From 27b3ff316a1be982afeecd91f28ce5aa89f803e1 Mon Sep 17 00:00:00 2001 From: Julien Plu Date: Thu, 12 Nov 2020 19:43:00 +0100 Subject: [PATCH] Try to understand and apply Sylvain's comments (#8458) --- examples/adversarial/run_hans.py | 7 +++++++ examples/bert-loses-patience/run_glue_with_pabee.py | 8 +++++++- examples/bertology/run_bertology.py | 7 +++++++ examples/contrib/legacy/run_language_modeling.py | 7 +++++++ examples/contrib/mm-imdb/run_mmimdb.py | 8 +++++++- examples/contrib/run_swag.py | 7 +++++++ examples/deebert/run_glue_deebert.py | 8 +++++++- examples/distillation/run_squad_w_distillation.py | 8 +++++++- examples/language-modeling/run_clm.py | 2 ++ examples/language-modeling/run_mlm.py | 2 ++ examples/language-modeling/run_mlm_wwm.py | 2 ++ examples/language-modeling/run_plm.py | 2 ++ examples/multiple-choice/run_multiple_choice.py | 7 +++++++ examples/multiple-choice/run_tf_multiple_choice.py | 6 ++++++ examples/question-answering/run_squad.py | 8 +++++++- examples/question-answering/run_squad_trainer.py | 7 +++++++ examples/question-answering/run_tf_squad.py | 6 ++++++ examples/seq2seq/finetune_trainer.py | 8 +++++++- examples/text-classification/run_glue.py | 2 ++ examples/text-classification/run_tf_glue.py | 6 ++++++ .../text-classification/run_tf_text_classification.py | 6 ++++++ examples/text-classification/run_xnli.py | 8 +++++++- examples/token-classification/run_ner.py | 2 ++ examples/token-classification/run_ner_old.py | 7 +++++++ examples/token-classification/run_tf_ner.py | 6 ++++++ 25 files changed, 140 insertions(+), 7 deletions(-) diff --git a/examples/adversarial/run_hans.py b/examples/adversarial/run_hans.py index 1bb6a12d15..6bde583282 100644 --- a/examples/adversarial/run_hans.py +++ b/examples/adversarial/run_hans.py @@ -23,6 +23,7 @@ from typing import Dict, List, Optional import numpy as np import torch +import transformers from transformers import ( AutoConfig, AutoModelForSequenceClassification, @@ -33,6 +34,7 @@ from transformers import ( default_data_collator, set_seed, ) +from transformers.trainer_utils import is_main_process from utils_hans import HansDataset, InputFeatures, hans_processors, hans_tasks_num_labels @@ -124,6 +126,11 @@ def main(): bool(training_args.local_rank != -1), training_args.fp16, ) + # 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", training_args) # Set seed diff --git a/examples/bert-loses-patience/run_glue_with_pabee.py b/examples/bert-loses-patience/run_glue_with_pabee.py index d20bc34e0e..37eb76a511 100755 --- a/examples/bert-loses-patience/run_glue_with_pabee.py +++ b/examples/bert-loses-patience/run_glue_with_pabee.py @@ -29,6 +29,7 @@ from torch.utils.data import DataLoader, RandomSampler, SequentialSampler, Tenso from torch.utils.data.distributed import DistributedSampler from tqdm import tqdm, trange +import transformers from pabee.modeling_pabee_albert import AlbertForSequenceClassificationWithPabee from pabee.modeling_pabee_bert import BertForSequenceClassificationWithPabee from transformers import ( @@ -44,6 +45,7 @@ from transformers import glue_compute_metrics as compute_metrics from transformers import glue_convert_examples_to_features as convert_examples_to_features from transformers import glue_output_modes as output_modes from transformers import glue_processors as processors +from transformers.trainer_utils import is_main_process try: @@ -630,7 +632,11 @@ def main(): bool(args.local_rank != -1), args.fp16, ) - + # Set the verbosity to info of the Transformers logger (on main process only): + if is_main_process(args.local_rank): + transformers.utils.logging.set_verbosity_info() + transformers.utils.logging.enable_default_handler() + transformers.utils.logging.enable_explicit_format() # Set seed set_seed(args) diff --git a/examples/bertology/run_bertology.py b/examples/bertology/run_bertology.py index 92653da4dd..340b2f2cc2 100644 --- a/examples/bertology/run_bertology.py +++ b/examples/bertology/run_bertology.py @@ -30,6 +30,7 @@ from torch.utils.data import DataLoader, SequentialSampler, Subset from torch.utils.data.distributed import DistributedSampler from tqdm import tqdm +import transformers from transformers import ( AutoConfig, AutoModelForSequenceClassification, @@ -41,6 +42,7 @@ from transformers import ( glue_processors, set_seed, ) +from transformers.trainer_utils import is_main_process logger = logging.getLogger(__name__) @@ -368,6 +370,11 @@ def main(): # Setup logging logging.basicConfig(level=logging.INFO if args.local_rank in [-1, 0] else logging.WARN) logger.info("device: {} n_gpu: {}, distributed: {}".format(args.device, args.n_gpu, bool(args.local_rank != -1))) + # Set the verbosity to info of the Transformers logger (on main process only): + if is_main_process(args.local_rank): + transformers.utils.logging.set_verbosity_info() + transformers.utils.logging.enable_default_handler() + transformers.utils.logging.enable_explicit_format() # Set seeds set_seed(args.seed) diff --git a/examples/contrib/legacy/run_language_modeling.py b/examples/contrib/legacy/run_language_modeling.py index e1b6ff0c96..3b52a52229 100644 --- a/examples/contrib/legacy/run_language_modeling.py +++ b/examples/contrib/legacy/run_language_modeling.py @@ -29,6 +29,7 @@ from typing import Optional from torch.utils.data import ConcatDataset +import transformers from transformers import ( CONFIG_MAPPING, MODEL_WITH_LM_HEAD_MAPPING, @@ -47,6 +48,7 @@ from transformers import ( TrainingArguments, set_seed, ) +from transformers.trainer_utils import is_main_process logger = logging.getLogger(__name__) @@ -219,6 +221,11 @@ def main(): bool(training_args.local_rank != -1), training_args.fp16, ) + # 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", training_args) # Set seed diff --git a/examples/contrib/mm-imdb/run_mmimdb.py b/examples/contrib/mm-imdb/run_mmimdb.py index 0ae6303134..cd2d47f6fe 100644 --- a/examples/contrib/mm-imdb/run_mmimdb.py +++ b/examples/contrib/mm-imdb/run_mmimdb.py @@ -31,6 +31,7 @@ from torch.utils.data import DataLoader, RandomSampler, SequentialSampler from torch.utils.data.distributed import DistributedSampler from tqdm import tqdm, trange +import transformers from transformers import ( WEIGHTS_NAME, AdamW, @@ -41,6 +42,7 @@ from transformers import ( MMBTForClassification, get_linear_schedule_with_warmup, ) +from transformers.trainer_utils import is_main_process from utils_mmimdb import ImageEncoder, JsonlDataset, collate_fn, get_image_transforms, get_mmimdb_labels @@ -476,7 +478,11 @@ def main(): bool(args.local_rank != -1), args.fp16, ) - + # Set the verbosity to info of the Transformers logger (on main process only): + if is_main_process(args.local_rank): + transformers.utils.logging.set_verbosity_info() + transformers.utils.logging.enable_default_handler() + transformers.utils.logging.enable_explicit_format() # Set seed set_seed(args) diff --git a/examples/contrib/run_swag.py b/examples/contrib/run_swag.py index f39041cd68..e2b49a4128 100644 --- a/examples/contrib/run_swag.py +++ b/examples/contrib/run_swag.py @@ -31,8 +31,10 @@ from torch.utils.data import DataLoader, RandomSampler, SequentialSampler, Tenso from torch.utils.data.distributed import DistributedSampler from tqdm import tqdm, trange +import transformers from transformers import WEIGHTS_NAME, AdamW, AutoConfig, AutoTokenizer, get_linear_schedule_with_warmup from transformers.modeling_auto import AutoModelForMultipleChoice +from transformers.trainer_utils import is_main_process try: @@ -620,6 +622,11 @@ def main(): bool(args.local_rank != -1), args.fp16, ) + # Set the verbosity to info of the Transformers logger (on main process only): + if is_main_process(args.local_rank): + transformers.utils.logging.set_verbosity_info() + transformers.utils.logging.enable_default_handler() + transformers.utils.logging.enable_explicit_format() # Set seed set_seed(args) diff --git a/examples/deebert/run_glue_deebert.py b/examples/deebert/run_glue_deebert.py index a215dcb1a3..8361553984 100644 --- a/examples/deebert/run_glue_deebert.py +++ b/examples/deebert/run_glue_deebert.py @@ -13,6 +13,7 @@ from torch.utils.data import DataLoader, RandomSampler, SequentialSampler, Tenso from torch.utils.data.distributed import DistributedSampler from tqdm import tqdm, trange +import transformers from src.modeling_highway_bert import DeeBertForSequenceClassification from src.modeling_highway_roberta import DeeRobertaForSequenceClassification from transformers import ( @@ -28,6 +29,7 @@ from transformers import glue_compute_metrics as compute_metrics from transformers import glue_convert_examples_to_features as convert_examples_to_features from transformers import glue_output_modes as output_modes from transformers import glue_processors as processors +from transformers.trainer_utils import is_main_process try: @@ -580,7 +582,11 @@ def main(): bool(args.local_rank != -1), args.fp16, ) - + # Set the verbosity to info of the Transformers logger (on main process only): + if is_main_process(args.local_rank): + transformers.utils.logging.set_verbosity_info() + transformers.utils.logging.enable_default_handler() + transformers.utils.logging.enable_explicit_format() # Set seed set_seed(args) diff --git a/examples/distillation/run_squad_w_distillation.py b/examples/distillation/run_squad_w_distillation.py index e1d7b75873..3a8af71267 100644 --- a/examples/distillation/run_squad_w_distillation.py +++ b/examples/distillation/run_squad_w_distillation.py @@ -30,6 +30,7 @@ from torch.utils.data import DataLoader, RandomSampler, SequentialSampler from torch.utils.data.distributed import DistributedSampler from tqdm import tqdm, trange +import transformers from transformers import ( WEIGHTS_NAME, AdamW, @@ -57,6 +58,7 @@ from transformers.data.metrics.squad_metrics import ( squad_evaluate, ) from transformers.data.processors.squad import SquadResult, SquadV1Processor, SquadV2Processor +from transformers.trainer_utils import is_main_process try: @@ -745,7 +747,11 @@ def main(): bool(args.local_rank != -1), args.fp16, ) - + # Set the verbosity to info of the Transformers logger (on main process only): + if is_main_process(args.local_rank): + transformers.utils.logging.set_verbosity_info() + transformers.utils.logging.enable_default_handler() + transformers.utils.logging.enable_explicit_format() # Set seed set_seed(args) diff --git a/examples/language-modeling/run_clm.py b/examples/language-modeling/run_clm.py index 9d9fd30b63..be75a4ee38 100644 --- a/examples/language-modeling/run_clm.py +++ b/examples/language-modeling/run_clm.py @@ -168,6 +168,8 @@ def main(): # 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", training_args) # Set seed before initializing model. diff --git a/examples/language-modeling/run_mlm.py b/examples/language-modeling/run_mlm.py index bf15b00d83..35e8a03df2 100644 --- a/examples/language-modeling/run_mlm.py +++ b/examples/language-modeling/run_mlm.py @@ -179,6 +179,8 @@ def main(): # 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", training_args) # Set seed before initializing model. diff --git a/examples/language-modeling/run_mlm_wwm.py b/examples/language-modeling/run_mlm_wwm.py index ecc4c55e7c..557282a742 100644 --- a/examples/language-modeling/run_mlm_wwm.py +++ b/examples/language-modeling/run_mlm_wwm.py @@ -186,6 +186,8 @@ def main(): # 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", training_args) # Set seed before initializing model. diff --git a/examples/language-modeling/run_plm.py b/examples/language-modeling/run_plm.py index bc1c3fd28e..5969fd98b5 100644 --- a/examples/language-modeling/run_plm.py +++ b/examples/language-modeling/run_plm.py @@ -176,6 +176,8 @@ def main(): # 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", training_args) # Set seed before initializing model. diff --git a/examples/multiple-choice/run_multiple_choice.py b/examples/multiple-choice/run_multiple_choice.py index f2147c44f0..b4b77f347e 100644 --- a/examples/multiple-choice/run_multiple_choice.py +++ b/examples/multiple-choice/run_multiple_choice.py @@ -23,6 +23,7 @@ from typing import Dict, Optional import numpy as np +import transformers from transformers import ( AutoConfig, AutoModelForMultipleChoice, @@ -33,6 +34,7 @@ from transformers import ( TrainingArguments, set_seed, ) +from transformers.trainer_utils import is_main_process from utils_multiple_choice import MultipleChoiceDataset, Split, processors @@ -115,6 +117,11 @@ def main(): bool(training_args.local_rank != -1), training_args.fp16, ) + # 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", training_args) # Set seed diff --git a/examples/multiple-choice/run_tf_multiple_choice.py b/examples/multiple-choice/run_tf_multiple_choice.py index 1eb19e32fe..9bdc4c6d03 100644 --- a/examples/multiple-choice/run_tf_multiple_choice.py +++ b/examples/multiple-choice/run_tf_multiple_choice.py @@ -33,9 +33,15 @@ from transformers import ( TFTrainingArguments, set_seed, ) +from transformers.utils import logging as hf_logging from utils_multiple_choice import Split, TFMultipleChoiceDataset, processors +hf_logging.set_verbosity_info() +hf_logging.enable_default_handler() +hf_logging.enable_explicit_format() + + logger = logging.getLogger(__name__) diff --git a/examples/question-answering/run_squad.py b/examples/question-answering/run_squad.py index 59550347c2..4063c5e5f5 100644 --- a/examples/question-answering/run_squad.py +++ b/examples/question-answering/run_squad.py @@ -29,6 +29,7 @@ from torch.utils.data import DataLoader, RandomSampler, SequentialSampler from torch.utils.data.distributed import DistributedSampler from tqdm import tqdm, trange +import transformers from transformers import ( MODEL_FOR_QUESTION_ANSWERING_MAPPING, WEIGHTS_NAME, @@ -45,6 +46,7 @@ from transformers.data.metrics.squad_metrics import ( squad_evaluate, ) from transformers.data.processors.squad import SquadResult, SquadV1Processor, SquadV2Processor +from transformers.trainer_utils import is_main_process try: @@ -712,7 +714,11 @@ def main(): bool(args.local_rank != -1), args.fp16, ) - + # Set the verbosity to info of the Transformers logger (on main process only): + if is_main_process(args.local_rank): + transformers.utils.logging.set_verbosity_info() + transformers.utils.logging.enable_default_handler() + transformers.utils.logging.enable_explicit_format() # Set seed set_seed(args) diff --git a/examples/question-answering/run_squad_trainer.py b/examples/question-answering/run_squad_trainer.py index d5fc072316..cb52f46df5 100644 --- a/examples/question-answering/run_squad_trainer.py +++ b/examples/question-answering/run_squad_trainer.py @@ -22,9 +22,11 @@ import sys from dataclasses import dataclass, field from typing import Optional +import transformers from transformers import AutoConfig, AutoModelForQuestionAnswering, AutoTokenizer, HfArgumentParser, SquadDataset from transformers import SquadDataTrainingArguments as DataTrainingArguments from transformers import Trainer, TrainingArguments +from transformers.trainer_utils import is_main_process logger = logging.getLogger(__name__) @@ -91,6 +93,11 @@ def main(): bool(training_args.local_rank != -1), training_args.fp16, ) + # 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", training_args) # Prepare Question-Answering task diff --git a/examples/question-answering/run_tf_squad.py b/examples/question-answering/run_tf_squad.py index 1382e7f033..fb712964c6 100644 --- a/examples/question-answering/run_tf_squad.py +++ b/examples/question-answering/run_tf_squad.py @@ -33,6 +33,12 @@ from transformers import ( squad_convert_examples_to_features, ) from transformers.data.processors.squad import SquadV1Processor, SquadV2Processor +from transformers.utils import logging as hf_logging + + +hf_logging.set_verbosity_info() +hf_logging.enable_default_handler() +hf_logging.enable_explicit_format() logger = logging.getLogger(__name__) diff --git a/examples/seq2seq/finetune_trainer.py b/examples/seq2seq/finetune_trainer.py index dd394365e9..a37344958c 100644 --- a/examples/seq2seq/finetune_trainer.py +++ b/examples/seq2seq/finetune_trainer.py @@ -4,10 +4,11 @@ import sys from dataclasses import dataclass, field from typing import Optional +import transformers from seq2seq_trainer import Seq2SeqTrainer from seq2seq_training_args import Seq2SeqTrainingArguments from transformers import AutoConfig, AutoModelForSeq2SeqLM, AutoTokenizer, HfArgumentParser, MBartTokenizer, set_seed -from transformers.trainer_utils import EvaluationStrategy +from transformers.trainer_utils import EvaluationStrategy, is_main_process from utils import ( Seq2SeqDataCollator, Seq2SeqDataset, @@ -131,6 +132,11 @@ def main(): bool(training_args.local_rank != -1), training_args.fp16, ) + # 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", training_args) # Set seed diff --git a/examples/text-classification/run_glue.py b/examples/text-classification/run_glue.py index 9ae039d111..d15bc5a9f1 100644 --- a/examples/text-classification/run_glue.py +++ b/examples/text-classification/run_glue.py @@ -171,6 +171,8 @@ def main(): # 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(f"Training/evaluation parameters {training_args}") # Set seed before initializing model. diff --git a/examples/text-classification/run_tf_glue.py b/examples/text-classification/run_tf_glue.py index 5477447040..538134ed6d 100644 --- a/examples/text-classification/run_tf_glue.py +++ b/examples/text-classification/run_tf_glue.py @@ -27,6 +27,12 @@ from transformers import ( glue_processors, glue_tasks_num_labels, ) +from transformers.utils import logging as hf_logging + + +hf_logging.set_verbosity_info() +hf_logging.enable_default_handler() +hf_logging.enable_explicit_format() class Split(Enum): diff --git a/examples/text-classification/run_tf_text_classification.py b/examples/text-classification/run_tf_text_classification.py index cb3b75da7b..4dfea9f894 100644 --- a/examples/text-classification/run_tf_text_classification.py +++ b/examples/text-classification/run_tf_text_classification.py @@ -21,6 +21,12 @@ from transformers import ( TFTrainer, TFTrainingArguments, ) +from transformers.utils import logging as hf_logging + + +hf_logging.set_verbosity_info() +hf_logging.enable_default_handler() +hf_logging.enable_explicit_format() def get_tfds( diff --git a/examples/text-classification/run_xnli.py b/examples/text-classification/run_xnli.py index 69103708d0..c7482814e0 100644 --- a/examples/text-classification/run_xnli.py +++ b/examples/text-classification/run_xnli.py @@ -29,6 +29,7 @@ from torch.utils.data import DataLoader, RandomSampler, SequentialSampler, Tenso from torch.utils.data.distributed import DistributedSampler from tqdm import tqdm, trange +import transformers from transformers import ( WEIGHTS_NAME, AdamW, @@ -41,6 +42,7 @@ from transformers import glue_convert_examples_to_features as convert_examples_t from transformers import xnli_compute_metrics as compute_metrics from transformers import xnli_output_modes as output_modes from transformers import xnli_processors as processors +from transformers.trainer_utils import is_main_process try: @@ -526,7 +528,11 @@ def main(): bool(args.local_rank != -1), args.fp16, ) - + # Set the verbosity to info of the Transformers logger (on main process only): + if is_main_process(args.local_rank): + transformers.utils.logging.set_verbosity_info() + transformers.utils.logging.enable_default_handler() + transformers.utils.logging.enable_explicit_format() # Set seed set_seed(args) diff --git a/examples/token-classification/run_ner.py b/examples/token-classification/run_ner.py index 7054dea78c..62cd47a0ba 100644 --- a/examples/token-classification/run_ner.py +++ b/examples/token-classification/run_ner.py @@ -163,6 +163,8 @@ def main(): # 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", training_args) # Set seed before initializing model. diff --git a/examples/token-classification/run_ner_old.py b/examples/token-classification/run_ner_old.py index a2981415f6..bcb92c5dee 100644 --- a/examples/token-classification/run_ner_old.py +++ b/examples/token-classification/run_ner_old.py @@ -25,6 +25,7 @@ import numpy as np from seqeval.metrics import accuracy_score, f1_score, precision_score, recall_score from torch import nn +import transformers from transformers import ( AutoConfig, AutoModelForTokenClassification, @@ -35,6 +36,7 @@ from transformers import ( TrainingArguments, set_seed, ) +from transformers.trainer_utils import is_main_process from utils_ner import Split, TokenClassificationDataset, TokenClassificationTask @@ -139,6 +141,11 @@ def main(): bool(training_args.local_rank != -1), training_args.fp16, ) + # 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", training_args) # Set seed diff --git a/examples/token-classification/run_tf_ner.py b/examples/token-classification/run_tf_ner.py index 27aa48e905..adb9094b9c 100644 --- a/examples/token-classification/run_tf_ner.py +++ b/examples/token-classification/run_tf_ner.py @@ -33,9 +33,15 @@ from transformers import ( TFTrainer, TFTrainingArguments, ) +from transformers.utils import logging as hf_logging from utils_ner import Split, TFTokenClassificationDataset, TokenClassificationTask +hf_logging.set_verbosity_info() +hf_logging.enable_default_handler() +hf_logging.enable_explicit_format() + + logger = logging.getLogger(__name__)