| { | |
| "best_metric": 8.153910529519171, | |
| "best_model_checkpoint": "../Whisper-squeezeformer-v6\\checkpoint-30000", | |
| "epoch": 12.0, | |
| "eval_steps": 2500, | |
| "global_step": 30000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 12.83377456665039, | |
| "learning_rate": 9.972e-06, | |
| "loss": 4.7944, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": 3.786200523376465, | |
| "eval_runtime": 867.735, | |
| "eval_samples_per_second": 3.019, | |
| "eval_steps_per_second": 0.378, | |
| "eval_wer": 136.5585057821059, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 14.687960624694824, | |
| "learning_rate": 8.575428571428573e-06, | |
| "loss": 3.8012, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 2.992347478866577, | |
| "eval_runtime": 676.1432, | |
| "eval_samples_per_second": 3.875, | |
| "eval_steps_per_second": 0.485, | |
| "eval_wer": 107.02982349360926, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 8.66879653930664, | |
| "learning_rate": 7.146857142857143e-06, | |
| "loss": 0.7205, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_loss": 0.2924005389213562, | |
| "eval_runtime": 675.8874, | |
| "eval_samples_per_second": 3.876, | |
| "eval_steps_per_second": 0.485, | |
| "eval_wer": 17.247413268411442, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 5.629953861236572, | |
| "learning_rate": 5.718285714285715e-06, | |
| "loss": 0.1473, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_loss": 0.23985154926776886, | |
| "eval_runtime": 594.3096, | |
| "eval_samples_per_second": 4.408, | |
| "eval_steps_per_second": 0.552, | |
| "eval_wer": 13.827601947656726, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "grad_norm": 5.151638031005859, | |
| "learning_rate": 4.289714285714286e-06, | |
| "loss": 0.0847, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_loss": 0.22282332181930542, | |
| "eval_runtime": 585.3758, | |
| "eval_samples_per_second": 4.476, | |
| "eval_steps_per_second": 0.56, | |
| "eval_wer": 11.767726719415702, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "grad_norm": 3.3535683155059814, | |
| "learning_rate": 2.861142857142857e-06, | |
| "loss": 0.0505, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_loss": 0.2200464904308319, | |
| "eval_runtime": 645.3171, | |
| "eval_samples_per_second": 4.06, | |
| "eval_steps_per_second": 0.508, | |
| "eval_wer": 12.20709068776628, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "grad_norm": 6.204869747161865, | |
| "learning_rate": 4.548727272727273e-06, | |
| "loss": 0.1802, | |
| "step": 17500 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_loss": 0.17816144227981567, | |
| "eval_runtime": 603.0644, | |
| "eval_samples_per_second": 4.344, | |
| "eval_steps_per_second": 0.544, | |
| "eval_wer": 9.755401704199635, | |
| "step": 17500 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "grad_norm": 4.365288734436035, | |
| "learning_rate": 3.6400000000000003e-06, | |
| "loss": 0.107, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_loss": 0.1680830866098404, | |
| "eval_runtime": 599.7782, | |
| "eval_samples_per_second": 4.368, | |
| "eval_steps_per_second": 0.547, | |
| "eval_wer": 9.196211199026171, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "grad_norm": 3.9741170406341553, | |
| "learning_rate": 2.7309090909090914e-06, | |
| "loss": 0.0693, | |
| "step": 22500 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_loss": 0.1669154316186905, | |
| "eval_runtime": 646.0215, | |
| "eval_samples_per_second": 4.056, | |
| "eval_steps_per_second": 0.508, | |
| "eval_wer": 8.220480827754107, | |
| "step": 22500 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "grad_norm": 3.504108428955078, | |
| "learning_rate": 1.8221818181818182e-06, | |
| "loss": 0.0462, | |
| "step": 25000 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_loss": 0.16997282207012177, | |
| "eval_runtime": 581.1384, | |
| "eval_samples_per_second": 4.508, | |
| "eval_steps_per_second": 0.564, | |
| "eval_wer": 8.9603621424224, | |
| "step": 25000 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "grad_norm": 3.334559679031372, | |
| "learning_rate": 9.130909090909091e-07, | |
| "loss": 0.032, | |
| "step": 27500 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_loss": 0.17454640567302704, | |
| "eval_runtime": 579.2297, | |
| "eval_samples_per_second": 4.523, | |
| "eval_steps_per_second": 0.566, | |
| "eval_wer": 8.595176506390748, | |
| "step": 27500 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "grad_norm": 2.283846616744995, | |
| "learning_rate": 4.363636363636364e-09, | |
| "loss": 0.024, | |
| "step": 30000 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_loss": 0.17751750349998474, | |
| "eval_runtime": 574.699, | |
| "eval_samples_per_second": 4.559, | |
| "eval_steps_per_second": 0.571, | |
| "eval_wer": 8.153910529519171, | |
| "step": 30000 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "step": 30000, | |
| "total_flos": 2.07995682816e+20, | |
| "train_loss": 0.03824057401021322, | |
| "train_runtime": 72024.4779, | |
| "train_samples_per_second": 8.331, | |
| "train_steps_per_second": 0.417 | |
| } | |
| ], | |
| "logging_steps": 2500, | |
| "max_steps": 30000, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 12, | |
| "save_steps": 2500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 2.07995682816e+20, | |
| "train_batch_size": 20, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |