| { | |
| "best_metric": 8.26887341422169, | |
| "best_model_checkpoint": "./fine-tuned-whisper/openai/whisper-medium.en/lr_1e-05_warmup_500_epochs_0_batch_32_grad_acc_1_max_steps_10000/dataset_myst/2023-09-05_18-38-56/checkpoint-500", | |
| "epoch": 2.4, | |
| "eval_steps": 500, | |
| "global_step": 1500, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.16, | |
| "learning_rate": 1.9000000000000002e-06, | |
| "loss": 1.4099, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "learning_rate": 3.88e-06, | |
| "loss": 0.7502, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "learning_rate": 5.8800000000000005e-06, | |
| "loss": 0.5411, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.64, | |
| "learning_rate": 7.88e-06, | |
| "loss": 0.4287, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "learning_rate": 9.88e-06, | |
| "loss": 0.4344, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "eval_loss": 0.24607381224632263, | |
| "eval_runtime": 1060.4953, | |
| "eval_samples_per_second": 2.98, | |
| "eval_steps_per_second": 0.187, | |
| "eval_wer": 8.26887341422169, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "learning_rate": 9.901052631578948e-06, | |
| "loss": 0.3805, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 1.12, | |
| "learning_rate": 9.795789473684213e-06, | |
| "loss": 0.3336, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 1.28, | |
| "learning_rate": 9.690526315789474e-06, | |
| "loss": 0.3087, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 1.44, | |
| "learning_rate": 9.585263157894737e-06, | |
| "loss": 0.3378, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "learning_rate": 9.48e-06, | |
| "loss": 0.3344, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "eval_loss": 0.2452411949634552, | |
| "eval_runtime": 1073.0532, | |
| "eval_samples_per_second": 2.945, | |
| "eval_steps_per_second": 0.185, | |
| "eval_wer": 8.76847797431891, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.76, | |
| "learning_rate": 9.374736842105264e-06, | |
| "loss": 0.3402, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 1.92, | |
| "learning_rate": 9.269473684210527e-06, | |
| "loss": 0.3224, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 2.08, | |
| "learning_rate": 9.16421052631579e-06, | |
| "loss": 0.2762, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 2.24, | |
| "learning_rate": 9.058947368421054e-06, | |
| "loss": 0.223, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 2.4, | |
| "learning_rate": 8.953684210526317e-06, | |
| "loss": 0.2329, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 2.4, | |
| "eval_loss": 0.2578390836715698, | |
| "eval_runtime": 1063.6558, | |
| "eval_samples_per_second": 2.971, | |
| "eval_steps_per_second": 0.186, | |
| "eval_wer": 8.311953858914759, | |
| "step": 1500 | |
| } | |
| ], | |
| "logging_steps": 100, | |
| "max_steps": 10000, | |
| "num_train_epochs": 16, | |
| "save_steps": 500, | |
| "total_flos": 4.896249983336448e+19, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |