| { | |
| "best_metric": 0.31682197155468367, | |
| "best_model_checkpoint": "/content/drive/My Drive/Generated_asr/whisper-small-akan-s900a-5e-5/checkpoint-800", | |
| "epoch": 14.222222222222221, | |
| "eval_steps": 100, | |
| "global_step": 800, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.8888888888888888, | |
| "grad_norm": 2.3911514282226562, | |
| "learning_rate": 2.5e-05, | |
| "loss": 0.2289, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 1.7777777777777777, | |
| "grad_norm": 1.2916463613510132, | |
| "learning_rate": 5e-05, | |
| "loss": 0.0983, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 1.7777777777777777, | |
| "eval_loss": 0.8214797973632812, | |
| "eval_runtime": 76.7879, | |
| "eval_samples_per_second": 2.344, | |
| "eval_steps_per_second": 1.172, | |
| "eval_wer": 0.3704430276279222, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 2.6666666666666665, | |
| "grad_norm": 1.2745870351791382, | |
| "learning_rate": 4.962019382530521e-05, | |
| "loss": 0.0491, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 3.5555555555555554, | |
| "grad_norm": 0.887119710445404, | |
| "learning_rate": 4.849231551964771e-05, | |
| "loss": 0.0235, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 3.5555555555555554, | |
| "eval_loss": 0.8763074278831482, | |
| "eval_runtime": 73.6156, | |
| "eval_samples_per_second": 2.445, | |
| "eval_steps_per_second": 1.223, | |
| "eval_wer": 0.3472290338401177, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 4.444444444444445, | |
| "grad_norm": 0.6900314688682556, | |
| "learning_rate": 4.665063509461097e-05, | |
| "loss": 0.0195, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 5.333333333333333, | |
| "grad_norm": 0.3968714773654938, | |
| "learning_rate": 4.415111107797445e-05, | |
| "loss": 0.0123, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 5.333333333333333, | |
| "eval_loss": 0.8793646097183228, | |
| "eval_runtime": 72.8057, | |
| "eval_samples_per_second": 2.472, | |
| "eval_steps_per_second": 1.236, | |
| "eval_wer": 0.33480464279875755, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 6.222222222222222, | |
| "grad_norm": 0.30950671434402466, | |
| "learning_rate": 4.1069690242163484e-05, | |
| "loss": 0.0081, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 7.111111111111111, | |
| "grad_norm": 0.12230431288480759, | |
| "learning_rate": 3.7500000000000003e-05, | |
| "loss": 0.005, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 7.111111111111111, | |
| "eval_loss": 0.8951243758201599, | |
| "eval_runtime": 72.0733, | |
| "eval_samples_per_second": 2.497, | |
| "eval_steps_per_second": 1.249, | |
| "eval_wer": 0.33480464279875755, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "grad_norm": 0.08047913014888763, | |
| "learning_rate": 3.355050358314172e-05, | |
| "loss": 0.0029, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 8.88888888888889, | |
| "grad_norm": 0.05442878231406212, | |
| "learning_rate": 2.9341204441673266e-05, | |
| "loss": 0.0021, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 8.88888888888889, | |
| "eval_loss": 0.8946982622146606, | |
| "eval_runtime": 72.3669, | |
| "eval_samples_per_second": 2.487, | |
| "eval_steps_per_second": 1.244, | |
| "eval_wer": 0.32663070132417854, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 9.777777777777779, | |
| "grad_norm": 0.1563282012939453, | |
| "learning_rate": 2.5e-05, | |
| "loss": 0.0021, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 10.666666666666666, | |
| "grad_norm": 0.01462751254439354, | |
| "learning_rate": 2.0658795558326743e-05, | |
| "loss": 0.0012, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 10.666666666666666, | |
| "eval_loss": 0.9011955857276917, | |
| "eval_runtime": 73.1155, | |
| "eval_samples_per_second": 2.462, | |
| "eval_steps_per_second": 1.231, | |
| "eval_wer": 0.3217263364394311, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 11.555555555555555, | |
| "grad_norm": 0.012643957510590553, | |
| "learning_rate": 1.6449496416858284e-05, | |
| "loss": 0.0023, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 12.444444444444445, | |
| "grad_norm": 0.010622282512485981, | |
| "learning_rate": 1.2500000000000006e-05, | |
| "loss": 0.0006, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 12.444444444444445, | |
| "eval_loss": 0.8978831171989441, | |
| "eval_runtime": 72.6692, | |
| "eval_samples_per_second": 2.477, | |
| "eval_steps_per_second": 1.238, | |
| "eval_wer": 0.31747588687265, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 13.333333333333334, | |
| "grad_norm": 0.009823041036725044, | |
| "learning_rate": 8.930309757836517e-06, | |
| "loss": 0.0004, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 14.222222222222221, | |
| "grad_norm": 0.007314720656722784, | |
| "learning_rate": 5.848888922025553e-06, | |
| "loss": 0.0004, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 14.222222222222221, | |
| "eval_loss": 0.9011634588241577, | |
| "eval_runtime": 72.4323, | |
| "eval_samples_per_second": 2.485, | |
| "eval_steps_per_second": 1.243, | |
| "eval_wer": 0.31682197155468367, | |
| "step": 800 | |
| } | |
| ], | |
| "logging_steps": 50, | |
| "max_steps": 1000, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 18, | |
| "save_steps": 100, | |
| "stateful_callbacks": { | |
| "EarlyStoppingCallback": { | |
| "args": { | |
| "early_stopping_patience": 3, | |
| "early_stopping_threshold": 0.001 | |
| }, | |
| "attributes": { | |
| "early_stopping_patience_counter": 0 | |
| } | |
| }, | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 3.693893124096e+18, | |
| "train_batch_size": 4, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |