerCamelCase__ ): '''simple docstring''' A_ : Optional[Any] = len(lowerCamelCase__ ) A_ : Union[str, Any] = [[0 for x in range(lowerCamelCase__ )] for x in range(lowerCamelCase__ )] A_ : int = [[0 for x in range(lowerCamelCase__ )] for x in range(lowerCamelCase__ )] for chain_length in range(2 , lowerCamelCase__ ): for a in range(1 , n - chain_length + 1 ): A_ : Any = a + chain_length - 1 A_ : int = sys.maxsize for c in range(lowerCamelCase__ , lowerCamelCase__ ): A_ : str = ( matrix[a][c] + matrix[c + 1][b] + array[a - 1] * array[c] * array[b] ) if cost < matrix[a][b]: A_ : List[str] = cost A_ : List[Any] = c return matrix, sol def a ( lowerCamelCase__ , lowerCamelCase__ , lowerCamelCase__ ): '''simple docstring''' if i == j: print("""A""" + str(lowerCamelCase__ ) , end=""" """ ) else: print("""(""" , end=""" """ ) print_optiomal_solution(lowerCamelCase__ , lowerCamelCase__ , optimal_solution[i][j] ) print_optiomal_solution(lowerCamelCase__ , optimal_solution[i][j] + 1 , lowerCamelCase__ ) print(""")""" , end=""" """ ) def a ( ): '''simple docstring''' A_ : Dict = [30, 35, 15, 5, 10, 20, 25] A_ : Tuple = len(lowerCamelCase__ ) # Size of matrix created from above array will be # 30*35 35*15 15*5 5*10 10*20 20*25 A_, A_ : Any = matrix_chain_order(lowerCamelCase__ ) print("""No. of Operation required: """ + str(matrix[1][n - 1] ) ) print_optiomal_solution(lowerCamelCase__ , 1 , n - 1 ) if __name__ == "__main__": main()
135
'''simple docstring''' 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 _lowerCAmelCase : @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 , lowercase=True , lowercase=False , lowercase=False , lowercase=False , ): A_ : List[str] = 4 A_ : int = 32 A_ : Optional[int] = (32, 32) A_ : Optional[Any] = torch.manual_seed(0 ) A_ : int = torch.device(lowercase ) A_ : int = (batch_size, num_channels) + sizes A_ : Optional[int] = randn_tensor(lowercase , generator=lowercase , device=lowercase ) A_ : Union[str, Any] = {"""hidden_states""": hidden_states} if include_temb: A_ : str = 128 A_ : List[Any] = randn_tensor((batch_size, temb_channels) , generator=lowercase , device=lowercase ) if include_res_hidden_states_tuple: A_ : List[str] = torch.manual_seed(1 ) A_ : int = (randn_tensor(lowercase , generator=lowercase , device=lowercase ),) if include_encoder_hidden_states: A_ : List[str] = floats_tensor((batch_size, 32, 32) ).to(lowercase ) if include_skip_sample: A_ : Dict = randn_tensor(((batch_size, 3) + sizes) , generator=lowercase , device=lowercase ) return dummy_input def _a (self ): A_ : Tuple = { """in_channels""": 32, """out_channels""": 32, """temb_channels""": 128, } if self.block_type == "up": A_ : Any = 32 if self.block_type == "mid": init_dict.pop("""out_channels""" ) A_ : Optional[int] = self.dummy_input return init_dict, inputs_dict def _a (self , lowercase ): A_, A_ : Optional[Any] = self.prepare_init_args_and_inputs_for_common() A_ : int = self.block_class(**lowercase ) unet_block.to(lowercase ) unet_block.eval() with torch.no_grad(): A_ : List[str] = unet_block(**lowercase ) if isinstance(lowercase , lowercase ): A_ : Union[str, Any] = output[0] self.assertEqual(output.shape , self.output_shape ) A_ : int = output[0, -1, -3:, -3:] A_ : List[Any] = torch.tensor(lowercase ).to(lowercase ) assert torch_all_close(output_slice.flatten() , lowercase , atol=5E-3 ) @unittest.skipIf(torch_device == """mps""" , """Training is not supported in mps""" ) def _a (self ): A_, A_ : Tuple = self.prepare_init_args_and_inputs_for_common() A_ : List[str] = self.block_class(**lowercase ) model.to(lowercase ) model.train() A_ : Any = model(**lowercase ) if isinstance(lowercase , lowercase ): A_ : str = output[0] A_ : Union[str, Any] = torch.device(lowercase ) A_ : Tuple = randn_tensor(output.shape , device=lowercase ) A_ : List[str] = torch.nn.functional.mse_loss(lowercase , lowercase ) loss.backward()
135
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 _UpperCAmelCase ( unittest.TestCase ): """simple docstring""" def lowercase ( self : str ) -> Optional[int]: __lowerCAmelCase = inspect.getfile(accelerate.test_utils ) __lowerCAmelCase = os.path.sep.join(mod_file.split(os.path.sep )[:-1] + ['scripts', 'test_script.py'] ) __lowerCAmelCase = os.path.sep.join( mod_file.split(os.path.sep )[:-1] + ['scripts', 'test_distributed_data_loop.py'] ) __lowerCAmelCase = os.path.sep.join(mod_file.split(os.path.sep )[:-1] + ['scripts', 'test_ops.py'] ) @require_multi_gpu def lowercase ( self : Optional[Any] ) -> List[Any]: print(f"""Found {torch.cuda.device_count()} devices.""" ) __lowerCAmelCase = ['torchrun', f"""--nproc_per_node={torch.cuda.device_count()}""", self.test_file_path] with patch_environment(omp_num_threads=1 ): execute_subprocess_async(_lowerCAmelCase , env=os.environ.copy() ) @require_multi_gpu def lowercase ( self : Tuple ) -> Tuple: print(f"""Found {torch.cuda.device_count()} devices.""" ) __lowerCAmelCase = ['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(_lowerCAmelCase , env=os.environ.copy() ) @require_multi_gpu def lowercase ( self : Union[str, Any] ) -> List[str]: __lowerCAmelCase = ['torchrun', f"""--nproc_per_node={torch.cuda.device_count()}""", inspect.getfile(self.__class__ )] with patch_environment(omp_num_threads=1 ): execute_subprocess_async(_lowerCAmelCase , env=os.environ.copy() ) @require_multi_gpu def lowercase ( self : Optional[int] ) -> Tuple: print(f"""Found {torch.cuda.device_count()} devices, using 2 devices only""" ) __lowerCAmelCase = ['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(_lowerCAmelCase , env=os.environ.copy() ) if __name__ == "__main__": _snake_case : Tuple = Accelerator() _snake_case : int = (accelerator.state.process_index + 2, 10) _snake_case : Union[str, Any] = torch.randint(0, 10, shape).to(accelerator.device) _snake_case : Optional[int] = '' _snake_case : Optional[Any] = 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)." _snake_case : Optional[Any] = 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." _snake_case : int = 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)
284
import itertools import json import linecache import os import pickle import re import socket import string from collections import Counter from logging import getLogger from pathlib import Path from typing import Callable, Dict, Iterable, List import git import torch from torch.utils.data import Dataset from transformers import BartTokenizer, RagTokenizer, TaTokenizer def UpperCAmelCase_ ( __UpperCAmelCase : Dict , __UpperCAmelCase : Dict , __UpperCAmelCase : Union[str, Any] , __UpperCAmelCase : int , __UpperCAmelCase : Optional[Any]=True , __UpperCAmelCase : Any="pt" ) -> List[str]: SCREAMING_SNAKE_CASE_ = {'add_prefix_space': True} if isinstance(__UpperCAmelCase , __UpperCAmelCase ) and not line.startswith(' ' ) else {} SCREAMING_SNAKE_CASE_ = padding_side return tokenizer( [line] , max_length=__UpperCAmelCase , padding='max_length' if pad_to_max_length else None , truncation=__UpperCAmelCase , return_tensors=__UpperCAmelCase , add_special_tokens=__UpperCAmelCase , **__UpperCAmelCase , ) def UpperCAmelCase_ ( __UpperCAmelCase : Union[str, Any] , __UpperCAmelCase : Optional[Any] , __UpperCAmelCase : Dict=None , ) -> Optional[Any]: SCREAMING_SNAKE_CASE_ = input_ids.ne(__UpperCAmelCase ).any(dim=0 ) if attention_mask is None: return input_ids[:, keep_column_mask] else: return (input_ids[:, keep_column_mask], attention_mask[:, keep_column_mask]) class lowerCamelCase_ ( _SCREAMING_SNAKE_CASE ): '''simple docstring''' def __init__( self : List[Any] , _lowerCAmelCase : Tuple , _lowerCAmelCase : int , _lowerCAmelCase : int , _lowerCAmelCase : Dict , _lowerCAmelCase : Union[str, Any]="train" , _lowerCAmelCase : Optional[Any]=None , _lowerCAmelCase : List[str]=None , _lowerCAmelCase : Optional[int]=None , _lowerCAmelCase : Optional[int]="" , ): super().__init__() SCREAMING_SNAKE_CASE_ = Path(_lowerCAmelCase ).joinpath(type_path + '.source' ) SCREAMING_SNAKE_CASE_ = Path(_lowerCAmelCase ).joinpath(type_path + '.target' ) SCREAMING_SNAKE_CASE_ = self.get_char_lens(self.src_file ) SCREAMING_SNAKE_CASE_ = max_source_length SCREAMING_SNAKE_CASE_ = max_target_length assert min(self.src_lens ) > 0, F"found empty line in {self.src_file}" SCREAMING_SNAKE_CASE_ = tokenizer SCREAMING_SNAKE_CASE_ = prefix if n_obs is not None: SCREAMING_SNAKE_CASE_ = self.src_lens[:n_obs] SCREAMING_SNAKE_CASE_ = src_lang SCREAMING_SNAKE_CASE_ = tgt_lang def __len__( self : Tuple ): return len(self.src_lens ) def __getitem__( self : List[str] , _lowerCAmelCase : Any ): SCREAMING_SNAKE_CASE_ = index + 1 # linecache starts at 1 SCREAMING_SNAKE_CASE_ = self.prefix + linecache.getline(str(self.src_file ) , _lowerCAmelCase ).rstrip('\n' ) SCREAMING_SNAKE_CASE_ = linecache.getline(str(self.tgt_file ) , _lowerCAmelCase ).rstrip('\n' ) assert source_line, F"empty source line for index {index}" assert tgt_line, F"empty tgt line for index {index}" # Need to add eos token manually for T5 if isinstance(self.tokenizer , _lowerCAmelCase ): source_line += self.tokenizer.eos_token tgt_line += self.tokenizer.eos_token # Pad source and target to the right SCREAMING_SNAKE_CASE_ = ( self.tokenizer.question_encoder if isinstance(self.tokenizer , _lowerCAmelCase ) else self.tokenizer ) SCREAMING_SNAKE_CASE_ = self.tokenizer.generator if isinstance(self.tokenizer , _lowerCAmelCase ) else self.tokenizer SCREAMING_SNAKE_CASE_ = encode_line(_lowerCAmelCase , _lowerCAmelCase , self.max_source_length , 'right' ) SCREAMING_SNAKE_CASE_ = encode_line(_lowerCAmelCase , _lowerCAmelCase , self.max_target_length , 'right' ) SCREAMING_SNAKE_CASE_ = source_inputs['input_ids'].squeeze() SCREAMING_SNAKE_CASE_ = target_inputs['input_ids'].squeeze() SCREAMING_SNAKE_CASE_ = source_inputs['attention_mask'].squeeze() return { "input_ids": source_ids, "attention_mask": src_mask, "decoder_input_ids": target_ids, } @staticmethod def lowerCAmelCase_ ( _lowerCAmelCase : Optional[int] ): return [len(_lowerCAmelCase ) for x in Path(_lowerCAmelCase ).open().readlines()] def lowerCAmelCase_ ( self : int , _lowerCAmelCase : Optional[int] ): SCREAMING_SNAKE_CASE_ = torch.stack([x['input_ids'] for x in batch] ) SCREAMING_SNAKE_CASE_ = torch.stack([x['attention_mask'] for x in batch] ) SCREAMING_SNAKE_CASE_ = torch.stack([x['decoder_input_ids'] for x in batch] ) SCREAMING_SNAKE_CASE_ = ( self.tokenizer.generator.pad_token_id if isinstance(self.tokenizer , _lowerCAmelCase ) else self.tokenizer.pad_token_id ) SCREAMING_SNAKE_CASE_ = ( self.tokenizer.question_encoder.pad_token_id if isinstance(self.tokenizer , _lowerCAmelCase ) else self.tokenizer.pad_token_id ) SCREAMING_SNAKE_CASE_ = trim_batch(_lowerCAmelCase , _lowerCAmelCase ) SCREAMING_SNAKE_CASE_ , SCREAMING_SNAKE_CASE_ = trim_batch(_lowerCAmelCase , _lowerCAmelCase , attention_mask=_lowerCAmelCase ) SCREAMING_SNAKE_CASE_ = { 'input_ids': source_ids, 'attention_mask': source_mask, 'decoder_input_ids': y, } return batch lowerCamelCase__ : List[str] = getLogger(__name__) def UpperCAmelCase_ ( __UpperCAmelCase : List[List] ) -> Tuple: return list(itertools.chain.from_iterable(__UpperCAmelCase ) ) def UpperCAmelCase_ ( __UpperCAmelCase : str ) -> None: SCREAMING_SNAKE_CASE_ = get_git_info() save_json(__UpperCAmelCase , os.path.join(__UpperCAmelCase , 'git_log.json' ) ) def UpperCAmelCase_ ( __UpperCAmelCase : Dict , __UpperCAmelCase : Tuple , __UpperCAmelCase : Optional[int]=4 , **__UpperCAmelCase : Tuple ) -> str: with open(__UpperCAmelCase , 'w' ) as f: json.dump(__UpperCAmelCase , __UpperCAmelCase , indent=__UpperCAmelCase , **__UpperCAmelCase ) def UpperCAmelCase_ ( __UpperCAmelCase : List[str] ) -> int: with open(__UpperCAmelCase ) as f: return json.load(__UpperCAmelCase ) def UpperCAmelCase_ ( ) -> Tuple: SCREAMING_SNAKE_CASE_ = git.Repo(search_parent_directories=__UpperCAmelCase ) SCREAMING_SNAKE_CASE_ = { 'repo_id': str(__UpperCAmelCase ), 'repo_sha': str(repo.head.object.hexsha ), 'repo_branch': str(repo.active_branch ), 'hostname': str(socket.gethostname() ), } return repo_infos def UpperCAmelCase_ ( __UpperCAmelCase : Callable , __UpperCAmelCase : Iterable ) -> List: return list(map(__UpperCAmelCase , __UpperCAmelCase ) ) def UpperCAmelCase_ ( __UpperCAmelCase : str , __UpperCAmelCase : Dict ) -> Dict: with open(__UpperCAmelCase , 'wb' ) as f: return pickle.dump(__UpperCAmelCase , __UpperCAmelCase ) def UpperCAmelCase_ ( __UpperCAmelCase : List[Any] ) -> Any: def remove_articles(__UpperCAmelCase : Any ): return re.sub(r'\b(a|an|the)\b' , ' ' , __UpperCAmelCase ) def white_space_fix(__UpperCAmelCase : List[str] ): return " ".join(text.split() ) def remove_punc(__UpperCAmelCase : List[Any] ): SCREAMING_SNAKE_CASE_ = set(string.punctuation ) return "".join(ch for ch in text if ch not in exclude ) def lower(__UpperCAmelCase : Union[str, Any] ): return text.lower() return white_space_fix(remove_articles(remove_punc(lower(__UpperCAmelCase ) ) ) ) def UpperCAmelCase_ ( __UpperCAmelCase : int , __UpperCAmelCase : Any ) -> List[str]: SCREAMING_SNAKE_CASE_ = normalize_answer(__UpperCAmelCase ).split() SCREAMING_SNAKE_CASE_ = normalize_answer(__UpperCAmelCase ).split() SCREAMING_SNAKE_CASE_ = Counter(__UpperCAmelCase ) & Counter(__UpperCAmelCase ) SCREAMING_SNAKE_CASE_ = sum(common.values() ) if num_same == 0: return 0 SCREAMING_SNAKE_CASE_ = 1.0 * num_same / len(__UpperCAmelCase ) SCREAMING_SNAKE_CASE_ = 1.0 * num_same / len(__UpperCAmelCase ) SCREAMING_SNAKE_CASE_ = (2 * precision * recall) / (precision + recall) return fa def UpperCAmelCase_ ( __UpperCAmelCase : str , __UpperCAmelCase : Optional[int] ) -> Any: return normalize_answer(__UpperCAmelCase ) == normalize_answer(__UpperCAmelCase ) def UpperCAmelCase_ ( __UpperCAmelCase : List[str] , __UpperCAmelCase : List[str] ) -> Dict: assert len(__UpperCAmelCase ) == len(__UpperCAmelCase ) SCREAMING_SNAKE_CASE_ = 0 for hypo, pred in zip(__UpperCAmelCase , __UpperCAmelCase ): em += exact_match_score(__UpperCAmelCase , __UpperCAmelCase ) if len(__UpperCAmelCase ) > 0: em /= len(__UpperCAmelCase ) return {"em": em} def UpperCAmelCase_ ( __UpperCAmelCase : Union[str, Any] ) -> Dict: return model_prefix.startswith('rag' ) def UpperCAmelCase_ ( __UpperCAmelCase : Optional[Any] , __UpperCAmelCase : Any , __UpperCAmelCase : List[str] ) -> Tuple: SCREAMING_SNAKE_CASE_ = {p: p for p in extra_params} # T5 models don't have `dropout` param, they have `dropout_rate` instead SCREAMING_SNAKE_CASE_ = 'dropout_rate' for p in extra_params: if getattr(__UpperCAmelCase , __UpperCAmelCase , __UpperCAmelCase ): if not hasattr(__UpperCAmelCase , __UpperCAmelCase ) and not hasattr(__UpperCAmelCase , equivalent_param[p] ): logger.info('config doesn\'t have a `{}` attribute'.format(__UpperCAmelCase ) ) delattr(__UpperCAmelCase , __UpperCAmelCase ) continue SCREAMING_SNAKE_CASE_ = p if hasattr(__UpperCAmelCase , __UpperCAmelCase ) else equivalent_param[p] setattr(__UpperCAmelCase , __UpperCAmelCase , getattr(__UpperCAmelCase , __UpperCAmelCase ) ) delattr(__UpperCAmelCase , __UpperCAmelCase ) return hparams, config
225
0
'''simple docstring''' from __future__ import annotations import numpy as np from numpy import floataa from numpy.typing import NDArray def lowerCamelCase ( UpperCAmelCase__ : NDArray[floataa] , UpperCAmelCase__ : NDArray[floataa] , UpperCAmelCase__ : list[int] , UpperCAmelCase__ : int , ) -> list[float]: lowercase_ , lowercase_ : Optional[Any] = coefficient_matrix.shape lowercase_ , lowercase_ : Optional[int] = constant_matrix.shape if rowsa != colsa: lowercase_ : Optional[int] = F'''Coefficient matrix dimensions must be nxn but received {rowsa}x{colsa}''' raise ValueError(UpperCAmelCase__ ) if colsa != 1: lowercase_ : Any = F'''Constant matrix must be nx1 but received {rowsa}x{colsa}''' raise ValueError(UpperCAmelCase__ ) if rowsa != rowsa: lowercase_ : Tuple = ( """Coefficient and constant matrices dimensions must be nxn and nx1 but """ F'''received {rowsa}x{colsa} and {rowsa}x{colsa}''' ) raise ValueError(UpperCAmelCase__ ) if len(UpperCAmelCase__ ) != rowsa: lowercase_ : Dict = ( """Number of initial values must be equal to number of rows in coefficient """ F'''matrix but received {len(UpperCAmelCase__ )} and {rowsa}''' ) raise ValueError(UpperCAmelCase__ ) if iterations <= 0: raise ValueError("""Iterations must be at least 1""" ) lowercase_ : NDArray[floataa] = np.concatenate( (coefficient_matrix, constant_matrix) , axis=1 ) lowercase_ , lowercase_ : str = table.shape strictly_diagonally_dominant(UpperCAmelCase__ ) # Iterates the whole matrix for given number of times for _ in range(UpperCAmelCase__ ): lowercase_ : List[Any] = [] for row in range(UpperCAmelCase__ ): lowercase_ : Optional[Any] = 0 for col in range(UpperCAmelCase__ ): if col == row: lowercase_ : Tuple = table[row][col] elif col == cols - 1: lowercase_ : Optional[int] = table[row][col] else: temp += (-1) * table[row][col] * init_val[col] lowercase_ : Dict = (temp + val) / denom new_val.append(UpperCAmelCase__ ) lowercase_ : List[str] = new_val return [float(UpperCAmelCase__ ) for i in new_val] def lowerCamelCase ( UpperCAmelCase__ : NDArray[floataa] ) -> bool: lowercase_ , lowercase_ : Union[str, Any] = table.shape lowercase_ : Optional[int] = True for i in range(0 , UpperCAmelCase__ ): lowercase_ : List[str] = 0 for j in range(0 , cols - 1 ): if i == j: continue else: total += table[i][j] if table[i][i] <= total: raise ValueError("""Coefficient matrix is not strictly diagonally dominant""" ) return is_diagonally_dominant # Test Cases if __name__ == "__main__": import doctest doctest.testmod()
21
'''simple docstring''' import os import sys from contextlib import contextmanager # Windows only if os.name == "nt": import ctypes import msvcrt # noqa class __magic_name__ ( ctypes.Structure): # _fields is a specific attr expected by ctypes UpperCamelCase__ = [('''size''', ctypes.c_int), ('''visible''', ctypes.c_byte)] def lowerCamelCase ( ) -> List[Any]: if os.name == "nt": lowercase_ : List[Any] = CursorInfo() lowercase_ : int = ctypes.windll.kernelaa.GetStdHandle(-11 ) ctypes.windll.kernelaa.GetConsoleCursorInfo(UpperCAmelCase__ , ctypes.byref(UpperCAmelCase__ ) ) lowercase_ : List[str] = False ctypes.windll.kernelaa.SetConsoleCursorInfo(UpperCAmelCase__ , ctypes.byref(UpperCAmelCase__ ) ) elif os.name == "posix": sys.stdout.write("""\033[?25l""" ) sys.stdout.flush() def lowerCamelCase ( ) -> str: if os.name == "nt": lowercase_ : int = CursorInfo() lowercase_ : Optional[Any] = ctypes.windll.kernelaa.GetStdHandle(-11 ) ctypes.windll.kernelaa.GetConsoleCursorInfo(UpperCAmelCase__ , ctypes.byref(UpperCAmelCase__ ) ) lowercase_ : Optional[int] = True ctypes.windll.kernelaa.SetConsoleCursorInfo(UpperCAmelCase__ , ctypes.byref(UpperCAmelCase__ ) ) elif os.name == "posix": sys.stdout.write("""\033[?25h""" ) sys.stdout.flush() @contextmanager def lowerCamelCase ( ) -> Any: try: hide_cursor() yield finally: show_cursor()
21
1
import argparse import os import evaluate import torch from datasets import load_dataset from torch.optim import AdamW from torch.utils.data import DataLoader from transformers import AutoModelForSequenceClassification, AutoTokenizer, get_linear_schedule_with_warmup, set_seed from accelerate import Accelerator, DistributedType from accelerate.local_sgd import LocalSGD ######################################################################## # This is a fully working simple example to use Accelerate # with LocalSGD, which is a method to synchronize model # parameters every K batches. It is different, but complementary # to gradient accumulation. # # This example trains a Bert base model on GLUE MRPC # in any of the following settings (with the same script): # - single CPU or single GPU # - multi GPUS (using PyTorch distributed mode) # - (multi) TPUs # - fp16 (mixed-precision) or fp32 (normal precision) # # To run it in each of these various modes, follow the instructions # in the readme for examples: # https://github.com/huggingface/accelerate/tree/main/examples # ######################################################################## UpperCAmelCase__ = 16 UpperCAmelCase__ = 32 def UpperCAmelCase_ ( __snake_case , __snake_case = 16 ) -> str: """simple docstring""" _lowercase =AutoTokenizer.from_pretrained('''bert-base-cased''' ) _lowercase =load_dataset('''glue''' , '''mrpc''' ) def tokenize_function(__snake_case ): # max_length=None => use the model max length (it's actually the default) _lowercase =tokenizer(examples['''sentence1'''] , examples['''sentence2'''] , truncation=__snake_case , max_length=__snake_case ) return outputs # Apply the method we just defined to all the examples in all the splits of the dataset # starting with the main process first: with accelerator.main_process_first(): _lowercase =datasets.map( __snake_case , batched=__snake_case , remove_columns=['''idx''', '''sentence1''', '''sentence2'''] , ) # We also rename the 'label' column to 'labels' which is the expected name for labels by the models of the # transformers library _lowercase =tokenized_datasets.rename_column('''label''' , '''labels''' ) def collate_fn(__snake_case ): # On TPU it's best to pad everything to the same length or training will be very slow. _lowercase =128 if accelerator.distributed_type == DistributedType.TPU else None # When using mixed precision we want round multiples of 8/16 if accelerator.mixed_precision == "fp8": _lowercase =16 elif accelerator.mixed_precision != "no": _lowercase =8 else: _lowercase =None return tokenizer.pad( __snake_case , padding='''longest''' , max_length=__snake_case , pad_to_multiple_of=__snake_case , return_tensors='''pt''' , ) # Instantiate dataloaders. _lowercase =DataLoader( tokenized_datasets['''train'''] , shuffle=__snake_case , collate_fn=__snake_case , batch_size=__snake_case ) _lowercase =DataLoader( tokenized_datasets['''validation'''] , shuffle=__snake_case , collate_fn=__snake_case , batch_size=__snake_case ) return train_dataloader, eval_dataloader # For testing only if os.environ.get('''TESTING_MOCKED_DATALOADERS''', None) == "1": from accelerate.test_utils.training import mocked_dataloaders UpperCAmelCase__ = mocked_dataloaders # noqa: F811 def UpperCAmelCase_ ( __snake_case , __snake_case ) -> List[Any]: """simple docstring""" if os.environ.get('''TESTING_MOCKED_DATALOADERS''' , __snake_case ) == "1": _lowercase =2 # New Code # _lowercase =int(args.gradient_accumulation_steps ) _lowercase =int(args.local_sgd_steps ) # Initialize accelerator _lowercase =Accelerator( cpu=args.cpu , mixed_precision=args.mixed_precision , gradient_accumulation_steps=__snake_case ) if accelerator.distributed_type not in [DistributedType.NO, DistributedType.MULTI_CPU, DistributedType.MULTI_GPU]: raise NotImplementedError('''LocalSGD is supported only for CPUs and GPUs (no DeepSpeed or MegatronLM)''' ) # Sample hyper-parameters for learning rate, batch size, seed and a few other HPs _lowercase =config['''lr'''] _lowercase =int(config['''num_epochs'''] ) _lowercase =int(config['''seed'''] ) _lowercase =int(config['''batch_size'''] ) _lowercase =evaluate.load('''glue''' , '''mrpc''' ) set_seed(__snake_case ) _lowercase , _lowercase =get_dataloaders(__snake_case , __snake_case ) # Instantiate the model (we build the model here so that the seed also control new weights initialization) _lowercase =AutoModelForSequenceClassification.from_pretrained('''bert-base-cased''' , return_dict=__snake_case ) # We could avoid this line since the accelerator is set with `device_placement=True` (default value). # Note that if you are placing tensors on devices manually, this line absolutely needs to be before the optimizer # creation otherwise training will not work on TPU (`accelerate` will kindly throw an error to make us aware of that). _lowercase =model.to(accelerator.device ) # Instantiate optimizer _lowercase =AdamW(params=model.parameters() , lr=__snake_case ) # Instantiate scheduler _lowercase =get_linear_schedule_with_warmup( optimizer=__snake_case , num_warmup_steps=100 , num_training_steps=(len(__snake_case ) * num_epochs) , ) # Prepare everything # There is no specific order to remember, we just need to unpack the objects in the same order we gave them to the # prepare method. _lowercase , _lowercase , _lowercase , _lowercase , _lowercase =accelerator.prepare( __snake_case , __snake_case , __snake_case , __snake_case , __snake_case ) # Now we train the model for epoch in range(__snake_case ): model.train() with LocalSGD( accelerator=__snake_case , model=__snake_case , local_sgd_steps=__snake_case , enabled=local_sgd_steps is not None ) as local_sgd: for step, batch in enumerate(__snake_case ): # We could avoid this line since we set the accelerator with `device_placement=True`. batch.to(accelerator.device ) # New code # # We use the new `accumulate` context manager to perform gradient accumulation # We also currently do not support TPUs nor advise it as bugs were found on the XLA side when running our tests. with accelerator.accumulate(__snake_case ): _lowercase =model(**__snake_case ) _lowercase =output.loss accelerator.backward(__snake_case ) optimizer.step() lr_scheduler.step() optimizer.zero_grad() # LocalSGD-specific line local_sgd.step() model.eval() for step, batch in enumerate(__snake_case ): # We could avoid this line since we set the accelerator with `device_placement=True`. batch.to(accelerator.device ) with torch.no_grad(): _lowercase =model(**__snake_case ) _lowercase =outputs.logits.argmax(dim=-1 ) _lowercase , _lowercase =accelerator.gather_for_metrics((predictions, batch['''labels''']) ) metric.add_batch( predictions=__snake_case , references=__snake_case , ) _lowercase =metric.compute() # Use accelerator.print to print only on the main process. accelerator.print(F"epoch {epoch}:" , __snake_case ) def UpperCAmelCase_ ( ) -> Optional[int]: """simple docstring""" _lowercase =argparse.ArgumentParser(description='''Simple example of training script.''' ) parser.add_argument( '''--mixed_precision''' , type=__snake_case , default=__snake_case , choices=['''no''', '''fp16''', '''bf16''', '''fp8'''] , help='''Whether to use mixed precision. Choose''' '''between fp16 and bf16 (bfloat16). Bf16 requires PyTorch >= 1.10.''' '''and an Nvidia Ampere GPU.''' , ) # New Code # parser.add_argument( '''--gradient_accumulation_steps''' , type=__snake_case , default=1 , help='''The number of minibatches to be ran before gradients are accumulated.''' , ) parser.add_argument( '''--local_sgd_steps''' , type=__snake_case , default=8 , help='''Number of local SGD steps or None to disable local SGD''' ) parser.add_argument('''--cpu''' , action='''store_true''' , help='''If passed, will train on the CPU.''' ) _lowercase =parser.parse_args() _lowercase ={'''lr''': 2e-5, '''num_epochs''': 3, '''seed''': 42, '''batch_size''': 16} training_function(__snake_case , __snake_case ) if __name__ == "__main__": main()
5
from typing import Any def UpperCAmelCase_ ( __snake_case , __snake_case , __snake_case , __snake_case , __snake_case , ) -> list: """simple docstring""" _validation( __snake_case , __snake_case , __snake_case , __snake_case , __snake_case , ) # Creates data structures and fill initial step _lowercase ={} _lowercase ={} for state in states_space: _lowercase =observations_space[0] _lowercase =( initial_probabilities[state] * emission_probabilities[state][observation] ) _lowercase =None # Fills the data structure with the probabilities of # different transitions and pointers to previous states for o in range(1 , len(__snake_case ) ): _lowercase =observations_space[o] _lowercase =observations_space[o - 1] for state in states_space: # Calculates the argmax for probability function _lowercase ='''''' _lowercase =-1 for k_state in states_space: _lowercase =( probabilities[(k_state, prior_observation)] * transition_probabilities[k_state][state] * emission_probabilities[state][observation] ) if probability > max_probability: _lowercase =probability _lowercase =k_state # Update probabilities and pointers dicts _lowercase =( probabilities[(arg_max, prior_observation)] * transition_probabilities[arg_max][state] * emission_probabilities[state][observation] ) _lowercase =arg_max # The final observation _lowercase =observations_space[len(__snake_case ) - 1] # argmax for given final observation _lowercase ='''''' _lowercase =-1 for k_state in states_space: _lowercase =probabilities[(k_state, final_observation)] if probability > max_probability: _lowercase =probability _lowercase =k_state _lowercase =arg_max # Process pointers backwards _lowercase =last_state _lowercase =[] for o in range(len(__snake_case ) - 1 , -1 , -1 ): result.append(__snake_case ) _lowercase =pointers[previous, observations_space[o]] result.reverse() return result def UpperCAmelCase_ ( __snake_case , __snake_case , __snake_case , __snake_case , __snake_case , ) -> None: """simple docstring""" _validate_not_empty( __snake_case , __snake_case , __snake_case , __snake_case , __snake_case , ) _validate_lists(__snake_case , __snake_case ) _validate_dicts( __snake_case , __snake_case , __snake_case ) def UpperCAmelCase_ ( __snake_case , __snake_case , __snake_case , __snake_case , __snake_case , ) -> None: """simple docstring""" if not all( [ observations_space, states_space, initial_probabilities, transition_probabilities, emission_probabilities, ] ): raise ValueError('''There\'s an empty parameter''' ) def UpperCAmelCase_ ( __snake_case , __snake_case ) -> None: """simple docstring""" _validate_list(__snake_case , '''observations_space''' ) _validate_list(__snake_case , '''states_space''' ) def UpperCAmelCase_ ( __snake_case , __snake_case ) -> None: """simple docstring""" if not isinstance(_object , __snake_case ): _lowercase =F"{var_name} must be a list" raise ValueError(__snake_case ) else: for x in _object: if not isinstance(__snake_case , __snake_case ): _lowercase =F"{var_name} must be a list of strings" raise ValueError(__snake_case ) def UpperCAmelCase_ ( __snake_case , __snake_case , __snake_case , ) -> None: """simple docstring""" _validate_dict(__snake_case , '''initial_probabilities''' , __snake_case ) _validate_nested_dict(__snake_case , '''transition_probabilities''' ) _validate_nested_dict(__snake_case , '''emission_probabilities''' ) def UpperCAmelCase_ ( __snake_case , __snake_case ) -> None: """simple docstring""" _validate_dict(_object , __snake_case , __snake_case ) for x in _object.values(): _validate_dict(__snake_case , __snake_case , __snake_case , __snake_case ) def UpperCAmelCase_ ( __snake_case , __snake_case , __snake_case , __snake_case = False ) -> None: """simple docstring""" if not isinstance(_object , __snake_case ): _lowercase =F"{var_name} must be a dict" raise ValueError(__snake_case ) if not all(isinstance(__snake_case , __snake_case ) for x in _object ): _lowercase =F"{var_name} all keys must be strings" raise ValueError(__snake_case ) if not all(isinstance(__snake_case , __snake_case ) for x in _object.values() ): _lowercase ='''nested dictionary ''' if nested else '''''' _lowercase =F"{var_name} {nested_text}all values must be {value_type.__name__}" raise ValueError(__snake_case ) if __name__ == "__main__": from doctest import testmod testmod()
5
1
from typing import TYPE_CHECKING from ...utils import ( OptionalDependencyNotAvailable, _LazyModule, is_tf_available, is_tokenizers_available, is_torch_available, ) _lowerCamelCase = { """configuration_convbert""": ["""CONVBERT_PRETRAINED_CONFIG_ARCHIVE_MAP""", """ConvBertConfig""", """ConvBertOnnxConfig"""], """tokenization_convbert""": ["""ConvBertTokenizer"""], } try: if not is_tokenizers_available(): raise OptionalDependencyNotAvailable() except OptionalDependencyNotAvailable: pass else: _lowerCamelCase = ["""ConvBertTokenizerFast"""] try: if not is_torch_available(): raise OptionalDependencyNotAvailable() except OptionalDependencyNotAvailable: pass else: _lowerCamelCase = [ """CONVBERT_PRETRAINED_MODEL_ARCHIVE_LIST""", """ConvBertForMaskedLM""", """ConvBertForMultipleChoice""", """ConvBertForQuestionAnswering""", """ConvBertForSequenceClassification""", """ConvBertForTokenClassification""", """ConvBertLayer""", """ConvBertModel""", """ConvBertPreTrainedModel""", """load_tf_weights_in_convbert""", ] try: if not is_tf_available(): raise OptionalDependencyNotAvailable() except OptionalDependencyNotAvailable: pass else: _lowerCamelCase = [ """TF_CONVBERT_PRETRAINED_MODEL_ARCHIVE_LIST""", """TFConvBertForMaskedLM""", """TFConvBertForMultipleChoice""", """TFConvBertForQuestionAnswering""", """TFConvBertForSequenceClassification""", """TFConvBertForTokenClassification""", """TFConvBertLayer""", """TFConvBertModel""", """TFConvBertPreTrainedModel""", ] if TYPE_CHECKING: from .configuration_convbert import CONVBERT_PRETRAINED_CONFIG_ARCHIVE_MAP, ConvBertConfig, ConvBertOnnxConfig from .tokenization_convbert import ConvBertTokenizer try: if not is_tokenizers_available(): raise OptionalDependencyNotAvailable() except OptionalDependencyNotAvailable: pass else: from .tokenization_convbert_fast import ConvBertTokenizerFast try: if not is_torch_available(): raise OptionalDependencyNotAvailable() except OptionalDependencyNotAvailable: pass else: from .modeling_convbert import ( CONVBERT_PRETRAINED_MODEL_ARCHIVE_LIST, ConvBertForMaskedLM, ConvBertForMultipleChoice, ConvBertForQuestionAnswering, ConvBertForSequenceClassification, ConvBertForTokenClassification, ConvBertLayer, ConvBertModel, ConvBertPreTrainedModel, load_tf_weights_in_convbert, ) try: if not is_tf_available(): raise OptionalDependencyNotAvailable() except OptionalDependencyNotAvailable: pass else: from .modeling_tf_convbert import ( TF_CONVBERT_PRETRAINED_MODEL_ARCHIVE_LIST, TFConvBertForMaskedLM, TFConvBertForMultipleChoice, TFConvBertForQuestionAnswering, TFConvBertForSequenceClassification, TFConvBertForTokenClassification, TFConvBertLayer, TFConvBertModel, TFConvBertPreTrainedModel, ) else: import sys _lowerCamelCase = _LazyModule(__name__, globals()["""__file__"""], _import_structure, module_spec=__spec__)
354
'''simple docstring''' from __future__ import annotations from itertools import permutations from random import randint from timeit import repeat def a__ ( ) -> tuple[list[int], int]: """simple docstring""" UpperCAmelCase_ : Tuple = [randint(-10_00 , 10_00 ) for i in range(10 )] UpperCAmelCase_ : str = randint(-50_00 , 50_00 ) return (arr, r) _lowerCamelCase = make_dataset() def a__ ( _SCREAMING_SNAKE_CASE : list[int] , _SCREAMING_SNAKE_CASE : int ) -> tuple[int, ...]: """simple docstring""" for triplet in permutations(_SCREAMING_SNAKE_CASE , 3 ): if sum(_SCREAMING_SNAKE_CASE ) == target: return tuple(sorted(_SCREAMING_SNAKE_CASE ) ) return (0, 0, 0) def a__ ( _SCREAMING_SNAKE_CASE : list[int] , _SCREAMING_SNAKE_CASE : int ) -> tuple[int, int, int]: """simple docstring""" arr.sort() UpperCAmelCase_ : Optional[int] = len(_SCREAMING_SNAKE_CASE ) for i in range(n - 1 ): UpperCAmelCase_ , UpperCAmelCase_ : Union[str, Any] = i + 1, n - 1 while left < right: if arr[i] + arr[left] + arr[right] == target: return (arr[i], arr[left], arr[right]) elif arr[i] + arr[left] + arr[right] < target: left += 1 elif arr[i] + arr[left] + arr[right] > target: right -= 1 return (0, 0, 0) def a__ ( ) -> tuple[float, float]: """simple docstring""" UpperCAmelCase_ : Tuple = "\nfrom __main__ import dataset, triplet_sum1, triplet_sum2\n" UpperCAmelCase_ : Optional[Any] = "\ntriplet_sum1(*dataset)\n" UpperCAmelCase_ : str = "\ntriplet_sum2(*dataset)\n" UpperCAmelCase_ : Dict = repeat(setup=_SCREAMING_SNAKE_CASE , stmt=_SCREAMING_SNAKE_CASE , repeat=5 , number=1_00_00 ) UpperCAmelCase_ : Dict = repeat(setup=_SCREAMING_SNAKE_CASE , stmt=_SCREAMING_SNAKE_CASE , repeat=5 , number=1_00_00 ) return (min(_SCREAMING_SNAKE_CASE ), min(_SCREAMING_SNAKE_CASE )) if __name__ == "__main__": from doctest import testmod testmod() _lowerCamelCase = solution_times() print(f"""The time for naive implementation is {times[0]}.""") print(f"""The time for optimized implementation is {times[1]}.""")
67
0
import re from typing import Callable, List, Optional, Union import tensorflow as tf try: from tensorflow.keras.optimizers.legacy import Adam except ImportError: from tensorflow.keras.optimizers import Adam class _lowerCAmelCase ( tf.keras.optimizers.schedules.LearningRateSchedule ): def __init__( self , _UpperCamelCase , _UpperCamelCase , _UpperCamelCase , _UpperCamelCase = 1.0 , _UpperCamelCase = None , ) -> Tuple: super().__init__() lowerCAmelCase_ = initial_learning_rate lowerCAmelCase_ = warmup_steps lowerCAmelCase_ = power lowerCAmelCase_ = decay_schedule_fn lowerCAmelCase_ = name def __call__( self , _UpperCamelCase ) -> List[Any]: with tf.name_scope(self.name or "WarmUp" ) as name: # Implements polynomial warmup. i.e., if global_step < warmup_steps, the # learning rate will be `global_step/num_warmup_steps * init_lr`. lowerCAmelCase_ = tf.cast(__lowercase , tf.floataa ) lowerCAmelCase_ = tf.cast(self.warmup_steps , tf.floataa ) lowerCAmelCase_ = global_step_float / warmup_steps_float lowerCAmelCase_ = self.initial_learning_rate * tf.math.pow(__lowercase , self.power ) return tf.cond( global_step_float < warmup_steps_float , lambda: warmup_learning_rate , lambda: self.decay_schedule_fn(step - self.warmup_steps ) , name=__lowercase , ) def __a ( self ) -> Optional[Any]: return { "initial_learning_rate": self.initial_learning_rate, "decay_schedule_fn": self.decay_schedule_fn, "warmup_steps": self.warmup_steps, "power": self.power, "name": self.name, } def lowerCamelCase__ ( __lowerCAmelCase : float , __lowerCAmelCase : int , __lowerCAmelCase : int , __lowerCAmelCase : float = 0.0 , __lowerCAmelCase : float = 0.9 , __lowerCAmelCase : float = 0.999 , __lowerCAmelCase : float = 1e-8 , __lowerCAmelCase : Optional[float] = None , __lowerCAmelCase : Optional[float] = None , __lowerCAmelCase : float = 0.0 , __lowerCAmelCase : float = 1.0 , __lowerCAmelCase : Optional[List[str]] = None , ): """simple docstring""" lowerCAmelCase_ = tf.keras.optimizers.schedules.PolynomialDecay( initial_learning_rate=_lowercase , decay_steps=num_train_steps - num_warmup_steps , end_learning_rate=init_lr * min_lr_ratio , power=_lowercase , ) if num_warmup_steps: lowerCAmelCase_ = WarmUp( initial_learning_rate=_lowercase , decay_schedule_fn=_lowercase , warmup_steps=_lowercase , ) if weight_decay_rate > 0.0: lowerCAmelCase_ = AdamWeightDecay( learning_rate=_lowercase , weight_decay_rate=_lowercase , beta_a=_lowercase , beta_a=_lowercase , epsilon=_lowercase , clipnorm=_lowercase , global_clipnorm=_lowercase , exclude_from_weight_decay=["LayerNorm", "layer_norm", "bias"] , include_in_weight_decay=_lowercase , ) else: lowerCAmelCase_ = tf.keras.optimizers.Adam( learning_rate=_lowercase , beta_a=_lowercase , beta_a=_lowercase , epsilon=_lowercase , clipnorm=_lowercase , global_clipnorm=_lowercase , ) # We return the optimizer and the LR scheduler in order to better track the # evolution of the LR independently of the optimizer. return optimizer, lr_schedule class _lowerCAmelCase ( A__ ): def __init__( self , _UpperCamelCase = 0.001 , _UpperCamelCase = 0.9 , _UpperCamelCase = 0.999 , _UpperCamelCase = 1e-7 , _UpperCamelCase = False , _UpperCamelCase = 0.0 , _UpperCamelCase = None , _UpperCamelCase = None , _UpperCamelCase = "AdamWeightDecay" , **_UpperCamelCase , ) -> Dict: super().__init__(__lowercase , __lowercase , __lowercase , __lowercase , __lowercase , __lowercase , **__lowercase ) lowerCAmelCase_ = weight_decay_rate lowerCAmelCase_ = include_in_weight_decay lowerCAmelCase_ = exclude_from_weight_decay @classmethod def __a ( cls , _UpperCamelCase ) -> List[Any]: lowerCAmelCase_ = {"""WarmUp""": WarmUp} return super(__lowercase , cls ).from_config(__lowercase , custom_objects=__lowercase ) def __a ( self , _UpperCamelCase , _UpperCamelCase , _UpperCamelCase ) -> Any: super(__lowercase , self )._prepare_local(__lowercase , __lowercase , __lowercase ) lowerCAmelCase_ = tf.constant( self.weight_decay_rate , name="adam_weight_decay_rate" ) def __a ( self , _UpperCamelCase , _UpperCamelCase , _UpperCamelCase ) -> Tuple: lowerCAmelCase_ = self._do_use_weight_decay(var.name ) if do_decay: return var.assign_sub( learning_rate * var * apply_state[(var.device, var.dtype.base_dtype)]["weight_decay_rate"] , use_locking=self._use_locking , ) return tf.no_op() def __a ( self , _UpperCamelCase , _UpperCamelCase=None , **_UpperCamelCase ) -> Tuple: lowerCAmelCase_ = list(zip(*__lowercase ) ) return super(__lowercase , self ).apply_gradients(zip(__lowercase , __lowercase ) , name=__lowercase , **__lowercase ) def __a ( self , _UpperCamelCase , _UpperCamelCase , _UpperCamelCase ) -> Dict: if apply_state is None: return self._decayed_lr_t[var_dtype], {} lowerCAmelCase_ = apply_state or {} lowerCAmelCase_ = apply_state.get((var_device, var_dtype) ) if coefficients is None: lowerCAmelCase_ = self._fallback_apply_state(__lowercase , __lowercase ) lowerCAmelCase_ = coefficients return coefficients["lr_t"], {"apply_state": apply_state} def __a ( self , _UpperCamelCase , _UpperCamelCase , _UpperCamelCase=None ) -> List[str]: lowerCAmelCase_ = self._get_lr(var.device , var.dtype.base_dtype , __lowercase ) lowerCAmelCase_ = self._decay_weights_op(__lowercase , __lowercase , __lowercase ) with tf.control_dependencies([decay] ): return super(__lowercase , self )._resource_apply_dense(__lowercase , __lowercase , **__lowercase ) def __a ( self , _UpperCamelCase , _UpperCamelCase , _UpperCamelCase , _UpperCamelCase=None ) -> Tuple: lowerCAmelCase_ = self._get_lr(var.device , var.dtype.base_dtype , __lowercase ) lowerCAmelCase_ = self._decay_weights_op(__lowercase , __lowercase , __lowercase ) with tf.control_dependencies([decay] ): return super(__lowercase , self )._resource_apply_sparse(__lowercase , __lowercase , __lowercase , **__lowercase ) def __a ( self ) -> int: lowerCAmelCase_ = super().get_config() config.update({"weight_decay_rate": self.weight_decay_rate} ) return config def __a ( self , _UpperCamelCase ) -> str: if self.weight_decay_rate == 0: return False if self._include_in_weight_decay: for r in self._include_in_weight_decay: if re.search(__lowercase , __lowercase ) is not None: return True if self._exclude_from_weight_decay: for r in self._exclude_from_weight_decay: if re.search(__lowercase , __lowercase ) is not None: return False return True class _lowerCAmelCase ( A__ ): def __init__( self ) -> Tuple: lowerCAmelCase_ = [] lowerCAmelCase_ = None @property def __a ( self ) -> List[str]: if self._accum_steps is None: lowerCAmelCase_ = tf.Variable( tf.constant(0 , dtype=tf.intaa ) , trainable=__lowercase , synchronization=tf.VariableSynchronization.ON_READ , aggregation=tf.VariableAggregation.ONLY_FIRST_REPLICA , ) return self._accum_steps.value() @property def __a ( self ) -> int: if not self._gradients: raise ValueError("The accumulator should be called first to initialize the gradients" ) return [gradient.value() if gradient is not None else gradient for gradient in self._gradients] def __call__( self , _UpperCamelCase ) -> List[Any]: if not self._gradients: lowerCAmelCase_ = self.step # Create the step variable. self._gradients.extend( [ tf.Variable( tf.zeros_like(__lowercase ) , trainable=__lowercase , synchronization=tf.VariableSynchronization.ON_READ , aggregation=tf.VariableAggregation.ONLY_FIRST_REPLICA , ) if gradient is not None else gradient for gradient in gradients ] ) if len(__lowercase ) != len(self._gradients ): raise ValueError(f"""Expected {len(self._gradients )} gradients, but got {len(__lowercase )}""" ) for accum_gradient, gradient in zip(self._gradients , __lowercase ): if accum_gradient is not None and gradient is not None: accum_gradient.assign_add(__lowercase ) self._accum_steps.assign_add(1 ) def __a ( self ) -> Dict: if not self._gradients: return self._accum_steps.assign(0 ) for gradient in self._gradients: if gradient is not None: gradient.assign(tf.zeros_like(__lowercase ) )
231
from __future__ import annotations import math def lowerCAmelCase_ ( _lowercase : int) -> bool: """simple docstring""" if 1 < number < 4: # 2 and 3 are primes return True elif number < 2 or number % 2 == 0 or number % 3 == 0: # Negatives, 0, 1, all even numbers, all multiples of 3 are not primes return False # All primes number are in format of 6k +/- 1 for i in range(5 , int(math.sqrt(_lowercase) + 1) , 6): if number % i == 0 or number % (i + 2) == 0: return False return True _lowercase : str =[num for num in range(3, 10_0001, 2) if not is_prime(num)] def lowerCAmelCase_ ( _lowercase : int) -> list[int]: """simple docstring""" if not isinstance(_lowercase , _lowercase): raise ValueError("""n must be an integer""") if n <= 0: raise ValueError("""n must be >= 0""") a__ : int = [] for num in range(len(_lowercase)): a__ : Any = 0 while 2 * i * i <= odd_composites[num]: a__ : Union[str, Any] = odd_composites[num] - 2 * i * i if is_prime(_lowercase): break i += 1 else: list_nums.append(odd_composites[num]) if len(_lowercase) == n: return list_nums return [] def lowerCAmelCase_ ( ) -> int: """simple docstring""" return compute_nums(1)[0] if __name__ == "__main__": print(f'{solution() = }')
170
0
import argparse import copy def lowerCamelCase__ (_UpperCAmelCase): SCREAMING_SNAKE_CASE = {} with open(_UpperCAmelCase) as f: for line in f: if line.split()[0] not in dict_of_neighbours: SCREAMING_SNAKE_CASE = [] _list.append([line.split()[1], line.split()[2]]) SCREAMING_SNAKE_CASE = _list else: dict_of_neighbours[line.split()[0]].append( [line.split()[1], line.split()[2]]) if line.split()[1] not in dict_of_neighbours: SCREAMING_SNAKE_CASE = [] _list.append([line.split()[0], line.split()[2]]) SCREAMING_SNAKE_CASE = _list else: dict_of_neighbours[line.split()[1]].append( [line.split()[0], line.split()[2]]) return dict_of_neighbours def lowerCamelCase__ (_UpperCAmelCase , _UpperCAmelCase): with open(_UpperCAmelCase) as f: SCREAMING_SNAKE_CASE = f.read(1) SCREAMING_SNAKE_CASE = start_node SCREAMING_SNAKE_CASE = [] SCREAMING_SNAKE_CASE = start_node SCREAMING_SNAKE_CASE = 0 while visiting not in first_solution: SCREAMING_SNAKE_CASE = 1_0000 for k in dict_of_neighbours[visiting]: if int(k[1]) < int(_UpperCAmelCase) and k[0] not in first_solution: SCREAMING_SNAKE_CASE = k[1] SCREAMING_SNAKE_CASE = k[0] first_solution.append(_UpperCAmelCase) SCREAMING_SNAKE_CASE = distance_of_first_solution + int(_UpperCAmelCase) SCREAMING_SNAKE_CASE = best_node first_solution.append(_UpperCAmelCase) SCREAMING_SNAKE_CASE = 0 for k in dict_of_neighbours[first_solution[-2]]: if k[0] == start_node: break position += 1 SCREAMING_SNAKE_CASE = ( distance_of_first_solution + int(dict_of_neighbours[first_solution[-2]][position][1]) - 1_0000 ) return first_solution, distance_of_first_solution def lowerCamelCase__ (_UpperCAmelCase , _UpperCAmelCase): SCREAMING_SNAKE_CASE = [] for n in solution[1:-1]: SCREAMING_SNAKE_CASE = solution.index(_UpperCAmelCase) for kn in solution[1:-1]: SCREAMING_SNAKE_CASE = solution.index(_UpperCAmelCase) if n == kn: continue SCREAMING_SNAKE_CASE = copy.deepcopy(_UpperCAmelCase) SCREAMING_SNAKE_CASE = kn SCREAMING_SNAKE_CASE = n SCREAMING_SNAKE_CASE = 0 for k in _tmp[:-1]: SCREAMING_SNAKE_CASE = _tmp[_tmp.index(_UpperCAmelCase) + 1] for i in dict_of_neighbours[k]: if i[0] == next_node: SCREAMING_SNAKE_CASE = distance + int(i[1]) _tmp.append(_UpperCAmelCase) if _tmp not in neighborhood_of_solution: neighborhood_of_solution.append(_tmp) SCREAMING_SNAKE_CASE = len(neighborhood_of_solution[0]) - 1 neighborhood_of_solution.sort(key=lambda _UpperCAmelCase: x[index_of_last_item_in_the_list]) return neighborhood_of_solution def lowerCamelCase__ (_UpperCAmelCase , _UpperCAmelCase , _UpperCAmelCase , _UpperCAmelCase , _UpperCAmelCase): SCREAMING_SNAKE_CASE = 1 SCREAMING_SNAKE_CASE = first_solution SCREAMING_SNAKE_CASE = [] SCREAMING_SNAKE_CASE = distance_of_first_solution SCREAMING_SNAKE_CASE = solution while count <= iters: SCREAMING_SNAKE_CASE = find_neighborhood(_UpperCAmelCase , _UpperCAmelCase) SCREAMING_SNAKE_CASE = 0 SCREAMING_SNAKE_CASE = neighborhood[index_of_best_solution] SCREAMING_SNAKE_CASE = len(_UpperCAmelCase) - 1 SCREAMING_SNAKE_CASE = False while not found: SCREAMING_SNAKE_CASE = 0 while i < len(_UpperCAmelCase): if best_solution[i] != solution[i]: SCREAMING_SNAKE_CASE = best_solution[i] SCREAMING_SNAKE_CASE = solution[i] break SCREAMING_SNAKE_CASE = i + 1 if [first_exchange_node, second_exchange_node] not in tabu_list and [ second_exchange_node, first_exchange_node, ] not in tabu_list: tabu_list.append([first_exchange_node, second_exchange_node]) SCREAMING_SNAKE_CASE = True SCREAMING_SNAKE_CASE = best_solution[:-1] SCREAMING_SNAKE_CASE = neighborhood[index_of_best_solution][best_cost_index] if cost < best_cost: SCREAMING_SNAKE_CASE = cost SCREAMING_SNAKE_CASE = solution else: SCREAMING_SNAKE_CASE = index_of_best_solution + 1 SCREAMING_SNAKE_CASE = neighborhood[index_of_best_solution] if len(_UpperCAmelCase) >= size: tabu_list.pop(0) SCREAMING_SNAKE_CASE = count + 1 return best_solution_ever, best_cost def lowerCamelCase__ (_UpperCAmelCase=None): SCREAMING_SNAKE_CASE = generate_neighbours(args.File) SCREAMING_SNAKE_CASE , SCREAMING_SNAKE_CASE = generate_first_solution( args.File , _UpperCAmelCase) SCREAMING_SNAKE_CASE , SCREAMING_SNAKE_CASE = tabu_search( _UpperCAmelCase , _UpperCAmelCase , _UpperCAmelCase , args.Iterations , args.Size , ) print(F'''Best solution: {best_sol}, with total distance: {best_cost}.''') if __name__ == "__main__": a_ : Optional[Any] = argparse.ArgumentParser(description='Tabu Search') parser.add_argument( '-f', '--File', type=str, help='Path to the file containing the data', required=True, ) parser.add_argument( '-i', '--Iterations', type=int, help='How many iterations the algorithm should perform', required=True, ) parser.add_argument( '-s', '--Size', type=int, help='Size of the tabu list', required=True ) # Pass the arguments to main method main(parser.parse_args())
350
import argparse import torch from torch import nn from transformers import MaMaaaConfig, MaMaaaForConditionalGeneration def lowerCamelCase__ (_UpperCAmelCase): SCREAMING_SNAKE_CASE = [ 'encoder.version', 'decoder.version', 'model.encoder.version', 'model.decoder.version', 'decoder.output_projection.weight', '_float_tensor', 'encoder.embed_positions._float_tensor', 'decoder.embed_positions._float_tensor', ] for k in ignore_keys: state_dict.pop(_UpperCAmelCase , _UpperCAmelCase) def lowerCamelCase__ (_UpperCAmelCase): SCREAMING_SNAKE_CASE , SCREAMING_SNAKE_CASE = emb.weight.shape SCREAMING_SNAKE_CASE = nn.Linear(_UpperCAmelCase , _UpperCAmelCase , bias=_UpperCAmelCase) SCREAMING_SNAKE_CASE = emb.weight.data return lin_layer def lowerCamelCase__ (_UpperCAmelCase): SCREAMING_SNAKE_CASE = torch.load(_UpperCAmelCase , map_location='cpu') SCREAMING_SNAKE_CASE = mam_aaa['args'] or mam_aaa['cfg']['model'] SCREAMING_SNAKE_CASE = mam_aaa['model'] remove_ignore_keys_(_UpperCAmelCase) SCREAMING_SNAKE_CASE = state_dict['encoder.embed_tokens.weight'].shape[0] SCREAMING_SNAKE_CASE = MaMaaaConfig( vocab_size=_UpperCAmelCase , max_position_embeddings=1024 , encoder_layers=args.encoder_layers , decoder_layers=args.decoder_layers , encoder_attention_heads=args.encoder_attention_heads , decoder_attention_heads=args.decoder_attention_heads , encoder_ffn_dim=args.encoder_ffn_embed_dim , decoder_ffn_dim=args.decoder_ffn_embed_dim , d_model=args.encoder_embed_dim , encoder_layerdrop=args.encoder_layerdrop , decoder_layerdrop=args.decoder_layerdrop , dropout=args.dropout , attention_dropout=args.attention_dropout , activation_dropout=args.activation_dropout , activation_function='relu' , ) SCREAMING_SNAKE_CASE = state_dict['decoder.embed_tokens.weight'] SCREAMING_SNAKE_CASE = MaMaaaForConditionalGeneration(_UpperCAmelCase) model.model.load_state_dict(_UpperCAmelCase , strict=_UpperCAmelCase) SCREAMING_SNAKE_CASE = make_linear_from_emb(model.model.shared) return model if __name__ == "__main__": a_ : int = argparse.ArgumentParser() # Required parameters parser.add_argument('fairseq_path', type=str, help='path to a model.pt on local filesystem.') parser.add_argument('pytorch_dump_folder_path', default=None, type=str, help='Path to the output PyTorch model.') a_ : List[str] = parser.parse_args() a_ : Dict = convert_fairseq_mamaaa_checkpoint_from_disk(args.fairseq_pathß) model.save_pretrained(args.pytorch_dump_folder_path)
327
0
import copy from typing import Any, Dict, List, Optional, Union import numpy as np import torch from ...audio_utils import mel_filter_bank, spectrogram, window_function from ...feature_extraction_sequence_utils import SequenceFeatureExtractor from ...feature_extraction_utils import BatchFeature from ...utils import TensorType, logging lowerCamelCase = logging.get_logger(__name__) class _a ( _lowercase): _a : List[str] = ['input_features', 'is_longer'] def __init__( self : Dict , _SCREAMING_SNAKE_CASE : Tuple=64 , _SCREAMING_SNAKE_CASE : List[str]=4_8000 , _SCREAMING_SNAKE_CASE : List[str]=480 , _SCREAMING_SNAKE_CASE : Union[str, Any]=10 , _SCREAMING_SNAKE_CASE : Optional[int]=1024 , _SCREAMING_SNAKE_CASE : Union[str, Any]=0.0 , _SCREAMING_SNAKE_CASE : Optional[int]=False , _SCREAMING_SNAKE_CASE : Any = 0 , _SCREAMING_SNAKE_CASE : List[Any] = 1_4000 , _SCREAMING_SNAKE_CASE : Optional[int] = None , _SCREAMING_SNAKE_CASE : Optional[int] = "fusion" , _SCREAMING_SNAKE_CASE : Tuple = "repeatpad" , **_SCREAMING_SNAKE_CASE : Dict , )-> Any: super().__init__( feature_size=SCREAMING_SNAKE_CASE_ , sampling_rate=SCREAMING_SNAKE_CASE_ , padding_value=SCREAMING_SNAKE_CASE_ , return_attention_mask=SCREAMING_SNAKE_CASE_ , **SCREAMING_SNAKE_CASE_ , ) lowerCAmelCase__ : Tuple = top_db lowerCAmelCase__ : Union[str, Any] = truncation lowerCAmelCase__ : Dict = padding lowerCAmelCase__ : List[Any] = fft_window_size lowerCAmelCase__ : List[Any] = (fft_window_size >> 1) + 1 lowerCAmelCase__ : Tuple = hop_length lowerCAmelCase__ : List[Any] = max_length_s lowerCAmelCase__ : str = max_length_s * sampling_rate lowerCAmelCase__ : List[Any] = sampling_rate lowerCAmelCase__ : Any = frequency_min lowerCAmelCase__ : Dict = frequency_max lowerCAmelCase__ : Optional[Any] = mel_filter_bank( num_frequency_bins=self.nb_frequency_bins , num_mel_filters=SCREAMING_SNAKE_CASE_ , min_frequency=SCREAMING_SNAKE_CASE_ , max_frequency=SCREAMING_SNAKE_CASE_ , sampling_rate=SCREAMING_SNAKE_CASE_ , norm=SCREAMING_SNAKE_CASE_ , mel_scale='''htk''' , ) lowerCAmelCase__ : Any = mel_filter_bank( num_frequency_bins=self.nb_frequency_bins , num_mel_filters=SCREAMING_SNAKE_CASE_ , min_frequency=SCREAMING_SNAKE_CASE_ , max_frequency=SCREAMING_SNAKE_CASE_ , sampling_rate=SCREAMING_SNAKE_CASE_ , norm='''slaney''' , mel_scale='''slaney''' , ) def UpperCAmelCase__( self : Tuple )-> Dict[str, Any]: lowerCAmelCase__ : int = copy.deepcopy(self.__dict__ ) lowerCAmelCase__ : Optional[Any] = self.__class__.__name__ if "mel_filters" in output: del output["mel_filters"] if "mel_filters_slaney" in output: del output["mel_filters_slaney"] return output def UpperCAmelCase__( self : Dict , _SCREAMING_SNAKE_CASE : Dict , _SCREAMING_SNAKE_CASE : Tuple = None )-> np.ndarray: lowerCAmelCase__ : Tuple = spectrogram( SCREAMING_SNAKE_CASE_ , window_function(self.fft_window_size , '''hann''' ) , frame_length=self.fft_window_size , hop_length=self.hop_length , power=2.0 , mel_filters=SCREAMING_SNAKE_CASE_ , log_mel='''dB''' , ) return log_mel_spectrogram.T def UpperCAmelCase__( self : Optional[Any] , _SCREAMING_SNAKE_CASE : Tuple , _SCREAMING_SNAKE_CASE : Union[str, Any] , _SCREAMING_SNAKE_CASE : Optional[int] )-> Dict: lowerCAmelCase__ : int = np.array_split(list(range(0 , total_frames - chunk_frames + 1 ) ) , 3 ) if len(ranges[1] ) == 0: # if the audio is too short, we just use the first chunk lowerCAmelCase__ : Tuple = [0] if len(ranges[2] ) == 0: # if the audio is too short, we just use the first chunk lowerCAmelCase__ : Optional[Any] = [0] # randomly choose index for each part lowerCAmelCase__ : Optional[int] = np.random.choice(ranges[0] ) lowerCAmelCase__ : Union[str, Any] = np.random.choice(ranges[1] ) lowerCAmelCase__ : List[Any] = np.random.choice(ranges[2] ) lowerCAmelCase__ : List[str] = mel[idx_front : idx_front + chunk_frames, :] lowerCAmelCase__ : List[Any] = mel[idx_middle : idx_middle + chunk_frames, :] lowerCAmelCase__ : Tuple = mel[idx_back : idx_back + chunk_frames, :] lowerCAmelCase__ : int = torch.tensor(mel[None, None, :] ) lowerCAmelCase__ : Tuple = torch.nn.functional.interpolate( SCREAMING_SNAKE_CASE_ , size=[chunk_frames, 64] , mode='''bilinear''' , align_corners=SCREAMING_SNAKE_CASE_ ) lowerCAmelCase__ : Optional[int] = mel_shrink[0][0].numpy() lowerCAmelCase__ : List[Any] = np.stack([mel_shrink, mel_chunk_front, mel_chunk_middle, mel_chunk_back] , axis=0 ) return mel_fusion def UpperCAmelCase__( self : Dict , _SCREAMING_SNAKE_CASE : Union[str, Any] , _SCREAMING_SNAKE_CASE : List[str] , _SCREAMING_SNAKE_CASE : List[Any] , _SCREAMING_SNAKE_CASE : Union[str, Any] )-> np.array: if waveform.shape[0] > max_length: if truncation == "rand_trunc": lowerCAmelCase__ : List[Any] = True # random crop to max_length (for compatibility) -> this should be handled by self.pad lowerCAmelCase__ : List[str] = len(SCREAMING_SNAKE_CASE_ ) - max_length lowerCAmelCase__ : Dict = np.random.randint(0 , overflow + 1 ) lowerCAmelCase__ : List[str] = waveform[idx : idx + max_length] lowerCAmelCase__ : str = self._np_extract_fbank_features(SCREAMING_SNAKE_CASE_ , self.mel_filters_slaney )[None, :] elif truncation == "fusion": lowerCAmelCase__ : Any = self._np_extract_fbank_features(SCREAMING_SNAKE_CASE_ , self.mel_filters ) lowerCAmelCase__ : Any = max_length // self.hop_length + 1 # the +1 related to how the spectrogram is computed lowerCAmelCase__ : str = mel.shape[0] if chunk_frames == total_frames: # there is a corner case where the audio length is larger than max_length but smaller than max_length+hop_length. # In this case, we just use the whole audio. lowerCAmelCase__ : Dict = np.stack([mel, mel, mel, mel] , axis=0 ) lowerCAmelCase__ : Optional[int] = False else: lowerCAmelCase__ : Optional[Any] = self._random_mel_fusion(SCREAMING_SNAKE_CASE_ , SCREAMING_SNAKE_CASE_ , SCREAMING_SNAKE_CASE_ ) lowerCAmelCase__ : Optional[int] = True else: raise NotImplementedError(F'data_truncating {truncation} not implemented' ) else: lowerCAmelCase__ : List[Any] = False # only use repeat as a new possible value for padding. you repeat the audio before applying the usual max_length padding if waveform.shape[0] < max_length: if padding == "repeat": lowerCAmelCase__ : int = int(max_length / len(SCREAMING_SNAKE_CASE_ ) ) lowerCAmelCase__ : Optional[Any] = np.stack(np.tile(SCREAMING_SNAKE_CASE_ , n_repeat + 1 ) )[:max_length] if padding == "repeatpad": lowerCAmelCase__ : Optional[int] = int(max_length / len(SCREAMING_SNAKE_CASE_ ) ) lowerCAmelCase__ : str = np.stack(np.tile(SCREAMING_SNAKE_CASE_ , SCREAMING_SNAKE_CASE_ ) ) lowerCAmelCase__ : Any = np.pad(SCREAMING_SNAKE_CASE_ , (0, max_length - waveform.shape[0]) , mode='''constant''' , constant_values=0 ) if truncation == "fusion": lowerCAmelCase__ : str = self._np_extract_fbank_features(SCREAMING_SNAKE_CASE_ , self.mel_filters ) lowerCAmelCase__ : Union[str, Any] = np.stack([input_mel, input_mel, input_mel, input_mel] , axis=0 ) else: lowerCAmelCase__ : Union[str, Any] = self._np_extract_fbank_features(SCREAMING_SNAKE_CASE_ , self.mel_filters_slaney )[None, :] return input_mel, longer def __call__( self : int , _SCREAMING_SNAKE_CASE : List[Any] , _SCREAMING_SNAKE_CASE : Optional[int] = None , _SCREAMING_SNAKE_CASE : Optional[Any] = None , _SCREAMING_SNAKE_CASE : Any = None , _SCREAMING_SNAKE_CASE : Optional[Any] = None , _SCREAMING_SNAKE_CASE : List[str] = None , **_SCREAMING_SNAKE_CASE : str , )-> BatchFeature: lowerCAmelCase__ : Dict = truncation if truncation is not None else self.truncation lowerCAmelCase__ : Dict = padding if padding else self.padding if sampling_rate is not None: if sampling_rate != self.sampling_rate: raise ValueError( F'The model corresponding to this feature extractor: {self.__class__.__name__} was trained using a' F' sampling rate of {self.sampling_rate}. Please make sure that the provided `raw_speech` input' F' was sampled with {self.sampling_rate} and not {sampling_rate}.' ) else: logger.warning( '''It is strongly recommended to pass the `sampling_rate` argument to this function. ''' '''Failing to do so can result in silent errors that might be hard to debug.''' ) lowerCAmelCase__ : Dict = isinstance(SCREAMING_SNAKE_CASE_ , np.ndarray ) and len(raw_speech.shape ) > 1 if is_batched_numpy and len(raw_speech.shape ) > 2: raise ValueError(F'Only mono-channel audio is supported for input to {self}' ) lowerCAmelCase__ : str = is_batched_numpy or ( isinstance(SCREAMING_SNAKE_CASE_ , (list, tuple) ) and (isinstance(raw_speech[0] , (np.ndarray, tuple, list) )) ) if is_batched: lowerCAmelCase__ : Optional[Any] = [np.asarray(SCREAMING_SNAKE_CASE_ , dtype=np.floataa ) for speech in raw_speech] elif not is_batched and not isinstance(SCREAMING_SNAKE_CASE_ , np.ndarray ): lowerCAmelCase__ : str = np.asarray(SCREAMING_SNAKE_CASE_ , dtype=np.floataa ) elif isinstance(SCREAMING_SNAKE_CASE_ , np.ndarray ) and raw_speech.dtype is np.dtype(np.floataa ): lowerCAmelCase__ : str = raw_speech.astype(np.floataa ) # always return batch if not is_batched: lowerCAmelCase__ : Optional[Any] = [np.asarray(SCREAMING_SNAKE_CASE_ )] # convert to mel spectrogram, truncate and pad if needed. lowerCAmelCase__ : str = [ self._get_input_mel(SCREAMING_SNAKE_CASE_ , max_length if max_length else self.nb_max_samples , SCREAMING_SNAKE_CASE_ , SCREAMING_SNAKE_CASE_ ) for waveform in raw_speech ] lowerCAmelCase__ : Union[str, Any] = [] lowerCAmelCase__ : List[Any] = [] for mel, longer in padded_inputs: input_mel.append(SCREAMING_SNAKE_CASE_ ) is_longer.append(SCREAMING_SNAKE_CASE_ ) if truncation == "fusion" and sum(SCREAMING_SNAKE_CASE_ ) == 0: # if no audio is longer than 10s, then randomly select one audio to be longer lowerCAmelCase__ : Any = np.random.randint(0 , len(SCREAMING_SNAKE_CASE_ ) ) lowerCAmelCase__ : Tuple = True if isinstance(input_mel[0] , SCREAMING_SNAKE_CASE_ ): lowerCAmelCase__ : int = [np.asarray(SCREAMING_SNAKE_CASE_ , dtype=np.floataa ) for feature in input_mel] # is_longer is a list of bool lowerCAmelCase__ : Optional[Any] = [[longer] for longer in is_longer] lowerCAmelCase__ : Optional[Any] = {'''input_features''': input_mel, '''is_longer''': is_longer} lowerCAmelCase__ : Optional[int] = BatchFeature(SCREAMING_SNAKE_CASE_ ) if return_tensors is not None: lowerCAmelCase__ : Tuple = input_features.convert_to_tensors(SCREAMING_SNAKE_CASE_ ) return input_features
131
def _A ( SCREAMING_SNAKE_CASE__ : list[int] , SCREAMING_SNAKE_CASE__ : list[int] ): UpperCamelCase :Tuple = len(SCREAMING_SNAKE_CASE__ ) print('''The following activities are selected:''' ) # The first activity is always selected UpperCamelCase :Dict = 0 print(SCREAMING_SNAKE_CASE__ , end=''',''' ) # Consider rest of the activities for j in range(SCREAMING_SNAKE_CASE__ ): # If this activity has start time greater than # or equal to the finish time of previously # selected activity, then select it if start[j] >= finish[i]: print(SCREAMING_SNAKE_CASE__ , end=''',''' ) UpperCamelCase :List[str] = j if __name__ == "__main__": import doctest doctest.testmod() __snake_case = [1, 3, 0, 5, 8, 5] __snake_case = [2, 4, 6, 7, 9, 9] print_max_activities(start, finish)
259
0
import os import sys from contextlib import contextmanager # Windows only if os.name == "nt": import ctypes import msvcrt # noqa class A__ ( ctypes.Structure ): # _fields is a specific attr expected by ctypes lowercase = [('size', ctypes.c_int), ('visible', ctypes.c_byte)] def lowerCAmelCase__ ( ) -> Optional[int]: if os.name == "nt": lowerCAmelCase__ : List[str] = CursorInfo() lowerCAmelCase__ : int = ctypes.windll.kernelaa.GetStdHandle(-11 ) ctypes.windll.kernelaa.GetConsoleCursorInfo(SCREAMING_SNAKE_CASE_ , ctypes.byref(SCREAMING_SNAKE_CASE_ ) ) lowerCAmelCase__ : Union[str, Any] = False ctypes.windll.kernelaa.SetConsoleCursorInfo(SCREAMING_SNAKE_CASE_ , ctypes.byref(SCREAMING_SNAKE_CASE_ ) ) elif os.name == "posix": sys.stdout.write('\033[?25l' ) sys.stdout.flush() def lowerCAmelCase__ ( ) -> List[Any]: if os.name == "nt": lowerCAmelCase__ : Any = CursorInfo() lowerCAmelCase__ : str = ctypes.windll.kernelaa.GetStdHandle(-11 ) ctypes.windll.kernelaa.GetConsoleCursorInfo(SCREAMING_SNAKE_CASE_ , ctypes.byref(SCREAMING_SNAKE_CASE_ ) ) lowerCAmelCase__ : List[str] = True ctypes.windll.kernelaa.SetConsoleCursorInfo(SCREAMING_SNAKE_CASE_ , ctypes.byref(SCREAMING_SNAKE_CASE_ ) ) elif os.name == "posix": sys.stdout.write('\033[?25h' ) sys.stdout.flush() @contextmanager def lowerCAmelCase__ ( ) -> List[Any]: try: hide_cursor() yield finally: show_cursor()
368
import gc import unittest import numpy as np import torch from diffusers import DanceDiffusionPipeline, IPNDMScheduler, UNetaDModel from diffusers.utils import slow, torch_device from diffusers.utils.testing_utils import enable_full_determinism, require_torch_gpu, skip_mps from ..pipeline_params import UNCONDITIONAL_AUDIO_GENERATION_BATCH_PARAMS, UNCONDITIONAL_AUDIO_GENERATION_PARAMS from ..test_pipelines_common import PipelineTesterMixin enable_full_determinism() class A__ ( __magic_name__ , unittest.TestCase ): lowercase = DanceDiffusionPipeline lowercase = UNCONDITIONAL_AUDIO_GENERATION_PARAMS lowercase = PipelineTesterMixin.required_optional_params - { 'callback', 'latents', 'callback_steps', 'output_type', 'num_images_per_prompt', } lowercase = UNCONDITIONAL_AUDIO_GENERATION_BATCH_PARAMS lowercase = False lowercase = False def _lowerCamelCase ( self : List[str] ): '''simple docstring''' torch.manual_seed(0 ) lowerCAmelCase__ : Optional[int] = UNetaDModel( block_out_channels=(32, 32, 64) , extra_in_channels=16 , sample_size=512 , sample_rate=16_000 , in_channels=2 , out_channels=2 , flip_sin_to_cos=a , use_timestep_embedding=a , time_embedding_type='fourier' , mid_block_type='UNetMidBlock1D' , down_block_types=('DownBlock1DNoSkip', 'DownBlock1D', 'AttnDownBlock1D') , up_block_types=('AttnUpBlock1D', 'UpBlock1D', 'UpBlock1DNoSkip') , ) lowerCAmelCase__ : Tuple = IPNDMScheduler() lowerCAmelCase__ : str = { 'unet': unet, 'scheduler': scheduler, } return components def _lowerCamelCase ( self : int , a : Dict , a : List[str]=0 ): '''simple docstring''' if str(a ).startswith('mps' ): lowerCAmelCase__ : Union[str, Any] = torch.manual_seed(a ) else: lowerCAmelCase__ : Optional[Any] = torch.Generator(device=a ).manual_seed(a ) lowerCAmelCase__ : Optional[Any] = { 'batch_size': 1, 'generator': generator, 'num_inference_steps': 4, } return inputs def _lowerCamelCase ( self : Optional[Any] ): '''simple docstring''' lowerCAmelCase__ : Optional[Any] = 'cpu' # ensure determinism for the device-dependent torch.Generator lowerCAmelCase__ : int = self.get_dummy_components() lowerCAmelCase__ : List[str] = DanceDiffusionPipeline(**a ) lowerCAmelCase__ : Any = pipe.to(a ) pipe.set_progress_bar_config(disable=a ) lowerCAmelCase__ : List[str] = self.get_dummy_inputs(a ) lowerCAmelCase__ : List[Any] = pipe(**a ) lowerCAmelCase__ : List[str] = output.audios lowerCAmelCase__ : Optional[Any] = audio[0, -3:, -3:] assert audio.shape == (1, 2, components["unet"].sample_size) lowerCAmelCase__ : List[Any] = np.array([-0.7_2_6_5, 1.0_0_0_0, -0.8_3_8_8, 0.1_1_7_5, 0.9_4_9_8, -1.0_0_0_0] ) assert np.abs(audio_slice.flatten() - expected_slice ).max() < 1E-2 @skip_mps def _lowerCamelCase ( self : str ): '''simple docstring''' return super().test_save_load_local() @skip_mps def _lowerCamelCase ( self : Tuple ): '''simple docstring''' return super().test_dict_tuple_outputs_equivalent(expected_max_difference=3E-3 ) @skip_mps def _lowerCamelCase ( self : List[str] ): '''simple docstring''' return super().test_save_load_optional_components() @skip_mps def _lowerCamelCase ( self : Tuple ): '''simple docstring''' return super().test_attention_slicing_forward_pass() def _lowerCamelCase ( self : List[str] ): '''simple docstring''' super().test_inference_batch_single_identical(expected_max_diff=3E-3 ) @slow @require_torch_gpu class A__ ( unittest.TestCase ): def _lowerCamelCase ( self : Dict ): '''simple docstring''' super().tearDown() gc.collect() torch.cuda.empty_cache() def _lowerCamelCase ( self : Union[str, Any] ): '''simple docstring''' lowerCAmelCase__ : Tuple = torch_device lowerCAmelCase__ : List[str] = DanceDiffusionPipeline.from_pretrained('harmonai/maestro-150k' ) lowerCAmelCase__ : List[str] = pipe.to(a ) pipe.set_progress_bar_config(disable=a ) lowerCAmelCase__ : Optional[int] = torch.manual_seed(0 ) lowerCAmelCase__ : Optional[int] = pipe(generator=a , num_inference_steps=100 , audio_length_in_s=4.0_9_6 ) lowerCAmelCase__ : int = output.audios lowerCAmelCase__ : List[Any] = audio[0, -3:, -3:] assert audio.shape == (1, 2, pipe.unet.sample_size) lowerCAmelCase__ : Dict = np.array([-0.0_1_9_2, -0.0_2_3_1, -0.0_3_1_8, -0.0_0_5_9, 0.0_0_0_2, -0.0_0_2_0] ) assert np.abs(audio_slice.flatten() - expected_slice ).max() < 1E-2 def _lowerCamelCase ( self : Tuple ): '''simple docstring''' lowerCAmelCase__ : str = torch_device lowerCAmelCase__ : List[Any] = DanceDiffusionPipeline.from_pretrained('harmonai/maestro-150k' , torch_dtype=torch.floataa ) lowerCAmelCase__ : Optional[int] = pipe.to(a ) pipe.set_progress_bar_config(disable=a ) lowerCAmelCase__ : str = torch.manual_seed(0 ) lowerCAmelCase__ : Optional[int] = pipe(generator=a , num_inference_steps=100 , audio_length_in_s=4.0_9_6 ) lowerCAmelCase__ : str = output.audios lowerCAmelCase__ : Tuple = audio[0, -3:, -3:] assert audio.shape == (1, 2, pipe.unet.sample_size) lowerCAmelCase__ : int = np.array([-0.0_3_6_7, -0.0_4_8_8, -0.0_7_7_1, -0.0_5_2_5, -0.0_4_4_4, -0.0_3_4_1] ) assert np.abs(audio_slice.flatten() - expected_slice ).max() < 1E-2
307
0
'''simple docstring''' import re from typing import Callable, List, Optional, Union import tensorflow as tf try: from tensorflow.keras.optimizers.legacy import Adam except ImportError: from tensorflow.keras.optimizers import Adam class lowerCAmelCase__ ( tf.keras.optimizers.schedules.LearningRateSchedule ): def __init__( self : int , lowerCamelCase__ : float , lowerCamelCase__ : Callable , lowerCamelCase__ : int , lowerCamelCase__ : float = 1.0 , lowerCamelCase__ : str = None , ) ->int: '''simple docstring''' super().__init__() _UpperCAmelCase : int = initial_learning_rate _UpperCAmelCase : Any = warmup_steps _UpperCAmelCase : Dict = power _UpperCAmelCase : int = decay_schedule_fn _UpperCAmelCase : List[str] = name def __call__( self : Dict , lowerCamelCase__ : Tuple ) ->Any: '''simple docstring''' with tf.name_scope(self.name or "WarmUp" ) as name: # Implements polynomial warmup. i.e., if global_step < warmup_steps, the # learning rate will be `global_step/num_warmup_steps * init_lr`. _UpperCAmelCase : Dict = tf.cast(lowerCamelCase__ , tf.floataa ) _UpperCAmelCase : Union[str, Any] = tf.cast(self.warmup_steps , tf.floataa ) _UpperCAmelCase : int = global_step_float / warmup_steps_float _UpperCAmelCase : Optional[Any] = self.initial_learning_rate * tf.math.pow(lowerCamelCase__ , self.power ) return tf.cond( global_step_float < warmup_steps_float , lambda: warmup_learning_rate , lambda: self.decay_schedule_fn(step - self.warmup_steps ) , name=lowerCamelCase__ , ) def lowerCAmelCase__ ( self : Tuple ) ->List[Any]: '''simple docstring''' return { "initial_learning_rate": self.initial_learning_rate, "decay_schedule_fn": self.decay_schedule_fn, "warmup_steps": self.warmup_steps, "power": self.power, "name": self.name, } def __lowerCAmelCase (__lowerCAmelCase , __lowerCAmelCase , __lowerCAmelCase , __lowerCAmelCase = 0.0 , __lowerCAmelCase = 0.9 , __lowerCAmelCase = 0.9_9_9 , __lowerCAmelCase = 1e-8 , __lowerCAmelCase = None , __lowerCAmelCase = None , __lowerCAmelCase = 0.0 , __lowerCAmelCase = 1.0 , __lowerCAmelCase = None , ): _UpperCAmelCase : Union[str, Any] = tf.keras.optimizers.schedules.PolynomialDecay( initial_learning_rate=__lowerCAmelCase , decay_steps=num_train_steps - num_warmup_steps , end_learning_rate=init_lr * min_lr_ratio , power=__lowerCAmelCase , ) if num_warmup_steps: _UpperCAmelCase : Union[str, Any] = WarmUp( initial_learning_rate=__lowerCAmelCase , decay_schedule_fn=__lowerCAmelCase , warmup_steps=__lowerCAmelCase , ) if weight_decay_rate > 0.0: _UpperCAmelCase : int = AdamWeightDecay( learning_rate=__lowerCAmelCase , weight_decay_rate=__lowerCAmelCase , beta_a=__lowerCAmelCase , beta_a=__lowerCAmelCase , epsilon=__lowerCAmelCase , clipnorm=__lowerCAmelCase , global_clipnorm=__lowerCAmelCase , exclude_from_weight_decay=["LayerNorm", "layer_norm", "bias"] , include_in_weight_decay=__lowerCAmelCase , ) else: _UpperCAmelCase : Optional[Any] = tf.keras.optimizers.Adam( learning_rate=__lowerCAmelCase , beta_a=__lowerCAmelCase , beta_a=__lowerCAmelCase , epsilon=__lowerCAmelCase , clipnorm=__lowerCAmelCase , global_clipnorm=__lowerCAmelCase , ) # We return the optimizer and the LR scheduler in order to better track the # evolution of the LR independently of the optimizer. return optimizer, lr_schedule class lowerCAmelCase__ ( UpperCAmelCase__ ): def __init__( self : Dict , lowerCamelCase__ : Union[float, tf.keras.optimizers.schedules.LearningRateSchedule] = 0.0_0_1 , lowerCamelCase__ : float = 0.9 , lowerCamelCase__ : float = 0.9_9_9 , lowerCamelCase__ : float = 1E-7 , lowerCamelCase__ : bool = False , lowerCamelCase__ : float = 0.0 , lowerCamelCase__ : Optional[List[str]] = None , lowerCamelCase__ : Optional[List[str]] = None , lowerCamelCase__ : str = "AdamWeightDecay" , **lowerCamelCase__ : List[Any] , ) ->Tuple: '''simple docstring''' super().__init__(lowerCamelCase__ , lowerCamelCase__ , lowerCamelCase__ , lowerCamelCase__ , lowerCamelCase__ , lowerCamelCase__ , **lowerCamelCase__ ) _UpperCAmelCase : Optional[int] = weight_decay_rate _UpperCAmelCase : int = include_in_weight_decay _UpperCAmelCase : int = exclude_from_weight_decay @classmethod def lowerCAmelCase__ ( cls : Tuple , lowerCamelCase__ : List[Any] ) ->List[str]: '''simple docstring''' _UpperCAmelCase : str = {"WarmUp": WarmUp} return super(lowerCamelCase__ , cls ).from_config(lowerCamelCase__ , custom_objects=lowerCamelCase__ ) def lowerCAmelCase__ ( self : Optional[int] , lowerCamelCase__ : List[str] , lowerCamelCase__ : Optional[Any] , lowerCamelCase__ : Tuple ) ->int: '''simple docstring''' super(lowerCamelCase__ , self )._prepare_local(lowerCamelCase__ , lowerCamelCase__ , lowerCamelCase__ ) _UpperCAmelCase : str = tf.constant( self.weight_decay_rate , name="adam_weight_decay_rate" ) def lowerCAmelCase__ ( self : Dict , lowerCamelCase__ : Tuple , lowerCamelCase__ : Optional[Any] , lowerCamelCase__ : Optional[int] ) ->List[str]: '''simple docstring''' _UpperCAmelCase : str = self._do_use_weight_decay(var.name ) if do_decay: return var.assign_sub( learning_rate * var * apply_state[(var.device, var.dtype.base_dtype)]["weight_decay_rate"] , use_locking=self._use_locking , ) return tf.no_op() def lowerCAmelCase__ ( self : List[str] , lowerCamelCase__ : Any , lowerCamelCase__ : List[Any]=None , **lowerCamelCase__ : Any ) ->Dict: '''simple docstring''' _UpperCAmelCase , _UpperCAmelCase : Any = list(zip(*lowerCamelCase__ ) ) return super(lowerCamelCase__ , self ).apply_gradients(zip(lowerCamelCase__ , lowerCamelCase__ ) , name=lowerCamelCase__ , **lowerCamelCase__ ) def lowerCAmelCase__ ( self : Dict , lowerCamelCase__ : Any , lowerCamelCase__ : List[Any] , lowerCamelCase__ : int ) ->Tuple: '''simple docstring''' if apply_state is None: return self._decayed_lr_t[var_dtype], {} _UpperCAmelCase : Tuple = apply_state or {} _UpperCAmelCase : Optional[int] = apply_state.get((var_device, var_dtype) ) if coefficients is None: _UpperCAmelCase : List[str] = self._fallback_apply_state(lowerCamelCase__ , lowerCamelCase__ ) _UpperCAmelCase : Any = coefficients return coefficients["lr_t"], {"apply_state": apply_state} def lowerCAmelCase__ ( self : Union[str, Any] , lowerCamelCase__ : Optional[int] , lowerCamelCase__ : Tuple , lowerCamelCase__ : Any=None ) ->Dict: '''simple docstring''' _UpperCAmelCase , _UpperCAmelCase : int = self._get_lr(var.device , var.dtype.base_dtype , lowerCamelCase__ ) _UpperCAmelCase : Any = self._decay_weights_op(lowerCamelCase__ , lowerCamelCase__ , lowerCamelCase__ ) with tf.control_dependencies([decay] ): return super(lowerCamelCase__ , self )._resource_apply_dense(lowerCamelCase__ , lowerCamelCase__ , **lowerCamelCase__ ) def lowerCAmelCase__ ( self : List[Any] , lowerCamelCase__ : int , lowerCamelCase__ : Union[str, Any] , lowerCamelCase__ : str , lowerCamelCase__ : Any=None ) ->Any: '''simple docstring''' _UpperCAmelCase , _UpperCAmelCase : int = self._get_lr(var.device , var.dtype.base_dtype , lowerCamelCase__ ) _UpperCAmelCase : int = self._decay_weights_op(lowerCamelCase__ , lowerCamelCase__ , lowerCamelCase__ ) with tf.control_dependencies([decay] ): return super(lowerCamelCase__ , self )._resource_apply_sparse(lowerCamelCase__ , lowerCamelCase__ , lowerCamelCase__ , **lowerCamelCase__ ) def lowerCAmelCase__ ( self : Union[str, Any] ) ->Any: '''simple docstring''' _UpperCAmelCase : List[str] = super().get_config() config.update({"weight_decay_rate": self.weight_decay_rate} ) return config def lowerCAmelCase__ ( self : str , lowerCamelCase__ : str ) ->Optional[Any]: '''simple docstring''' if self.weight_decay_rate == 0: return False if self._include_in_weight_decay: for r in self._include_in_weight_decay: if re.search(lowerCamelCase__ , lowerCamelCase__ ) is not None: return True if self._exclude_from_weight_decay: for r in self._exclude_from_weight_decay: if re.search(lowerCamelCase__ , lowerCamelCase__ ) is not None: return False return True class lowerCAmelCase__ ( UpperCAmelCase__ ): def __init__( self : Union[str, Any] ) ->int: '''simple docstring''' _UpperCAmelCase : Dict = [] _UpperCAmelCase : int = None @property def lowerCAmelCase__ ( self : Optional[Any] ) ->Tuple: '''simple docstring''' if self._accum_steps is None: _UpperCAmelCase : str = tf.Variable( tf.constant(0 , dtype=tf.intaa ) , trainable=lowerCamelCase__ , synchronization=tf.VariableSynchronization.ON_READ , aggregation=tf.VariableAggregation.ONLY_FIRST_REPLICA , ) return self._accum_steps.value() @property def lowerCAmelCase__ ( self : Union[str, Any] ) ->Optional[int]: '''simple docstring''' if not self._gradients: raise ValueError("The accumulator should be called first to initialize the gradients" ) return [gradient.value() if gradient is not None else gradient for gradient in self._gradients] def __call__( self : Optional[Any] , lowerCamelCase__ : Optional[int] ) ->str: '''simple docstring''' if not self._gradients: _UpperCAmelCase : Tuple = self.step # Create the step variable. self._gradients.extend( [ tf.Variable( tf.zeros_like(lowerCamelCase__ ) , trainable=lowerCamelCase__ , synchronization=tf.VariableSynchronization.ON_READ , aggregation=tf.VariableAggregation.ONLY_FIRST_REPLICA , ) if gradient is not None else gradient for gradient in gradients ] ) if len(lowerCamelCase__ ) != len(self._gradients ): raise ValueError(F"""Expected {len(self._gradients )} gradients, but got {len(lowerCamelCase__ )}""" ) for accum_gradient, gradient in zip(self._gradients , lowerCamelCase__ ): if accum_gradient is not None and gradient is not None: accum_gradient.assign_add(lowerCamelCase__ ) self._accum_steps.assign_add(1 ) def lowerCAmelCase__ ( self : Dict ) ->Tuple: '''simple docstring''' if not self._gradients: return self._accum_steps.assign(0 ) for gradient in self._gradients: if gradient is not None: gradient.assign(tf.zeros_like(lowerCamelCase__ ) )
234
'''simple docstring''' import os from shutil import copyfile from typing import List, Optional, Tuple from ...tokenization_utils import AddedToken from ...tokenization_utils_fast import PreTrainedTokenizerFast from ...utils import is_sentencepiece_available, logging if is_sentencepiece_available(): from .tokenization_fnet import FNetTokenizer else: lowerCamelCase__ = None lowerCamelCase__ = logging.get_logger(__name__) lowerCamelCase__ = {'vocab_file': 'spiece.model', 'tokenizer_file': 'tokenizer.json'} lowerCamelCase__ = { 'vocab_file': { 'google/fnet-base': 'https://huggingface.co/google/fnet-base/resolve/main/spiece.model', 'google/fnet-large': 'https://huggingface.co/google/fnet-large/resolve/main/spiece.model', }, 'tokenizer_file': { 'google/fnet-base': 'https://huggingface.co/google/fnet-base/resolve/main/tokenizer.json', 'google/fnet-large': 'https://huggingface.co/google/fnet-large/resolve/main/tokenizer.json', }, } lowerCamelCase__ = { 'google/fnet-base': 512, 'google/fnet-large': 512, } lowerCamelCase__ = '▁' class lowerCAmelCase__ ( UpperCAmelCase__ ): lowerCAmelCase : Dict = VOCAB_FILES_NAMES lowerCAmelCase : Optional[Any] = PRETRAINED_VOCAB_FILES_MAP lowerCAmelCase : List[str] = PRETRAINED_POSITIONAL_EMBEDDINGS_SIZES lowerCAmelCase : Optional[int] = ["input_ids", "token_type_ids"] lowerCAmelCase : Optional[Any] = FNetTokenizer def __init__( self : Dict , lowerCamelCase__ : int=None , lowerCamelCase__ : List[Any]=None , lowerCamelCase__ : Any=False , lowerCamelCase__ : List[str]=True , lowerCamelCase__ : Dict=True , lowerCamelCase__ : str="<unk>" , lowerCamelCase__ : List[str]="[SEP]" , lowerCamelCase__ : Union[str, Any]="<pad>" , lowerCamelCase__ : Optional[Any]="[CLS]" , lowerCamelCase__ : Any="[MASK]" , **lowerCamelCase__ : Any , ) ->List[Any]: '''simple docstring''' _UpperCAmelCase : Dict = ( AddedToken(lowerCamelCase__ , lstrip=lowerCamelCase__ , rstrip=lowerCamelCase__ , normalized=lowerCamelCase__ ) if isinstance(lowerCamelCase__ , lowerCamelCase__ ) else mask_token ) super().__init__( lowerCamelCase__ , tokenizer_file=lowerCamelCase__ , do_lower_case=lowerCamelCase__ , remove_space=lowerCamelCase__ , keep_accents=lowerCamelCase__ , unk_token=lowerCamelCase__ , sep_token=lowerCamelCase__ , pad_token=lowerCamelCase__ , cls_token=lowerCamelCase__ , mask_token=lowerCamelCase__ , **lowerCamelCase__ , ) _UpperCAmelCase : Optional[Any] = do_lower_case _UpperCAmelCase : Tuple = remove_space _UpperCAmelCase : List[Any] = keep_accents _UpperCAmelCase : Tuple = vocab_file _UpperCAmelCase : str = False if not self.vocab_file else True def lowerCAmelCase__ ( self : str , lowerCamelCase__ : List[int] , lowerCamelCase__ : Optional[List[int]] = None ) ->List[int]: '''simple docstring''' _UpperCAmelCase : Optional[Any] = [self.sep_token_id] _UpperCAmelCase : Tuple = [self.cls_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 lowerCAmelCase__ ( self : int , lowerCamelCase__ : List[int] , lowerCamelCase__ : Optional[List[int]] = None ) ->List[int]: '''simple docstring''' _UpperCAmelCase : Optional[int] = [self.sep_token_id] _UpperCAmelCase : Optional[Any] = [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 lowerCAmelCase__ ( self : Dict , lowerCamelCase__ : str , lowerCamelCase__ : Optional[str] = None ) ->Tuple[str]: '''simple docstring''' if not os.path.isdir(lowerCamelCase__ ): logger.error(F"""Vocabulary path ({save_directory}) should be a directory""" ) return _UpperCAmelCase : Union[str, Any] = os.path.join( lowerCamelCase__ , (filename_prefix + "-" if filename_prefix else "") + VOCAB_FILES_NAMES["vocab_file"] ) if os.path.abspath(self.vocab_file ) != os.path.abspath(lowerCamelCase__ ): copyfile(self.vocab_file , lowerCamelCase__ ) return (out_vocab_file,)
234
1
"""simple docstring""" import functools import gc import inspect import torch from .imports import is_npu_available, is_xpu_available def A (*__A : str ) -> Optional[int]: """simple docstring""" if not isinstance(a_ , a_ ): UpperCAmelCase_ = list(a_ ) for i in range(len(a_ ) ): UpperCAmelCase_ = None gc.collect() if is_xpu_available(): torch.xpu.empty_cache() elif is_npu_available(): torch.npu.empty_cache() else: torch.cuda.empty_cache() return objects def A (__A : Exception ) -> Optional[Any]: """simple docstring""" UpperCAmelCase_ = [ "CUDA out of memory.", # CUDA OOM "cuDNN error: CUDNN_STATUS_NOT_SUPPORTED.", # CUDNN SNAFU "DefaultCPUAllocator: can't allocate memory", # CPU OOM ] if isinstance(a_ , a_ ) and len(exception.args ) == 1: return any(err in exception.args[0] for err in _statements ) return False def A (__A : callable = None , __A : int = 128 ) -> List[str]: """simple docstring""" if function is None: return functools.partial(a_ , starting_batch_size=a_ ) UpperCAmelCase_ = starting_batch_size def decorator(*__A : Optional[int] , **__A : List[str] ): nonlocal batch_size gc.collect() if is_xpu_available(): torch.xpu.empty_cache() elif is_npu_available(): torch.npu.empty_cache() else: torch.cuda.empty_cache() UpperCAmelCase_ = list(inspect.signature(a_ ).parameters.keys() ) # Guard against user error if len(a_ ) < (len(a_ ) + 1): UpperCAmelCase_ = ", ".join([F"""{arg}={value}""" for arg, value in zip(params[1:] , args[1:] )] ) raise TypeError( F"""Batch size was passed into `{function.__name__}` as the first argument when called.""" F"""Remove this as the decorator already does so: `{function.__name__}({arg_str})`""" ) while True: if batch_size == 0: raise RuntimeError('''No executable batch size found, reached zero.''' ) try: return function(a_ , *a_ , **a_ ) except Exception as e: if should_reduce_batch_size(a_ ): gc.collect() if is_xpu_available(): torch.xpu.empty_cache() elif is_npu_available(): torch.npu.empty_cache() else: torch.cuda.empty_cache() batch_size //= 2 else: raise return decorator
366
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 __snake_case ( unittest.TestCase ): def lowerCamelCase ( self : Optional[int] , _snake_case : Union[str, Any]): """simple docstring""" for model_result in results.values(): for batch_size, sequence_length in zip(model_result['''bs'''] , model_result['''ss''']): UpperCAmelCase_ = model_result['''result'''][batch_size][sequence_length] self.assertIsNotNone(_snake_case) def lowerCamelCase ( self : Optional[Any]): """simple docstring""" UpperCAmelCase_ = '''sshleifer/tiny-gpt2''' UpperCAmelCase_ = PyTorchBenchmarkArguments( models=[MODEL_ID] , training=_snake_case , inference=_snake_case , sequence_lengths=[8] , batch_sizes=[1] , multi_process=_snake_case , ) UpperCAmelCase_ = PyTorchBenchmark(_snake_case) UpperCAmelCase_ = benchmark.run() self.check_results_dict_not_empty(results.time_inference_result) self.check_results_dict_not_empty(results.memory_inference_result) def lowerCamelCase ( self : str): """simple docstring""" UpperCAmelCase_ = '''sgugger/tiny-distilbert-classification''' UpperCAmelCase_ = PyTorchBenchmarkArguments( models=[MODEL_ID] , training=_snake_case , inference=_snake_case , sequence_lengths=[8] , batch_sizes=[1] , multi_process=_snake_case , only_pretrain_model=_snake_case , ) UpperCAmelCase_ = PyTorchBenchmark(_snake_case) UpperCAmelCase_ = benchmark.run() self.check_results_dict_not_empty(results.time_inference_result) self.check_results_dict_not_empty(results.memory_inference_result) def lowerCamelCase ( self : List[str]): """simple docstring""" UpperCAmelCase_ = '''sshleifer/tiny-gpt2''' UpperCAmelCase_ = PyTorchBenchmarkArguments( models=[MODEL_ID] , training=_snake_case , inference=_snake_case , torchscript=_snake_case , sequence_lengths=[8] , batch_sizes=[1] , multi_process=_snake_case , ) UpperCAmelCase_ = PyTorchBenchmark(_snake_case) UpperCAmelCase_ = 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 lowerCamelCase ( self : List[str]): """simple docstring""" UpperCAmelCase_ = '''sshleifer/tiny-gpt2''' UpperCAmelCase_ = PyTorchBenchmarkArguments( models=[MODEL_ID] , training=_snake_case , inference=_snake_case , fpaa=_snake_case , sequence_lengths=[8] , batch_sizes=[1] , multi_process=_snake_case , ) UpperCAmelCase_ = PyTorchBenchmark(_snake_case) UpperCAmelCase_ = benchmark.run() self.check_results_dict_not_empty(results.time_inference_result) self.check_results_dict_not_empty(results.memory_inference_result) def lowerCamelCase ( self : Optional[Any]): """simple docstring""" UpperCAmelCase_ = '''sshleifer/tiny-gpt2''' UpperCAmelCase_ = AutoConfig.from_pretrained(_snake_case) # set architectures equal to `None` UpperCAmelCase_ = None UpperCAmelCase_ = PyTorchBenchmarkArguments( models=[MODEL_ID] , training=_snake_case , inference=_snake_case , sequence_lengths=[8] , batch_sizes=[1] , multi_process=_snake_case , ) UpperCAmelCase_ = PyTorchBenchmark(_snake_case , configs=[config]) UpperCAmelCase_ = benchmark.run() self.check_results_dict_not_empty(results.time_inference_result) self.check_results_dict_not_empty(results.memory_inference_result) def lowerCamelCase ( self : Dict): """simple docstring""" UpperCAmelCase_ = '''sshleifer/tiny-gpt2''' UpperCAmelCase_ = PyTorchBenchmarkArguments( models=[MODEL_ID] , training=_snake_case , inference=_snake_case , sequence_lengths=[8] , batch_sizes=[1] , multi_process=_snake_case , ) UpperCAmelCase_ = PyTorchBenchmark(_snake_case) UpperCAmelCase_ = 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 lowerCamelCase ( self : Optional[Any]): """simple docstring""" UpperCAmelCase_ = '''sshleifer/tiny-gpt2''' UpperCAmelCase_ = PyTorchBenchmarkArguments( models=[MODEL_ID] , training=_snake_case , inference=_snake_case , sequence_lengths=[8] , batch_sizes=[1] , fpaa=_snake_case , multi_process=_snake_case , ) UpperCAmelCase_ = PyTorchBenchmark(_snake_case) UpperCAmelCase_ = benchmark.run() self.check_results_dict_not_empty(results.time_train_result) self.check_results_dict_not_empty(results.memory_train_result) def lowerCamelCase ( self : List[Any]): """simple docstring""" UpperCAmelCase_ = '''sshleifer/tiny-gpt2''' UpperCAmelCase_ = AutoConfig.from_pretrained(_snake_case) UpperCAmelCase_ = PyTorchBenchmarkArguments( models=[MODEL_ID] , training=_snake_case , inference=_snake_case , sequence_lengths=[8] , batch_sizes=[1] , multi_process=_snake_case , ) UpperCAmelCase_ = PyTorchBenchmark(_snake_case , configs=[config]) UpperCAmelCase_ = benchmark.run() self.check_results_dict_not_empty(results.time_inference_result) self.check_results_dict_not_empty(results.memory_inference_result) def lowerCamelCase ( self : Dict): """simple docstring""" UpperCAmelCase_ = '''sshleifer/tinier_bart''' UpperCAmelCase_ = AutoConfig.from_pretrained(_snake_case) UpperCAmelCase_ = PyTorchBenchmarkArguments( models=[MODEL_ID] , training=_snake_case , inference=_snake_case , sequence_lengths=[8] , batch_sizes=[1] , multi_process=_snake_case , ) UpperCAmelCase_ = PyTorchBenchmark(_snake_case , configs=[config]) UpperCAmelCase_ = benchmark.run() self.check_results_dict_not_empty(results.time_inference_result) self.check_results_dict_not_empty(results.memory_inference_result) def lowerCamelCase ( self : Dict): """simple docstring""" UpperCAmelCase_ = '''sshleifer/tiny-gpt2''' UpperCAmelCase_ = AutoConfig.from_pretrained(_snake_case) UpperCAmelCase_ = PyTorchBenchmarkArguments( models=[MODEL_ID] , training=_snake_case , inference=_snake_case , sequence_lengths=[8] , batch_sizes=[1] , multi_process=_snake_case , ) UpperCAmelCase_ = PyTorchBenchmark(_snake_case , configs=[config]) UpperCAmelCase_ = benchmark.run() self.check_results_dict_not_empty(results.time_train_result) self.check_results_dict_not_empty(results.memory_train_result) def lowerCamelCase ( self : List[Any]): """simple docstring""" UpperCAmelCase_ = '''sshleifer/tinier_bart''' UpperCAmelCase_ = AutoConfig.from_pretrained(_snake_case) UpperCAmelCase_ = PyTorchBenchmarkArguments( models=[MODEL_ID] , training=_snake_case , inference=_snake_case , sequence_lengths=[8] , batch_sizes=[1] , multi_process=_snake_case , ) UpperCAmelCase_ = PyTorchBenchmark(_snake_case , configs=[config]) UpperCAmelCase_ = benchmark.run() self.check_results_dict_not_empty(results.time_train_result) self.check_results_dict_not_empty(results.memory_train_result) def lowerCamelCase ( self : List[Any]): """simple docstring""" UpperCAmelCase_ = '''sshleifer/tiny-gpt2''' with tempfile.TemporaryDirectory() as tmp_dir: UpperCAmelCase_ = PyTorchBenchmarkArguments( models=[MODEL_ID] , training=_snake_case , inference=_snake_case , save_to_csv=_snake_case , sequence_lengths=[8] , batch_sizes=[1] , inference_time_csv_file=os.path.join(_snake_case , '''inf_time.csv''') , train_memory_csv_file=os.path.join(_snake_case , '''train_mem.csv''') , inference_memory_csv_file=os.path.join(_snake_case , '''inf_mem.csv''') , train_time_csv_file=os.path.join(_snake_case , '''train_time.csv''') , env_info_csv_file=os.path.join(_snake_case , '''env.csv''') , multi_process=_snake_case , ) UpperCAmelCase_ = PyTorchBenchmark(_snake_case) benchmark.run() self.assertTrue(Path(os.path.join(_snake_case , '''inf_time.csv''')).exists()) self.assertTrue(Path(os.path.join(_snake_case , '''train_time.csv''')).exists()) self.assertTrue(Path(os.path.join(_snake_case , '''inf_mem.csv''')).exists()) self.assertTrue(Path(os.path.join(_snake_case , '''train_mem.csv''')).exists()) self.assertTrue(Path(os.path.join(_snake_case , '''env.csv''')).exists()) def lowerCamelCase ( self : List[str]): """simple docstring""" UpperCAmelCase_ = '''sshleifer/tiny-gpt2''' def _check_summary_is_not_empty(_snake_case : Tuple): self.assertTrue(hasattr(_snake_case , '''sequential''')) self.assertTrue(hasattr(_snake_case , '''cumulative''')) self.assertTrue(hasattr(_snake_case , '''current''')) self.assertTrue(hasattr(_snake_case , '''total''')) with tempfile.TemporaryDirectory() as tmp_dir: UpperCAmelCase_ = PyTorchBenchmarkArguments( models=[MODEL_ID] , training=_snake_case , inference=_snake_case , sequence_lengths=[8] , batch_sizes=[1] , log_filename=os.path.join(_snake_case , '''log.txt''') , log_print=_snake_case , trace_memory_line_by_line=_snake_case , multi_process=_snake_case , ) UpperCAmelCase_ = PyTorchBenchmark(_snake_case) UpperCAmelCase_ = benchmark.run() _check_summary_is_not_empty(result.inference_summary) _check_summary_is_not_empty(result.train_summary) self.assertTrue(Path(os.path.join(_snake_case , '''log.txt''')).exists())
7
0
from .imports import is_tqdm_available if is_tqdm_available(): from tqdm.auto import tqdm as _tqdm from ..state import PartialState def lowercase_ ( _lowerCamelCase : bool = True , *_lowerCamelCase : int , **_lowerCamelCase : Optional[Any]): if not is_tqdm_available(): raise ImportError("Accelerate's `tqdm` module requires `tqdm` to be installed. Please run `pip install tqdm`.") lowercase__ : Dict = False if main_process_only: lowercase__ : int = PartialState().local_process_index == 0 return _tqdm(*_lowerCamelCase , **_lowerCamelCase , disable=_lowerCamelCase)
87
import time import warnings from abc import ABC from copy import deepcopy from typing import Optional import torch from ..utils import add_start_docstrings, logging UpperCamelCase = logging.get_logger(__name__) UpperCamelCase = R''' Args: input_ids (`torch.LongTensor` of shape `(batch_size, sequence_length)`): Indices of input sequence tokens in the vocabulary. Indices can be obtained using [`AutoTokenizer`]. See [`PreTrainedTokenizer.encode`] and [`PreTrainedTokenizer.__call__`] for details. [What are input IDs?](../glossary#input-ids) scores (`torch.FloatTensor` of shape `(batch_size, config.vocab_size)`): Prediction scores of a language modeling head. These can be scores for each vocabulary token before SoftMax or scores for each vocabulary token after SoftMax. kwargs (`Dict[str, Any]`, *optional*): Additional stopping criteria specific kwargs. Return: `bool`. `False` indicates we should continue, `True` indicates we should stop. ''' class snake_case_ ( __A ): @add_start_docstrings(lowercase_ ) def __call__( self : Optional[Any] , lowercase_ : torch.LongTensor , lowercase_ : torch.FloatTensor , **lowercase_ : List[str] ) -> bool: raise NotImplementedError("StoppingCriteria needs to be subclassed" ) class snake_case_ ( __A ): def __init__( self : Dict , lowercase_ : int , lowercase_ : Optional[int] = None ) -> List[str]: lowercase__ : str = max_length lowercase__ : Optional[int] = max_position_embeddings @add_start_docstrings(lowercase_ ) def __call__( self : Tuple , lowercase_ : torch.LongTensor , lowercase_ : torch.FloatTensor , **lowercase_ : Union[str, Any] ) -> bool: lowercase__ : str = input_ids.shape[-1] lowercase__ : Any = cur_len >= self.max_length if self.max_position_embeddings is not None and not is_done and cur_len >= self.max_position_embeddings: logger.warning_once( "This is a friendly reminder - the current text generation call will exceed the model's predefined " F'''maximum length ({self.max_position_embeddings}). Depending on the model, you may observe ''' "exceptions, performance degradation, or nothing at all." ) return is_done class snake_case_ ( __A ): def __init__( self : Tuple , lowercase_ : int , lowercase_ : int ) -> List[str]: warnings.warn( "The class `MaxNewTokensCriteria` is deprecated. " F'''Please use `MaxLengthCriteria(max_length={start_length + max_new_tokens})` ''' "with `max_length = start_length + max_new_tokens` instead." , lowercase_ , ) lowercase__ : Optional[int] = start_length lowercase__ : str = max_new_tokens lowercase__ : Tuple = start_length + max_new_tokens @add_start_docstrings(lowercase_ ) def __call__( self : List[Any] , lowercase_ : torch.LongTensor , lowercase_ : torch.FloatTensor , **lowercase_ : Dict ) -> bool: return input_ids.shape[-1] >= self.max_length class snake_case_ ( __A ): def __init__( self : Tuple , lowercase_ : float , lowercase_ : Optional[float] = None ) -> Dict: lowercase__ : List[str] = max_time lowercase__ : Tuple = time.time() if initial_timestamp is None else initial_timestamp @add_start_docstrings(lowercase_ ) def __call__( self : int , lowercase_ : torch.LongTensor , lowercase_ : torch.FloatTensor , **lowercase_ : Union[str, Any] ) -> bool: return time.time() - self.initial_timestamp > self.max_time class snake_case_ ( __A ): @add_start_docstrings(lowercase_ ) def __call__( self : str , lowercase_ : torch.LongTensor , lowercase_ : torch.FloatTensor , **lowercase_ : List[str] ) -> bool: return any(criteria(lowercase_ , lowercase_ ) for criteria in self ) @property def __UpperCamelCase ( self : Optional[Any] ) -> Optional[int]: for stopping_criterium in self: if isinstance(lowercase_ , lowercase_ ): return stopping_criterium.max_length elif isinstance(lowercase_ , lowercase_ ): return stopping_criterium.max_length return None def lowercase_ ( _lowerCamelCase : StoppingCriteriaList , _lowerCamelCase : int): lowercase__ : Optional[int] = stopping_criteria.max_length lowercase__ : str = deepcopy(_lowerCamelCase) if stopping_max_length is not None and stopping_max_length != max_length: warnings.warn("You set different `max_length` for stopping criteria and `max_length` parameter" , _lowerCamelCase) elif stopping_max_length is None: new_stopping_criteria.append(MaxLengthCriteria(max_length=_lowerCamelCase)) return new_stopping_criteria
87
1
from typing import TYPE_CHECKING from ...utils import ( OptionalDependencyNotAvailable, _LazyModule, is_flax_available, is_tf_available, is_tokenizers_available, is_torch_available, is_vision_available, ) __lowerCamelCase : List[Any] = { """configuration_owlvit""": [ """OWLVIT_PRETRAINED_CONFIG_ARCHIVE_MAP""", """OwlViTConfig""", """OwlViTOnnxConfig""", """OwlViTTextConfig""", """OwlViTVisionConfig""", ], """processing_owlvit""": ["""OwlViTProcessor"""], } try: if not is_vision_available(): raise OptionalDependencyNotAvailable() except OptionalDependencyNotAvailable: pass else: __lowerCamelCase : Dict = ["""OwlViTFeatureExtractor"""] __lowerCamelCase : int = ["""OwlViTImageProcessor"""] try: if not is_torch_available(): raise OptionalDependencyNotAvailable() except OptionalDependencyNotAvailable: pass else: __lowerCamelCase : int = [ """OWLVIT_PRETRAINED_MODEL_ARCHIVE_LIST""", """OwlViTModel""", """OwlViTPreTrainedModel""", """OwlViTTextModel""", """OwlViTVisionModel""", """OwlViTForObjectDetection""", ] if TYPE_CHECKING: from .configuration_owlvit import ( OWLVIT_PRETRAINED_CONFIG_ARCHIVE_MAP, OwlViTConfig, OwlViTOnnxConfig, OwlViTTextConfig, OwlViTVisionConfig, ) from .processing_owlvit import OwlViTProcessor try: if not is_vision_available(): raise OptionalDependencyNotAvailable() except OptionalDependencyNotAvailable: pass else: from .feature_extraction_owlvit import OwlViTFeatureExtractor from .image_processing_owlvit import OwlViTImageProcessor try: if not is_torch_available(): raise OptionalDependencyNotAvailable() except OptionalDependencyNotAvailable: pass else: from .modeling_owlvit import ( OWLVIT_PRETRAINED_MODEL_ARCHIVE_LIST, OwlViTForObjectDetection, OwlViTModel, OwlViTPreTrainedModel, OwlViTTextModel, OwlViTVisionModel, ) else: import sys __lowerCamelCase : str = _LazyModule(__name__, globals()["""__file__"""], _import_structure, module_spec=__spec__)
352
from typing import Dict, List, Optional, Union import numpy as np from ...image_processing_utils import BaseImageProcessor, BatchFeature, get_size_dict from ...image_transforms import rescale, resize, to_channel_dimension_format from ...image_utils import ( ChannelDimension, ImageInput, PILImageResampling, make_list_of_images, to_numpy_array, valid_images, ) from ...utils import TensorType, is_vision_available, logging if is_vision_available(): import PIL __lowerCamelCase : int = logging.get_logger(__name__) def SCREAMING_SNAKE_CASE ( snake_case_ : int , snake_case_ : Any ): snake_case__ : List[str] = b.T snake_case__ : Union[str, Any] = np.sum(np.square(snake_case_ ) , axis=1 ) snake_case__ : Dict = np.sum(np.square(snake_case_ ) , axis=0 ) snake_case__ : Dict = np.matmul(snake_case_ , snake_case_ ) snake_case__ : Any = aa[:, None] - 2 * ab + ba[None, :] return d def SCREAMING_SNAKE_CASE ( snake_case_ : Optional[int] , snake_case_ : Tuple ): snake_case__ : Tuple = x.reshape(-1 , 3 ) snake_case__ : int = squared_euclidean_distance(snake_case_ , snake_case_ ) return np.argmin(snake_case_ , axis=1 ) class SCREAMING_SNAKE_CASE__ ( UpperCamelCase_ ): """simple docstring""" a_ = ["pixel_values"] def __init__( self : str , __A : Optional[Union[List[List[int]], np.ndarray]] = None , __A : bool = True , __A : Dict[str, int] = None , __A : PILImageResampling = PILImageResampling.BILINEAR , __A : bool = True , __A : bool = True , **__A : Union[str, Any] , ): super().__init__(**__A ) snake_case__ : Optional[int] = size if size is not None else {"height": 2_5_6, "width": 2_5_6} snake_case__ : List[Any] = get_size_dict(__A ) snake_case__ : Any = np.array(__A ) if clusters is not None else None snake_case__ : Optional[Any] = do_resize snake_case__ : Any = size snake_case__ : List[Any] = resample snake_case__ : List[Any] = do_normalize snake_case__ : Dict = do_color_quantize def _lowercase ( self : List[Any] , __A : np.ndarray , __A : Dict[str, int] , __A : PILImageResampling = PILImageResampling.BILINEAR , __A : Optional[Union[str, ChannelDimension]] = None , **__A : int , ): snake_case__ : List[Any] = get_size_dict(__A ) if "height" not in size or "width" not in size: raise ValueError(f'''Size dictionary must contain both height and width keys. Got {size.keys()}''' ) return resize( __A , size=(size["height"], size["width"]) , resample=__A , data_format=__A , **__A ) def _lowercase ( self : List[Any] , __A : np.ndarray , __A : Optional[Union[str, ChannelDimension]] = None , ): snake_case__ : List[str] = rescale(image=__A , scale=1 / 1_2_7.5 , data_format=__A ) snake_case__ : List[Any] = image - 1 return image def _lowercase ( self : Dict , __A : ImageInput , __A : bool = None , __A : Dict[str, int] = None , __A : PILImageResampling = None , __A : bool = None , __A : Optional[bool] = None , __A : Optional[Union[List[List[int]], np.ndarray]] = None , __A : Optional[Union[str, TensorType]] = None , __A : Optional[Union[str, ChannelDimension]] = ChannelDimension.FIRST , **__A : Optional[int] , ): snake_case__ : Any = do_resize if do_resize is not None else self.do_resize snake_case__ : Union[str, Any] = size if size is not None else self.size snake_case__ : Union[str, Any] = get_size_dict(__A ) snake_case__ : Optional[Any] = resample if resample is not None else self.resample snake_case__ : Tuple = do_normalize if do_normalize is not None else self.do_normalize snake_case__ : Optional[Any] = do_color_quantize if do_color_quantize is not None else self.do_color_quantize snake_case__ : Union[str, Any] = clusters if clusters is not None else self.clusters snake_case__ : Union[str, Any] = np.array(__A ) snake_case__ : Any = 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 or resample is None: raise ValueError("Size and resample must be specified if do_resize is True." ) if do_color_quantize and clusters is None: raise ValueError("Clusters must be specified if do_color_quantize is True." ) # All transformations expect numpy arrays. snake_case__ : Optional[Any] = [to_numpy_array(__A ) for image in images] if do_resize: snake_case__ : List[str] = [self.resize(image=__A , size=__A , resample=__A ) for image in images] if do_normalize: snake_case__ : Union[str, Any] = [self.normalize(image=__A ) for image in images] if do_color_quantize: snake_case__ : int = [to_channel_dimension_format(__A , ChannelDimension.LAST ) for image in images] # color quantize from (batch_size, height, width, 3) to (batch_size, height, width) snake_case__ : int = np.array(__A ) snake_case__ : Dict = color_quantize(__A , __A ).reshape(images.shape[:-1] ) # flatten to (batch_size, height*width) snake_case__ : str = images.shape[0] snake_case__ : str = images.reshape(__A , -1 ) # We need to convert back to a list of images to keep consistent behaviour across processors. snake_case__ : Union[str, Any] = list(__A ) else: snake_case__ : Any = [to_channel_dimension_format(__A , __A ) for image in images] snake_case__ : Optional[int] = {"input_ids": images} return BatchFeature(data=__A , tensor_type=__A )
286
0
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 UpperCAmelCase_ : str = datasets.utils.logging.get_logger(__name__) @dataclass class _SCREAMING_SNAKE_CASE ( datasets.BuilderConfig ): snake_case__ : Optional[datasets.Features] = None snake_case__ : str = "utf-8" snake_case__ : Optional[str] = None snake_case__ : Optional[str] = None snake_case__ : bool = True # deprecated snake_case__ : Optional[int] = None # deprecated snake_case__ : int = 1_0 << 2_0 # 10MB snake_case__ : Optional[bool] = None class _SCREAMING_SNAKE_CASE ( datasets.ArrowBasedBuilder ): snake_case__ : int = JsonConfig def _A ( self : List[Any] ): if self.config.block_size is not None: logger.warning("""The JSON loader parameter `block_size` is deprecated. Please use `chunksize` instead""" ) UpperCamelCase :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 : int , __lowerCamelCase : Tuple ): 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}""" ) UpperCamelCase :List[Any] = dl_manager.download_and_extract(self.config.data_files ) if isinstance(__lowerCamelCase , (str, list, tuple) ): UpperCamelCase :Optional[int] = data_files if isinstance(__lowerCamelCase , __lowerCamelCase ): UpperCamelCase :Optional[int] = [files] UpperCamelCase :Optional[Any] = [dl_manager.iter_files(__lowerCamelCase ) for file in files] return [datasets.SplitGenerator(name=datasets.Split.TRAIN , gen_kwargs={"""files""": files} )] UpperCamelCase :Tuple = [] for split_name, files in data_files.items(): if isinstance(__lowerCamelCase , __lowerCamelCase ): UpperCamelCase :Dict = [files] UpperCamelCase :List[Any] = [dl_manager.iter_files(__lowerCamelCase ) for file in files] splits.append(datasets.SplitGenerator(name=__lowerCamelCase , gen_kwargs={"""files""": files} ) ) return splits def _A ( self : List[str] , __lowerCamelCase : pa.Table ): if self.config.features is not None: # adding missing columns for column_name in set(self.config.features ) - set(pa_table.column_names ): UpperCamelCase :List[str] = self.config.features.arrow_schema.field(__lowerCamelCase ).type UpperCamelCase :Union[str, Any] = pa_table.append_column(__lowerCamelCase , pa.array([None] * len(__lowerCamelCase ) , type=__lowerCamelCase ) ) # more expensive cast to support nested structures with keys in a different order # allows str <-> int/float or str to Audio for example UpperCamelCase :Any = table_cast(__lowerCamelCase , self.config.features.arrow_schema ) return pa_table def _A ( self : Tuple , __lowerCamelCase : int ): for file_idx, file in enumerate(itertools.chain.from_iterable(__lowerCamelCase ) ): # 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(__lowerCamelCase , encoding=self.config.encoding , errors=self.config.encoding_errors ) as f: UpperCamelCase :Dict = json.load(__lowerCamelCase ) # We keep only the field we are interested in UpperCamelCase :int = dataset[self.config.field] # We accept two format: a list of dicts or a dict of lists if isinstance(__lowerCamelCase , (list, tuple) ): UpperCamelCase :Optional[int] = set().union(*[row.keys() for row in dataset] ) UpperCamelCase :Any = {col: [row.get(__lowerCamelCase ) for row in dataset] for col in keys} else: UpperCamelCase :Optional[int] = dataset UpperCamelCase :Optional[Any] = pa.Table.from_pydict(__lowerCamelCase ) yield file_idx, self._cast_table(__lowerCamelCase ) # If the file has one json object per line else: with open(__lowerCamelCase , """rb""" ) as f: UpperCamelCase :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 UpperCamelCase :str = max(self.config.chunksize // 32 , 16 << 10 ) UpperCamelCase :Optional[int] = ( self.config.encoding_errors if self.config.encoding_errors is not None else """strict""" ) while True: UpperCamelCase :int = f.read(self.config.chunksize ) if not batch: break # Finish current line try: batch += f.readline() except (AttributeError, io.UnsupportedOperation): batch += readline(__lowerCamelCase ) # PyArrow only accepts utf-8 encoded bytes if self.config.encoding != "utf-8": UpperCamelCase :List[str] = batch.decode(self.config.encoding , errors=__lowerCamelCase ).encode("""utf-8""" ) try: while True: try: UpperCamelCase :int = paj.read_json( io.BytesIO(__lowerCamelCase ) , read_options=paj.ReadOptions(block_size=__lowerCamelCase ) ) break except (pa.ArrowInvalid, pa.ArrowNotImplementedError) as e: if ( isinstance(__lowerCamelCase , pa.ArrowInvalid ) and "straddling" not in str(__lowerCamelCase ) or block_size > len(__lowerCamelCase ) ): 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(__lowerCamelCase )} 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( __lowerCamelCase , encoding=self.config.encoding , errors=self.config.encoding_errors ) as f: UpperCamelCase :Optional[Any] = json.load(__lowerCamelCase ) except json.JSONDecodeError: logger.error(F"""Failed to read file '{file}' with error {type(__lowerCamelCase )}: {e}""" ) raise e # If possible, parse the file as a list of json objects and exit the loop if isinstance(__lowerCamelCase , __lowerCamelCase ): # list is the only sequence type supported in JSON try: UpperCamelCase :Tuple = set().union(*[row.keys() for row in dataset] ) UpperCamelCase :Dict = {col: [row.get(__lowerCamelCase ) for row in dataset] for col in keys} UpperCamelCase :Any = pa.Table.from_pydict(__lowerCamelCase ) except (pa.ArrowInvalid, AttributeError) as e: logger.error(F"""Failed to read file '{file}' with error {type(__lowerCamelCase )}: {e}""" ) raise ValueError(F"""Not able to read records in the JSON file at {file}.""" ) from None yield file_idx, self._cast_table(__lowerCamelCase ) break else: logger.error(F"""Failed to read file '{file}' with error {type(__lowerCamelCase )}: {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(__lowerCamelCase ) batch_idx += 1
38
from ...utils import ( OptionalDependencyNotAvailable, is_torch_available, is_transformers_available, is_transformers_version, ) try: if not (is_transformers_available() and is_torch_available()): raise OptionalDependencyNotAvailable() except OptionalDependencyNotAvailable: from ...utils.dummy_torch_and_transformers_objects import ShapEPipeline else: from .camera import create_pan_cameras from .pipeline_shap_e import ShapEPipeline from .pipeline_shap_e_img2img import ShapEImgaImgPipeline from .renderer import ( BoundingBoxVolume, ImportanceRaySampler, MLPNeRFModelOutput, MLPNeRSTFModel, ShapEParamsProjModel, ShapERenderer, StratifiedRaySampler, VoidNeRFModel, )
38
1
from math import pi, sqrt def snake_case_ ( lowerCAmelCase_ )-> float: '''simple docstring''' if num <= 0: raise ValueError("""math domain error""" ) if num > 171.5: raise OverflowError("""math range error""" ) elif num - int(lowerCAmelCase_ ) not in (0, 0.5): raise NotImplementedError("""num must be an integer or a half-integer""" ) elif num == 0.5: return sqrt(lowerCAmelCase_ ) else: return 1.0 if num == 1 else (num - 1) * gamma(num - 1 ) def snake_case_ ( )-> None: '''simple docstring''' assert gamma(0.5 ) == sqrt(lowerCAmelCase_ ) assert gamma(1 ) == 1.0 assert gamma(2 ) == 1.0 if __name__ == "__main__": from doctest import testmod testmod() A_ : Union[str, Any] = 1.0 while num: A_ : Optional[Any] = float(input("""Gamma of: """)) print(f"""gamma({num}) = {gamma(num)}""") print("""\nEnter 0 to exit...""")
365
'''simple docstring''' from __future__ import annotations from collections import deque from collections.abc import Iterator from dataclasses import dataclass @dataclass class lowercase : """simple docstring""" UpperCAmelCase = 42 UpperCAmelCase = 42 class lowercase : """simple docstring""" def __init__( self ,a_ ) -> List[str]: _UpperCAmelCase : list[list[Edge]] = [[] for _ in range(a_ )] _UpperCAmelCase : int = size def __getitem__( self ,a_ ) -> Iterator[Edge]: return iter(self._graph[vertex] ) @property def _snake_case ( self ) -> List[Any]: return self._size def _snake_case ( self ,a_ ,a_ ,a_ ) -> Tuple: if weight not in (0, 1): raise ValueError("""Edge weight must be either 0 or 1.""" ) if to_vertex < 0 or to_vertex >= self.size: raise ValueError("""Vertex indexes must be in [0; size).""" ) self._graph[from_vertex].append(Edge(a_ ,a_ ) ) def _snake_case ( self ,a_ ,a_ ) -> int | None: _UpperCAmelCase : Union[str, Any] = deque([start_vertex] ) _UpperCAmelCase : list[int | None] = [None] * self.size _UpperCAmelCase : Union[str, Any] = 0 while queue: _UpperCAmelCase : Union[str, Any] = queue.popleft() _UpperCAmelCase : Union[str, Any] = distances[current_vertex] if current_distance is None: continue for edge in self[current_vertex]: _UpperCAmelCase : List[Any] = current_distance + edge.weight _UpperCAmelCase : List[Any] = distances[edge.destination_vertex] if ( isinstance(a_ ,a_ ) and new_distance >= dest_vertex_distance ): continue _UpperCAmelCase : Tuple = new_distance if edge.weight == 0: queue.appendleft(edge.destination_vertex ) else: queue.append(edge.destination_vertex ) if distances[finish_vertex] is None: raise ValueError("""No path from start_vertex to finish_vertex.""" ) return distances[finish_vertex] if __name__ == "__main__": import doctest doctest.testmod()
349
0
'''simple docstring''' def SCREAMING_SNAKE_CASE( __lowercase ) -> int: assert ( isinstance(__lowercase , __lowercase ) and number_of_steps > 0 ), F"""number_of_steps needs to be positive integer, your input {number_of_steps}""" if number_of_steps == 1: return 1 A , A: int = 1, 1 for _ in range(number_of_steps - 1 ): A , A: Any = current + previous, current return current if __name__ == "__main__": import doctest doctest.testmod()
319
'''simple docstring''' from __future__ import annotations import math def SCREAMING_SNAKE_CASE( __lowercase , __lowercase ) -> list: if len(__lowercase ) != 2 or len(a[0] ) != 2 or len(__lowercase ) != 2 or len(b[0] ) != 2: raise Exception('''Matrices are not 2x2''' ) A: str = [ [a[0][0] * b[0][0] + a[0][1] * b[1][0], a[0][0] * b[0][1] + a[0][1] * b[1][1]], [a[1][0] * b[0][0] + a[1][1] * b[1][0], a[1][0] * b[0][1] + a[1][1] * b[1][1]], ] return new_matrix def SCREAMING_SNAKE_CASE( __lowercase , __lowercase ) -> Union[str, Any]: return [ [matrix_a[row][col] + matrix_b[row][col] for col in range(len(matrix_a[row] ) )] for row in range(len(__lowercase ) ) ] def SCREAMING_SNAKE_CASE( __lowercase , __lowercase ) -> Union[str, Any]: return [ [matrix_a[row][col] - matrix_b[row][col] for col in range(len(matrix_a[row] ) )] for row in range(len(__lowercase ) ) ] def SCREAMING_SNAKE_CASE( __lowercase ) -> tuple[list, list, list, list]: if len(__lowercase ) % 2 != 0 or len(a[0] ) % 2 != 0: raise Exception('''Odd matrices are not supported!''' ) A: Union[str, Any] = len(__lowercase ) A: str = matrix_length // 2 A: Optional[int] = [[a[i][j] for j in range(__lowercase , __lowercase )] for i in range(__lowercase )] A: Optional[Any] = [ [a[i][j] for j in range(__lowercase , __lowercase )] for i in range(__lowercase , __lowercase ) ] A: Union[str, Any] = [[a[i][j] for j in range(__lowercase )] for i in range(__lowercase )] A: int = [[a[i][j] for j in range(__lowercase )] for i in range(__lowercase , __lowercase )] return top_left, top_right, bot_left, bot_right def SCREAMING_SNAKE_CASE( __lowercase ) -> tuple[int, int]: return len(__lowercase ), len(matrix[0] ) def SCREAMING_SNAKE_CASE( __lowercase ) -> None: print('''\n'''.join(str(__lowercase ) for line in matrix ) ) def SCREAMING_SNAKE_CASE( __lowercase , __lowercase ) -> list: if matrix_dimensions(__lowercase ) == (2, 2): return default_matrix_multiplication(__lowercase , __lowercase ) A , A , A , A: Union[str, Any] = split_matrix(__lowercase ) A , A , A , A: List[Any] = split_matrix(__lowercase ) A: Optional[int] = actual_strassen(__lowercase , matrix_subtraction(__lowercase , __lowercase ) ) A: Any = actual_strassen(matrix_addition(__lowercase , __lowercase ) , __lowercase ) A: Tuple = actual_strassen(matrix_addition(__lowercase , __lowercase ) , __lowercase ) A: Optional[int] = actual_strassen(__lowercase , matrix_subtraction(__lowercase , __lowercase ) ) A: Tuple = actual_strassen(matrix_addition(__lowercase , __lowercase ) , matrix_addition(__lowercase , __lowercase ) ) A: Union[str, Any] = actual_strassen(matrix_subtraction(__lowercase , __lowercase ) , matrix_addition(__lowercase , __lowercase ) ) A: List[str] = actual_strassen(matrix_subtraction(__lowercase , __lowercase ) , matrix_addition(__lowercase , __lowercase ) ) A: int = matrix_addition(matrix_subtraction(matrix_addition(__lowercase , __lowercase ) , __lowercase ) , __lowercase ) A: Any = matrix_addition(__lowercase , __lowercase ) A: List[Any] = matrix_addition(__lowercase , __lowercase ) A: List[str] = matrix_subtraction(matrix_subtraction(matrix_addition(__lowercase , __lowercase ) , __lowercase ) , __lowercase ) # construct the new matrix from our 4 quadrants A: Union[str, Any] = [] for i in range(len(__lowercase ) ): new_matrix.append(top_left[i] + top_right[i] ) for i in range(len(__lowercase ) ): new_matrix.append(bot_left[i] + bot_right[i] ) return new_matrix def SCREAMING_SNAKE_CASE( __lowercase , __lowercase ) -> list: if matrix_dimensions(__lowercase )[1] != matrix_dimensions(__lowercase )[0]: A: int = ( '''Unable to multiply these matrices, please check the dimensions.\n''' F"""Matrix A: {matrixa}\n""" F"""Matrix B: {matrixa}""" ) raise Exception(__lowercase ) A: str = matrix_dimensions(__lowercase ) A: str = matrix_dimensions(__lowercase ) if dimensiona[0] == dimensiona[1] and dimensiona[0] == dimensiona[1]: return [matrixa, matrixa] A: Union[str, Any] = max(*__lowercase , *__lowercase ) A: Optional[int] = int(math.pow(2 , math.ceil(math.loga(__lowercase ) ) ) ) A: List[Any] = matrixa A: Tuple = matrixa # Adding zeros to the matrices so that the arrays dimensions are the same and also # power of 2 for i in range(0 , __lowercase ): if i < dimensiona[0]: for _ in range(dimensiona[1] , __lowercase ): new_matrixa[i].append(0 ) else: new_matrixa.append([0] * maxim ) if i < dimensiona[0]: for _ in range(dimensiona[1] , __lowercase ): new_matrixa[i].append(0 ) else: new_matrixa.append([0] * maxim ) A: Any = actual_strassen(__lowercase , __lowercase ) # Removing the additional zeros for i in range(0 , __lowercase ): if i < dimensiona[0]: for _ in range(dimensiona[1] , __lowercase ): final_matrix[i].pop() else: final_matrix.pop() return final_matrix if __name__ == "__main__": UpperCamelCase = [ [2, 3, 4, 5], [6, 4, 3, 1], [2, 3, 6, 7], [3, 1, 2, 4], [2, 3, 4, 5], [6, 4, 3, 1], [2, 3, 6, 7], [3, 1, 2, 4], [2, 3, 4, 5], [6, 2, 3, 1], ] UpperCamelCase = [[0, 2, 1, 1], [16, 2, 3, 3], [2, 2, 7, 7], [13, 11, 22, 4]] print(strassen(matrixa, matrixa))
319
1
"""simple docstring""" __A : Optional[Any] = { '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', }
57
"""simple docstring""" import argparse import json import os import torch from transformers import LukeConfig, LukeModel, LukeTokenizer, RobertaTokenizer from transformers.tokenization_utils_base import AddedToken @torch.no_grad() def __SCREAMING_SNAKE_CASE ( lowercase__ , lowercase__ , lowercase__ , lowercase__ , lowercase__ ): """simple docstring""" # Load configuration defined in the metadata file with open(lowercase__ ) as metadata_file: A = json.load(lowercase__ ) A = LukeConfig(use_entity_aware_attention=lowercase__ , **metadata["model_config"] ) # Load in the weights from the checkpoint_path A = torch.load(lowercase__ , map_location="cpu" ) # Load the entity vocab file A = load_entity_vocab(lowercase__ ) A = RobertaTokenizer.from_pretrained(metadata["model_config"]["bert_model_name"] ) # Add special tokens to the token vocabulary for downstream tasks A = AddedToken("<ent>" , lstrip=lowercase__ , rstrip=lowercase__ ) A = AddedToken("<ent2>" , lstrip=lowercase__ , rstrip=lowercase__ ) tokenizer.add_special_tokens({"additional_special_tokens": [entity_token_a, entity_token_a]} ) config.vocab_size += 2 print(F"""Saving tokenizer to {pytorch_dump_folder_path}""" ) tokenizer.save_pretrained(lowercase__ ) with open(os.path.join(lowercase__ , LukeTokenizer.vocab_files_names["entity_vocab_file"] ) , "w" ) as f: json.dump(lowercase__ , lowercase__ ) A = LukeTokenizer.from_pretrained(lowercase__ ) # Initialize the embeddings of the special tokens A = state_dict["embeddings.word_embeddings.weight"] A = word_emb[tokenizer.convert_tokens_to_ids(["@"] )[0]].unsqueeze(0 ) A = word_emb[tokenizer.convert_tokens_to_ids(["#"] )[0]].unsqueeze(0 ) A = torch.cat([word_emb, ent_emb, enta_emb] ) # Initialize the query layers of the entity-aware self-attention mechanism for layer_index in range(config.num_hidden_layers ): for matrix_name in ["query.weight", "query.bias"]: A = F"""encoder.layer.{layer_index}.attention.self.""" A = state_dict[prefix + matrix_name] A = state_dict[prefix + matrix_name] A = state_dict[prefix + matrix_name] # Initialize the embedding of the [MASK2] entity using that of the [MASK] entity for downstream tasks A = state_dict["entity_embeddings.entity_embeddings.weight"] A = entity_emb[entity_vocab["[MASK]"]] A = LukeModel(config=lowercase__ ).eval() A , A = model.load_state_dict(lowercase__ , strict=lowercase__ ) if not (len(lowercase__ ) == 1 and missing_keys[0] == "embeddings.position_ids"): raise ValueError(F"""Missing keys {", ".join(lowercase__ )}. Expected only missing embeddings.position_ids""" ) if not (all(key.startswith("entity_predictions" ) or key.startswith("lm_head" ) for key in unexpected_keys )): raise ValueError( "Unexpected keys" F""" {", ".join([key for key in unexpected_keys if not (key.startswith("entity_predictions" ) or key.startswith("lm_head" ))] )}""" ) # Check outputs A = LukeTokenizer.from_pretrained(lowercase__ , task="entity_classification" ) A = ( "Top seed Ana Ivanovic said on Thursday she could hardly believe her luck as a fortuitous netcord helped the" " new world number one avoid a humiliating second- round exit at Wimbledon ." ) A = (39, 42) A = tokenizer(lowercase__ , entity_spans=[span] , add_prefix_space=lowercase__ , return_tensors="pt" ) A = model(**lowercase__ ) # Verify word hidden states if model_size == "large": A = torch.Size((1, 42, 1_024) ) A = torch.tensor( [[0.01_33, 0.08_65, 0.00_95], [0.30_93, -0.25_76, -0.74_18], [-0.17_20, -0.21_17, -0.28_69]] ) else: # base A = torch.Size((1, 42, 768) ) A = torch.tensor([[0.00_37, 0.13_68, -0.00_91], [0.10_99, 0.33_29, -0.10_95], [0.07_65, 0.53_35, 0.11_79]] ) if not (outputs.last_hidden_state.shape == expected_shape): raise ValueError( F"""Outputs.last_hidden_state.shape is {outputs.last_hidden_state.shape}, Expected shape is {expected_shape}""" ) if not torch.allclose(outputs.last_hidden_state[0, :3, :3] , lowercase__ , atol=1e-4 ): raise ValueError # Verify entity hidden states if model_size == "large": A = torch.Size((1, 1, 1_024) ) A = torch.tensor([[0.04_66, -0.01_06, -0.01_79]] ) else: # base A = torch.Size((1, 1, 768) ) A = torch.tensor([[0.14_57, 0.10_44, 0.01_74]] ) if not (outputs.entity_last_hidden_state.shape != expected_shape): raise ValueError( F"""Outputs.entity_last_hidden_state.shape is {outputs.entity_last_hidden_state.shape}, Expected shape is""" F""" {expected_shape}""" ) if not torch.allclose(outputs.entity_last_hidden_state[0, :3, :3] , lowercase__ , atol=1e-4 ): raise ValueError # Finally, save our PyTorch model and tokenizer print("Saving PyTorch model to {}".format(lowercase__ ) ) model.save_pretrained(lowercase__ ) def __SCREAMING_SNAKE_CASE ( lowercase__ ): """simple docstring""" A = {} with open(lowercase__ , "r" , encoding="utf-8" ) as f: for index, line in enumerate(lowercase__ ): A , A = line.rstrip().split("\t" ) A = index return entity_vocab if __name__ == "__main__": __A : Optional[Any] = argparse.ArgumentParser() # Required parameters parser.add_argument('--checkpoint_path', type=str, help='Path to a pytorch_model.bin file.') parser.add_argument( '--metadata_path', default=None, type=str, help='Path to a metadata.json file, defining the configuration.' ) parser.add_argument( '--entity_vocab_path', default=None, type=str, help='Path to an entity_vocab.tsv file, containing the entity vocabulary.', ) parser.add_argument( '--pytorch_dump_folder_path', default=None, type=str, help='Path to where to dump the output PyTorch model.' ) parser.add_argument( '--model_size', default='base', type=str, choices=['base', 'large'], help='Size of the model to be converted.' ) __A : int = parser.parse_args() convert_luke_checkpoint( args.checkpoint_path, args.metadata_path, args.entity_vocab_path, args.pytorch_dump_folder_path, args.model_size, )
57
1
'''simple docstring''' import argparse import os import torch from transformers import FlavaImageCodebook, FlavaImageCodebookConfig def _UpperCamelCase ( __A , __A , __A , __A ) -> str: '''simple docstring''' UpperCamelCase__ = s.rsplit(__A , __A ) return new.join(__A ) def _UpperCamelCase ( __A ) -> Any: '''simple docstring''' return sum(param.float().sum() if "encoder.embeddings" not in key else 0 for key, param in state_dict.items() ) def _UpperCamelCase ( __A ) -> Optional[int]: '''simple docstring''' UpperCamelCase__ = {} UpperCamelCase__ = ["group_1", "group_2", "group_3", "group_4"] for key, value in state_dict.items(): for group_key in group_keys: if group_key in key: UpperCamelCase__ = key.replace(F'''{group_key}.''' , F'''{group_key}.group.''' ) if "res_path" in key: UpperCamelCase__ = key.replace("res_path." , "res_path.path." ) if key.endswith(".w" ): UpperCamelCase__ = rreplace(__A , ".w" , ".weight" , 1 ) if key.endswith(".b" ): UpperCamelCase__ = rreplace(__A , ".b" , ".bias" , 1 ) UpperCamelCase__ = value.float() return upgrade @torch.no_grad() def _UpperCamelCase ( __A , __A , __A=None , __A=True ) -> Optional[int]: '''simple docstring''' from dall_e import Encoder UpperCamelCase__ = Encoder() if os.path.exists(__A ): UpperCamelCase__ = torch.load(__A ) else: UpperCamelCase__ = torch.hub.load_state_dict_from_url(__A ) if isinstance(__A , __A ): UpperCamelCase__ = ckpt.state_dict() encoder.load_state_dict(__A ) if config_path is not None: UpperCamelCase__ = FlavaImageCodebookConfig.from_pretrained(__A ) else: UpperCamelCase__ = FlavaImageCodebookConfig() UpperCamelCase__ = FlavaImageCodebook(__A ).eval() UpperCamelCase__ = encoder.state_dict() UpperCamelCase__ = upgrade_state_dict(__A ) hf_model.load_state_dict(__A ) UpperCamelCase__ = hf_model.state_dict() UpperCamelCase__ = count_parameters(__A ) UpperCamelCase__ = count_parameters(__A ) assert torch.allclose(__A , __A , atol=1E-3 ) if save_checkpoint: hf_model.save_pretrained(__A ) else: return hf_state_dict if __name__ == "__main__": a__ : Dict = argparse.ArgumentParser() parser.add_argument('--pytorch_dump_folder_path', default=None, type=str, help='Path to the output PyTorch model.') parser.add_argument('--checkpoint_path', default=None, type=str, help='Path to flava checkpoint') parser.add_argument('--config_path', default=None, type=str, help='Path to hf config.json of model to convert') a__ : Any = parser.parse_args() convert_dalle_checkpoint(args.checkpoint_path, args.pytorch_dump_folder_path, args.config_path)
80
'''simple docstring''' from __future__ import annotations def _UpperCamelCase ( __A ) -> float: '''simple docstring''' UpperCamelCase__ = 0.00 UpperCamelCase__ = 0 for resistor in resistors: if resistor <= 0: UpperCamelCase__ = F'''Resistor at index {index} has a negative or zero value!''' raise ValueError(__A ) first_sum += 1 / float(__A ) index += 1 return 1 / first_sum def _UpperCamelCase ( __A ) -> float: '''simple docstring''' UpperCamelCase__ = 0.00 UpperCamelCase__ = 0 for resistor in resistors: sum_r += resistor if resistor < 0: UpperCamelCase__ = F'''Resistor at index {index} has a negative value!''' raise ValueError(__A ) index += 1 return sum_r if __name__ == "__main__": import doctest doctest.testmod()
80
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 _lowercase: int = logging.get_logger(__name__) class _lowercase ( lowerCAmelCase ): """simple docstring""" __A = ["pixel_values"] def __init__(self , lowerCamelCase_ = True , lowerCamelCase_ = 32 , lowerCamelCase_=PILImageResampling.BILINEAR , lowerCamelCase_ = True , **lowerCamelCase_ , ): """simple docstring""" a = do_resize a = do_rescale a = size_divisor a = resample super().__init__(**lowerCamelCase_ ) def UpperCamelCase_ (self , lowerCamelCase_ , lowerCamelCase_ , lowerCamelCase_ , lowerCamelCase_ = None , **lowerCamelCase_ ): """simple docstring""" a , a = get_image_size(lowerCamelCase_ ) # Rounds the height and width down to the closest multiple of size_divisor a = height // size_divisor * size_divisor a = width // size_divisor * size_divisor a = resize(lowerCamelCase_ , (new_h, new_w) , resample=lowerCamelCase_ , data_format=lowerCamelCase_ , **lowerCamelCase_ ) return image def UpperCamelCase_ (self , lowerCamelCase_ , lowerCamelCase_ , lowerCamelCase_ = None , **lowerCamelCase_ ): """simple docstring""" return rescale(image=lowerCamelCase_ , scale=lowerCamelCase_ , data_format=lowerCamelCase_ , **lowerCamelCase_ ) def UpperCamelCase_ (self , lowerCamelCase_ , lowerCamelCase_ = None , lowerCamelCase_ = None , lowerCamelCase_=None , lowerCamelCase_ = None , lowerCamelCase_ = None , lowerCamelCase_ = ChannelDimension.FIRST , **lowerCamelCase_ , ): """simple docstring""" a = do_resize if do_resize is not None else self.do_resize a = do_rescale if do_rescale is not None else self.do_rescale a = size_divisor if size_divisor is not None else self.size_divisor a = 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 = make_list_of_images(lowerCamelCase_ ) if not valid_images(lowerCamelCase_ ): raise ValueError("Invalid image(s)" ) # All transformations expect numpy arrays. a = [to_numpy_array(lowerCamelCase_ ) for img in images] if do_resize: a = [self.resize(lowerCamelCase_ , size_divisor=lowerCamelCase_ , resample=lowerCamelCase_ ) for image in images] if do_rescale: a = [self.rescale(lowerCamelCase_ , scale=1 / 255 ) for image in images] a = [to_channel_dimension_format(lowerCamelCase_ , lowerCamelCase_ ) for image in images] a = {"pixel_values": images} return BatchFeature(data=lowerCamelCase_ , tensor_type=lowerCamelCase_ )
71
import unittest from parameterized import parameterized from transformers import LlamaConfig, is_torch_available, set_seed 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 LlamaForCausalLM, LlamaForSequenceClassification, LlamaModel, LlamaTokenizer class _lowercase : """simple docstring""" def __init__(self , lowerCamelCase_ , lowerCamelCase_=13 , lowerCamelCase_=7 , lowerCamelCase_=True , lowerCamelCase_=True , lowerCamelCase_=False , lowerCamelCase_=True , lowerCamelCase_=99 , lowerCamelCase_=32 , lowerCamelCase_=5 , lowerCamelCase_=4 , lowerCamelCase_=37 , lowerCamelCase_="gelu" , lowerCamelCase_=0.1 , lowerCamelCase_=0.1 , lowerCamelCase_=512 , lowerCamelCase_=16 , lowerCamelCase_=2 , lowerCamelCase_=0.02 , lowerCamelCase_=3 , lowerCamelCase_=4 , lowerCamelCase_=None , ): """simple docstring""" a = parent a = batch_size a = seq_length a = is_training a = use_input_mask a = use_token_type_ids a = use_labels a = vocab_size a = hidden_size a = num_hidden_layers a = num_attention_heads a = intermediate_size a = hidden_act a = hidden_dropout_prob a = attention_probs_dropout_prob a = max_position_embeddings a = type_vocab_size a = type_sequence_label_size a = initializer_range a = num_labels a = num_choices a = scope def UpperCamelCase_ (self ): """simple docstring""" a = ids_tensor([self.batch_size, self.seq_length] , self.vocab_size ) a = None if self.use_input_mask: a = random_attention_mask([self.batch_size, self.seq_length] ) a = None if self.use_token_type_ids: a = ids_tensor([self.batch_size, self.seq_length] , self.type_vocab_size ) a = None a = None a = None if self.use_labels: a = ids_tensor([self.batch_size] , self.type_sequence_label_size ) a = ids_tensor([self.batch_size, self.seq_length] , self.num_labels ) a = ids_tensor([self.batch_size] , self.num_choices ) a = self.get_config() return config, input_ids, token_type_ids, input_mask, sequence_labels, token_labels, choice_labels def UpperCamelCase_ (self ): """simple docstring""" return LlamaConfig( 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=lowerCamelCase_ , initializer_range=self.initializer_range , ) def UpperCamelCase_ (self , lowerCamelCase_ , lowerCamelCase_ , lowerCamelCase_ , lowerCamelCase_ , lowerCamelCase_ , lowerCamelCase_ , lowerCamelCase_ ): """simple docstring""" a = LlamaModel(config=lowerCamelCase_ ) model.to(lowerCamelCase_ ) model.eval() a = model(lowerCamelCase_ , attention_mask=lowerCamelCase_ ) a = model(lowerCamelCase_ ) self.parent.assertEqual(result.last_hidden_state.shape , (self.batch_size, self.seq_length, self.hidden_size) ) def UpperCamelCase_ (self , lowerCamelCase_ , lowerCamelCase_ , lowerCamelCase_ , lowerCamelCase_ , lowerCamelCase_ , lowerCamelCase_ , lowerCamelCase_ , lowerCamelCase_ , lowerCamelCase_ , ): """simple docstring""" a = True a = LlamaModel(lowerCamelCase_ ) model.to(lowerCamelCase_ ) model.eval() a = model( lowerCamelCase_ , attention_mask=lowerCamelCase_ , encoder_hidden_states=lowerCamelCase_ , encoder_attention_mask=lowerCamelCase_ , ) a = model( lowerCamelCase_ , attention_mask=lowerCamelCase_ , encoder_hidden_states=lowerCamelCase_ , ) a = model(lowerCamelCase_ , attention_mask=lowerCamelCase_ ) self.parent.assertEqual(result.last_hidden_state.shape , (self.batch_size, self.seq_length, self.hidden_size) ) def UpperCamelCase_ (self , lowerCamelCase_ , lowerCamelCase_ , lowerCamelCase_ , lowerCamelCase_ , lowerCamelCase_ , lowerCamelCase_ , lowerCamelCase_ , lowerCamelCase_ , lowerCamelCase_ , ): """simple docstring""" a = LlamaForCausalLM(config=lowerCamelCase_ ) model.to(lowerCamelCase_ ) model.eval() a = model(lowerCamelCase_ , attention_mask=lowerCamelCase_ , labels=lowerCamelCase_ ) self.parent.assertEqual(result.logits.shape , (self.batch_size, self.seq_length, self.vocab_size) ) def UpperCamelCase_ (self , lowerCamelCase_ , lowerCamelCase_ , lowerCamelCase_ , lowerCamelCase_ , lowerCamelCase_ , lowerCamelCase_ , lowerCamelCase_ , lowerCamelCase_ , lowerCamelCase_ , ): """simple docstring""" a = True a = True a = LlamaForCausalLM(config=lowerCamelCase_ ) model.to(lowerCamelCase_ ) model.eval() # first forward pass a = model( lowerCamelCase_ , attention_mask=lowerCamelCase_ , encoder_hidden_states=lowerCamelCase_ , encoder_attention_mask=lowerCamelCase_ , use_cache=lowerCamelCase_ , ) a = outputs.past_key_values # create hypothetical multiple next token and extent to next_input_ids a = ids_tensor((self.batch_size, 3) , config.vocab_size ) a = ids_tensor((self.batch_size, 3) , vocab_size=2 ) # append to next input_ids and a = torch.cat([input_ids, next_tokens] , dim=-1 ) a = torch.cat([input_mask, next_mask] , dim=-1 ) a = model( lowerCamelCase_ , attention_mask=lowerCamelCase_ , encoder_hidden_states=lowerCamelCase_ , encoder_attention_mask=lowerCamelCase_ , output_hidden_states=lowerCamelCase_ , )["hidden_states"][0] a = model( lowerCamelCase_ , attention_mask=lowerCamelCase_ , encoder_hidden_states=lowerCamelCase_ , encoder_attention_mask=lowerCamelCase_ , past_key_values=lowerCamelCase_ , output_hidden_states=lowerCamelCase_ , )["hidden_states"][0] # select random slice a = ids_tensor((1,) , output_from_past.shape[-1] ).item() a = output_from_no_past[:, -3:, random_slice_idx].detach() a = 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(lowerCamelCase_ , lowerCamelCase_ , atol=1E-3 ) ) def UpperCamelCase_ (self ): """simple docstring""" a = self.prepare_config_and_inputs() ( ( a ) , ( a ) , ( a ) , ( a ) , ( a ) , ( a ) , ( a ) , ) = config_and_inputs a = {"input_ids": input_ids, "attention_mask": input_mask} return config, inputs_dict @require_torch class _lowercase ( lowerCAmelCase, lowerCAmelCase, lowerCAmelCase, unittest.TestCase ): """simple docstring""" __A = (LlamaModel, LlamaForCausalLM, LlamaForSequenceClassification) if is_torch_available() else () __A = (LlamaForCausalLM,) if is_torch_available() else () __A = ( { "feature-extraction": LlamaModel, "text-classification": LlamaForSequenceClassification, "text-generation": LlamaForCausalLM, "zero-shot": LlamaForSequenceClassification, } if is_torch_available() else {} ) __A = False __A = False def UpperCamelCase_ (self ): """simple docstring""" a = LlamaModelTester(self ) a = ConfigTester(self , config_class=lowerCamelCase_ , hidden_size=37 ) def UpperCamelCase_ (self ): """simple docstring""" self.config_tester.run_common_tests() def UpperCamelCase_ (self ): """simple docstring""" a = self.model_tester.prepare_config_and_inputs() self.model_tester.create_and_check_model(*lowerCamelCase_ ) def UpperCamelCase_ (self ): """simple docstring""" a = self.model_tester.prepare_config_and_inputs() for type in ["absolute", "relative_key", "relative_key_query"]: a = type self.model_tester.create_and_check_model(*lowerCamelCase_ ) def UpperCamelCase_ (self ): """simple docstring""" a , a = self.model_tester.prepare_config_and_inputs_for_common() a = 3 a = input_dict["input_ids"] a = input_ids.ne(1 ).to(lowerCamelCase_ ) a = ids_tensor([self.model_tester.batch_size] , self.model_tester.type_sequence_label_size ) a = LlamaForSequenceClassification(lowerCamelCase_ ) model.to(lowerCamelCase_ ) model.eval() a = model(lowerCamelCase_ , attention_mask=lowerCamelCase_ , labels=lowerCamelCase_ ) self.assertEqual(result.logits.shape , (self.model_tester.batch_size, self.model_tester.num_labels) ) def UpperCamelCase_ (self ): """simple docstring""" a , a = self.model_tester.prepare_config_and_inputs_for_common() a = 3 a = "single_label_classification" a = input_dict["input_ids"] a = input_ids.ne(1 ).to(lowerCamelCase_ ) a = ids_tensor([self.model_tester.batch_size] , self.model_tester.type_sequence_label_size ) a = LlamaForSequenceClassification(lowerCamelCase_ ) model.to(lowerCamelCase_ ) model.eval() a = model(lowerCamelCase_ , attention_mask=lowerCamelCase_ , labels=lowerCamelCase_ ) self.assertEqual(result.logits.shape , (self.model_tester.batch_size, self.model_tester.num_labels) ) def UpperCamelCase_ (self ): """simple docstring""" a , a = self.model_tester.prepare_config_and_inputs_for_common() a = 3 a = "multi_label_classification" a = input_dict["input_ids"] a = input_ids.ne(1 ).to(lowerCamelCase_ ) a = ids_tensor( [self.model_tester.batch_size, config.num_labels] , self.model_tester.type_sequence_label_size ).to(torch.float ) a = LlamaForSequenceClassification(lowerCamelCase_ ) model.to(lowerCamelCase_ ) model.eval() a = model(lowerCamelCase_ , attention_mask=lowerCamelCase_ , labels=lowerCamelCase_ ) self.assertEqual(result.logits.shape , (self.model_tester.batch_size, self.model_tester.num_labels) ) @unittest.skip("LLaMA buffers include complex numbers, which breaks this test" ) def UpperCamelCase_ (self ): """simple docstring""" pass @parameterized.expand([("linear",), ("dynamic",)] ) def UpperCamelCase_ (self , lowerCamelCase_ ): """simple docstring""" a , a = self.model_tester.prepare_config_and_inputs_for_common() a = ids_tensor([1, 10] , config.vocab_size ) a = ids_tensor([1, int(config.max_position_embeddings * 1.5 )] , config.vocab_size ) set_seed(42 ) # Fixed seed at init time so the two models get the same random weights a = LlamaModel(lowerCamelCase_ ) original_model.to(lowerCamelCase_ ) original_model.eval() a = original_model(lowerCamelCase_ ).last_hidden_state a = original_model(lowerCamelCase_ ).last_hidden_state set_seed(42 ) # Fixed seed at init time so the two models get the same random weights a = {"type": scaling_type, "factor": 10.0} a = LlamaModel(lowerCamelCase_ ) scaled_model.to(lowerCamelCase_ ) scaled_model.eval() a = scaled_model(lowerCamelCase_ ).last_hidden_state a = scaled_model(lowerCamelCase_ ).last_hidden_state # Dynamic scaling does not change the RoPE embeddings until it receives an input longer than the original # maximum sequence length, so the outputs for the short input should match. if scaling_type == "dynamic": self.assertTrue(torch.allclose(lowerCamelCase_ , lowerCamelCase_ , atol=1E-5 ) ) else: self.assertFalse(torch.allclose(lowerCamelCase_ , lowerCamelCase_ , atol=1E-5 ) ) # The output should be different for long inputs self.assertFalse(torch.allclose(lowerCamelCase_ , lowerCamelCase_ , atol=1E-5 ) ) @require_torch class _lowercase ( unittest.TestCase ): """simple docstring""" @unittest.skip("Logits are not exactly the same, once we fix the instabalities somehow, will update!" ) @slow def UpperCamelCase_ (self ): """simple docstring""" a = [1, 306, 4658, 278, 6593, 310, 2834, 338] a = LlamaForCausalLM.from_pretrained("meta-llama/Llama-2-7b-hf" , device_map="auto" ) a = model(torch.tensor([input_ids] ) ) # Expected mean on dim = -1 a = torch.tensor([[-6.6550, -4.1227, -4.9859, -3.2406, 0.8262, -3.0033, 1.2964, -3.3699]] ) torch.testing.assert_close(out.mean(-1 ) , lowerCamelCase_ , atol=1E-2 , rtol=1E-2 ) # slicing logits[0, 0, 0:30] # fmt: off a = torch.tensor([-12.8281, -7.4453, -0.4639, -8.0625, -7.2500, -8.0000, -6.4883, -7.7695, -7.8438, -7.0312, -6.2188, -7.1328, -1.8496, 1.9961, -8.6250, -6.7227, -12.8281, -6.9492, -7.0742, -7.7852, -7.5820, -7.9062, -6.9375, -7.9805, -8.3438, -8.1562, -8.0469, -7.6250, -7.7422, -7.3398,] ) # fmt: on torch.testing.assert_close(out[0, 0, :30] , lowerCamelCase_ , atol=1E-5 , rtol=1E-5 ) @unittest.skip("Logits are not exactly the same, once we fix the instabalities somehow, will update!" ) @slow def UpperCamelCase_ (self ): """simple docstring""" a = [1, 306, 4658, 278, 6593, 310, 2834, 338] a = LlamaForCausalLM.from_pretrained("meta-llama/Llama-2-13b-hf" , device_map="auto" ) a = model(torch.tensor(lowerCamelCase_ ) ) # Expected mean on dim = -1 a = torch.tensor([[-2.0622, -1.2794, -1.1638, -0.9788, -1.4603, -1.0238, -1.7893, -1.4411]] ) torch.testing.assert_close(out.mean(-1 ) , lowerCamelCase_ , atol=1E-2 , rtol=1E-2 ) # slicing logits[0, 0, 0:30] # fmt: off a = torch.tensor([-8.1406, -8.0547, 2.7461, -1.2344, -0.1448, -1.8262, -1.0020, -1.8154, -1.6895, -1.8516, -2.3574, -0.9277, 3.7598, 6.5742, -1.2998, -0.1177, -8.1406, -2.9688, -2.9199, -3.1699, -3.5254, -2.3555, -2.7988, -3.4141, -2.8262, -4.5195, -3.3379, -3.3164, -2.7832, -3.0273] ) # fmt: on torch.testing.assert_close(out[0, 0, :30] , lowerCamelCase_ , atol=1E-5 , rtol=1E-5 ) @unittest.skip("Logits are not exactly the same, once we fix the instabalities somehow, will update!" ) @slow def UpperCamelCase_ (self ): """simple docstring""" a = [1, 306, 4658, 278, 6593, 310, 2834, 338] a = LlamaForCausalLM.from_pretrained("meta-llama/Llama-2-13b-chat-hf" , device_map="auto" ) a = model(torch.tensor(lowerCamelCase_ ) ) # Expected mean on dim = -1 a = torch.tensor([[-0.8562, -1.8520, -0.7551, -0.4162, -1.5161, -1.2038, -2.4823, -2.3254]] ) torch.testing.assert_close(out.mean(-1 ) , lowerCamelCase_ , atol=1E-2 , rtol=1E-2 ) # slicing logits[0, 0, 0:30] # fmt: off a = torch.tensor([-2.2227, 4.8828, 0.9023, -0.4578, -0.7871, -0.1033, -0.6221, -0.5786, -0.7803, -1.0674, -1.2920, -0.1570, 0.8008, 2.0723, -0.9497, 0.2771, -2.2227, -0.7612, -1.4346, -1.2061, -1.6426, -0.3000, -0.7139, -1.1934, -1.8691, -1.6973, -1.5947, -1.2705, -0.3523, -0.5513] ) # fmt: on torch.testing.assert_close(out.mean(-1 ) , lowerCamelCase_ , atol=1E-2 , rtol=1E-2 ) @unittest.skip( "Logits are not exactly the same, once we fix the instabalities somehow, will update! Also it is gonna be a `too_slow` test" ) @slow def UpperCamelCase_ (self ): """simple docstring""" a = [1, 306, 4658, 278, 6593, 310, 2834, 338] a = LlamaForCausalLM.from_pretrained("meta-llama/Llama-2-70b-hf" , device_map="auto" ) a = model(torch.tensor(lowerCamelCase_ ) ) a = torch.tensor( [[-4.2327, -3.3360, -4.6665, -4.7631, -1.8180, -3.4170, -1.4211, -3.1810]] , dtype=torch.floataa ) torch.testing.assert_close(out.mean(-1 ) , lowerCamelCase_ , atol=1E-2 , rtol=1E-2 ) # fmt: off a = torch.tensor([-9.4922, -3.9551, 1.7998, -5.6758, -5.1055, -5.8984, -4.8320, -6.8086, -6.5391, -5.6172, -5.5820, -5.5352, 1.7881, 3.6289, -6.5117, -3.4785, -9.5000, -6.0352, -6.8125, -6.0195, -6.6836, -5.4727, -6.2812, -6.0391, -7.3398, -7.4297, -7.4844, -6.5820, -5.8789, -5.5312] ) # fmt: on torch.testing.assert_close(out[0, 0, :30] , lowerCamelCase_ , atol=1E-5 , rtol=1E-5 ) @unittest.skip("Model is curently gated" ) @slow def UpperCamelCase_ (self ): """simple docstring""" a = "Simply put, the theory of relativity states that 1) the laws of physics are the same everywhere in the universe and 2) the passage of time and the length of objects can vary depending on the observer\'s frame of reference.\n\nThe first part of the theory, that the laws of physics are the same everywhere, is known as the \"princi" a = "Simply put, the theory of relativity states that " a = LlamaTokenizer.from_pretrained("meta-llama/Llama-2-13b-chat-hf" ) a = tokenizer.encode(lowerCamelCase_ , return_tensors="pt" ) a = LlamaForCausalLM.from_pretrained( "meta-llama/Llama-2-13b-chat-hf" , device_map="sequential" , use_safetensors=lowerCamelCase_ ) # greedy generation outputs a = model.generate(lowerCamelCase_ , max_new_tokens=64 , top_p=lowerCamelCase_ , temperature=1 , do_sample=lowerCamelCase_ ) a = tokenizer.decode(generated_ids[0] , skip_special_tokens=lowerCamelCase_ ) self.assertEqual(lowerCamelCase_ , lowerCamelCase_ )
71
1
import argparse import torch from transformers import GPTaLMHeadModel, RobertaForMaskedLM if __name__ == "__main__": a_ = argparse.ArgumentParser( description=( 'Extraction some layers of the full RobertaForMaskedLM or GPT2LMHeadModel for Transfer Learned' ' Distillation' ) ) parser.add_argument('--model_type', default='roberta', choices=['roberta', 'gpt2']) parser.add_argument('--model_name', default='roberta-large', type=str) parser.add_argument('--dump_checkpoint', default='serialization_dir/tf_roberta_048131723.pth', type=str) parser.add_argument('--vocab_transform', action='store_true') a_ = parser.parse_args() if args.model_type == "roberta": a_ = RobertaForMaskedLM.from_pretrained(args.model_name) a_ = 'roberta' elif args.model_type == "gpt2": a_ = GPTaLMHeadModel.from_pretrained(args.model_name) a_ = 'transformer' a_ = model.state_dict() a_ = {} # Embeddings # if args.model_type == "gpt2": for param_name in ["wte.weight", "wpe.weight"]: a_ = state_dict[F"""{prefix}.{param_name}"""] else: for w in ["word_embeddings", "position_embeddings", "token_type_embeddings"]: a_ = F"""{prefix}.embeddings.{w}.weight""" a_ = state_dict[param_name] for w in ["weight", "bias"]: a_ = F"""{prefix}.embeddings.LayerNorm.{w}""" a_ = state_dict[param_name] # Transformer Blocks # a_ = 0 for teacher_idx in [0, 2, 4, 7, 9, 11]: if args.model_type == "gpt2": for layer in ["ln_1", "attn.c_attn", "attn.c_proj", "ln_2", "mlp.c_fc", "mlp.c_proj"]: for w in ["weight", "bias"]: a_ = state_dict[ F"""{prefix}.h.{teacher_idx}.{layer}.{w}""" ] a_ = state_dict[F"""{prefix}.h.{teacher_idx}.attn.bias"""] else: for layer in [ "attention.self.query", "attention.self.key", "attention.self.value", "attention.output.dense", "attention.output.LayerNorm", "intermediate.dense", "output.dense", "output.LayerNorm", ]: for w in ["weight", "bias"]: a_ = state_dict[ F"""{prefix}.encoder.layer.{teacher_idx}.{layer}.{w}""" ] std_idx += 1 # Language Modeling Head ###s if args.model_type == "roberta": for layer in ["lm_head.decoder.weight", "lm_head.bias"]: a_ = state_dict[F"""{layer}"""] if args.vocab_transform: for w in ["weight", "bias"]: a_ = state_dict[F"""lm_head.dense.{w}"""] a_ = state_dict[F"""lm_head.layer_norm.{w}"""] elif args.model_type == "gpt2": for w in ["weight", "bias"]: a_ = state_dict[F"""{prefix}.ln_f.{w}"""] a_ = state_dict['lm_head.weight'] print(F"""N layers selected for distillation: {std_idx}""") print(F"""Number of params transferred for distillation: {len(compressed_sd.keys())}""") print(F"""Save transferred checkpoint to {args.dump_checkpoint}.""") torch.save(compressed_sd, args.dump_checkpoint)
175
import os import posixpath import uuid from dataclasses import dataclass from typing import TYPE_CHECKING, Iterable, List, Optional, Tuple, Union import numpy as np import pyarrow as pa import datasets from datasets.arrow_writer import ArrowWriter, ParquetWriter from datasets.config import MAX_SHARD_SIZE from datasets.filesystems import ( is_remote_filesystem, rename, ) from datasets.iterable_dataset import _BaseExamplesIterable from datasets.utils.py_utils import convert_file_size_to_int a_ = datasets.utils.logging.get_logger(__name__) if TYPE_CHECKING: import pyspark @dataclass class _lowercase ( datasets.BuilderConfig ): lowercase = None def __lowercase ( lowerCamelCase : "pyspark.sql.DataFrame" , lowerCamelCase : List[int] , ): import pyspark def generate_fn(): UpperCamelCase_ : Dict = df.select('*' , pyspark.sql.functions.spark_partition_id().alias('part_id' ) ) for partition_id in partition_order: UpperCamelCase_ : Tuple = df_with_partition_id.select('*' ).where(F"part_id = {partition_id}" ).drop('part_id' ) UpperCamelCase_ : Union[str, Any] = partition_df.collect() UpperCamelCase_ : Any = 0 for row in rows: yield F"{partition_id}_{row_id}", row.asDict() row_id += 1 return generate_fn class _lowercase ( _BaseExamplesIterable ): def __init__( self : Optional[int] , snake_case : "pyspark.sql.DataFrame" , snake_case : Tuple=None , ) -> Tuple: """simple docstring""" UpperCamelCase_ : Dict = df UpperCamelCase_ : int = partition_order or range(self.df.rdd.getNumPartitions() ) UpperCamelCase_ : Optional[Any] = _generate_iterable_examples(self.df , self.partition_order ) def __iter__( self : Optional[int] ) -> Any: """simple docstring""" yield from self.generate_examples_fn() def SCREAMING_SNAKE_CASE__ ( self : Any , snake_case : np.random.Generator ) -> "SparkExamplesIterable": """simple docstring""" UpperCamelCase_ : Optional[Any] = list(range(self.df.rdd.getNumPartitions() ) ) generator.shuffle(snake_case ) return SparkExamplesIterable(self.df , partition_order=snake_case ) def SCREAMING_SNAKE_CASE__ ( self : int , snake_case : int , snake_case : int ) -> "SparkExamplesIterable": """simple docstring""" UpperCamelCase_ : Tuple = self.split_shard_indices_by_worker(snake_case , snake_case ) return SparkExamplesIterable(self.df , partition_order=snake_case ) @property def SCREAMING_SNAKE_CASE__ ( self : int ) -> int: """simple docstring""" return len(self.partition_order ) class _lowercase ( datasets.DatasetBuilder ): lowercase = SparkConfig def __init__( self : List[Any] , snake_case : "pyspark.sql.DataFrame" , snake_case : str = None , snake_case : str = None , **snake_case : Optional[Any] , ) -> List[str]: """simple docstring""" import pyspark UpperCamelCase_ : List[Any] = pyspark.sql.SparkSession.builder.getOrCreate() UpperCamelCase_ : str = df UpperCamelCase_ : Tuple = working_dir super().__init__( cache_dir=snake_case , config_name=str(self.df.semanticHash() ) , **snake_case , ) def SCREAMING_SNAKE_CASE__ ( self : Any ) -> Dict: """simple docstring""" def create_cache_and_write_probe(snake_case : str ): # makedirs with exist_ok will recursively create the directory. It will not throw an error if directories # already exist. os.makedirs(self._cache_dir , exist_ok=snake_case ) UpperCamelCase_ : Tuple = os.path.join(self._cache_dir , 'fs_test' + uuid.uuida().hex ) # Opening the file in append mode will create a new file unless it already exists, in which case it will not # change the file contents. open(snake_case , 'a' ) return [probe_file] if self._spark.conf.get('spark.master' , '' ).startswith('local' ): return # If the cluster is multi-node, make sure that the user provided a cache_dir and that it is on an NFS # accessible to the driver. # TODO: Stream batches to the driver using ArrowCollectSerializer instead of throwing an error. if self._cache_dir: UpperCamelCase_ : Tuple = ( self._spark.sparkContext.parallelize(range(1 ) , 1 ).mapPartitions(snake_case ).collect() ) if os.path.isfile(probe[0] ): return raise ValueError( 'When using Dataset.from_spark on a multi-node cluster, the driver and all workers should be able to access cache_dir' ) def SCREAMING_SNAKE_CASE__ ( self : int ) -> Union[str, Any]: """simple docstring""" return datasets.DatasetInfo(features=self.config.features ) def SCREAMING_SNAKE_CASE__ ( self : Optional[Any] , snake_case : datasets.download.download_manager.DownloadManager ) -> Optional[int]: """simple docstring""" return [datasets.SplitGenerator(name=datasets.Split.TRAIN )] def SCREAMING_SNAKE_CASE__ ( self : Optional[Any] , snake_case : Optional[int] ) -> List[Any]: """simple docstring""" import pyspark def get_arrow_batch_size(snake_case : Dict ): for batch in it: yield pa.RecordBatch.from_pydict({'batch_bytes': [batch.nbytes]} ) UpperCamelCase_ : List[str] = self.df.count() UpperCamelCase_ : Union[str, Any] = df_num_rows if df_num_rows <= 1_0_0 else 1_0_0 # Approximate the size of each row (in Arrow format) by averaging over a max-100-row sample. UpperCamelCase_ : str = ( self.df.limit(snake_case ) .repartition(1 ) .mapInArrow(snake_case , 'batch_bytes: long' ) .agg(pyspark.sql.functions.sum('batch_bytes' ).alias('sample_bytes' ) ) .collect()[0] .sample_bytes / sample_num_rows ) UpperCamelCase_ : Optional[int] = approx_bytes_per_row * df_num_rows if approx_total_size > max_shard_size: # Make sure there is at least one row per partition. UpperCamelCase_ : Optional[Any] = min(snake_case , int(approx_total_size / max_shard_size ) ) UpperCamelCase_ : int = self.df.repartition(snake_case ) def SCREAMING_SNAKE_CASE__ ( self : Union[str, Any] , snake_case : str , snake_case : str , snake_case : int , ) -> Iterable[Tuple[int, bool, Union[int, tuple]]]: """simple docstring""" import pyspark UpperCamelCase_ : List[Any] = ParquetWriter if file_format == 'parquet' else ArrowWriter UpperCamelCase_ : List[str] = os.path.join(self._working_dir , os.path.basename(snake_case ) ) if self._working_dir else fpath UpperCamelCase_ : Union[str, Any] = file_format == 'parquet' # Define these so that we don't reference self in write_arrow, which will result in a pickling error due to # pickling the SparkContext. UpperCamelCase_ : Union[str, Any] = self.config.features UpperCamelCase_ : Any = self._writer_batch_size UpperCamelCase_ : Dict = self._fs.storage_options def write_arrow(snake_case : List[str] ): # Within the same SparkContext, no two task attempts will share the same attempt ID. UpperCamelCase_ : Any = pyspark.TaskContext().taskAttemptId() UpperCamelCase_ : str = next(snake_case , snake_case ) if first_batch is None: # Some partitions might not receive any data. return pa.RecordBatch.from_arrays( [[task_id], [0], [0]] , names=['task_id', 'num_examples', 'num_bytes'] , ) UpperCamelCase_ : Any = 0 UpperCamelCase_ : Optional[Any] = writer_class( features=snake_case , path=working_fpath.replace('SSSSS' , f"{shard_id:05d}" ).replace('TTTTT' , f"{task_id:05d}" ) , writer_batch_size=snake_case , storage_options=snake_case , embed_local_files=snake_case , ) UpperCamelCase_ : str = pa.Table.from_batches([first_batch] ) writer.write_table(snake_case ) for batch in it: if max_shard_size is not None and writer._num_bytes >= max_shard_size: UpperCamelCase_, UpperCamelCase_ : str = writer.finalize() writer.close() yield pa.RecordBatch.from_arrays( [[task_id], [num_examples], [num_bytes]] , names=['task_id', 'num_examples', 'num_bytes'] , ) shard_id += 1 UpperCamelCase_ : Union[str, Any] = writer_class( features=writer._features , path=working_fpath.replace('SSSSS' , f"{shard_id:05d}" ).replace('TTTTT' , f"{task_id:05d}" ) , writer_batch_size=snake_case , storage_options=snake_case , embed_local_files=snake_case , ) UpperCamelCase_ : Optional[Any] = pa.Table.from_batches([batch] ) writer.write_table(snake_case ) if writer._num_bytes > 0: UpperCamelCase_, UpperCamelCase_ : str = writer.finalize() writer.close() yield pa.RecordBatch.from_arrays( [[task_id], [num_examples], [num_bytes]] , names=['task_id', 'num_examples', 'num_bytes'] , ) if working_fpath != fpath: for file in os.listdir(os.path.dirname(snake_case ) ): UpperCamelCase_ : Dict = os.path.join(os.path.dirname(snake_case ) , os.path.basename(snake_case ) ) shutil.move(snake_case , snake_case ) UpperCamelCase_ : int = ( self.df.mapInArrow(snake_case , 'task_id: long, num_examples: long, num_bytes: long' ) .groupBy('task_id' ) .agg( pyspark.sql.functions.sum('num_examples' ).alias('total_num_examples' ) , pyspark.sql.functions.sum('num_bytes' ).alias('total_num_bytes' ) , pyspark.sql.functions.count('num_bytes' ).alias('num_shards' ) , pyspark.sql.functions.collect_list('num_examples' ).alias('shard_lengths' ) , ) .collect() ) for row in stats: yield row.task_id, (row.total_num_examples, row.total_num_bytes, row.num_shards, row.shard_lengths) def SCREAMING_SNAKE_CASE__ ( self : int , snake_case : "datasets.SplitGenerator" , snake_case : str = "arrow" , snake_case : Optional[Union[str, int]] = None , snake_case : Optional[int] = None , **snake_case : Any , ) -> int: """simple docstring""" self._validate_cache_dir() UpperCamelCase_ : Optional[int] = convert_file_size_to_int(max_shard_size or MAX_SHARD_SIZE ) self._repartition_df_if_needed(snake_case ) UpperCamelCase_ : List[str] = not is_remote_filesystem(self._fs ) UpperCamelCase_ : List[Any] = os.path.join if is_local else posixpath.join UpperCamelCase_ : Optional[int] = '-TTTTT-SSSSS-of-NNNNN' UpperCamelCase_ : Dict = f"{self.name}-{split_generator.name}{SUFFIX}.{file_format}" UpperCamelCase_ : int = path_join(self._output_dir , snake_case ) UpperCamelCase_ : int = 0 UpperCamelCase_ : Optional[int] = 0 UpperCamelCase_ : Union[str, Any] = 0 UpperCamelCase_ : Optional[Any] = [] UpperCamelCase_ : Any = [] for task_id, content in self._prepare_split_single(snake_case , snake_case , snake_case ): ( ( UpperCamelCase_ ), ( UpperCamelCase_ ), ( UpperCamelCase_ ), ( UpperCamelCase_ ), ) : Optional[Any] = content if num_bytes > 0: total_num_examples += num_examples total_num_bytes += num_bytes total_shards += num_shards task_id_and_num_shards.append((task_id, num_shards) ) all_shard_lengths.extend(snake_case ) UpperCamelCase_ : Optional[Any] = total_num_examples UpperCamelCase_ : Any = total_num_bytes # should rename everything at the end logger.debug(f"Renaming {total_shards} shards." ) if total_shards > 1: UpperCamelCase_ : List[Any] = all_shard_lengths # Define fs outside of _rename_shard so that we don't reference self in the function, which will result in a # pickling error due to pickling the SparkContext. UpperCamelCase_ : int = self._fs # use the -SSSSS-of-NNNNN pattern def _rename_shard( snake_case : int , snake_case : int , snake_case : int , ): rename( snake_case , fpath.replace('SSSSS' , f"{shard_id:05d}" ).replace('TTTTT' , f"{task_id:05d}" ) , fpath.replace('TTTTT-SSSSS' , f"{global_shard_id:05d}" ).replace('NNNNN' , f"{total_shards:05d}" ) , ) UpperCamelCase_ : Any = [] UpperCamelCase_ : Optional[int] = 0 for i in range(len(snake_case ) ): UpperCamelCase_, UpperCamelCase_ : Union[str, Any] = task_id_and_num_shards[i] for shard_id in range(snake_case ): args.append([task_id, shard_id, global_shard_id] ) global_shard_id += 1 self._spark.sparkContext.parallelize(snake_case , len(snake_case ) ).map(lambda snake_case : _rename_shard(*snake_case ) ).collect() else: # don't use any pattern UpperCamelCase_ : Tuple = 0 UpperCamelCase_ : Optional[Any] = task_id_and_num_shards[0][0] self._rename( fpath.replace('SSSSS' , f"{shard_id:05d}" ).replace('TTTTT' , f"{task_id:05d}" ) , fpath.replace(snake_case , '' ) , ) def SCREAMING_SNAKE_CASE__ ( self : Tuple , snake_case : "datasets.SplitGenerator" , ) -> SparkExamplesIterable: """simple docstring""" return SparkExamplesIterable(self.df )
175
1
lowercase__ : List[Any] = { "a": "AAAAA", "b": "AAAAB", "c": "AAABA", "d": "AAABB", "e": "AABAA", "f": "AABAB", "g": "AABBA", "h": "AABBB", "i": "ABAAA", "j": "BBBAA", "k": "ABAAB", "l": "ABABA", "m": "ABABB", "n": "ABBAA", "o": "ABBAB", "p": "ABBBA", "q": "ABBBB", "r": "BAAAA", "s": "BAAAB", "t": "BAABA", "u": "BAABB", "v": "BBBAB", "w": "BABAA", "x": "BABAB", "y": "BABBA", "z": "BABBB", " ": " ", } lowercase__ : str = {value: key for key, value in encode_dict.items()} def SCREAMING_SNAKE_CASE ( __UpperCamelCase) -> str: a = "" for letter in word.lower(): if letter.isalpha() or letter == " ": encoded += encode_dict[letter] else: raise Exception("encode() accepts only letters of the alphabet and spaces") return encoded def SCREAMING_SNAKE_CASE ( __UpperCamelCase) -> str: if set(__UpperCamelCase) - {"A", "B", " "} != set(): raise Exception("decode() accepts only 'A', 'B' and spaces") a = "" for word in coded.split(): while len(__UpperCamelCase) != 0: decoded += decode_dict[word[:5]] a = word[5:] decoded += " " return decoded.strip() if __name__ == "__main__": from doctest import testmod testmod()
358
import os import tempfile import unittest from transformers.models.marian.convert_marian_tatoeba_to_pytorch import DEFAULT_REPO, TatoebaConverter from transformers.testing_utils import slow from transformers.utils import cached_property @unittest.skipUnless(os.path.exists(UpperCamelCase__ ) , """Tatoeba directory does not exist.""" ) class a__ ( unittest.TestCase ): @cached_property def lowerCAmelCase_ ( self ) -> List[str]: '''simple docstring''' a = tempfile.mkdtemp() return TatoebaConverter(save_dir=A ) @slow def lowerCAmelCase_ ( self ) -> int: '''simple docstring''' self.resolver.convert_models(["heb-eng"] ) @slow def lowerCAmelCase_ ( self ) -> Dict: '''simple docstring''' a , a = self.resolver.write_model_card("opus-mt-he-en" , dry_run=A ) assert mmeta["long_pair"] == "heb-eng"
180
0
from math import sqrt def lowerCAmelCase_ ( _lowercase : Tuple) -> int: """simple docstring""" a__ : str = 0 for i in range(1 , int(sqrt(_lowercase) + 1)): if n % i == 0 and i != sqrt(_lowercase): total += i + n // i elif i == sqrt(_lowercase): total += i return total - n def lowerCAmelCase_ ( _lowercase : Optional[Any] = 1_0000) -> int: """simple docstring""" a__ : Optional[int] = sum( i for i in range(1 , _lowercase) if sum_of_divisors(sum_of_divisors(_lowercase)) == i and sum_of_divisors(_lowercase) != i) return total if __name__ == "__main__": print(solution(int(str(input()).strip())))
170
"""simple docstring""" from typing import Dict, List, Optional, Tuple, Union import numpy as np from ...image_processing_utils import BaseImageProcessor, BatchFeature, get_size_dict from ...image_transforms import ( center_crop, flip_channel_order, get_resize_output_image_size, rescale, resize, to_channel_dimension_format, ) from ...image_utils import ( ChannelDimension, ImageInput, PILImageResampling, make_list_of_images, to_numpy_array, valid_images, ) from ...utils import TensorType, is_torch_available, is_torch_tensor, is_vision_available, logging if is_vision_available(): import PIL if is_torch_available(): import torch _A = logging.get_logger(__name__) class lowerCamelCase ( lowerCAmelCase__ ): '''simple docstring''' SCREAMING_SNAKE_CASE = ['pixel_values'] def __init__(self , _lowerCamelCase = True , _lowerCamelCase = None , _lowerCamelCase = PILImageResampling.BILINEAR , _lowerCamelCase = True , _lowerCamelCase = 1 / 255 , _lowerCamelCase = True , _lowerCamelCase = None , _lowerCamelCase = True , **_lowerCamelCase , ): """simple docstring""" super().__init__(**_lowerCamelCase ) UpperCAmelCase__ : Optional[Any] = size if size is not None else {"""shortest_edge""": 224} UpperCAmelCase__ : List[Any] = get_size_dict(_lowerCamelCase , default_to_square=_lowerCamelCase ) UpperCAmelCase__ : str = crop_size if crop_size is not None else {"""height""": 256, """width""": 256} UpperCAmelCase__ : str = get_size_dict(_lowerCamelCase , param_name="""crop_size""" ) UpperCAmelCase__ : int = do_resize UpperCAmelCase__ : Any = size UpperCAmelCase__ : int = resample UpperCAmelCase__ : Union[str, Any] = do_rescale UpperCAmelCase__ : Union[str, Any] = rescale_factor UpperCAmelCase__ : str = do_center_crop UpperCAmelCase__ : Dict = crop_size UpperCAmelCase__ : List[str] = do_flip_channel_order def _a (self , _lowerCamelCase , _lowerCamelCase , _lowerCamelCase = PIL.Image.BILINEAR , _lowerCamelCase = None , **_lowerCamelCase , ): """simple docstring""" UpperCAmelCase__ : List[str] = get_size_dict(_lowerCamelCase , default_to_square=_lowerCamelCase ) if "shortest_edge" not in size: raise ValueError(F"""The `size` dictionary must contain the key `shortest_edge`. Got {size.keys()}""" ) UpperCAmelCase__ : Union[str, Any] = get_resize_output_image_size(_lowerCamelCase , size=size["""shortest_edge"""] , default_to_square=_lowerCamelCase ) return resize(_lowerCamelCase , size=_lowerCamelCase , resample=_lowerCamelCase , data_format=_lowerCamelCase , **_lowerCamelCase ) def _a (self , _lowerCamelCase , _lowerCamelCase , _lowerCamelCase = None , **_lowerCamelCase , ): """simple docstring""" UpperCAmelCase__ : Union[str, Any] = get_size_dict(_lowerCamelCase ) 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()}""" ) return center_crop(_lowerCamelCase , size=(size["""height"""], size["""width"""]) , data_format=_lowerCamelCase , **_lowerCamelCase ) def _a (self , _lowerCamelCase , _lowerCamelCase , _lowerCamelCase = None , **_lowerCamelCase , ): """simple docstring""" return rescale(_lowerCamelCase , scale=_lowerCamelCase , data_format=_lowerCamelCase , **_lowerCamelCase ) def _a (self , _lowerCamelCase , _lowerCamelCase = None ): """simple docstring""" return flip_channel_order(_lowerCamelCase , data_format=_lowerCamelCase ) def _a (self , _lowerCamelCase , _lowerCamelCase = None , _lowerCamelCase = None , _lowerCamelCase = None , _lowerCamelCase = None , _lowerCamelCase = None , _lowerCamelCase = None , _lowerCamelCase = None , _lowerCamelCase = None , _lowerCamelCase = None , _lowerCamelCase = ChannelDimension.FIRST , **_lowerCamelCase , ): """simple docstring""" UpperCAmelCase__ : Any = do_resize if do_resize is not None else self.do_resize UpperCAmelCase__ : List[str] = resample if resample is not None else self.resample UpperCAmelCase__ : Tuple = do_rescale if do_rescale is not None else self.do_rescale UpperCAmelCase__ : Optional[Any] = rescale_factor if rescale_factor is not None else self.rescale_factor UpperCAmelCase__ : List[Any] = do_center_crop if do_center_crop is not None else self.do_center_crop UpperCAmelCase__ : int = ( do_flip_channel_order if do_flip_channel_order is not None else self.do_flip_channel_order ) UpperCAmelCase__ : List[str] = size if size is not None else self.size UpperCAmelCase__ : Tuple = get_size_dict(_lowerCamelCase , default_to_square=_lowerCamelCase ) UpperCAmelCase__ : Optional[int] = crop_size if crop_size is not None else self.crop_size UpperCAmelCase__ : str = get_size_dict(_lowerCamelCase , param_name="""crop_size""" ) UpperCAmelCase__ : List[str] = make_list_of_images(_lowerCamelCase ) if not valid_images(_lowerCamelCase ): 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_center_crop and crop_size is None: raise ValueError("""Crop size must be specified if do_center_crop is True.""" ) # All transformations expect numpy arrays. UpperCAmelCase__ : Union[str, Any] = [to_numpy_array(_lowerCamelCase ) for image in images] if do_resize: UpperCAmelCase__ : Tuple = [self.resize(image=_lowerCamelCase , size=_lowerCamelCase , resample=_lowerCamelCase ) for image in images] if do_center_crop: UpperCAmelCase__ : Optional[Any] = [self.center_crop(image=_lowerCamelCase , size=_lowerCamelCase ) for image in images] if do_rescale: UpperCAmelCase__ : List[Any] = [self.rescale(image=_lowerCamelCase , scale=_lowerCamelCase ) for image in images] # the pretrained checkpoints assume images are BGR, not RGB if do_flip_channel_order: UpperCAmelCase__ : Any = [self.flip_channel_order(image=_lowerCamelCase ) for image in images] UpperCAmelCase__ : int = [to_channel_dimension_format(_lowerCamelCase , _lowerCamelCase ) for image in images] UpperCAmelCase__ : Optional[Any] = {"""pixel_values""": images} return BatchFeature(data=_lowerCamelCase , tensor_type=_lowerCamelCase ) def _a (self , _lowerCamelCase , _lowerCamelCase = None ): """simple docstring""" UpperCAmelCase__ : List[str] = outputs.logits # Resize logits and compute semantic segmentation maps if target_sizes is not None: if len(_lowerCamelCase ) != len(_lowerCamelCase ): raise ValueError( """Make sure that you pass in as many target sizes as the batch dimension of the logits""" ) if is_torch_tensor(_lowerCamelCase ): UpperCAmelCase__ : Optional[int] = target_sizes.numpy() UpperCAmelCase__ : Tuple = [] for idx in range(len(_lowerCamelCase ) ): UpperCAmelCase__ : Union[str, Any] = torch.nn.functional.interpolate( logits[idx].unsqueeze(dim=0 ) , size=target_sizes[idx] , mode="""bilinear""" , align_corners=_lowerCamelCase ) UpperCAmelCase__ : str = resized_logits[0].argmax(dim=0 ) semantic_segmentation.append(_lowerCamelCase ) else: UpperCAmelCase__ : str = logits.argmax(dim=1 ) UpperCAmelCase__ : int = [semantic_segmentation[i] for i in range(semantic_segmentation.shape[0] )] return semantic_segmentation
171
0
import logging import os import sys from dataclasses import dataclass, field from importlib import import_module from typing import Dict, List, Optional, Tuple import numpy as np from seqeval.metrics import accuracy_score, fa_score, precision_score, recall_score from torch import nn from utils_ner import Split, TokenClassificationDataset, TokenClassificationTask import transformers from transformers import ( AutoConfig, AutoModelForTokenClassification, AutoTokenizer, DataCollatorWithPadding, EvalPrediction, HfArgumentParser, Trainer, TrainingArguments, set_seed, ) from transformers.trainer_utils import is_main_process __lowerCamelCase : Tuple = logging.getLogger(__name__) @dataclass class A__ : _UpperCAmelCase :str = field( metadata={'help': 'Path to pretrained model or model identifier from huggingface.co/models'} ) _UpperCAmelCase :Optional[str] = field( default=__SCREAMING_SNAKE_CASE , metadata={'help': 'Pretrained config name or path if not the same as model_name'} ) _UpperCAmelCase :Optional[str] = field( default='NER' , metadata={'help': 'Task type to fine tune in training (e.g. NER, POS, etc)'} ) _UpperCAmelCase :Optional[str] = field( default=__SCREAMING_SNAKE_CASE , metadata={'help': 'Pretrained tokenizer name or path if not the same as model_name'} ) _UpperCAmelCase :bool = field(default=__SCREAMING_SNAKE_CASE , metadata={'help': 'Set this flag to use fast tokenization.'} ) # If you want to tweak more attributes on your tokenizer, you should do it in a distinct script, # or just modify its tokenizer_config.json. _UpperCAmelCase :Optional[str] = field( default=__SCREAMING_SNAKE_CASE , metadata={'help': 'Where do you want to store the pretrained models downloaded from huggingface.co'} , ) @dataclass class A__ : _UpperCAmelCase :str = field( metadata={'help': 'The input data dir. Should contain the .txt files for a CoNLL-2003-formatted task.'} ) _UpperCAmelCase :Optional[str] = field( default=__SCREAMING_SNAKE_CASE , metadata={'help': 'Path to a file containing all labels. If not specified, CoNLL-2003 labels are used.'} , ) _UpperCAmelCase :int = field( default=1_2_8 , metadata={ 'help': ( 'The maximum total input sequence length after tokenization. Sequences longer ' 'than this will be truncated, sequences shorter will be padded.' ) } , ) _UpperCAmelCase :bool = field( default=__SCREAMING_SNAKE_CASE , metadata={'help': 'Overwrite the cached training and evaluation sets'} ) def A_ ( ) -> Optional[int]: UpperCamelCase : 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. UpperCamelCase : Dict = parser.parse_json_file(json_file=os.path.abspath(sys.argv[1] ) ) else: UpperCamelCase : Tuple = parser.parse_args_into_dataclasses() if ( os.path.exists(training_args.output_dir ) and os.listdir(training_args.output_dir ) and training_args.do_train and not training_args.overwrite_output_dir ): raise ValueError( F"""Output directory ({training_args.output_dir}) already exists and is not empty. Use""" " --overwrite_output_dir to overcome." ) UpperCamelCase : Tuple = import_module("tasks" ) try: UpperCamelCase : Union[str, Any] = getattr(_SCREAMING_SNAKE_CASE , model_args.task_type ) UpperCamelCase : TokenClassificationTask = token_classification_task_clazz() except AttributeError: raise ValueError( F"""Task {model_args.task_type} needs to be defined as a TokenClassificationTask subclass in {module}. """ F"""Available tasks classes are: {TokenClassificationTask.__subclasses__()}""" ) # Setup logging logging.basicConfig( format="%(asctime)s - %(levelname)s - %(name)s - %(message)s" , datefmt="%m/%d/%Y %H:%M:%S" , level=logging.INFO if training_args.local_rank in [-1, 0] else logging.WARN , ) logger.warning( "Process rank: %s, device: %s, n_gpu: %s, distributed training: %s, 16-bits training: %s" , training_args.local_rank , training_args.device , training_args.n_gpu , bool(training_args.local_rank != -1 ) , 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() transformers.utils.logging.enable_default_handler() transformers.utils.logging.enable_explicit_format() logger.info("Training/evaluation parameters %s" , _SCREAMING_SNAKE_CASE ) # Set seed set_seed(training_args.seed ) # Prepare CONLL-2003 task UpperCamelCase : Optional[int] = token_classification_task.get_labels(data_args.labels ) UpperCamelCase : Dict[int, str] = dict(enumerate(_SCREAMING_SNAKE_CASE ) ) UpperCamelCase : Optional[Any] = len(_SCREAMING_SNAKE_CASE ) # Load pretrained model and tokenizer # # Distributed training: # The .from_pretrained methods guarantee that only one local process can concurrently # download model & vocab. UpperCamelCase : Dict = AutoConfig.from_pretrained( model_args.config_name if model_args.config_name else model_args.model_name_or_path , num_labels=_SCREAMING_SNAKE_CASE , idalabel=_SCREAMING_SNAKE_CASE , labelaid={label: i for i, label in enumerate(_SCREAMING_SNAKE_CASE )} , cache_dir=model_args.cache_dir , ) UpperCamelCase : Optional[Any] = 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 , ) UpperCamelCase : Dict = AutoModelForTokenClassification.from_pretrained( model_args.model_name_or_path , from_tf=bool(".ckpt" in model_args.model_name_or_path ) , config=_SCREAMING_SNAKE_CASE , cache_dir=model_args.cache_dir , ) # Get datasets UpperCamelCase : Optional[Any] = ( TokenClassificationDataset( token_classification_task=_SCREAMING_SNAKE_CASE , data_dir=data_args.data_dir , tokenizer=_SCREAMING_SNAKE_CASE , labels=_SCREAMING_SNAKE_CASE , model_type=config.model_type , max_seq_length=data_args.max_seq_length , overwrite_cache=data_args.overwrite_cache , mode=Split.train , ) if training_args.do_train else None ) UpperCamelCase : Tuple = ( TokenClassificationDataset( token_classification_task=_SCREAMING_SNAKE_CASE , data_dir=data_args.data_dir , tokenizer=_SCREAMING_SNAKE_CASE , labels=_SCREAMING_SNAKE_CASE , model_type=config.model_type , max_seq_length=data_args.max_seq_length , overwrite_cache=data_args.overwrite_cache , mode=Split.dev , ) if training_args.do_eval else None ) def align_predictions(_lowerCAmelCase , _lowerCAmelCase ) -> Tuple[List[int], List[int]]: UpperCamelCase : Optional[int] = np.argmax(_SCREAMING_SNAKE_CASE , axis=2 ) UpperCamelCase : Optional[Any] = preds.shape UpperCamelCase : str = [[] for _ in range(_SCREAMING_SNAKE_CASE )] UpperCamelCase : Any = [[] for _ in range(_SCREAMING_SNAKE_CASE )] for i in range(_SCREAMING_SNAKE_CASE ): for j in range(_SCREAMING_SNAKE_CASE ): if label_ids[i, j] != nn.CrossEntropyLoss().ignore_index: out_label_list[i].append(label_map[label_ids[i][j]] ) preds_list[i].append(label_map[preds[i][j]] ) return preds_list, out_label_list def compute_metrics(_lowerCAmelCase ) -> Dict: UpperCamelCase : Optional[Any] = align_predictions(p.predictions , p.label_ids ) return { "accuracy_score": accuracy_score(_SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE ), "precision": precision_score(_SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE ), "recall": recall_score(_SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE ), "f1": fa_score(_SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE ), } # Data collator UpperCamelCase : List[Any] = DataCollatorWithPadding(_SCREAMING_SNAKE_CASE , pad_to_multiple_of=8 ) if training_args.fpaa else None # Initialize our Trainer UpperCamelCase : Tuple = Trainer( model=_SCREAMING_SNAKE_CASE , args=_SCREAMING_SNAKE_CASE , train_dataset=_SCREAMING_SNAKE_CASE , eval_dataset=_SCREAMING_SNAKE_CASE , compute_metrics=_SCREAMING_SNAKE_CASE , data_collator=_SCREAMING_SNAKE_CASE , ) # Training if training_args.do_train: trainer.train( model_path=model_args.model_name_or_path if os.path.isdir(model_args.model_name_or_path ) else None ) trainer.save_model() # For convenience, we also re-save the tokenizer to the same directory, # so that you can share your model easily on huggingface.co/models =) if trainer.is_world_process_zero(): tokenizer.save_pretrained(training_args.output_dir ) # Evaluation UpperCamelCase : str = {} if training_args.do_eval: logger.info("*** Evaluate ***" ) UpperCamelCase : List[Any] = trainer.evaluate() UpperCamelCase : Any = os.path.join(training_args.output_dir , "eval_results.txt" ) if trainer.is_world_process_zero(): with open(_SCREAMING_SNAKE_CASE , "w" ) as writer: logger.info("***** Eval results *****" ) for key, value in result.items(): logger.info(" %s = %s" , _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE ) writer.write("%s = %s\n" % (key, value) ) results.update(_SCREAMING_SNAKE_CASE ) # Predict if training_args.do_predict: UpperCamelCase : str = TokenClassificationDataset( token_classification_task=_SCREAMING_SNAKE_CASE , data_dir=data_args.data_dir , tokenizer=_SCREAMING_SNAKE_CASE , labels=_SCREAMING_SNAKE_CASE , model_type=config.model_type , max_seq_length=data_args.max_seq_length , overwrite_cache=data_args.overwrite_cache , mode=Split.test , ) UpperCamelCase : List[Any] = trainer.predict(_SCREAMING_SNAKE_CASE ) UpperCamelCase : int = align_predictions(_SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE ) UpperCamelCase : Optional[int] = os.path.join(training_args.output_dir , "test_results.txt" ) if trainer.is_world_process_zero(): with open(_SCREAMING_SNAKE_CASE , "w" ) as writer: for key, value in metrics.items(): logger.info(" %s = %s" , _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE ) writer.write("%s = %s\n" % (key, value) ) # Save predictions UpperCamelCase : List[Any] = os.path.join(training_args.output_dir , "test_predictions.txt" ) if trainer.is_world_process_zero(): with open(_SCREAMING_SNAKE_CASE , "w" ) as writer: with open(os.path.join(data_args.data_dir , "test.txt" ) , "r" ) as f: token_classification_task.write_predictions_to_file(_SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE ) return results def A_ ( _lowerCAmelCase ) -> Optional[int]: main() if __name__ == "__main__": main()
351
import argparse import numpy as np import torch from transformers import SpeechTaHifiGan, SpeechTaHifiGanConfig, logging logging.set_verbosity_info() __lowerCamelCase : Dict = logging.get_logger("""transformers.models.speecht5""") def A_ ( _lowerCAmelCase , _lowerCAmelCase , _lowerCAmelCase ) -> Tuple: hf_model.apply_weight_norm() UpperCamelCase : int = checkpoint["input_conv.weight_g"] UpperCamelCase : Dict = checkpoint["input_conv.weight_v"] UpperCamelCase : List[Any] = checkpoint["input_conv.bias"] for i in range(len(config.upsample_rates ) ): UpperCamelCase : Any = checkpoint[F"""upsamples.{i}.1.weight_g"""] UpperCamelCase : List[Any] = checkpoint[F"""upsamples.{i}.1.weight_v"""] UpperCamelCase : Optional[Any] = checkpoint[F"""upsamples.{i}.1.bias"""] for i in range(len(config.upsample_rates ) * len(config.resblock_kernel_sizes ) ): for j in range(len(config.resblock_dilation_sizes ) ): UpperCamelCase : Union[str, Any] = checkpoint[F"""blocks.{i}.convs1.{j}.1.weight_g"""] UpperCamelCase : int = checkpoint[F"""blocks.{i}.convs1.{j}.1.weight_v"""] UpperCamelCase : str = checkpoint[F"""blocks.{i}.convs1.{j}.1.bias"""] UpperCamelCase : Union[str, Any] = checkpoint[F"""blocks.{i}.convs2.{j}.1.weight_g"""] UpperCamelCase : int = checkpoint[F"""blocks.{i}.convs2.{j}.1.weight_v"""] UpperCamelCase : Optional[Any] = checkpoint[F"""blocks.{i}.convs2.{j}.1.bias"""] UpperCamelCase : Tuple = checkpoint["output_conv.1.weight_g"] UpperCamelCase : Tuple = checkpoint["output_conv.1.weight_v"] UpperCamelCase : int = checkpoint["output_conv.1.bias"] hf_model.remove_weight_norm() @torch.no_grad() def A_ ( _lowerCAmelCase , _lowerCAmelCase , _lowerCAmelCase , _lowerCAmelCase=None , _lowerCAmelCase=None , ) -> Tuple: if config_path is not None: UpperCamelCase : List[Any] = SpeechTaHifiGanConfig.from_pretrained(_lowerCAmelCase ) else: UpperCamelCase : Optional[int] = SpeechTaHifiGanConfig() UpperCamelCase : List[str] = SpeechTaHifiGan(_lowerCAmelCase ) UpperCamelCase : str = torch.load(_lowerCAmelCase ) load_weights(orig_checkpoint["model"]["generator"] , _lowerCAmelCase , _lowerCAmelCase ) UpperCamelCase : List[Any] = np.load(_lowerCAmelCase ) UpperCamelCase : List[str] = stats[0].reshape(-1 ) UpperCamelCase : Tuple = stats[1].reshape(-1 ) UpperCamelCase : Any = torch.from_numpy(_lowerCAmelCase ).float() UpperCamelCase : Any = torch.from_numpy(_lowerCAmelCase ).float() model.save_pretrained(_lowerCAmelCase ) if repo_id: print("Pushing to the hub..." ) model.push_to_hub(_lowerCAmelCase ) if __name__ == "__main__": __lowerCamelCase : Union[str, Any] = argparse.ArgumentParser() parser.add_argument("""--checkpoint_path""", required=True, default=None, type=str, help="""Path to original checkpoint""") parser.add_argument("""--stats_path""", required=True, default=None, type=str, help="""Path to stats.npy file""") 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.""" ) __lowerCamelCase : Tuple = parser.parse_args() convert_hifigan_checkpoint( args.checkpoint_path, args.stats_path, args.pytorch_dump_folder_path, args.config_path, args.push_to_hub, )
140
0
import unittest import numpy as np import torch from transformers import CLIPTextConfig, CLIPTextModel from diffusers import DDIMScheduler, LDMPipeline, UNetaDModel, VQModel from diffusers.utils.testing_utils import enable_full_determinism, require_torch, slow, torch_device enable_full_determinism() class a__ ( unittest.TestCase ): """simple docstring""" @property def UpperCamelCase ( self ) -> List[str]: '''simple docstring''' torch.manual_seed(0 ) A__ = UNetaDModel( block_out_channels=(32, 64) , layers_per_block=2 , sample_size=32 , in_channels=3 , out_channels=3 , down_block_types=("DownBlock2D", "AttnDownBlock2D") , up_block_types=("AttnUpBlock2D", "UpBlock2D") , ) return model @property def UpperCamelCase ( self ) -> Any: '''simple docstring''' torch.manual_seed(0 ) A__ = VQModel( block_out_channels=[32, 64] , in_channels=3 , out_channels=3 , down_block_types=["DownEncoderBlock2D", "DownEncoderBlock2D"] , up_block_types=["UpDecoderBlock2D", "UpDecoderBlock2D"] , latent_channels=3 , ) return model @property def UpperCamelCase ( self ) -> Any: '''simple docstring''' torch.manual_seed(0 ) A__ = 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 , ) return CLIPTextModel(lowercase ) def UpperCamelCase ( self ) -> int: '''simple docstring''' A__ = self.dummy_uncond_unet A__ = DDIMScheduler() A__ = self.dummy_vq_model A__ = LDMPipeline(unet=lowercase , vqvae=lowercase , scheduler=lowercase ) ldm.to(lowercase ) ldm.set_progress_bar_config(disable=lowercase ) A__ = torch.manual_seed(0 ) A__ = ldm(generator=lowercase , num_inference_steps=2 , output_type="numpy" ).images A__ = torch.manual_seed(0 ) A__ = ldm(generator=lowercase , num_inference_steps=2 , output_type="numpy" , return_dict=lowercase )[0] A__ = image[0, -3:, -3:, -1] A__ = image_from_tuple[0, -3:, -3:, -1] assert image.shape == (1, 64, 64, 3) A__ = np.array([0.8512, 0.818, 0.6411, 0.6808, 0.4465, 0.5618, 0.46, 0.6231, 0.5172] ) A__ = 1e-2 if torch_device != "mps" else 3e-2 assert np.abs(image_slice.flatten() - expected_slice ).max() < tolerance assert np.abs(image_from_tuple_slice.flatten() - expected_slice ).max() < tolerance @slow @require_torch class a__ ( unittest.TestCase ): """simple docstring""" def UpperCamelCase ( self ) -> Optional[Any]: '''simple docstring''' A__ = LDMPipeline.from_pretrained("CompVis/ldm-celebahq-256" ) ldm.to(lowercase ) ldm.set_progress_bar_config(disable=lowercase ) A__ = torch.manual_seed(0 ) A__ = ldm(generator=lowercase , num_inference_steps=5 , output_type="numpy" ).images A__ = image[0, -3:, -3:, -1] assert image.shape == (1, 256, 256, 3) A__ = np.array([0.4399, 0.4_4975, 0.4_6825, 0.474, 0.4359, 0.4581, 0.4_5095, 0.4341, 0.4447] ) A__ = 1e-2 if torch_device != "mps" else 3e-2 assert np.abs(image_slice.flatten() - expected_slice ).max() < tolerance
68
import io import json import unittest from parameterized import parameterized from transformers import FSMTForConditionalGeneration, FSMTTokenizer from transformers.testing_utils import get_tests_dir, require_torch, slow, torch_device from utils import calculate_bleu lowerCAmelCase__ = get_tests_dir() + """/test_data/fsmt/fsmt_val_data.json""" with io.open(filename, """r""", encoding="""utf-8""") as f: lowerCAmelCase__ = json.load(f) @require_torch class a__ ( unittest.TestCase ): """simple docstring""" def UpperCamelCase ( self , lowercase ) -> int: '''simple docstring''' return FSMTTokenizer.from_pretrained(lowercase ) def UpperCamelCase ( self , lowercase ) -> Optional[int]: '''simple docstring''' A__ = FSMTForConditionalGeneration.from_pretrained(lowercase ).to(lowercase ) if torch_device == "cuda": model.half() return model @parameterized.expand( [ ["en-ru", 26.0], ["ru-en", 22.0], ["en-de", 22.0], ["de-en", 29.0], ] ) @slow def UpperCamelCase ( self , lowercase , lowercase ) -> List[Any]: '''simple docstring''' A__ = F'facebook/wmt19-{pair}' A__ = self.get_tokenizer(lowercase ) A__ = self.get_model(lowercase ) A__ = bleu_data[pair]["src"] A__ = bleu_data[pair]["tgt"] A__ = tokenizer(lowercase , return_tensors="pt" , truncation=lowercase , padding="longest" ).to(lowercase ) A__ = model.generate( input_ids=batch.input_ids , num_beams=8 , ) A__ = tokenizer.batch_decode( lowercase , skip_special_tokens=lowercase , clean_up_tokenization_spaces=lowercase ) A__ = calculate_bleu(lowercase , lowercase ) print(lowercase ) self.assertGreaterEqual(scores["bleu"] , lowercase )
68
1
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 ( lowerCAmelCase , lowerCAmelCase , lowerCAmelCase ): lowerCAmelCase_ : Any = [R"h\.\d+\.attn\.bias", R"h\.\d+\.attn\.masked_bias"] @register_to_config def __init__( self : List[Any] , UpperCAmelCase_ : int , UpperCAmelCase_ : int , UpperCAmelCase_ : Optional[int] = None , UpperCAmelCase_ : int = 50257 , UpperCAmelCase_ : int = 1024 , UpperCAmelCase_ : int = 768 , UpperCAmelCase_ : int = 12 , UpperCAmelCase_ : int = 12 , UpperCAmelCase_ : Optional[int] = None , UpperCAmelCase_ : str = "gelu_new" , UpperCAmelCase_ : float = 0.1 , UpperCAmelCase_ : float = 0.1 , UpperCAmelCase_ : float = 0.1 , UpperCAmelCase_ : float = 1E-5 , UpperCAmelCase_ : float = 0.02 , UpperCAmelCase_ : bool = True , UpperCAmelCase_ : bool = True , UpperCAmelCase_ : bool = False , UpperCAmelCase_ : bool = False , ): super().__init__() lowerCAmelCase : Tuple = 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." ) lowerCAmelCase : str = prefix_inner_dim lowerCAmelCase : Any = prefix_hidden_dim lowerCAmelCase : int = ( nn.Linear(self.prefix_inner_dim , self.prefix_hidden_dim ) if self.prefix_hidden_dim is not None else nn.Identity() ) lowerCAmelCase : Optional[Any] = ( nn.Linear(self.prefix_hidden_dim , UpperCAmelCase_ ) if self.prefix_hidden_dim is not None else nn.Identity() ) lowerCAmelCase : Union[str, Any] = GPTaConfig( vocab_size=UpperCAmelCase_ , n_positions=UpperCAmelCase_ , n_embd=UpperCAmelCase_ , n_layer=UpperCAmelCase_ , n_head=UpperCAmelCase_ , n_inner=UpperCAmelCase_ , activation_function=UpperCAmelCase_ , resid_pdrop=UpperCAmelCase_ , embd_pdrop=UpperCAmelCase_ , attn_pdrop=UpperCAmelCase_ , layer_norm_epsilon=UpperCAmelCase_ , initializer_range=UpperCAmelCase_ , scale_attn_weights=UpperCAmelCase_ , use_cache=UpperCAmelCase_ , scale_attn_by_inverse_layer_idx=UpperCAmelCase_ , reorder_and_upcast_attn=UpperCAmelCase_ , ) lowerCAmelCase : Union[str, Any] = GPTaLMHeadModel(UpperCAmelCase_ ) def lowercase__ ( self : Any , UpperCAmelCase_ : torch.Tensor , UpperCAmelCase_ : torch.Tensor , UpperCAmelCase_ : Optional[torch.Tensor] = None , UpperCAmelCase_ : Optional[torch.Tensor] = None , ): lowerCAmelCase : Dict = self.transformer.transformer.wte(UpperCAmelCase_ ) lowerCAmelCase : Union[str, Any] = self.encode_prefix(UpperCAmelCase_ ) lowerCAmelCase : Optional[int] = self.decode_prefix(UpperCAmelCase_ ) lowerCAmelCase : Optional[Any] = torch.cat((prefix_embeds, embedding_text) , dim=1 ) if labels is not None: lowerCAmelCase : Any = self.get_dummy_token(input_ids.shape[0] , input_ids.device ) lowerCAmelCase : Optional[int] = torch.cat((dummy_token, input_ids) , dim=1 ) lowerCAmelCase : int = self.transformer(inputs_embeds=UpperCAmelCase_ , labels=UpperCAmelCase_ , attention_mask=UpperCAmelCase_ ) if self.prefix_hidden_dim is not None: return out, hidden else: return out def lowercase__ ( self : Union[str, Any] , UpperCAmelCase_ : int , UpperCAmelCase_ : torch.device ): return torch.zeros(UpperCAmelCase_ , self.prefix_length , dtype=torch.intaa , device=UpperCAmelCase_ ) def lowercase__ ( self : Union[str, Any] , UpperCAmelCase_ : str ): return self.encode_prefix(UpperCAmelCase_ ) @torch.no_grad() def lowercase__ ( self : Optional[int] , UpperCAmelCase_ : str , UpperCAmelCase_ : List[str] , UpperCAmelCase_ : str ): lowerCAmelCase : str = torch.split(UpperCAmelCase_ , 1 , dim=0 ) lowerCAmelCase : int = [] lowerCAmelCase : Optional[Any] = [] for feature in features: lowerCAmelCase : Dict = self.decode_prefix(feature.to(UpperCAmelCase_ ) ) # back to the clip feature # Only support beam search for now lowerCAmelCase , lowerCAmelCase : Tuple = self.generate_beam( input_embeds=UpperCAmelCase_ , device=UpperCAmelCase_ , eos_token_id=UpperCAmelCase_ ) generated_tokens.append(output_tokens[0] ) generated_seq_lengths.append(seq_lengths[0] ) lowerCAmelCase : Dict = torch.stack(UpperCAmelCase_ ) lowerCAmelCase : Union[str, Any] = torch.stack(UpperCAmelCase_ ) return generated_tokens, generated_seq_lengths @torch.no_grad() def lowercase__ ( self : List[str] , UpperCAmelCase_ : Optional[int]=None , UpperCAmelCase_ : Tuple=None , UpperCAmelCase_ : int=None , UpperCAmelCase_ : int = 5 , UpperCAmelCase_ : int = 67 , UpperCAmelCase_ : float = 1.0 , UpperCAmelCase_ : Optional[int] = None , ): lowerCAmelCase : Tuple = eos_token_id lowerCAmelCase : int = None lowerCAmelCase : Union[str, Any] = None lowerCAmelCase : str = torch.ones(UpperCAmelCase_ , device=UpperCAmelCase_ , dtype=torch.int ) lowerCAmelCase : Any = torch.zeros(UpperCAmelCase_ , device=UpperCAmelCase_ , dtype=torch.bool ) if input_embeds is not None: lowerCAmelCase : int = input_embeds else: lowerCAmelCase : Union[str, Any] = self.transformer.transformer.wte(UpperCAmelCase_ ) for i in range(UpperCAmelCase_ ): lowerCAmelCase : Union[str, Any] = self.transformer(inputs_embeds=UpperCAmelCase_ ) lowerCAmelCase : Any = outputs.logits lowerCAmelCase : Optional[int] = logits[:, -1, :] / (temperature if temperature > 0 else 1.0) lowerCAmelCase : Optional[Any] = logits.softmax(-1 ).log() if scores is None: lowerCAmelCase , lowerCAmelCase : List[str] = logits.topk(UpperCAmelCase_ , -1 ) lowerCAmelCase : Union[str, Any] = generated.expand(UpperCAmelCase_ , *generated.shape[1:] ) lowerCAmelCase , lowerCAmelCase : str = next_tokens.permute(1 , 0 ), scores.squeeze(0 ) if tokens is None: lowerCAmelCase : int = next_tokens else: lowerCAmelCase : Optional[int] = tokens.expand(UpperCAmelCase_ , *tokens.shape[1:] ) lowerCAmelCase : Any = torch.cat((tokens, next_tokens) , dim=1 ) else: lowerCAmelCase : str = -float(np.inf ) lowerCAmelCase : List[str] = 0 lowerCAmelCase : Optional[int] = scores[:, None] + logits seq_lengths[~is_stopped] += 1 lowerCAmelCase : Optional[int] = scores_sum / seq_lengths[:, None] lowerCAmelCase , lowerCAmelCase : str = scores_sum_average.view(-1 ).topk(UpperCAmelCase_ , -1 ) lowerCAmelCase : List[Any] = next_tokens // scores_sum.shape[1] lowerCAmelCase : List[str] = seq_lengths[next_tokens_source] lowerCAmelCase : List[str] = next_tokens % scores_sum.shape[1] lowerCAmelCase : Optional[Any] = next_tokens.unsqueeze(1 ) lowerCAmelCase : int = tokens[next_tokens_source] lowerCAmelCase : Dict = torch.cat((tokens, next_tokens) , dim=1 ) lowerCAmelCase : List[str] = generated[next_tokens_source] lowerCAmelCase : Optional[int] = scores_sum_average * seq_lengths lowerCAmelCase : Union[str, Any] = is_stopped[next_tokens_source] lowerCAmelCase : int = self.transformer.transformer.wte(next_tokens.squeeze() ).view(generated.shape[0] , 1 , -1 ) lowerCAmelCase : Any = torch.cat((generated, next_token_embed) , dim=1 ) lowerCAmelCase : List[Any] = is_stopped + next_tokens.eq(UpperCAmelCase_ ).squeeze() if is_stopped.all(): break lowerCAmelCase : Tuple = scores / seq_lengths lowerCAmelCase : List[str] = scores.argsort(descending=UpperCAmelCase_ ) # tokens tensors are already padded to max_seq_length lowerCAmelCase : Tuple = [tokens[i] for i in order] lowerCAmelCase : str = torch.stack(UpperCAmelCase_ , dim=0 ) lowerCAmelCase : Union[str, Any] = torch.tensor([seq_lengths[i] for i in order] , dtype=seq_lengths.dtype ) return output_texts, seq_lengths
323
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 MobileNetVaImageProcessor class __A ( unittest.TestCase ): def __init__( self : List[str] , UpperCAmelCase_ : int , UpperCAmelCase_ : Optional[int]=7 , UpperCAmelCase_ : Tuple=3 , UpperCAmelCase_ : int=18 , UpperCAmelCase_ : List[str]=30 , UpperCAmelCase_ : str=400 , UpperCAmelCase_ : Union[str, Any]=True , UpperCAmelCase_ : Any=None , UpperCAmelCase_ : List[str]=True , UpperCAmelCase_ : Union[str, Any]=None , ): lowerCAmelCase : Any = size if size is not None else {'shortest_edge': 20} lowerCAmelCase : str = crop_size if crop_size is not None else {'height': 18, 'width': 18} lowerCAmelCase : List[Any] = parent lowerCAmelCase : Optional[Any] = batch_size lowerCAmelCase : int = num_channels lowerCAmelCase : int = image_size lowerCAmelCase : Tuple = min_resolution lowerCAmelCase : Any = max_resolution lowerCAmelCase : int = do_resize lowerCAmelCase : Dict = size lowerCAmelCase : int = do_center_crop lowerCAmelCase : str = crop_size def lowercase__ ( self : Optional[int] ): return { "do_resize": self.do_resize, "size": self.size, "do_center_crop": self.do_center_crop, "crop_size": self.crop_size, } @require_torch @require_vision class __A ( lowerCAmelCase , unittest.TestCase ): lowerCAmelCase_ : Optional[Any] = MobileNetVaImageProcessor if is_vision_available() else None def lowercase__ ( self : int ): lowerCAmelCase : List[str] = MobileNetVaImageProcessingTester(self ) @property def lowercase__ ( self : int ): return self.image_processor_tester.prepare_image_processor_dict() def lowercase__ ( self : Dict ): lowerCAmelCase : Any = self.image_processing_class(**self.image_processor_dict ) self.assertTrue(hasattr(UpperCAmelCase_ , 'do_resize' ) ) self.assertTrue(hasattr(UpperCAmelCase_ , 'size' ) ) self.assertTrue(hasattr(UpperCAmelCase_ , 'do_center_crop' ) ) self.assertTrue(hasattr(UpperCAmelCase_ , 'crop_size' ) ) def lowercase__ ( self : int ): lowerCAmelCase : Union[str, Any] = self.image_processing_class.from_dict(self.image_processor_dict ) self.assertEqual(image_processor.size , {'shortest_edge': 20} ) self.assertEqual(image_processor.crop_size , {'height': 18, 'width': 18} ) lowerCAmelCase : List[str] = self.image_processing_class.from_dict(self.image_processor_dict , size=42 , crop_size=84 ) self.assertEqual(image_processor.size , {'shortest_edge': 42} ) self.assertEqual(image_processor.crop_size , {'height': 84, 'width': 84} ) def lowercase__ ( self : str ): pass def lowercase__ ( self : List[str] ): # Initialize image_processing lowerCAmelCase : List[str] = self.image_processing_class(**self.image_processor_dict ) # create random PIL images lowerCAmelCase : Any = prepare_image_inputs(self.image_processor_tester , equal_resolution=UpperCAmelCase_ ) for image in image_inputs: self.assertIsInstance(UpperCAmelCase_ , Image.Image ) # Test not batched input lowerCAmelCase : str = 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.crop_size['height'], self.image_processor_tester.crop_size['width'], ) , ) # Test batched lowerCAmelCase : Dict = image_processing(UpperCAmelCase_ , 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.crop_size['height'], self.image_processor_tester.crop_size['width'], ) , ) def lowercase__ ( self : Optional[Any] ): # Initialize image_processing lowerCAmelCase : Any = self.image_processing_class(**self.image_processor_dict ) # create random numpy tensors lowerCAmelCase : Optional[int] = prepare_image_inputs(self.image_processor_tester , equal_resolution=UpperCAmelCase_ , numpify=UpperCAmelCase_ ) for image in image_inputs: self.assertIsInstance(UpperCAmelCase_ , np.ndarray ) # Test not batched input lowerCAmelCase : 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.crop_size['height'], self.image_processor_tester.crop_size['width'], ) , ) # Test batched lowerCAmelCase : Optional[int] = image_processing(UpperCAmelCase_ , 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.crop_size['height'], self.image_processor_tester.crop_size['width'], ) , ) def lowercase__ ( self : Dict ): # Initialize image_processing lowerCAmelCase : Optional[int] = self.image_processing_class(**self.image_processor_dict ) # create random PyTorch tensors lowerCAmelCase : str = prepare_image_inputs(self.image_processor_tester , equal_resolution=UpperCAmelCase_ , torchify=UpperCAmelCase_ ) for image in image_inputs: self.assertIsInstance(UpperCAmelCase_ , torch.Tensor ) # Test not batched input lowerCAmelCase : 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.crop_size['height'], self.image_processor_tester.crop_size['width'], ) , ) # Test batched lowerCAmelCase : List[str] = image_processing(UpperCAmelCase_ , 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.crop_size['height'], self.image_processor_tester.crop_size['width'], ) , )
323
1
from typing import TYPE_CHECKING from ...utils import OptionalDependencyNotAvailable, _LazyModule, is_tf_available, is_torch_available __lowerCAmelCase : Any = { 'configuration_data2vec_audio': ['DATA2VEC_AUDIO_PRETRAINED_CONFIG_ARCHIVE_MAP', 'Data2VecAudioConfig'], 'configuration_data2vec_text': [ 'DATA2VEC_TEXT_PRETRAINED_CONFIG_ARCHIVE_MAP', 'Data2VecTextConfig', 'Data2VecTextOnnxConfig', ], 'configuration_data2vec_vision': [ 'DATA2VEC_VISION_PRETRAINED_CONFIG_ARCHIVE_MAP', 'Data2VecVisionConfig', 'Data2VecVisionOnnxConfig', ], } try: if not is_torch_available(): raise OptionalDependencyNotAvailable() except OptionalDependencyNotAvailable: pass else: __lowerCAmelCase : str = [ 'DATA2VEC_AUDIO_PRETRAINED_MODEL_ARCHIVE_LIST', 'Data2VecAudioForAudioFrameClassification', 'Data2VecAudioForCTC', 'Data2VecAudioForSequenceClassification', 'Data2VecAudioForXVector', 'Data2VecAudioModel', 'Data2VecAudioPreTrainedModel', ] __lowerCAmelCase : Dict = [ 'DATA2VEC_TEXT_PRETRAINED_MODEL_ARCHIVE_LIST', 'Data2VecTextForCausalLM', 'Data2VecTextForMaskedLM', 'Data2VecTextForMultipleChoice', 'Data2VecTextForQuestionAnswering', 'Data2VecTextForSequenceClassification', 'Data2VecTextForTokenClassification', 'Data2VecTextModel', 'Data2VecTextPreTrainedModel', ] __lowerCAmelCase : Dict = [ 'DATA2VEC_VISION_PRETRAINED_MODEL_ARCHIVE_LIST', 'Data2VecVisionForImageClassification', 'Data2VecVisionForMaskedImageModeling', 'Data2VecVisionForSemanticSegmentation', 'Data2VecVisionModel', 'Data2VecVisionPreTrainedModel', ] if is_tf_available(): __lowerCAmelCase : Optional[Any] = [ 'TFData2VecVisionForImageClassification', 'TFData2VecVisionForSemanticSegmentation', 'TFData2VecVisionModel', 'TFData2VecVisionPreTrainedModel', ] if TYPE_CHECKING: from .configuration_dataavec_audio import DATA2VEC_AUDIO_PRETRAINED_CONFIG_ARCHIVE_MAP, DataaVecAudioConfig from .configuration_dataavec_text import ( DATA2VEC_TEXT_PRETRAINED_CONFIG_ARCHIVE_MAP, DataaVecTextConfig, DataaVecTextOnnxConfig, ) from .configuration_dataavec_vision import ( DATA2VEC_VISION_PRETRAINED_CONFIG_ARCHIVE_MAP, DataaVecVisionConfig, DataaVecVisionOnnxConfig, ) try: if not is_torch_available(): raise OptionalDependencyNotAvailable() except OptionalDependencyNotAvailable: pass else: from .modeling_dataavec_audio import ( DATA2VEC_AUDIO_PRETRAINED_MODEL_ARCHIVE_LIST, DataaVecAudioForAudioFrameClassification, DataaVecAudioForCTC, DataaVecAudioForSequenceClassification, DataaVecAudioForXVector, DataaVecAudioModel, DataaVecAudioPreTrainedModel, ) from .modeling_dataavec_text import ( DATA2VEC_TEXT_PRETRAINED_MODEL_ARCHIVE_LIST, DataaVecTextForCausalLM, DataaVecTextForMaskedLM, DataaVecTextForMultipleChoice, DataaVecTextForQuestionAnswering, DataaVecTextForSequenceClassification, DataaVecTextForTokenClassification, DataaVecTextModel, DataaVecTextPreTrainedModel, ) from .modeling_dataavec_vision import ( DATA2VEC_VISION_PRETRAINED_MODEL_ARCHIVE_LIST, DataaVecVisionForImageClassification, DataaVecVisionForMaskedImageModeling, DataaVecVisionForSemanticSegmentation, DataaVecVisionModel, DataaVecVisionPreTrainedModel, ) if is_tf_available(): from .modeling_tf_dataavec_vision import ( TFDataaVecVisionForImageClassification, TFDataaVecVisionForSemanticSegmentation, TFDataaVecVisionModel, TFDataaVecVisionPreTrainedModel, ) else: import sys __lowerCAmelCase : Dict = _LazyModule(__name__, globals()['__file__'], _import_structure, module_spec=__spec__)
107
'''simple docstring''' from collections import OrderedDict from typing import Mapping from ...configuration_utils import PretrainedConfig from ...onnx import OnnxConfig from ...utils import logging __snake_case = logging.get_logger(__name__) __snake_case = { '''camembert-base''': '''https://huggingface.co/camembert-base/resolve/main/config.json''', '''umberto-commoncrawl-cased-v1''': ( '''https://huggingface.co/Musixmatch/umberto-commoncrawl-cased-v1/resolve/main/config.json''' ), '''umberto-wikipedia-uncased-v1''': ( '''https://huggingface.co/Musixmatch/umberto-wikipedia-uncased-v1/resolve/main/config.json''' ), } class lowercase ( A__ ): """simple docstring""" _a = 'camembert' def __init__( self , UpperCamelCase_=30522 , UpperCamelCase_=768 , UpperCamelCase_=12 , UpperCamelCase_=12 , UpperCamelCase_=3072 , UpperCamelCase_="gelu" , UpperCamelCase_=0.1 , UpperCamelCase_=0.1 , UpperCamelCase_=512 , UpperCamelCase_=2 , UpperCamelCase_=0.02 , UpperCamelCase_=1e-12 , UpperCamelCase_=1 , UpperCamelCase_=0 , UpperCamelCase_=2 , UpperCamelCase_="absolute" , UpperCamelCase_=True , UpperCamelCase_=None , **UpperCamelCase_ , ): '''simple docstring''' super().__init__(pad_token_id=UpperCamelCase_ , bos_token_id=UpperCamelCase_ , eos_token_id=UpperCamelCase_ , **UpperCamelCase_ ) UpperCamelCase__ :int = vocab_size UpperCamelCase__ :Optional[int] = hidden_size UpperCamelCase__ :Optional[int] = num_hidden_layers UpperCamelCase__ :List[Any] = num_attention_heads UpperCamelCase__ :Union[str, Any] = hidden_act UpperCamelCase__ :List[Any] = intermediate_size UpperCamelCase__ :int = hidden_dropout_prob UpperCamelCase__ :Tuple = attention_probs_dropout_prob UpperCamelCase__ :Union[str, Any] = max_position_embeddings UpperCamelCase__ :Tuple = type_vocab_size UpperCamelCase__ :int = initializer_range UpperCamelCase__ :List[str] = layer_norm_eps UpperCamelCase__ :int = position_embedding_type UpperCamelCase__ :Any = use_cache UpperCamelCase__ :Any = classifier_dropout class lowercase ( A__ ): """simple docstring""" @property def lowerCAmelCase__ ( self ): '''simple docstring''' if self.task == "multiple-choice": UpperCamelCase__ :List[str] = {0: '''batch''', 1: '''choice''', 2: '''sequence'''} else: UpperCamelCase__ :Tuple = {0: '''batch''', 1: '''sequence'''} return OrderedDict( [ ('''input_ids''', dynamic_axis), ('''attention_mask''', dynamic_axis), ] )
97
0
"""simple docstring""" import unittest from transformers import is_torch_available from transformers.testing_utils import require_torch if is_torch_available(): import torch from transformers.generation import DisjunctiveConstraint @require_torch class _UpperCAmelCase ( unittest.TestCase ): '''simple docstring''' def __lowerCAmelCase ( self ) -> Dict: # For consistency across different places the DisjunctiveConstraint is called, # dc.token_ids is a list of integers. It is also initialized only by integers. _UpperCAmelCase : List[str] = [[1, 2, 4], [1, 2, 3, 4]] _UpperCAmelCase : List[str] = DisjunctiveConstraint(A ) self.assertTrue(isinstance(dc.token_ids , A ) ) with self.assertRaises(A ): DisjunctiveConstraint(torch.LongTensor([[1, 2, 4], [1, 2, 3]] ) ) with self.assertRaises(A ): DisjunctiveConstraint([torch.LongTensor([1, 2, 4] ), torch.LongTensor([1, 2, 3, 4, 5] )] ) def __lowerCAmelCase ( self ) -> List[Any]: # We can't have constraints that are complete subsets of another. This leads to a preverse # interpretation of "constraint fulfillment": does generating [1,2,3] fulfill the constraint? # It would mean that it generated [1,2] which fulfills it, but it's in the middle of potentially # fulfilling [1,2,3,4]. If we believe that [1,2,3] does fulfill the constraint, then the algorithm # will necessarily never reach [1,2,3,4], giving users a false sense of control (better to just not allow it). _UpperCAmelCase : Optional[int] = [[1, 2], [1, 2, 3, 4]] with self.assertRaises(A ): DisjunctiveConstraint(A ) # fails here def __lowerCAmelCase ( self ) -> Tuple: _UpperCAmelCase : Optional[Any] = [[1, 2, 3], [1, 2, 4]] _UpperCAmelCase : Optional[int] = DisjunctiveConstraint(A ) _UpperCAmelCase , _UpperCAmelCase , _UpperCAmelCase : Tuple = dc.update(1 ) _UpperCAmelCase : Dict = stepped is True and completed is False and reset is False self.assertTrue(A ) self.assertTrue(not dc.completed ) self.assertTrue(dc.current_seq == [1] ) _UpperCAmelCase , _UpperCAmelCase , _UpperCAmelCase : Optional[int] = dc.update(2 ) _UpperCAmelCase : Any = stepped is True and completed is False and reset is False self.assertTrue(A ) self.assertTrue(not dc.completed ) self.assertTrue(dc.current_seq == [1, 2] ) _UpperCAmelCase , _UpperCAmelCase , _UpperCAmelCase : Any = dc.update(3 ) _UpperCAmelCase : Optional[int] = stepped is True and completed is True and reset is False self.assertTrue(A ) self.assertTrue(dc.completed ) # Completed! self.assertTrue(dc.current_seq == [1, 2, 3] ) def __lowerCAmelCase ( self ) -> Optional[int]: _UpperCAmelCase : Tuple = [[1, 2, 3], [1, 2, 4, 5], [1, 2, 5]] _UpperCAmelCase : Any = DisjunctiveConstraint(A ) _UpperCAmelCase , _UpperCAmelCase , _UpperCAmelCase : Optional[int] = dc.update(1 ) self.assertTrue(not dc.completed ) self.assertTrue(dc.current_seq == [1] ) _UpperCAmelCase , _UpperCAmelCase , _UpperCAmelCase : Optional[int] = dc.update(2 ) self.assertTrue(not dc.completed ) self.assertTrue(dc.current_seq == [1, 2] ) _UpperCAmelCase , _UpperCAmelCase , _UpperCAmelCase : Any = dc.update(4 ) self.assertTrue(not dc.completed ) self.assertTrue(dc.current_seq == [1, 2, 4] ) _UpperCAmelCase , _UpperCAmelCase , _UpperCAmelCase : Optional[int] = dc.update(5 ) self.assertTrue(dc.completed ) # Completed! self.assertTrue(dc.current_seq == [1, 2, 4, 5] ) dc.reset() _UpperCAmelCase , _UpperCAmelCase , _UpperCAmelCase : Union[str, Any] = dc.update(1 ) self.assertTrue(not dc.completed ) self.assertTrue(dc.remaining() == 3 ) self.assertTrue(dc.current_seq == [1] ) _UpperCAmelCase , _UpperCAmelCase , _UpperCAmelCase : List[Any] = dc.update(2 ) self.assertTrue(not dc.completed ) self.assertTrue(dc.remaining() == 2 ) self.assertTrue(dc.current_seq == [1, 2] ) _UpperCAmelCase , _UpperCAmelCase , _UpperCAmelCase : Optional[int] = dc.update(5 ) self.assertTrue(dc.completed ) # Completed! self.assertTrue(dc.remaining() == 0 ) self.assertTrue(dc.current_seq == [1, 2, 5] )
68
"""simple docstring""" import os import time import warnings from dataclasses import dataclass, field from enum import Enum from typing import List, Optional, Union import torch from filelock import FileLock from torch.utils.data import Dataset from ...tokenization_utils_base import PreTrainedTokenizerBase from ...utils import logging from ..processors.glue import glue_convert_examples_to_features, glue_output_modes, glue_processors from ..processors.utils import InputFeatures _lowerCAmelCase :str = logging.get_logger(__name__) @dataclass class _UpperCAmelCase : '''simple docstring''' a__ =field(metadata={'''help''': '''The name of the task to train on: ''' + ''', '''.join(glue_processors.keys() )} ) a__ =field( metadata={'''help''': '''The input data dir. Should contain the .tsv files (or other data files) for the task.'''} ) a__ =field( default=1_2_8 ,metadata={ '''help''': ( '''The maximum total input sequence length after tokenization. Sequences longer ''' '''than this will be truncated, sequences shorter will be padded.''' ) } ,) a__ =field( default=a ,metadata={'''help''': '''Overwrite the cached training and evaluation sets'''} ) def __lowerCAmelCase ( self ) -> int: _UpperCAmelCase : str = self.task_name.lower() class _UpperCAmelCase ( a ): '''simple docstring''' a__ ='''train''' a__ ='''dev''' a__ ='''test''' class _UpperCAmelCase ( a ): '''simple docstring''' a__ =42 a__ =42 a__ =42 def __init__( self , A , A , A = None , A = Split.train , A = None , ) -> Dict: warnings.warn( '''This dataset will be removed from the library soon, preprocessing should be handled with the 🤗 Datasets ''' '''library. You can have a look at this example script for pointers: ''' '''https://github.com/huggingface/transformers/blob/main/examples/pytorch/text-classification/run_glue.py''' , A , ) _UpperCAmelCase : Dict = args _UpperCAmelCase : int = glue_processors[args.task_name]() _UpperCAmelCase : Any = glue_output_modes[args.task_name] if isinstance(A , A ): try: _UpperCAmelCase : int = Split[mode] except KeyError: raise KeyError('''mode is not a valid split name''' ) # Load data features from cache or dataset file _UpperCAmelCase : Optional[int] = os.path.join( cache_dir if cache_dir is not None else args.data_dir , f'cached_{mode.value}_{tokenizer.__class__.__name__}_{args.max_seq_length}_{args.task_name}' , ) _UpperCAmelCase : List[Any] = self.processor.get_labels() if args.task_name in ["mnli", "mnli-mm"] and tokenizer.__class__.__name__ in ( "RobertaTokenizer", "RobertaTokenizerFast", "XLMRobertaTokenizer", "BartTokenizer", "BartTokenizerFast", ): # HACK(label indices are swapped in RoBERTa pretrained model) _UpperCAmelCase , _UpperCAmelCase : Optional[Any] = label_list[2], label_list[1] _UpperCAmelCase : Optional[Any] = label_list # Make sure only the first process in distributed training processes the dataset, # and the others will use the cache. _UpperCAmelCase : List[Any] = cached_features_file + '''.lock''' with FileLock(A ): if os.path.exists(A ) and not args.overwrite_cache: _UpperCAmelCase : str = time.time() _UpperCAmelCase : Dict = torch.load(A ) logger.info( f'Loading features from cached file {cached_features_file} [took %.3f s]' , time.time() - start ) else: logger.info(f'Creating features from dataset file at {args.data_dir}' ) if mode == Split.dev: _UpperCAmelCase : Union[str, Any] = self.processor.get_dev_examples(args.data_dir ) elif mode == Split.test: _UpperCAmelCase : List[Any] = self.processor.get_test_examples(args.data_dir ) else: _UpperCAmelCase : Optional[Any] = self.processor.get_train_examples(args.data_dir ) if limit_length is not None: _UpperCAmelCase : Tuple = examples[:limit_length] _UpperCAmelCase : str = glue_convert_examples_to_features( A , A , max_length=args.max_seq_length , label_list=A , output_mode=self.output_mode , ) _UpperCAmelCase : Optional[int] = time.time() torch.save(self.features , A ) # ^ This seems to take a lot of time so I want to investigate why and how we can improve. logger.info( f'Saving features into cached file {cached_features_file} [took {time.time() - start:.3f} s]' ) def __len__( self ) -> Union[str, Any]: return len(self.features ) def __getitem__( self , A ) -> InputFeatures: return self.features[i] def __lowerCAmelCase ( self ) -> List[Any]: return self.label_list
68
1
"""simple docstring""" # 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 UpperCAmelCase__ = 'facebook/wmt19-en-de' UpperCAmelCase__ = FSMTTokenizer.from_pretrained(mname) # get the correct vocab sizes, etc. from the master model UpperCAmelCase__ = 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, ) ) UpperCAmelCase__ = FSMTForConditionalGeneration(config) print(F"num of params {tiny_model.num_parameters()}") # Test UpperCAmelCase__ = tokenizer(['Making tiny model'], return_tensors='pt') UpperCAmelCase__ = tiny_model(**batch) print('test output:', len(outputs.logits[0])) # Save UpperCAmelCase__ = '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
288
import pprint import requests UpperCAmelCase = '''https://zenquotes.io/api''' def UpperCAmelCase_ ( ): return requests.get(API_ENDPOINT_URL + '/today' ).json() def UpperCAmelCase_ ( ): return requests.get(API_ENDPOINT_URL + '/random' ).json() if __name__ == "__main__": UpperCAmelCase = random_quotes() pprint.pprint(response)
195
0
import argparse import logging import os from pathlib import Path from typing import Any, Dict import pytorch_lightning as pl from pytorch_lightning.utilities import rank_zero_info from transformers import ( AdamW, AutoConfig, AutoModel, AutoModelForPreTraining, AutoModelForQuestionAnswering, AutoModelForSeqaSeqLM, AutoModelForSequenceClassification, AutoModelForTokenClassification, AutoModelWithLMHead, AutoTokenizer, PretrainedConfig, PreTrainedTokenizer, ) from transformers.optimization import ( Adafactor, get_cosine_schedule_with_warmup, get_cosine_with_hard_restarts_schedule_with_warmup, get_linear_schedule_with_warmup, get_polynomial_decay_schedule_with_warmup, ) from transformers.utils.versions import require_version _SCREAMING_SNAKE_CASE : int = logging.getLogger(__name__) require_version('''pytorch_lightning>=1.0.4''') _SCREAMING_SNAKE_CASE : List[Any] = { 'base': AutoModel, 'sequence-classification': AutoModelForSequenceClassification, 'question-answering': AutoModelForQuestionAnswering, 'pretraining': AutoModelForPreTraining, 'token-classification': AutoModelForTokenClassification, 'language-modeling': AutoModelWithLMHead, 'summarization': AutoModelForSeqaSeqLM, 'translation': AutoModelForSeqaSeqLM, } # update this and the import above to support new schedulers from transformers.optimization _SCREAMING_SNAKE_CASE : int = { 'linear': get_linear_schedule_with_warmup, 'cosine': get_cosine_schedule_with_warmup, 'cosine_w_restarts': get_cosine_with_hard_restarts_schedule_with_warmup, 'polynomial': get_polynomial_decay_schedule_with_warmup, # '': get_constant_schedule, # not supported for now # '': get_constant_schedule_with_warmup, # not supported for now } _SCREAMING_SNAKE_CASE : Any = sorted(arg_to_scheduler.keys()) _SCREAMING_SNAKE_CASE : Optional[Any] = '{' + ', '.join(arg_to_scheduler_choices) + '}' class UpperCAmelCase__ ( pl.LightningModule ): """simple docstring""" def __init__( self : List[str] , __lowerCamelCase : Tuple , __lowerCamelCase : Optional[int]=None , __lowerCamelCase : int="base" , __lowerCamelCase : Tuple=None , __lowerCamelCase : str=None , __lowerCamelCase : List[Any]=None , **__lowerCamelCase : str , ) -> List[Any]: super().__init__() # TODO: move to self.save_hyperparameters() # self.save_hyperparameters() # can also expand arguments into trainer signature for easier reading self.save_hyperparameters(SCREAMING_SNAKE_CASE_ ) SCREAMING_SNAKE_CASE__ = 0 SCREAMING_SNAKE_CASE__ = Path(self.hparams.output_dir ) SCREAMING_SNAKE_CASE__ = self.hparams.cache_dir if self.hparams.cache_dir else None if config is None: SCREAMING_SNAKE_CASE__ = AutoConfig.from_pretrained( self.hparams.config_name if self.hparams.config_name else self.hparams.model_name_or_path , **({'''num_labels''': num_labels} if num_labels is not None else {}) , cache_dir=SCREAMING_SNAKE_CASE_ , **SCREAMING_SNAKE_CASE_ , ) else: SCREAMING_SNAKE_CASE__ = config SCREAMING_SNAKE_CASE__ = ("""encoder_layerdrop""", """decoder_layerdrop""", """dropout""", """attention_dropout""") for p in extra_model_params: if getattr(self.hparams , SCREAMING_SNAKE_CASE_ , SCREAMING_SNAKE_CASE_ ): assert hasattr(self.config , SCREAMING_SNAKE_CASE_ ), f'''model config doesn\'t have a `{p}` attribute''' setattr(self.config , SCREAMING_SNAKE_CASE_ , getattr(self.hparams , SCREAMING_SNAKE_CASE_ ) ) if tokenizer is None: SCREAMING_SNAKE_CASE__ = AutoTokenizer.from_pretrained( self.hparams.tokenizer_name if self.hparams.tokenizer_name else self.hparams.model_name_or_path , cache_dir=SCREAMING_SNAKE_CASE_ , ) else: SCREAMING_SNAKE_CASE__ = tokenizer SCREAMING_SNAKE_CASE__ = MODEL_MODES[mode] if model is None: SCREAMING_SNAKE_CASE__ = self.model_type.from_pretrained( self.hparams.model_name_or_path , from_tf=bool('''.ckpt''' in self.hparams.model_name_or_path ) , config=self.config , cache_dir=SCREAMING_SNAKE_CASE_ , ) else: SCREAMING_SNAKE_CASE__ = model def lowercase_ ( self : Optional[Any] , *__lowerCamelCase : Tuple , **__lowerCamelCase : List[Any] ) -> int: SCREAMING_SNAKE_CASE__ = self.model_type.from_pretrained(*SCREAMING_SNAKE_CASE_ , **SCREAMING_SNAKE_CASE_ ) def lowercase_ ( self : str ) -> List[str]: SCREAMING_SNAKE_CASE__ = arg_to_scheduler[self.hparams.lr_scheduler] SCREAMING_SNAKE_CASE__ = get_schedule_func( self.opt , num_warmup_steps=self.hparams.warmup_steps , num_training_steps=self.total_steps() ) SCREAMING_SNAKE_CASE__ = {"""scheduler""": scheduler, """interval""": """step""", """frequency""": 1} return scheduler def lowercase_ ( self : Dict ) -> str: SCREAMING_SNAKE_CASE__ = self.model SCREAMING_SNAKE_CASE__ = ["""bias""", """LayerNorm.weight"""] SCREAMING_SNAKE_CASE__ = [ { """params""": [ p for n, p in model.named_parameters() if not any(nd in n for nd in no_decay ) ], # check this named paramters """weight_decay""": self.hparams.weight_decay, }, { """params""": [p for n, p in model.named_parameters() if any(nd in n for nd in no_decay )], """weight_decay""": 0.0, }, ] if self.hparams.adafactor: SCREAMING_SNAKE_CASE__ = Adafactor( SCREAMING_SNAKE_CASE_ , lr=self.hparams.learning_rate , scale_parameter=SCREAMING_SNAKE_CASE_ , relative_step=SCREAMING_SNAKE_CASE_ ) else: SCREAMING_SNAKE_CASE__ = AdamW( SCREAMING_SNAKE_CASE_ , lr=self.hparams.learning_rate , eps=self.hparams.adam_epsilon ) SCREAMING_SNAKE_CASE__ = optimizer SCREAMING_SNAKE_CASE__ = self.get_lr_scheduler() return [optimizer], [scheduler] def lowercase_ ( self : str , __lowerCamelCase : int , __lowerCamelCase : List[str] ) -> Dict: return self.validation_step(SCREAMING_SNAKE_CASE_ , SCREAMING_SNAKE_CASE_ ) def lowercase_ ( self : List[Any] , __lowerCamelCase : Union[str, Any] ) -> Union[str, Any]: return self.validation_end(SCREAMING_SNAKE_CASE_ ) def lowercase_ ( self : List[Any] ) -> int: SCREAMING_SNAKE_CASE__ = max(1 , self.hparams.gpus ) # TODO: consider num_tpu_cores SCREAMING_SNAKE_CASE__ = self.hparams.train_batch_size * self.hparams.accumulate_grad_batches * num_devices return (self.dataset_size / effective_batch_size) * self.hparams.max_epochs def lowercase_ ( self : Dict , __lowerCamelCase : Tuple ) -> Any: if stage == "test": SCREAMING_SNAKE_CASE__ = len(self.test_dataloader().dataset ) else: SCREAMING_SNAKE_CASE__ = self.get_dataloader('''train''' , self.hparams.train_batch_size , shuffle=SCREAMING_SNAKE_CASE_ ) SCREAMING_SNAKE_CASE__ = len(self.train_dataloader().dataset ) def lowercase_ ( self : Optional[Any] , __lowerCamelCase : Optional[Any] , __lowerCamelCase : str , __lowerCamelCase : Optional[Any] = False ) -> str: raise NotImplementedError('''You must implement this for your task''' ) def lowercase_ ( self : Dict ) -> Union[str, Any]: return self.train_loader def lowercase_ ( self : Tuple ) -> Dict: return self.get_dataloader('''dev''' , self.hparams.eval_batch_size , shuffle=SCREAMING_SNAKE_CASE_ ) def lowercase_ ( self : Any ) -> List[str]: return self.get_dataloader('''test''' , self.hparams.eval_batch_size , shuffle=SCREAMING_SNAKE_CASE_ ) def lowercase_ ( self : Union[str, Any] , __lowerCamelCase : int ) -> Dict: return os.path.join( self.hparams.data_dir , '''cached_{}_{}_{}'''.format( SCREAMING_SNAKE_CASE_ , list(filter(SCREAMING_SNAKE_CASE_ , self.hparams.model_name_or_path.split('''/''' ) ) ).pop() , str(self.hparams.max_seq_length ) , ) , ) @pl.utilities.rank_zero_only def lowercase_ ( self : Union[str, Any] , __lowerCamelCase : Dict ) -> None: SCREAMING_SNAKE_CASE__ = self.output_dir.joinpath('''best_tfmr''' ) SCREAMING_SNAKE_CASE__ = self.step_count self.model.save_pretrained(SCREAMING_SNAKE_CASE_ ) self.tokenizer.save_pretrained(SCREAMING_SNAKE_CASE_ ) @staticmethod def lowercase_ ( __lowerCamelCase : Tuple , __lowerCamelCase : str ) -> int: parser.add_argument( '''--model_name_or_path''' , default=SCREAMING_SNAKE_CASE_ , type=SCREAMING_SNAKE_CASE_ , required=SCREAMING_SNAKE_CASE_ , help='''Path to pretrained model or model identifier from huggingface.co/models''' , ) parser.add_argument( '''--config_name''' , default='''''' , type=SCREAMING_SNAKE_CASE_ , help='''Pretrained config name or path if not the same as model_name''' ) parser.add_argument( '''--tokenizer_name''' , default=SCREAMING_SNAKE_CASE_ , type=SCREAMING_SNAKE_CASE_ , help='''Pretrained tokenizer name or path if not the same as model_name''' , ) parser.add_argument( '''--cache_dir''' , default=str(Path(SCREAMING_SNAKE_CASE_ ).parent / '''test_run''' / '''cache''' ) , type=SCREAMING_SNAKE_CASE_ , help='''Where do you want to store the pre-trained models downloaded from huggingface.co''' , ) parser.add_argument( '''--encoder_layerdrop''' , type=SCREAMING_SNAKE_CASE_ , help='''Encoder layer dropout probability (Optional). Goes into model.config''' , ) parser.add_argument( '''--decoder_layerdrop''' , type=SCREAMING_SNAKE_CASE_ , help='''Decoder layer dropout probability (Optional). Goes into model.config''' , ) parser.add_argument( '''--dropout''' , type=SCREAMING_SNAKE_CASE_ , help='''Dropout probability (Optional). Goes into model.config''' , ) parser.add_argument( '''--attention_dropout''' , type=SCREAMING_SNAKE_CASE_ , help='''Attention dropout probability (Optional). Goes into model.config''' , ) parser.add_argument('''--learning_rate''' , default=5e-5 , type=SCREAMING_SNAKE_CASE_ , help='''The initial learning rate for Adam.''' ) parser.add_argument( '''--lr_scheduler''' , default='''linear''' , choices=SCREAMING_SNAKE_CASE_ , metavar=SCREAMING_SNAKE_CASE_ , type=SCREAMING_SNAKE_CASE_ , help='''Learning rate scheduler''' , ) parser.add_argument('''--weight_decay''' , default=0.0 , type=SCREAMING_SNAKE_CASE_ , help='''Weight decay if we apply some.''' ) parser.add_argument('''--adam_epsilon''' , default=1e-8 , type=SCREAMING_SNAKE_CASE_ , help='''Epsilon for Adam optimizer.''' ) parser.add_argument('''--warmup_steps''' , default=0 , type=SCREAMING_SNAKE_CASE_ , help='''Linear warmup over warmup_steps.''' ) parser.add_argument('''--num_workers''' , default=4 , type=SCREAMING_SNAKE_CASE_ , help='''kwarg passed to DataLoader''' ) parser.add_argument('''--num_train_epochs''' , dest='''max_epochs''' , default=3 , type=SCREAMING_SNAKE_CASE_ ) parser.add_argument('''--train_batch_size''' , default=32 , type=SCREAMING_SNAKE_CASE_ ) parser.add_argument('''--eval_batch_size''' , default=32 , type=SCREAMING_SNAKE_CASE_ ) parser.add_argument('''--adafactor''' , action='''store_true''' ) class UpperCAmelCase__ ( pl.Callback ): """simple docstring""" def lowercase_ ( self : int , __lowerCamelCase : List[str] , __lowerCamelCase : Tuple ) -> Optional[Any]: if ( trainer.is_global_zero and trainer.global_rank == 0 ): # we initialize the retriever only on master worker with RAY. In new pytorch-lightning accelorators are removed. pl_module.model.rag.retriever.init_retrieval() # better to use hook functions. class UpperCAmelCase__ ( pl.Callback ): """simple docstring""" def lowercase_ ( self : Optional[int] , __lowerCamelCase : Tuple , __lowerCamelCase : Union[str, Any] ) -> Optional[int]: # print(pl_module.model.rag) for name, param in pl_module.model.rag.named_parameters(): if param.grad is None: print(SCREAMING_SNAKE_CASE_ ) class UpperCAmelCase__ ( pl.Callback ): """simple docstring""" def lowercase_ ( self : int , __lowerCamelCase : List[str] , __lowerCamelCase : Union[str, Any] ) -> Dict: SCREAMING_SNAKE_CASE__ = trainer.lr_schedulers[0]["""scheduler"""] SCREAMING_SNAKE_CASE__ = {f'''lr_group_{i}''': lr for i, lr in enumerate(lr_scheduler.get_lr() )} pl_module.logger.log_metrics(SCREAMING_SNAKE_CASE_ ) def lowercase_ ( self : Dict , __lowerCamelCase : Dict , __lowerCamelCase : Any ) -> Tuple: rank_zero_info('''***** Validation results *****''' ) SCREAMING_SNAKE_CASE__ = trainer.callback_metrics # Log results for key in sorted(SCREAMING_SNAKE_CASE_ ): if key not in ["log", "progress_bar"]: rank_zero_info('''{} = {}\n'''.format(SCREAMING_SNAKE_CASE_ , str(metrics[key] ) ) ) def lowercase_ ( self : List[str] , __lowerCamelCase : Optional[Any] , __lowerCamelCase : Optional[Any] ) -> Tuple: rank_zero_info('''***** Test results *****''' ) SCREAMING_SNAKE_CASE__ = trainer.callback_metrics # Log and save results to file SCREAMING_SNAKE_CASE__ = os.path.join(pl_module.hparams.output_dir , '''test_results.txt''' ) with open(SCREAMING_SNAKE_CASE_ , '''w''' ) as writer: for key in sorted(SCREAMING_SNAKE_CASE_ ): if key not in ["log", "progress_bar"]: rank_zero_info('''{} = {}\n'''.format(SCREAMING_SNAKE_CASE_ , str(metrics[key] ) ) ) writer.write('''{} = {}\n'''.format(SCREAMING_SNAKE_CASE_ , str(metrics[key] ) ) ) def UpperCAmelCase_ ( _A , _A ): '''simple docstring''' parser.add_argument( '''--output_dir''' , default=str(Path(lowerCAmelCase__ ).parent / '''test_run''' / '''model_checkpoints''' ) , type=lowerCAmelCase__ , help='''The output directory where the model predictions and checkpoints will be written.''' , ) parser.add_argument( '''--fp16''' , action='''store_true''' , help='''Whether to use 16-bit (mixed) precision (through NVIDIA apex) instead of 32-bit''' , ) parser.add_argument( '''--fp16_opt_level''' , type=lowerCAmelCase__ , default='''O2''' , help=( '''For fp16: Apex AMP optimization level selected in [\'O0\', \'O1\', \'O2\', and \'O3\'].''' '''See details at https://nvidia.github.io/apex/amp.html''' ) , ) parser.add_argument('''--n_tpu_cores''' , dest='''tpu_cores''' , type=lowerCAmelCase__ ) parser.add_argument('''--max_grad_norm''' , dest='''gradient_clip_val''' , default=1.0 , type=lowerCAmelCase__ , help='''Max gradient norm''' ) parser.add_argument('''--do_train''' , action='''store_true''' , help='''Whether to run training.''' ) parser.add_argument('''--do_predict''' , action='''store_true''' , help='''Whether to run predictions on the test set.''' ) parser.add_argument( '''--gradient_accumulation_steps''' , dest='''accumulate_grad_batches''' , type=lowerCAmelCase__ , default=1 , help='''Number of updates steps to accumulate before performing a backward/update pass.''' , ) parser.add_argument('''--seed''' , type=lowerCAmelCase__ , default=42 , help='''random seed for initialization''' ) parser.add_argument( '''--data_dir''' , default=str(Path(lowerCAmelCase__ ).parent / '''test_run''' / '''dummy-train-data''' ) , type=lowerCAmelCase__ , help='''The input data dir. Should contain the training files for the CoNLL-2003 NER task.''' , ) def UpperCAmelCase_ ( _A , _A , _A=None , _A=True , _A=[] , _A=None , _A=None , **_A , ): '''simple docstring''' pl.seed_everything(args.seed ) # init model SCREAMING_SNAKE_CASE__ = Path(model.hparams.output_dir ) odir.mkdir(exist_ok=lowerCAmelCase__ ) # add custom checkpoints if checkpoint_callback is None: SCREAMING_SNAKE_CASE__ = pl.callbacks.ModelCheckpoint( filepath=args.output_dir , prefix='''checkpoint''' , monitor='''val_loss''' , mode='''min''' , save_top_k=1 ) if early_stopping_callback: extra_callbacks.append(lowerCAmelCase__ ) if logging_callback is None: SCREAMING_SNAKE_CASE__ = LoggingCallback() SCREAMING_SNAKE_CASE__ = {} if args.fpaa: SCREAMING_SNAKE_CASE__ = 16 if args.gpus > 1: SCREAMING_SNAKE_CASE__ = """auto""" SCREAMING_SNAKE_CASE__ = """ddp""" SCREAMING_SNAKE_CASE__ = args.accumulate_grad_batches SCREAMING_SNAKE_CASE__ = None SCREAMING_SNAKE_CASE__ = """auto""" SCREAMING_SNAKE_CASE__ = pl.Trainer.from_argparse_args( lowerCAmelCase__ , weights_summary=lowerCAmelCase__ , callbacks=[logging_callback] + extra_callbacks + [InitCallback()] + [checkpoint_callback] , logger=lowerCAmelCase__ , val_check_interval=1 , num_sanity_val_steps=2 , **lowerCAmelCase__ , ) if args.do_train: trainer.fit(lowerCAmelCase__ ) else: print('''RAG modeling tests with new set functions successfuly executed!''' ) return trainer
366
from ...configuration_utils import PretrainedConfig from ...utils import logging _SCREAMING_SNAKE_CASE : List[str] = logging.get_logger(__name__) _SCREAMING_SNAKE_CASE : Tuple = { '''transfo-xl-wt103''': '''https://huggingface.co/transfo-xl-wt103/resolve/main/config.json''', } class UpperCAmelCase__ ( A__ ): """simple docstring""" a = "transfo-xl" a = ["mems"] a = { "n_token": "vocab_size", "hidden_size": "d_model", "num_attention_heads": "n_head", "num_hidden_layers": "n_layer", } def __init__( self : Any , __lowerCamelCase : int=26_7735 , __lowerCamelCase : Any=[2_0000, 4_0000, 20_0000] , __lowerCamelCase : Dict=1024 , __lowerCamelCase : Optional[int]=1024 , __lowerCamelCase : List[Any]=16 , __lowerCamelCase : Union[str, Any]=64 , __lowerCamelCase : Dict=4096 , __lowerCamelCase : int=4 , __lowerCamelCase : Dict=False , __lowerCamelCase : Tuple=18 , __lowerCamelCase : Optional[int]=1600 , __lowerCamelCase : str=1000 , __lowerCamelCase : Any=True , __lowerCamelCase : Dict=True , __lowerCamelCase : List[Any]=0 , __lowerCamelCase : int=-1 , __lowerCamelCase : Tuple=True , __lowerCamelCase : List[Any]=0.1 , __lowerCamelCase : int=0.0 , __lowerCamelCase : int=True , __lowerCamelCase : str="normal" , __lowerCamelCase : List[str]=0.01 , __lowerCamelCase : Any=0.01 , __lowerCamelCase : List[str]=0.02 , __lowerCamelCase : List[str]=1e-5 , __lowerCamelCase : Union[str, Any]=0 , **__lowerCamelCase : int , ) -> Optional[Any]: SCREAMING_SNAKE_CASE__ = vocab_size SCREAMING_SNAKE_CASE__ = [] self.cutoffs.extend(__lowerCamelCase ) if proj_share_all_but_first: SCREAMING_SNAKE_CASE__ = [False] + [True] * len(self.cutoffs ) else: SCREAMING_SNAKE_CASE__ = [False] + [False] * len(self.cutoffs ) SCREAMING_SNAKE_CASE__ = d_model SCREAMING_SNAKE_CASE__ = d_embed SCREAMING_SNAKE_CASE__ = d_head SCREAMING_SNAKE_CASE__ = d_inner SCREAMING_SNAKE_CASE__ = div_val SCREAMING_SNAKE_CASE__ = pre_lnorm SCREAMING_SNAKE_CASE__ = n_layer SCREAMING_SNAKE_CASE__ = n_head SCREAMING_SNAKE_CASE__ = mem_len SCREAMING_SNAKE_CASE__ = same_length SCREAMING_SNAKE_CASE__ = attn_type SCREAMING_SNAKE_CASE__ = clamp_len SCREAMING_SNAKE_CASE__ = sample_softmax SCREAMING_SNAKE_CASE__ = adaptive SCREAMING_SNAKE_CASE__ = dropout SCREAMING_SNAKE_CASE__ = dropatt SCREAMING_SNAKE_CASE__ = untie_r SCREAMING_SNAKE_CASE__ = init SCREAMING_SNAKE_CASE__ = init_range SCREAMING_SNAKE_CASE__ = proj_init_std SCREAMING_SNAKE_CASE__ = init_std SCREAMING_SNAKE_CASE__ = layer_norm_epsilon super().__init__(eos_token_id=__lowerCamelCase , **__lowerCamelCase ) @property def lowercase_ ( self : str ) -> Dict: # Message copied from Transformer-XL documentation logger.info(f'''The model {self.model_type} is one of the few models that has no sequence length limit.''' ) return -1 @max_position_embeddings.setter def lowercase_ ( self : List[str] , __lowerCamelCase : Any ) -> List[Any]: # Message copied from Transformer-XL documentation raise NotImplementedError( f'''The model {self.model_type} is one of the few models that has no sequence length limit.''' )
218
0
from math import isclose, sqrt def a_ ( _A , _A , _A ) -> tuple[float, float, float]: """simple docstring""" snake_case__ = point_y / 4 / point_x snake_case__ = 2 * normal_gradient / (1 + normal_gradient * normal_gradient) snake_case__ = (1 - normal_gradient * normal_gradient) / ( 1 + normal_gradient * normal_gradient ) snake_case__ = (sa - ca * incoming_gradient) / (ca + sa * incoming_gradient) # to find the next point, solve the simultaeneous equations: # y^2 + 4x^2 = 100 # y - b = m * (x - a) # ==> A x^2 + B x + C = 0 snake_case__ = outgoing_gradient**2 + 4 snake_case__ = 2 * outgoing_gradient * (point_y - outgoing_gradient * point_x) snake_case__ = (point_y - outgoing_gradient * point_x) ** 2 - 100 snake_case__ = ( -linear_term - sqrt(linear_term**2 - 4 * quadratic_term * constant_term ) ) / (2 * quadratic_term) snake_case__ = ( -linear_term + sqrt(linear_term**2 - 4 * quadratic_term * constant_term ) ) / (2 * quadratic_term) # two solutions, one of which is our input point snake_case__ = x_minus if isclose(_A , _A ) else x_plus snake_case__ = point_y + outgoing_gradient * (next_x - point_x) return next_x, next_y, outgoing_gradient def a_ ( _A = 1.4 , _A = -9.6 ) -> int: """simple docstring""" snake_case__ = 0 snake_case__ = first_x_coord snake_case__ = first_y_coord snake_case__ = (10.1 - point_y) / (0.0 - point_x) while not (-0.01 <= point_x <= 0.01 and point_y > 0): snake_case__ , snake_case__ , snake_case__ = next_point(_A , _A , _A ) num_reflections += 1 return num_reflections if __name__ == "__main__": print(f'''{solution() = }''')
307
# Lint as: python3 import sys from collections.abc import Mapping from typing import TYPE_CHECKING import numpy as np import pyarrow as pa from .. import config from ..utils.py_utils import map_nested from .formatting import TensorFormatter if TYPE_CHECKING: import torch class __SCREAMING_SNAKE_CASE( TensorFormatter[Mapping, "torch.Tensor", Mapping] ): def __init__( self: Any , UpperCamelCase: Optional[int]=None , **UpperCamelCase: Union[str, Any] ) -> int: super().__init__(features=UpperCamelCase ) snake_case__ = torch_tensor_kwargs import torch # noqa import torch at initialization def lowerCAmelCase_ ( self: Any , UpperCamelCase: Any ) -> List[str]: import torch if isinstance(UpperCamelCase , UpperCamelCase ) and column: if all( isinstance(UpperCamelCase , torch.Tensor ) and x.shape == column[0].shape and x.dtype == column[0].dtype for x in column ): return torch.stack(UpperCamelCase ) return column def lowerCAmelCase_ ( self: str , UpperCamelCase: Dict ) -> Union[str, Any]: import torch if isinstance(UpperCamelCase , (str, bytes, type(UpperCamelCase )) ): return value elif isinstance(UpperCamelCase , (np.character, np.ndarray) ) and np.issubdtype(value.dtype , np.character ): return value.tolist() snake_case__ = {} if isinstance(UpperCamelCase , (np.number, np.ndarray) ) and np.issubdtype(value.dtype , np.integer ): snake_case__ = {'dtype': torch.intaa} elif isinstance(UpperCamelCase , (np.number, np.ndarray) ) and np.issubdtype(value.dtype , np.floating ): snake_case__ = {'dtype': torch.floataa} elif config.PIL_AVAILABLE and "PIL" in sys.modules: import PIL.Image if isinstance(UpperCamelCase , PIL.Image.Image ): snake_case__ = np.asarray(UpperCamelCase ) return torch.tensor(UpperCamelCase , **{**default_dtype, **self.torch_tensor_kwargs} ) def lowerCAmelCase_ ( self: Any , UpperCamelCase: str ) -> Any: import torch # support for torch, tf, jax etc. if hasattr(UpperCamelCase , '__array__' ) and not isinstance(UpperCamelCase , torch.Tensor ): snake_case__ = data_struct.__array__() # support for nested types like struct of list of struct if isinstance(UpperCamelCase , np.ndarray ): if data_struct.dtype == object: # torch tensors cannot be instantied from an array of objects return self._consolidate([self.recursive_tensorize(UpperCamelCase ) for substruct in data_struct] ) elif isinstance(UpperCamelCase , (list, tuple) ): return self._consolidate([self.recursive_tensorize(UpperCamelCase ) for substruct in data_struct] ) return self._tensorize(UpperCamelCase ) def lowerCAmelCase_ ( self: List[Any] , UpperCamelCase: dict ) -> List[str]: return map_nested(self._recursive_tensorize , UpperCamelCase , map_list=UpperCamelCase ) def lowerCAmelCase_ ( self: Tuple , UpperCamelCase: pa.Table ) -> Mapping: snake_case__ = self.numpy_arrow_extractor().extract_row(UpperCamelCase ) snake_case__ = self.python_features_decoder.decode_row(UpperCamelCase ) return self.recursive_tensorize(UpperCamelCase ) def lowerCAmelCase_ ( self: List[str] , UpperCamelCase: pa.Table ) -> "torch.Tensor": snake_case__ = self.numpy_arrow_extractor().extract_column(UpperCamelCase ) snake_case__ = self.python_features_decoder.decode_column(UpperCamelCase , pa_table.column_names[0] ) snake_case__ = self.recursive_tensorize(UpperCamelCase ) snake_case__ = self._consolidate(UpperCamelCase ) return column def lowerCAmelCase_ ( self: Union[str, Any] , UpperCamelCase: pa.Table ) -> Mapping: snake_case__ = self.numpy_arrow_extractor().extract_batch(UpperCamelCase ) snake_case__ = self.python_features_decoder.decode_batch(UpperCamelCase ) snake_case__ = self.recursive_tensorize(UpperCamelCase ) for column_name in batch: snake_case__ = self._consolidate(batch[column_name] ) return batch
307
1
import os from pathlib import Path from unittest.mock import patch import pytest import zstandard as zstd from datasets.download.download_config import DownloadConfig from datasets.utils.file_utils import ( OfflineModeIsEnabled, cached_path, fsspec_get, fsspec_head, ftp_get, ftp_head, get_from_cache, http_get, http_head, ) _UpperCAmelCase : List[str] = "\\n Text data.\n Second line of data." _UpperCAmelCase : str = "file" @pytest.fixture(scope='session' ) def A ( lowercase ) -> List[Any]: '''simple docstring''' UpperCamelCase = tmp_path_factory.mktemp('data' ) / (FILE_PATH + '.zstd') UpperCamelCase = bytes(lowercase , 'utf-8' ) with zstd.open(lowercase , 'wb' ) as f: f.write(lowercase ) return path @pytest.fixture def A ( lowercase ) -> Tuple: '''simple docstring''' with open(os.path.join(tmpfs.local_root_dir , lowercase ) , 'w' ) as f: f.write(lowercase ) return FILE_PATH @pytest.mark.parametrize('compression_format' , ['gzip', 'xz', 'zstd'] ) def A ( lowercase , lowercase , lowercase , lowercase , lowercase , lowercase ) -> Union[str, Any]: '''simple docstring''' UpperCamelCase = {'gzip': gz_file, 'xz': xz_file, 'zstd': zstd_path} UpperCamelCase = input_paths[compression_format] UpperCamelCase = tmp_path / 'cache' UpperCamelCase = DownloadConfig(cache_dir=lowercase , extract_compressed_file=lowercase ) UpperCamelCase = cached_path(lowercase , download_config=lowercase ) with open(lowercase ) as f: UpperCamelCase = f.read() with open(lowercase ) as f: UpperCamelCase = f.read() assert extracted_file_content == expected_file_content @pytest.mark.parametrize('default_extracted' , [True, False] ) @pytest.mark.parametrize('default_cache_dir' , [True, False] ) def A ( lowercase , lowercase , lowercase , lowercase , lowercase ) -> List[str]: '''simple docstring''' UpperCamelCase = 'custom_cache' UpperCamelCase = 'custom_extracted_dir' UpperCamelCase = tmp_path / 'custom_extracted_path' if default_extracted: UpperCamelCase = ('downloads' if default_cache_dir else custom_cache_dir, 'extracted') else: monkeypatch.setattr('datasets.config.EXTRACTED_DATASETS_DIR' , lowercase ) monkeypatch.setattr('datasets.config.EXTRACTED_DATASETS_PATH' , str(lowercase ) ) UpperCamelCase = custom_extracted_path.parts[-2:] if default_cache_dir else (custom_cache_dir, custom_extracted_dir) UpperCamelCase = xz_file UpperCamelCase = ( DownloadConfig(extract_compressed_file=lowercase ) if default_cache_dir else DownloadConfig(cache_dir=tmp_path / custom_cache_dir , extract_compressed_file=lowercase ) ) UpperCamelCase = cached_path(lowercase , download_config=lowercase ) assert Path(lowercase ).parent.parts[-2:] == expected def A ( lowercase ) -> Tuple: '''simple docstring''' UpperCamelCase = str(Path(lowercase ).resolve() ) assert cached_path(lowercase ) == text_file # relative path UpperCamelCase = str(Path(lowercase ).resolve().relative_to(Path(os.getcwd() ) ) ) assert cached_path(lowercase ) == text_file def A ( lowercase ) -> Dict: '''simple docstring''' UpperCamelCase = str(tmp_path.resolve() / '__missing_file__.txt' ) with pytest.raises(lowercase ): cached_path(lowercase ) # relative path UpperCamelCase = './__missing_file__.txt' with pytest.raises(lowercase ): cached_path(lowercase ) def A ( lowercase ) -> Optional[Any]: '''simple docstring''' UpperCamelCase = get_from_cache(f'''tmp://{tmpfs_file}''' ) with open(lowercase ) as f: UpperCamelCase = f.read() assert output_file_content == FILE_CONTENT @patch('datasets.config.HF_DATASETS_OFFLINE' , lowercase ) def A ( ) -> Tuple: '''simple docstring''' with pytest.raises(lowercase ): cached_path('https://huggingface.co' ) @patch('datasets.config.HF_DATASETS_OFFLINE' , lowercase ) def A ( lowercase ) -> List[Any]: '''simple docstring''' UpperCamelCase = tmp_path_factory.mktemp('data' ) / 'file.html' with pytest.raises(lowercase ): http_get('https://huggingface.co' , temp_file=lowercase ) with pytest.raises(lowercase ): http_head('https://huggingface.co' ) @patch('datasets.config.HF_DATASETS_OFFLINE' , lowercase ) def A ( lowercase ) -> str: '''simple docstring''' UpperCamelCase = tmp_path_factory.mktemp('data' ) / 'file.html' with pytest.raises(lowercase ): ftp_get('ftp://huggingface.co' , temp_file=lowercase ) with pytest.raises(lowercase ): ftp_head('ftp://huggingface.co' ) @patch('datasets.config.HF_DATASETS_OFFLINE' , lowercase ) def A ( lowercase ) -> Optional[int]: '''simple docstring''' UpperCamelCase = tmp_path_factory.mktemp('data' ) / 'file.html' with pytest.raises(lowercase ): fsspec_get('s3://huggingface.co' , temp_file=lowercase ) with pytest.raises(lowercase ): fsspec_head('s3://huggingface.co' )
368
from __future__ import annotations def A ( lowercase , lowercase ) -> tuple[int, int]: '''simple docstring''' if b == 0: return (1, 0) ((UpperCamelCase) , (UpperCamelCase)) = extended_euclid(lowercase , a % b ) UpperCamelCase = a // b return (y, x - k * y) def A ( lowercase , lowercase , lowercase , lowercase ) -> int: '''simple docstring''' ((UpperCamelCase) , (UpperCamelCase)) = extended_euclid(lowercase , lowercase ) UpperCamelCase = na * na UpperCamelCase = ra * x * na + ra * y * na return (n % m + m) % m def A ( lowercase , lowercase ) -> int: '''simple docstring''' ((UpperCamelCase) , (UpperCamelCase)) = extended_euclid(lowercase , lowercase ) if b < 0: UpperCamelCase = (b % n + n) % n return b def A ( lowercase , lowercase , lowercase , lowercase ) -> int: '''simple docstring''' UpperCamelCase , UpperCamelCase = invert_modulo(lowercase , lowercase ), invert_modulo(lowercase , lowercase ) UpperCamelCase = na * na UpperCamelCase = ra * x * na + ra * y * na return (n % m + m) % m if __name__ == "__main__": from doctest import testmod testmod(name="chinese_remainder_theorem", verbose=True) testmod(name="chinese_remainder_theorem2", verbose=True) testmod(name="invert_modulo", verbose=True) testmod(name="extended_euclid", verbose=True)
110
0
"""simple docstring""" from string import ascii_uppercase _a = {str(ord(c) - 55): c for c in ascii_uppercase} def __a ( __lowerCamelCase, __lowerCamelCase ): if isinstance(__lowerCamelCase, __lowerCamelCase ): raise TypeError("int() can't convert non-string with explicit base" ) if num < 0: raise ValueError("parameter must be positive int" ) if isinstance(__lowerCamelCase, __lowerCamelCase ): raise TypeError("'str' object cannot be interpreted as an integer" ) if isinstance(__lowerCamelCase, __lowerCamelCase ): raise TypeError("'float' object cannot be interpreted as an integer" ) if base in (0, 1): raise ValueError("base must be >= 2" ) if base > 36: raise ValueError("base must be <= 36" ) UpperCAmelCase_ : Optional[Any] = "" UpperCAmelCase_ : int = 0 UpperCAmelCase_ : str = 0 while div != 1: UpperCAmelCase_ , UpperCAmelCase_ : str = divmod(__lowerCamelCase, __lowerCamelCase ) if base >= 11 and 9 < mod < 36: UpperCAmelCase_ : Tuple = ALPHABET_VALUES[str(__lowerCamelCase )] else: UpperCAmelCase_ : int = str(__lowerCamelCase ) new_value += actual_value UpperCAmelCase_ : Any = num // base UpperCAmelCase_ : Optional[int] = div if div == 0: return str(new_value[::-1] ) elif div == 1: new_value += str(__lowerCamelCase ) return str(new_value[::-1] ) return new_value[::-1] if __name__ == "__main__": import doctest doctest.testmod() for base in range(2, 37): for num in range(1_000): assert int(decimal_to_any(num, base), base) == num, ( num, base, decimal_to_any(num, base), int(decimal_to_any(num, base), base), )
61
import sys from .dependency_versions_table import deps from .utils.versions import require_version, require_version_core # define which module versions we always want to check at run time # (usually the ones defined in `install_requires` in setup.py) # # order specific notes: # - tqdm must be checked before tokenizers snake_case : List[str] = '''python tqdm regex requests packaging filelock numpy tokenizers'''.split() if sys.version_info < (3, 7): pkgs_to_check_at_runtime.append('''dataclasses''') if sys.version_info < (3, 8): pkgs_to_check_at_runtime.append('''importlib_metadata''') for pkg in pkgs_to_check_at_runtime: if pkg in deps: if pkg == "tokenizers": # must be loaded here, or else tqdm check may fail from .utils import is_tokenizers_available if not is_tokenizers_available(): continue # not required, check version only if installed require_version_core(deps[pkg]) else: raise ValueError(f"""can't find {pkg} in {deps.keys()}, check dependency_versions_table.py""") def __lowercase ( __lowerCAmelCase : Optional[Any] , __lowerCAmelCase : Tuple=None ): require_version(deps[pkg] , __lowerCAmelCase )
240
0
from __future__ import annotations from statistics import mean def __lowerCAmelCase ( a__ , a__ , a__ ) -> list[int]: __a = [0] * no_of_processes __a = [0] * no_of_processes # Initialize remaining_time to waiting_time. for i in range(__lowerCAmelCase ): __a = burst_time[i] __a = [] __a = 0 __a = 0 # When processes are not completed, # A process whose arrival time has passed \ # and has remaining execution time is put into the ready_process. # The shortest process in the ready_process, target_process is executed. while completed != no_of_processes: __a = [] __a = -1 for i in range(__lowerCAmelCase ): if (arrival_time[i] <= total_time) and (remaining_time[i] > 0): ready_process.append(__lowerCAmelCase ) if len(__lowerCAmelCase ) > 0: __a = ready_process[0] for i in ready_process: if remaining_time[i] < remaining_time[target_process]: __a = i total_time += burst_time[target_process] completed += 1 __a = 0 __a = ( total_time - arrival_time[target_process] - burst_time[target_process] ) else: total_time += 1 return waiting_time def __lowerCAmelCase ( a__ , a__ , a__ ) -> list[int]: __a = [0] * no_of_processes for i in range(__lowerCAmelCase ): __a = burst_time[i] + waiting_time[i] return turn_around_time if __name__ == "__main__": print('[TEST CASE 01]') A : int = 4 A : Union[str, Any] = [2, 5, 3, 7] A : Dict = [0, 0, 0, 0] A : Optional[Any] = calculate_waitingtime(arrival_time, burst_time, no_of_processes) A : Dict = calculate_turnaroundtime( burst_time, no_of_processes, waiting_time ) # Printing the Result print('PID\tBurst Time\tArrival Time\tWaiting Time\tTurnaround Time') for i, process_id in enumerate(list(range(1, 5))): print( F"{process_id}\t{burst_time[i]}\t\t\t{arrival_time[i]}\t\t\t\t" F"{waiting_time[i]}\t\t\t\t{turn_around_time[i]}" ) print(F"\nAverage waiting time = {mean(waiting_time):.5f}") print(F"Average turnaround time = {mean(turn_around_time):.5f}")
371
from typing import List from .keymap import KEYMAP, get_character def __lowerCAmelCase ( a__ ) -> List[str]: def decorator(a__ ): __a = getattr(a__ , '''handle_key''' , [] ) handle += [key] setattr(a__ , '''handle_key''' , a__ ) return func return decorator def __lowerCAmelCase ( *a__ ) -> str: def decorator(a__ ): __a = getattr(a__ , '''handle_key''' , [] ) handle += keys setattr(a__ , '''handle_key''' , a__ ) return func return decorator class __A( a ): def __new__( cls , _snake_case , _snake_case , _snake_case ) -> Union[str, Any]: '''simple docstring''' __a = super().__new__(cls , _snake_case , _snake_case , _snake_case ) if not hasattr(_snake_case , '''key_handler''' ): setattr(_snake_case , '''key_handler''' , {} ) setattr(_snake_case , '''handle_input''' , KeyHandler.handle_input ) for value in attrs.values(): __a = getattr(_snake_case , '''handle_key''' , [] ) for key in handled_keys: __a = value return new_cls @staticmethod def SCREAMING_SNAKE_CASE_ ( cls ) -> List[str]: '''simple docstring''' __a = get_character() if char != KEYMAP["undefined"]: __a = ord(_snake_case ) __a = cls.key_handler.get(_snake_case ) if handler: __a = char return handler(cls ) else: return None def __lowerCAmelCase ( cls ) -> Union[str, Any]: return KeyHandler(cls.__name__ , cls.__bases__ , cls.__dict__.copy() )
33
0
"""simple docstring""" 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 __lowercase ( unittest.TestCase ): '''simple docstring''' def __init__( self , _UpperCAmelCase , _UpperCAmelCase=7 , _UpperCAmelCase=3 , _UpperCAmelCase=18 , _UpperCAmelCase=30 , _UpperCAmelCase=400 , _UpperCAmelCase=True , _UpperCAmelCase=32 , _UpperCAmelCase=True , ): __a : Dict = parent __a : List[str] = batch_size __a : str = num_channels __a : Optional[int] = image_size __a : Tuple = min_resolution __a : str = max_resolution __a : List[str] = do_resize __a : Dict = size_divisor __a : Dict = do_rescale def _lowerCamelCase ( self ): return { "do_resize": self.do_resize, "size_divisor": self.size_divisor, "do_rescale": self.do_rescale, } @require_torch @require_vision class __lowercase ( _UpperCamelCase , unittest.TestCase ): '''simple docstring''' __lowerCAmelCase = GLPNImageProcessor if is_vision_available() else None def _lowerCamelCase ( self ): __a : str = GLPNImageProcessingTester(self ) @property def _lowerCamelCase ( self ): return self.image_processor_tester.prepare_image_processor_dict() def _lowerCamelCase ( self ): __a : Optional[int] = self.image_processing_class(**self.image_processor_dict ) self.assertTrue(hasattr(_UpperCAmelCase , '''do_resize''' ) ) self.assertTrue(hasattr(_UpperCAmelCase , '''size_divisor''' ) ) self.assertTrue(hasattr(_UpperCAmelCase , '''resample''' ) ) self.assertTrue(hasattr(_UpperCAmelCase , '''do_rescale''' ) ) def _lowerCamelCase ( self ): pass def _lowerCamelCase ( self ): # Initialize image_processing __a : Any = self.image_processing_class(**self.image_processor_dict ) # create random PIL images __a : Any = prepare_image_inputs(self.image_processor_tester , equal_resolution=_UpperCAmelCase ) for image in image_inputs: self.assertIsInstance(_UpperCAmelCase , Image.Image ) # Test not batched input (GLPNImageProcessor doesn't support batching) __a : Optional[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 _lowerCamelCase ( self ): # Initialize image_processing __a : Optional[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=_UpperCAmelCase , numpify=_UpperCAmelCase ) for image in image_inputs: self.assertIsInstance(_UpperCAmelCase , np.ndarray ) # 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 _lowerCamelCase ( self ): # Initialize image_processing __a : int = self.image_processing_class(**self.image_processor_dict ) # create random PyTorch tensors __a : Dict = prepare_image_inputs(self.image_processor_tester , equal_resolution=_UpperCAmelCase , torchify=_UpperCAmelCase ) for image in image_inputs: self.assertIsInstance(_UpperCAmelCase , torch.Tensor ) # Test not batched input (GLPNImageProcessor doesn't support batching) __a : List[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 )
160
"""simple docstring""" from statistics import mean import numpy as np def __A ( a_ :list , a_ :list , a_ :list , a_ :int) -> list: __a : Any = 0 # Number of processes finished __a : Union[str, Any] = 0 # Displays the finished process. # If it is 0, the performance is completed if it is 1, before the performance. __a : Any = [0] * no_of_process # List to include calculation results __a : str = [0] * no_of_process # Sort by arrival time. __a : List[Any] = [burst_time[i] for i in np.argsort(a_)] __a : Tuple = [process_name[i] for i in np.argsort(a_)] arrival_time.sort() while no_of_process > finished_process_count: __a : Optional[Any] = 0 while finished_process[i] == 1: i += 1 if current_time < arrival_time[i]: __a : Dict = arrival_time[i] __a : Dict = 0 # Index showing the location of the process being performed __a : Tuple = 0 # Saves the current response ratio. __a : List[str] = 0 for i in range(0 , a_): if finished_process[i] == 0 and arrival_time[i] <= current_time: __a : Tuple = (burst_time[i] + (current_time - arrival_time[i])) / burst_time[ i ] if response_ratio < temp: __a : Tuple = temp __a : Optional[Any] = i # Calculate the turn around time __a : Optional[int] = current_time + burst_time[loc] - arrival_time[loc] current_time += burst_time[loc] # Indicates that the process has been performed. __a : int = 1 # Increase finished_process_count by 1 finished_process_count += 1 return turn_around_time def __A ( a_ :list , a_ :list , a_ :list , a_ :int) -> list: __a : Dict = [0] * no_of_process for i in range(0 , a_): __a : Optional[Any] = turn_around_time[i] - burst_time[i] return waiting_time if __name__ == "__main__": A = 5 A = ['''A''', '''B''', '''C''', '''D''', '''E'''] A = [1, 2, 3, 4, 5] A = [1, 2, 3, 4, 5] A = calculate_turn_around_time( process_name, arrival_time, burst_time, no_of_process ) A = calculate_waiting_time( process_name, turn_around_time, burst_time, no_of_process ) print('''Process name \tArrival time \tBurst time \tTurn around time \tWaiting time''') for i in range(0, no_of_process): print( F'{process_name[i]}\t\t{arrival_time[i]}\t\t{burst_time[i]}\t\t' F'{turn_around_time[i]}\t\t\t{waiting_time[i]}' ) print(F'average waiting time : {mean(waiting_time):.5f}') print(F'average turn around time : {mean(turn_around_time):.5f}')
160
1
'''simple docstring''' import unittest from transformers import MobileBertConfig, is_torch_available from transformers.models.auto import get_values from transformers.testing_utils import require_sentencepiece, require_tokenizers, require_torch, slow, torch_device 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 ( MODEL_FOR_PRETRAINING_MAPPING, MobileBertForMaskedLM, MobileBertForMultipleChoice, MobileBertForNextSentencePrediction, MobileBertForPreTraining, MobileBertForQuestionAnswering, MobileBertForSequenceClassification, MobileBertForTokenClassification, MobileBertModel, ) class A : def __init__( self , SCREAMING_SNAKE_CASE , SCREAMING_SNAKE_CASE=13 , SCREAMING_SNAKE_CASE=7 , SCREAMING_SNAKE_CASE=True , SCREAMING_SNAKE_CASE=True , SCREAMING_SNAKE_CASE=True , SCREAMING_SNAKE_CASE=True , SCREAMING_SNAKE_CASE=99 , SCREAMING_SNAKE_CASE=64 , SCREAMING_SNAKE_CASE=32 , SCREAMING_SNAKE_CASE=5 , SCREAMING_SNAKE_CASE=4 , SCREAMING_SNAKE_CASE=37 , SCREAMING_SNAKE_CASE="gelu" , SCREAMING_SNAKE_CASE=0.1 , SCREAMING_SNAKE_CASE=0.1 , SCREAMING_SNAKE_CASE=512 , SCREAMING_SNAKE_CASE=16 , SCREAMING_SNAKE_CASE=2 , SCREAMING_SNAKE_CASE=0.02 , SCREAMING_SNAKE_CASE=3 , SCREAMING_SNAKE_CASE=4 , SCREAMING_SNAKE_CASE=None , ) -> List[Any]: """simple docstring""" A : Optional[int] = parent A : Optional[Any] = batch_size A : Optional[int] = seq_length A : Optional[Any] = is_training A : Tuple = use_input_mask A : Any = use_token_type_ids A : Union[str, Any] = use_labels A : List[str] = vocab_size A : Any = hidden_size A : Union[str, Any] = embedding_size A : Dict = num_hidden_layers A : Tuple = num_attention_heads A : Dict = intermediate_size A : Optional[int] = hidden_act A : Optional[Any] = hidden_dropout_prob A : List[Any] = attention_probs_dropout_prob A : List[str] = max_position_embeddings A : Tuple = type_vocab_size A : str = type_sequence_label_size A : Any = initializer_range A : Union[str, Any] = num_labels A : List[Any] = num_choices A : Union[str, Any] = scope def __lowerCAmelCase ( self ) -> Union[str, Any]: """simple docstring""" A : Optional[int] = ids_tensor([self.batch_size, self.seq_length] , self.vocab_size ) A : int = None if self.use_input_mask: A : List[Any] = random_attention_mask([self.batch_size, self.seq_length] ) A : int = None if self.use_token_type_ids: A : Tuple = ids_tensor([self.batch_size, self.seq_length] , self.type_vocab_size ) A : Union[str, Any] = None A : List[Any] = None A : Tuple = None if self.use_labels: A : Optional[int] = ids_tensor([self.batch_size] , self.type_sequence_label_size ) A : int = ids_tensor([self.batch_size, self.seq_length] , self.num_labels ) A : Any = ids_tensor([self.batch_size] , self.num_choices ) A : List[str] = self.get_config() return config, input_ids, token_type_ids, input_mask, sequence_labels, token_labels, choice_labels def __lowerCAmelCase ( self ) -> List[str]: """simple docstring""" return MobileBertConfig( 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 , embedding_size=self.embedding_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=SCREAMING_SNAKE_CASE , initializer_range=self.initializer_range , ) def __lowerCAmelCase ( self , SCREAMING_SNAKE_CASE , SCREAMING_SNAKE_CASE , SCREAMING_SNAKE_CASE , SCREAMING_SNAKE_CASE , SCREAMING_SNAKE_CASE , SCREAMING_SNAKE_CASE , SCREAMING_SNAKE_CASE ) -> Tuple: """simple docstring""" A : Optional[Any] = MobileBertModel(config=SCREAMING_SNAKE_CASE ) model.to(SCREAMING_SNAKE_CASE ) model.eval() A : Optional[int] = model(SCREAMING_SNAKE_CASE , attention_mask=SCREAMING_SNAKE_CASE , token_type_ids=SCREAMING_SNAKE_CASE ) A : str = model(SCREAMING_SNAKE_CASE , token_type_ids=SCREAMING_SNAKE_CASE ) A : str = model(SCREAMING_SNAKE_CASE ) self.parent.assertEqual(result.last_hidden_state.shape , (self.batch_size, self.seq_length, self.hidden_size) ) self.parent.assertEqual(result.pooler_output.shape , (self.batch_size, self.hidden_size) ) def __lowerCAmelCase ( self , SCREAMING_SNAKE_CASE , SCREAMING_SNAKE_CASE , SCREAMING_SNAKE_CASE , SCREAMING_SNAKE_CASE , SCREAMING_SNAKE_CASE , SCREAMING_SNAKE_CASE , SCREAMING_SNAKE_CASE ) -> List[Any]: """simple docstring""" A : Optional[Any] = MobileBertForMaskedLM(config=SCREAMING_SNAKE_CASE ) model.to(SCREAMING_SNAKE_CASE ) model.eval() A : Optional[Any] = model(SCREAMING_SNAKE_CASE , attention_mask=SCREAMING_SNAKE_CASE , token_type_ids=SCREAMING_SNAKE_CASE , labels=SCREAMING_SNAKE_CASE ) self.parent.assertEqual(result.logits.shape , (self.batch_size, self.seq_length, self.vocab_size) ) def __lowerCAmelCase ( self , SCREAMING_SNAKE_CASE , SCREAMING_SNAKE_CASE , SCREAMING_SNAKE_CASE , SCREAMING_SNAKE_CASE , SCREAMING_SNAKE_CASE , SCREAMING_SNAKE_CASE , SCREAMING_SNAKE_CASE ) -> Any: """simple docstring""" A : Dict = MobileBertForNextSentencePrediction(config=SCREAMING_SNAKE_CASE ) model.to(SCREAMING_SNAKE_CASE ) model.eval() A : Any = model( SCREAMING_SNAKE_CASE , attention_mask=SCREAMING_SNAKE_CASE , token_type_ids=SCREAMING_SNAKE_CASE , labels=SCREAMING_SNAKE_CASE , ) self.parent.assertEqual(result.logits.shape , (self.batch_size, 2) ) def __lowerCAmelCase ( self , SCREAMING_SNAKE_CASE , SCREAMING_SNAKE_CASE , SCREAMING_SNAKE_CASE , SCREAMING_SNAKE_CASE , SCREAMING_SNAKE_CASE , SCREAMING_SNAKE_CASE , SCREAMING_SNAKE_CASE ) -> Tuple: """simple docstring""" A : int = MobileBertForPreTraining(config=SCREAMING_SNAKE_CASE ) model.to(SCREAMING_SNAKE_CASE ) model.eval() A : Any = model( SCREAMING_SNAKE_CASE , attention_mask=SCREAMING_SNAKE_CASE , token_type_ids=SCREAMING_SNAKE_CASE , labels=SCREAMING_SNAKE_CASE , next_sentence_label=SCREAMING_SNAKE_CASE , ) self.parent.assertEqual(result.prediction_logits.shape , (self.batch_size, self.seq_length, self.vocab_size) ) self.parent.assertEqual(result.seq_relationship_logits.shape , (self.batch_size, 2) ) def __lowerCAmelCase ( self , SCREAMING_SNAKE_CASE , SCREAMING_SNAKE_CASE , SCREAMING_SNAKE_CASE , SCREAMING_SNAKE_CASE , SCREAMING_SNAKE_CASE , SCREAMING_SNAKE_CASE , SCREAMING_SNAKE_CASE ) -> Union[str, Any]: """simple docstring""" A : str = MobileBertForQuestionAnswering(config=SCREAMING_SNAKE_CASE ) model.to(SCREAMING_SNAKE_CASE ) model.eval() A : int = model( SCREAMING_SNAKE_CASE , attention_mask=SCREAMING_SNAKE_CASE , token_type_ids=SCREAMING_SNAKE_CASE , start_positions=SCREAMING_SNAKE_CASE , end_positions=SCREAMING_SNAKE_CASE , ) self.parent.assertEqual(result.start_logits.shape , (self.batch_size, self.seq_length) ) self.parent.assertEqual(result.end_logits.shape , (self.batch_size, self.seq_length) ) def __lowerCAmelCase ( self , SCREAMING_SNAKE_CASE , SCREAMING_SNAKE_CASE , SCREAMING_SNAKE_CASE , SCREAMING_SNAKE_CASE , SCREAMING_SNAKE_CASE , SCREAMING_SNAKE_CASE , SCREAMING_SNAKE_CASE ) -> Tuple: """simple docstring""" A : List[str] = self.num_labels A : Dict = MobileBertForSequenceClassification(SCREAMING_SNAKE_CASE ) model.to(SCREAMING_SNAKE_CASE ) model.eval() A : Union[str, Any] = model(SCREAMING_SNAKE_CASE , attention_mask=SCREAMING_SNAKE_CASE , token_type_ids=SCREAMING_SNAKE_CASE , labels=SCREAMING_SNAKE_CASE ) self.parent.assertEqual(result.logits.shape , (self.batch_size, self.num_labels) ) def __lowerCAmelCase ( self , SCREAMING_SNAKE_CASE , SCREAMING_SNAKE_CASE , SCREAMING_SNAKE_CASE , SCREAMING_SNAKE_CASE , SCREAMING_SNAKE_CASE , SCREAMING_SNAKE_CASE , SCREAMING_SNAKE_CASE ) -> Optional[int]: """simple docstring""" A : Any = self.num_labels A : str = MobileBertForTokenClassification(config=SCREAMING_SNAKE_CASE ) model.to(SCREAMING_SNAKE_CASE ) model.eval() A : Union[str, Any] = model(SCREAMING_SNAKE_CASE , attention_mask=SCREAMING_SNAKE_CASE , token_type_ids=SCREAMING_SNAKE_CASE , labels=SCREAMING_SNAKE_CASE ) self.parent.assertEqual(result.logits.shape , (self.batch_size, self.seq_length, self.num_labels) ) def __lowerCAmelCase ( self , SCREAMING_SNAKE_CASE , SCREAMING_SNAKE_CASE , SCREAMING_SNAKE_CASE , SCREAMING_SNAKE_CASE , SCREAMING_SNAKE_CASE , SCREAMING_SNAKE_CASE , SCREAMING_SNAKE_CASE ) -> Union[str, Any]: """simple docstring""" A : List[Any] = self.num_choices A : Tuple = MobileBertForMultipleChoice(config=SCREAMING_SNAKE_CASE ) model.to(SCREAMING_SNAKE_CASE ) model.eval() A : List[Any] = input_ids.unsqueeze(1 ).expand(-1 , self.num_choices , -1 ).contiguous() A : Dict = token_type_ids.unsqueeze(1 ).expand(-1 , self.num_choices , -1 ).contiguous() A : List[str] = input_mask.unsqueeze(1 ).expand(-1 , self.num_choices , -1 ).contiguous() A : Union[str, Any] = model( SCREAMING_SNAKE_CASE , attention_mask=SCREAMING_SNAKE_CASE , token_type_ids=SCREAMING_SNAKE_CASE , labels=SCREAMING_SNAKE_CASE , ) self.parent.assertEqual(result.logits.shape , (self.batch_size, self.num_choices) ) def __lowerCAmelCase ( self ) -> List[str]: """simple docstring""" A : List[str] = self.prepare_config_and_inputs() ( ( A ), ( A ), ( A ), ( A ), ( A ), ( A ), ( A ), ) : Dict = config_and_inputs A : Tuple = {'''input_ids''': input_ids, '''token_type_ids''': token_type_ids, '''attention_mask''': input_mask} return config, inputs_dict @require_torch class A ( __snake_case , __snake_case , unittest.TestCase ): __magic_name__ = ( ( MobileBertModel, MobileBertForMaskedLM, MobileBertForMultipleChoice, MobileBertForNextSentencePrediction, MobileBertForPreTraining, MobileBertForQuestionAnswering, MobileBertForSequenceClassification, MobileBertForTokenClassification, ) if is_torch_available() else () ) __magic_name__ = ( { '''feature-extraction''': MobileBertModel, '''fill-mask''': MobileBertForMaskedLM, '''question-answering''': MobileBertForQuestionAnswering, '''text-classification''': MobileBertForSequenceClassification, '''token-classification''': MobileBertForTokenClassification, '''zero-shot''': MobileBertForSequenceClassification, } if is_torch_available() else {} ) __magic_name__ = True def __lowerCAmelCase ( self , SCREAMING_SNAKE_CASE , SCREAMING_SNAKE_CASE , SCREAMING_SNAKE_CASE=False ) -> str: """simple docstring""" A : int = super()._prepare_for_class(SCREAMING_SNAKE_CASE , SCREAMING_SNAKE_CASE , return_labels=SCREAMING_SNAKE_CASE ) if return_labels: if model_class in get_values(SCREAMING_SNAKE_CASE ): A : Optional[Any] = torch.zeros( (self.model_tester.batch_size, self.model_tester.seq_length) , dtype=torch.long , device=SCREAMING_SNAKE_CASE ) A : List[str] = torch.zeros( self.model_tester.batch_size , dtype=torch.long , device=SCREAMING_SNAKE_CASE ) return inputs_dict def __lowerCAmelCase ( self ) -> Union[str, Any]: """simple docstring""" A : Dict = MobileBertModelTester(self ) A : List[Any] = ConfigTester(self , config_class=SCREAMING_SNAKE_CASE , hidden_size=37 ) def __lowerCAmelCase ( self ) -> str: """simple docstring""" self.config_tester.run_common_tests() def __lowerCAmelCase ( self ) -> Any: """simple docstring""" A : int = self.model_tester.prepare_config_and_inputs() self.model_tester.create_and_check_mobilebert_model(*SCREAMING_SNAKE_CASE ) def __lowerCAmelCase ( self ) -> List[Any]: """simple docstring""" A : Optional[Any] = self.model_tester.prepare_config_and_inputs() self.model_tester.create_and_check_mobilebert_for_masked_lm(*SCREAMING_SNAKE_CASE ) def __lowerCAmelCase ( self ) -> Tuple: """simple docstring""" A : List[str] = self.model_tester.prepare_config_and_inputs() self.model_tester.create_and_check_mobilebert_for_multiple_choice(*SCREAMING_SNAKE_CASE ) def __lowerCAmelCase ( self ) -> Union[str, Any]: """simple docstring""" A : Optional[Any] = self.model_tester.prepare_config_and_inputs() self.model_tester.create_and_check_mobilebert_for_next_sequence_prediction(*SCREAMING_SNAKE_CASE ) def __lowerCAmelCase ( self ) -> Tuple: """simple docstring""" A : Union[str, Any] = self.model_tester.prepare_config_and_inputs() self.model_tester.create_and_check_mobilebert_for_pretraining(*SCREAMING_SNAKE_CASE ) def __lowerCAmelCase ( self ) -> Optional[int]: """simple docstring""" A : Optional[int] = self.model_tester.prepare_config_and_inputs() self.model_tester.create_and_check_mobilebert_for_question_answering(*SCREAMING_SNAKE_CASE ) def __lowerCAmelCase ( self ) -> List[Any]: """simple docstring""" A : List[str] = self.model_tester.prepare_config_and_inputs() self.model_tester.create_and_check_mobilebert_for_sequence_classification(*SCREAMING_SNAKE_CASE ) def __lowerCAmelCase ( self ) -> int: """simple docstring""" A : int = self.model_tester.prepare_config_and_inputs() self.model_tester.create_and_check_mobilebert_for_token_classification(*SCREAMING_SNAKE_CASE ) def lowerCAmelCase_ ( snake_case__ ): '''simple docstring''' return torch.tensor( snake_case__ , dtype=torch.long , device=snake_case__ , ) lowercase : Dict = 1e-3 @require_torch @require_sentencepiece @require_tokenizers class A ( unittest.TestCase ): @slow def __lowerCAmelCase ( self ) -> Optional[int]: """simple docstring""" A : Optional[Any] = MobileBertModel.from_pretrained('''google/mobilebert-uncased''' ).to(SCREAMING_SNAKE_CASE ) A : Optional[Any] = _long_tensor([[101, 7110, 1005, 1056, 2023, 11333, 17413, 1029, 102]] ) with torch.no_grad(): A : List[Any] = model(SCREAMING_SNAKE_CASE )[0] A : Optional[Any] = torch.Size((1, 9, 512) ) self.assertEqual(output.shape , SCREAMING_SNAKE_CASE ) A : Union[str, Any] = torch.tensor( [ [ [-2.4_73_65_26e07, 8.2_69_16_56e04, 1.6_52_18_38e05], [-5.7_54_17_04e-01, 3.9_05_60_22e00, 4.4_01_15_07e00], [2.6_04_73_59e00, 1.5_67_76_52e00, -1.7_32_41_88e-01], ] ] , device=SCREAMING_SNAKE_CASE , ) # MobileBERT results range from 10e0 to 10e8. Even a 0.0000001% difference with a value of 10e8 results in a # ~1 difference, it's therefore not a good idea to measure using addition. # Here, we instead divide the expected result with the result in order to obtain ~1. We then check that the # result is held between bounds: 1 - TOLERANCE < expected_result / result < 1 + TOLERANCE A : Union[str, Any] = torch.all((expected_slice / output[..., :3, :3]) >= 1 - TOLERANCE ) A : Tuple = torch.all((expected_slice / output[..., :3, :3]) <= 1 + TOLERANCE ) self.assertTrue(lower_bound and upper_bound )
311
'''simple docstring''' import tempfile import torch from diffusers import ( DEISMultistepScheduler, DPMSolverMultistepScheduler, DPMSolverSinglestepScheduler, UniPCMultistepScheduler, ) from .test_schedulers import SchedulerCommonTest class A ( __snake_case ): __magic_name__ = (UniPCMultistepScheduler,) __magic_name__ = (('''num_inference_steps''', 25),) def __lowerCAmelCase ( self , **SCREAMING_SNAKE_CASE ) -> List[str]: """simple docstring""" A : str = { '''num_train_timesteps''': 1000, '''beta_start''': 0.0_001, '''beta_end''': 0.02, '''beta_schedule''': '''linear''', '''solver_order''': 2, '''solver_type''': '''bh2''', } config.update(**SCREAMING_SNAKE_CASE ) return config def __lowerCAmelCase ( self , SCREAMING_SNAKE_CASE=0 , **SCREAMING_SNAKE_CASE ) -> Any: """simple docstring""" A : List[Any] = dict(self.forward_default_kwargs ) A : Union[str, Any] = kwargs.pop('''num_inference_steps''' , SCREAMING_SNAKE_CASE ) A : Optional[Any] = self.dummy_sample A : int = 0.1 * sample A : Union[str, Any] = [residual + 0.2, residual + 0.15, residual + 0.10] for scheduler_class in self.scheduler_classes: A : Optional[Any] = self.get_scheduler_config(**SCREAMING_SNAKE_CASE ) A : Optional[int] = scheduler_class(**SCREAMING_SNAKE_CASE ) scheduler.set_timesteps(SCREAMING_SNAKE_CASE ) # copy over dummy past residuals A : List[Any] = dummy_past_residuals[: scheduler.config.solver_order] with tempfile.TemporaryDirectory() as tmpdirname: scheduler.save_config(SCREAMING_SNAKE_CASE ) A : List[Any] = scheduler_class.from_pretrained(SCREAMING_SNAKE_CASE ) new_scheduler.set_timesteps(SCREAMING_SNAKE_CASE ) # copy over dummy past residuals A : Dict = dummy_past_residuals[: new_scheduler.config.solver_order] A, A : Tuple = sample, sample for t in range(SCREAMING_SNAKE_CASE , time_step + scheduler.config.solver_order + 1 ): A : Any = scheduler.step(SCREAMING_SNAKE_CASE , SCREAMING_SNAKE_CASE , SCREAMING_SNAKE_CASE , **SCREAMING_SNAKE_CASE ).prev_sample A : Optional[Any] = new_scheduler.step(SCREAMING_SNAKE_CASE , SCREAMING_SNAKE_CASE , SCREAMING_SNAKE_CASE , **SCREAMING_SNAKE_CASE ).prev_sample assert torch.sum(torch.abs(output - new_output ) ) < 1e-5, "Scheduler outputs are not identical" def __lowerCAmelCase ( self , SCREAMING_SNAKE_CASE=0 , **SCREAMING_SNAKE_CASE ) -> Tuple: """simple docstring""" A : Optional[Any] = dict(self.forward_default_kwargs ) A : Tuple = kwargs.pop('''num_inference_steps''' , SCREAMING_SNAKE_CASE ) A : List[Any] = self.dummy_sample A : int = 0.1 * sample A : Optional[Any] = [residual + 0.2, residual + 0.15, residual + 0.10] for scheduler_class in self.scheduler_classes: A : Optional[int] = self.get_scheduler_config() A : Any = scheduler_class(**SCREAMING_SNAKE_CASE ) scheduler.set_timesteps(SCREAMING_SNAKE_CASE ) # copy over dummy past residuals (must be after setting timesteps) A : int = dummy_past_residuals[: scheduler.config.solver_order] with tempfile.TemporaryDirectory() as tmpdirname: scheduler.save_config(SCREAMING_SNAKE_CASE ) A : int = scheduler_class.from_pretrained(SCREAMING_SNAKE_CASE ) # copy over dummy past residuals new_scheduler.set_timesteps(SCREAMING_SNAKE_CASE ) # copy over dummy past residual (must be after setting timesteps) A : Optional[Any] = dummy_past_residuals[: new_scheduler.config.solver_order] A : List[Any] = scheduler.step(SCREAMING_SNAKE_CASE , SCREAMING_SNAKE_CASE , SCREAMING_SNAKE_CASE , **SCREAMING_SNAKE_CASE ).prev_sample A : List[Any] = new_scheduler.step(SCREAMING_SNAKE_CASE , SCREAMING_SNAKE_CASE , SCREAMING_SNAKE_CASE , **SCREAMING_SNAKE_CASE ).prev_sample assert torch.sum(torch.abs(output - new_output ) ) < 1e-5, "Scheduler outputs are not identical" def __lowerCAmelCase ( self , SCREAMING_SNAKE_CASE=None , **SCREAMING_SNAKE_CASE ) -> Optional[int]: """simple docstring""" if scheduler is None: A : Dict = self.scheduler_classes[0] A : Union[str, Any] = self.get_scheduler_config(**SCREAMING_SNAKE_CASE ) A : List[Any] = scheduler_class(**SCREAMING_SNAKE_CASE ) A : Tuple = self.scheduler_classes[0] A : Union[str, Any] = self.get_scheduler_config(**SCREAMING_SNAKE_CASE ) A : List[str] = scheduler_class(**SCREAMING_SNAKE_CASE ) A : int = 10 A : Tuple = self.dummy_model() A : Any = self.dummy_sample_deter scheduler.set_timesteps(SCREAMING_SNAKE_CASE ) for i, t in enumerate(scheduler.timesteps ): A : int = model(SCREAMING_SNAKE_CASE , SCREAMING_SNAKE_CASE ) A : Optional[Any] = scheduler.step(SCREAMING_SNAKE_CASE , SCREAMING_SNAKE_CASE , SCREAMING_SNAKE_CASE ).prev_sample return sample def __lowerCAmelCase ( self ) -> Optional[int]: """simple docstring""" A : Tuple = dict(self.forward_default_kwargs ) A : List[Any] = kwargs.pop('''num_inference_steps''' , SCREAMING_SNAKE_CASE ) for scheduler_class in self.scheduler_classes: A : Dict = self.get_scheduler_config() A : Dict = scheduler_class(**SCREAMING_SNAKE_CASE ) A : Optional[Any] = self.dummy_sample A : Optional[int] = 0.1 * sample if num_inference_steps is not None and hasattr(SCREAMING_SNAKE_CASE , '''set_timesteps''' ): scheduler.set_timesteps(SCREAMING_SNAKE_CASE ) elif num_inference_steps is not None and not hasattr(SCREAMING_SNAKE_CASE , '''set_timesteps''' ): A : Tuple = num_inference_steps # copy over dummy past residuals (must be done after set_timesteps) A : Dict = [residual + 0.2, residual + 0.15, residual + 0.10] A : List[str] = dummy_past_residuals[: scheduler.config.solver_order] A : List[Any] = scheduler.timesteps[5] A : Dict = scheduler.timesteps[6] A : List[Any] = scheduler.step(SCREAMING_SNAKE_CASE , SCREAMING_SNAKE_CASE , SCREAMING_SNAKE_CASE , **SCREAMING_SNAKE_CASE ).prev_sample A : List[Any] = scheduler.step(SCREAMING_SNAKE_CASE , SCREAMING_SNAKE_CASE , SCREAMING_SNAKE_CASE , **SCREAMING_SNAKE_CASE ).prev_sample self.assertEqual(output_a.shape , sample.shape ) self.assertEqual(output_a.shape , output_a.shape ) def __lowerCAmelCase ( self ) -> Tuple: """simple docstring""" A : Union[str, Any] = UniPCMultistepScheduler(**self.get_scheduler_config() ) A : List[Any] = self.full_loop(scheduler=SCREAMING_SNAKE_CASE ) A : List[str] = torch.mean(torch.abs(SCREAMING_SNAKE_CASE ) ) assert abs(result_mean.item() - 0.2_464 ) < 1e-3 A : Dict = DPMSolverSinglestepScheduler.from_config(scheduler.config ) A : Optional[int] = DEISMultistepScheduler.from_config(scheduler.config ) A : List[Any] = DPMSolverMultistepScheduler.from_config(scheduler.config ) A : List[Any] = UniPCMultistepScheduler.from_config(scheduler.config ) A : Optional[Any] = self.full_loop(scheduler=SCREAMING_SNAKE_CASE ) A : Optional[Any] = torch.mean(torch.abs(SCREAMING_SNAKE_CASE ) ) assert abs(result_mean.item() - 0.2_464 ) < 1e-3 def __lowerCAmelCase ( self ) -> Union[str, Any]: """simple docstring""" for timesteps in [25, 50, 100, 999, 1000]: self.check_over_configs(num_train_timesteps=SCREAMING_SNAKE_CASE ) def __lowerCAmelCase ( self ) -> List[str]: """simple docstring""" self.check_over_configs(thresholding=SCREAMING_SNAKE_CASE ) for order in [1, 2, 3]: for solver_type in ["bh1", "bh2"]: for threshold in [0.5, 1.0, 2.0]: for prediction_type in ["epsilon", "sample"]: self.check_over_configs( thresholding=SCREAMING_SNAKE_CASE , prediction_type=SCREAMING_SNAKE_CASE , sample_max_value=SCREAMING_SNAKE_CASE , solver_order=SCREAMING_SNAKE_CASE , solver_type=SCREAMING_SNAKE_CASE , ) def __lowerCAmelCase ( self ) -> Union[str, Any]: """simple docstring""" for prediction_type in ["epsilon", "v_prediction"]: self.check_over_configs(prediction_type=SCREAMING_SNAKE_CASE ) def __lowerCAmelCase ( self ) -> List[Any]: """simple docstring""" for solver_type in ["bh1", "bh2"]: for order in [1, 2, 3]: for prediction_type in ["epsilon", "sample"]: self.check_over_configs( solver_order=SCREAMING_SNAKE_CASE , solver_type=SCREAMING_SNAKE_CASE , prediction_type=SCREAMING_SNAKE_CASE , ) A : Dict = self.full_loop( solver_order=SCREAMING_SNAKE_CASE , solver_type=SCREAMING_SNAKE_CASE , prediction_type=SCREAMING_SNAKE_CASE , ) assert not torch.isnan(SCREAMING_SNAKE_CASE ).any(), "Samples have nan numbers" def __lowerCAmelCase ( self ) -> Tuple: """simple docstring""" self.check_over_configs(lower_order_final=SCREAMING_SNAKE_CASE ) self.check_over_configs(lower_order_final=SCREAMING_SNAKE_CASE ) def __lowerCAmelCase ( self ) -> List[str]: """simple docstring""" for num_inference_steps in [1, 2, 3, 5, 10, 50, 100, 999, 1000]: self.check_over_forward(num_inference_steps=SCREAMING_SNAKE_CASE , time_step=0 ) def __lowerCAmelCase ( self ) -> List[str]: """simple docstring""" A : int = self.full_loop() A : int = torch.mean(torch.abs(SCREAMING_SNAKE_CASE ) ) assert abs(result_mean.item() - 0.2_464 ) < 1e-3 def __lowerCAmelCase ( self ) -> List[str]: """simple docstring""" A : List[Any] = self.full_loop(prediction_type='''v_prediction''' ) A : Any = torch.mean(torch.abs(SCREAMING_SNAKE_CASE ) ) assert abs(result_mean.item() - 0.1_014 ) < 1e-3 def __lowerCAmelCase ( self ) -> Union[str, Any]: """simple docstring""" A : Dict = self.scheduler_classes[0] A : List[Any] = self.get_scheduler_config(thresholding=SCREAMING_SNAKE_CASE , dynamic_thresholding_ratio=0 ) A : List[str] = scheduler_class(**SCREAMING_SNAKE_CASE ) A : Tuple = 10 A : Union[str, Any] = self.dummy_model() A : Dict = self.dummy_sample_deter.half() scheduler.set_timesteps(SCREAMING_SNAKE_CASE ) for i, t in enumerate(scheduler.timesteps ): A : Dict = model(SCREAMING_SNAKE_CASE , SCREAMING_SNAKE_CASE ) A : Optional[Any] = scheduler.step(SCREAMING_SNAKE_CASE , SCREAMING_SNAKE_CASE , SCREAMING_SNAKE_CASE ).prev_sample assert sample.dtype == torch.floataa def __lowerCAmelCase ( self , **SCREAMING_SNAKE_CASE ) -> str: """simple docstring""" for scheduler_class in self.scheduler_classes: A : Dict = self.get_scheduler_config(**SCREAMING_SNAKE_CASE ) A : Tuple = scheduler_class(**SCREAMING_SNAKE_CASE ) scheduler.set_timesteps(scheduler.config.num_train_timesteps ) assert len(scheduler.timesteps.unique() ) == scheduler.num_inference_steps
311
1
from typing import TYPE_CHECKING from ...utils import OptionalDependencyNotAvailable, _LazyModule, is_tokenizers_available, is_torch_available UpperCAmelCase__ = { "configuration_nezha": ["NEZHA_PRETRAINED_CONFIG_ARCHIVE_MAP", "NezhaConfig"], } try: if not is_torch_available(): raise OptionalDependencyNotAvailable() except OptionalDependencyNotAvailable: pass else: UpperCAmelCase__ = [ "NEZHA_PRETRAINED_MODEL_ARCHIVE_LIST", "NezhaForNextSentencePrediction", "NezhaForMaskedLM", "NezhaForPreTraining", "NezhaForMultipleChoice", "NezhaForQuestionAnswering", "NezhaForSequenceClassification", "NezhaForTokenClassification", "NezhaModel", "NezhaPreTrainedModel", ] if TYPE_CHECKING: from .configuration_nezha import NEZHA_PRETRAINED_CONFIG_ARCHIVE_MAP, NezhaConfig try: if not is_torch_available(): raise OptionalDependencyNotAvailable() except OptionalDependencyNotAvailable: pass else: from .modeling_nezha import ( NEZHA_PRETRAINED_MODEL_ARCHIVE_LIST, NezhaForMaskedLM, NezhaForMultipleChoice, NezhaForNextSentencePrediction, NezhaForPreTraining, NezhaForQuestionAnswering, NezhaForSequenceClassification, NezhaForTokenClassification, NezhaModel, NezhaPreTrainedModel, ) else: import sys UpperCAmelCase__ = _LazyModule(__name__, globals()["__file__"], _import_structure, module_spec=__spec__)
339
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 UpperCAmelCase__ = { "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 A ( _UpperCAmelCase : Optional[int] ) -> str: '''simple docstring''' _UpperCAmelCase = ['layers', 'blocks'] for k in ignore_keys: state_dict.pop(_UpperCAmelCase , _UpperCAmelCase ) UpperCAmelCase__ = { "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 A ( _UpperCAmelCase : Dict ) -> Optional[int]: '''simple docstring''' _UpperCAmelCase = list(s_dict.keys() ) for key in keys: _UpperCAmelCase = key for k, v in WHISPER_MAPPING.items(): if k in key: _UpperCAmelCase = new_key.replace(_UpperCAmelCase , _UpperCAmelCase ) print(F"{key} -> {new_key}" ) _UpperCAmelCase = s_dict.pop(_UpperCAmelCase ) return s_dict def A ( _UpperCAmelCase : List[Any] ) -> Union[str, Any]: '''simple docstring''' _UpperCAmelCase , _UpperCAmelCase = emb.weight.shape _UpperCAmelCase = nn.Linear(_UpperCAmelCase , _UpperCAmelCase , bias=_UpperCAmelCase ) _UpperCAmelCase = emb.weight.data return lin_layer def A ( _UpperCAmelCase : str , _UpperCAmelCase : str ) -> bytes: '''simple docstring''' os.makedirs(_UpperCAmelCase , exist_ok=_UpperCAmelCase ) _UpperCAmelCase = os.path.basename(_UpperCAmelCase ) _UpperCAmelCase = url.split('/' )[-2] _UpperCAmelCase = os.path.join(_UpperCAmelCase , _UpperCAmelCase ) if os.path.exists(_UpperCAmelCase ) and not os.path.isfile(_UpperCAmelCase ): raise RuntimeError(F"{download_target} exists and is not a regular file" ) if os.path.isfile(_UpperCAmelCase ): _UpperCAmelCase = open(_UpperCAmelCase , 'rb' ).read() if hashlib.shaaaa(_UpperCAmelCase ).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(_UpperCAmelCase ) as source, open(_UpperCAmelCase , 'wb' ) as output: with tqdm( total=int(source.info().get('Content-Length' ) ) , ncols=80 , unit='iB' , unit_scale=_UpperCAmelCase , unit_divisor=1_024 ) as loop: while True: _UpperCAmelCase = source.read(8_192 ) if not buffer: break output.write(_UpperCAmelCase ) loop.update(len(_UpperCAmelCase ) ) _UpperCAmelCase = open(_UpperCAmelCase , 'rb' ).read() if hashlib.shaaaa(_UpperCAmelCase ).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 A ( _UpperCAmelCase : List[Any] , _UpperCAmelCase : Any ) -> Optional[int]: '''simple docstring''' if ".pt" not in checkpoint_path: _UpperCAmelCase = _download(_MODELS[checkpoint_path] ) else: _UpperCAmelCase = torch.load(_UpperCAmelCase , map_location='cpu' ) _UpperCAmelCase = original_checkpoint['dims'] _UpperCAmelCase = original_checkpoint['model_state_dict'] _UpperCAmelCase = state_dict['decoder.token_embedding.weight'] remove_ignore_keys_(_UpperCAmelCase ) rename_keys(_UpperCAmelCase ) _UpperCAmelCase = True _UpperCAmelCase = state_dict['decoder.layers.0.fc1.weight'].shape[0] _UpperCAmelCase = WhisperConfig( vocab_size=dimensions['n_vocab'] , encoder_ffn_dim=_UpperCAmelCase , decoder_ffn_dim=_UpperCAmelCase , 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'] , ) _UpperCAmelCase = WhisperForConditionalGeneration(_UpperCAmelCase ) _UpperCAmelCase , _UpperCAmelCase = model.model.load_state_dict(_UpperCAmelCase , strict=_UpperCAmelCase ) if len(_UpperCAmelCase ) > 0 and not set(_UpperCAmelCase ) <= { "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: _UpperCAmelCase = make_linear_from_emb(model.model.decoder.embed_tokens ) else: _UpperCAmelCase = proj_out_weights model.save_pretrained(_UpperCAmelCase ) if __name__ == "__main__": UpperCAmelCase__ = 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.") UpperCAmelCase__ = parser.parse_args() convert_openai_whisper_to_tfms(args.checkpoint_path, args.pytorch_dump_folder_path)
339
1
'''simple docstring''' import argparse import torch from transformers import OpenAIGPTConfig, OpenAIGPTModel, load_tf_weights_in_openai_gpt from transformers.utils import CONFIG_NAME, WEIGHTS_NAME, logging logging.set_verbosity_info() def _SCREAMING_SNAKE_CASE ( UpperCamelCase , UpperCamelCase , UpperCamelCase ): """simple docstring""" if openai_config_file == "": lowerCAmelCase__ : str = OpenAIGPTConfig() else: lowerCAmelCase__ : Union[str, Any] = OpenAIGPTConfig.from_json_file(UpperCamelCase ) lowerCAmelCase__ : Dict = OpenAIGPTModel(UpperCamelCase ) # Load weights from numpy load_tf_weights_in_openai_gpt(UpperCamelCase , UpperCamelCase , UpperCamelCase ) # Save pytorch-model lowerCAmelCase__ : Dict = pytorch_dump_folder_path + """/""" + WEIGHTS_NAME lowerCAmelCase__ : Optional[Any] = pytorch_dump_folder_path + """/""" + CONFIG_NAME print(f"""Save PyTorch model to {pytorch_weights_dump_path}""" ) torch.save(model.state_dict() , UpperCamelCase ) print(f"""Save configuration file to {pytorch_config_dump_path}""" ) with open(UpperCamelCase , """w""" , encoding="""utf-8""" ) as f: f.write(config.to_json_string() ) if __name__ == "__main__": _lowerCAmelCase = argparse.ArgumentParser() # Required parameters parser.add_argument( '''--openai_checkpoint_folder_path''', default=None, type=str, required=True, help='''Path to the TensorFlow checkpoint path.''', ) parser.add_argument( '''--pytorch_dump_folder_path''', default=None, type=str, required=True, help='''Path to the output PyTorch model.''' ) parser.add_argument( '''--openai_config_file''', default='''''', type=str, help=( '''An optional config json file corresponding to the pre-trained OpenAI model. \n''' '''This specifies the model architecture.''' ), ) _lowerCAmelCase = parser.parse_args() convert_openai_checkpoint_to_pytorch( args.openai_checkpoint_folder_path, args.openai_config_file, args.pytorch_dump_folder_path )
356
'''simple docstring''' from __future__ import annotations # This is the precision for this function which can be altered. # It is recommended for users to keep this number greater than or equal to 10. _lowerCAmelCase = 10 def _SCREAMING_SNAKE_CASE ( UpperCamelCase , UpperCamelCase , UpperCamelCase , UpperCamelCase ): """simple docstring""" for i in range(UpperCamelCase , UpperCamelCase ): if array[i] == target: return i return -1 def _SCREAMING_SNAKE_CASE ( UpperCamelCase , UpperCamelCase ): """simple docstring""" lowerCAmelCase__ : Optional[Any] = 0 lowerCAmelCase__ : Optional[int] = len(UpperCamelCase ) while left <= right: if right - left < precision: return lin_search(UpperCamelCase , UpperCamelCase , UpperCamelCase , UpperCamelCase ) lowerCAmelCase__ : Tuple = (left + right) // 3 + 1 lowerCAmelCase__ : List[str] = 2 * (left + right) // 3 + 1 if array[one_third] == target: return one_third elif array[two_third] == target: return two_third elif target < array[one_third]: lowerCAmelCase__ : int = one_third - 1 elif array[two_third] < target: lowerCAmelCase__ : Union[str, Any] = two_third + 1 else: lowerCAmelCase__ : List[Any] = one_third + 1 lowerCAmelCase__ : List[str] = two_third - 1 else: return -1 def _SCREAMING_SNAKE_CASE ( UpperCamelCase , UpperCamelCase , UpperCamelCase , UpperCamelCase ): """simple docstring""" if left < right: if right - left < precision: return lin_search(UpperCamelCase , UpperCamelCase , UpperCamelCase , UpperCamelCase ) lowerCAmelCase__ : Dict = (left + right) // 3 + 1 lowerCAmelCase__ : Tuple = 2 * (left + right) // 3 + 1 if array[one_third] == target: return one_third elif array[two_third] == target: return two_third elif target < array[one_third]: return rec_ternary_search(UpperCamelCase , one_third - 1 , UpperCamelCase , UpperCamelCase ) elif array[two_third] < target: return rec_ternary_search(two_third + 1 , UpperCamelCase , UpperCamelCase , UpperCamelCase ) else: return rec_ternary_search(one_third + 1 , two_third - 1 , UpperCamelCase , UpperCamelCase ) else: return -1 if __name__ == "__main__": import doctest doctest.testmod() _lowerCAmelCase = input('''Enter numbers separated by comma:\n''').strip() _lowerCAmelCase = [int(item.strip()) for item in user_input.split(''',''')] assert collection == sorted(collection), F"List must be ordered.\n{collection}." _lowerCAmelCase = int(input('''Enter the number to be found in the list:\n''').strip()) _lowerCAmelCase = ite_ternary_search(collection, target) _lowerCAmelCase = rec_ternary_search(0, len(collection) - 1, collection, target) if resulta != -1: print(F"""Iterative search: {target} found at positions: {resulta}""") print(F"""Recursive search: {target} found at positions: {resulta}""") else: print('''Not found''')
184
0