Model save

README.md

@@ -1,33 +1,40 @@
 ---
 library_name: transformers
-language:
-- eu
 license: apache-2.0
 base_model: openai/whisper-large-v3
 tags:
-- whisper-event
 - generated_from_trainer
 datasets:
-- mozilla-foundation/common_voice_17_0
+- common_voice_17_0
+metrics:
+- wer
 model-index:
-- name: Whisper Large Basque
-  results: []
+- name: openai/whisper-large-v3
+  results:
+  - task:
+      name: Automatic Speech Recognition
+      type: automatic-speech-recognition
+    dataset:
+      name: common_voice_17_0
+      type: common_voice_17_0
+      config: eu
+      split: test
+      args: eu
+    metrics:
+    - name: Wer
+      type: wer
+      value: 7.215361500971087
 ---
 
 <!-- This model card has been generated automatically according to the information the Trainer had access to. You
 should probably proofread and complete it, then remove this comment. -->
 
-# Whisper Large Basque
+# openai/whisper-large-v3
 
-This model is a fine-tuned version of [openai/whisper-large-v3](https://huggingface.co/openai/whisper-large-v3) on the mozilla-foundation/common_voice_17_0 eu dataset.
+This model is a fine-tuned version of [openai/whisper-large-v3](https://huggingface.co/openai/whisper-large-v3) on the common_voice_17_0 dataset.
 It achieves the following results on the evaluation set:
-- eval_loss: 0.9278
-- eval_model_preparation_time: 0.0102
-- eval_wer: 44.2953
-- eval_runtime: 4165.1595
-- eval_samples_per_second: 3.272
-- eval_steps_per_second: 0.409
-- step: 0
+- Loss: 0.1259
+- Wer: 7.2154
 
 ## Model description
 
@@ -56,6 +63,30 @@ The following hyperparameters were used during training:
 - training_steps: 10000
 - mixed_precision_training: Native AMP
 
+### Training results
+
+| Training Loss | Epoch  | Step | Validation Loss | Wer     |
+|:-------------:|:------:|:----:|:---------------:|:-------:|
+| 0.2208        | 0.05   | 500  | 0.2592          | 20.6915 |
+| 0.1489        | 0.1    | 1000 | 0.1971          | 14.6827 |
+| 0.1973        | 0.15   | 1500 | 0.1747          | 12.3777 |
+| 0.1353        | 1.0296 | 2000 | 0.1527          | 10.7195 |
+| 0.1065        | 1.0796 | 2500 | 0.1456          | 9.8694  |
+| 0.106         | 1.1296 | 3000 | 0.1362          | 9.0925  |
+| 0.0718        | 2.0092 | 3500 | 0.1326          | 8.5428  |
+| 0.0683        | 2.0592 | 4000 | 0.1343          | 8.4851  |
+| 0.0482        | 2.1092 | 4500 | 0.1336          | 8.1049  |
+| 0.0548        | 2.1592 | 5000 | 0.1316          | 7.9244  |
+| 0.0282        | 3.0388 | 5500 | 0.1391          | 7.8182  |
+| 0.025         | 3.0888 | 6000 | 0.1425          | 7.9409  |
+| 0.0274        | 3.1388 | 6500 | 0.1391          | 7.7311  |
+| 0.0155        | 4.0184 | 7000 | 0.1492          | 7.6972  |
+| 0.0189        | 4.0684 | 7500 | 0.1517          | 7.6569  |
+| 0.0139        | 4.1184 | 8000 | 0.1539          | 7.6267  |
+| 0.0141        | 4.1684 | 8500 | 0.1550          | 7.5424  |
+| 0.0368        | 5.048  | 9000 | 0.1259          | 7.2154  |
+
+
 ### Framework versions
 
 - Transformers 4.46.0.dev0

model-00001-of-00002.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:08e0005225b3dbaf55dd13ac62926cc7e02c1025d66fa375e6fb305ff79cd4f9
+oid sha256:70e95881b347aab16bfadb8b06b24d389bceb51ce8b9894db264ed67fc19f29e
 size 4993448880

model-00002-of-00002.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:630ca774672856d2e0e39a702e590f635a1cfc5726a64b6578ab46dd367369a9
+oid sha256:52bad486fd09fef8dae53b1e1e2b13ea27a7a645786a7a9920fe7344b96bf162
 size 1180663192

run.sh

@@ -31,6 +31,8 @@ WANDB_PROJECT=whisper-medium-eu \
 	--gradient_checkpointing \
 	--fp16 \
 	--overwrite_output_dir \
+	--resume_from_checkpoint="checkpoint-9000" \
+	--do_train \
 	--do_eval \
 	--predict_with_generate \
 	--do_normalize_eval \

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:4b703135451bdb3fdf1b0263595ef460845b9b248a97e458a32874babc3e4138
+oid sha256:bf7a986abd3107068a3a0d5d4f994d9792d101aaf76cd5472057cde7be944e18
 size 5368

wandb/debug-internal.log

@@ -1,10 +1,10 @@
-{"time":"2024-10-07T12:56:15.257353437Z","level":"INFO","msg":"using version","core version":"0.18.3"}
-{"time":"2024-10-07T12:56:15.257380326Z","level":"INFO","msg":"created symlink","path":"/home/tknika/whisper-large-eu/wandb/run-20241007_125615-a3z1jk8c/logs/debug-core.log"}
-{"time":"2024-10-07T12:56:15.259721418Z","level":"ERROR","msg":"dialing: google: could not find default credentials. See https://cloud.google.com/docs/authentication/external/set-up-adc for more information"}
-{"time":"2024-10-07T12:56:15.26442537Z","level":"INFO","msg":"created new stream","id":"a3z1jk8c"}
-{"time":"2024-10-07T12:56:15.264442509Z","level":"INFO","msg":"stream: started","id":"a3z1jk8c"}
-{"time":"2024-10-07T12:56:15.264458959Z","level":"INFO","msg":"handler: started","stream_id":{"value":"a3z1jk8c"}}
-{"time":"2024-10-07T12:56:15.264475109Z","level":"INFO","msg":"sender: started","stream_id":{"value":"a3z1jk8c"}}
-{"time":"2024-10-07T12:56:15.264497739Z","level":"INFO","msg":"writer: Do: started","stream_id":{"value":"a3z1jk8c"}}
-{"time":"2024-10-07T12:56:15.681557119Z","level":"INFO","msg":"wandb-core","!BADKEY":null}
-{"time":"2024-10-07T12:56:15.68260129Z","level":"INFO","msg":"Starting system monitor"}
+{"time":"2024-10-07T13:18:49.427383059Z","level":"INFO","msg":"using version","core version":"0.18.3"}
+{"time":"2024-10-07T13:18:49.427406249Z","level":"INFO","msg":"created symlink","path":"/home/tknika/whisper-large-eu/wandb/run-20241007_131849-0rbzerob/logs/debug-core.log"}
+{"time":"2024-10-07T13:18:49.429699121Z","level":"ERROR","msg":"dialing: google: could not find default credentials. See https://cloud.google.com/docs/authentication/external/set-up-adc for more information"}
+{"time":"2024-10-07T13:18:49.436074781Z","level":"INFO","msg":"created new stream","id":"0rbzerob"}
+{"time":"2024-10-07T13:18:49.436114921Z","level":"INFO","msg":"stream: started","id":"0rbzerob"}
+{"time":"2024-10-07T13:18:49.43615474Z","level":"INFO","msg":"writer: Do: started","stream_id":{"value":"0rbzerob"}}
+{"time":"2024-10-07T13:18:49.43615079Z","level":"INFO","msg":"sender: started","stream_id":{"value":"0rbzerob"}}
+{"time":"2024-10-07T13:18:49.43618894Z","level":"INFO","msg":"handler: started","stream_id":{"value":"0rbzerob"}}
+{"time":"2024-10-07T13:18:49.857246919Z","level":"INFO","msg":"wandb-core","!BADKEY":null}
+{"time":"2024-10-07T13:18:49.859623571Z","level":"INFO","msg":"Starting system monitor"}

wandb/debug.log

@@ -1,28 +1,28 @@
-2024-10-07 12:56:15,251 INFO    MainThread:20958 [wandb_setup.py:_flush():79] Current SDK version is 0.18.3
-2024-10-07 12:56:15,252 INFO    MainThread:20958 [wandb_setup.py:_flush():79] Configure stats pid to 20958
-2024-10-07 12:56:15,252 INFO    MainThread:20958 [wandb_setup.py:_flush():79] Loading settings from /home/tknika/.config/wandb/settings
-2024-10-07 12:56:15,252 INFO    MainThread:20958 [wandb_setup.py:_flush():79] Loading settings from /home/tknika/whisper-large-eu/wandb/settings
-2024-10-07 12:56:15,252 INFO    MainThread:20958 [wandb_setup.py:_flush():79] Loading settings from environment variables: {'project': 'whisper-medium-eu'}
-2024-10-07 12:56:15,252 INFO    MainThread:20958 [wandb_setup.py:_flush():79] Applying setup settings: {'mode': None, '_disable_service': None}
-2024-10-07 12:56:15,252 INFO    MainThread:20958 [wandb_setup.py:_flush():79] Inferring run settings from compute environment: {'program_relpath': 'run_speech_recognition_seq2seq_streaming.py', 'program_abspath': '/home/tknika/whisper-large-eu/run_speech_recognition_seq2seq_streaming.py', 'program': '/home/tknika/whisper-large-eu/run_speech_recognition_seq2seq_streaming.py'}
-2024-10-07 12:56:15,252 INFO    MainThread:20958 [wandb_setup.py:_flush():79] Applying login settings: {}
-2024-10-07 12:56:15,252 INFO    MainThread:20958 [wandb_init.py:_log_setup():532] Logging user logs to /home/tknika/whisper-large-eu/wandb/run-20241007_125615-a3z1jk8c/logs/debug.log
-2024-10-07 12:56:15,252 INFO    MainThread:20958 [wandb_init.py:_log_setup():533] Logging internal logs to /home/tknika/whisper-large-eu/wandb/run-20241007_125615-a3z1jk8c/logs/debug-internal.log
-2024-10-07 12:56:15,252 INFO    MainThread:20958 [wandb_init.py:init():617] calling init triggers
-2024-10-07 12:56:15,252 INFO    MainThread:20958 [wandb_init.py:init():624] wandb.init called with sweep_config: {}
+2024-10-07 13:18:49,422 INFO    MainThread:21209 [wandb_setup.py:_flush():79] Current SDK version is 0.18.3
+2024-10-07 13:18:49,422 INFO    MainThread:21209 [wandb_setup.py:_flush():79] Configure stats pid to 21209
+2024-10-07 13:18:49,422 INFO    MainThread:21209 [wandb_setup.py:_flush():79] Loading settings from /home/tknika/.config/wandb/settings
+2024-10-07 13:18:49,422 INFO    MainThread:21209 [wandb_setup.py:_flush():79] Loading settings from /home/tknika/whisper-large-eu/wandb/settings
+2024-10-07 13:18:49,422 INFO    MainThread:21209 [wandb_setup.py:_flush():79] Loading settings from environment variables: {'project': 'whisper-medium-eu'}
+2024-10-07 13:18:49,422 INFO    MainThread:21209 [wandb_setup.py:_flush():79] Applying setup settings: {'mode': None, '_disable_service': None}
+2024-10-07 13:18:49,422 INFO    MainThread:21209 [wandb_setup.py:_flush():79] Inferring run settings from compute environment: {'program_relpath': 'run_speech_recognition_seq2seq_streaming.py', 'program_abspath': '/home/tknika/whisper-large-eu/run_speech_recognition_seq2seq_streaming.py', 'program': '/home/tknika/whisper-large-eu/run_speech_recognition_seq2seq_streaming.py'}
+2024-10-07 13:18:49,422 INFO    MainThread:21209 [wandb_setup.py:_flush():79] Applying login settings: {}
+2024-10-07 13:18:49,422 INFO    MainThread:21209 [wandb_init.py:_log_setup():532] Logging user logs to /home/tknika/whisper-large-eu/wandb/run-20241007_131849-0rbzerob/logs/debug.log
+2024-10-07 13:18:49,422 INFO    MainThread:21209 [wandb_init.py:_log_setup():533] Logging internal logs to /home/tknika/whisper-large-eu/wandb/run-20241007_131849-0rbzerob/logs/debug-internal.log
+2024-10-07 13:18:49,422 INFO    MainThread:21209 [wandb_init.py:init():617] calling init triggers
+2024-10-07 13:18:49,422 INFO    MainThread:21209 [wandb_init.py:init():624] wandb.init called with sweep_config: {}
 config: {}
-2024-10-07 12:56:15,252 INFO    MainThread:20958 [wandb_init.py:init():667] starting backend
-2024-10-07 12:56:15,252 INFO    MainThread:20958 [wandb_init.py:init():671] sending inform_init request
-2024-10-07 12:56:15,254 INFO    MainThread:20958 [backend.py:_multiprocessing_setup():104] multiprocessing start_methods=fork,spawn,forkserver, using: spawn
-2024-10-07 12:56:15,254 INFO    MainThread:20958 [wandb_init.py:init():684] backend started and connected
-2024-10-07 12:56:15,258 INFO    MainThread:20958 [wandb_init.py:init():779] updated telemetry
-2024-10-07 12:56:15,265 INFO    MainThread:20958 [wandb_init.py:init():812] communicating run to backend with 90.0 second timeout
-2024-10-07 12:56:15,676 INFO    MainThread:20958 [wandb_init.py:init():863] starting run threads in backend
-2024-10-07 12:56:15,774 INFO    MainThread:20958 [wandb_run.py:_console_start():2465] atexit reg
-2024-10-07 12:56:15,774 INFO    MainThread:20958 [wandb_run.py:_redirect():2313] redirect: wrap_raw
-2024-10-07 12:56:15,774 INFO    MainThread:20958 [wandb_run.py:_redirect():2378] Wrapping output streams.
-2024-10-07 12:56:15,774 INFO    MainThread:20958 [wandb_run.py:_redirect():2403] Redirects installed.
-2024-10-07 12:56:15,775 INFO    MainThread:20958 [wandb_init.py:init():907] run started, returning control to user process
-2024-10-07 12:56:15,777 INFO    MainThread:20958 [wandb_run.py:_config_callback():1394] config_cb None None {'vocab_size': 51866, 'num_mel_bins': 128, 'd_model': 1280, 'encoder_layers': 32, 'encoder_attention_heads': 20, 'decoder_layers': 32, 'decoder_attention_heads': 20, 'decoder_ffn_dim': 5120, 'encoder_ffn_dim': 5120, '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': 32, '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': 'float16', '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': 50256, '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-large-v3', 'transformers_version': '4.46.0.dev0', 'model_type': 'whisper', 'forced_decoder_ids': None, 'output_dir': './', 'overwrite_output_dir': True, 'do_train': False, 'do_eval': True, 'do_predict': False, 'eval_strategy': 'steps', 'prediction_loss_only': False, 'per_device_train_batch_size': 16, 'per_device_eval_batch_size': 8, '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': 4.375e-06, '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': 10000, '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/Oct07_11-46-39_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': 500, 'dataloader_num_workers': 0, 'dataloader_prefetch_factor': None, 'past_index': -1, 'run_name': 'whisper-large-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': False, '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, 'sortish_sampler': False, 'predict_with_generate': True, 'generation_max_length': 228, 'generation_num_beams': None, 'generation_config': None}
-2024-10-07 12:56:15,780 INFO    MainThread:20958 [wandb_config.py:__setitem__():154] config set model/num_parameters = 1543490560 - <bound method Run._config_callback of <wandb.sdk.wandb_run.Run object at 0x748ced2ceae0>>
-2024-10-07 12:56:15,780 INFO    MainThread:20958 [wandb_run.py:_config_callback():1394] config_cb model/num_parameters 1543490560 None
+2024-10-07 13:18:49,422 INFO    MainThread:21209 [wandb_init.py:init():667] starting backend
+2024-10-07 13:18:49,422 INFO    MainThread:21209 [wandb_init.py:init():671] sending inform_init request
+2024-10-07 13:18:49,424 INFO    MainThread:21209 [backend.py:_multiprocessing_setup():104] multiprocessing start_methods=fork,spawn,forkserver, using: spawn
+2024-10-07 13:18:49,425 INFO    MainThread:21209 [wandb_init.py:init():684] backend started and connected
+2024-10-07 13:18:49,428 INFO    MainThread:21209 [wandb_init.py:init():779] updated telemetry
+2024-10-07 13:18:49,436 INFO    MainThread:21209 [wandb_init.py:init():812] communicating run to backend with 90.0 second timeout
+2024-10-07 13:18:49,852 INFO    MainThread:21209 [wandb_init.py:init():863] starting run threads in backend
+2024-10-07 13:18:49,971 INFO    MainThread:21209 [wandb_run.py:_console_start():2465] atexit reg
+2024-10-07 13:18:49,972 INFO    MainThread:21209 [wandb_run.py:_redirect():2313] redirect: wrap_raw
+2024-10-07 13:18:49,972 INFO    MainThread:21209 [wandb_run.py:_redirect():2378] Wrapping output streams.
+2024-10-07 13:18:49,972 INFO    MainThread:21209 [wandb_run.py:_redirect():2403] Redirects installed.
+2024-10-07 13:18:49,973 INFO    MainThread:21209 [wandb_init.py:init():907] run started, returning control to user process
+2024-10-07 13:18:49,974 INFO    MainThread:21209 [wandb_run.py:_config_callback():1394] config_cb None None {'vocab_size': 51866, 'num_mel_bins': 128, 'd_model': 1280, 'encoder_layers': 32, 'encoder_attention_heads': 20, 'decoder_layers': 32, 'decoder_attention_heads': 20, 'decoder_ffn_dim': 5120, 'encoder_ffn_dim': 5120, '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': 32, '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': 'float16', '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': 50256, '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-large-v3', 'transformers_version': '4.46.0.dev0', 'model_type': 'whisper', 'forced_decoder_ids': None, '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': 16, 'per_device_eval_batch_size': 8, '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': 4.375e-06, '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': 10000, '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/Oct07_13-18-21_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': 500, 'dataloader_num_workers': 0, 'dataloader_prefetch_factor': None, 'past_index': -1, 'run_name': 'whisper-large-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': 'checkpoint-9000', 'hub_model_id': None, 'hub_strategy': 'every_save', 'hub_token': '<HUB_TOKEN>', 'hub_private_repo': False, '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, 'sortish_sampler': False, 'predict_with_generate': True, 'generation_max_length': 228, 'generation_num_beams': None, 'generation_config': None}
+2024-10-07 13:18:49,978 INFO    MainThread:21209 [wandb_config.py:__setitem__():154] config set model/num_parameters = 1543490560 - <bound method Run._config_callback of <wandb.sdk.wandb_run.Run object at 0x7a0b4e8a1df0>>
+2024-10-07 13:18:49,978 INFO    MainThread:21209 [wandb_run.py:_config_callback():1394] config_cb model/num_parameters 1543490560 None

wandb/run-20241007_125615-a3z1jk8c/files/config.yaml

@@ -0,0 +1,544 @@
+_name_or_path:
+    value: openai/whisper-large-v3
+_wandb:
+    value:
+        cli_version: 0.18.3
+        m:
+            - "1": train/global_step
+              "6":
+                - 3
+              "7": []
+            - "1": eval/model_preparation_time
+              "5": 1
+              "6":
+                - 1
+                - 3
+              "7": []
+            - "1": eval/runtime
+              "5": 1
+              "6":
+                - 1
+                - 3
+              "7": []
+            - "1": eval/samples_per_second
+              "5": 1
+              "6":
+                - 1
+                - 3
+              "7": []
+            - "1": eval/steps_per_second
+              "5": 1
+              "6":
+                - 1
+                - 3
+              "7": []
+            - "1": eval/loss
+              "5": 1
+              "6":
+                - 1
+                - 3
+              "7": []
+            - "1": eval/wer
+              "5": 1
+              "6":
+                - 1
+                - 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": 0.18.3
+            "6": 4.46.0.dev0
+            "8":
+                - 5
+            "9":
+                "1": transformers_trainer
+            "12": 0.18.3
+            "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
+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: 1280
+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: 20
+decoder_ffn_dim:
+    value: 5120
+decoder_layerdrop:
+    value: 0
+decoder_layers:
+    value: 32
+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: false
+dropout:
+    value: 0
+early_stopping:
+    value: false
+encoder_attention_heads:
+    value: 20
+encoder_ffn_dim:
+    value: 5120
+encoder_layerdrop:
+    value: 0
+encoder_layers:
+    value: 32
+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: 500
+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: 228
+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: false
+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: 4.375e-06
+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/Oct07_11-46-39_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: 10000
+max_target_positions:
+    value: 448
+median_filter_width:
+    value: 7
+metric_for_best_model:
+    value: wer
+min_length:
+    value: 0
+model/num_parameters:
+    value: 1543490560
+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: 32
+num_mel_bins:
+    value: 128
+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: 50256
+past_index:
+    value: -1
+per_device_eval_batch_size:
+    value: 8
+per_device_train_batch_size:
+    value: 16
+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-large-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: float16
+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: 4.46.0.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: 51866
+warmup_ratio:
+    value: 0
+warmup_steps:
+    value: 500
+weight_decay:
+    value: 0

wandb/run-20241007_125615-a3z1jk8c/files/wandb-summary.json

@@ -0,0 +1 @@
+{"_step":0,"_wandb":{"runtime":266},"eval/runtime":4165.1595,"_timestamp":1.7283057757810946e+09,"eval/model_preparation_time":0.0102,"eval/steps_per_second":0.409,"train/global_step":0,"eval/wer":44.29532045879292,"_runtime":0.559212086,"eval/samples_per_second":3.272,"eval/loss":0.9277587532997131}

wandb/run-20241007_125615-a3z1jk8c/logs/debug-core.log

@@ -5,3 +5,10 @@
 {"time":"2024-10-07T12:56:14.698116167Z","level":"INFO","msg":"connection: ManageConnectionData: new connection created","id":"127.0.0.1:49446"}
 {"time":"2024-10-07T12:56:15.257025559Z","level":"INFO","msg":"handleInformInit: received","streamId":"a3z1jk8c","id":"127.0.0.1:49446"}
 {"time":"2024-10-07T12:56:15.264445669Z","level":"INFO","msg":"handleInformInit: stream started","streamId":"a3z1jk8c","id":"127.0.0.1:49446"}
+{"time":"2024-10-07T13:00:42.167739631Z","level":"INFO","msg":"handleInformTeardown: server teardown initiated","id":"127.0.0.1:49446"}
+{"time":"2024-10-07T13:00:42.167972019Z","level":"INFO","msg":"server is shutting down"}
+{"time":"2024-10-07T13:00:42.167958909Z","level":"INFO","msg":"connection: Close: initiating connection closure","id":"127.0.0.1:49446"}
+{"time":"2024-10-07T13:00:42.168215837Z","level":"INFO","msg":"connection: Close: connection successfully closed","id":"127.0.0.1:49446"}
+{"time":"2024-10-07T13:00:45.394906949Z","level":"INFO","msg":"handleInformTeardown: server shutdown complete","id":"127.0.0.1:49446"}
+{"time":"2024-10-07T13:00:45.394950439Z","level":"INFO","msg":"connection: ManageConnectionData: connection closed","id":"127.0.0.1:49446"}
+{"time":"2024-10-07T13:00:45.394977679Z","level":"INFO","msg":"server is closed"}

wandb/run-20241007_125615-a3z1jk8c/logs/debug-internal.log

@@ -8,3 +8,11 @@
 {"time":"2024-10-07T12:56:15.264497739Z","level":"INFO","msg":"writer: Do: started","stream_id":{"value":"a3z1jk8c"}}
 {"time":"2024-10-07T12:56:15.681557119Z","level":"INFO","msg":"wandb-core","!BADKEY":null}
 {"time":"2024-10-07T12:56:15.68260129Z","level":"INFO","msg":"Starting system monitor"}
+{"time":"2024-10-07T13:00:42.167889239Z","level":"INFO","msg":"stream: closing","id":"a3z1jk8c"}
+{"time":"2024-10-07T13:00:42.167946829Z","level":"INFO","msg":"Stopping system monitor"}
+{"time":"2024-10-07T13:00:42.173849946Z","level":"INFO","msg":"Stopped system monitor"}
+{"time":"2024-10-07T13:00:45.087931056Z","level":"INFO","msg":"fileTransfer: Close: file transfer manager closed"}
+{"time":"2024-10-07T13:00:45.394585792Z","level":"INFO","msg":"handler: closed","stream_id":{"value":"a3z1jk8c"}}
+{"time":"2024-10-07T13:00:45.394682341Z","level":"INFO","msg":"sender: closed","stream_id":{"value":"a3z1jk8c"}}
+{"time":"2024-10-07T13:00:45.394644421Z","level":"INFO","msg":"writer: Close: closed","stream_id":{"value":"a3z1jk8c"}}
+{"time":"2024-10-07T13:00:45.39479126Z","level":"INFO","msg":"stream: closed","id":"a3z1jk8c"}

wandb/run-20241007_125615-a3z1jk8c/logs/debug.log

@@ -26,3 +26,4 @@ config: {}
 2024-10-07 12:56:15,777 INFO    MainThread:20958 [wandb_run.py:_config_callback():1394] config_cb None None {'vocab_size': 51866, 'num_mel_bins': 128, 'd_model': 1280, 'encoder_layers': 32, 'encoder_attention_heads': 20, 'decoder_layers': 32, 'decoder_attention_heads': 20, 'decoder_ffn_dim': 5120, 'encoder_ffn_dim': 5120, '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': 32, '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': 'float16', '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': 50256, '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-large-v3', 'transformers_version': '4.46.0.dev0', 'model_type': 'whisper', 'forced_decoder_ids': None, 'output_dir': './', 'overwrite_output_dir': True, 'do_train': False, 'do_eval': True, 'do_predict': False, 'eval_strategy': 'steps', 'prediction_loss_only': False, 'per_device_train_batch_size': 16, 'per_device_eval_batch_size': 8, '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': 4.375e-06, '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': 10000, '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/Oct07_11-46-39_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': 500, 'dataloader_num_workers': 0, 'dataloader_prefetch_factor': None, 'past_index': -1, 'run_name': 'whisper-large-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': False, '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, 'sortish_sampler': False, 'predict_with_generate': True, 'generation_max_length': 228, 'generation_num_beams': None, 'generation_config': None}
 2024-10-07 12:56:15,780 INFO    MainThread:20958 [wandb_config.py:__setitem__():154] config set model/num_parameters = 1543490560 - <bound method Run._config_callback of <wandb.sdk.wandb_run.Run object at 0x748ced2ceae0>>
 2024-10-07 12:56:15,780 INFO    MainThread:20958 [wandb_run.py:_config_callback():1394] config_cb model/num_parameters 1543490560 None
+2024-10-07 13:00:42,168 WARNING MsgRouterThr:20958 [router.py:message_loop():77] message_loop has been closed

wandb/run-20241007_131849-0rbzerob/files/output.log

@@ -0,0 +1,34 @@
+Reading metadata...: 75336it [00:03, 21675.20it/s]                                                                                                 | 0/10000 [00:00<?, ?it/s]
+Reading metadata...: 13630it [00:00, 18317.27it/s]
+[INFO|trainer_utils.py:830] 2024-10-07 13:19:00,935 >> The following columns in the training set don't have a corresponding argument in `WhisperForConditionalGeneration.forward` and have been ignored: input_length. If input_length are not expected by `WhisperForConditionalGeneration.forward`,  you can safely ignore this message.
+[WARNING|trainer.py:2506] 2024-10-07 13:23:37,058 >> There seems not to be a single sample in your epoch_iterator, stopping training at step 9000! This is expected if you're using an IterableDataset and set num_steps (10000) higher than the number of available samples.
+[INFO|trainer.py:2532] 2024-10-07 13:23:37,059 >>
+
+Training completed. Do not forget to share your model on huggingface.co/models =)
+
+
+[INFO|trainer.py:2770] 2024-10-07 13:23:37,059 >> Loading best model from ./checkpoint-9000 (score: 7.215361500971087).
+[WARNING|trainer.py:2892] 2024-10-07 13:23:38,416 >> There were missing keys in the checkpoint model loaded: ['proj_out.weight'].
+  0%|                                                                                                                                              | 0/10000 [04:48<?, ?it/s]
+{'train_runtime': 289.8068, 'train_samples_per_second': 552.092, 'train_steps_per_second': 34.506, 'train_loss': 0.0, 'epoch': 5.05}
+[INFO|trainer.py:3738] 2024-10-07 13:23:38,418 >> Saving model checkpoint to ./
+/home/tknika/whisper-large-eu/.venv/lib/python3.12/site-packages/transformers/modeling_utils.py:2774: UserWarning: Moving the following attributes in the config to the generation config: {'max_length': 448, 'begin_suppress_tokens': [220, 50257]}. You are seeing this warning because you've set generation parameters in the model config, as opposed to in the generation config.
+  warnings.warn(
+[INFO|configuration_utils.py:410] 2024-10-07 13:23:38,421 >> Configuration saved in ./config.json
+[INFO|configuration_utils.py:868] 2024-10-07 13:23:38,422 >> Configuration saved in ./generation_config.json
+[INFO|modeling_utils.py:3000] 2024-10-07 13:23:45,845 >> The model is bigger than the maximum size per checkpoint (5GB) and is going to be split in 2 checkpoint shards. You can find where each parameters has been saved in the index located at ./model.safetensors.index.json.
+[INFO|feature_extraction_utils.py:435] 2024-10-07 13:23:45,849 >> Feature extractor saved in ./preprocessor_config.json
+[INFO|trainer.py:3738] 2024-10-07 13:23:45,850 >> Saving model checkpoint to ./
+[INFO|configuration_utils.py:410] 2024-10-07 13:23:45,852 >> Configuration saved in ./config.json
+[INFO|configuration_utils.py:868] 2024-10-07 13:23:45,852 >> Configuration saved in ./generation_config.json
+[INFO|modeling_utils.py:3000] 2024-10-07 13:23:53,998 >> The model is bigger than the maximum size per checkpoint (5GB) and is going to be split in 2 checkpoint shards. You can find where each parameters has been saved in the index located at ./model.safetensors.index.json.
+[INFO|feature_extraction_utils.py:435] 2024-10-07 13:23:53,999 >> Feature extractor saved in ./preprocessor_config.json
+/home/tknika/whisper-large-eu/.venv/lib/python3.12/site-packages/huggingface_hub/hf_api.py:3889: UserWarning: It seems that you are about to commit a data file (.venv/lib/python3.12/site-packages/pyarrow/tests/data/parquet/v0.7.1.all-named-index.parquet) to a model repository. You are sure this is intended? If you are trying to upload a dataset, please set `repo_type='dataset'` or `--repo-type=dataset` in a CLI.
+  warnings.warn(
+/home/tknika/whisper-large-eu/.venv/lib/python3.12/site-packages/huggingface_hub/hf_api.py:3889: UserWarning: It seems that you are about to commit a data file (.venv/lib/python3.12/site-packages/pyarrow/tests/data/parquet/v0.7.1.column-metadata-handling.parquet) to a model repository. You are sure this is intended? If you are trying to upload a dataset, please set `repo_type='dataset'` or `--repo-type=dataset` in a CLI.
+  warnings.warn(
+/home/tknika/whisper-large-eu/.venv/lib/python3.12/site-packages/huggingface_hub/hf_api.py:3889: UserWarning: It seems that you are about to commit a data file (.venv/lib/python3.12/site-packages/pyarrow/tests/data/parquet/v0.7.1.parquet) to a model repository. You are sure this is intended? If you are trying to upload a dataset, please set `repo_type='dataset'` or `--repo-type=dataset` in a CLI.
+  warnings.warn(
+/home/tknika/whisper-large-eu/.venv/lib/python3.12/site-packages/huggingface_hub/hf_api.py:3889: UserWarning: It seems that you are about to commit a data file (.venv/lib/python3.12/site-packages/pyarrow/tests/data/parquet/v0.7.1.some-named-index.parquet) to a model repository. You are sure this is intended? If you are trying to upload a dataset, please set `repo_type='dataset'` or `--repo-type=dataset` in a CLI.
+  warnings.warn(
+training_args.bin: 100%|██████████████���█████████████████████████████████████████████████████████████████████████████████████████████████| 5.37k/5.37k [00:00<00:00, 15.6kB/s]

wandb/run-20241007_131849-0rbzerob/files/requirements.txt

@@ -0,0 +1,94 @@
+Markdown==3.7
+requests==2.32.3
+RapidFuzz==3.10.0
+yarl==1.13.1
+pyarrow==17.0.0
+docker-pycreds==0.4.0
+nvidia-cufft-cu12==11.0.2.54
+PyYAML==6.0.2
+packaging==24.1
+librosa==0.10.2.post1
+soxr==0.5.0.post1
+multiprocess==0.70.16
+nvidia-nvjitlink-cu12==12.6.77
+safetensors==0.4.5
+joblib==1.4.2
+pip==24.0
+wandb==0.18.3
+networkx==3.3
+numba==0.60.0
+scipy==1.14.1
+MarkupSafe==2.1.5
+GitPython==3.1.43
+aiohttp==3.10.9
+msgpack==1.1.0
+mpmath==1.3.0
+tzdata==2024.2
+nvidia-cudnn-cu12==9.1.0.70
+scikit-learn==1.5.2
+pytz==2024.2
+dill==0.3.8
+nvidia-cusparse-cu12==12.1.0.106
+soundfile==0.12.1
+aiosignal==1.3.1
+gitdb==4.0.11
+Jinja2==3.1.4
+jiwer==3.0.4
+decorator==5.1.1
+nvidia-cusolver-cu12==11.4.5.107
+protobuf==5.28.2
+idna==3.10
+tqdm==4.66.5
+pandas==2.2.3
+python-dateutil==2.9.0.post0
+Werkzeug==3.0.4
+click==8.1.7
+regex==2024.9.11
+typing_extensions==4.12.2
+nvidia-cublas-cu12==12.1.3.1
+transformers==4.46.0.dev0
+nvidia-nccl-cu12==2.20.5
+nvidia-cuda-cupti-cu12==12.1.105
+triton==3.0.0
+pooch==1.8.2
+smmap==5.0.1
+grpcio==1.66.2
+setuptools==75.1.0
+setproctitle==1.3.3
+accelerate==0.34.2
+nvidia-cuda-nvrtc-cu12==12.1.105
+tensorboard==2.18.0
+absl-py==2.1.0
+nvidia-nvtx-cu12==12.1.105
+fsspec==2024.6.1
+pycparser==2.22
+lazy_loader==0.4
+tensorboard-data-server==0.7.2
+urllib3==2.2.3
+threadpoolctl==3.5.0
+llvmlite==0.43.0
+sympy==1.13.3
+audioread==3.0.1
+tokenizers==0.20.0
+more-itertools==10.5.0
+cffi==1.17.1
+evaluate==0.4.3
+nvidia-curand-cu12==10.3.2.106
+psutil==6.0.0
+filelock==3.16.1
+attrs==24.2.0
+six==1.16.0
+frozenlist==1.4.1
+sentry-sdk==2.15.0
+nvidia-cuda-runtime-cu12==12.1.105
+xxhash==3.5.0
+platformdirs==4.3.6
+multidict==6.1.0
+aiohappyeyeballs==2.4.3
+torch==2.4.1
+huggingface-hub==0.25.1
+numpy==2.0.2
+datasets==3.0.2.dev0
+torchaudio==2.4.1
+charset-normalizer==3.3.2
+certifi==2024.8.30

wandb/run-20241007_131849-0rbzerob/files/wandb-metadata.json

@@ -0,0 +1,87 @@
+{
+  "os":  "Linux-6.8.0-45-generic-x86_64-with-glibc2.39",
+  "python":  "3.12.3",
+  "startedAt":  "2024-10-07T13:18:49.425338Z",
+  "args":  [
+    "--model_name_or_path=openai/whisper-large-v3",
+    "--dataset_name=mozilla-foundation/common_voice_17_0",
+    "--dataset_config_name=eu",
+    "--language=basque",
+    "--train_split_name=train+validation",
+    "--eval_split_name=test",
+    "--model_index_name=Whisper Large Basque",
+    "--max_steps=10000",
+    "--output_dir=./",
+    "--per_device_train_batch_size=16",
+    "--per_device_eval_batch_size=8",
+    "--gradient_accumulation_steps=1",
+    "--logging_steps=25",
+    "--learning_rate=4.375e-6",
+    "--warmup_steps=500",
+    "--evaluation_strategy=steps",
+    "--eval_steps=500",
+    "--save_strategy=steps",
+    "--save_steps=1000",
+    "--generation_max_length=228",
+    "--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",
+    "--resume_from_checkpoint=checkpoint-9000",
+    "--do_train",
+    "--do_eval",
+    "--predict_with_generate",
+    "--do_normalize_eval",
+    "--streaming",
+    "--push_to_hub",
+    "--report_to",
+    "wandb",
+    "--run_name",
+    "whisper-large-eu"
+  ],
+  "program":  "/home/tknika/whisper-large-eu/run_speech_recognition_seq2seq_streaming.py",
+  "codePath":  "run_speech_recognition_seq2seq_streaming.py",
+  "git":  {
+    "remote":  "https://huggingface.co/xezpeleta/whisper-large-eu",
+    "commit":  "45227421df6af8836af459c374361e7303a68aea"
+  },
+  "email":  "[email protected]",
+  "root":  "/home/tknika/whisper-large-eu",
+  "host":  "tknika",
+  "username":  "tknika",
+  "executable":  "/home/tknika/whisper-large-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":  "314615791616",
+      "used":  "265684557824"
+    }
+  },
+  "memory":  {
+    "total":  "33654026240"
+  },
+  "cpu":  {
+    "count":  8,
+    "countLogical":  8
+  },
+  "gpu_nvidia":  [
+    {
+      "name":  "NVIDIA L40-48Q",
+      "memoryTotal":  "51539607552",
+      "cudaCores":  18176,
+      "architecture":  "Ada"
+    }
+  ],
+  "cudaVersion":  "12.4"
+}

wandb/run-20241007_131849-0rbzerob/logs/debug-core.log

@@ -0,0 +1,7 @@
+{"time":"2024-10-07T13:18:48.681266604Z","level":"INFO","msg":"started logging, with flags","port-filename":"/tmp/tmptcl7jc73/port-21209.txt","pid":21209,"debug":false,"disable-analytics":false}
+{"time":"2024-10-07T13:18:48.681313674Z","level":"INFO","msg":"FeatureState","shutdownOnParentExitEnabled":false}
+{"time":"2024-10-07T13:18:48.697970073Z","level":"INFO","msg":"Will exit if parent process dies.","ppid":21209}
+{"time":"2024-10-07T13:18:48.697953933Z","level":"INFO","msg":"server is running","addr":{"IP":"127.0.0.1","Port":33313,"Zone":""}}
+{"time":"2024-10-07T13:18:48.868568838Z","level":"INFO","msg":"connection: ManageConnectionData: new connection created","id":"127.0.0.1:55398"}
+{"time":"2024-10-07T13:18:49.427121931Z","level":"INFO","msg":"handleInformInit: received","streamId":"0rbzerob","id":"127.0.0.1:55398"}
+{"time":"2024-10-07T13:18:49.436122661Z","level":"INFO","msg":"handleInformInit: stream started","streamId":"0rbzerob","id":"127.0.0.1:55398"}

wandb/run-20241007_131849-0rbzerob/logs/debug-internal.log

@@ -0,0 +1,10 @@
+{"time":"2024-10-07T13:18:49.427383059Z","level":"INFO","msg":"using version","core version":"0.18.3"}
+{"time":"2024-10-07T13:18:49.427406249Z","level":"INFO","msg":"created symlink","path":"/home/tknika/whisper-large-eu/wandb/run-20241007_131849-0rbzerob/logs/debug-core.log"}
+{"time":"2024-10-07T13:18:49.429699121Z","level":"ERROR","msg":"dialing: google: could not find default credentials. See https://cloud.google.com/docs/authentication/external/set-up-adc for more information"}
+{"time":"2024-10-07T13:18:49.436074781Z","level":"INFO","msg":"created new stream","id":"0rbzerob"}
+{"time":"2024-10-07T13:18:49.436114921Z","level":"INFO","msg":"stream: started","id":"0rbzerob"}
+{"time":"2024-10-07T13:18:49.43615474Z","level":"INFO","msg":"writer: Do: started","stream_id":{"value":"0rbzerob"}}
+{"time":"2024-10-07T13:18:49.43615079Z","level":"INFO","msg":"sender: started","stream_id":{"value":"0rbzerob"}}
+{"time":"2024-10-07T13:18:49.43618894Z","level":"INFO","msg":"handler: started","stream_id":{"value":"0rbzerob"}}
+{"time":"2024-10-07T13:18:49.857246919Z","level":"INFO","msg":"wandb-core","!BADKEY":null}
+{"time":"2024-10-07T13:18:49.859623571Z","level":"INFO","msg":"Starting system monitor"}

wandb/run-20241007_131849-0rbzerob/logs/debug.log

@@ -0,0 +1,28 @@
+2024-10-07 13:18:49,422 INFO    MainThread:21209 [wandb_setup.py:_flush():79] Current SDK version is 0.18.3
+2024-10-07 13:18:49,422 INFO    MainThread:21209 [wandb_setup.py:_flush():79] Configure stats pid to 21209
+2024-10-07 13:18:49,422 INFO    MainThread:21209 [wandb_setup.py:_flush():79] Loading settings from /home/tknika/.config/wandb/settings
+2024-10-07 13:18:49,422 INFO    MainThread:21209 [wandb_setup.py:_flush():79] Loading settings from /home/tknika/whisper-large-eu/wandb/settings
+2024-10-07 13:18:49,422 INFO    MainThread:21209 [wandb_setup.py:_flush():79] Loading settings from environment variables: {'project': 'whisper-medium-eu'}
+2024-10-07 13:18:49,422 INFO    MainThread:21209 [wandb_setup.py:_flush():79] Applying setup settings: {'mode': None, '_disable_service': None}
+2024-10-07 13:18:49,422 INFO    MainThread:21209 [wandb_setup.py:_flush():79] Inferring run settings from compute environment: {'program_relpath': 'run_speech_recognition_seq2seq_streaming.py', 'program_abspath': '/home/tknika/whisper-large-eu/run_speech_recognition_seq2seq_streaming.py', 'program': '/home/tknika/whisper-large-eu/run_speech_recognition_seq2seq_streaming.py'}
+2024-10-07 13:18:49,422 INFO    MainThread:21209 [wandb_setup.py:_flush():79] Applying login settings: {}
+2024-10-07 13:18:49,422 INFO    MainThread:21209 [wandb_init.py:_log_setup():532] Logging user logs to /home/tknika/whisper-large-eu/wandb/run-20241007_131849-0rbzerob/logs/debug.log
+2024-10-07 13:18:49,422 INFO    MainThread:21209 [wandb_init.py:_log_setup():533] Logging internal logs to /home/tknika/whisper-large-eu/wandb/run-20241007_131849-0rbzerob/logs/debug-internal.log
+2024-10-07 13:18:49,422 INFO    MainThread:21209 [wandb_init.py:init():617] calling init triggers
+2024-10-07 13:18:49,422 INFO    MainThread:21209 [wandb_init.py:init():624] wandb.init called with sweep_config: {}
+config: {}
+2024-10-07 13:18:49,422 INFO    MainThread:21209 [wandb_init.py:init():667] starting backend
+2024-10-07 13:18:49,422 INFO    MainThread:21209 [wandb_init.py:init():671] sending inform_init request
+2024-10-07 13:18:49,424 INFO    MainThread:21209 [backend.py:_multiprocessing_setup():104] multiprocessing start_methods=fork,spawn,forkserver, using: spawn
+2024-10-07 13:18:49,425 INFO    MainThread:21209 [wandb_init.py:init():684] backend started and connected
+2024-10-07 13:18:49,428 INFO    MainThread:21209 [wandb_init.py:init():779] updated telemetry
+2024-10-07 13:18:49,436 INFO    MainThread:21209 [wandb_init.py:init():812] communicating run to backend with 90.0 second timeout
+2024-10-07 13:18:49,852 INFO    MainThread:21209 [wandb_init.py:init():863] starting run threads in backend
+2024-10-07 13:18:49,971 INFO    MainThread:21209 [wandb_run.py:_console_start():2465] atexit reg
+2024-10-07 13:18:49,972 INFO    MainThread:21209 [wandb_run.py:_redirect():2313] redirect: wrap_raw
+2024-10-07 13:18:49,972 INFO    MainThread:21209 [wandb_run.py:_redirect():2378] Wrapping output streams.
+2024-10-07 13:18:49,972 INFO    MainThread:21209 [wandb_run.py:_redirect():2403] Redirects installed.
+2024-10-07 13:18:49,973 INFO    MainThread:21209 [wandb_init.py:init():907] run started, returning control to user process
+2024-10-07 13:18:49,974 INFO    MainThread:21209 [wandb_run.py:_config_callback():1394] config_cb None None {'vocab_size': 51866, 'num_mel_bins': 128, 'd_model': 1280, 'encoder_layers': 32, 'encoder_attention_heads': 20, 'decoder_layers': 32, 'decoder_attention_heads': 20, 'decoder_ffn_dim': 5120, 'encoder_ffn_dim': 5120, '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': 32, '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': 'float16', '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': 50256, '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-large-v3', 'transformers_version': '4.46.0.dev0', 'model_type': 'whisper', 'forced_decoder_ids': None, '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': 16, 'per_device_eval_batch_size': 8, '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': 4.375e-06, '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': 10000, '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/Oct07_13-18-21_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': 500, 'dataloader_num_workers': 0, 'dataloader_prefetch_factor': None, 'past_index': -1, 'run_name': 'whisper-large-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': 'checkpoint-9000', 'hub_model_id': None, 'hub_strategy': 'every_save', 'hub_token': '<HUB_TOKEN>', 'hub_private_repo': False, '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, 'sortish_sampler': False, 'predict_with_generate': True, 'generation_max_length': 228, 'generation_num_beams': None, 'generation_config': None}
+2024-10-07 13:18:49,978 INFO    MainThread:21209 [wandb_config.py:__setitem__():154] config set model/num_parameters = 1543490560 - <bound method Run._config_callback of <wandb.sdk.wandb_run.Run object at 0x7a0b4e8a1df0>>
+2024-10-07 13:18:49,978 INFO    MainThread:21209 [wandb_run.py:_config_callback():1394] config_cb model/num_parameters 1543490560 None