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=, 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=, 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