| { |
| "best_global_step": 1084, |
| "best_metric": 79.94547700093382, |
| "best_model_checkpoint": "./kaggle/working/whisper-tiny-dar/checkpoint-1084", |
| "epoch": 4.0, |
| "eval_steps": 500, |
| "global_step": 1084, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "epoch": 0.18450184501845018, |
| "grad_norm": 8.362913131713867, |
| "learning_rate": 4.800000000000001e-06, |
| "loss": 1.939, |
| "step": 50 |
| }, |
| { |
| "epoch": 0.36900369003690037, |
| "grad_norm": 6.892248630523682, |
| "learning_rate": 9.800000000000001e-06, |
| "loss": 1.5598, |
| "step": 100 |
| }, |
| { |
| "epoch": 0.5535055350553506, |
| "grad_norm": 6.687413215637207, |
| "learning_rate": 9.617529880478089e-06, |
| "loss": 1.3182, |
| "step": 150 |
| }, |
| { |
| "epoch": 0.7380073800738007, |
| "grad_norm": 6.872180938720703, |
| "learning_rate": 9.219123505976096e-06, |
| "loss": 1.2207, |
| "step": 200 |
| }, |
| { |
| "epoch": 0.922509225092251, |
| "grad_norm": 6.462919235229492, |
| "learning_rate": 8.820717131474104e-06, |
| "loss": 1.1472, |
| "step": 250 |
| }, |
| { |
| "epoch": 1.0, |
| "eval_loss": 1.1004470586776733, |
| "eval_runtime": 467.7367, |
| "eval_samples_per_second": 2.277, |
| "eval_steps_per_second": 0.286, |
| "eval_wer": 85.98367322348405, |
| "step": 271 |
| }, |
| { |
| "epoch": 1.1070110701107012, |
| "grad_norm": 7.008378505706787, |
| "learning_rate": 8.422310756972111e-06, |
| "loss": 1.068, |
| "step": 300 |
| }, |
| { |
| "epoch": 1.2915129151291513, |
| "grad_norm": 7.158851146697998, |
| "learning_rate": 8.02390438247012e-06, |
| "loss": 1.029, |
| "step": 350 |
| }, |
| { |
| "epoch": 1.4760147601476015, |
| "grad_norm": 6.748999118804932, |
| "learning_rate": 7.625498007968128e-06, |
| "loss": 0.993, |
| "step": 400 |
| }, |
| { |
| "epoch": 1.6605166051660518, |
| "grad_norm": 6.34915018081665, |
| "learning_rate": 7.227091633466136e-06, |
| "loss": 0.9657, |
| "step": 450 |
| }, |
| { |
| "epoch": 1.8450184501845017, |
| "grad_norm": 6.932847023010254, |
| "learning_rate": 6.828685258964144e-06, |
| "loss": 0.938, |
| "step": 500 |
| }, |
| { |
| "epoch": 2.0, |
| "eval_loss": 0.9617473483085632, |
| "eval_runtime": 467.5279, |
| "eval_samples_per_second": 2.278, |
| "eval_steps_per_second": 0.287, |
| "eval_wer": 82.57523270175015, |
| "step": 542 |
| }, |
| { |
| "epoch": 2.029520295202952, |
| "grad_norm": 5.872282028198242, |
| "learning_rate": 6.430278884462152e-06, |
| "loss": 0.9207, |
| "step": 550 |
| }, |
| { |
| "epoch": 2.2140221402214024, |
| "grad_norm": 6.210878372192383, |
| "learning_rate": 6.031872509960159e-06, |
| "loss": 0.8734, |
| "step": 600 |
| }, |
| { |
| "epoch": 2.3985239852398523, |
| "grad_norm": 6.416970252990723, |
| "learning_rate": 5.633466135458168e-06, |
| "loss": 0.8515, |
| "step": 650 |
| }, |
| { |
| "epoch": 2.5830258302583027, |
| "grad_norm": 5.866384029388428, |
| "learning_rate": 5.235059760956176e-06, |
| "loss": 0.852, |
| "step": 700 |
| }, |
| { |
| "epoch": 2.767527675276753, |
| "grad_norm": 6.105711936950684, |
| "learning_rate": 4.836653386454184e-06, |
| "loss": 0.859, |
| "step": 750 |
| }, |
| { |
| "epoch": 2.952029520295203, |
| "grad_norm": 5.599572658538818, |
| "learning_rate": 4.438247011952192e-06, |
| "loss": 0.8405, |
| "step": 800 |
| }, |
| { |
| "epoch": 3.0, |
| "eval_loss": 0.9079960584640503, |
| "eval_runtime": 467.7935, |
| "eval_samples_per_second": 2.277, |
| "eval_steps_per_second": 0.286, |
| "eval_wer": 81.04798626381901, |
| "step": 813 |
| }, |
| { |
| "epoch": 3.1365313653136533, |
| "grad_norm": 5.818225383758545, |
| "learning_rate": 4.0398406374501995e-06, |
| "loss": 0.8054, |
| "step": 850 |
| }, |
| { |
| "epoch": 3.321033210332103, |
| "grad_norm": 5.295748233795166, |
| "learning_rate": 3.6414342629482076e-06, |
| "loss": 0.7887, |
| "step": 900 |
| }, |
| { |
| "epoch": 3.5055350553505535, |
| "grad_norm": 6.146871089935303, |
| "learning_rate": 3.2430278884462153e-06, |
| "loss": 0.795, |
| "step": 950 |
| }, |
| { |
| "epoch": 3.6900369003690034, |
| "grad_norm": 5.9709649085998535, |
| "learning_rate": 2.8446215139442235e-06, |
| "loss": 0.7919, |
| "step": 1000 |
| }, |
| { |
| "epoch": 3.874538745387454, |
| "grad_norm": 6.177392482757568, |
| "learning_rate": 2.446215139442231e-06, |
| "loss": 0.7801, |
| "step": 1050 |
| }, |
| { |
| "epoch": 4.0, |
| "eval_loss": 0.88407963514328, |
| "eval_runtime": 468.1847, |
| "eval_samples_per_second": 2.275, |
| "eval_steps_per_second": 0.286, |
| "eval_wer": 79.94547700093382, |
| "step": 1084 |
| } |
| ], |
| "logging_steps": 50, |
| "max_steps": 1355, |
| "num_input_tokens_seen": 0, |
| "num_train_epochs": 5, |
| "save_steps": 500, |
| "stateful_callbacks": { |
| "TrainerControl": { |
| "args": { |
| "should_epoch_stop": false, |
| "should_evaluate": false, |
| "should_log": false, |
| "should_save": true, |
| "should_training_stop": false |
| }, |
| "attributes": {} |
| } |
| }, |
| "total_flos": 4.2698994352128e+17, |
| "train_batch_size": 8, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
