Remove local_rank arg from TrainingArguments (#41382)

SunMarc · web-flow · commit 0419ff881d7b · 2025-10-09T18:54:12.000+02:00
diff --git a/examples/legacy/multiple_choice/run_multiple_choice.py b/examples/legacy/multiple_choice/run_multiple_choice.py
@@ -99,18 +99,18 @@ def main():
     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,
+        level=logging.INFO if training_args.local_process_index 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.local_process_index,
         training_args.device,
         training_args.n_gpu,
-        bool(training_args.local_rank != -1),
+        bool(training_args.parallel_mode.value == "distributed"),
         training_args.fp16,
     )
     # Set the verbosity to info of the Transformers logger (on main process only):
-    if is_main_process(training_args.local_rank):
+    if is_main_process(training_args.local_process_index):
         transformers.utils.logging.set_verbosity_info()
         transformers.utils.logging.enable_default_handler()
         transformers.utils.logging.enable_explicit_format()
diff --git a/examples/legacy/question-answering/run_squad_trainer.py b/examples/legacy/question-answering/run_squad_trainer.py
@@ -80,18 +80,18 @@ def main():
     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,
+        level=logging.INFO if training_args.local_process_index 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.local_process_index,
         training_args.device,
         training_args.n_gpu,
-        bool(training_args.local_rank != -1),
+        bool(training_args.parallel_mode.value == "distributed"),
         training_args.fp16,
     )
     # Set the verbosity to info of the Transformers logger (on main process only):
-    if is_main_process(training_args.local_rank):
+    if is_main_process(training_args.local_process_index):
         transformers.utils.logging.set_verbosity_info()
         transformers.utils.logging.enable_default_handler()
         transformers.utils.logging.enable_explicit_format()
diff --git a/examples/legacy/run_language_modeling.py b/examples/legacy/run_language_modeling.py
@@ -212,18 +212,18 @@ def main():
     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,
+        level=logging.INFO if training_args.local_process_index 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.local_process_index,
         training_args.device,
         training_args.n_gpu,
-        bool(training_args.local_rank != -1),
+        bool(training_args.parallel_mode.value == "distributed"),
         training_args.fp16,
     )
     # Set the verbosity to info of the Transformers logger (on main process only):
-    if is_main_process(training_args.local_rank):
+    if is_main_process(training_args.local_process_index):
         transformers.utils.logging.set_verbosity_info()
         transformers.utils.logging.enable_default_handler()
         transformers.utils.logging.enable_explicit_format()
diff --git a/examples/legacy/seq2seq/finetune_trainer.py b/examples/legacy/seq2seq/finetune_trainer.py
@@ -171,11 +171,11 @@ def main():
     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,
+        level=logging.INFO if training_args.local_process_index 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.local_process_index,
         training_args.device,
         training_args.n_gpu,
         bool(training_args.parallel_mode == ParallelMode.DISTRIBUTED),
@@ -184,7 +184,7 @@ def main():
     transformers.utils.logging.enable_default_handler()
     transformers.utils.logging.enable_explicit_format()
     # Set the verbosity to info of the Transformers logger (on main process only):
-    if is_main_process(training_args.local_rank):
+    if is_main_process(training_args.local_process_index):
         transformers.utils.logging.set_verbosity_info()
     logger.info("Training/evaluation parameters %s", training_args)
 
diff --git a/examples/legacy/token-classification/run_ner.py b/examples/legacy/token-classification/run_ner.py
@@ -125,18 +125,18 @@ def main():
     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,
+        level=logging.INFO if training_args.local_process_index 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.local_process_index,
         training_args.device,
         training_args.n_gpu,
-        bool(training_args.local_rank != -1),
+        bool(training_args.parallel_mode.value == "distributed"),
         training_args.fp16,
     )
     # Set the verbosity to info of the Transformers logger (on main process only):
-    if is_main_process(training_args.local_rank):
+    if is_main_process(training_args.local_process_index):
         transformers.utils.logging.set_verbosity_info()
         transformers.utils.logging.enable_default_handler()
         transformers.utils.logging.enable_explicit_format()
diff --git a/examples/pytorch/audio-classification/run_audio_classification.py b/examples/pytorch/audio-classification/run_audio_classification.py
@@ -236,7 +236,7 @@ def main():
 
     # Log on each process the small summary:
     logger.warning(
-        f"Process rank: {training_args.local_rank}, device: {training_args.device}, n_gpu: {training_args.n_gpu}, "
+        f"Process rank: {training_args.local_process_index}, device: {training_args.device}, n_gpu: {training_args.n_gpu}, "
         + f"distributed training: {training_args.parallel_mode.value == 'distributed'}, 16-bits training: {training_args.fp16}"
     )
     logger.info(f"Training/evaluation parameters {training_args}")
diff --git a/examples/pytorch/contrastive-image-text/run_clip.py b/examples/pytorch/contrastive-image-text/run_clip.py
@@ -265,7 +265,7 @@ def main():
 
     # Log on each process the small summary:
     logger.warning(
-        f"Process rank: {training_args.local_rank}, device: {training_args.device}, n_gpu: {training_args.n_gpu}, "
+        f"Process rank: {training_args.local_process_index}, device: {training_args.device}, n_gpu: {training_args.n_gpu}, "
         + f"distributed training: {training_args.parallel_mode.value == 'distributed'}, 16-bits training: {training_args.fp16}"
     )
     logger.info(f"Training/evaluation parameters {training_args}")
diff --git a/examples/pytorch/image-classification/run_image_classification.py b/examples/pytorch/image-classification/run_image_classification.py
@@ -219,7 +219,7 @@ def main():
 
     # Log on each process the small summary:
     logger.warning(
-        f"Process rank: {training_args.local_rank}, device: {training_args.device}, n_gpu: {training_args.n_gpu}, "
+        f"Process rank: {training_args.local_process_index}, device: {training_args.device}, n_gpu: {training_args.n_gpu}, "
         + f"distributed training: {training_args.parallel_mode.value == 'distributed'}, 16-bits training: {training_args.fp16}"
     )
     logger.info(f"Training/evaluation parameters {training_args}")
diff --git a/examples/pytorch/image-pretraining/run_mae.py b/examples/pytorch/image-pretraining/run_mae.py
@@ -211,7 +211,7 @@ def main():
 
     # Log on each process the small summary:
     logger.warning(
-        f"Process rank: {training_args.local_rank}, device: {training_args.device}, n_gpu: {training_args.n_gpu}, "
+        f"Process rank: {training_args.local_process_index}, device: {training_args.device}, n_gpu: {training_args.n_gpu}, "
         + f"distributed training: {training_args.parallel_mode.value == 'distributed'}, 16-bits training: {training_args.fp16}"
     )
     logger.info(f"Training/evaluation parameters {training_args}")
diff --git a/examples/pytorch/image-pretraining/run_mim.py b/examples/pytorch/image-pretraining/run_mim.py
@@ -275,7 +275,7 @@ def main():
 
     # Log on each process the small summary:
     logger.warning(
-        f"Process rank: {training_args.local_rank}, device: {training_args.device}, n_gpu: {training_args.n_gpu}, "
+        f"Process rank: {training_args.local_process_index}, device: {training_args.device}, n_gpu: {training_args.n_gpu}, "
         + f"distributed training: {training_args.parallel_mode.value == 'distributed'}, 16-bits training: {training_args.fp16}"
     )
     logger.info(f"Training/evaluation parameters {training_args}")
diff --git a/examples/pytorch/instance-segmentation/run_instance_segmentation.py b/examples/pytorch/instance-segmentation/run_instance_segmentation.py
@@ -357,7 +357,7 @@ def main():
     # Setup logging and log on each process the small summary:
     setup_logging(training_args)
     logger.warning(
-        f"Process rank: {training_args.local_rank}, device: {training_args.device}, n_gpu: {training_args.n_gpu}, "
+        f"Process rank: {training_args.local_process_index}, device: {training_args.device}, n_gpu: {training_args.n_gpu}, "
         + f"distributed training: {training_args.parallel_mode.value == 'distributed'}, 16-bits training: {training_args.fp16}"
     )
     logger.info(f"Training/evaluation parameters {training_args}")
diff --git a/examples/pytorch/language-modeling/run_clm.py b/examples/pytorch/language-modeling/run_clm.py
@@ -311,7 +311,7 @@ def main():
 
     # Log on each process the small summary:
     logger.warning(
-        f"Process rank: {training_args.local_rank}, device: {training_args.device}, n_gpu: {training_args.n_gpu}, "
+        f"Process rank: {training_args.local_process_index}, device: {training_args.device}, n_gpu: {training_args.n_gpu}, "
         + f"distributed training: {training_args.parallel_mode.value == 'distributed'}, 16-bits training: {training_args.fp16}"
     )
     logger.info(f"Training/evaluation parameters {training_args}")
diff --git a/examples/pytorch/language-modeling/run_fim.py b/examples/pytorch/language-modeling/run_fim.py
@@ -338,7 +338,7 @@ def main():
 
     # Log on each process the small summary:
     logger.warning(
-        f"Process rank: {training_args.local_rank}, device: {training_args.device}, n_gpu: {training_args.n_gpu}, "
+        f"Process rank: {training_args.local_process_index}, device: {training_args.device}, n_gpu: {training_args.n_gpu}, "
         + f"distributed training: {training_args.parallel_mode.value == 'distributed'}, 16-bits training: {training_args.fp16}"
     )
     logger.info(f"Training/evaluation parameters {training_args}")
diff --git a/examples/pytorch/language-modeling/run_mlm.py b/examples/pytorch/language-modeling/run_mlm.py
@@ -283,7 +283,7 @@ def main():
 
     # Log on each process the small summary:
     logger.warning(
-        f"Process rank: {training_args.local_rank}, device: {training_args.device}, n_gpu: {training_args.n_gpu}, "
+        f"Process rank: {training_args.local_process_index}, device: {training_args.device}, n_gpu: {training_args.n_gpu}, "
         + f"distributed training: {training_args.parallel_mode.value == 'distributed'}, 16-bits training: {training_args.fp16}"
     )
     # Set the verbosity to info of the Transformers logger (on main process only):
diff --git a/examples/pytorch/language-modeling/run_plm.py b/examples/pytorch/language-modeling/run_plm.py
@@ -263,7 +263,7 @@ def main():
 
     # Log on each process the small summary:
     logger.warning(
-        f"Process rank: {training_args.local_rank}, device: {training_args.device}, n_gpu: {training_args.n_gpu}, "
+        f"Process rank: {training_args.local_process_index}, device: {training_args.device}, n_gpu: {training_args.n_gpu}, "
         + f"distributed training: {training_args.parallel_mode.value == 'distributed'}, 16-bits training: {training_args.fp16}"
     )
     logger.info(f"Training/evaluation parameters {training_args}")
diff --git a/examples/pytorch/multiple-choice/run_swag.py b/examples/pytorch/multiple-choice/run_swag.py
@@ -207,7 +207,7 @@ def main():
 
     # Log on each process the small summary:
     logger.warning(
-        f"Process rank: {training_args.local_rank}, device: {training_args.device}, n_gpu: {training_args.n_gpu}, "
+        f"Process rank: {training_args.local_process_index}, device: {training_args.device}, n_gpu: {training_args.n_gpu}, "
         + f"distributed training: {training_args.parallel_mode.value == 'distributed'}, 16-bits training: {training_args.fp16}"
     )
     logger.info(f"Training/evaluation parameters {training_args}")
diff --git a/examples/pytorch/object-detection/run_object_detection.py b/examples/pytorch/object-detection/run_object_detection.py
@@ -367,7 +367,7 @@ def main():
 
     # Log on each process the small summary:
     logger.warning(
-        f"Process rank: {training_args.local_rank}, device: {training_args.device}, n_gpu: {training_args.n_gpu}, "
+        f"Process rank: {training_args.local_process_index}, device: {training_args.device}, n_gpu: {training_args.n_gpu}, "
         + f"distributed training: {training_args.parallel_mode.value == 'distributed'}, 16-bits training: {training_args.fp16}"
     )
     logger.info(f"Training/evaluation parameters {training_args}")
diff --git a/examples/pytorch/question-answering/run_qa.py b/examples/pytorch/question-answering/run_qa.py
@@ -256,7 +256,7 @@ def main():
 
     # Log on each process the small summary:
     logger.warning(
-        f"Process rank: {training_args.local_rank}, device: {training_args.device}, n_gpu: {training_args.n_gpu}, "
+        f"Process rank: {training_args.local_process_index}, device: {training_args.device}, n_gpu: {training_args.n_gpu}, "
         + f"distributed training: {training_args.parallel_mode.value == 'distributed'}, 16-bits training: {training_args.fp16}"
     )
     logger.info(f"Training/evaluation parameters {training_args}")
diff --git a/examples/pytorch/question-answering/run_qa_beam_search.py b/examples/pytorch/question-answering/run_qa_beam_search.py
@@ -254,7 +254,7 @@ def main():
 
     # Log on each process the small summary:
     logger.warning(
-        f"Process rank: {training_args.local_rank}, device: {training_args.device}, n_gpu: {training_args.n_gpu}, "
+        f"Process rank: {training_args.local_process_index}, device: {training_args.device}, n_gpu: {training_args.n_gpu}, "
         + f"distributed training: {training_args.parallel_mode.value == 'distributed'}, 16-bits training: {training_args.fp16}"
     )
     logger.info(f"Training/evaluation parameters {training_args}")
diff --git a/examples/pytorch/question-answering/run_seq2seq_qa.py b/examples/pytorch/question-answering/run_seq2seq_qa.py
@@ -302,7 +302,7 @@ def main():
 
     # Log on each process the small summary:
     logger.warning(
-        f"Process rank: {training_args.local_rank}, device: {training_args.device}, n_gpu: {training_args.n_gpu}, "
+        f"Process rank: {training_args.local_process_index}, device: {training_args.device}, n_gpu: {training_args.n_gpu}, "
         + f"distributed training: {training_args.parallel_mode.value == 'distributed'}, 16-bits training: {training_args.fp16}"
     )
     logger.info(f"Training/evaluation parameters {training_args}")
diff --git a/examples/pytorch/semantic-segmentation/run_semantic_segmentation.py b/examples/pytorch/semantic-segmentation/run_semantic_segmentation.py
@@ -215,7 +215,7 @@ def main():
 
     # Log on each process the small summary:
     logger.warning(
-        f"Process rank: {training_args.local_rank}, device: {training_args.device}, n_gpu: {training_args.n_gpu}, "
+        f"Process rank: {training_args.local_process_index}, device: {training_args.device}, n_gpu: {training_args.n_gpu}, "
         + f"distributed training: {training_args.parallel_mode.value == 'distributed'}, 16-bits training: {training_args.fp16}"
     )
     logger.info(f"Training/evaluation parameters {training_args}")
diff --git a/examples/pytorch/speech-recognition/run_speech_recognition_ctc.py b/examples/pytorch/speech-recognition/run_speech_recognition_ctc.py
@@ -435,15 +435,15 @@ def main():
         datefmt="%m/%d/%Y %H:%M:%S",
         handlers=[logging.StreamHandler(sys.stdout)],
     )
-    logger.setLevel(logging.INFO if is_main_process(training_args.local_rank) else logging.WARN)
+    logger.setLevel(logging.INFO if is_main_process(training_args.local_process_index) else logging.WARN)
 
     # Log on each process the small summary:
     logger.warning(
-        f"Process rank: {training_args.local_rank}, device: {training_args.device}, n_gpu: {training_args.n_gpu}, "
+        f"Process rank: {training_args.local_process_index}, device: {training_args.device}, n_gpu: {training_args.n_gpu}, "
         f"distributed training: {training_args.parallel_mode.value == 'distributed'}, 16-bits training: {training_args.fp16}"
     )
     # Set the verbosity to info of the Transformers logger (on main process only):
-    if is_main_process(training_args.local_rank):
+    if is_main_process(training_args.local_process_index):
         transformers.utils.logging.set_verbosity_info()
     logger.info("Training/evaluation parameters %s", training_args)
 
@@ -726,7 +726,7 @@ def compute_metrics(pred):
     # make sure all processes wait until data is saved
     with training_args.main_process_first():
         # only the main process saves them
-        if is_main_process(training_args.local_rank):
+        if is_main_process(training_args.local_process_index):
             # save feature extractor, tokenizer and config
             feature_extractor.save_pretrained(training_args.output_dir)
             tokenizer.save_pretrained(training_args.output_dir)
diff --git a/examples/pytorch/speech-recognition/run_speech_recognition_ctc_adapter.py b/examples/pytorch/speech-recognition/run_speech_recognition_ctc_adapter.py
@@ -412,15 +412,15 @@ def main():
         datefmt="%m/%d/%Y %H:%M:%S",
         handlers=[logging.StreamHandler(sys.stdout)],
     )
-    logger.setLevel(logging.INFO if is_main_process(training_args.local_rank) else logging.WARN)
+    logger.setLevel(logging.INFO if is_main_process(training_args.local_process_index) else logging.WARN)
 
     # Log on each process the small summary:
     logger.warning(
-        f"Process rank: {training_args.local_rank}, device: {training_args.device}, n_gpu: {training_args.n_gpu}, "
+        f"Process rank: {training_args.local_process_index}, device: {training_args.device}, n_gpu: {training_args.n_gpu}, "
         f"distributed training: {training_args.parallel_mode.value == 'distributed'}, 16-bits training: {training_args.fp16}"
     )
     # Set the verbosity to info of the Transformers logger (on main process only):
-    if is_main_process(training_args.local_rank):
+    if is_main_process(training_args.local_process_index):
         transformers.utils.logging.set_verbosity_info()
     logger.info("Training/evaluation parameters %s", training_args)
 
@@ -721,7 +721,7 @@ def compute_metrics(pred):
     # make sure all processes wait until data is saved
     with training_args.main_process_first():
         # only the main process saves them
-        if is_main_process(training_args.local_rank):
+        if is_main_process(training_args.local_process_index):
             # save feature extractor, tokenizer and config
             feature_extractor.save_pretrained(training_args.output_dir)
             tokenizer.save_pretrained(training_args.output_dir)
diff --git a/examples/pytorch/speech-recognition/run_speech_recognition_seq2seq.py b/examples/pytorch/speech-recognition/run_speech_recognition_seq2seq.py
@@ -326,17 +326,17 @@ def main():
     transformers.utils.logging.enable_default_handler()
     transformers.utils.logging.enable_explicit_format()
 
-    logger.setLevel(logging.INFO if is_main_process(training_args.local_rank) else logging.WARN)
+    logger.setLevel(logging.INFO if is_main_process(training_args.local_process_index) else logging.WARN)
 
     # Log on each process the small summary:
     logger.warning(
-        f"Process rank: {training_args.local_rank}, device: {training_args.device}, n_gpu: {training_args.n_gpu}, "
+        f"Process rank: {training_args.local_process_index}, device: {training_args.device}, n_gpu: {training_args.n_gpu}, "
         f"distributed training: {training_args.parallel_mode.value == 'distributed'}, 16-bits training: {training_args.fp16}"
     )
     logger.info(f"Training/evaluation parameters {training_args}")
 
     # Set the verbosity to info of the Transformers logger (on main process only):
-    if is_main_process(training_args.local_rank):
+    if is_main_process(training_args.local_process_index):
         transformers.utils.logging.set_verbosity_info()
     logger.info("Training/evaluation parameters %s", training_args)
 
@@ -557,7 +557,7 @@ def compute_metrics(pred):
     # make sure all processes wait until data is saved
     with training_args.main_process_first():
         # only the main process saves them
-        if is_main_process(training_args.local_rank):
+        if is_main_process(training_args.local_process_index):
             # save feature extractor, tokenizer and config
             feature_extractor.save_pretrained(training_args.output_dir)
             tokenizer.save_pretrained(training_args.output_dir)
diff --git a/examples/pytorch/summarization/run_summarization.py b/examples/pytorch/summarization/run_summarization.py
@@ -356,7 +356,7 @@ def main():
 
     # Log on each process the small summary:
     logger.warning(
-        f"Process rank: {training_args.local_rank}, device: {training_args.device}, n_gpu: {training_args.n_gpu}, "
+        f"Process rank: {training_args.local_process_index}, device: {training_args.device}, n_gpu: {training_args.n_gpu}, "
         + f"distributed training: {training_args.parallel_mode.value == 'distributed'}, 16-bits training: {training_args.fp16}"
     )
     logger.info(f"Training/evaluation parameters {training_args}")
diff --git a/examples/pytorch/text-classification/run_classification.py b/examples/pytorch/text-classification/run_classification.py
diff --git a/examples/pytorch/text-classification/run_glue.py b/examples/pytorch/text-classification/run_glue.py
diff --git a/examples/pytorch/text-classification/run_xnli.py b/examples/pytorch/text-classification/run_xnli.py
diff --git a/examples/pytorch/token-classification/run_ner.py b/examples/pytorch/token-classification/run_ner.py
diff --git a/examples/pytorch/translation/run_translation.py b/examples/pytorch/translation/run_translation.py
diff --git a/src/transformers/trainer.py b/src/transformers/trainer.py
diff --git a/src/transformers/training_args.py b/src/transformers/training_args.py
diff --git a/templates/adding_a_new_example_script/{{cookiecutter.directory_name}}/run_{{cookiecutter.example_shortcut}}.py b/templates/adding_a_new_example_script/{{cookiecutter.directory_name}}/run_{{cookiecutter.example_shortcut}}.py
diff --git a/tests/sagemaker/scripts/pytorch/run_glue_model_parallelism.py b/tests/sagemaker/scripts/pytorch/run_glue_model_parallelism.py
diff --git a/tests/trainer/test_trainer_distributed.py b/tests/trainer/test_trainer_distributed.py
diff --git a/tests/trainer/test_trainer_tpu.py b/tests/trainer/test_trainer_tpu.py

Original file line number	Diff line number	Diff line change
`@@ -236,7 +236,7 @@ def main():`
`236`	`236`
`237`	`237`	`# Log on each process the small summary:`
`238`	`238`	`logger.warning(`
`239`		`- f"Process rank: {training_args.local_rank}, device: {training_args.device}, n_gpu: {training_args.n_gpu}, "`
	`239`	`+ f"Process rank: {training_args.local_process_index}, device: {training_args.device}, n_gpu: {training_args.n_gpu}, "`
`240`	`240`	`+ f"distributed training: {training_args.parallel_mode.value == 'distributed'}, 16-bits training: {training_args.fp16}"`
`241`	`241`	`)`
`242`	`242`	`logger.info(f"Training/evaluation parameters {training_args}")`
Original file line number	Diff line number	Diff line change
`@@ -265,7 +265,7 @@ def main():`
`265`	`265`
`266`	`266`	`# Log on each process the small summary:`
`267`	`267`	`logger.warning(`
`268`		`- f"Process rank: {training_args.local_rank}, device: {training_args.device}, n_gpu: {training_args.n_gpu}, "`
	`268`	`+ f"Process rank: {training_args.local_process_index}, device: {training_args.device}, n_gpu: {training_args.n_gpu}, "`
`269`	`269`	`+ f"distributed training: {training_args.parallel_mode.value == 'distributed'}, 16-bits training: {training_args.fp16}"`
`270`	`270`	`)`
`271`	`271`	`logger.info(f"Training/evaluation parameters {training_args}")`
Original file line number	Diff line number	Diff line change
`@@ -219,7 +219,7 @@ def main():`
`219`	`219`
`220`	`220`	`# Log on each process the small summary:`
`221`	`221`	`logger.warning(`
`222`		`- f"Process rank: {training_args.local_rank}, device: {training_args.device}, n_gpu: {training_args.n_gpu}, "`
	`222`	`+ f"Process rank: {training_args.local_process_index}, device: {training_args.device}, n_gpu: {training_args.n_gpu}, "`
`223`	`223`	`+ f"distributed training: {training_args.parallel_mode.value == 'distributed'}, 16-bits training: {training_args.fp16}"`
`224`	`224`	`)`
`225`	`225`	`logger.info(f"Training/evaluation parameters {training_args}")`
Original file line number	Diff line number	Diff line change
`@@ -211,7 +211,7 @@ def main():`
`211`	`211`
`212`	`212`	`# Log on each process the small summary:`
`213`	`213`	`logger.warning(`
`214`		`- f"Process rank: {training_args.local_rank}, device: {training_args.device}, n_gpu: {training_args.n_gpu}, "`
	`214`	`+ f"Process rank: {training_args.local_process_index}, device: {training_args.device}, n_gpu: {training_args.n_gpu}, "`
`215`	`215`	`+ f"distributed training: {training_args.parallel_mode.value == 'distributed'}, 16-bits training: {training_args.fp16}"`
`216`	`216`	`)`
`217`	`217`	`logger.info(f"Training/evaluation parameters {training_args}")`
Original file line number	Diff line number	Diff line change
`@@ -275,7 +275,7 @@ def main():`
`275`	`275`
`276`	`276`	`# Log on each process the small summary:`
`277`	`277`	`logger.warning(`
`278`		`- f"Process rank: {training_args.local_rank}, device: {training_args.device}, n_gpu: {training_args.n_gpu}, "`
	`278`	`+ f"Process rank: {training_args.local_process_index}, device: {training_args.device}, n_gpu: {training_args.n_gpu}, "`
`279`	`279`	`+ f"distributed training: {training_args.parallel_mode.value == 'distributed'}, 16-bits training: {training_args.fp16}"`
`280`	`280`	`)`
`281`	`281`	`logger.info(f"Training/evaluation parameters {training_args}")`