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/data/datasets/glue.py

@@ -1,4 +1,3 @@
-import logging
 import os
 import time
 from dataclasses import dataclass, field
@@ -14,11 +13,12 @@ from ...tokenization_bart import BartTokenizer, BartTokenizerFast
 from ...tokenization_roberta import RobertaTokenizer, RobertaTokenizerFast
 from ...tokenization_utils import PreTrainedTokenizer
 from ...tokenization_xlm_roberta import XLMRobertaTokenizer
+from ...utils import logging
 from ..processors.glue import glue_convert_examples_to_features, glue_output_modes, glue_processors
 from ..processors.utils import InputFeatures
 
 
-logger = logging.getLogger(__name__)
+logger = logging.get_logger(__name__)
 
 
 @dataclass

src/transformers/data/datasets/language_modeling.py

@@ -1,4 +1,3 @@
-import logging
 import os
 import pickle
 import time
@@ -9,9 +8,10 @@ from torch.utils.data.dataset import Dataset
 from filelock import FileLock
 
 from ...tokenization_utils import PreTrainedTokenizer
+from ...utils import logging
 
 
-logger = logging.getLogger(__name__)
+logger = logging.get_logger(__name__)
 
 
 class TextDataset(Dataset):

src/transformers/data/datasets/squad.py

@@ -1,4 +1,3 @@
-import logging
 import os
 import time
 from dataclasses import dataclass, field
@@ -12,10 +11,11 @@ from filelock import FileLock
 
 from ...modeling_auto import MODEL_FOR_QUESTION_ANSWERING_MAPPING
 from ...tokenization_utils import PreTrainedTokenizer
+from ...utils import logging
 from ..processors.squad import SquadFeatures, SquadV1Processor, SquadV2Processor, squad_convert_examples_to_features
 
 
-logger = logging.getLogger(__name__)
+logger = logging.get_logger(__name__)
 
 MODEL_CONFIG_CLASSES = list(MODEL_FOR_QUESTION_ANSWERING_MAPPING.keys())
 MODEL_TYPES = tuple(conf.model_type for conf in MODEL_CONFIG_CLASSES)

src/transformers/data/metrics/squad_metrics.py

@@ -10,15 +10,16 @@ that a question is unanswerable.
 
 import collections
 import json
-import logging
 import math
 import re
 import string
 
 from transformers.tokenization_bert import BasicTokenizer
 
+from ...utils import logging
 
-logger = logging.getLogger(__name__)
+
+logger = logging.get_logger(__name__)
 
 
 def normalize_answer(s):

src/transformers/data/processors/glue.py

@@ -15,7 +15,6 @@
 # limitations under the License.
 """ GLUE processors and helpers """
 
-import logging
 import os
 from dataclasses import asdict
 from enum import Enum
@@ -23,13 +22,14 @@ from typing import List, Optional, Union
 
 from ...file_utils import is_tf_available
 from ...tokenization_utils import PreTrainedTokenizer
+from ...utils import logging
 from .utils import DataProcessor, InputExample, InputFeatures
 
 
 if is_tf_available():
     import tensorflow as tf
 
-logger = logging.getLogger(__name__)
+logger = logging.get_logger(__name__)
 
 
 def glue_convert_examples_to_features(

src/transformers/data/processors/squad.py

@@ -1,5 +1,4 @@
 import json
-import logging
 import os
 from functools import partial
 from multiprocessing import Pool, cpu_count
@@ -10,6 +9,7 @@ from tqdm import tqdm
 from ...file_utils import is_tf_available, is_torch_available
 from ...tokenization_bert import whitespace_tokenize
 from ...tokenization_utils_base import TruncationStrategy
+from ...utils import logging
 from .utils import DataProcessor
 
 
@@ -24,7 +24,7 @@ if is_torch_available():
 if is_tf_available():
     import tensorflow as tf
 
-logger = logging.getLogger(__name__)
+logger = logging.get_logger(__name__)
 
 
 def _improve_answer_span(doc_tokens, input_start, input_end, tokenizer, orig_answer_text):

src/transformers/data/processors/utils.py

@@ -17,14 +17,14 @@
 import csv
 import dataclasses
 import json
-import logging
 from dataclasses import dataclass
 from typing import List, Optional, Union
 
 from ...file_utils import is_tf_available, is_torch_available
+from ...utils import logging
 
 
-logger = logging.getLogger(__name__)
+logger = logging.get_logger(__name__)
 
 
 @dataclass

src/transformers/data/processors/xnli.py

@@ -16,13 +16,13 @@
 """ XNLI utils (dataset loading and evaluation) """
 
 
-import logging
 import os
 
+from ...utils import logging
 from .utils import DataProcessor, InputExample
 
 
-logger = logging.getLogger(__name__)
+logger = logging.get_logger(__name__)
 
 
 class XnliProcessor(DataProcessor):