upd run_2 logs

run_2/train_20221217-224651.log

@@ -1,130 +0,0 @@
-12/17/2022 22:46:51 - WARNING - __main__ - Process rank: -1, device: cuda:0, n_gpu: 1distributed training: False, 16-bits training: True
-12/17/2022 22:46:51 - INFO - __main__ - Training/evaluation parameters Seq2SeqTrainingArguments(
-_n_gpu=1,
-adafactor=False,
-adam_beta1=0.9,
-adam_beta2=0.999,
-adam_epsilon=1e-08,
-auto_find_batch_size=False,
-bf16=False,
-bf16_full_eval=False,
-data_seed=None,
-dataloader_drop_last=False,
-dataloader_num_workers=0,
-dataloader_pin_memory=True,
-ddp_bucket_cap_mb=None,
-ddp_find_unused_parameters=None,
-ddp_timeout=1800,
-debug=[],
-deepspeed=None,
-disable_tqdm=False,
-do_eval=True,
-do_predict=False,
-do_train=True,
-eval_accumulation_steps=None,
-eval_delay=0,
-eval_steps=1000,
-evaluation_strategy=steps,
-fp16=True,
-fp16_backend=auto,
-fp16_full_eval=False,
-fp16_opt_level=O1,
-fsdp=[],
-fsdp_min_num_params=0,
-fsdp_transformer_layer_cls_to_wrap=None,
-full_determinism=False,
-generation_max_length=225,
-generation_num_beams=None,
-gradient_accumulation_steps=1,
-gradient_checkpointing=True,
-greater_is_better=False,
-group_by_length=False,
-half_precision_backend=auto,
-hub_model_id=None,
-hub_private_repo=False,
-hub_strategy=every_save,
-hub_token=<HUB_TOKEN>,
-ignore_data_skip=True,
-include_inputs_for_metrics=False,
-jit_mode_eval=False,
-label_names=None,
-label_smoothing_factor=0.0,
-learning_rate=3.5e-05,
-length_column_name=length,
-load_best_model_at_end=True,
-local_rank=-1,
-log_level=passive,
-log_level_replica=passive,
-log_on_each_node=True,
-logging_dir=./runs/Dec17_22-46-51_129-213-88-66,
-logging_first_step=True,
-logging_nan_inf_filter=True,
-logging_steps=50,
-logging_strategy=steps,
-lr_scheduler_type=linear,
-max_grad_norm=1.0,
-max_steps=6000,
-metric_for_best_model=wer,
-mp_parameters=,
-no_cuda=False,
-num_train_epochs=3.0,
-optim=adamw_hf,
-optim_args=None,
-output_dir=./,
-overwrite_output_dir=False,
-past_index=-1,
-per_device_eval_batch_size=32,
-per_device_train_batch_size=64,
-predict_with_generate=True,
-prediction_loss_only=False,
-push_to_hub=False,
-push_to_hub_model_id=None,
-push_to_hub_organization=None,
-push_to_hub_token=<PUSH_TO_HUB_TOKEN>,
-ray_scope=last,
-remove_unused_columns=True,
-report_to=['tensorboard'],
-resume_from_checkpoint=None,
-run_name=./,
-save_on_each_node=False,
-save_steps=1000,
-save_strategy=steps,
-save_total_limit=None,
-seed=43,
-sharded_ddp=[],
-skip_memory_metrics=True,
-sortish_sampler=False,
-tf32=None,
-torch_compile=False,
-torch_compile_backend=None,
-torch_compile_mode=None,
-torchdynamo=None,
-tpu_metrics_debug=False,
-tpu_num_cores=None,
-use_ipex=False,
-use_legacy_prediction_loop=False,
-use_mps_device=False,
-warmup_ratio=0.0,
-warmup_steps=0,
-weight_decay=0.0,
-xpu_backend=None,
-)
-12/17/2022 22:46:51 - INFO - __main__ - Data parameters: DataTrainingArguments(dataset_name='mozilla-foundation/common_voice_11_0', dataset_config_name='be', max_train_samples=None, max_eval_samples=None, audio_column_name='audio', text_column_name='sentence', max_duration_in_seconds=30.0, min_duration_in_seconds=0.0, train_split_name='train', eval_split_name='validation', do_lower_case=False, do_remove_punctuation=False, do_normalize_eval=True, language='be', task='transcribe', shuffle_buffer_size=500, streaming_train=True, streaming_eval=False)
-12/17/2022 22:46:51 - INFO - __main__ - Model parameters: ModelArguments(model_name_or_path='ales/whisper-small-belarusian', config_name=None, tokenizer_name=None, feature_extractor_name=None, cache_dir=None, use_fast_tokenizer=True, model_revision='main', use_auth_token=True, freeze_feature_encoder=False, freeze_encoder=False, forced_decoder_ids=None, suppress_tokens=None, model_index_name='Whisper Small Belarusian')
-12/17/2022 22:46:51 - INFO - __main__ - output_dir already exists. will try to load last checkpoint.
-12/17/2022 22:46:51 - INFO - __main__ - last_checkpoint is None. will try to read from training_args.resume_from_checkpoint
-12/17/2022 22:46:51 - INFO - __main__ - last_checkpoint is None. resume_from_checkpoint is either None or not existing dir. will try to read from the model saved in the root of output_dir.
-12/17/2022 22:46:51 - INFO - __main__ - dir is not empty, but contains only: ['src', 'train_20221217-224651.log', 'train_run_2.log']. it is OK - will start training
-12/17/2022 22:46:51 - INFO - datasets.info - Loading Dataset Infos from /home/ubuntu/.cache/huggingface/modules/datasets_modules/datasets/mozilla-foundation--common_voice_11_0/f8e47235d9b4e68fa24ed71d63266a02018ccf7194b2a8c9c598a5f3ab304d9f
-12/17/2022 22:46:51 - INFO - datasets.builder - Overwrite dataset info from restored data version.
-12/17/2022 22:46:51 - INFO - datasets.info - Loading Dataset info from /home/ubuntu/.cache/huggingface/datasets/mozilla-foundation___common_voice_11_0/be/11.0.0/f8e47235d9b4e68fa24ed71d63266a02018ccf7194b2a8c9c598a5f3ab304d9f
-12/17/2022 22:46:51 - INFO - datasets.info - Loading Dataset Infos from /home/ubuntu/.cache/huggingface/modules/datasets_modules/datasets/mozilla-foundation--common_voice_11_0/f8e47235d9b4e68fa24ed71d63266a02018ccf7194b2a8c9c598a5f3ab304d9f
-12/17/2022 22:46:51 - INFO - datasets.builder - Overwrite dataset info from restored data version.
-12/17/2022 22:46:51 - INFO - datasets.info - Loading Dataset info from /home/ubuntu/.cache/huggingface/datasets/mozilla-foundation___common_voice_11_0/be/11.0.0/f8e47235d9b4e68fa24ed71d63266a02018ccf7194b2a8c9c598a5f3ab304d9f
-12/17/2022 22:46:51 - WARNING - datasets.builder - Found cached dataset common_voice_11_0 (/home/ubuntu/.cache/huggingface/datasets/mozilla-foundation___common_voice_11_0/be/11.0.0/f8e47235d9b4e68fa24ed71d63266a02018ccf7194b2a8c9c598a5f3ab304d9f)
-12/17/2022 22:46:51 - INFO - datasets.info - Loading Dataset info from /home/ubuntu/.cache/huggingface/datasets/mozilla-foundation___common_voice_11_0/be/11.0.0/f8e47235d9b4e68fa24ed71d63266a02018ccf7194b2a8c9c598a5f3ab304d9f
-12/17/2022 22:47:06 - INFO - __main__ - vectorizing dataset
-12/17/2022 22:47:06 - INFO - __main__ - will preprocess data using None processes.
-12/17/2022 22:47:08 - INFO - datasets.arrow_dataset - Caching processed dataset at /home/ubuntu/.cache/huggingface/datasets/mozilla-foundation___common_voice_11_0/be/11.0.0/f8e47235d9b4e68fa24ed71d63266a02018ccf7194b2a8c9c598a5f3ab304d9f/cache-796bd4b577aed289.arrow
-12/17/2022 23:35:22 - INFO - __main__ - will launch training and pass resume_from_checkpoint=None
-12/17/2022 23:35:22 - INFO - __main__ - ShuffleCallback. shuffling train dataset. seed: 43. dataset epoch: 0