Ignore wanb logs

.gitignore

@@ -1,3 +1,4 @@
 checkpoint-*/
 nohup.out
 .venv/
+wandb/

wandb/debug-internal.log

@@ -1,7 +0,0 @@
-{"time":"2025-02-12T15:27:10.115999744Z","level":"INFO","msg":"stream: starting","core version":"0.19.6","symlink path":"/home/tknika/xezpeleta/whisper/whisper-small-eu/wandb/run-20250212_152709-lejyafmi/logs/debug-core.log"}
-{"time":"2025-02-12T15:27:10.219975741Z","level":"INFO","msg":"created new stream","id":"lejyafmi"}
-{"time":"2025-02-12T15:27:10.220016671Z","level":"INFO","msg":"stream: started","id":"lejyafmi"}
-{"time":"2025-02-12T15:27:10.220140679Z","level":"INFO","msg":"writer: Do: started","stream_id":"lejyafmi"}
-{"time":"2025-02-12T15:27:10.220197299Z","level":"INFO","msg":"handler: started","stream_id":"lejyafmi"}
-{"time":"2025-02-12T15:27:10.220285178Z","level":"INFO","msg":"sender: started","stream_id":"lejyafmi"}
-{"time":"2025-02-12T15:27:10.587185852Z","level":"INFO","msg":"Starting system monitor"}

wandb/debug.log

@@ -1,25 +0,0 @@
-2025-02-12 15:27:09,901 INFO    MainThread:243546 [wandb_setup.py:_flush():68] Current SDK version is 0.19.6
-2025-02-12 15:27:09,901 INFO    MainThread:243546 [wandb_setup.py:_flush():68] Configure stats pid to 243546
-2025-02-12 15:27:09,901 INFO    MainThread:243546 [wandb_setup.py:_flush():68] Loading settings from /home/tknika/.config/wandb/settings
-2025-02-12 15:27:09,901 INFO    MainThread:243546 [wandb_setup.py:_flush():68] Loading settings from /home/tknika/xezpeleta/whisper/whisper-small-eu/wandb/settings
-2025-02-12 15:27:09,901 INFO    MainThread:243546 [wandb_setup.py:_flush():68] Loading settings from environment variables
-2025-02-12 15:27:09,901 INFO    MainThread:243546 [wandb_init.py:setup_run_log_directory():637] Logging user logs to /home/tknika/xezpeleta/whisper/whisper-small-eu/wandb/run-20250212_152709-lejyafmi/logs/debug.log
-2025-02-12 15:27:09,901 INFO    MainThread:243546 [wandb_init.py:setup_run_log_directory():638] Logging internal logs to /home/tknika/xezpeleta/whisper/whisper-small-eu/wandb/run-20250212_152709-lejyafmi/logs/debug-internal.log
-2025-02-12 15:27:09,901 INFO    MainThread:243546 [wandb_init.py:init():756] calling init triggers
-2025-02-12 15:27:09,901 INFO    MainThread:243546 [wandb_init.py:init():761] wandb.init called with sweep_config: {}
-config: {'_wandb': {}}
-2025-02-12 15:27:09,901 INFO    MainThread:243546 [wandb_init.py:init():789] starting backend
-2025-02-12 15:27:10,107 INFO    MainThread:243546 [wandb_init.py:init():793] sending inform_init request
-2025-02-12 15:27:10,112 INFO    MainThread:243546 [backend.py:_multiprocessing_setup():97] multiprocessing start_methods=fork,spawn,forkserver, using: spawn
-2025-02-12 15:27:10,113 INFO    MainThread:243546 [wandb_init.py:init():808] backend started and connected
-2025-02-12 15:27:10,115 INFO    MainThread:243546 [wandb_init.py:init():901] updated telemetry
-2025-02-12 15:27:10,122 INFO    MainThread:243546 [wandb_init.py:init():936] communicating run to backend with 90.0 second timeout
-2025-02-12 15:27:10,584 INFO    MainThread:243546 [wandb_init.py:init():994] starting run threads in backend
-2025-02-12 15:27:10,691 INFO    MainThread:243546 [wandb_run.py:_console_start():2385] atexit reg
-2025-02-12 15:27:10,692 INFO    MainThread:243546 [wandb_run.py:_redirect():2235] redirect: wrap_raw
-2025-02-12 15:27:10,692 INFO    MainThread:243546 [wandb_run.py:_redirect():2300] Wrapping output streams.
-2025-02-12 15:27:10,692 INFO    MainThread:243546 [wandb_run.py:_redirect():2325] Redirects installed.
-2025-02-12 15:27:10,694 INFO    MainThread:243546 [wandb_init.py:init():1036] run started, returning control to user process
-2025-02-12 15:27:10,698 INFO    MainThread:243546 [wandb_run.py:_config_callback():1253] config_cb None None {'vocab_size': 51865, 'num_mel_bins': 80, 'd_model': 768, 'encoder_layers': 12, 'encoder_attention_heads': 12, 'decoder_layers': 12, 'decoder_attention_heads': 12, 'decoder_ffn_dim': 3072, 'encoder_ffn_dim': 3072, 'dropout': 0.0, 'attention_dropout': 0.0, 'activation_dropout': 0.0, 'activation_function': 'gelu', 'init_std': 0.02, 'encoder_layerdrop': 0.0, 'decoder_layerdrop': 0.0, 'use_cache': False, 'num_hidden_layers': 12, 'scale_embedding': False, 'max_source_positions': 1500, 'max_target_positions': 448, 'classifier_proj_size': 256, 'use_weighted_layer_sum': False, 'apply_spec_augment': False, 'mask_time_prob': 0.05, 'mask_time_length': 10, 'mask_time_min_masks': 2, 'mask_feature_prob': 0.0, 'mask_feature_length': 10, 'mask_feature_min_masks': 0, 'median_filter_width': 7, 'return_dict': True, 'output_hidden_states': False, 'output_attentions': False, 'torchscript': False, 'torch_dtype': 'float32', 'use_bfloat16': False, 'tf_legacy_loss': False, 'pruned_heads': {}, 'tie_word_embeddings': True, 'chunk_size_feed_forward': 0, 'is_encoder_decoder': True, 'is_decoder': False, 'cross_attention_hidden_size': None, 'add_cross_attention': False, 'tie_encoder_decoder': False, 'max_length': 448, 'min_length': 0, 'do_sample': False, 'early_stopping': False, 'num_beams': 1, 'num_beam_groups': 1, 'diversity_penalty': 0.0, 'temperature': 1.0, 'top_k': 50, 'top_p': 1.0, 'typical_p': 1.0, 'repetition_penalty': 1.0, 'length_penalty': 1.0, 'no_repeat_ngram_size': 0, 'encoder_no_repeat_ngram_size': 0, 'bad_words_ids': None, 'num_return_sequences': 1, 'output_scores': False, 'return_dict_in_generate': False, 'forced_bos_token_id': None, 'forced_eos_token_id': None, 'remove_invalid_values': False, 'exponential_decay_length_penalty': None, 'suppress_tokens': None, 'begin_suppress_tokens': [220, 50257], 'architectures': ['WhisperForConditionalGeneration'], 'finetuning_task': None, 'id2label': {0: 'LABEL_0', 1: 'LABEL_1'}, 'label2id': {'LABEL_0': 0, 'LABEL_1': 1}, 'tokenizer_class': None, 'prefix': None, 'bos_token_id': 50257, 'pad_token_id': 50257, 'eos_token_id': 50257, 'sep_token_id': None, 'decoder_start_token_id': 50258, 'task_specific_params': None, 'problem_type': None, '_name_or_path': 'openai/whisper-small', '_attn_implementation_autoset': True, 'transformers_version': '4.49.0.dev0', 'forced_decoder_ids': None, 'model_type': 'whisper', 'output_dir': './', 'overwrite_output_dir': True, 'do_train': True, 'do_eval': True, 'do_predict': False, 'eval_strategy': 'steps', 'prediction_loss_only': False, 'per_device_train_batch_size': 32, 'per_device_eval_batch_size': 16, 'per_gpu_train_batch_size': None, 'per_gpu_eval_batch_size': None, 'gradient_accumulation_steps': 1, 'eval_accumulation_steps': None, 'eval_delay': 0, 'torch_empty_cache_steps': None, 'learning_rate': 1e-05, 'weight_decay': 0.0, 'adam_beta1': 0.9, 'adam_beta2': 0.999, 'adam_epsilon': 1e-08, 'max_grad_norm': 1.0, 'num_train_epochs': 3.0, 'max_steps': 8000, 'lr_scheduler_type': 'linear', 'lr_scheduler_kwargs': {}, 'warmup_ratio': 0.0, 'warmup_steps': 500, 'log_level': 'passive', 'log_level_replica': 'warning', 'log_on_each_node': True, 'logging_dir': './runs/Feb12_15-26-19_tknika', 'logging_strategy': 'steps', 'logging_first_step': False, 'logging_steps': 25, 'logging_nan_inf_filter': True, 'save_strategy': 'steps', 'save_steps': 1000, 'save_total_limit': None, 'save_safetensors': True, 'save_on_each_node': False, 'save_only_model': False, 'restore_callback_states_from_checkpoint': False, 'no_cuda': False, 'use_cpu': False, 'use_mps_device': False, 'seed': 42, 'data_seed': None, 'jit_mode_eval': False, 'use_ipex': False, 'bf16': False, 'fp16': True, 'fp16_opt_level': 'O1', 'half_precision_backend': 'auto', 'bf16_full_eval': False, 'fp16_full_eval': False, 'tf32': None, 'local_rank': 0, 'ddp_backend': None, 'tpu_num_cores': None, 'tpu_metrics_debug': False, 'debug': [], 'dataloader_drop_last': False, 'eval_steps': 1000, 'dataloader_num_workers': 0, 'dataloader_prefetch_factor': None, 'past_index': -1, 'run_name': 'whisper-small-eu', 'disable_tqdm': False, 'remove_unused_columns': True, 'label_names': None, 'load_best_model_at_end': True, 'metric_for_best_model': 'wer', 'greater_is_better': False, 'ignore_data_skip': False, 'fsdp': [], 'fsdp_min_num_params': 0, 'fsdp_config': {'min_num_params': 0, 'xla': False, 'xla_fsdp_v2': False, 'xla_fsdp_grad_ckpt': False}, 'fsdp_transformer_layer_cls_to_wrap': None, 'accelerator_config': {'split_batches': False, 'dispatch_batches': None, 'even_batches': True, 'use_seedable_sampler': True, 'non_blocking': False, 'gradient_accumulation_kwargs': None}, 'deepspeed': None, 'label_smoothing_factor': 0.0, 'optim': 'adamw_torch', 'optim_args': None, 'adafactor': False, 'group_by_length': False, 'length_column_name': 'input_length', 'report_to': ['wandb'], 'ddp_find_unused_parameters': None, 'ddp_bucket_cap_mb': None, 'ddp_broadcast_buffers': None, 'dataloader_pin_memory': True, 'dataloader_persistent_workers': False, 'skip_memory_metrics': True, 'use_legacy_prediction_loop': False, 'push_to_hub': True, 'resume_from_checkpoint': None, 'hub_model_id': None, 'hub_strategy': 'every_save', 'hub_token': '<HUB_TOKEN>', 'hub_private_repo': None, 'hub_always_push': False, 'gradient_checkpointing': True, 'gradient_checkpointing_kwargs': None, 'include_inputs_for_metrics': False, 'include_for_metrics': [], 'eval_do_concat_batches': True, 'fp16_backend': 'auto', 'evaluation_strategy': 'steps', 'push_to_hub_model_id': None, 'push_to_hub_organization': None, 'push_to_hub_token': '<PUSH_TO_HUB_TOKEN>', 'mp_parameters': '', 'auto_find_batch_size': False, 'full_determinism': False, 'torchdynamo': None, 'ray_scope': 'last', 'ddp_timeout': 1800, 'torch_compile': False, 'torch_compile_backend': None, 'torch_compile_mode': None, 'dispatch_batches': None, 'split_batches': None, 'include_tokens_per_second': False, 'include_num_input_tokens_seen': False, 'neftune_noise_alpha': None, 'optim_target_modules': None, 'batch_eval_metrics': False, 'eval_on_start': False, 'use_liger_kernel': False, 'eval_use_gather_object': False, 'average_tokens_across_devices': False, 'sortish_sampler': False, 'predict_with_generate': True, 'generation_max_length': 225, 'generation_num_beams': None, 'generation_config': None}
-2025-02-12 15:27:10,704 INFO    MainThread:243546 [wandb_config.py:__setitem__():154] config set model/num_parameters = 241734912 - <bound method Run._config_callback of <wandb.sdk.wandb_run.Run object at 0x7eb0a4c1e180>>
-2025-02-12 15:27:10,704 INFO    MainThread:243546 [wandb_run.py:_config_callback():1253] config_cb model/num_parameters 241734912 None

wandb/run-20250212_121751-d4i88lzt/files/config.yaml

@@ -1,512 +0,0 @@
-_attn_implementation_autoset:
-    value: true
-_name_or_path:
-    value: openai/whisper-small
-_wandb:
-    value:
-        cli_version: 0.19.6
-        m:
-            - "1": train/global_step
-              "6":
-                - 3
-              "7": []
-        python_version: 3.12.3
-        t:
-            "1":
-                - 1
-                - 5
-                - 11
-                - 49
-                - 51
-                - 53
-                - 55
-                - 71
-                - 100
-            "2":
-                - 1
-                - 5
-                - 11
-                - 49
-                - 51
-                - 53
-                - 55
-                - 71
-                - 100
-            "3":
-                - 7
-                - 13
-                - 19
-                - 23
-                - 55
-                - 66
-            "4": 3.12.3
-            "5": 0.19.6
-            "6": 4.49.0.dev0
-            "8":
-                - 5
-            "9":
-                "1": transformers_trainer
-            "12": 0.19.6
-            "13": linux-x86_64
-accelerator_config:
-    value:
-        dispatch_batches: null
-        even_batches: true
-        gradient_accumulation_kwargs: null
-        non_blocking: false
-        split_batches: false
-        use_seedable_sampler: true
-activation_dropout:
-    value: 0
-activation_function:
-    value: gelu
-adafactor:
-    value: false
-adam_beta1:
-    value: 0.9
-adam_beta2:
-    value: 0.999
-adam_epsilon:
-    value: 1e-08
-add_cross_attention:
-    value: false
-apply_spec_augment:
-    value: false
-architectures:
-    value:
-        - WhisperForConditionalGeneration
-attention_dropout:
-    value: 0
-auto_find_batch_size:
-    value: false
-average_tokens_across_devices:
-    value: false
-bad_words_ids:
-    value: null
-batch_eval_metrics:
-    value: false
-begin_suppress_tokens:
-    value:
-        - 220
-        - 50257
-bf16:
-    value: false
-bf16_full_eval:
-    value: false
-bos_token_id:
-    value: 50257
-chunk_size_feed_forward:
-    value: 0
-classifier_proj_size:
-    value: 256
-cross_attention_hidden_size:
-    value: null
-d_model:
-    value: 768
-data_seed:
-    value: null
-dataloader_drop_last:
-    value: false
-dataloader_num_workers:
-    value: 0
-dataloader_persistent_workers:
-    value: false
-dataloader_pin_memory:
-    value: true
-dataloader_prefetch_factor:
-    value: null
-ddp_backend:
-    value: null
-ddp_broadcast_buffers:
-    value: null
-ddp_bucket_cap_mb:
-    value: null
-ddp_find_unused_parameters:
-    value: null
-ddp_timeout:
-    value: 1800
-debug:
-    value: []
-decoder_attention_heads:
-    value: 12
-decoder_ffn_dim:
-    value: 3072
-decoder_layerdrop:
-    value: 0
-decoder_layers:
-    value: 12
-decoder_start_token_id:
-    value: 50258
-deepspeed:
-    value: null
-disable_tqdm:
-    value: false
-dispatch_batches:
-    value: null
-diversity_penalty:
-    value: 0
-do_eval:
-    value: true
-do_predict:
-    value: false
-do_sample:
-    value: false
-do_train:
-    value: true
-dropout:
-    value: 0
-early_stopping:
-    value: false
-encoder_attention_heads:
-    value: 12
-encoder_ffn_dim:
-    value: 3072
-encoder_layerdrop:
-    value: 0
-encoder_layers:
-    value: 12
-encoder_no_repeat_ngram_size:
-    value: 0
-eos_token_id:
-    value: 50257
-eval_accumulation_steps:
-    value: null
-eval_delay:
-    value: 0
-eval_do_concat_batches:
-    value: true
-eval_on_start:
-    value: false
-eval_steps:
-    value: 1000
-eval_strategy:
-    value: steps
-eval_use_gather_object:
-    value: false
-evaluation_strategy:
-    value: steps
-exponential_decay_length_penalty:
-    value: null
-finetuning_task:
-    value: null
-forced_bos_token_id:
-    value: null
-forced_decoder_ids:
-    value: null
-forced_eos_token_id:
-    value: null
-fp16:
-    value: true
-fp16_backend:
-    value: auto
-fp16_full_eval:
-    value: false
-fp16_opt_level:
-    value: O1
-fsdp:
-    value: []
-fsdp_config:
-    value:
-        min_num_params: 0
-        xla: false
-        xla_fsdp_grad_ckpt: false
-        xla_fsdp_v2: false
-fsdp_min_num_params:
-    value: 0
-fsdp_transformer_layer_cls_to_wrap:
-    value: null
-full_determinism:
-    value: false
-generation_config:
-    value: null
-generation_max_length:
-    value: 225
-generation_num_beams:
-    value: null
-gradient_accumulation_steps:
-    value: 1
-gradient_checkpointing:
-    value: true
-gradient_checkpointing_kwargs:
-    value: null
-greater_is_better:
-    value: false
-group_by_length:
-    value: false
-half_precision_backend:
-    value: auto
-hub_always_push:
-    value: false
-hub_model_id:
-    value: null
-hub_private_repo:
-    value: null
-hub_strategy:
-    value: every_save
-hub_token:
-    value: <HUB_TOKEN>
-id2label:
-    value:
-        "0": LABEL_0
-        "1": LABEL_1
-ignore_data_skip:
-    value: false
-include_for_metrics:
-    value: []
-include_inputs_for_metrics:
-    value: false
-include_num_input_tokens_seen:
-    value: false
-include_tokens_per_second:
-    value: false
-init_std:
-    value: 0.02
-is_decoder:
-    value: false
-is_encoder_decoder:
-    value: true
-jit_mode_eval:
-    value: false
-label_names:
-    value: null
-label_smoothing_factor:
-    value: 0
-label2id:
-    value:
-        LABEL_0: 0
-        LABEL_1: 1
-learning_rate:
-    value: 1e-05
-length_column_name:
-    value: input_length
-length_penalty:
-    value: 1
-load_best_model_at_end:
-    value: true
-local_rank:
-    value: 0
-log_level:
-    value: passive
-log_level_replica:
-    value: warning
-log_on_each_node:
-    value: true
-logging_dir:
-    value: ./runs/Feb12_12-17-27_tknika
-logging_first_step:
-    value: false
-logging_nan_inf_filter:
-    value: true
-logging_steps:
-    value: 25
-logging_strategy:
-    value: steps
-lr_scheduler_type:
-    value: linear
-mask_feature_length:
-    value: 10
-mask_feature_min_masks:
-    value: 0
-mask_feature_prob:
-    value: 0
-mask_time_length:
-    value: 10
-mask_time_min_masks:
-    value: 2
-mask_time_prob:
-    value: 0.05
-max_grad_norm:
-    value: 1
-max_length:
-    value: 448
-max_source_positions:
-    value: 1500
-max_steps:
-    value: 8000
-max_target_positions:
-    value: 448
-median_filter_width:
-    value: 7
-metric_for_best_model:
-    value: wer
-min_length:
-    value: 0
-model/num_parameters:
-    value: 241734912
-model_type:
-    value: whisper
-mp_parameters:
-    value: ""
-neftune_noise_alpha:
-    value: null
-no_cuda:
-    value: false
-no_repeat_ngram_size:
-    value: 0
-num_beam_groups:
-    value: 1
-num_beams:
-    value: 1
-num_hidden_layers:
-    value: 12
-num_mel_bins:
-    value: 80
-num_return_sequences:
-    value: 1
-num_train_epochs:
-    value: 3
-optim:
-    value: adamw_torch
-optim_args:
-    value: null
-optim_target_modules:
-    value: null
-output_attentions:
-    value: false
-output_dir:
-    value: ./
-output_hidden_states:
-    value: false
-output_scores:
-    value: false
-overwrite_output_dir:
-    value: true
-pad_token_id:
-    value: 50257
-past_index:
-    value: -1
-per_device_eval_batch_size:
-    value: 16
-per_device_train_batch_size:
-    value: 32
-per_gpu_eval_batch_size:
-    value: null
-per_gpu_train_batch_size:
-    value: null
-predict_with_generate:
-    value: true
-prediction_loss_only:
-    value: false
-prefix:
-    value: null
-problem_type:
-    value: null
-push_to_hub:
-    value: true
-push_to_hub_model_id:
-    value: null
-push_to_hub_organization:
-    value: null
-push_to_hub_token:
-    value: <PUSH_TO_HUB_TOKEN>
-ray_scope:
-    value: last
-remove_invalid_values:
-    value: false
-remove_unused_columns:
-    value: true
-repetition_penalty:
-    value: 1
-report_to:
-    value:
-        - wandb
-restore_callback_states_from_checkpoint:
-    value: false
-resume_from_checkpoint:
-    value: null
-return_dict:
-    value: true
-return_dict_in_generate:
-    value: false
-run_name:
-    value: whisper-small-eu
-save_on_each_node:
-    value: false
-save_only_model:
-    value: false
-save_safetensors:
-    value: true
-save_steps:
-    value: 1000
-save_strategy:
-    value: steps
-save_total_limit:
-    value: null
-scale_embedding:
-    value: false
-seed:
-    value: 42
-sep_token_id:
-    value: null
-skip_memory_metrics:
-    value: true
-sortish_sampler:
-    value: false
-split_batches:
-    value: null
-suppress_tokens:
-    value: null
-task_specific_params:
-    value: null
-temperature:
-    value: 1
-tf_legacy_loss:
-    value: false
-tf32:
-    value: null
-tie_encoder_decoder:
-    value: false
-tie_word_embeddings:
-    value: true
-tokenizer_class:
-    value: null
-top_k:
-    value: 50
-top_p:
-    value: 1
-torch_compile:
-    value: false
-torch_compile_backend:
-    value: null
-torch_compile_mode:
-    value: null
-torch_dtype:
-    value: float32
-torch_empty_cache_steps:
-    value: null
-torchdynamo:
-    value: null
-torchscript:
-    value: false
-tpu_metrics_debug:
-    value: false
-tpu_num_cores:
-    value: null
-transformers_version:
-    value: 4.49.0.dev0
-typical_p:
-    value: 1
-use_bfloat16:
-    value: false
-use_cache:
-    value: false
-use_cpu:
-    value: false
-use_ipex:
-    value: false
-use_legacy_prediction_loop:
-    value: false
-use_liger_kernel:
-    value: false
-use_mps_device:
-    value: false
-use_weighted_layer_sum:
-    value: false
-vocab_size:
-    value: 51865
-warmup_ratio:
-    value: 0
-warmup_steps:
-    value: 500
-weight_decay:
-    value: 0

wandb/run-20250212_121751-d4i88lzt/files/output.log

@@ -1,22 +0,0 @@
-  0%|                                                                                                                                               | 0/8000 [00:00<?, ?it/s]Traceback (most recent call last):
-  File "/home/tknika/xezpeleta/whisper/whisper-small-eu/run_speech_recognition_seq2seq_streaming.py", line 630, in <module>
-    main()
-  File "/home/tknika/xezpeleta/whisper/whisper-small-eu/run_speech_recognition_seq2seq_streaming.py", line 579, in main
-    train_result = trainer.train(resume_from_checkpoint=checkpoint)
-                   ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
-  File "/home/tknika/xezpeleta/whisper/whisper-small-eu/.venv/lib/python3.12/site-packages/transformers/trainer.py", line 2197, in train
-    return inner_training_loop(
-           ^^^^^^^^^^^^^^^^^^^^
-  File "/home/tknika/xezpeleta/whisper/whisper-small-eu/.venv/lib/python3.12/site-packages/transformers/trainer.py", line 2438, in _inner_training_loop
-    self.control = self.callback_handler.on_epoch_begin(args, self.state, self.control)
-                   ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
-  File "/home/tknika/xezpeleta/whisper/whisper-small-eu/.venv/lib/python3.12/site-packages/transformers/trainer_callback.py", line 514, in on_epoch_begin
-    return self.call_event("on_epoch_begin", args, state, control)
-           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
-  File "/home/tknika/xezpeleta/whisper/whisper-small-eu/.venv/lib/python3.12/site-packages/transformers/trainer_callback.py", line 557, in call_event
-    result = getattr(callback, event)(
-             ^^^^^^^^^^^^^^^^^^^^^^^^^
-  File "/home/tknika/xezpeleta/whisper/whisper-small-eu/run_speech_recognition_seq2seq_streaming.py", line 555, in on_epoch_begin
-    if isinstance(train_dataloader.dataset, IterableDatasetShard):
-                  ^^^^^^^^^^^^^^^^^^^^^^^^
-AttributeError: 'NoneType' object has no attribute 'dataset'

wandb/run-20250212_121751-d4i88lzt/files/requirements.txt

@@ -1,115 +0,0 @@
-aiosignal==1.3.2
-Markdown==3.7
-more-itertools==10.6.0
-requests==2.32.3
-sentry-sdk==2.21.0
-torchaudio==2.6.0
-charset-normalizer==3.4.1
-docker-pycreds==0.4.0
-nvidia-cusolver-cu12==11.6.1.9
-PyYAML==6.0.2
-librosa==0.10.2.post1
-soxr==0.5.0.post1
-multiprocess==0.70.16
-setuptools==75.8.0
-nvidia-cufft-cu12==11.2.1.3
-joblib==1.4.2
-pytz==2025.1
-pip==24.0
-scikit-learn==1.6.1
-certifi==2025.1.31
-jiwer==3.1.0
-regex==2024.11.6
-annotated-types==0.7.0
-grpcio==1.70.0
-msgpack==1.1.0
-mpmath==1.3.0
-nvidia-cudnn-cu12==9.1.0.70
-soundfile==0.13.1
-dill==0.3.8
-nvidia-nvtx-cu12==12.4.127
-six==1.17.0
-nvidia-cuda-cupti-cu12==12.4.127
-pyarrow==19.0.0
-nvidia-nccl-cu12==2.21.5
-psutil==6.1.1
-decorator==5.1.1
-llvmlite==0.44.0
-frozenlist==1.5.0
-pydantic==2.10.6
-networkx==3.4.2
-idna==3.10
-wandb==0.19.6
-aiohttp==3.11.12
-RapidFuzz==3.12.1
-pandas==2.2.3
-python-dateutil==2.9.0.post0
-numpy==2.1.3
-tokenizers==0.21.0
-nvidia-cusparselt-cu12==0.6.2
-typing_extensions==4.12.2
-urllib3==2.3.0
-setproctitle==1.3.4
-tzdata==2025.1
-sympy==1.13.1
-pooch==1.8.2
-click==8.1.8
-pydantic_core==2.27.2
-MarkupSafe==3.0.2
-scipy==1.15.1
-accelerate==1.3.0
-tensorboard==2.19.0
-protobuf==5.29.3
-gitdb==4.0.12
-smmap==5.0.2
-absl-py==2.1.0
-tqdm==4.67.1
-yarl==1.18.3
-pycparser==2.22
-nvidia-cusparse-cu12==12.3.1.170
-attrs==25.1.0
-lazy_loader==0.4
-tensorboard-data-server==0.7.2
-threadpoolctl==3.5.0
-GitPython==3.1.44
-safetensors==0.5.2
-fsspec==2024.12.0
-nvidia-cuda-nvrtc-cu12==12.4.127
-filelock==3.17.0
-aiohappyeyeballs==2.4.6
-packaging==24.2
-datasets==3.2.1.dev0
-audioread==3.0.1
-propcache==0.2.1
-transformers==4.49.0.dev0
-nvidia-cuda-runtime-cu12==12.4.127
-cffi==1.17.1
-evaluate==0.4.3
-Werkzeug==3.1.3
-huggingface-hub==0.28.1
-Jinja2==3.1.5
-torch==2.6.0
-nvidia-curand-cu12==10.3.5.147
-xxhash==3.5.0
-platformdirs==4.3.6
-multidict==6.1.0
-nvidia-cublas-cu12==12.4.5.8
-nvidia-nvjitlink-cu12==12.4.127
-triton==3.2.0
-numba==0.61.0
-importlib_metadata==8.0.0
-platformdirs==4.2.2
-typeguard==4.3.0
-more-itertools==10.3.0
-tomli==2.0.1
-autocommand==2.2.2
-zipp==3.19.2
-typing_extensions==4.12.2
-backports.tarfile==1.2.0
-inflect==7.3.1
-jaraco.text==3.12.1
-wheel==0.43.0
-packaging==24.2
-jaraco.collections==5.1.0
-jaraco.functools==4.0.1
-jaraco.context==5.3.0

wandb/run-20250212_121751-d4i88lzt/files/wandb-metadata.json

@@ -1,85 +0,0 @@
-{
-  "os":  "Linux-6.8.0-48-generic-x86_64-with-glibc2.39",
-  "python":  "CPython 3.12.3",
-  "startedAt":  "2025-02-12T12:17:51.527114Z",
-  "args":  [
-    "--model_name_or_path=openai/whisper-small",
-    "--dataset_name=asierhv/composite_corpus_eu_v2.1",
-    "--language=basque",
-    "--train_split_name=train",
-    "--eval_split_name=dev_parl+test_parl+test_cv+test_oslr",
-    "--model_index_name=Whisper Small Basque",
-    "--max_steps=8000",
-    "--output_dir=./",
-    "--per_device_train_batch_size=32",
-    "--per_device_eval_batch_size=16",
-    "--gradient_accumulation_steps=1",
-    "--logging_steps=25",
-    "--learning_rate=1e-5",
-    "--warmup_steps=500",
-    "--evaluation_strategy=steps",
-    "--eval_steps=1000",
-    "--save_strategy=steps",
-    "--save_steps=1000",
-    "--generation_max_length=225",
-    "--length_column_name=input_length",
-    "--max_duration_in_seconds=30",
-    "--text_column_name=sentence",
-    "--freeze_feature_encoder=False",
-    "--report_to=tensorboard",
-    "--metric_for_best_model=wer",
-    "--greater_is_better=False",
-    "--load_best_model_at_end",
-    "--gradient_checkpointing",
-    "--fp16",
-    "--overwrite_output_dir",
-    "--do_train",
-    "--do_eval",
-    "--predict_with_generate",
-    "--do_normalize_eval",
-    "--streaming",
-    "--use_auth_token",
-    "--push_to_hub",
-    "--report_to",
-    "wandb",
-    "--run_name",
-    "whisper-small-eu"
-  ],
-  "program":  "/home/tknika/xezpeleta/whisper/whisper-small-eu/run_speech_recognition_seq2seq_streaming.py",
-  "codePath":  "run_speech_recognition_seq2seq_streaming.py",
-  "git":  {
-    "remote":  "https://huggingface.co/xezpeleta/whisper-small-eu",
-    "commit":  "9c975864b20b4df94398a870e97cad2934253ec3"
-  },
-  "email":  "[email protected]",
-  "root":  "/home/tknika/xezpeleta/whisper/whisper-small-eu",
-  "host":  "tknika",
-  "executable":  "/home/tknika/xezpeleta/whisper/whisper-small-eu/.venv/bin/python",
-  "codePathLocal":  "run_speech_recognition_seq2seq_streaming.py",
-  "cpu_count":  8,
-  "cpu_count_logical":  8,
-  "gpu":  "NVIDIA L40-48Q",
-  "gpu_count":  1,
-  "disk":  {
-    "/":  {
-      "total":  "525987168256",
-      "used":  "297346564096"
-    }
-  },
-  "memory":  {
-    "total":  "33654022144"
-  },
-  "cpu":  {
-    "count":  8,
-    "countLogical":  8
-  },
-  "gpu_nvidia":  [
-    {
-      "name":  "NVIDIA L40-48Q",
-      "memoryTotal":  "51539607552",
-      "cudaCores":  18176,
-      "architecture":  "Ada"
-    }
-  ],
-  "cudaVersion":  "12.4"
-}

wandb/run-20250212_121751-d4i88lzt/files/wandb-summary.json

@@ -1 +0,0 @@
-{"_wandb":{"runtime":0}}

wandb/run-20250212_121751-d4i88lzt/logs/debug-core.log

@@ -1,14 +0,0 @@
-{"time":"2025-02-12T12:17:51.340771692Z","level":"INFO","msg":"main: starting server","port-filename":"/tmp/tmpu_kqxp5v/port-223392.txt","pid":223392,"log-level":0,"disable-analytics":false,"shutdown-on-parent-exit":false}
-{"time":"2025-02-12T12:17:51.391525122Z","level":"INFO","msg":"Will exit if parent process dies.","ppid":223392}
-{"time":"2025-02-12T12:17:51.391505422Z","level":"INFO","msg":"server is running","addr":{"IP":"127.0.0.1","Port":35377,"Zone":""}}
-{"time":"2025-02-12T12:17:51.521026758Z","level":"INFO","msg":"connection: ManageConnectionData: new connection created","id":"127.0.0.1:42852"}
-{"time":"2025-02-12T12:17:51.529437253Z","level":"INFO","msg":"handleInformInit: received","streamId":"d4i88lzt","id":"127.0.0.1:42852"}
-{"time":"2025-02-12T12:17:51.635683608Z","level":"INFO","msg":"handleInformInit: stream started","streamId":"d4i88lzt","id":"127.0.0.1:42852"}
-{"time":"2025-02-12T12:17:52.089736796Z","level":"INFO","msg":"handleInformTeardown: server teardown initiated","id":"127.0.0.1:42852"}
-{"time":"2025-02-12T12:17:52.089842845Z","level":"INFO","msg":"connection: closing","id":"127.0.0.1:42852"}
-{"time":"2025-02-12T12:17:52.089890025Z","level":"INFO","msg":"connection: closed successfully","id":"127.0.0.1:42852"}
-{"time":"2025-02-12T12:17:52.089878375Z","level":"INFO","msg":"server is shutting down"}
-{"time":"2025-02-12T12:17:52.241493374Z","level":"ERROR","msg":"processOutgoingData: flush error","error":"write tcp 127.0.0.1:35377->127.0.0.1:42852: use of closed network connection","id":"127.0.0.1:42852"}
-{"time":"2025-02-12T12:17:53.244042129Z","level":"INFO","msg":"handleInformTeardown: server shutdown complete","id":"127.0.0.1:42852"}
-{"time":"2025-02-12T12:17:53.244065929Z","level":"INFO","msg":"connection: ManageConnectionData: connection closed","id":"127.0.0.1:42852"}
-{"time":"2025-02-12T12:17:53.244128968Z","level":"INFO","msg":"server is closed"}

wandb/run-20250212_121751-d4i88lzt/logs/debug-internal.log

@@ -1,15 +0,0 @@
-{"time":"2025-02-12T12:17:51.5298133Z","level":"INFO","msg":"stream: starting","core version":"0.19.6","symlink path":"/home/tknika/xezpeleta/whisper/whisper-small-eu/wandb/run-20250212_121751-d4i88lzt/logs/debug-core.log"}
-{"time":"2025-02-12T12:17:51.635607299Z","level":"INFO","msg":"created new stream","id":"d4i88lzt"}
-{"time":"2025-02-12T12:17:51.635674098Z","level":"INFO","msg":"stream: started","id":"d4i88lzt"}
-{"time":"2025-02-12T12:17:51.635773898Z","level":"INFO","msg":"writer: Do: started","stream_id":"d4i88lzt"}
-{"time":"2025-02-12T12:17:51.635842217Z","level":"INFO","msg":"sender: started","stream_id":"d4i88lzt"}
-{"time":"2025-02-12T12:17:51.635963186Z","level":"INFO","msg":"handler: started","stream_id":"d4i88lzt"}
-{"time":"2025-02-12T12:17:51.947487454Z","level":"INFO","msg":"Starting system monitor"}
-{"time":"2025-02-12T12:17:52.089832235Z","level":"INFO","msg":"stream: closing","id":"d4i88lzt"}
-{"time":"2025-02-12T12:17:52.089860885Z","level":"INFO","msg":"Stopping system monitor"}
-{"time":"2025-02-12T12:17:52.090422051Z","level":"INFO","msg":"Stopped system monitor"}
-{"time":"2025-02-12T12:17:53.018559862Z","level":"INFO","msg":"fileTransfer: Close: file transfer manager closed"}
-{"time":"2025-02-12T12:17:53.24378817Z","level":"INFO","msg":"handler: closed","stream_id":"d4i88lzt"}
-{"time":"2025-02-12T12:17:53.24383994Z","level":"INFO","msg":"writer: Close: closed","stream_id":"d4i88lzt"}
-{"time":"2025-02-12T12:17:53.24386653Z","level":"INFO","msg":"sender: closed","stream_id":"d4i88lzt"}
-{"time":"2025-02-12T12:17:53.243926789Z","level":"INFO","msg":"stream: closed","id":"d4i88lzt"}

wandb/run-20250212_121751-d4i88lzt/logs/debug.log

@@ -1,26 +0,0 @@
-2025-02-12 12:17:51,311 INFO    MainThread:223392 [wandb_setup.py:_flush():68] Current SDK version is 0.19.6
-2025-02-12 12:17:51,311 INFO    MainThread:223392 [wandb_setup.py:_flush():68] Configure stats pid to 223392
-2025-02-12 12:17:51,311 INFO    MainThread:223392 [wandb_setup.py:_flush():68] Loading settings from /home/tknika/.config/wandb/settings
-2025-02-12 12:17:51,311 INFO    MainThread:223392 [wandb_setup.py:_flush():68] Loading settings from /home/tknika/xezpeleta/whisper/whisper-small-eu/wandb/settings
-2025-02-12 12:17:51,311 INFO    MainThread:223392 [wandb_setup.py:_flush():68] Loading settings from environment variables
-2025-02-12 12:17:51,311 INFO    MainThread:223392 [wandb_init.py:setup_run_log_directory():637] Logging user logs to /home/tknika/xezpeleta/whisper/whisper-small-eu/wandb/run-20250212_121751-d4i88lzt/logs/debug.log
-2025-02-12 12:17:51,311 INFO    MainThread:223392 [wandb_init.py:setup_run_log_directory():638] Logging internal logs to /home/tknika/xezpeleta/whisper/whisper-small-eu/wandb/run-20250212_121751-d4i88lzt/logs/debug-internal.log
-2025-02-12 12:17:51,311 INFO    MainThread:223392 [wandb_init.py:init():756] calling init triggers
-2025-02-12 12:17:51,312 INFO    MainThread:223392 [wandb_init.py:init():761] wandb.init called with sweep_config: {}
-config: {'_wandb': {}}
-2025-02-12 12:17:51,312 INFO    MainThread:223392 [wandb_init.py:init():789] starting backend
-2025-02-12 12:17:51,521 INFO    MainThread:223392 [wandb_init.py:init():793] sending inform_init request
-2025-02-12 12:17:51,526 INFO    MainThread:223392 [backend.py:_multiprocessing_setup():97] multiprocessing start_methods=fork,spawn,forkserver, using: spawn
-2025-02-12 12:17:51,526 INFO    MainThread:223392 [wandb_init.py:init():808] backend started and connected
-2025-02-12 12:17:51,528 INFO    MainThread:223392 [wandb_init.py:init():901] updated telemetry
-2025-02-12 12:17:51,535 INFO    MainThread:223392 [wandb_init.py:init():936] communicating run to backend with 90.0 second timeout
-2025-02-12 12:17:51,944 INFO    MainThread:223392 [wandb_init.py:init():994] starting run threads in backend
-2025-02-12 12:17:52,049 INFO    MainThread:223392 [wandb_run.py:_console_start():2385] atexit reg
-2025-02-12 12:17:52,049 INFO    MainThread:223392 [wandb_run.py:_redirect():2235] redirect: wrap_raw
-2025-02-12 12:17:52,049 INFO    MainThread:223392 [wandb_run.py:_redirect():2300] Wrapping output streams.
-2025-02-12 12:17:52,049 INFO    MainThread:223392 [wandb_run.py:_redirect():2325] Redirects installed.
-2025-02-12 12:17:52,051 INFO    MainThread:223392 [wandb_init.py:init():1036] run started, returning control to user process
-2025-02-12 12:17:52,052 INFO    MainThread:223392 [wandb_run.py:_config_callback():1253] config_cb None None {'vocab_size': 51865, 'num_mel_bins': 80, 'd_model': 768, 'encoder_layers': 12, 'encoder_attention_heads': 12, 'decoder_layers': 12, 'decoder_attention_heads': 12, 'decoder_ffn_dim': 3072, 'encoder_ffn_dim': 3072, 'dropout': 0.0, 'attention_dropout': 0.0, 'activation_dropout': 0.0, 'activation_function': 'gelu', 'init_std': 0.02, 'encoder_layerdrop': 0.0, 'decoder_layerdrop': 0.0, 'use_cache': False, 'num_hidden_layers': 12, 'scale_embedding': False, 'max_source_positions': 1500, 'max_target_positions': 448, 'classifier_proj_size': 256, 'use_weighted_layer_sum': False, 'apply_spec_augment': False, 'mask_time_prob': 0.05, 'mask_time_length': 10, 'mask_time_min_masks': 2, 'mask_feature_prob': 0.0, 'mask_feature_length': 10, 'mask_feature_min_masks': 0, 'median_filter_width': 7, 'return_dict': True, 'output_hidden_states': False, 'output_attentions': False, 'torchscript': False, 'torch_dtype': 'float32', 'use_bfloat16': False, 'tf_legacy_loss': False, 'pruned_heads': {}, 'tie_word_embeddings': True, 'chunk_size_feed_forward': 0, 'is_encoder_decoder': True, 'is_decoder': False, 'cross_attention_hidden_size': None, 'add_cross_attention': False, 'tie_encoder_decoder': False, 'max_length': 448, 'min_length': 0, 'do_sample': False, 'early_stopping': False, 'num_beams': 1, 'num_beam_groups': 1, 'diversity_penalty': 0.0, 'temperature': 1.0, 'top_k': 50, 'top_p': 1.0, 'typical_p': 1.0, 'repetition_penalty': 1.0, 'length_penalty': 1.0, 'no_repeat_ngram_size': 0, 'encoder_no_repeat_ngram_size': 0, 'bad_words_ids': None, 'num_return_sequences': 1, 'output_scores': False, 'return_dict_in_generate': False, 'forced_bos_token_id': None, 'forced_eos_token_id': None, 'remove_invalid_values': False, 'exponential_decay_length_penalty': None, 'suppress_tokens': None, 'begin_suppress_tokens': [220, 50257], 'architectures': ['WhisperForConditionalGeneration'], 'finetuning_task': None, 'id2label': {0: 'LABEL_0', 1: 'LABEL_1'}, 'label2id': {'LABEL_0': 0, 'LABEL_1': 1}, 'tokenizer_class': None, 'prefix': None, 'bos_token_id': 50257, 'pad_token_id': 50257, 'eos_token_id': 50257, 'sep_token_id': None, 'decoder_start_token_id': 50258, 'task_specific_params': None, 'problem_type': None, '_name_or_path': 'openai/whisper-small', '_attn_implementation_autoset': True, 'transformers_version': '4.49.0.dev0', 'forced_decoder_ids': None, 'model_type': 'whisper', 'output_dir': './', 'overwrite_output_dir': True, 'do_train': True, 'do_eval': True, 'do_predict': False, 'eval_strategy': 'steps', 'prediction_loss_only': False, 'per_device_train_batch_size': 32, 'per_device_eval_batch_size': 16, 'per_gpu_train_batch_size': None, 'per_gpu_eval_batch_size': None, 'gradient_accumulation_steps': 1, 'eval_accumulation_steps': None, 'eval_delay': 0, 'torch_empty_cache_steps': None, 'learning_rate': 1e-05, 'weight_decay': 0.0, 'adam_beta1': 0.9, 'adam_beta2': 0.999, 'adam_epsilon': 1e-08, 'max_grad_norm': 1.0, 'num_train_epochs': 3.0, 'max_steps': 8000, 'lr_scheduler_type': 'linear', 'lr_scheduler_kwargs': {}, 'warmup_ratio': 0.0, 'warmup_steps': 500, 'log_level': 'passive', 'log_level_replica': 'warning', 'log_on_each_node': True, 'logging_dir': './runs/Feb12_12-17-27_tknika', 'logging_strategy': 'steps', 'logging_first_step': False, 'logging_steps': 25, 'logging_nan_inf_filter': True, 'save_strategy': 'steps', 'save_steps': 1000, 'save_total_limit': None, 'save_safetensors': True, 'save_on_each_node': False, 'save_only_model': False, 'restore_callback_states_from_checkpoint': False, 'no_cuda': False, 'use_cpu': False, 'use_mps_device': False, 'seed': 42, 'data_seed': None, 'jit_mode_eval': False, 'use_ipex': False, 'bf16': False, 'fp16': True, 'fp16_opt_level': 'O1', 'half_precision_backend': 'auto', 'bf16_full_eval': False, 'fp16_full_eval': False, 'tf32': None, 'local_rank': 0, 'ddp_backend': None, 'tpu_num_cores': None, 'tpu_metrics_debug': False, 'debug': [], 'dataloader_drop_last': False, 'eval_steps': 1000, 'dataloader_num_workers': 0, 'dataloader_prefetch_factor': None, 'past_index': -1, 'run_name': 'whisper-small-eu', 'disable_tqdm': False, 'remove_unused_columns': True, 'label_names': None, 'load_best_model_at_end': True, 'metric_for_best_model': 'wer', 'greater_is_better': False, 'ignore_data_skip': False, 'fsdp': [], 'fsdp_min_num_params': 0, 'fsdp_config': {'min_num_params': 0, 'xla': False, 'xla_fsdp_v2': False, 'xla_fsdp_grad_ckpt': False}, 'fsdp_transformer_layer_cls_to_wrap': None, 'accelerator_config': {'split_batches': False, 'dispatch_batches': None, 'even_batches': True, 'use_seedable_sampler': True, 'non_blocking': False, 'gradient_accumulation_kwargs': None}, 'deepspeed': None, 'label_smoothing_factor': 0.0, 'optim': 'adamw_torch', 'optim_args': None, 'adafactor': False, 'group_by_length': False, 'length_column_name': 'input_length', 'report_to': ['wandb'], 'ddp_find_unused_parameters': None, 'ddp_bucket_cap_mb': None, 'ddp_broadcast_buffers': None, 'dataloader_pin_memory': True, 'dataloader_persistent_workers': False, 'skip_memory_metrics': True, 'use_legacy_prediction_loop': False, 'push_to_hub': True, 'resume_from_checkpoint': None, 'hub_model_id': None, 'hub_strategy': 'every_save', 'hub_token': '<HUB_TOKEN>', 'hub_private_repo': None, 'hub_always_push': False, 'gradient_checkpointing': True, 'gradient_checkpointing_kwargs': None, 'include_inputs_for_metrics': False, 'include_for_metrics': [], 'eval_do_concat_batches': True, 'fp16_backend': 'auto', 'evaluation_strategy': 'steps', 'push_to_hub_model_id': None, 'push_to_hub_organization': None, 'push_to_hub_token': '<PUSH_TO_HUB_TOKEN>', 'mp_parameters': '', 'auto_find_batch_size': False, 'full_determinism': False, 'torchdynamo': None, 'ray_scope': 'last', 'ddp_timeout': 1800, 'torch_compile': False, 'torch_compile_backend': None, 'torch_compile_mode': None, 'dispatch_batches': None, 'split_batches': None, 'include_tokens_per_second': False, 'include_num_input_tokens_seen': False, 'neftune_noise_alpha': None, 'optim_target_modules': None, 'batch_eval_metrics': False, 'eval_on_start': False, 'use_liger_kernel': False, 'eval_use_gather_object': False, 'average_tokens_across_devices': False, 'sortish_sampler': False, 'predict_with_generate': True, 'generation_max_length': 225, 'generation_num_beams': None, 'generation_config': None}
-2025-02-12 12:17:52,054 INFO    MainThread:223392 [wandb_config.py:__setitem__():154] config set model/num_parameters = 241734912 - <bound method Run._config_callback of <wandb.sdk.wandb_run.Run object at 0x75ef87e92c00>>
-2025-02-12 12:17:52,055 INFO    MainThread:223392 [wandb_run.py:_config_callback():1253] config_cb model/num_parameters 241734912 None
-2025-02-12 12:17:52,089 WARNING MsgRouterThr:223392 [router.py:message_loop():75] message_loop has been closed

wandb/run-20250212_122637-v3d3ouvn/files/config.yaml

@@ -1,512 +0,0 @@
-_attn_implementation_autoset:
-    value: true
-_name_or_path:
-    value: openai/whisper-small
-_wandb:
-    value:
-        cli_version: 0.19.6
-        m:
-            - "1": train/global_step
-              "6":
-                - 3
-              "7": []
-        python_version: 3.12.3
-        t:
-            "1":
-                - 1
-                - 5
-                - 11
-                - 49
-                - 51
-                - 53
-                - 55
-                - 71
-                - 100
-            "2":
-                - 1
-                - 5
-                - 11
-                - 49
-                - 51
-                - 53
-                - 55
-                - 71
-                - 100
-            "3":
-                - 7
-                - 13
-                - 19
-                - 23
-                - 55
-                - 66
-            "4": 3.12.3
-            "5": 0.19.6
-            "6": 4.49.0.dev0
-            "8":
-                - 5
-            "9":
-                "1": transformers_trainer
-            "12": 0.19.6
-            "13": linux-x86_64
-accelerator_config:
-    value:
-        dispatch_batches: null
-        even_batches: true
-        gradient_accumulation_kwargs: null
-        non_blocking: false
-        split_batches: false
-        use_seedable_sampler: true
-activation_dropout:
-    value: 0
-activation_function:
-    value: gelu
-adafactor:
-    value: false
-adam_beta1:
-    value: 0.9
-adam_beta2:
-    value: 0.999
-adam_epsilon:
-    value: 1e-08
-add_cross_attention:
-    value: false
-apply_spec_augment:
-    value: false
-architectures:
-    value:
-        - WhisperForConditionalGeneration
-attention_dropout:
-    value: 0
-auto_find_batch_size:
-    value: false
-average_tokens_across_devices:
-    value: false
-bad_words_ids:
-    value: null
-batch_eval_metrics:
-    value: false
-begin_suppress_tokens:
-    value:
-        - 220
-        - 50257
-bf16:
-    value: false
-bf16_full_eval:
-    value: false
-bos_token_id:
-    value: 50257
-chunk_size_feed_forward:
-    value: 0
-classifier_proj_size:
-    value: 256
-cross_attention_hidden_size:
-    value: null
-d_model:
-    value: 768
-data_seed:
-    value: null
-dataloader_drop_last:
-    value: false
-dataloader_num_workers:
-    value: 0
-dataloader_persistent_workers:
-    value: false
-dataloader_pin_memory:
-    value: true
-dataloader_prefetch_factor:
-    value: null
-ddp_backend:
-    value: null
-ddp_broadcast_buffers:
-    value: null
-ddp_bucket_cap_mb:
-    value: null
-ddp_find_unused_parameters:
-    value: null
-ddp_timeout:
-    value: 1800
-debug:
-    value: []
-decoder_attention_heads:
-    value: 12
-decoder_ffn_dim:
-    value: 3072
-decoder_layerdrop:
-    value: 0
-decoder_layers:
-    value: 12
-decoder_start_token_id:
-    value: 50258
-deepspeed:
-    value: null
-disable_tqdm:
-    value: false
-dispatch_batches:
-    value: null
-diversity_penalty:
-    value: 0
-do_eval:
-    value: true
-do_predict:
-    value: false
-do_sample:
-    value: false
-do_train:
-    value: true
-dropout:
-    value: 0
-early_stopping:
-    value: false
-encoder_attention_heads:
-    value: 12
-encoder_ffn_dim:
-    value: 3072
-encoder_layerdrop:
-    value: 0
-encoder_layers:
-    value: 12
-encoder_no_repeat_ngram_size:
-    value: 0
-eos_token_id:
-    value: 50257
-eval_accumulation_steps:
-    value: null
-eval_delay:
-    value: 0
-eval_do_concat_batches:
-    value: true
-eval_on_start:
-    value: false
-eval_steps:
-    value: 1000
-eval_strategy:
-    value: steps
-eval_use_gather_object:
-    value: false
-evaluation_strategy:
-    value: steps
-exponential_decay_length_penalty:
-    value: null
-finetuning_task:
-    value: null
-forced_bos_token_id:
-    value: null
-forced_decoder_ids:
-    value: null
-forced_eos_token_id:
-    value: null
-fp16:
-    value: true
-fp16_backend:
-    value: auto
-fp16_full_eval:
-    value: false
-fp16_opt_level:
-    value: O1
-fsdp:
-    value: []
-fsdp_config:
-    value:
-        min_num_params: 0
-        xla: false
-        xla_fsdp_grad_ckpt: false
-        xla_fsdp_v2: false
-fsdp_min_num_params:
-    value: 0
-fsdp_transformer_layer_cls_to_wrap:
-    value: null
-full_determinism:
-    value: false
-generation_config:
-    value: null
-generation_max_length:
-    value: 225
-generation_num_beams:
-    value: null
-gradient_accumulation_steps:
-    value: 1
-gradient_checkpointing:
-    value: true
-gradient_checkpointing_kwargs:
-    value: null
-greater_is_better:
-    value: false
-group_by_length:
-    value: false
-half_precision_backend:
-    value: auto
-hub_always_push:
-    value: false
-hub_model_id:
-    value: null
-hub_private_repo:
-    value: null
-hub_strategy:
-    value: every_save
-hub_token:
-    value: <HUB_TOKEN>
-id2label:
-    value:
-        "0": LABEL_0
-        "1": LABEL_1
-ignore_data_skip:
-    value: false
-include_for_metrics:
-    value: []
-include_inputs_for_metrics:
-    value: false
-include_num_input_tokens_seen:
-    value: false
-include_tokens_per_second:
-    value: false
-init_std:
-    value: 0.02
-is_decoder:
-    value: false
-is_encoder_decoder:
-    value: true
-jit_mode_eval:
-    value: false
-label_names:
-    value: null
-label_smoothing_factor:
-    value: 0
-label2id:
-    value:
-        LABEL_0: 0
-        LABEL_1: 1
-learning_rate:
-    value: 1e-05
-length_column_name:
-    value: input_length
-length_penalty:
-    value: 1
-load_best_model_at_end:
-    value: true
-local_rank:
-    value: 0
-log_level:
-    value: passive
-log_level_replica:
-    value: warning
-log_on_each_node:
-    value: true
-logging_dir:
-    value: ./runs/Feb12_12-26-11_tknika
-logging_first_step:
-    value: false
-logging_nan_inf_filter:
-    value: true
-logging_steps:
-    value: 25
-logging_strategy:
-    value: steps
-lr_scheduler_type:
-    value: linear
-mask_feature_length:
-    value: 10
-mask_feature_min_masks:
-    value: 0
-mask_feature_prob:
-    value: 0
-mask_time_length:
-    value: 10
-mask_time_min_masks:
-    value: 2
-mask_time_prob:
-    value: 0.05
-max_grad_norm:
-    value: 1
-max_length:
-    value: 448
-max_source_positions:
-    value: 1500
-max_steps:
-    value: 8000
-max_target_positions:
-    value: 448
-median_filter_width:
-    value: 7
-metric_for_best_model:
-    value: wer
-min_length:
-    value: 0
-model/num_parameters:
-    value: 241734912
-model_type:
-    value: whisper
-mp_parameters:
-    value: ""
-neftune_noise_alpha:
-    value: null
-no_cuda:
-    value: false
-no_repeat_ngram_size:
-    value: 0
-num_beam_groups:
-    value: 1
-num_beams:
-    value: 1
-num_hidden_layers:
-    value: 12
-num_mel_bins:
-    value: 80
-num_return_sequences:
-    value: 1
-num_train_epochs:
-    value: 3
-optim:
-    value: adamw_torch
-optim_args:
-    value: null
-optim_target_modules:
-    value: null
-output_attentions:
-    value: false
-output_dir:
-    value: ./
-output_hidden_states:
-    value: false
-output_scores:
-    value: false
-overwrite_output_dir:
-    value: true
-pad_token_id:
-    value: 50257
-past_index:
-    value: -1
-per_device_eval_batch_size:
-    value: 16
-per_device_train_batch_size:
-    value: 32
-per_gpu_eval_batch_size:
-    value: null
-per_gpu_train_batch_size:
-    value: null
-predict_with_generate:
-    value: true
-prediction_loss_only:
-    value: false
-prefix:
-    value: null
-problem_type:
-    value: null
-push_to_hub:
-    value: true
-push_to_hub_model_id:
-    value: null
-push_to_hub_organization:
-    value: null
-push_to_hub_token:
-    value: <PUSH_TO_HUB_TOKEN>
-ray_scope:
-    value: last
-remove_invalid_values:
-    value: false
-remove_unused_columns:
-    value: true
-repetition_penalty:
-    value: 1
-report_to:
-    value:
-        - wandb
-restore_callback_states_from_checkpoint:
-    value: false
-resume_from_checkpoint:
-    value: null
-return_dict:
-    value: true
-return_dict_in_generate:
-    value: false
-run_name:
-    value: whisper-small-eu
-save_on_each_node:
-    value: false
-save_only_model:
-    value: false
-save_safetensors:
-    value: true
-save_steps:
-    value: 1000
-save_strategy:
-    value: steps
-save_total_limit:
-    value: null
-scale_embedding:
-    value: false
-seed:
-    value: 42
-sep_token_id:
-    value: null
-skip_memory_metrics:
-    value: true
-sortish_sampler:
-    value: false
-split_batches:
-    value: null
-suppress_tokens:
-    value: null
-task_specific_params:
-    value: null
-temperature:
-    value: 1
-tf_legacy_loss:
-    value: false
-tf32:
-    value: null
-tie_encoder_decoder:
-    value: false
-tie_word_embeddings:
-    value: true
-tokenizer_class:
-    value: null
-top_k:
-    value: 50
-top_p:
-    value: 1
-torch_compile:
-    value: false
-torch_compile_backend:
-    value: null
-torch_compile_mode:
-    value: null
-torch_dtype:
-    value: float32
-torch_empty_cache_steps:
-    value: null
-torchdynamo:
-    value: null
-torchscript:
-    value: false
-tpu_metrics_debug:
-    value: false
-tpu_num_cores:
-    value: null
-transformers_version:
-    value: 4.49.0.dev0
-typical_p:
-    value: 1
-use_bfloat16:
-    value: false
-use_cache:
-    value: false
-use_cpu:
-    value: false
-use_ipex:
-    value: false
-use_legacy_prediction_loop:
-    value: false
-use_liger_kernel:
-    value: false
-use_mps_device:
-    value: false
-use_weighted_layer_sum:
-    value: false
-vocab_size:
-    value: 51865
-warmup_ratio:
-    value: 0
-warmup_steps:
-    value: 500
-weight_decay:
-    value: 0

wandb/run-20250212_122637-v3d3ouvn/files/output.log

@@ -1,22 +0,0 @@
-  0%|                                                                                                                                               | 0/8000 [00:00<?, ?it/s]Traceback (most recent call last):
-  File "/home/tknika/xezpeleta/whisper/whisper-small-eu/run_speech_recognition_seq2seq_streaming.py", line 630, in <module>
-    main()
-  File "/home/tknika/xezpeleta/whisper/whisper-small-eu/run_speech_recognition_seq2seq_streaming.py", line 579, in main
-    train_result = trainer.train(resume_from_checkpoint=checkpoint)
-                   ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
-  File "/home/tknika/xezpeleta/whisper/whisper-small-eu/.venv/lib/python3.12/site-packages/transformers/trainer.py", line 2197, in train
-    return inner_training_loop(
-           ^^^^^^^^^^^^^^^^^^^^
-  File "/home/tknika/xezpeleta/whisper/whisper-small-eu/.venv/lib/python3.12/site-packages/transformers/trainer.py", line 2438, in _inner_training_loop
-    self.control = self.callback_handler.on_epoch_begin(args, self.state, self.control)
-                   ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
-  File "/home/tknika/xezpeleta/whisper/whisper-small-eu/.venv/lib/python3.12/site-packages/transformers/trainer_callback.py", line 514, in on_epoch_begin
-    return self.call_event("on_epoch_begin", args, state, control)
-           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
-  File "/home/tknika/xezpeleta/whisper/whisper-small-eu/.venv/lib/python3.12/site-packages/transformers/trainer_callback.py", line 557, in call_event
-    result = getattr(callback, event)(
-             ^^^^^^^^^^^^^^^^^^^^^^^^^
-  File "/home/tknika/xezpeleta/whisper/whisper-small-eu/run_speech_recognition_seq2seq_streaming.py", line 555, in on_epoch_begin
-    if isinstance(train_dataloader.dataset, IterableDatasetShard):
-                  ^^^^^^^^^^^^^^^^^^^^^^^^
-AttributeError: 'NoneType' object has no attribute 'dataset'

wandb/run-20250212_122637-v3d3ouvn/files/requirements.txt

@@ -1,115 +0,0 @@
-aiosignal==1.3.2
-Markdown==3.7
-more-itertools==10.6.0
-requests==2.32.3
-sentry-sdk==2.21.0
-torchaudio==2.6.0
-charset-normalizer==3.4.1
-docker-pycreds==0.4.0
-nvidia-cusolver-cu12==11.6.1.9
-PyYAML==6.0.2
-librosa==0.10.2.post1
-soxr==0.5.0.post1
-multiprocess==0.70.16
-setuptools==75.8.0
-nvidia-cufft-cu12==11.2.1.3
-joblib==1.4.2
-pytz==2025.1
-pip==24.0
-scikit-learn==1.6.1
-certifi==2025.1.31
-jiwer==3.1.0
-regex==2024.11.6
-annotated-types==0.7.0
-grpcio==1.70.0
-msgpack==1.1.0
-mpmath==1.3.0
-nvidia-cudnn-cu12==9.1.0.70
-soundfile==0.13.1
-dill==0.3.8
-nvidia-nvtx-cu12==12.4.127
-six==1.17.0
-nvidia-cuda-cupti-cu12==12.4.127
-pyarrow==19.0.0
-nvidia-nccl-cu12==2.21.5
-psutil==6.1.1
-decorator==5.1.1
-llvmlite==0.44.0
-frozenlist==1.5.0
-pydantic==2.10.6
-networkx==3.4.2
-idna==3.10
-wandb==0.19.6
-aiohttp==3.11.12
-RapidFuzz==3.12.1
-pandas==2.2.3
-python-dateutil==2.9.0.post0
-numpy==2.1.3
-tokenizers==0.21.0
-nvidia-cusparselt-cu12==0.6.2
-typing_extensions==4.12.2
-urllib3==2.3.0
-setproctitle==1.3.4
-tzdata==2025.1
-sympy==1.13.1
-pooch==1.8.2
-click==8.1.8
-pydantic_core==2.27.2
-MarkupSafe==3.0.2
-scipy==1.15.1
-accelerate==1.3.0
-tensorboard==2.19.0
-protobuf==5.29.3
-gitdb==4.0.12
-smmap==5.0.2
-absl-py==2.1.0
-tqdm==4.67.1
-yarl==1.18.3
-pycparser==2.22
-nvidia-cusparse-cu12==12.3.1.170
-attrs==25.1.0
-lazy_loader==0.4
-tensorboard-data-server==0.7.2
-threadpoolctl==3.5.0
-GitPython==3.1.44
-safetensors==0.5.2
-fsspec==2024.12.0
-nvidia-cuda-nvrtc-cu12==12.4.127
-filelock==3.17.0
-aiohappyeyeballs==2.4.6
-packaging==24.2
-datasets==3.2.1.dev0
-audioread==3.0.1
-propcache==0.2.1
-transformers==4.49.0.dev0
-nvidia-cuda-runtime-cu12==12.4.127
-cffi==1.17.1
-evaluate==0.4.3
-Werkzeug==3.1.3
-huggingface-hub==0.28.1
-Jinja2==3.1.5
-torch==2.6.0
-nvidia-curand-cu12==10.3.5.147
-xxhash==3.5.0
-platformdirs==4.3.6
-multidict==6.1.0
-nvidia-cublas-cu12==12.4.5.8
-nvidia-nvjitlink-cu12==12.4.127
-triton==3.2.0
-numba==0.61.0
-importlib_metadata==8.0.0
-platformdirs==4.2.2
-typeguard==4.3.0
-more-itertools==10.3.0
-tomli==2.0.1
-autocommand==2.2.2
-zipp==3.19.2
-typing_extensions==4.12.2
-backports.tarfile==1.2.0
-inflect==7.3.1
-jaraco.text==3.12.1
-wheel==0.43.0
-packaging==24.2
-jaraco.collections==5.1.0
-jaraco.functools==4.0.1
-jaraco.context==5.3.0

wandb/run-20250212_122637-v3d3ouvn/files/wandb-metadata.json

@@ -1,85 +0,0 @@
-{
-  "os":  "Linux-6.8.0-48-generic-x86_64-with-glibc2.39",
-  "python":  "CPython 3.12.3",
-  "startedAt":  "2025-02-12T12:26:37.277902Z",
-  "args":  [
-    "--model_name_or_path=openai/whisper-small",
-    "--dataset_name=asierhv/composite_corpus_eu_v2.1",
-    "--language=basque",
-    "--train_split_name=train",
-    "--eval_split_name=dev_parl+test_parl+test_cv+test_oslr",
-    "--model_index_name=Whisper Small Basque",
-    "--max_steps=8000",
-    "--output_dir=./",
-    "--per_device_train_batch_size=32",
-    "--per_device_eval_batch_size=16",
-    "--gradient_accumulation_steps=1",
-    "--logging_steps=25",
-    "--learning_rate=1e-5",
-    "--warmup_steps=500",
-    "--evaluation_strategy=steps",
-    "--eval_steps=1000",
-    "--save_strategy=steps",
-    "--save_steps=1000",
-    "--generation_max_length=225",
-    "--length_column_name=input_length",
-    "--max_duration_in_seconds=30",
-    "--text_column_name=sentence",
-    "--freeze_feature_encoder=False",
-    "--report_to=tensorboard",
-    "--metric_for_best_model=wer",
-    "--greater_is_better=False",
-    "--load_best_model_at_end",
-    "--gradient_checkpointing",
-    "--fp16",
-    "--overwrite_output_dir",
-    "--do_train",
-    "--do_eval",
-    "--predict_with_generate",
-    "--do_normalize_eval",
-    "--streaming",
-    "--use_auth_token",
-    "--push_to_hub",
-    "--report_to",
-    "wandb",
-    "--run_name",
-    "whisper-small-eu"
-  ],
-  "program":  "/home/tknika/xezpeleta/whisper/whisper-small-eu/run_speech_recognition_seq2seq_streaming.py",
-  "codePath":  "run_speech_recognition_seq2seq_streaming.py",
-  "git":  {
-    "remote":  "https://huggingface.co/xezpeleta/whisper-small-eu",
-    "commit":  "9c975864b20b4df94398a870e97cad2934253ec3"
-  },
-  "email":  "[email protected]",
-  "root":  "/home/tknika/xezpeleta/whisper/whisper-small-eu",
-  "host":  "tknika",
-  "executable":  "/home/tknika/xezpeleta/whisper/whisper-small-eu/.venv/bin/python",
-  "codePathLocal":  "run_speech_recognition_seq2seq_streaming.py",
-  "cpu_count":  8,
-  "cpu_count_logical":  8,
-  "gpu":  "NVIDIA L40-48Q",
-  "gpu_count":  1,
-  "disk":  {
-    "/":  {
-      "total":  "525987168256",
-      "used":  "297346666496"
-    }
-  },
-  "memory":  {
-    "total":  "33654022144"
-  },
-  "cpu":  {
-    "count":  8,
-    "countLogical":  8
-  },
-  "gpu_nvidia":  [
-    {
-      "name":  "NVIDIA L40-48Q",
-      "memoryTotal":  "51539607552",
-      "cudaCores":  18176,
-      "architecture":  "Ada"
-    }
-  ],
-  "cudaVersion":  "12.4"
-}

wandb/run-20250212_122637-v3d3ouvn/files/wandb-summary.json

@@ -1 +0,0 @@
-{"_wandb":{"runtime":0}}

wandb/run-20250212_122637-v3d3ouvn/logs/debug-core.log

@@ -1,14 +0,0 @@
-{"time":"2025-02-12T12:26:37.096402413Z","level":"INFO","msg":"main: starting server","port-filename":"/tmp/tmpcjtnmyy4/port-224110.txt","pid":224110,"log-level":0,"disable-analytics":false,"shutdown-on-parent-exit":false}
-{"time":"2025-02-12T12:26:37.136235603Z","level":"INFO","msg":"Will exit if parent process dies.","ppid":224110}
-{"time":"2025-02-12T12:26:37.136202753Z","level":"INFO","msg":"server is running","addr":{"IP":"127.0.0.1","Port":34237,"Zone":""}}
-{"time":"2025-02-12T12:26:37.272154204Z","level":"INFO","msg":"connection: ManageConnectionData: new connection created","id":"127.0.0.1:48156"}
-{"time":"2025-02-12T12:26:37.280104802Z","level":"INFO","msg":"handleInformInit: received","streamId":"v3d3ouvn","id":"127.0.0.1:48156"}
-{"time":"2025-02-12T12:26:37.385176776Z","level":"INFO","msg":"handleInformInit: stream started","streamId":"v3d3ouvn","id":"127.0.0.1:48156"}
-{"time":"2025-02-12T12:26:37.805006529Z","level":"INFO","msg":"handleInformTeardown: server teardown initiated","id":"127.0.0.1:48156"}
-{"time":"2025-02-12T12:26:37.805113068Z","level":"INFO","msg":"server is shutting down"}
-{"time":"2025-02-12T12:26:37.805096358Z","level":"INFO","msg":"connection: closing","id":"127.0.0.1:48156"}
-{"time":"2025-02-12T12:26:37.805232397Z","level":"INFO","msg":"connection: closed successfully","id":"127.0.0.1:48156"}
-{"time":"2025-02-12T12:26:37.995286135Z","level":"ERROR","msg":"processOutgoingData: flush error","error":"write tcp 127.0.0.1:34237->127.0.0.1:48156: use of closed network connection","id":"127.0.0.1:48156"}
-{"time":"2025-02-12T12:26:39.120464204Z","level":"INFO","msg":"handleInformTeardown: server shutdown complete","id":"127.0.0.1:48156"}
-{"time":"2025-02-12T12:26:39.120492104Z","level":"INFO","msg":"connection: ManageConnectionData: connection closed","id":"127.0.0.1:48156"}
-{"time":"2025-02-12T12:26:39.120507034Z","level":"INFO","msg":"server is closed"}

wandb/run-20250212_122637-v3d3ouvn/logs/debug-internal.log

@@ -1,15 +0,0 @@
-{"time":"2025-02-12T12:26:37.280430379Z","level":"INFO","msg":"stream: starting","core version":"0.19.6","symlink path":"/home/tknika/xezpeleta/whisper/whisper-small-eu/wandb/run-20250212_122637-v3d3ouvn/logs/debug-core.log"}
-{"time":"2025-02-12T12:26:37.385120447Z","level":"INFO","msg":"created new stream","id":"v3d3ouvn"}
-{"time":"2025-02-12T12:26:37.385167976Z","level":"INFO","msg":"stream: started","id":"v3d3ouvn"}
-{"time":"2025-02-12T12:26:37.385225046Z","level":"INFO","msg":"writer: Do: started","stream_id":"v3d3ouvn"}
-{"time":"2025-02-12T12:26:37.385310785Z","level":"INFO","msg":"sender: started","stream_id":"v3d3ouvn"}
-{"time":"2025-02-12T12:26:37.385358905Z","level":"INFO","msg":"handler: started","stream_id":"v3d3ouvn"}
-{"time":"2025-02-12T12:26:37.656629021Z","level":"INFO","msg":"Starting system monitor"}
-{"time":"2025-02-12T12:26:37.805164318Z","level":"INFO","msg":"stream: closing","id":"v3d3ouvn"}
-{"time":"2025-02-12T12:26:37.805220128Z","level":"INFO","msg":"Stopping system monitor"}
-{"time":"2025-02-12T12:26:37.805952593Z","level":"INFO","msg":"Stopped system monitor"}
-{"time":"2025-02-12T12:26:38.904190518Z","level":"INFO","msg":"fileTransfer: Close: file transfer manager closed"}
-{"time":"2025-02-12T12:26:39.120209166Z","level":"INFO","msg":"handler: closed","stream_id":"v3d3ouvn"}
-{"time":"2025-02-12T12:26:39.120281046Z","level":"INFO","msg":"writer: Close: closed","stream_id":"v3d3ouvn"}
-{"time":"2025-02-12T12:26:39.120312915Z","level":"INFO","msg":"sender: closed","stream_id":"v3d3ouvn"}
-{"time":"2025-02-12T12:26:39.120355495Z","level":"INFO","msg":"stream: closed","id":"v3d3ouvn"}

wandb/run-20250212_122637-v3d3ouvn/logs/debug.log

@@ -1,26 +0,0 @@
-2025-02-12 12:26:37,062 INFO    MainThread:224110 [wandb_setup.py:_flush():68] Current SDK version is 0.19.6
-2025-02-12 12:26:37,062 INFO    MainThread:224110 [wandb_setup.py:_flush():68] Configure stats pid to 224110
-2025-02-12 12:26:37,062 INFO    MainThread:224110 [wandb_setup.py:_flush():68] Loading settings from /home/tknika/.config/wandb/settings
-2025-02-12 12:26:37,062 INFO    MainThread:224110 [wandb_setup.py:_flush():68] Loading settings from /home/tknika/xezpeleta/whisper/whisper-small-eu/wandb/settings
-2025-02-12 12:26:37,062 INFO    MainThread:224110 [wandb_setup.py:_flush():68] Loading settings from environment variables
-2025-02-12 12:26:37,062 INFO    MainThread:224110 [wandb_init.py:setup_run_log_directory():637] Logging user logs to /home/tknika/xezpeleta/whisper/whisper-small-eu/wandb/run-20250212_122637-v3d3ouvn/logs/debug.log
-2025-02-12 12:26:37,062 INFO    MainThread:224110 [wandb_init.py:setup_run_log_directory():638] Logging internal logs to /home/tknika/xezpeleta/whisper/whisper-small-eu/wandb/run-20250212_122637-v3d3ouvn/logs/debug-internal.log
-2025-02-12 12:26:37,062 INFO    MainThread:224110 [wandb_init.py:init():756] calling init triggers
-2025-02-12 12:26:37,062 INFO    MainThread:224110 [wandb_init.py:init():761] wandb.init called with sweep_config: {}
-config: {'_wandb': {}}
-2025-02-12 12:26:37,062 INFO    MainThread:224110 [wandb_init.py:init():789] starting backend
-2025-02-12 12:26:37,272 INFO    MainThread:224110 [wandb_init.py:init():793] sending inform_init request
-2025-02-12 12:26:37,277 INFO    MainThread:224110 [backend.py:_multiprocessing_setup():97] multiprocessing start_methods=fork,spawn,forkserver, using: spawn
-2025-02-12 12:26:37,277 INFO    MainThread:224110 [wandb_init.py:init():808] backend started and connected
-2025-02-12 12:26:37,279 INFO    MainThread:224110 [wandb_init.py:init():901] updated telemetry
-2025-02-12 12:26:37,285 INFO    MainThread:224110 [wandb_init.py:init():936] communicating run to backend with 90.0 second timeout
-2025-02-12 12:26:37,653 INFO    MainThread:224110 [wandb_init.py:init():994] starting run threads in backend
-2025-02-12 12:26:37,764 INFO    MainThread:224110 [wandb_run.py:_console_start():2385] atexit reg
-2025-02-12 12:26:37,765 INFO    MainThread:224110 [wandb_run.py:_redirect():2235] redirect: wrap_raw
-2025-02-12 12:26:37,765 INFO    MainThread:224110 [wandb_run.py:_redirect():2300] Wrapping output streams.
-2025-02-12 12:26:37,765 INFO    MainThread:224110 [wandb_run.py:_redirect():2325] Redirects installed.
-2025-02-12 12:26:37,766 INFO    MainThread:224110 [wandb_init.py:init():1036] run started, returning control to user process
-2025-02-12 12:26:37,767 INFO    MainThread:224110 [wandb_run.py:_config_callback():1253] config_cb None None {'vocab_size': 51865, 'num_mel_bins': 80, 'd_model': 768, 'encoder_layers': 12, 'encoder_attention_heads': 12, 'decoder_layers': 12, 'decoder_attention_heads': 12, 'decoder_ffn_dim': 3072, 'encoder_ffn_dim': 3072, 'dropout': 0.0, 'attention_dropout': 0.0, 'activation_dropout': 0.0, 'activation_function': 'gelu', 'init_std': 0.02, 'encoder_layerdrop': 0.0, 'decoder_layerdrop': 0.0, 'use_cache': False, 'num_hidden_layers': 12, 'scale_embedding': False, 'max_source_positions': 1500, 'max_target_positions': 448, 'classifier_proj_size': 256, 'use_weighted_layer_sum': False, 'apply_spec_augment': False, 'mask_time_prob': 0.05, 'mask_time_length': 10, 'mask_time_min_masks': 2, 'mask_feature_prob': 0.0, 'mask_feature_length': 10, 'mask_feature_min_masks': 0, 'median_filter_width': 7, 'return_dict': True, 'output_hidden_states': False, 'output_attentions': False, 'torchscript': False, 'torch_dtype': 'float32', 'use_bfloat16': False, 'tf_legacy_loss': False, 'pruned_heads': {}, 'tie_word_embeddings': True, 'chunk_size_feed_forward': 0, 'is_encoder_decoder': True, 'is_decoder': False, 'cross_attention_hidden_size': None, 'add_cross_attention': False, 'tie_encoder_decoder': False, 'max_length': 448, 'min_length': 0, 'do_sample': False, 'early_stopping': False, 'num_beams': 1, 'num_beam_groups': 1, 'diversity_penalty': 0.0, 'temperature': 1.0, 'top_k': 50, 'top_p': 1.0, 'typical_p': 1.0, 'repetition_penalty': 1.0, 'length_penalty': 1.0, 'no_repeat_ngram_size': 0, 'encoder_no_repeat_ngram_size': 0, 'bad_words_ids': None, 'num_return_sequences': 1, 'output_scores': False, 'return_dict_in_generate': False, 'forced_bos_token_id': None, 'forced_eos_token_id': None, 'remove_invalid_values': False, 'exponential_decay_length_penalty': None, 'suppress_tokens': None, 'begin_suppress_tokens': [220, 50257], 'architectures': ['WhisperForConditionalGeneration'], 'finetuning_task': None, 'id2label': {0: 'LABEL_0', 1: 'LABEL_1'}, 'label2id': {'LABEL_0': 0, 'LABEL_1': 1}, 'tokenizer_class': None, 'prefix': None, 'bos_token_id': 50257, 'pad_token_id': 50257, 'eos_token_id': 50257, 'sep_token_id': None, 'decoder_start_token_id': 50258, 'task_specific_params': None, 'problem_type': None, '_name_or_path': 'openai/whisper-small', '_attn_implementation_autoset': True, 'transformers_version': '4.49.0.dev0', 'forced_decoder_ids': None, 'model_type': 'whisper', 'output_dir': './', 'overwrite_output_dir': True, 'do_train': True, 'do_eval': True, 'do_predict': False, 'eval_strategy': 'steps', 'prediction_loss_only': False, 'per_device_train_batch_size': 32, 'per_device_eval_batch_size': 16, 'per_gpu_train_batch_size': None, 'per_gpu_eval_batch_size': None, 'gradient_accumulation_steps': 1, 'eval_accumulation_steps': None, 'eval_delay': 0, 'torch_empty_cache_steps': None, 'learning_rate': 1e-05, 'weight_decay': 0.0, 'adam_beta1': 0.9, 'adam_beta2': 0.999, 'adam_epsilon': 1e-08, 'max_grad_norm': 1.0, 'num_train_epochs': 3.0, 'max_steps': 8000, 'lr_scheduler_type': 'linear', 'lr_scheduler_kwargs': {}, 'warmup_ratio': 0.0, 'warmup_steps': 500, 'log_level': 'passive', 'log_level_replica': 'warning', 'log_on_each_node': True, 'logging_dir': './runs/Feb12_12-26-11_tknika', 'logging_strategy': 'steps', 'logging_first_step': False, 'logging_steps': 25, 'logging_nan_inf_filter': True, 'save_strategy': 'steps', 'save_steps': 1000, 'save_total_limit': None, 'save_safetensors': True, 'save_on_each_node': False, 'save_only_model': False, 'restore_callback_states_from_checkpoint': False, 'no_cuda': False, 'use_cpu': False, 'use_mps_device': False, 'seed': 42, 'data_seed': None, 'jit_mode_eval': False, 'use_ipex': False, 'bf16': False, 'fp16': True, 'fp16_opt_level': 'O1', 'half_precision_backend': 'auto', 'bf16_full_eval': False, 'fp16_full_eval': False, 'tf32': None, 'local_rank': 0, 'ddp_backend': None, 'tpu_num_cores': None, 'tpu_metrics_debug': False, 'debug': [], 'dataloader_drop_last': False, 'eval_steps': 1000, 'dataloader_num_workers': 0, 'dataloader_prefetch_factor': None, 'past_index': -1, 'run_name': 'whisper-small-eu', 'disable_tqdm': False, 'remove_unused_columns': True, 'label_names': None, 'load_best_model_at_end': True, 'metric_for_best_model': 'wer', 'greater_is_better': False, 'ignore_data_skip': False, 'fsdp': [], 'fsdp_min_num_params': 0, 'fsdp_config': {'min_num_params': 0, 'xla': False, 'xla_fsdp_v2': False, 'xla_fsdp_grad_ckpt': False}, 'fsdp_transformer_layer_cls_to_wrap': None, 'accelerator_config': {'split_batches': False, 'dispatch_batches': None, 'even_batches': True, 'use_seedable_sampler': True, 'non_blocking': False, 'gradient_accumulation_kwargs': None}, 'deepspeed': None, 'label_smoothing_factor': 0.0, 'optim': 'adamw_torch', 'optim_args': None, 'adafactor': False, 'group_by_length': False, 'length_column_name': 'input_length', 'report_to': ['wandb'], 'ddp_find_unused_parameters': None, 'ddp_bucket_cap_mb': None, 'ddp_broadcast_buffers': None, 'dataloader_pin_memory': True, 'dataloader_persistent_workers': False, 'skip_memory_metrics': True, 'use_legacy_prediction_loop': False, 'push_to_hub': True, 'resume_from_checkpoint': None, 'hub_model_id': None, 'hub_strategy': 'every_save', 'hub_token': '<HUB_TOKEN>', 'hub_private_repo': None, 'hub_always_push': False, 'gradient_checkpointing': True, 'gradient_checkpointing_kwargs': None, 'include_inputs_for_metrics': False, 'include_for_metrics': [], 'eval_do_concat_batches': True, 'fp16_backend': 'auto', 'evaluation_strategy': 'steps', 'push_to_hub_model_id': None, 'push_to_hub_organization': None, 'push_to_hub_token': '<PUSH_TO_HUB_TOKEN>', 'mp_parameters': '', 'auto_find_batch_size': False, 'full_determinism': False, 'torchdynamo': None, 'ray_scope': 'last', 'ddp_timeout': 1800, 'torch_compile': False, 'torch_compile_backend': None, 'torch_compile_mode': None, 'dispatch_batches': None, 'split_batches': None, 'include_tokens_per_second': False, 'include_num_input_tokens_seen': False, 'neftune_noise_alpha': None, 'optim_target_modules': None, 'batch_eval_metrics': False, 'eval_on_start': False, 'use_liger_kernel': False, 'eval_use_gather_object': False, 'average_tokens_across_devices': False, 'sortish_sampler': False, 'predict_with_generate': True, 'generation_max_length': 225, 'generation_num_beams': None, 'generation_config': None}
-2025-02-12 12:26:37,770 INFO    MainThread:224110 [wandb_config.py:__setitem__():154] config set model/num_parameters = 241734912 - <bound method Run._config_callback of <wandb.sdk.wandb_run.Run object at 0x7a5cbc15a330>>
-2025-02-12 12:26:37,770 INFO    MainThread:224110 [wandb_run.py:_config_callback():1253] config_cb model/num_parameters 241734912 None
-2025-02-12 12:26:37,805 WARNING MsgRouterThr:224110 [router.py:message_loop():75] message_loop has been closed

wandb/run-20250212_122854-4m048f5s/files/config.yaml

@@ -1,512 +0,0 @@
-_attn_implementation_autoset:
-    value: true
-_name_or_path:
-    value: openai/whisper-small
-_wandb:
-    value:
-        cli_version: 0.19.6
-        m:
-            - "1": train/global_step
-              "6":
-                - 3
-              "7": []
-        python_version: 3.12.3
-        t:
-            "1":
-                - 1
-                - 5
-                - 11
-                - 49
-                - 51
-                - 53
-                - 55
-                - 71
-                - 100
-            "2":
-                - 1
-                - 5
-                - 11
-                - 49
-                - 51
-                - 53
-                - 55
-                - 71
-                - 100
-            "3":
-                - 7
-                - 13
-                - 19
-                - 23
-                - 55
-                - 66
-            "4": 3.12.3
-            "5": 0.19.6
-            "6": 4.49.0.dev0
-            "8":
-                - 5
-            "9":
-                "1": transformers_trainer
-            "12": 0.19.6
-            "13": linux-x86_64
-accelerator_config:
-    value:
-        dispatch_batches: null
-        even_batches: true
-        gradient_accumulation_kwargs: null
-        non_blocking: false
-        split_batches: false
-        use_seedable_sampler: true
-activation_dropout:
-    value: 0
-activation_function:
-    value: gelu
-adafactor:
-    value: false
-adam_beta1:
-    value: 0.9
-adam_beta2:
-    value: 0.999
-adam_epsilon:
-    value: 1e-08
-add_cross_attention:
-    value: false
-apply_spec_augment:
-    value: false
-architectures:
-    value:
-        - WhisperForConditionalGeneration
-attention_dropout:
-    value: 0
-auto_find_batch_size:
-    value: false
-average_tokens_across_devices:
-    value: false
-bad_words_ids:
-    value: null
-batch_eval_metrics:
-    value: false
-begin_suppress_tokens:
-    value:
-        - 220
-        - 50257
-bf16:
-    value: false
-bf16_full_eval:
-    value: false
-bos_token_id:
-    value: 50257
-chunk_size_feed_forward:
-    value: 0
-classifier_proj_size:
-    value: 256
-cross_attention_hidden_size:
-    value: null
-d_model:
-    value: 768
-data_seed:
-    value: null
-dataloader_drop_last:
-    value: false
-dataloader_num_workers:
-    value: 0
-dataloader_persistent_workers:
-    value: false
-dataloader_pin_memory:
-    value: true
-dataloader_prefetch_factor:
-    value: null
-ddp_backend:
-    value: null
-ddp_broadcast_buffers:
-    value: null
-ddp_bucket_cap_mb:
-    value: null
-ddp_find_unused_parameters:
-    value: null
-ddp_timeout:
-    value: 1800
-debug:
-    value: []
-decoder_attention_heads:
-    value: 12
-decoder_ffn_dim:
-    value: 3072
-decoder_layerdrop:
-    value: 0
-decoder_layers:
-    value: 12
-decoder_start_token_id:
-    value: 50258
-deepspeed:
-    value: null
-disable_tqdm:
-    value: false
-dispatch_batches:
-    value: null
-diversity_penalty:
-    value: 0
-do_eval:
-    value: true
-do_predict:
-    value: false
-do_sample:
-    value: false
-do_train:
-    value: true
-dropout:
-    value: 0
-early_stopping:
-    value: false
-encoder_attention_heads:
-    value: 12
-encoder_ffn_dim:
-    value: 3072
-encoder_layerdrop:
-    value: 0
-encoder_layers:
-    value: 12
-encoder_no_repeat_ngram_size:
-    value: 0
-eos_token_id:
-    value: 50257
-eval_accumulation_steps:
-    value: null
-eval_delay:
-    value: 0
-eval_do_concat_batches:
-    value: true
-eval_on_start:
-    value: false
-eval_steps:
-    value: 1000
-eval_strategy:
-    value: steps
-eval_use_gather_object:
-    value: false
-evaluation_strategy:
-    value: steps
-exponential_decay_length_penalty:
-    value: null
-finetuning_task:
-    value: null
-forced_bos_token_id:
-    value: null
-forced_decoder_ids:
-    value: null
-forced_eos_token_id:
-    value: null
-fp16:
-    value: true
-fp16_backend:
-    value: auto
-fp16_full_eval:
-    value: false
-fp16_opt_level:
-    value: O1
-fsdp:
-    value: []
-fsdp_config:
-    value:
-        min_num_params: 0
-        xla: false
-        xla_fsdp_grad_ckpt: false
-        xla_fsdp_v2: false
-fsdp_min_num_params:
-    value: 0
-fsdp_transformer_layer_cls_to_wrap:
-    value: null
-full_determinism:
-    value: false
-generation_config:
-    value: null
-generation_max_length:
-    value: 225
-generation_num_beams:
-    value: null
-gradient_accumulation_steps:
-    value: 1
-gradient_checkpointing:
-    value: true
-gradient_checkpointing_kwargs:
-    value: null
-greater_is_better:
-    value: false
-group_by_length:
-    value: false
-half_precision_backend:
-    value: auto
-hub_always_push:
-    value: false
-hub_model_id:
-    value: null
-hub_private_repo:
-    value: null
-hub_strategy:
-    value: every_save
-hub_token:
-    value: <HUB_TOKEN>
-id2label:
-    value:
-        "0": LABEL_0
-        "1": LABEL_1
-ignore_data_skip:
-    value: false
-include_for_metrics:
-    value: []
-include_inputs_for_metrics:
-    value: false
-include_num_input_tokens_seen:
-    value: false
-include_tokens_per_second:
-    value: false
-init_std:
-    value: 0.02
-is_decoder:
-    value: false
-is_encoder_decoder:
-    value: true
-jit_mode_eval:
-    value: false
-label_names:
-    value: null
-label_smoothing_factor:
-    value: 0
-label2id:
-    value:
-        LABEL_0: 0
-        LABEL_1: 1
-learning_rate:
-    value: 1e-05
-length_column_name:
-    value: input_length
-length_penalty:
-    value: 1
-load_best_model_at_end:
-    value: true
-local_rank:
-    value: 0
-log_level:
-    value: passive
-log_level_replica:
-    value: warning
-log_on_each_node:
-    value: true
-logging_dir:
-    value: ./runs/Feb12_12-28-29_tknika
-logging_first_step:
-    value: false
-logging_nan_inf_filter:
-    value: true
-logging_steps:
-    value: 25
-logging_strategy:
-    value: steps
-lr_scheduler_type:
-    value: linear
-mask_feature_length:
-    value: 10
-mask_feature_min_masks:
-    value: 0
-mask_feature_prob:
-    value: 0
-mask_time_length:
-    value: 10
-mask_time_min_masks:
-    value: 2
-mask_time_prob:
-    value: 0.05
-max_grad_norm:
-    value: 1
-max_length:
-    value: 448
-max_source_positions:
-    value: 1500
-max_steps:
-    value: 8000
-max_target_positions:
-    value: 448
-median_filter_width:
-    value: 7
-metric_for_best_model:
-    value: wer
-min_length:
-    value: 0
-model/num_parameters:
-    value: 241734912
-model_type:
-    value: whisper
-mp_parameters:
-    value: ""
-neftune_noise_alpha:
-    value: null
-no_cuda:
-    value: false
-no_repeat_ngram_size:
-    value: 0
-num_beam_groups:
-    value: 1
-num_beams:
-    value: 1
-num_hidden_layers:
-    value: 12
-num_mel_bins:
-    value: 80
-num_return_sequences:
-    value: 1
-num_train_epochs:
-    value: 3
-optim:
-    value: adamw_torch
-optim_args:
-    value: null
-optim_target_modules:
-    value: null
-output_attentions:
-    value: false
-output_dir:
-    value: ./
-output_hidden_states:
-    value: false
-output_scores:
-    value: false
-overwrite_output_dir:
-    value: true
-pad_token_id:
-    value: 50257
-past_index:
-    value: -1
-per_device_eval_batch_size:
-    value: 16
-per_device_train_batch_size:
-    value: 32
-per_gpu_eval_batch_size:
-    value: null
-per_gpu_train_batch_size:
-    value: null
-predict_with_generate:
-    value: true
-prediction_loss_only:
-    value: false
-prefix:
-    value: null
-problem_type:
-    value: null
-push_to_hub:
-    value: true
-push_to_hub_model_id:
-    value: null
-push_to_hub_organization:
-    value: null
-push_to_hub_token:
-    value: <PUSH_TO_HUB_TOKEN>
-ray_scope:
-    value: last
-remove_invalid_values:
-    value: false
-remove_unused_columns:
-    value: true
-repetition_penalty:
-    value: 1
-report_to:
-    value:
-        - wandb
-restore_callback_states_from_checkpoint:
-    value: false
-resume_from_checkpoint:
-    value: null
-return_dict:
-    value: true
-return_dict_in_generate:
-    value: false
-run_name:
-    value: whisper-small-eu
-save_on_each_node:
-    value: false
-save_only_model:
-    value: false
-save_safetensors:
-    value: true
-save_steps:
-    value: 1000
-save_strategy:
-    value: steps
-save_total_limit:
-    value: null
-scale_embedding:
-    value: false
-seed:
-    value: 42
-sep_token_id:
-    value: null
-skip_memory_metrics:
-    value: true
-sortish_sampler:
-    value: false
-split_batches:
-    value: null
-suppress_tokens:
-    value: null
-task_specific_params:
-    value: null
-temperature:
-    value: 1
-tf_legacy_loss:
-    value: false
-tf32:
-    value: null
-tie_encoder_decoder:
-    value: false
-tie_word_embeddings:
-    value: true
-tokenizer_class:
-    value: null
-top_k:
-    value: 50
-top_p:
-    value: 1
-torch_compile:
-    value: false
-torch_compile_backend:
-    value: null
-torch_compile_mode:
-    value: null
-torch_dtype:
-    value: float32
-torch_empty_cache_steps:
-    value: null
-torchdynamo:
-    value: null
-torchscript:
-    value: false
-tpu_metrics_debug:
-    value: false
-tpu_num_cores:
-    value: null
-transformers_version:
-    value: 4.49.0.dev0
-typical_p:
-    value: 1
-use_bfloat16:
-    value: false
-use_cache:
-    value: false
-use_cpu:
-    value: false
-use_ipex:
-    value: false
-use_legacy_prediction_loop:
-    value: false
-use_liger_kernel:
-    value: false
-use_mps_device:
-    value: false
-use_weighted_layer_sum:
-    value: false
-vocab_size:
-    value: 51865
-warmup_ratio:
-    value: 0
-warmup_steps:
-    value: 500
-weight_decay:
-    value: 0

wandb/run-20250212_122854-4m048f5s/files/output.log

@@ -1,22 +0,0 @@
-  0%|                                                                                                                                               | 0/8000 [00:00<?, ?it/s]Traceback (most recent call last):
-  File "/home/tknika/xezpeleta/whisper/whisper-small-eu/run_speech_recognition_seq2seq_streaming.py", line 630, in <module>
-    main()
-  File "/home/tknika/xezpeleta/whisper/whisper-small-eu/run_speech_recognition_seq2seq_streaming.py", line 579, in main
-    train_result = trainer.train(resume_from_checkpoint=checkpoint)
-                   ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
-  File "/home/tknika/xezpeleta/whisper/whisper-small-eu/.venv/lib/python3.12/site-packages/transformers/trainer.py", line 2197, in train
-    return inner_training_loop(
-           ^^^^^^^^^^^^^^^^^^^^
-  File "/home/tknika/xezpeleta/whisper/whisper-small-eu/.venv/lib/python3.12/site-packages/transformers/trainer.py", line 2438, in _inner_training_loop
-    self.control = self.callback_handler.on_epoch_begin(args, self.state, self.control)
-                   ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
-  File "/home/tknika/xezpeleta/whisper/whisper-small-eu/.venv/lib/python3.12/site-packages/transformers/trainer_callback.py", line 514, in on_epoch_begin
-    return self.call_event("on_epoch_begin", args, state, control)
-           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
-  File "/home/tknika/xezpeleta/whisper/whisper-small-eu/.venv/lib/python3.12/site-packages/transformers/trainer_callback.py", line 557, in call_event
-    result = getattr(callback, event)(
-             ^^^^^^^^^^^^^^^^^^^^^^^^^
-  File "/home/tknika/xezpeleta/whisper/whisper-small-eu/run_speech_recognition_seq2seq_streaming.py", line 555, in on_epoch_begin
-    if isinstance(train_dataloader.dataset, IterableDatasetShard):
-                  ^^^^^^^^^^^^^^^^^^^^^^^^
-AttributeError: 'NoneType' object has no attribute 'dataset'

wandb/run-20250212_122854-4m048f5s/files/requirements.txt

@@ -1,115 +0,0 @@
-aiosignal==1.3.2
-Markdown==3.7
-more-itertools==10.6.0
-requests==2.32.3
-sentry-sdk==2.21.0
-torchaudio==2.6.0
-charset-normalizer==3.4.1
-docker-pycreds==0.4.0
-nvidia-cusolver-cu12==11.6.1.9
-PyYAML==6.0.2
-librosa==0.10.2.post1
-soxr==0.5.0.post1
-multiprocess==0.70.16
-setuptools==75.8.0
-nvidia-cufft-cu12==11.2.1.3
-joblib==1.4.2
-pytz==2025.1
-pip==24.0
-scikit-learn==1.6.1
-certifi==2025.1.31
-jiwer==3.1.0
-regex==2024.11.6
-annotated-types==0.7.0
-grpcio==1.70.0
-msgpack==1.1.0
-mpmath==1.3.0
-nvidia-cudnn-cu12==9.1.0.70
-soundfile==0.13.1
-dill==0.3.8
-nvidia-nvtx-cu12==12.4.127
-six==1.17.0
-nvidia-cuda-cupti-cu12==12.4.127
-pyarrow==19.0.0
-nvidia-nccl-cu12==2.21.5
-psutil==6.1.1
-decorator==5.1.1
-llvmlite==0.44.0
-frozenlist==1.5.0
-pydantic==2.10.6
-networkx==3.4.2
-idna==3.10
-wandb==0.19.6
-aiohttp==3.11.12
-RapidFuzz==3.12.1
-pandas==2.2.3
-python-dateutil==2.9.0.post0
-numpy==2.1.3
-tokenizers==0.21.0
-nvidia-cusparselt-cu12==0.6.2
-typing_extensions==4.12.2
-urllib3==2.3.0
-setproctitle==1.3.4
-tzdata==2025.1
-sympy==1.13.1
-pooch==1.8.2
-click==8.1.8
-pydantic_core==2.27.2
-MarkupSafe==3.0.2
-scipy==1.15.1
-accelerate==1.3.0
-tensorboard==2.19.0
-protobuf==5.29.3
-gitdb==4.0.12
-smmap==5.0.2
-absl-py==2.1.0
-tqdm==4.67.1
-yarl==1.18.3
-pycparser==2.22
-nvidia-cusparse-cu12==12.3.1.170
-attrs==25.1.0
-lazy_loader==0.4
-tensorboard-data-server==0.7.2
-threadpoolctl==3.5.0
-GitPython==3.1.44
-safetensors==0.5.2
-fsspec==2024.12.0
-nvidia-cuda-nvrtc-cu12==12.4.127
-filelock==3.17.0
-aiohappyeyeballs==2.4.6
-packaging==24.2
-datasets==3.2.1.dev0
-audioread==3.0.1
-propcache==0.2.1
-transformers==4.49.0.dev0
-nvidia-cuda-runtime-cu12==12.4.127
-cffi==1.17.1
-evaluate==0.4.3
-Werkzeug==3.1.3
-huggingface-hub==0.28.1
-Jinja2==3.1.5
-torch==2.6.0
-nvidia-curand-cu12==10.3.5.147
-xxhash==3.5.0
-platformdirs==4.3.6
-multidict==6.1.0
-nvidia-cublas-cu12==12.4.5.8
-nvidia-nvjitlink-cu12==12.4.127
-triton==3.2.0
-numba==0.61.0
-importlib_metadata==8.0.0
-platformdirs==4.2.2
-typeguard==4.3.0
-more-itertools==10.3.0
-tomli==2.0.1
-autocommand==2.2.2
-zipp==3.19.2
-typing_extensions==4.12.2
-backports.tarfile==1.2.0
-inflect==7.3.1
-jaraco.text==3.12.1
-wheel==0.43.0
-packaging==24.2
-jaraco.collections==5.1.0
-jaraco.functools==4.0.1
-jaraco.context==5.3.0

wandb/run-20250212_122854-4m048f5s/files/wandb-metadata.json

@@ -1,85 +0,0 @@
-{
-  "os":  "Linux-6.8.0-48-generic-x86_64-with-glibc2.39",
-  "python":  "CPython 3.12.3",
-  "startedAt":  "2025-02-12T12:28:54.528397Z",
-  "args":  [
-    "--model_name_or_path=openai/whisper-small",
-    "--dataset_name=asierhv/composite_corpus_eu_v2.1",
-    "--language=basque",
-    "--train_split_name=train",
-    "--eval_split_name=dev_parl+test_parl+test_cv+test_oslr",
-    "--model_index_name=Whisper Small Basque",
-    "--max_steps=8000",
-    "--output_dir=./",
-    "--per_device_train_batch_size=32",
-    "--per_device_eval_batch_size=16",
-    "--gradient_accumulation_steps=1",
-    "--logging_steps=25",
-    "--learning_rate=1e-5",
-    "--warmup_steps=500",
-    "--evaluation_strategy=steps",
-    "--eval_steps=1000",
-    "--save_strategy=steps",
-    "--save_steps=1000",
-    "--generation_max_length=225",
-    "--length_column_name=input_length",
-    "--max_duration_in_seconds=30",
-    "--text_column_name=sentence",
-    "--freeze_feature_encoder=False",
-    "--report_to=tensorboard",
-    "--metric_for_best_model=wer",
-    "--greater_is_better=False",
-    "--load_best_model_at_end",
-    "--gradient_checkpointing",
-    "--fp16",
-    "--overwrite_output_dir",
-    "--do_train",
-    "--do_eval",
-    "--predict_with_generate",
-    "--do_normalize_eval",
-    "--streaming",
-    "--use_auth_token",
-    "--push_to_hub",
-    "--report_to",
-    "wandb",
-    "--run_name",
-    "whisper-small-eu"
-  ],
-  "program":  "/home/tknika/xezpeleta/whisper/whisper-small-eu/run_speech_recognition_seq2seq_streaming.py",
-  "codePath":  "run_speech_recognition_seq2seq_streaming.py",
-  "git":  {
-    "remote":  "https://huggingface.co/xezpeleta/whisper-small-eu",
-    "commit":  "9c975864b20b4df94398a870e97cad2934253ec3"
-  },
-  "email":  "[email protected]",
-  "root":  "/home/tknika/xezpeleta/whisper/whisper-small-eu",
-  "host":  "tknika",
-  "executable":  "/home/tknika/xezpeleta/whisper/whisper-small-eu/.venv/bin/python",
-  "codePathLocal":  "run_speech_recognition_seq2seq_streaming.py",
-  "cpu_count":  8,
-  "cpu_count_logical":  8,
-  "gpu":  "NVIDIA L40-48Q",
-  "gpu_count":  1,
-  "disk":  {
-    "/":  {
-      "total":  "525987168256",
-      "used":  "297346756608"
-    }
-  },
-  "memory":  {
-    "total":  "33654022144"
-  },
-  "cpu":  {
-    "count":  8,
-    "countLogical":  8
-  },
-  "gpu_nvidia":  [
-    {
-      "name":  "NVIDIA L40-48Q",
-      "memoryTotal":  "51539607552",
-      "cudaCores":  18176,
-      "architecture":  "Ada"
-    }
-  ],
-  "cudaVersion":  "12.4"
-}

wandb/run-20250212_122854-4m048f5s/files/wandb-summary.json

@@ -1 +0,0 @@
-{"_wandb":{"runtime":0}}

wandb/run-20250212_122854-4m048f5s/logs/debug-core.log

@@ -1,14 +0,0 @@
-{"time":"2025-02-12T12:28:54.343223143Z","level":"INFO","msg":"main: starting server","port-filename":"/tmp/tmpmg8o5mqm/port-224528.txt","pid":224528,"log-level":0,"disable-analytics":false,"shutdown-on-parent-exit":false}
-{"time":"2025-02-12T12:28:54.34827505Z","level":"INFO","msg":"Will exit if parent process dies.","ppid":224528}
-{"time":"2025-02-12T12:28:54.34821581Z","level":"INFO","msg":"server is running","addr":{"IP":"127.0.0.1","Port":44981,"Zone":""}}
-{"time":"2025-02-12T12:28:54.521681286Z","level":"INFO","msg":"connection: ManageConnectionData: new connection created","id":"127.0.0.1:51116"}
-{"time":"2025-02-12T12:28:54.53173104Z","level":"INFO","msg":"handleInformInit: received","streamId":"4m048f5s","id":"127.0.0.1:51116"}
-{"time":"2025-02-12T12:28:54.636478984Z","level":"INFO","msg":"handleInformInit: stream started","streamId":"4m048f5s","id":"127.0.0.1:51116"}
-{"time":"2025-02-12T12:28:55.028718067Z","level":"INFO","msg":"handleInformTeardown: server teardown initiated","id":"127.0.0.1:51116"}
-{"time":"2025-02-12T12:28:55.028819337Z","level":"INFO","msg":"connection: closing","id":"127.0.0.1:51116"}
-{"time":"2025-02-12T12:28:55.028818347Z","level":"INFO","msg":"server is shutting down"}
-{"time":"2025-02-12T12:28:55.028912476Z","level":"INFO","msg":"connection: closed successfully","id":"127.0.0.1:51116"}
-{"time":"2025-02-12T12:28:55.368512133Z","level":"ERROR","msg":"processOutgoingData: flush error","error":"write tcp 127.0.0.1:44981->127.0.0.1:51116: use of closed network connection","id":"127.0.0.1:51116"}
-{"time":"2025-02-12T12:28:56.249016671Z","level":"INFO","msg":"handleInformTeardown: server shutdown complete","id":"127.0.0.1:51116"}
-{"time":"2025-02-12T12:28:56.249048031Z","level":"INFO","msg":"connection: ManageConnectionData: connection closed","id":"127.0.0.1:51116"}
-{"time":"2025-02-12T12:28:56.249109521Z","level":"INFO","msg":"server is closed"}

wandb/run-20250212_122854-4m048f5s/logs/debug-internal.log

@@ -1,15 +0,0 @@
-{"time":"2025-02-12T12:28:54.532033248Z","level":"INFO","msg":"stream: starting","core version":"0.19.6","symlink path":"/home/tknika/xezpeleta/whisper/whisper-small-eu/wandb/run-20250212_122854-4m048f5s/logs/debug-core.log"}
-{"time":"2025-02-12T12:28:54.636425775Z","level":"INFO","msg":"created new stream","id":"4m048f5s"}
-{"time":"2025-02-12T12:28:54.636473304Z","level":"INFO","msg":"stream: started","id":"4m048f5s"}
-{"time":"2025-02-12T12:28:54.636556744Z","level":"INFO","msg":"writer: Do: started","stream_id":"4m048f5s"}
-{"time":"2025-02-12T12:28:54.636597734Z","level":"INFO","msg":"handler: started","stream_id":"4m048f5s"}
-{"time":"2025-02-12T12:28:54.636670993Z","level":"INFO","msg":"sender: started","stream_id":"4m048f5s"}
-{"time":"2025-02-12T12:28:54.886030488Z","level":"INFO","msg":"Starting system monitor"}
-{"time":"2025-02-12T12:28:55.028853626Z","level":"INFO","msg":"stream: closing","id":"4m048f5s"}
-{"time":"2025-02-12T12:28:55.028891716Z","level":"INFO","msg":"Stopping system monitor"}
-{"time":"2025-02-12T12:28:55.029589382Z","level":"INFO","msg":"Stopped system monitor"}
-{"time":"2025-02-12T12:28:56.017176821Z","level":"INFO","msg":"fileTransfer: Close: file transfer manager closed"}
-{"time":"2025-02-12T12:28:56.248703933Z","level":"INFO","msg":"handler: closed","stream_id":"4m048f5s"}
-{"time":"2025-02-12T12:28:56.248768363Z","level":"INFO","msg":"writer: Close: closed","stream_id":"4m048f5s"}
-{"time":"2025-02-12T12:28:56.248802103Z","level":"INFO","msg":"sender: closed","stream_id":"4m048f5s"}
-{"time":"2025-02-12T12:28:56.248896982Z","level":"INFO","msg":"stream: closed","id":"4m048f5s"}

wandb/run-20250212_122854-4m048f5s/logs/debug.log

@@ -1,26 +0,0 @@
-2025-02-12 12:28:54,316 INFO    MainThread:224528 [wandb_setup.py:_flush():68] Current SDK version is 0.19.6
-2025-02-12 12:28:54,316 INFO    MainThread:224528 [wandb_setup.py:_flush():68] Configure stats pid to 224528
-2025-02-12 12:28:54,316 INFO    MainThread:224528 [wandb_setup.py:_flush():68] Loading settings from /home/tknika/.config/wandb/settings
-2025-02-12 12:28:54,316 INFO    MainThread:224528 [wandb_setup.py:_flush():68] Loading settings from /home/tknika/xezpeleta/whisper/whisper-small-eu/wandb/settings
-2025-02-12 12:28:54,316 INFO    MainThread:224528 [wandb_setup.py:_flush():68] Loading settings from environment variables
-2025-02-12 12:28:54,316 INFO    MainThread:224528 [wandb_init.py:setup_run_log_directory():637] Logging user logs to /home/tknika/xezpeleta/whisper/whisper-small-eu/wandb/run-20250212_122854-4m048f5s/logs/debug.log
-2025-02-12 12:28:54,316 INFO    MainThread:224528 [wandb_init.py:setup_run_log_directory():638] Logging internal logs to /home/tknika/xezpeleta/whisper/whisper-small-eu/wandb/run-20250212_122854-4m048f5s/logs/debug-internal.log
-2025-02-12 12:28:54,316 INFO    MainThread:224528 [wandb_init.py:init():756] calling init triggers
-2025-02-12 12:28:54,316 INFO    MainThread:224528 [wandb_init.py:init():761] wandb.init called with sweep_config: {}
-config: {'_wandb': {}}
-2025-02-12 12:28:54,316 INFO    MainThread:224528 [wandb_init.py:init():789] starting backend
-2025-02-12 12:28:54,521 INFO    MainThread:224528 [wandb_init.py:init():793] sending inform_init request
-2025-02-12 12:28:54,527 INFO    MainThread:224528 [backend.py:_multiprocessing_setup():97] multiprocessing start_methods=fork,spawn,forkserver, using: spawn
-2025-02-12 12:28:54,528 INFO    MainThread:224528 [wandb_init.py:init():808] backend started and connected
-2025-02-12 12:28:54,530 INFO    MainThread:224528 [wandb_init.py:init():901] updated telemetry
-2025-02-12 12:28:54,537 INFO    MainThread:224528 [wandb_init.py:init():936] communicating run to backend with 90.0 second timeout
-2025-02-12 12:28:54,883 INFO    MainThread:224528 [wandb_init.py:init():994] starting run threads in backend
-2025-02-12 12:28:54,988 INFO    MainThread:224528 [wandb_run.py:_console_start():2385] atexit reg
-2025-02-12 12:28:54,989 INFO    MainThread:224528 [wandb_run.py:_redirect():2235] redirect: wrap_raw
-2025-02-12 12:28:54,989 INFO    MainThread:224528 [wandb_run.py:_redirect():2300] Wrapping output streams.
-2025-02-12 12:28:54,989 INFO    MainThread:224528 [wandb_run.py:_redirect():2325] Redirects installed.
-2025-02-12 12:28:54,990 INFO    MainThread:224528 [wandb_init.py:init():1036] run started, returning control to user process
-2025-02-12 12:28:54,991 INFO    MainThread:224528 [wandb_run.py:_config_callback():1253] config_cb None None {'vocab_size': 51865, 'num_mel_bins': 80, 'd_model': 768, 'encoder_layers': 12, 'encoder_attention_heads': 12, 'decoder_layers': 12, 'decoder_attention_heads': 12, 'decoder_ffn_dim': 3072, 'encoder_ffn_dim': 3072, 'dropout': 0.0, 'attention_dropout': 0.0, 'activation_dropout': 0.0, 'activation_function': 'gelu', 'init_std': 0.02, 'encoder_layerdrop': 0.0, 'decoder_layerdrop': 0.0, 'use_cache': False, 'num_hidden_layers': 12, 'scale_embedding': False, 'max_source_positions': 1500, 'max_target_positions': 448, 'classifier_proj_size': 256, 'use_weighted_layer_sum': False, 'apply_spec_augment': False, 'mask_time_prob': 0.05, 'mask_time_length': 10, 'mask_time_min_masks': 2, 'mask_feature_prob': 0.0, 'mask_feature_length': 10, 'mask_feature_min_masks': 0, 'median_filter_width': 7, 'return_dict': True, 'output_hidden_states': False, 'output_attentions': False, 'torchscript': False, 'torch_dtype': 'float32', 'use_bfloat16': False, 'tf_legacy_loss': False, 'pruned_heads': {}, 'tie_word_embeddings': True, 'chunk_size_feed_forward': 0, 'is_encoder_decoder': True, 'is_decoder': False, 'cross_attention_hidden_size': None, 'add_cross_attention': False, 'tie_encoder_decoder': False, 'max_length': 448, 'min_length': 0, 'do_sample': False, 'early_stopping': False, 'num_beams': 1, 'num_beam_groups': 1, 'diversity_penalty': 0.0, 'temperature': 1.0, 'top_k': 50, 'top_p': 1.0, 'typical_p': 1.0, 'repetition_penalty': 1.0, 'length_penalty': 1.0, 'no_repeat_ngram_size': 0, 'encoder_no_repeat_ngram_size': 0, 'bad_words_ids': None, 'num_return_sequences': 1, 'output_scores': False, 'return_dict_in_generate': False, 'forced_bos_token_id': None, 'forced_eos_token_id': None, 'remove_invalid_values': False, 'exponential_decay_length_penalty': None, 'suppress_tokens': None, 'begin_suppress_tokens': [220, 50257], 'architectures': ['WhisperForConditionalGeneration'], 'finetuning_task': None, 'id2label': {0: 'LABEL_0', 1: 'LABEL_1'}, 'label2id': {'LABEL_0': 0, 'LABEL_1': 1}, 'tokenizer_class': None, 'prefix': None, 'bos_token_id': 50257, 'pad_token_id': 50257, 'eos_token_id': 50257, 'sep_token_id': None, 'decoder_start_token_id': 50258, 'task_specific_params': None, 'problem_type': None, '_name_or_path': 'openai/whisper-small', '_attn_implementation_autoset': True, 'transformers_version': '4.49.0.dev0', 'forced_decoder_ids': None, 'model_type': 'whisper', 'output_dir': './', 'overwrite_output_dir': True, 'do_train': True, 'do_eval': True, 'do_predict': False, 'eval_strategy': 'steps', 'prediction_loss_only': False, 'per_device_train_batch_size': 32, 'per_device_eval_batch_size': 16, 'per_gpu_train_batch_size': None, 'per_gpu_eval_batch_size': None, 'gradient_accumulation_steps': 1, 'eval_accumulation_steps': None, 'eval_delay': 0, 'torch_empty_cache_steps': None, 'learning_rate': 1e-05, 'weight_decay': 0.0, 'adam_beta1': 0.9, 'adam_beta2': 0.999, 'adam_epsilon': 1e-08, 'max_grad_norm': 1.0, 'num_train_epochs': 3.0, 'max_steps': 8000, 'lr_scheduler_type': 'linear', 'lr_scheduler_kwargs': {}, 'warmup_ratio': 0.0, 'warmup_steps': 500, 'log_level': 'passive', 'log_level_replica': 'warning', 'log_on_each_node': True, 'logging_dir': './runs/Feb12_12-28-29_tknika', 'logging_strategy': 'steps', 'logging_first_step': False, 'logging_steps': 25, 'logging_nan_inf_filter': True, 'save_strategy': 'steps', 'save_steps': 1000, 'save_total_limit': None, 'save_safetensors': True, 'save_on_each_node': False, 'save_only_model': False, 'restore_callback_states_from_checkpoint': False, 'no_cuda': False, 'use_cpu': False, 'use_mps_device': False, 'seed': 42, 'data_seed': None, 'jit_mode_eval': False, 'use_ipex': False, 'bf16': False, 'fp16': True, 'fp16_opt_level': 'O1', 'half_precision_backend': 'auto', 'bf16_full_eval': False, 'fp16_full_eval': False, 'tf32': None, 'local_rank': 0, 'ddp_backend': None, 'tpu_num_cores': None, 'tpu_metrics_debug': False, 'debug': [], 'dataloader_drop_last': False, 'eval_steps': 1000, 'dataloader_num_workers': 0, 'dataloader_prefetch_factor': None, 'past_index': -1, 'run_name': 'whisper-small-eu', 'disable_tqdm': False, 'remove_unused_columns': True, 'label_names': None, 'load_best_model_at_end': True, 'metric_for_best_model': 'wer', 'greater_is_better': False, 'ignore_data_skip': False, 'fsdp': [], 'fsdp_min_num_params': 0, 'fsdp_config': {'min_num_params': 0, 'xla': False, 'xla_fsdp_v2': False, 'xla_fsdp_grad_ckpt': False}, 'fsdp_transformer_layer_cls_to_wrap': None, 'accelerator_config': {'split_batches': False, 'dispatch_batches': None, 'even_batches': True, 'use_seedable_sampler': True, 'non_blocking': False, 'gradient_accumulation_kwargs': None}, 'deepspeed': None, 'label_smoothing_factor': 0.0, 'optim': 'adamw_torch', 'optim_args': None, 'adafactor': False, 'group_by_length': False, 'length_column_name': 'input_length', 'report_to': ['wandb'], 'ddp_find_unused_parameters': None, 'ddp_bucket_cap_mb': None, 'ddp_broadcast_buffers': None, 'dataloader_pin_memory': True, 'dataloader_persistent_workers': False, 'skip_memory_metrics': True, 'use_legacy_prediction_loop': False, 'push_to_hub': True, 'resume_from_checkpoint': None, 'hub_model_id': None, 'hub_strategy': 'every_save', 'hub_token': '<HUB_TOKEN>', 'hub_private_repo': None, 'hub_always_push': False, 'gradient_checkpointing': True, 'gradient_checkpointing_kwargs': None, 'include_inputs_for_metrics': False, 'include_for_metrics': [], 'eval_do_concat_batches': True, 'fp16_backend': 'auto', 'evaluation_strategy': 'steps', 'push_to_hub_model_id': None, 'push_to_hub_organization': None, 'push_to_hub_token': '<PUSH_TO_HUB_TOKEN>', 'mp_parameters': '', 'auto_find_batch_size': False, 'full_determinism': False, 'torchdynamo': None, 'ray_scope': 'last', 'ddp_timeout': 1800, 'torch_compile': False, 'torch_compile_backend': None, 'torch_compile_mode': None, 'dispatch_batches': None, 'split_batches': None, 'include_tokens_per_second': False, 'include_num_input_tokens_seen': False, 'neftune_noise_alpha': None, 'optim_target_modules': None, 'batch_eval_metrics': False, 'eval_on_start': False, 'use_liger_kernel': False, 'eval_use_gather_object': False, 'average_tokens_across_devices': False, 'sortish_sampler': False, 'predict_with_generate': True, 'generation_max_length': 225, 'generation_num_beams': None, 'generation_config': None}
-2025-02-12 12:28:54,995 INFO    MainThread:224528 [wandb_config.py:__setitem__():154] config set model/num_parameters = 241734912 - <bound method Run._config_callback of <wandb.sdk.wandb_run.Run object at 0x71c5f6c57cb0>>
-2025-02-12 12:28:54,995 INFO    MainThread:224528 [wandb_run.py:_config_callback():1253] config_cb model/num_parameters 241734912 None
-2025-02-12 12:28:55,029 WARNING MsgRouterThr:224528 [router.py:message_loop():75] message_loop has been closed

wandb/run-20250212_125202-c6xjc1gs/files/config.yaml

@@ -1,512 +0,0 @@
-_attn_implementation_autoset:
-    value: true
-_name_or_path:
-    value: openai/whisper-small
-_wandb:
-    value:
-        cli_version: 0.19.6
-        m:
-            - "1": train/global_step
-              "6":
-                - 3
-              "7": []
-        python_version: 3.12.3
-        t:
-            "1":
-                - 1
-                - 5
-                - 11
-                - 49
-                - 51
-                - 53
-                - 55
-                - 71
-                - 100
-            "2":
-                - 1
-                - 5
-                - 11
-                - 49
-                - 51
-                - 53
-                - 55
-                - 71
-                - 100
-            "3":
-                - 7
-                - 13
-                - 19
-                - 23
-                - 55
-                - 66
-            "4": 3.12.3
-            "5": 0.19.6
-            "6": 4.49.0.dev0
-            "8":
-                - 5
-            "9":
-                "1": transformers_trainer
-            "12": 0.19.6
-            "13": linux-x86_64
-accelerator_config:
-    value:
-        dispatch_batches: null
-        even_batches: true
-        gradient_accumulation_kwargs: null
-        non_blocking: false
-        split_batches: false
-        use_seedable_sampler: true
-activation_dropout:
-    value: 0
-activation_function:
-    value: gelu
-adafactor:
-    value: false
-adam_beta1:
-    value: 0.9
-adam_beta2:
-    value: 0.999
-adam_epsilon:
-    value: 1e-08
-add_cross_attention:
-    value: false
-apply_spec_augment:
-    value: false
-architectures:
-    value:
-        - WhisperForConditionalGeneration
-attention_dropout:
-    value: 0
-auto_find_batch_size:
-    value: false
-average_tokens_across_devices:
-    value: false
-bad_words_ids:
-    value: null
-batch_eval_metrics:
-    value: false
-begin_suppress_tokens:
-    value:
-        - 220
-        - 50257
-bf16:
-    value: false
-bf16_full_eval:
-    value: false
-bos_token_id:
-    value: 50257
-chunk_size_feed_forward:
-    value: 0
-classifier_proj_size:
-    value: 256
-cross_attention_hidden_size:
-    value: null
-d_model:
-    value: 768
-data_seed:
-    value: null
-dataloader_drop_last:
-    value: false
-dataloader_num_workers:
-    value: 0
-dataloader_persistent_workers:
-    value: false
-dataloader_pin_memory:
-    value: true
-dataloader_prefetch_factor:
-    value: null
-ddp_backend:
-    value: null
-ddp_broadcast_buffers:
-    value: null
-ddp_bucket_cap_mb:
-    value: null
-ddp_find_unused_parameters:
-    value: null
-ddp_timeout:
-    value: 1800
-debug:
-    value: []
-decoder_attention_heads:
-    value: 12
-decoder_ffn_dim:
-    value: 3072
-decoder_layerdrop:
-    value: 0
-decoder_layers:
-    value: 12
-decoder_start_token_id:
-    value: 50258
-deepspeed:
-    value: null
-disable_tqdm:
-    value: false
-dispatch_batches:
-    value: null
-diversity_penalty:
-    value: 0
-do_eval:
-    value: true
-do_predict:
-    value: false
-do_sample:
-    value: false
-do_train:
-    value: true
-dropout:
-    value: 0
-early_stopping:
-    value: false
-encoder_attention_heads:
-    value: 12
-encoder_ffn_dim:
-    value: 3072
-encoder_layerdrop:
-    value: 0
-encoder_layers:
-    value: 12
-encoder_no_repeat_ngram_size:
-    value: 0
-eos_token_id:
-    value: 50257
-eval_accumulation_steps:
-    value: null
-eval_delay:
-    value: 0
-eval_do_concat_batches:
-    value: true
-eval_on_start:
-    value: false
-eval_steps:
-    value: 1000
-eval_strategy:
-    value: steps
-eval_use_gather_object:
-    value: false
-evaluation_strategy:
-    value: steps
-exponential_decay_length_penalty:
-    value: null
-finetuning_task:
-    value: null
-forced_bos_token_id:
-    value: null
-forced_decoder_ids:
-    value: null
-forced_eos_token_id:
-    value: null
-fp16:
-    value: true
-fp16_backend:
-    value: auto
-fp16_full_eval:
-    value: false
-fp16_opt_level:
-    value: O1
-fsdp:
-    value: []
-fsdp_config:
-    value:
-        min_num_params: 0
-        xla: false
-        xla_fsdp_grad_ckpt: false
-        xla_fsdp_v2: false
-fsdp_min_num_params:
-    value: 0
-fsdp_transformer_layer_cls_to_wrap:
-    value: null
-full_determinism:
-    value: false
-generation_config:
-    value: null
-generation_max_length:
-    value: 225
-generation_num_beams:
-    value: null
-gradient_accumulation_steps:
-    value: 1
-gradient_checkpointing:
-    value: true
-gradient_checkpointing_kwargs:
-    value: null
-greater_is_better:
-    value: false
-group_by_length:
-    value: false
-half_precision_backend:
-    value: auto
-hub_always_push:
-    value: false
-hub_model_id:
-    value: null
-hub_private_repo:
-    value: null
-hub_strategy:
-    value: every_save
-hub_token:
-    value: <HUB_TOKEN>
-id2label:
-    value:
-        "0": LABEL_0
-        "1": LABEL_1
-ignore_data_skip:
-    value: false
-include_for_metrics:
-    value: []
-include_inputs_for_metrics:
-    value: false
-include_num_input_tokens_seen:
-    value: false
-include_tokens_per_second:
-    value: false
-init_std:
-    value: 0.02
-is_decoder:
-    value: false
-is_encoder_decoder:
-    value: true
-jit_mode_eval:
-    value: false
-label_names:
-    value: null
-label_smoothing_factor:
-    value: 0
-label2id:
-    value:
-        LABEL_0: 0
-        LABEL_1: 1
-learning_rate:
-    value: 1e-05
-length_column_name:
-    value: input_length
-length_penalty:
-    value: 1
-load_best_model_at_end:
-    value: true
-local_rank:
-    value: 0
-log_level:
-    value: passive
-log_level_replica:
-    value: warning
-log_on_each_node:
-    value: true
-logging_dir:
-    value: ./runs/Feb12_12-51-48_tknika
-logging_first_step:
-    value: false
-logging_nan_inf_filter:
-    value: true
-logging_steps:
-    value: 25
-logging_strategy:
-    value: steps
-lr_scheduler_type:
-    value: linear
-mask_feature_length:
-    value: 10
-mask_feature_min_masks:
-    value: 0
-mask_feature_prob:
-    value: 0
-mask_time_length:
-    value: 10
-mask_time_min_masks:
-    value: 2
-mask_time_prob:
-    value: 0.05
-max_grad_norm:
-    value: 1
-max_length:
-    value: 448
-max_source_positions:
-    value: 1500
-max_steps:
-    value: 8000
-max_target_positions:
-    value: 448
-median_filter_width:
-    value: 7
-metric_for_best_model:
-    value: wer
-min_length:
-    value: 0
-model/num_parameters:
-    value: 241734912
-model_type:
-    value: whisper
-mp_parameters:
-    value: ""
-neftune_noise_alpha:
-    value: null
-no_cuda:
-    value: false
-no_repeat_ngram_size:
-    value: 0
-num_beam_groups:
-    value: 1
-num_beams:
-    value: 1
-num_hidden_layers:
-    value: 12
-num_mel_bins:
-    value: 80
-num_return_sequences:
-    value: 1
-num_train_epochs:
-    value: 3
-optim:
-    value: adamw_torch
-optim_args:
-    value: null
-optim_target_modules:
-    value: null
-output_attentions:
-    value: false
-output_dir:
-    value: ./
-output_hidden_states:
-    value: false
-output_scores:
-    value: false
-overwrite_output_dir:
-    value: true
-pad_token_id:
-    value: 50257
-past_index:
-    value: -1
-per_device_eval_batch_size:
-    value: 16
-per_device_train_batch_size:
-    value: 32
-per_gpu_eval_batch_size:
-    value: null
-per_gpu_train_batch_size:
-    value: null
-predict_with_generate:
-    value: true
-prediction_loss_only:
-    value: false
-prefix:
-    value: null
-problem_type:
-    value: null
-push_to_hub:
-    value: true
-push_to_hub_model_id:
-    value: null
-push_to_hub_organization:
-    value: null
-push_to_hub_token:
-    value: <PUSH_TO_HUB_TOKEN>
-ray_scope:
-    value: last
-remove_invalid_values:
-    value: false
-remove_unused_columns:
-    value: true
-repetition_penalty:
-    value: 1
-report_to:
-    value:
-        - wandb
-restore_callback_states_from_checkpoint:
-    value: false
-resume_from_checkpoint:
-    value: null
-return_dict:
-    value: true
-return_dict_in_generate:
-    value: false
-run_name:
-    value: whisper-small-eu
-save_on_each_node:
-    value: false
-save_only_model:
-    value: false
-save_safetensors:
-    value: true
-save_steps:
-    value: 1000
-save_strategy:
-    value: steps
-save_total_limit:
-    value: null
-scale_embedding:
-    value: false
-seed:
-    value: 42
-sep_token_id:
-    value: null
-skip_memory_metrics:
-    value: true
-sortish_sampler:
-    value: false
-split_batches:
-    value: null
-suppress_tokens:
-    value: null
-task_specific_params:
-    value: null
-temperature:
-    value: 1
-tf_legacy_loss:
-    value: false
-tf32:
-    value: null
-tie_encoder_decoder:
-    value: false
-tie_word_embeddings:
-    value: true
-tokenizer_class:
-    value: null
-top_k:
-    value: 50
-top_p:
-    value: 1
-torch_compile:
-    value: false
-torch_compile_backend:
-    value: null
-torch_compile_mode:
-    value: null
-torch_dtype:
-    value: float32
-torch_empty_cache_steps:
-    value: null
-torchdynamo:
-    value: null
-torchscript:
-    value: false
-tpu_metrics_debug:
-    value: false
-tpu_num_cores:
-    value: null
-transformers_version:
-    value: 4.49.0.dev0
-typical_p:
-    value: 1
-use_bfloat16:
-    value: false
-use_cache:
-    value: false
-use_cpu:
-    value: false
-use_ipex:
-    value: false
-use_legacy_prediction_loop:
-    value: false
-use_liger_kernel:
-    value: false
-use_mps_device:
-    value: false
-use_weighted_layer_sum:
-    value: false
-vocab_size:
-    value: 51865
-warmup_ratio:
-    value: 0
-warmup_steps:
-    value: 500
-weight_decay:
-    value: 0

wandb/run-20250212_125202-c6xjc1gs/files/output.log

@@ -1,22 +0,0 @@
-  0%|                                                                                                                                               | 0/8000 [00:00<?, ?it/s]Traceback (most recent call last):
-  File "/home/tknika/xezpeleta/whisper/whisper-small-eu/run_speech_recognition_seq2seq_streaming.py", line 630, in <module>
-    main()
-  File "/home/tknika/xezpeleta/whisper/whisper-small-eu/run_speech_recognition_seq2seq_streaming.py", line 579, in main
-    train_result = trainer.train(resume_from_checkpoint=checkpoint)
-                   ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
-  File "/home/tknika/xezpeleta/whisper/whisper-small-eu/.venv/lib/python3.12/site-packages/transformers/trainer.py", line 2197, in train
-    return inner_training_loop(
-           ^^^^^^^^^^^^^^^^^^^^
-  File "/home/tknika/xezpeleta/whisper/whisper-small-eu/.venv/lib/python3.12/site-packages/transformers/trainer.py", line 2438, in _inner_training_loop
-    self.control = self.callback_handler.on_epoch_begin(args, self.state, self.control)
-                   ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
-  File "/home/tknika/xezpeleta/whisper/whisper-small-eu/.venv/lib/python3.12/site-packages/transformers/trainer_callback.py", line 514, in on_epoch_begin
-    return self.call_event("on_epoch_begin", args, state, control)
-           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
-  File "/home/tknika/xezpeleta/whisper/whisper-small-eu/.venv/lib/python3.12/site-packages/transformers/trainer_callback.py", line 557, in call_event
-    result = getattr(callback, event)(
-             ^^^^^^^^^^^^^^^^^^^^^^^^^
-  File "/home/tknika/xezpeleta/whisper/whisper-small-eu/run_speech_recognition_seq2seq_streaming.py", line 555, in on_epoch_begin
-    if isinstance(train_dataloader.dataset, IterableDatasetShard):
-                  ^^^^^^^^^^^^^^^^^^^^^^^^
-AttributeError: 'NoneType' object has no attribute 'dataset'

wandb/run-20250212_125202-c6xjc1gs/files/requirements.txt

@@ -1,115 +0,0 @@
-aiosignal==1.3.2
-Markdown==3.7
-more-itertools==10.6.0
-requests==2.32.3
-sentry-sdk==2.21.0
-torchaudio==2.6.0
-charset-normalizer==3.4.1
-docker-pycreds==0.4.0
-nvidia-cusolver-cu12==11.6.1.9
-PyYAML==6.0.2
-librosa==0.10.2.post1
-soxr==0.5.0.post1
-multiprocess==0.70.16
-setuptools==75.8.0
-nvidia-cufft-cu12==11.2.1.3
-joblib==1.4.2
-pytz==2025.1
-pip==24.0
-scikit-learn==1.6.1
-certifi==2025.1.31
-jiwer==3.1.0
-regex==2024.11.6
-annotated-types==0.7.0
-grpcio==1.70.0
-msgpack==1.1.0
-mpmath==1.3.0
-nvidia-cudnn-cu12==9.1.0.70
-soundfile==0.13.1
-dill==0.3.8
-nvidia-nvtx-cu12==12.4.127
-six==1.17.0
-nvidia-cuda-cupti-cu12==12.4.127
-pyarrow==19.0.0
-nvidia-nccl-cu12==2.21.5
-psutil==6.1.1
-decorator==5.1.1
-llvmlite==0.44.0
-frozenlist==1.5.0
-pydantic==2.10.6
-networkx==3.4.2
-idna==3.10
-wandb==0.19.6
-aiohttp==3.11.12
-RapidFuzz==3.12.1
-pandas==2.2.3
-python-dateutil==2.9.0.post0
-numpy==2.1.3
-tokenizers==0.21.0
-nvidia-cusparselt-cu12==0.6.2
-typing_extensions==4.12.2
-urllib3==2.3.0
-setproctitle==1.3.4
-tzdata==2025.1
-sympy==1.13.1
-pooch==1.8.2
-click==8.1.8
-pydantic_core==2.27.2
-MarkupSafe==3.0.2
-scipy==1.15.1
-accelerate==1.3.0
-tensorboard==2.19.0
-protobuf==5.29.3
-gitdb==4.0.12
-smmap==5.0.2
-absl-py==2.1.0
-tqdm==4.67.1
-yarl==1.18.3
-pycparser==2.22
-nvidia-cusparse-cu12==12.3.1.170
-attrs==25.1.0
-lazy_loader==0.4
-tensorboard-data-server==0.7.2
-threadpoolctl==3.5.0
-GitPython==3.1.44
-safetensors==0.5.2
-fsspec==2024.12.0
-nvidia-cuda-nvrtc-cu12==12.4.127
-filelock==3.17.0
-aiohappyeyeballs==2.4.6
-packaging==24.2
-datasets==3.2.1.dev0
-audioread==3.0.1
-propcache==0.2.1
-transformers==4.49.0.dev0
-nvidia-cuda-runtime-cu12==12.4.127
-cffi==1.17.1
-evaluate==0.4.3
-Werkzeug==3.1.3
-huggingface-hub==0.28.1
-Jinja2==3.1.5
-torch==2.6.0
-nvidia-curand-cu12==10.3.5.147
-xxhash==3.5.0
-platformdirs==4.3.6
-multidict==6.1.0
-nvidia-cublas-cu12==12.4.5.8
-nvidia-nvjitlink-cu12==12.4.127
-triton==3.2.0
-numba==0.61.0
-importlib_metadata==8.0.0
-platformdirs==4.2.2
-typeguard==4.3.0
-more-itertools==10.3.0
-tomli==2.0.1
-autocommand==2.2.2
-zipp==3.19.2
-typing_extensions==4.12.2
-backports.tarfile==1.2.0
-inflect==7.3.1
-jaraco.text==3.12.1
-wheel==0.43.0
-packaging==24.2
-jaraco.collections==5.1.0
-jaraco.functools==4.0.1
-jaraco.context==5.3.0

wandb/run-20250212_125202-c6xjc1gs/files/wandb-metadata.json

@@ -1,85 +0,0 @@
-{
-  "os":  "Linux-6.8.0-48-generic-x86_64-with-glibc2.39",
-  "python":  "CPython 3.12.3",
-  "startedAt":  "2025-02-12T12:52:03.105234Z",
-  "args":  [
-    "--model_name_or_path=openai/whisper-small",
-    "--dataset_name=asierhv/composite_corpus_eu_v2.1",
-    "--language=basque",
-    "--train_split_name=train",
-    "--eval_split_name=test_parl",
-    "--model_index_name=Whisper Small Basque",
-    "--max_steps=8000",
-    "--output_dir=./",
-    "--per_device_train_batch_size=32",
-    "--per_device_eval_batch_size=16",
-    "--gradient_accumulation_steps=1",
-    "--logging_steps=25",
-    "--learning_rate=1e-5",
-    "--warmup_steps=500",
-    "--evaluation_strategy=steps",
-    "--eval_steps=1000",
-    "--save_strategy=steps",
-    "--save_steps=1000",
-    "--generation_max_length=225",
-    "--length_column_name=input_length",
-    "--max_duration_in_seconds=30",
-    "--text_column_name=sentence",
-    "--freeze_feature_encoder=False",
-    "--report_to=tensorboard",
-    "--metric_for_best_model=wer",
-    "--greater_is_better=False",
-    "--load_best_model_at_end",
-    "--gradient_checkpointing",
-    "--fp16",
-    "--overwrite_output_dir",
-    "--do_train",
-    "--do_eval",
-    "--predict_with_generate",
-    "--do_normalize_eval",
-    "--streaming",
-    "--use_auth_token",
-    "--push_to_hub",
-    "--report_to",
-    "wandb",
-    "--run_name",
-    "whisper-small-eu"
-  ],
-  "program":  "/home/tknika/xezpeleta/whisper/whisper-small-eu/run_speech_recognition_seq2seq_streaming.py",
-  "codePath":  "run_speech_recognition_seq2seq_streaming.py",
-  "git":  {
-    "remote":  "https://huggingface.co/xezpeleta/whisper-small-eu",
-    "commit":  "9c975864b20b4df94398a870e97cad2934253ec3"
-  },
-  "email":  "[email protected]",
-  "root":  "/home/tknika/xezpeleta/whisper/whisper-small-eu",
-  "host":  "tknika",
-  "executable":  "/home/tknika/xezpeleta/whisper/whisper-small-eu/.venv/bin/python",
-  "codePathLocal":  "run_speech_recognition_seq2seq_streaming.py",
-  "cpu_count":  8,
-  "cpu_count_logical":  8,
-  "gpu":  "NVIDIA L40-48Q",
-  "gpu_count":  1,
-  "disk":  {
-    "/":  {
-      "total":  "525987168256",
-      "used":  "313777016832"
-    }
-  },
-  "memory":  {
-    "total":  "33654022144"
-  },
-  "cpu":  {
-    "count":  8,
-    "countLogical":  8
-  },
-  "gpu_nvidia":  [
-    {
-      "name":  "NVIDIA L40-48Q",
-      "memoryTotal":  "51539607552",
-      "cudaCores":  18176,
-      "architecture":  "Ada"
-    }
-  ],
-  "cudaVersion":  "12.4"
-}

wandb/run-20250212_125202-c6xjc1gs/files/wandb-summary.json

@@ -1 +0,0 @@
-{"_wandb":{"runtime":0}}

wandb/run-20250212_125202-c6xjc1gs/logs/debug-core.log

@@ -1,14 +0,0 @@
-{"time":"2025-02-12T12:52:02.919881508Z","level":"INFO","msg":"main: starting server","port-filename":"/tmp/tmpeae6bnaj/port-226112.txt","pid":226112,"log-level":0,"disable-analytics":false,"shutdown-on-parent-exit":false}
-{"time":"2025-02-12T12:52:02.924775623Z","level":"INFO","msg":"Will exit if parent process dies.","ppid":226112}
-{"time":"2025-02-12T12:52:02.924741833Z","level":"INFO","msg":"server is running","addr":{"IP":"127.0.0.1","Port":37305,"Zone":""}}
-{"time":"2025-02-12T12:52:03.098177175Z","level":"INFO","msg":"connection: ManageConnectionData: new connection created","id":"127.0.0.1:34596"}
-{"time":"2025-02-12T12:52:03.107916075Z","level":"INFO","msg":"handleInformInit: received","streamId":"c6xjc1gs","id":"127.0.0.1:34596"}
-{"time":"2025-02-12T12:52:03.213738528Z","level":"INFO","msg":"handleInformInit: stream started","streamId":"c6xjc1gs","id":"127.0.0.1:34596"}
-{"time":"2025-02-12T12:52:03.606976673Z","level":"INFO","msg":"handleInformTeardown: server teardown initiated","id":"127.0.0.1:34596"}
-{"time":"2025-02-12T12:52:03.607096473Z","level":"INFO","msg":"connection: closing","id":"127.0.0.1:34596"}
-{"time":"2025-02-12T12:52:03.607114372Z","level":"INFO","msg":"server is shutting down"}
-{"time":"2025-02-12T12:52:03.607218922Z","level":"INFO","msg":"connection: closed successfully","id":"127.0.0.1:34596"}
-{"time":"2025-02-12T12:52:03.804235797Z","level":"ERROR","msg":"processOutgoingData: flush error","error":"write tcp 127.0.0.1:37305->127.0.0.1:34596: use of closed network connection","id":"127.0.0.1:34596"}
-{"time":"2025-02-12T12:52:05.13436235Z","level":"INFO","msg":"handleInformTeardown: server shutdown complete","id":"127.0.0.1:34596"}
-{"time":"2025-02-12T12:52:05.13438775Z","level":"INFO","msg":"connection: ManageConnectionData: connection closed","id":"127.0.0.1:34596"}
-{"time":"2025-02-12T12:52:05.13441372Z","level":"INFO","msg":"server is closed"}

wandb/run-20250212_125202-c6xjc1gs/logs/debug-internal.log

@@ -1,15 +0,0 @@
-{"time":"2025-02-12T12:52:03.108316863Z","level":"INFO","msg":"stream: starting","core version":"0.19.6","symlink path":"/home/tknika/xezpeleta/whisper/whisper-small-eu/wandb/run-20250212_125202-c6xjc1gs/logs/debug-core.log"}
-{"time":"2025-02-12T12:52:03.213666458Z","level":"INFO","msg":"created new stream","id":"c6xjc1gs"}
-{"time":"2025-02-12T12:52:03.213728098Z","level":"INFO","msg":"stream: started","id":"c6xjc1gs"}
-{"time":"2025-02-12T12:52:03.213779117Z","level":"INFO","msg":"writer: Do: started","stream_id":"c6xjc1gs"}
-{"time":"2025-02-12T12:52:03.213809587Z","level":"INFO","msg":"handler: started","stream_id":"c6xjc1gs"}
-{"time":"2025-02-12T12:52:03.214018716Z","level":"INFO","msg":"sender: started","stream_id":"c6xjc1gs"}
-{"time":"2025-02-12T12:52:03.484749537Z","level":"INFO","msg":"Starting system monitor"}
-{"time":"2025-02-12T12:52:03.607062513Z","level":"INFO","msg":"stream: closing","id":"c6xjc1gs"}
-{"time":"2025-02-12T12:52:03.607125442Z","level":"INFO","msg":"Stopping system monitor"}
-{"time":"2025-02-12T12:52:03.607814399Z","level":"INFO","msg":"Stopped system monitor"}
-{"time":"2025-02-12T12:52:04.912814278Z","level":"INFO","msg":"fileTransfer: Close: file transfer manager closed"}
-{"time":"2025-02-12T12:52:05.133965652Z","level":"INFO","msg":"handler: closed","stream_id":"c6xjc1gs"}
-{"time":"2025-02-12T12:52:05.134024822Z","level":"INFO","msg":"sender: closed","stream_id":"c6xjc1gs"}
-{"time":"2025-02-12T12:52:05.134018042Z","level":"INFO","msg":"writer: Close: closed","stream_id":"c6xjc1gs"}
-{"time":"2025-02-12T12:52:05.134218211Z","level":"INFO","msg":"stream: closed","id":"c6xjc1gs"}

wandb/run-20250212_125202-c6xjc1gs/logs/debug.log

@@ -1,26 +0,0 @@
-2025-02-12 12:52:02,886 INFO    MainThread:226112 [wandb_setup.py:_flush():68] Current SDK version is 0.19.6
-2025-02-12 12:52:02,887 INFO    MainThread:226112 [wandb_setup.py:_flush():68] Configure stats pid to 226112
-2025-02-12 12:52:02,887 INFO    MainThread:226112 [wandb_setup.py:_flush():68] Loading settings from /home/tknika/.config/wandb/settings
-2025-02-12 12:52:02,887 INFO    MainThread:226112 [wandb_setup.py:_flush():68] Loading settings from /home/tknika/xezpeleta/whisper/whisper-small-eu/wandb/settings
-2025-02-12 12:52:02,887 INFO    MainThread:226112 [wandb_setup.py:_flush():68] Loading settings from environment variables
-2025-02-12 12:52:02,887 INFO    MainThread:226112 [wandb_init.py:setup_run_log_directory():637] Logging user logs to /home/tknika/xezpeleta/whisper/whisper-small-eu/wandb/run-20250212_125202-c6xjc1gs/logs/debug.log
-2025-02-12 12:52:02,887 INFO    MainThread:226112 [wandb_init.py:setup_run_log_directory():638] Logging internal logs to /home/tknika/xezpeleta/whisper/whisper-small-eu/wandb/run-20250212_125202-c6xjc1gs/logs/debug-internal.log
-2025-02-12 12:52:02,887 INFO    MainThread:226112 [wandb_init.py:init():756] calling init triggers
-2025-02-12 12:52:02,887 INFO    MainThread:226112 [wandb_init.py:init():761] wandb.init called with sweep_config: {}
-config: {'_wandb': {}}
-2025-02-12 12:52:02,887 INFO    MainThread:226112 [wandb_init.py:init():789] starting backend
-2025-02-12 12:52:03,097 INFO    MainThread:226112 [wandb_init.py:init():793] sending inform_init request
-2025-02-12 12:52:03,104 INFO    MainThread:226112 [backend.py:_multiprocessing_setup():97] multiprocessing start_methods=fork,spawn,forkserver, using: spawn
-2025-02-12 12:52:03,104 INFO    MainThread:226112 [wandb_init.py:init():808] backend started and connected
-2025-02-12 12:52:03,107 INFO    MainThread:226112 [wandb_init.py:init():901] updated telemetry
-2025-02-12 12:52:03,114 INFO    MainThread:226112 [wandb_init.py:init():936] communicating run to backend with 90.0 second timeout
-2025-02-12 12:52:03,483 INFO    MainThread:226112 [wandb_init.py:init():994] starting run threads in backend
-2025-02-12 12:52:03,566 INFO    MainThread:226112 [wandb_run.py:_console_start():2385] atexit reg
-2025-02-12 12:52:03,566 INFO    MainThread:226112 [wandb_run.py:_redirect():2235] redirect: wrap_raw
-2025-02-12 12:52:03,566 INFO    MainThread:226112 [wandb_run.py:_redirect():2300] Wrapping output streams.
-2025-02-12 12:52:03,566 INFO    MainThread:226112 [wandb_run.py:_redirect():2325] Redirects installed.
-2025-02-12 12:52:03,568 INFO    MainThread:226112 [wandb_init.py:init():1036] run started, returning control to user process
-2025-02-12 12:52:03,569 INFO    MainThread:226112 [wandb_run.py:_config_callback():1253] config_cb None None {'vocab_size': 51865, 'num_mel_bins': 80, 'd_model': 768, 'encoder_layers': 12, 'encoder_attention_heads': 12, 'decoder_layers': 12, 'decoder_attention_heads': 12, 'decoder_ffn_dim': 3072, 'encoder_ffn_dim': 3072, 'dropout': 0.0, 'attention_dropout': 0.0, 'activation_dropout': 0.0, 'activation_function': 'gelu', 'init_std': 0.02, 'encoder_layerdrop': 0.0, 'decoder_layerdrop': 0.0, 'use_cache': False, 'num_hidden_layers': 12, 'scale_embedding': False, 'max_source_positions': 1500, 'max_target_positions': 448, 'classifier_proj_size': 256, 'use_weighted_layer_sum': False, 'apply_spec_augment': False, 'mask_time_prob': 0.05, 'mask_time_length': 10, 'mask_time_min_masks': 2, 'mask_feature_prob': 0.0, 'mask_feature_length': 10, 'mask_feature_min_masks': 0, 'median_filter_width': 7, 'return_dict': True, 'output_hidden_states': False, 'output_attentions': False, 'torchscript': False, 'torch_dtype': 'float32', 'use_bfloat16': False, 'tf_legacy_loss': False, 'pruned_heads': {}, 'tie_word_embeddings': True, 'chunk_size_feed_forward': 0, 'is_encoder_decoder': True, 'is_decoder': False, 'cross_attention_hidden_size': None, 'add_cross_attention': False, 'tie_encoder_decoder': False, 'max_length': 448, 'min_length': 0, 'do_sample': False, 'early_stopping': False, 'num_beams': 1, 'num_beam_groups': 1, 'diversity_penalty': 0.0, 'temperature': 1.0, 'top_k': 50, 'top_p': 1.0, 'typical_p': 1.0, 'repetition_penalty': 1.0, 'length_penalty': 1.0, 'no_repeat_ngram_size': 0, 'encoder_no_repeat_ngram_size': 0, 'bad_words_ids': None, 'num_return_sequences': 1, 'output_scores': False, 'return_dict_in_generate': False, 'forced_bos_token_id': None, 'forced_eos_token_id': None, 'remove_invalid_values': False, 'exponential_decay_length_penalty': None, 'suppress_tokens': None, 'begin_suppress_tokens': [220, 50257], 'architectures': ['WhisperForConditionalGeneration'], 'finetuning_task': None, 'id2label': {0: 'LABEL_0', 1: 'LABEL_1'}, 'label2id': {'LABEL_0': 0, 'LABEL_1': 1}, 'tokenizer_class': None, 'prefix': None, 'bos_token_id': 50257, 'pad_token_id': 50257, 'eos_token_id': 50257, 'sep_token_id': None, 'decoder_start_token_id': 50258, 'task_specific_params': None, 'problem_type': None, '_name_or_path': 'openai/whisper-small', '_attn_implementation_autoset': True, 'transformers_version': '4.49.0.dev0', 'forced_decoder_ids': None, 'model_type': 'whisper', 'output_dir': './', 'overwrite_output_dir': True, 'do_train': True, 'do_eval': True, 'do_predict': False, 'eval_strategy': 'steps', 'prediction_loss_only': False, 'per_device_train_batch_size': 32, 'per_device_eval_batch_size': 16, 'per_gpu_train_batch_size': None, 'per_gpu_eval_batch_size': None, 'gradient_accumulation_steps': 1, 'eval_accumulation_steps': None, 'eval_delay': 0, 'torch_empty_cache_steps': None, 'learning_rate': 1e-05, 'weight_decay': 0.0, 'adam_beta1': 0.9, 'adam_beta2': 0.999, 'adam_epsilon': 1e-08, 'max_grad_norm': 1.0, 'num_train_epochs': 3.0, 'max_steps': 8000, 'lr_scheduler_type': 'linear', 'lr_scheduler_kwargs': {}, 'warmup_ratio': 0.0, 'warmup_steps': 500, 'log_level': 'passive', 'log_level_replica': 'warning', 'log_on_each_node': True, 'logging_dir': './runs/Feb12_12-51-48_tknika', 'logging_strategy': 'steps', 'logging_first_step': False, 'logging_steps': 25, 'logging_nan_inf_filter': True, 'save_strategy': 'steps', 'save_steps': 1000, 'save_total_limit': None, 'save_safetensors': True, 'save_on_each_node': False, 'save_only_model': False, 'restore_callback_states_from_checkpoint': False, 'no_cuda': False, 'use_cpu': False, 'use_mps_device': False, 'seed': 42, 'data_seed': None, 'jit_mode_eval': False, 'use_ipex': False, 'bf16': False, 'fp16': True, 'fp16_opt_level': 'O1', 'half_precision_backend': 'auto', 'bf16_full_eval': False, 'fp16_full_eval': False, 'tf32': None, 'local_rank': 0, 'ddp_backend': None, 'tpu_num_cores': None, 'tpu_metrics_debug': False, 'debug': [], 'dataloader_drop_last': False, 'eval_steps': 1000, 'dataloader_num_workers': 0, 'dataloader_prefetch_factor': None, 'past_index': -1, 'run_name': 'whisper-small-eu', 'disable_tqdm': False, 'remove_unused_columns': True, 'label_names': None, 'load_best_model_at_end': True, 'metric_for_best_model': 'wer', 'greater_is_better': False, 'ignore_data_skip': False, 'fsdp': [], 'fsdp_min_num_params': 0, 'fsdp_config': {'min_num_params': 0, 'xla': False, 'xla_fsdp_v2': False, 'xla_fsdp_grad_ckpt': False}, 'fsdp_transformer_layer_cls_to_wrap': None, 'accelerator_config': {'split_batches': False, 'dispatch_batches': None, 'even_batches': True, 'use_seedable_sampler': True, 'non_blocking': False, 'gradient_accumulation_kwargs': None}, 'deepspeed': None, 'label_smoothing_factor': 0.0, 'optim': 'adamw_torch', 'optim_args': None, 'adafactor': False, 'group_by_length': False, 'length_column_name': 'input_length', 'report_to': ['wandb'], 'ddp_find_unused_parameters': None, 'ddp_bucket_cap_mb': None, 'ddp_broadcast_buffers': None, 'dataloader_pin_memory': True, 'dataloader_persistent_workers': False, 'skip_memory_metrics': True, 'use_legacy_prediction_loop': False, 'push_to_hub': True, 'resume_from_checkpoint': None, 'hub_model_id': None, 'hub_strategy': 'every_save', 'hub_token': '<HUB_TOKEN>', 'hub_private_repo': None, 'hub_always_push': False, 'gradient_checkpointing': True, 'gradient_checkpointing_kwargs': None, 'include_inputs_for_metrics': False, 'include_for_metrics': [], 'eval_do_concat_batches': True, 'fp16_backend': 'auto', 'evaluation_strategy': 'steps', 'push_to_hub_model_id': None, 'push_to_hub_organization': None, 'push_to_hub_token': '<PUSH_TO_HUB_TOKEN>', 'mp_parameters': '', 'auto_find_batch_size': False, 'full_determinism': False, 'torchdynamo': None, 'ray_scope': 'last', 'ddp_timeout': 1800, 'torch_compile': False, 'torch_compile_backend': None, 'torch_compile_mode': None, 'dispatch_batches': None, 'split_batches': None, 'include_tokens_per_second': False, 'include_num_input_tokens_seen': False, 'neftune_noise_alpha': None, 'optim_target_modules': None, 'batch_eval_metrics': False, 'eval_on_start': False, 'use_liger_kernel': False, 'eval_use_gather_object': False, 'average_tokens_across_devices': False, 'sortish_sampler': False, 'predict_with_generate': True, 'generation_max_length': 225, 'generation_num_beams': None, 'generation_config': None}
-2025-02-12 12:52:03,571 INFO    MainThread:226112 [wandb_config.py:__setitem__():154] config set model/num_parameters = 241734912 - <bound method Run._config_callback of <wandb.sdk.wandb_run.Run object at 0x7d4830f2ddf0>>
-2025-02-12 12:52:03,571 INFO    MainThread:226112 [wandb_run.py:_config_callback():1253] config_cb model/num_parameters 241734912 None
-2025-02-12 12:52:03,607 WARNING MsgRouterThr:226112 [router.py:message_loop():75] message_loop has been closed

wandb/run-20250212_125924-xhsgsxqq/files/config.yaml

@@ -1,512 +0,0 @@
-_attn_implementation_autoset:
-    value: true
-_name_or_path:
-    value: openai/whisper-small
-_wandb:
-    value:
-        cli_version: 0.19.6
-        m:
-            - "1": train/global_step
-              "6":
-                - 3
-              "7": []
-        python_version: 3.12.3
-        t:
-            "1":
-                - 1
-                - 5
-                - 11
-                - 49
-                - 51
-                - 53
-                - 55
-                - 71
-                - 100
-            "2":
-                - 1
-                - 5
-                - 11
-                - 49
-                - 51
-                - 53
-                - 55
-                - 71
-                - 100
-            "3":
-                - 7
-                - 13
-                - 19
-                - 23
-                - 55
-                - 66
-            "4": 3.12.3
-            "5": 0.19.6
-            "6": 4.49.0.dev0
-            "8":
-                - 5
-            "9":
-                "1": transformers_trainer
-            "12": 0.19.6
-            "13": linux-x86_64
-accelerator_config:
-    value:
-        dispatch_batches: null
-        even_batches: true
-        gradient_accumulation_kwargs: null
-        non_blocking: false
-        split_batches: false
-        use_seedable_sampler: true
-activation_dropout:
-    value: 0
-activation_function:
-    value: gelu
-adafactor:
-    value: false
-adam_beta1:
-    value: 0.9
-adam_beta2:
-    value: 0.999
-adam_epsilon:
-    value: 1e-08
-add_cross_attention:
-    value: false
-apply_spec_augment:
-    value: false
-architectures:
-    value:
-        - WhisperForConditionalGeneration
-attention_dropout:
-    value: 0
-auto_find_batch_size:
-    value: false
-average_tokens_across_devices:
-    value: false
-bad_words_ids:
-    value: null
-batch_eval_metrics:
-    value: false
-begin_suppress_tokens:
-    value:
-        - 220
-        - 50257
-bf16:
-    value: false
-bf16_full_eval:
-    value: false
-bos_token_id:
-    value: 50257
-chunk_size_feed_forward:
-    value: 0
-classifier_proj_size:
-    value: 256
-cross_attention_hidden_size:
-    value: null
-d_model:
-    value: 768
-data_seed:
-    value: null
-dataloader_drop_last:
-    value: false
-dataloader_num_workers:
-    value: 0
-dataloader_persistent_workers:
-    value: false
-dataloader_pin_memory:
-    value: true
-dataloader_prefetch_factor:
-    value: null
-ddp_backend:
-    value: null
-ddp_broadcast_buffers:
-    value: null
-ddp_bucket_cap_mb:
-    value: null
-ddp_find_unused_parameters:
-    value: null
-ddp_timeout:
-    value: 1800
-debug:
-    value: []
-decoder_attention_heads:
-    value: 12
-decoder_ffn_dim:
-    value: 3072
-decoder_layerdrop:
-    value: 0
-decoder_layers:
-    value: 12
-decoder_start_token_id:
-    value: 50258
-deepspeed:
-    value: null
-disable_tqdm:
-    value: false
-dispatch_batches:
-    value: null
-diversity_penalty:
-    value: 0
-do_eval:
-    value: true
-do_predict:
-    value: false
-do_sample:
-    value: false
-do_train:
-    value: true
-dropout:
-    value: 0
-early_stopping:
-    value: false
-encoder_attention_heads:
-    value: 12
-encoder_ffn_dim:
-    value: 3072
-encoder_layerdrop:
-    value: 0
-encoder_layers:
-    value: 12
-encoder_no_repeat_ngram_size:
-    value: 0
-eos_token_id:
-    value: 50257
-eval_accumulation_steps:
-    value: null
-eval_delay:
-    value: 0
-eval_do_concat_batches:
-    value: true
-eval_on_start:
-    value: false
-eval_steps:
-    value: 1000
-eval_strategy:
-    value: steps
-eval_use_gather_object:
-    value: false
-evaluation_strategy:
-    value: steps
-exponential_decay_length_penalty:
-    value: null
-finetuning_task:
-    value: null
-forced_bos_token_id:
-    value: null
-forced_decoder_ids:
-    value: null
-forced_eos_token_id:
-    value: null
-fp16:
-    value: true
-fp16_backend:
-    value: auto
-fp16_full_eval:
-    value: false
-fp16_opt_level:
-    value: O1
-fsdp:
-    value: []
-fsdp_config:
-    value:
-        min_num_params: 0
-        xla: false
-        xla_fsdp_grad_ckpt: false
-        xla_fsdp_v2: false
-fsdp_min_num_params:
-    value: 0
-fsdp_transformer_layer_cls_to_wrap:
-    value: null
-full_determinism:
-    value: false
-generation_config:
-    value: null
-generation_max_length:
-    value: 225
-generation_num_beams:
-    value: null
-gradient_accumulation_steps:
-    value: 1
-gradient_checkpointing:
-    value: true
-gradient_checkpointing_kwargs:
-    value: null
-greater_is_better:
-    value: false
-group_by_length:
-    value: false
-half_precision_backend:
-    value: auto
-hub_always_push:
-    value: false
-hub_model_id:
-    value: null
-hub_private_repo:
-    value: null
-hub_strategy:
-    value: every_save
-hub_token:
-    value: <HUB_TOKEN>
-id2label:
-    value:
-        "0": LABEL_0
-        "1": LABEL_1
-ignore_data_skip:
-    value: false
-include_for_metrics:
-    value: []
-include_inputs_for_metrics:
-    value: false
-include_num_input_tokens_seen:
-    value: false
-include_tokens_per_second:
-    value: false
-init_std:
-    value: 0.02
-is_decoder:
-    value: false
-is_encoder_decoder:
-    value: true
-jit_mode_eval:
-    value: false
-label_names:
-    value: null
-label_smoothing_factor:
-    value: 0
-label2id:
-    value:
-        LABEL_0: 0
-        LABEL_1: 1
-learning_rate:
-    value: 1e-05
-length_column_name:
-    value: input_length
-length_penalty:
-    value: 1
-load_best_model_at_end:
-    value: true
-local_rank:
-    value: 0
-log_level:
-    value: passive
-log_level_replica:
-    value: warning
-log_on_each_node:
-    value: true
-logging_dir:
-    value: ./runs/Feb12_12-58-59_tknika
-logging_first_step:
-    value: false
-logging_nan_inf_filter:
-    value: true
-logging_steps:
-    value: 25
-logging_strategy:
-    value: steps
-lr_scheduler_type:
-    value: linear
-mask_feature_length:
-    value: 10
-mask_feature_min_masks:
-    value: 0
-mask_feature_prob:
-    value: 0
-mask_time_length:
-    value: 10
-mask_time_min_masks:
-    value: 2
-mask_time_prob:
-    value: 0.05
-max_grad_norm:
-    value: 1
-max_length:
-    value: 448
-max_source_positions:
-    value: 1500
-max_steps:
-    value: 8000
-max_target_positions:
-    value: 448
-median_filter_width:
-    value: 7
-metric_for_best_model:
-    value: wer
-min_length:
-    value: 0
-model/num_parameters:
-    value: 241734912
-model_type:
-    value: whisper
-mp_parameters:
-    value: ""
-neftune_noise_alpha:
-    value: null
-no_cuda:
-    value: false
-no_repeat_ngram_size:
-    value: 0
-num_beam_groups:
-    value: 1
-num_beams:
-    value: 1
-num_hidden_layers:
-    value: 12
-num_mel_bins:
-    value: 80
-num_return_sequences:
-    value: 1
-num_train_epochs:
-    value: 3
-optim:
-    value: adamw_torch
-optim_args:
-    value: null
-optim_target_modules:
-    value: null
-output_attentions:
-    value: false
-output_dir:
-    value: ./
-output_hidden_states:
-    value: false
-output_scores:
-    value: false
-overwrite_output_dir:
-    value: true
-pad_token_id:
-    value: 50257
-past_index:
-    value: -1
-per_device_eval_batch_size:
-    value: 16
-per_device_train_batch_size:
-    value: 32
-per_gpu_eval_batch_size:
-    value: null
-per_gpu_train_batch_size:
-    value: null
-predict_with_generate:
-    value: true
-prediction_loss_only:
-    value: false
-prefix:
-    value: null
-problem_type:
-    value: null
-push_to_hub:
-    value: true
-push_to_hub_model_id:
-    value: null
-push_to_hub_organization:
-    value: null
-push_to_hub_token:
-    value: <PUSH_TO_HUB_TOKEN>
-ray_scope:
-    value: last
-remove_invalid_values:
-    value: false
-remove_unused_columns:
-    value: true
-repetition_penalty:
-    value: 1
-report_to:
-    value:
-        - wandb
-restore_callback_states_from_checkpoint:
-    value: false
-resume_from_checkpoint:
-    value: null
-return_dict:
-    value: true
-return_dict_in_generate:
-    value: false
-run_name:
-    value: whisper-small-eu
-save_on_each_node:
-    value: false
-save_only_model:
-    value: false
-save_safetensors:
-    value: true
-save_steps:
-    value: 1000
-save_strategy:
-    value: steps
-save_total_limit:
-    value: null
-scale_embedding:
-    value: false
-seed:
-    value: 42
-sep_token_id:
-    value: null
-skip_memory_metrics:
-    value: true
-sortish_sampler:
-    value: false
-split_batches:
-    value: null
-suppress_tokens:
-    value: null
-task_specific_params:
-    value: null
-temperature:
-    value: 1
-tf_legacy_loss:
-    value: false
-tf32:
-    value: null
-tie_encoder_decoder:
-    value: false
-tie_word_embeddings:
-    value: true
-tokenizer_class:
-    value: null
-top_k:
-    value: 50
-top_p:
-    value: 1
-torch_compile:
-    value: false
-torch_compile_backend:
-    value: null
-torch_compile_mode:
-    value: null
-torch_dtype:
-    value: float32
-torch_empty_cache_steps:
-    value: null
-torchdynamo:
-    value: null
-torchscript:
-    value: false
-tpu_metrics_debug:
-    value: false
-tpu_num_cores:
-    value: null
-transformers_version:
-    value: 4.49.0.dev0
-typical_p:
-    value: 1
-use_bfloat16:
-    value: false
-use_cache:
-    value: false
-use_cpu:
-    value: false
-use_ipex:
-    value: false
-use_legacy_prediction_loop:
-    value: false
-use_liger_kernel:
-    value: false
-use_mps_device:
-    value: false
-use_weighted_layer_sum:
-    value: false
-vocab_size:
-    value: 51865
-warmup_ratio:
-    value: 0
-warmup_steps:
-    value: 500
-weight_decay:
-    value: 0

wandb/run-20250212_125924-xhsgsxqq/files/output.log

@@ -1,22 +0,0 @@
-  0%|                                                                                                                                               | 0/8000 [00:00<?, ?it/s]Traceback (most recent call last):
-  File "/home/tknika/xezpeleta/whisper/whisper-small-eu/run_speech_recognition_seq2seq_streaming.py", line 632, in <module>
-    main()
-  File "/home/tknika/xezpeleta/whisper/whisper-small-eu/run_speech_recognition_seq2seq_streaming.py", line 581, in main
-    train_result = trainer.train(resume_from_checkpoint=checkpoint)
-                   ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
-  File "/home/tknika/xezpeleta/whisper/whisper-small-eu/.venv/lib/python3.12/site-packages/transformers/trainer.py", line 2197, in train
-    return inner_training_loop(
-           ^^^^^^^^^^^^^^^^^^^^
-  File "/home/tknika/xezpeleta/whisper/whisper-small-eu/.venv/lib/python3.12/site-packages/transformers/trainer.py", line 2438, in _inner_training_loop
-    self.control = self.callback_handler.on_epoch_begin(args, self.state, self.control)
-                   ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
-  File "/home/tknika/xezpeleta/whisper/whisper-small-eu/.venv/lib/python3.12/site-packages/transformers/trainer_callback.py", line 514, in on_epoch_begin
-    return self.call_event("on_epoch_begin", args, state, control)
-           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
-  File "/home/tknika/xezpeleta/whisper/whisper-small-eu/.venv/lib/python3.12/site-packages/transformers/trainer_callback.py", line 557, in call_event
-    result = getattr(callback, event)(
-             ^^^^^^^^^^^^^^^^^^^^^^^^^
-  File "/home/tknika/xezpeleta/whisper/whisper-small-eu/run_speech_recognition_seq2seq_streaming.py", line 557, in on_epoch_begin
-    if isinstance(train_dataloader.dataset, IterableDatasetShard):
-                  ^^^^^^^^^^^^^^^^^^^^^^^^
-AttributeError: 'NoneType' object has no attribute 'dataset'

wandb/run-20250212_125924-xhsgsxqq/files/requirements.txt

@@ -1,115 +0,0 @@
-aiosignal==1.3.2
-Markdown==3.7
-more-itertools==10.6.0
-requests==2.32.3
-sentry-sdk==2.21.0
-torchaudio==2.6.0
-charset-normalizer==3.4.1
-docker-pycreds==0.4.0
-nvidia-cusolver-cu12==11.6.1.9
-PyYAML==6.0.2
-librosa==0.10.2.post1
-soxr==0.5.0.post1
-multiprocess==0.70.16
-setuptools==75.8.0
-nvidia-cufft-cu12==11.2.1.3
-joblib==1.4.2
-pytz==2025.1
-pip==24.0
-scikit-learn==1.6.1
-certifi==2025.1.31
-jiwer==3.1.0
-regex==2024.11.6
-annotated-types==0.7.0
-grpcio==1.70.0
-msgpack==1.1.0
-mpmath==1.3.0
-nvidia-cudnn-cu12==9.1.0.70
-soundfile==0.13.1
-dill==0.3.8
-nvidia-nvtx-cu12==12.4.127
-six==1.17.0
-nvidia-cuda-cupti-cu12==12.4.127
-pyarrow==19.0.0
-nvidia-nccl-cu12==2.21.5
-psutil==6.1.1
-decorator==5.1.1
-llvmlite==0.44.0
-frozenlist==1.5.0
-pydantic==2.10.6
-networkx==3.4.2
-idna==3.10
-wandb==0.19.6
-aiohttp==3.11.12
-RapidFuzz==3.12.1
-pandas==2.2.3
-python-dateutil==2.9.0.post0
-numpy==2.1.3
-tokenizers==0.21.0
-nvidia-cusparselt-cu12==0.6.2
-typing_extensions==4.12.2
-urllib3==2.3.0
-setproctitle==1.3.4
-tzdata==2025.1
-sympy==1.13.1
-pooch==1.8.2
-click==8.1.8
-pydantic_core==2.27.2
-MarkupSafe==3.0.2
-scipy==1.15.1
-accelerate==1.3.0
-tensorboard==2.19.0
-protobuf==5.29.3
-gitdb==4.0.12
-smmap==5.0.2
-absl-py==2.1.0
-tqdm==4.67.1
-yarl==1.18.3
-pycparser==2.22
-nvidia-cusparse-cu12==12.3.1.170
-attrs==25.1.0
-lazy_loader==0.4
-tensorboard-data-server==0.7.2
-threadpoolctl==3.5.0
-GitPython==3.1.44
-safetensors==0.5.2
-fsspec==2024.12.0
-nvidia-cuda-nvrtc-cu12==12.4.127
-filelock==3.17.0
-aiohappyeyeballs==2.4.6
-packaging==24.2
-datasets==3.2.1.dev0
-audioread==3.0.1
-propcache==0.2.1
-transformers==4.49.0.dev0
-nvidia-cuda-runtime-cu12==12.4.127
-cffi==1.17.1
-evaluate==0.4.3
-Werkzeug==3.1.3
-huggingface-hub==0.28.1
-Jinja2==3.1.5
-torch==2.6.0
-nvidia-curand-cu12==10.3.5.147
-xxhash==3.5.0
-platformdirs==4.3.6
-multidict==6.1.0
-nvidia-cublas-cu12==12.4.5.8
-nvidia-nvjitlink-cu12==12.4.127
-triton==3.2.0
-numba==0.61.0
-importlib_metadata==8.0.0
-platformdirs==4.2.2
-typeguard==4.3.0
-more-itertools==10.3.0
-tomli==2.0.1
-autocommand==2.2.2
-zipp==3.19.2
-typing_extensions==4.12.2
-backports.tarfile==1.2.0
-inflect==7.3.1
-jaraco.text==3.12.1
-wheel==0.43.0
-packaging==24.2
-jaraco.collections==5.1.0
-jaraco.functools==4.0.1
-jaraco.context==5.3.0

wandb/run-20250212_125924-xhsgsxqq/files/wandb-metadata.json

@@ -1,85 +0,0 @@
-{
-  "os":  "Linux-6.8.0-48-generic-x86_64-with-glibc2.39",
-  "python":  "CPython 3.12.3",
-  "startedAt":  "2025-02-12T12:59:24.816046Z",
-  "args":  [
-    "--model_name_or_path=openai/whisper-small",
-    "--dataset_name=asierhv/composite_corpus_eu_v2.1",
-    "--language=basque",
-    "--train_split_name=train",
-    "--eval_split_name=dev_parl+test_parl+test_cv+test_oslr",
-    "--model_index_name=Whisper Small Basque",
-    "--max_steps=8000",
-    "--output_dir=./",
-    "--per_device_train_batch_size=32",
-    "--per_device_eval_batch_size=16",
-    "--gradient_accumulation_steps=1",
-    "--logging_steps=25",
-    "--learning_rate=1e-5",
-    "--warmup_steps=500",
-    "--evaluation_strategy=steps",
-    "--eval_steps=1000",
-    "--save_strategy=steps",
-    "--save_steps=1000",
-    "--generation_max_length=225",
-    "--length_column_name=input_length",
-    "--max_duration_in_seconds=30",
-    "--text_column_name=sentence",
-    "--freeze_feature_encoder=False",
-    "--report_to=tensorboard",
-    "--metric_for_best_model=wer",
-    "--greater_is_better=False",
-    "--load_best_model_at_end",
-    "--gradient_checkpointing",
-    "--fp16",
-    "--overwrite_output_dir",
-    "--do_train",
-    "--do_eval",
-    "--predict_with_generate",
-    "--do_normalize_eval",
-    "--streaming",
-    "--use_auth_token",
-    "--push_to_hub",
-    "--report_to",
-    "wandb",
-    "--run_name",
-    "whisper-small-eu"
-  ],
-  "program":  "/home/tknika/xezpeleta/whisper/whisper-small-eu/run_speech_recognition_seq2seq_streaming.py",
-  "codePath":  "run_speech_recognition_seq2seq_streaming.py",
-  "git":  {
-    "remote":  "https://huggingface.co/xezpeleta/whisper-small-eu",
-    "commit":  "9c975864b20b4df94398a870e97cad2934253ec3"
-  },
-  "email":  "[email protected]",
-  "root":  "/home/tknika/xezpeleta/whisper/whisper-small-eu",
-  "host":  "tknika",
-  "executable":  "/home/tknika/xezpeleta/whisper/whisper-small-eu/.venv/bin/python",
-  "codePathLocal":  "run_speech_recognition_seq2seq_streaming.py",
-  "cpu_count":  8,
-  "cpu_count_logical":  8,
-  "gpu":  "NVIDIA L40-48Q",
-  "gpu_count":  1,
-  "disk":  {
-    "/":  {
-      "total":  "525987168256",
-      "used":  "313777115136"
-    }
-  },
-  "memory":  {
-    "total":  "33654022144"
-  },
-  "cpu":  {
-    "count":  8,
-    "countLogical":  8
-  },
-  "gpu_nvidia":  [
-    {
-      "name":  "NVIDIA L40-48Q",
-      "memoryTotal":  "51539607552",
-      "cudaCores":  18176,
-      "architecture":  "Ada"
-    }
-  ],
-  "cudaVersion":  "12.4"
-}

wandb/run-20250212_125924-xhsgsxqq/files/wandb-summary.json

@@ -1 +0,0 @@
-{"_wandb":{"runtime":0}}

wandb/run-20250212_125924-xhsgsxqq/logs/debug-core.log

@@ -1,14 +0,0 @@
-{"time":"2025-02-12T12:59:24.63359638Z","level":"INFO","msg":"main: starting server","port-filename":"/tmp/tmpa90v2n0h/port-226591.txt","pid":226591,"log-level":0,"disable-analytics":false,"shutdown-on-parent-exit":false}
-{"time":"2025-02-12T12:59:24.673351851Z","level":"INFO","msg":"Will exit if parent process dies.","ppid":226591}
-{"time":"2025-02-12T12:59:24.673324591Z","level":"INFO","msg":"server is running","addr":{"IP":"127.0.0.1","Port":41203,"Zone":""}}
-{"time":"2025-02-12T12:59:24.809390061Z","level":"INFO","msg":"connection: ManageConnectionData: new connection created","id":"127.0.0.1:53388"}
-{"time":"2025-02-12T12:59:24.819517706Z","level":"INFO","msg":"handleInformInit: received","streamId":"xhsgsxqq","id":"127.0.0.1:53388"}
-{"time":"2025-02-12T12:59:24.923364896Z","level":"INFO","msg":"handleInformInit: stream started","streamId":"xhsgsxqq","id":"127.0.0.1:53388"}
-{"time":"2025-02-12T12:59:25.341856618Z","level":"INFO","msg":"handleInformTeardown: server teardown initiated","id":"127.0.0.1:53388"}
-{"time":"2025-02-12T12:59:25.341962867Z","level":"INFO","msg":"server is shutting down"}
-{"time":"2025-02-12T12:59:25.341964847Z","level":"INFO","msg":"connection: closing","id":"127.0.0.1:53388"}
-{"time":"2025-02-12T12:59:25.342139496Z","level":"INFO","msg":"connection: closed successfully","id":"127.0.0.1:53388"}
-{"time":"2025-02-12T12:59:25.569637185Z","level":"ERROR","msg":"processOutgoingData: flush error","error":"write tcp 127.0.0.1:41203->127.0.0.1:53388: use of closed network connection","id":"127.0.0.1:53388"}
-{"time":"2025-02-12T12:59:26.643739482Z","level":"INFO","msg":"handleInformTeardown: server shutdown complete","id":"127.0.0.1:53388"}
-{"time":"2025-02-12T12:59:26.643783881Z","level":"INFO","msg":"connection: ManageConnectionData: connection closed","id":"127.0.0.1:53388"}
-{"time":"2025-02-12T12:59:26.643808411Z","level":"INFO","msg":"server is closed"}

wandb/run-20250212_125924-xhsgsxqq/logs/debug-internal.log

@@ -1,15 +0,0 @@
-{"time":"2025-02-12T12:59:24.819689255Z","level":"INFO","msg":"stream: starting","core version":"0.19.6","symlink path":"/home/tknika/xezpeleta/whisper/whisper-small-eu/wandb/run-20250212_125924-xhsgsxqq/logs/debug-core.log"}
-{"time":"2025-02-12T12:59:24.923303396Z","level":"INFO","msg":"created new stream","id":"xhsgsxqq"}
-{"time":"2025-02-12T12:59:24.923354596Z","level":"INFO","msg":"stream: started","id":"xhsgsxqq"}
-{"time":"2025-02-12T12:59:24.923472545Z","level":"INFO","msg":"writer: Do: started","stream_id":"xhsgsxqq"}
-{"time":"2025-02-12T12:59:24.923494475Z","level":"INFO","msg":"handler: started","stream_id":"xhsgsxqq"}
-{"time":"2025-02-12T12:59:24.923560215Z","level":"INFO","msg":"sender: started","stream_id":"xhsgsxqq"}
-{"time":"2025-02-12T12:59:25.192419842Z","level":"INFO","msg":"Starting system monitor"}
-{"time":"2025-02-12T12:59:25.341944447Z","level":"INFO","msg":"stream: closing","id":"xhsgsxqq"}
-{"time":"2025-02-12T12:59:25.341971537Z","level":"INFO","msg":"Stopping system monitor"}
-{"time":"2025-02-12T12:59:25.342739283Z","level":"INFO","msg":"Stopped system monitor"}
-{"time":"2025-02-12T12:59:26.408412135Z","level":"INFO","msg":"fileTransfer: Close: file transfer manager closed"}
-{"time":"2025-02-12T12:59:26.643441283Z","level":"INFO","msg":"handler: closed","stream_id":"xhsgsxqq"}
-{"time":"2025-02-12T12:59:26.643483513Z","level":"INFO","msg":"writer: Close: closed","stream_id":"xhsgsxqq"}
-{"time":"2025-02-12T12:59:26.643525403Z","level":"INFO","msg":"sender: closed","stream_id":"xhsgsxqq"}
-{"time":"2025-02-12T12:59:26.643566923Z","level":"INFO","msg":"stream: closed","id":"xhsgsxqq"}

wandb/run-20250212_125924-xhsgsxqq/logs/debug.log

@@ -1,26 +0,0 @@
-2025-02-12 12:59:24,598 INFO    MainThread:226591 [wandb_setup.py:_flush():68] Current SDK version is 0.19.6
-2025-02-12 12:59:24,598 INFO    MainThread:226591 [wandb_setup.py:_flush():68] Configure stats pid to 226591
-2025-02-12 12:59:24,598 INFO    MainThread:226591 [wandb_setup.py:_flush():68] Loading settings from /home/tknika/.config/wandb/settings
-2025-02-12 12:59:24,598 INFO    MainThread:226591 [wandb_setup.py:_flush():68] Loading settings from /home/tknika/xezpeleta/whisper/whisper-small-eu/wandb/settings
-2025-02-12 12:59:24,598 INFO    MainThread:226591 [wandb_setup.py:_flush():68] Loading settings from environment variables
-2025-02-12 12:59:24,598 INFO    MainThread:226591 [wandb_init.py:setup_run_log_directory():637] Logging user logs to /home/tknika/xezpeleta/whisper/whisper-small-eu/wandb/run-20250212_125924-xhsgsxqq/logs/debug.log
-2025-02-12 12:59:24,598 INFO    MainThread:226591 [wandb_init.py:setup_run_log_directory():638] Logging internal logs to /home/tknika/xezpeleta/whisper/whisper-small-eu/wandb/run-20250212_125924-xhsgsxqq/logs/debug-internal.log
-2025-02-12 12:59:24,598 INFO    MainThread:226591 [wandb_init.py:init():756] calling init triggers
-2025-02-12 12:59:24,598 INFO    MainThread:226591 [wandb_init.py:init():761] wandb.init called with sweep_config: {}
-config: {'_wandb': {}}
-2025-02-12 12:59:24,598 INFO    MainThread:226591 [wandb_init.py:init():789] starting backend
-2025-02-12 12:59:24,809 INFO    MainThread:226591 [wandb_init.py:init():793] sending inform_init request
-2025-02-12 12:59:24,815 INFO    MainThread:226591 [backend.py:_multiprocessing_setup():97] multiprocessing start_methods=fork,spawn,forkserver, using: spawn
-2025-02-12 12:59:24,815 INFO    MainThread:226591 [wandb_init.py:init():808] backend started and connected
-2025-02-12 12:59:24,818 INFO    MainThread:226591 [wandb_init.py:init():901] updated telemetry
-2025-02-12 12:59:24,823 INFO    MainThread:226591 [wandb_init.py:init():936] communicating run to backend with 90.0 second timeout
-2025-02-12 12:59:25,189 INFO    MainThread:226591 [wandb_init.py:init():994] starting run threads in backend
-2025-02-12 12:59:25,301 INFO    MainThread:226591 [wandb_run.py:_console_start():2385] atexit reg
-2025-02-12 12:59:25,301 INFO    MainThread:226591 [wandb_run.py:_redirect():2235] redirect: wrap_raw
-2025-02-12 12:59:25,301 INFO    MainThread:226591 [wandb_run.py:_redirect():2300] Wrapping output streams.
-2025-02-12 12:59:25,301 INFO    MainThread:226591 [wandb_run.py:_redirect():2325] Redirects installed.
-2025-02-12 12:59:25,303 INFO    MainThread:226591 [wandb_init.py:init():1036] run started, returning control to user process
-2025-02-12 12:59:25,304 INFO    MainThread:226591 [wandb_run.py:_config_callback():1253] config_cb None None {'vocab_size': 51865, 'num_mel_bins': 80, 'd_model': 768, 'encoder_layers': 12, 'encoder_attention_heads': 12, 'decoder_layers': 12, 'decoder_attention_heads': 12, 'decoder_ffn_dim': 3072, 'encoder_ffn_dim': 3072, 'dropout': 0.0, 'attention_dropout': 0.0, 'activation_dropout': 0.0, 'activation_function': 'gelu', 'init_std': 0.02, 'encoder_layerdrop': 0.0, 'decoder_layerdrop': 0.0, 'use_cache': False, 'num_hidden_layers': 12, 'scale_embedding': False, 'max_source_positions': 1500, 'max_target_positions': 448, 'classifier_proj_size': 256, 'use_weighted_layer_sum': False, 'apply_spec_augment': False, 'mask_time_prob': 0.05, 'mask_time_length': 10, 'mask_time_min_masks': 2, 'mask_feature_prob': 0.0, 'mask_feature_length': 10, 'mask_feature_min_masks': 0, 'median_filter_width': 7, 'return_dict': True, 'output_hidden_states': False, 'output_attentions': False, 'torchscript': False, 'torch_dtype': 'float32', 'use_bfloat16': False, 'tf_legacy_loss': False, 'pruned_heads': {}, 'tie_word_embeddings': True, 'chunk_size_feed_forward': 0, 'is_encoder_decoder': True, 'is_decoder': False, 'cross_attention_hidden_size': None, 'add_cross_attention': False, 'tie_encoder_decoder': False, 'max_length': 448, 'min_length': 0, 'do_sample': False, 'early_stopping': False, 'num_beams': 1, 'num_beam_groups': 1, 'diversity_penalty': 0.0, 'temperature': 1.0, 'top_k': 50, 'top_p': 1.0, 'typical_p': 1.0, 'repetition_penalty': 1.0, 'length_penalty': 1.0, 'no_repeat_ngram_size': 0, 'encoder_no_repeat_ngram_size': 0, 'bad_words_ids': None, 'num_return_sequences': 1, 'output_scores': False, 'return_dict_in_generate': False, 'forced_bos_token_id': None, 'forced_eos_token_id': None, 'remove_invalid_values': False, 'exponential_decay_length_penalty': None, 'suppress_tokens': None, 'begin_suppress_tokens': [220, 50257], 'architectures': ['WhisperForConditionalGeneration'], 'finetuning_task': None, 'id2label': {0: 'LABEL_0', 1: 'LABEL_1'}, 'label2id': {'LABEL_0': 0, 'LABEL_1': 1}, 'tokenizer_class': None, 'prefix': None, 'bos_token_id': 50257, 'pad_token_id': 50257, 'eos_token_id': 50257, 'sep_token_id': None, 'decoder_start_token_id': 50258, 'task_specific_params': None, 'problem_type': None, '_name_or_path': 'openai/whisper-small', '_attn_implementation_autoset': True, 'transformers_version': '4.49.0.dev0', 'forced_decoder_ids': None, 'model_type': 'whisper', 'output_dir': './', 'overwrite_output_dir': True, 'do_train': True, 'do_eval': True, 'do_predict': False, 'eval_strategy': 'steps', 'prediction_loss_only': False, 'per_device_train_batch_size': 32, 'per_device_eval_batch_size': 16, 'per_gpu_train_batch_size': None, 'per_gpu_eval_batch_size': None, 'gradient_accumulation_steps': 1, 'eval_accumulation_steps': None, 'eval_delay': 0, 'torch_empty_cache_steps': None, 'learning_rate': 1e-05, 'weight_decay': 0.0, 'adam_beta1': 0.9, 'adam_beta2': 0.999, 'adam_epsilon': 1e-08, 'max_grad_norm': 1.0, 'num_train_epochs': 3.0, 'max_steps': 8000, 'lr_scheduler_type': 'linear', 'lr_scheduler_kwargs': {}, 'warmup_ratio': 0.0, 'warmup_steps': 500, 'log_level': 'passive', 'log_level_replica': 'warning', 'log_on_each_node': True, 'logging_dir': './runs/Feb12_12-58-59_tknika', 'logging_strategy': 'steps', 'logging_first_step': False, 'logging_steps': 25, 'logging_nan_inf_filter': True, 'save_strategy': 'steps', 'save_steps': 1000, 'save_total_limit': None, 'save_safetensors': True, 'save_on_each_node': False, 'save_only_model': False, 'restore_callback_states_from_checkpoint': False, 'no_cuda': False, 'use_cpu': False, 'use_mps_device': False, 'seed': 42, 'data_seed': None, 'jit_mode_eval': False, 'use_ipex': False, 'bf16': False, 'fp16': True, 'fp16_opt_level': 'O1', 'half_precision_backend': 'auto', 'bf16_full_eval': False, 'fp16_full_eval': False, 'tf32': None, 'local_rank': 0, 'ddp_backend': None, 'tpu_num_cores': None, 'tpu_metrics_debug': False, 'debug': [], 'dataloader_drop_last': False, 'eval_steps': 1000, 'dataloader_num_workers': 0, 'dataloader_prefetch_factor': None, 'past_index': -1, 'run_name': 'whisper-small-eu', 'disable_tqdm': False, 'remove_unused_columns': True, 'label_names': None, 'load_best_model_at_end': True, 'metric_for_best_model': 'wer', 'greater_is_better': False, 'ignore_data_skip': False, 'fsdp': [], 'fsdp_min_num_params': 0, 'fsdp_config': {'min_num_params': 0, 'xla': False, 'xla_fsdp_v2': False, 'xla_fsdp_grad_ckpt': False}, 'fsdp_transformer_layer_cls_to_wrap': None, 'accelerator_config': {'split_batches': False, 'dispatch_batches': None, 'even_batches': True, 'use_seedable_sampler': True, 'non_blocking': False, 'gradient_accumulation_kwargs': None}, 'deepspeed': None, 'label_smoothing_factor': 0.0, 'optim': 'adamw_torch', 'optim_args': None, 'adafactor': False, 'group_by_length': False, 'length_column_name': 'input_length', 'report_to': ['wandb'], 'ddp_find_unused_parameters': None, 'ddp_bucket_cap_mb': None, 'ddp_broadcast_buffers': None, 'dataloader_pin_memory': True, 'dataloader_persistent_workers': False, 'skip_memory_metrics': True, 'use_legacy_prediction_loop': False, 'push_to_hub': True, 'resume_from_checkpoint': None, 'hub_model_id': None, 'hub_strategy': 'every_save', 'hub_token': '<HUB_TOKEN>', 'hub_private_repo': None, 'hub_always_push': False, 'gradient_checkpointing': True, 'gradient_checkpointing_kwargs': None, 'include_inputs_for_metrics': False, 'include_for_metrics': [], 'eval_do_concat_batches': True, 'fp16_backend': 'auto', 'evaluation_strategy': 'steps', 'push_to_hub_model_id': None, 'push_to_hub_organization': None, 'push_to_hub_token': '<PUSH_TO_HUB_TOKEN>', 'mp_parameters': '', 'auto_find_batch_size': False, 'full_determinism': False, 'torchdynamo': None, 'ray_scope': 'last', 'ddp_timeout': 1800, 'torch_compile': False, 'torch_compile_backend': None, 'torch_compile_mode': None, 'dispatch_batches': None, 'split_batches': None, 'include_tokens_per_second': False, 'include_num_input_tokens_seen': False, 'neftune_noise_alpha': None, 'optim_target_modules': None, 'batch_eval_metrics': False, 'eval_on_start': False, 'use_liger_kernel': False, 'eval_use_gather_object': False, 'average_tokens_across_devices': False, 'sortish_sampler': False, 'predict_with_generate': True, 'generation_max_length': 225, 'generation_num_beams': None, 'generation_config': None}
-2025-02-12 12:59:25,306 INFO    MainThread:226591 [wandb_config.py:__setitem__():154] config set model/num_parameters = 241734912 - <bound method Run._config_callback of <wandb.sdk.wandb_run.Run object at 0x76451d282f30>>
-2025-02-12 12:59:25,306 INFO    MainThread:226591 [wandb_run.py:_config_callback():1253] config_cb model/num_parameters 241734912 None
-2025-02-12 12:59:25,342 WARNING MsgRouterThr:226591 [router.py:message_loop():75] message_loop has been closed

wandb/run-20250212_130533-zeu6vay4/files/config.yaml

@@ -1,512 +0,0 @@
-_attn_implementation_autoset:
-    value: true
-_name_or_path:
-    value: openai/whisper-small
-_wandb:
-    value:
-        cli_version: 0.19.6
-        m:
-            - "1": train/global_step
-              "6":
-                - 3
-              "7": []
-        python_version: 3.12.3
-        t:
-            "1":
-                - 1
-                - 5
-                - 11
-                - 49
-                - 51
-                - 53
-                - 55
-                - 71
-                - 100
-            "2":
-                - 1
-                - 5
-                - 11
-                - 49
-                - 51
-                - 53
-                - 55
-                - 71
-                - 100
-            "3":
-                - 7
-                - 13
-                - 19
-                - 23
-                - 55
-                - 66
-            "4": 3.12.3
-            "5": 0.19.6
-            "6": 4.49.0.dev0
-            "8":
-                - 5
-            "9":
-                "1": transformers_trainer
-            "12": 0.19.6
-            "13": linux-x86_64
-accelerator_config:
-    value:
-        dispatch_batches: null
-        even_batches: true
-        gradient_accumulation_kwargs: null
-        non_blocking: false
-        split_batches: false
-        use_seedable_sampler: true
-activation_dropout:
-    value: 0
-activation_function:
-    value: gelu
-adafactor:
-    value: false
-adam_beta1:
-    value: 0.9
-adam_beta2:
-    value: 0.999
-adam_epsilon:
-    value: 1e-08
-add_cross_attention:
-    value: false
-apply_spec_augment:
-    value: false
-architectures:
-    value:
-        - WhisperForConditionalGeneration
-attention_dropout:
-    value: 0
-auto_find_batch_size:
-    value: false
-average_tokens_across_devices:
-    value: false
-bad_words_ids:
-    value: null
-batch_eval_metrics:
-    value: false
-begin_suppress_tokens:
-    value:
-        - 220
-        - 50257
-bf16:
-    value: false
-bf16_full_eval:
-    value: false
-bos_token_id:
-    value: 50257
-chunk_size_feed_forward:
-    value: 0
-classifier_proj_size:
-    value: 256
-cross_attention_hidden_size:
-    value: null
-d_model:
-    value: 768
-data_seed:
-    value: null
-dataloader_drop_last:
-    value: false
-dataloader_num_workers:
-    value: 0
-dataloader_persistent_workers:
-    value: false
-dataloader_pin_memory:
-    value: true
-dataloader_prefetch_factor:
-    value: null
-ddp_backend:
-    value: null
-ddp_broadcast_buffers:
-    value: null
-ddp_bucket_cap_mb:
-    value: null
-ddp_find_unused_parameters:
-    value: null
-ddp_timeout:
-    value: 1800
-debug:
-    value: []
-decoder_attention_heads:
-    value: 12
-decoder_ffn_dim:
-    value: 3072
-decoder_layerdrop:
-    value: 0
-decoder_layers:
-    value: 12
-decoder_start_token_id:
-    value: 50258
-deepspeed:
-    value: null
-disable_tqdm:
-    value: false
-dispatch_batches:
-    value: null
-diversity_penalty:
-    value: 0
-do_eval:
-    value: true
-do_predict:
-    value: false
-do_sample:
-    value: false
-do_train:
-    value: true
-dropout:
-    value: 0
-early_stopping:
-    value: false
-encoder_attention_heads:
-    value: 12
-encoder_ffn_dim:
-    value: 3072
-encoder_layerdrop:
-    value: 0
-encoder_layers:
-    value: 12
-encoder_no_repeat_ngram_size:
-    value: 0
-eos_token_id:
-    value: 50257
-eval_accumulation_steps:
-    value: null
-eval_delay:
-    value: 0
-eval_do_concat_batches:
-    value: true
-eval_on_start:
-    value: false
-eval_steps:
-    value: 1000
-eval_strategy:
-    value: steps
-eval_use_gather_object:
-    value: false
-evaluation_strategy:
-    value: steps
-exponential_decay_length_penalty:
-    value: null
-finetuning_task:
-    value: null
-forced_bos_token_id:
-    value: null
-forced_decoder_ids:
-    value: null
-forced_eos_token_id:
-    value: null
-fp16:
-    value: true
-fp16_backend:
-    value: auto
-fp16_full_eval:
-    value: false
-fp16_opt_level:
-    value: O1
-fsdp:
-    value: []
-fsdp_config:
-    value:
-        min_num_params: 0
-        xla: false
-        xla_fsdp_grad_ckpt: false
-        xla_fsdp_v2: false
-fsdp_min_num_params:
-    value: 0
-fsdp_transformer_layer_cls_to_wrap:
-    value: null
-full_determinism:
-    value: false
-generation_config:
-    value: null
-generation_max_length:
-    value: 225
-generation_num_beams:
-    value: null
-gradient_accumulation_steps:
-    value: 1
-gradient_checkpointing:
-    value: true
-gradient_checkpointing_kwargs:
-    value: null
-greater_is_better:
-    value: false
-group_by_length:
-    value: false
-half_precision_backend:
-    value: auto
-hub_always_push:
-    value: false
-hub_model_id:
-    value: null
-hub_private_repo:
-    value: null
-hub_strategy:
-    value: every_save
-hub_token:
-    value: <HUB_TOKEN>
-id2label:
-    value:
-        "0": LABEL_0
-        "1": LABEL_1
-ignore_data_skip:
-    value: false
-include_for_metrics:
-    value: []
-include_inputs_for_metrics:
-    value: false
-include_num_input_tokens_seen:
-    value: false
-include_tokens_per_second:
-    value: false
-init_std:
-    value: 0.02
-is_decoder:
-    value: false
-is_encoder_decoder:
-    value: true
-jit_mode_eval:
-    value: false
-label_names:
-    value: null
-label_smoothing_factor:
-    value: 0
-label2id:
-    value:
-        LABEL_0: 0
-        LABEL_1: 1
-learning_rate:
-    value: 1e-05
-length_column_name:
-    value: input_length
-length_penalty:
-    value: 1
-load_best_model_at_end:
-    value: true
-local_rank:
-    value: 0
-log_level:
-    value: passive
-log_level_replica:
-    value: warning
-log_on_each_node:
-    value: true
-logging_dir:
-    value: ./runs/Feb12_13-05-10_tknika
-logging_first_step:
-    value: false
-logging_nan_inf_filter:
-    value: true
-logging_steps:
-    value: 25
-logging_strategy:
-    value: steps
-lr_scheduler_type:
-    value: linear
-mask_feature_length:
-    value: 10
-mask_feature_min_masks:
-    value: 0
-mask_feature_prob:
-    value: 0
-mask_time_length:
-    value: 10
-mask_time_min_masks:
-    value: 2
-mask_time_prob:
-    value: 0.05
-max_grad_norm:
-    value: 1
-max_length:
-    value: 448
-max_source_positions:
-    value: 1500
-max_steps:
-    value: 8000
-max_target_positions:
-    value: 448
-median_filter_width:
-    value: 7
-metric_for_best_model:
-    value: wer
-min_length:
-    value: 0
-model/num_parameters:
-    value: 241734912
-model_type:
-    value: whisper
-mp_parameters:
-    value: ""
-neftune_noise_alpha:
-    value: null
-no_cuda:
-    value: false
-no_repeat_ngram_size:
-    value: 0
-num_beam_groups:
-    value: 1
-num_beams:
-    value: 1
-num_hidden_layers:
-    value: 12
-num_mel_bins:
-    value: 80
-num_return_sequences:
-    value: 1
-num_train_epochs:
-    value: 3
-optim:
-    value: adamw_torch
-optim_args:
-    value: null
-optim_target_modules:
-    value: null
-output_attentions:
-    value: false
-output_dir:
-    value: ./
-output_hidden_states:
-    value: false
-output_scores:
-    value: false
-overwrite_output_dir:
-    value: true
-pad_token_id:
-    value: 50257
-past_index:
-    value: -1
-per_device_eval_batch_size:
-    value: 16
-per_device_train_batch_size:
-    value: 32
-per_gpu_eval_batch_size:
-    value: null
-per_gpu_train_batch_size:
-    value: null
-predict_with_generate:
-    value: true
-prediction_loss_only:
-    value: false
-prefix:
-    value: null
-problem_type:
-    value: null
-push_to_hub:
-    value: true
-push_to_hub_model_id:
-    value: null
-push_to_hub_organization:
-    value: null
-push_to_hub_token:
-    value: <PUSH_TO_HUB_TOKEN>
-ray_scope:
-    value: last
-remove_invalid_values:
-    value: false
-remove_unused_columns:
-    value: true
-repetition_penalty:
-    value: 1
-report_to:
-    value:
-        - wandb
-restore_callback_states_from_checkpoint:
-    value: false
-resume_from_checkpoint:
-    value: null
-return_dict:
-    value: true
-return_dict_in_generate:
-    value: false
-run_name:
-    value: whisper-small-eu
-save_on_each_node:
-    value: false
-save_only_model:
-    value: false
-save_safetensors:
-    value: true
-save_steps:
-    value: 1000
-save_strategy:
-    value: steps
-save_total_limit:
-    value: null
-scale_embedding:
-    value: false
-seed:
-    value: 42
-sep_token_id:
-    value: null
-skip_memory_metrics:
-    value: true
-sortish_sampler:
-    value: false
-split_batches:
-    value: null
-suppress_tokens:
-    value: null
-task_specific_params:
-    value: null
-temperature:
-    value: 1
-tf_legacy_loss:
-    value: false
-tf32:
-    value: null
-tie_encoder_decoder:
-    value: false
-tie_word_embeddings:
-    value: true
-tokenizer_class:
-    value: null
-top_k:
-    value: 50
-top_p:
-    value: 1
-torch_compile:
-    value: false
-torch_compile_backend:
-    value: null
-torch_compile_mode:
-    value: null
-torch_dtype:
-    value: float32
-torch_empty_cache_steps:
-    value: null
-torchdynamo:
-    value: null
-torchscript:
-    value: false
-tpu_metrics_debug:
-    value: false
-tpu_num_cores:
-    value: null
-transformers_version:
-    value: 4.49.0.dev0
-typical_p:
-    value: 1
-use_bfloat16:
-    value: false
-use_cache:
-    value: false
-use_cpu:
-    value: false
-use_ipex:
-    value: false
-use_legacy_prediction_loop:
-    value: false
-use_liger_kernel:
-    value: false
-use_mps_device:
-    value: false
-use_weighted_layer_sum:
-    value: false
-vocab_size:
-    value: 51865
-warmup_ratio:
-    value: 0
-warmup_steps:
-    value: 500
-weight_decay:
-    value: 0

wandb/run-20250212_130533-zeu6vay4/files/output.log

@@ -1,22 +0,0 @@
-  0%|                                                                                                                                               | 0/8000 [00:00<?, ?it/s]Traceback (most recent call last):
-  File "/home/tknika/xezpeleta/whisper/whisper-small-eu/run_speech_recognition_seq2seq_streaming.py", line 633, in <module>
-    main()
-  File "/home/tknika/xezpeleta/whisper/whisper-small-eu/run_speech_recognition_seq2seq_streaming.py", line 582, in main
-    train_result = trainer.train(resume_from_checkpoint=checkpoint)
-                   ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
-  File "/home/tknika/xezpeleta/whisper/whisper-small-eu/.venv/lib/python3.12/site-packages/transformers/trainer.py", line 2197, in train
-    return inner_training_loop(
-           ^^^^^^^^^^^^^^^^^^^^
-  File "/home/tknika/xezpeleta/whisper/whisper-small-eu/.venv/lib/python3.12/site-packages/transformers/trainer.py", line 2438, in _inner_training_loop
-    self.control = self.callback_handler.on_epoch_begin(args, self.state, self.control)
-                   ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
-  File "/home/tknika/xezpeleta/whisper/whisper-small-eu/.venv/lib/python3.12/site-packages/transformers/trainer_callback.py", line 514, in on_epoch_begin
-    return self.call_event("on_epoch_begin", args, state, control)
-           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
-  File "/home/tknika/xezpeleta/whisper/whisper-small-eu/.venv/lib/python3.12/site-packages/transformers/trainer_callback.py", line 557, in call_event
-    result = getattr(callback, event)(
-             ^^^^^^^^^^^^^^^^^^^^^^^^^
-  File "/home/tknika/xezpeleta/whisper/whisper-small-eu/run_speech_recognition_seq2seq_streaming.py", line 557, in on_epoch_begin
-    if isinstance(train_dataloader.dataset, IterableDatasetShard):
-                  ^^^^^^^^^^^^^^^^^^^^^^^^
-AttributeError: 'NoneType' object has no attribute 'dataset'