| { |
| "best_global_step": 50, |
| "best_metric": 94.90034412693939, |
| "best_model_checkpoint": "./artifacts/whisper-small-bn/checkpoint-50", |
| "epoch": 0.05235602094240838, |
| "eval_steps": 5, |
| "global_step": 50, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "epoch": 0.005235602094240838, |
| "grad_norm": 15.15451717376709, |
| "learning_rate": 1.0000000000000002e-06, |
| "loss": 1.8518, |
| "step": 5 |
| }, |
| { |
| "epoch": 0.005235602094240838, |
| "eval_loss": 1.8115043640136719, |
| "eval_model_preparation_time": 0.0029, |
| "eval_runtime": 31064.0047, |
| "eval_samples_per_second": 0.3, |
| "eval_steps_per_second": 0.038, |
| "eval_wer": 147.3591487262605, |
| "step": 5 |
| }, |
| { |
| "epoch": 0.010471204188481676, |
| "grad_norm": 10.38752555847168, |
| "learning_rate": 2.0000000000000003e-06, |
| "loss": 1.7685, |
| "step": 10 |
| }, |
| { |
| "epoch": 0.010471204188481676, |
| "eval_loss": 1.7016992568969727, |
| "eval_model_preparation_time": 0.0029, |
| "eval_runtime": 30036.062, |
| "eval_samples_per_second": 0.311, |
| "eval_steps_per_second": 0.039, |
| "eval_wer": 134.91772664810966, |
| "step": 10 |
| }, |
| { |
| "epoch": 0.015706806282722512, |
| "grad_norm": 8.120345115661621, |
| "learning_rate": 3e-06, |
| "loss": 1.6483, |
| "step": 15 |
| }, |
| { |
| "epoch": 0.015706806282722512, |
| "eval_loss": 1.5776597261428833, |
| "eval_model_preparation_time": 0.0029, |
| "eval_runtime": 30603.6517, |
| "eval_samples_per_second": 0.305, |
| "eval_steps_per_second": 0.038, |
| "eval_wer": 133.56118530002465, |
| "step": 15 |
| }, |
| { |
| "epoch": 0.020942408376963352, |
| "grad_norm": 7.45015811920166, |
| "learning_rate": 4.000000000000001e-06, |
| "loss": 1.532, |
| "step": 20 |
| }, |
| { |
| "epoch": 0.020942408376963352, |
| "eval_loss": 1.4560885429382324, |
| "eval_model_preparation_time": 0.0029, |
| "eval_runtime": 31279.9657, |
| "eval_samples_per_second": 0.298, |
| "eval_steps_per_second": 0.037, |
| "eval_wer": 122.22613720446778, |
| "step": 20 |
| }, |
| { |
| "epoch": 0.02617801047120419, |
| "grad_norm": 7.290828704833984, |
| "learning_rate": 5e-06, |
| "loss": 1.4186, |
| "step": 25 |
| }, |
| { |
| "epoch": 0.02617801047120419, |
| "eval_loss": 1.3617514371871948, |
| "eval_model_preparation_time": 0.0029, |
| "eval_runtime": 31425.8283, |
| "eval_samples_per_second": 0.297, |
| "eval_steps_per_second": 0.037, |
| "eval_wer": 104.2305298145473, |
| "step": 25 |
| }, |
| { |
| "epoch": 0.031413612565445025, |
| "grad_norm": 7.941519737243652, |
| "learning_rate": 6e-06, |
| "loss": 1.3173, |
| "step": 30 |
| }, |
| { |
| "epoch": 0.031413612565445025, |
| "eval_loss": 1.2517695426940918, |
| "eval_model_preparation_time": 0.0029, |
| "eval_runtime": 31322.7629, |
| "eval_samples_per_second": 0.298, |
| "eval_steps_per_second": 0.037, |
| "eval_wer": 121.96070140821911, |
| "step": 30 |
| }, |
| { |
| "epoch": 0.03664921465968586, |
| "grad_norm": 7.520599365234375, |
| "learning_rate": 7e-06, |
| "loss": 1.1809, |
| "step": 35 |
| }, |
| { |
| "epoch": 0.03664921465968586, |
| "eval_loss": 1.0982742309570312, |
| "eval_model_preparation_time": 0.0029, |
| "eval_runtime": 31224.2913, |
| "eval_samples_per_second": 0.299, |
| "eval_steps_per_second": 0.037, |
| "eval_wer": 115.53151756456785, |
| "step": 35 |
| }, |
| { |
| "epoch": 0.041884816753926704, |
| "grad_norm": 6.726326942443848, |
| "learning_rate": 8.000000000000001e-06, |
| "loss": 1.0262, |
| "step": 40 |
| }, |
| { |
| "epoch": 0.041884816753926704, |
| "eval_loss": 0.9411609768867493, |
| "eval_model_preparation_time": 0.0029, |
| "eval_runtime": 31344.2102, |
| "eval_samples_per_second": 0.298, |
| "eval_steps_per_second": 0.037, |
| "eval_wer": 109.31726624619758, |
| "step": 40 |
| }, |
| { |
| "epoch": 0.04712041884816754, |
| "grad_norm": 7.976568698883057, |
| "learning_rate": 9e-06, |
| "loss": 0.8603, |
| "step": 45 |
| }, |
| { |
| "epoch": 0.04712041884816754, |
| "eval_loss": 0.7995896339416504, |
| "eval_model_preparation_time": 0.0029, |
| "eval_runtime": 29970.6388, |
| "eval_samples_per_second": 0.311, |
| "eval_steps_per_second": 0.039, |
| "eval_wer": 97.59463490833069, |
| "step": 45 |
| }, |
| { |
| "epoch": 0.05235602094240838, |
| "grad_norm": 7.062014579772949, |
| "learning_rate": 0.0, |
| "loss": 0.7294, |
| "step": 50 |
| }, |
| { |
| "epoch": 0.05235602094240838, |
| "eval_loss": 0.6859976053237915, |
| "eval_model_preparation_time": 0.0029, |
| "eval_runtime": 29101.6197, |
| "eval_samples_per_second": 0.32, |
| "eval_steps_per_second": 0.04, |
| "eval_wer": 94.90034412693939, |
| "step": 50 |
| } |
| ], |
| "logging_steps": 5, |
| "max_steps": 50, |
| "num_input_tokens_seen": 0, |
| "num_train_epochs": 1, |
| "save_steps": 5, |
| "stateful_callbacks": { |
| "TrainerControl": { |
| "args": { |
| "should_epoch_stop": false, |
| "should_evaluate": false, |
| "should_log": false, |
| "should_save": true, |
| "should_training_stop": true |
| }, |
| "attributes": {} |
| } |
| }, |
| "total_flos": 4.61736640512e+17, |
| "train_batch_size": 32, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
