Centralize logging (#6434)

* Logging

* Style

* hf_logging > utils.logging

* Address @thomwolf's comments

* Update test

* Update src/transformers/benchmark/benchmark_utils.py
Co-authored-by: Sylvain Gugger <35901082+sgugger@users.noreply.github.com>

* Revert bad change
Co-authored-by: Sylvain Gugger <35901082+sgugger@users.noreply.github.com>

src/transformers/tokenization_dpr.py

@@ -16,15 +16,15 @@
 
 
 import collections
-import logging
 from typing import List, Optional, Union
 
 from .file_utils import add_end_docstrings, add_start_docstrings
 from .tokenization_bert import BertTokenizer, BertTokenizerFast
 from .tokenization_utils_base import BatchEncoding, TensorType
+from .utils import logging
 
 
-logger = logging.getLogger(__name__)
+logger = logging.get_logger(__name__)
 
 VOCAB_FILES_NAMES = {"vocab_file": "vocab.txt"}
 

src/transformers/tokenization_flaubert.py

@@ -15,15 +15,15 @@
 """Tokenization classes for Flaubert, based on XLM."""
 
 
-import logging
 import unicodedata
 
 import six
 
 from .tokenization_xlm import XLMTokenizer
+from .utils import logging
 
 
-logger = logging.getLogger(__name__)
+logger = logging.get_logger(__name__)
 
 VOCAB_FILES_NAMES = {
     "vocab_file": "vocab.json",

src/transformers/tokenization_gpt2.py

@@ -16,7 +16,6 @@
 
 
 import json
-import logging
 import os
 from functools import lru_cache
 
@@ -26,9 +25,10 @@ from tokenizers import ByteLevelBPETokenizer
 from .tokenization_utils import AddedToken, PreTrainedTokenizer
 from .tokenization_utils_base import BatchEncoding
 from .tokenization_utils_fast import PreTrainedTokenizerFast
+from .utils import logging
 
 
-logger = logging.getLogger(__name__)
+logger = logging.get_logger(__name__)
 
 VOCAB_FILES_NAMES = {
     "vocab_file": "vocab.json",

src/transformers/tokenization_longformer.py

@@ -13,12 +13,11 @@
 # See the License for the specific language governing permissions and
 # limitations under the License.
 
-import logging
-
 from .tokenization_roberta import RobertaTokenizer, RobertaTokenizerFast
+from .utils import logging
 
 
-logger = logging.getLogger(__name__)
+logger = logging.get_logger(__name__)
 
 
 # vocab and merges same as roberta

src/transformers/tokenization_mbart.py

@@ -13,16 +13,16 @@
 # See the License for the specific language governing permissions and
 # limitations under the License.
 
-import logging
 from typing import List, Optional
 
 from .file_utils import add_start_docstrings_to_callable
 from .tokenization_utils import BatchEncoding
 from .tokenization_utils_base import PREPARE_SEQ2SEQ_BATCH_DOCSTRING
 from .tokenization_xlm_roberta import XLMRobertaTokenizer
+from .utils import logging
 
 
-logger = logging.getLogger(__name__)
+logger = logging.get_logger(__name__)
 
 _all_mbart_models = ["facebook/mbart-large-en-ro", "facebook/mbart-large-cc25"]
 SPM_URL = "https://s3.amazonaws.com/models.huggingface.co/bert/facebook/mbart-large-en-ro/sentence.bpe.model"

src/transformers/tokenization_mobilebert.py

@@ -13,13 +13,11 @@
 # limitations under the License.
 """Tokenization classes for MobileBERT."""
 
-
-import logging
-
 from .tokenization_bert import BertTokenizer, BertTokenizerFast
+from .utils import logging
 
 
-logger = logging.getLogger(__name__)
+logger = logging.get_logger(__name__)
 
 VOCAB_FILES_NAMES = {"vocab_file": "vocab.txt"}
 

src/transformers/tokenization_openai.py

@@ -16,7 +16,6 @@
 
 
 import json
-import logging
 import os
 import re
 
@@ -25,9 +24,10 @@ from tokenizers import CharBPETokenizer
 from .tokenization_bert import BasicTokenizer
 from .tokenization_utils import PreTrainedTokenizer
 from .tokenization_utils_fast import PreTrainedTokenizerFast
+from .utils import logging
 
 
-logger = logging.getLogger(__name__)
+logger = logging.get_logger(__name__)
 
 VOCAB_FILES_NAMES = {
     "vocab_file": "vocab.json",

src/transformers/tokenization_reformer.py

@@ -15,14 +15,14 @@
 """ Tokenization class for model Reformer."""
 
 
-import logging
 import os
 from shutil import copyfile
 
 from .tokenization_utils import PreTrainedTokenizer
+from .utils import logging
 
 
-logger = logging.getLogger(__name__)
+logger = logging.get_logger(__name__)
 
 SPIECE_UNDERLINE = "▁"
 

src/transformers/tokenization_retribert.py

@@ -14,13 +14,11 @@
 # limitations under the License.
 """Tokenization classes for RetriBERT."""
 
-
-import logging
-
 from .tokenization_bert import BertTokenizer, BertTokenizerFast
+from .utils import logging
 
 
-logger = logging.getLogger(__name__)
+logger = logging.get_logger(__name__)
 
 VOCAB_FILES_NAMES = {"vocab_file": "vocab.txt"}
 

src/transformers/tokenization_roberta.py

@@ -15,16 +15,16 @@
 """Tokenization classes for RoBERTa."""
 
 
-import logging
 from typing import List, Optional
 
 from tokenizers.processors import RobertaProcessing
 
 from .tokenization_gpt2 import GPT2Tokenizer, GPT2TokenizerFast
 from .tokenization_utils import AddedToken
+from .utils import logging
 
 
-logger = logging.getLogger(__name__)
+logger = logging.get_logger(__name__)
 
 VOCAB_FILES_NAMES = {
     "vocab_file": "vocab.json",

src/transformers/tokenization_t5.py

@@ -15,7 +15,6 @@
 """ Tokenization class for model T5."""
 
 
-import logging
 import os
 import re
 import warnings
@@ -23,9 +22,10 @@ from shutil import copyfile
 from typing import List, Optional
 
 from .tokenization_utils import BatchEncoding, PreTrainedTokenizer
+from .utils import logging
 
 
-logger = logging.getLogger(__name__)
+logger = logging.get_logger(__name__)
 
 SPIECE_UNDERLINE = "▁"
 

src/transformers/tokenization_transfo_xl.py

@@ -19,7 +19,6 @@
 
 
 import glob
-import logging
 import os
 import pickle
 import re
@@ -38,13 +37,14 @@ from tokenizers.processors import BertProcessing
 from .file_utils import cached_path, is_torch_available
 from .tokenization_utils import PreTrainedTokenizer
 from .tokenization_utils_fast import PreTrainedTokenizerFast
+from .utils import logging
 
 
 if is_torch_available():
     import torch
 
 
-logger = logging.getLogger(__name__)
+logger = logging.get_logger(__name__)
 
 VOCAB_FILES_NAMES = {"pretrained_vocab_file": "vocab.bin", "vocab_file": "vocab.txt"}
 VOCAB_FILES_NAMES_FAST = {"pretrained_vocab_file": "vocab.json", "vocab_file": "vocab.json"}

src/transformers/tokenization_utils.py

@@ -17,7 +17,6 @@
 """
 
 import itertools
-import logging
 import re
 import unicodedata
 from typing import Any, Dict, List, Optional, Tuple, Union, overload
@@ -40,9 +39,10 @@ from .tokenization_utils_base import (
     TextInputPair,
     TruncationStrategy,
 )
+from .utils import logging
 
 
-logger = logging.getLogger(__name__)
+logger = logging.get_logger(__name__)
 
 
 def _is_whitespace(char):

src/transformers/tokenization_utils_base.py

@@ -20,7 +20,6 @@
 
 import copy
 import json
-import logging
 import os
 import warnings
 from collections import OrderedDict, UserDict
@@ -41,6 +40,7 @@ from .file_utils import (
     is_torch_available,
     torch_required,
 )
+from .utils import logging
 
 
 if is_tf_available():
@@ -49,7 +49,7 @@ if is_torch_available():
     import torch
 
 
-logger = logging.getLogger(__name__)
+logger = logging.get_logger(__name__)
 
 VERY_LARGE_INTEGER = int(1e30)  # This is used to set the max input length for a model with infinite size input
 LARGE_INTEGER = int(1e20)  # This is used when we need something big but slightly smaller than VERY_LARGE_INTEGER

src/transformers/tokenization_utils_fast.py

@@ -16,7 +16,6 @@
     For slow (python) tokenizers see tokenization_utils.py
 """
 
-import logging
 import os
 from collections import defaultdict
 from typing import Any, Dict, List, Optional, Tuple, Union
@@ -38,9 +37,10 @@ from .tokenization_utils_base import (
     TextInputPair,
     TruncationStrategy,
 )
+from .utils import logging
 
 
-logger = logging.getLogger(__name__)
+logger = logging.get_logger(__name__)
 
 
 @add_end_docstrings(

src/transformers/tokenization_xlm.py

@@ -16,7 +16,6 @@
 
 
 import json
-import logging
 import os
 import re
 import sys
@@ -26,9 +25,10 @@ from typing import List, Optional
 import sacremoses as sm
 
 from .tokenization_utils import PreTrainedTokenizer
+from .utils import logging
 
 
-logger = logging.getLogger(__name__)
+logger = logging.get_logger(__name__)
 
 VOCAB_FILES_NAMES = {
     "vocab_file": "vocab.json",

src/transformers/tokenization_xlm_roberta.py

@@ -15,16 +15,16 @@
 """ Tokenization classes for XLM-RoBERTa model."""
 
 
-import logging
 import os
 from shutil import copyfile
 from typing import List, Optional
 
 from .tokenization_utils import PreTrainedTokenizer
 from .tokenization_xlnet import SPIECE_UNDERLINE
+from .utils import logging
 
 
-logger = logging.getLogger(__name__)
+logger = logging.get_logger(__name__)
 
 VOCAB_FILES_NAMES = {"vocab_file": "sentencepiece.bpe.model"}
 

src/transformers/tokenization_xlnet.py

@@ -15,16 +15,16 @@
 """ Tokenization classes for XLNet model."""
 
 
-import logging
 import os
 import unicodedata
 from shutil import copyfile
 from typing import List, Optional
 
 from .tokenization_utils import PreTrainedTokenizer
+from .utils import logging
 
 
-logger = logging.getLogger(__name__)
+logger = logging.get_logger(__name__)
 
 VOCAB_FILES_NAMES = {"vocab_file": "spiece.model"}
 

src/transformers/trainer.py

 import inspect
-import logging
 import math
 import os
 import re
@@ -44,6 +43,7 @@ from .trainer_utils import (
     set_seed,
 )
 from .training_args import TrainingArguments
+from .utils import logging
 
 
 _use_native_amp = False
@@ -86,7 +86,7 @@ if is_optuna_available():
 if is_ray_available():
     from ray import tune
 
-logger = logging.getLogger(__name__)
+logger = logging.get_logger(__name__)
 
 
 @contextmanager

src/transformers/trainer_tf.py

 """Tensorflow trainer class."""
 
 import datetime
-import logging
 import math
 import os
 import warnings
@@ -16,6 +15,7 @@ from .modeling_tf_utils import TFPreTrainedModel
 from .optimization_tf import GradientAccumulator, create_optimizer
 from .trainer_utils import PREFIX_CHECKPOINT_DIR, EvalPrediction, PredictionOutput, set_seed
 from .training_args_tf import TFTrainingArguments
+from .utils import logging
 
 
 if is_wandb_available():
@@ -24,7 +24,7 @@ if is_wandb_available():
 if is_comet_available():
     import comet_ml
 
-logger = logging.getLogger(__name__)
+logger = logging.get_logger(__name__)
 
 
 class TFTrainer: