| { | |
| "best_metric": 17.391605006569392, | |
| "best_model_checkpoint": "/content/drive/MyDrive/whisper_it/checkpoint-900", | |
| "epoch": 0.9433962264150944, | |
| "global_step": 900, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.03, | |
| "learning_rate": 4.2000000000000006e-07, | |
| "loss": 3.6389, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "learning_rate": 9.200000000000001e-07, | |
| "loss": 3.019, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "learning_rate": 1.42e-06, | |
| "loss": 2.2988, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "learning_rate": 1.9200000000000003e-06, | |
| "loss": 1.4744, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "eval_loss": 1.1852223873138428, | |
| "eval_runtime": 687.2583, | |
| "eval_samples_per_second": 2.183, | |
| "eval_steps_per_second": 0.274, | |
| "eval_wer": 117.60597469054699, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "learning_rate": 2.42e-06, | |
| "loss": 1.062, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "learning_rate": 2.92e-06, | |
| "loss": 0.931, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "learning_rate": 3.4200000000000007e-06, | |
| "loss": 0.8408, | |
| "step": 175 | |
| }, | |
| { | |
| "epoch": 0.21, | |
| "learning_rate": 3.920000000000001e-06, | |
| "loss": 0.7241, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.21, | |
| "eval_loss": 0.7452345490455627, | |
| "eval_runtime": 856.828, | |
| "eval_samples_per_second": 1.751, | |
| "eval_steps_per_second": 0.219, | |
| "eval_wer": 79.73860728856926, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "learning_rate": 4.42e-06, | |
| "loss": 0.6557, | |
| "step": 225 | |
| }, | |
| { | |
| "epoch": 0.26, | |
| "learning_rate": 4.92e-06, | |
| "loss": 0.5752, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "learning_rate": 5.420000000000001e-06, | |
| "loss": 0.5109, | |
| "step": 275 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "learning_rate": 5.92e-06, | |
| "loss": 0.3321, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "eval_loss": 0.32159581780433655, | |
| "eval_runtime": 644.4221, | |
| "eval_samples_per_second": 2.328, | |
| "eval_steps_per_second": 0.292, | |
| "eval_wer": 21.04971993638061, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.34, | |
| "learning_rate": 6.42e-06, | |
| "loss": 0.2987, | |
| "step": 325 | |
| }, | |
| { | |
| "epoch": 0.37, | |
| "learning_rate": 6.92e-06, | |
| "loss": 0.2577, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 0.39, | |
| "learning_rate": 7.420000000000001e-06, | |
| "loss": 0.2928, | |
| "step": 375 | |
| }, | |
| { | |
| "epoch": 0.42, | |
| "learning_rate": 7.92e-06, | |
| "loss": 0.293, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.42, | |
| "eval_loss": 0.3030182719230652, | |
| "eval_runtime": 618.4321, | |
| "eval_samples_per_second": 2.425, | |
| "eval_steps_per_second": 0.304, | |
| "eval_wer": 20.212986653758385, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.45, | |
| "learning_rate": 8.42e-06, | |
| "loss": 0.2734, | |
| "step": 425 | |
| }, | |
| { | |
| "epoch": 0.47, | |
| "learning_rate": 8.920000000000001e-06, | |
| "loss": 0.2863, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "learning_rate": 9.42e-06, | |
| "loss": 0.2847, | |
| "step": 475 | |
| }, | |
| { | |
| "epoch": 0.52, | |
| "learning_rate": 9.920000000000002e-06, | |
| "loss": 0.2698, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.52, | |
| "eval_loss": 0.2989492118358612, | |
| "eval_runtime": 616.9209, | |
| "eval_samples_per_second": 2.431, | |
| "eval_steps_per_second": 0.305, | |
| "eval_wer": 19.76350183251504, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.55, | |
| "learning_rate": 9.537444933920706e-06, | |
| "loss": 0.2828, | |
| "step": 525 | |
| }, | |
| { | |
| "epoch": 0.58, | |
| "learning_rate": 8.986784140969164e-06, | |
| "loss": 0.2609, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "learning_rate": 8.436123348017622e-06, | |
| "loss": 0.242, | |
| "step": 575 | |
| }, | |
| { | |
| "epoch": 0.63, | |
| "learning_rate": 7.88546255506608e-06, | |
| "loss": 0.2453, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.63, | |
| "eval_loss": 0.28984662890434265, | |
| "eval_runtime": 632.3877, | |
| "eval_samples_per_second": 2.372, | |
| "eval_steps_per_second": 0.297, | |
| "eval_wer": 19.009750363045434, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.66, | |
| "learning_rate": 7.334801762114538e-06, | |
| "loss": 0.2461, | |
| "step": 625 | |
| }, | |
| { | |
| "epoch": 0.68, | |
| "learning_rate": 6.784140969162997e-06, | |
| "loss": 0.2962, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 0.71, | |
| "learning_rate": 6.233480176211454e-06, | |
| "loss": 0.2454, | |
| "step": 675 | |
| }, | |
| { | |
| "epoch": 0.73, | |
| "learning_rate": 5.682819383259912e-06, | |
| "loss": 0.2338, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 0.73, | |
| "eval_loss": 0.27684447169303894, | |
| "eval_runtime": 636.9783, | |
| "eval_samples_per_second": 2.355, | |
| "eval_steps_per_second": 0.295, | |
| "eval_wer": 18.705483714819167, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 0.76, | |
| "learning_rate": 5.1321585903083705e-06, | |
| "loss": 0.2568, | |
| "step": 725 | |
| }, | |
| { | |
| "epoch": 0.79, | |
| "learning_rate": 4.581497797356828e-06, | |
| "loss": 0.2672, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 0.81, | |
| "learning_rate": 4.030837004405287e-06, | |
| "loss": 0.2256, | |
| "step": 775 | |
| }, | |
| { | |
| "epoch": 0.84, | |
| "learning_rate": 3.4801762114537445e-06, | |
| "loss": 0.2402, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 0.84, | |
| "eval_loss": 0.26460355520248413, | |
| "eval_runtime": 635.6494, | |
| "eval_samples_per_second": 2.36, | |
| "eval_steps_per_second": 0.296, | |
| "eval_wer": 18.221423138095567, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 0.86, | |
| "learning_rate": 2.9295154185022028e-06, | |
| "loss": 0.2636, | |
| "step": 825 | |
| }, | |
| { | |
| "epoch": 0.89, | |
| "learning_rate": 2.378854625550661e-06, | |
| "loss": 0.2354, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 0.92, | |
| "learning_rate": 1.828193832599119e-06, | |
| "loss": 0.2205, | |
| "step": 875 | |
| }, | |
| { | |
| "epoch": 0.94, | |
| "learning_rate": 1.2775330396475772e-06, | |
| "loss": 0.234, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 0.94, | |
| "eval_loss": 0.2581839859485626, | |
| "eval_runtime": 635.7247, | |
| "eval_samples_per_second": 2.36, | |
| "eval_steps_per_second": 0.296, | |
| "eval_wer": 17.391605006569392, | |
| "step": 900 | |
| } | |
| ], | |
| "max_steps": 954, | |
| "num_train_epochs": 1, | |
| "total_flos": 4.155629764608e+18, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |