| { | |
| "best_metric": 28.62849533954727, | |
| "best_model_checkpoint": "./whisper-small-ja\\checkpoint-170", | |
| "epoch": 6.0606060606060606, | |
| "global_step": 200, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.3, | |
| "eval_loss": 1.0626804828643799, | |
| "eval_runtime": 108.3983, | |
| "eval_samples_per_second": 0.185, | |
| "eval_steps_per_second": 0.028, | |
| "eval_wer": 123.16910785619174, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.61, | |
| "eval_loss": 0.6694555878639221, | |
| "eval_runtime": 138.6792, | |
| "eval_samples_per_second": 0.144, | |
| "eval_steps_per_second": 0.022, | |
| "eval_wer": 123.43541944074568, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.76, | |
| "learning_rate": 8.974358974358976e-06, | |
| "loss": 1.1477, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.91, | |
| "eval_loss": 0.5782793164253235, | |
| "eval_runtime": 133.633, | |
| "eval_samples_per_second": 0.15, | |
| "eval_steps_per_second": 0.022, | |
| "eval_wer": 102.6631158455393, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 1.21, | |
| "eval_loss": 0.5438617467880249, | |
| "eval_runtime": 132.1171, | |
| "eval_samples_per_second": 0.151, | |
| "eval_steps_per_second": 0.023, | |
| "eval_wer": 100.0, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 1.52, | |
| "learning_rate": 7.692307692307694e-06, | |
| "loss": 0.5121, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 1.52, | |
| "eval_loss": 0.49029508233070374, | |
| "eval_runtime": 130.7034, | |
| "eval_samples_per_second": 0.153, | |
| "eval_steps_per_second": 0.023, | |
| "eval_wer": 100.0, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 1.82, | |
| "eval_loss": 0.45088639855384827, | |
| "eval_runtime": 130.4494, | |
| "eval_samples_per_second": 0.153, | |
| "eval_steps_per_second": 0.023, | |
| "eval_wer": 100.0, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 2.12, | |
| "eval_loss": 0.4300904870033264, | |
| "eval_runtime": 210.4853, | |
| "eval_samples_per_second": 0.095, | |
| "eval_steps_per_second": 0.014, | |
| "eval_wer": 100.0, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 2.27, | |
| "learning_rate": 6.410256410256412e-06, | |
| "loss": 0.3039, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 2.42, | |
| "eval_loss": 0.40350860357284546, | |
| "eval_runtime": 132.7429, | |
| "eval_samples_per_second": 0.151, | |
| "eval_steps_per_second": 0.023, | |
| "eval_wer": 111.31824234354195, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 2.73, | |
| "eval_loss": 0.38837021589279175, | |
| "eval_runtime": 134.6115, | |
| "eval_samples_per_second": 0.149, | |
| "eval_steps_per_second": 0.022, | |
| "eval_wer": 106.52463382157124, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 3.03, | |
| "learning_rate": 5.128205128205128e-06, | |
| "loss": 0.1777, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 3.03, | |
| "eval_loss": 0.3667179048061371, | |
| "eval_runtime": 135.322, | |
| "eval_samples_per_second": 0.148, | |
| "eval_steps_per_second": 0.022, | |
| "eval_wer": 117.31025299600533, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 3.33, | |
| "eval_loss": 0.3598465919494629, | |
| "eval_runtime": 130.8128, | |
| "eval_samples_per_second": 0.153, | |
| "eval_steps_per_second": 0.023, | |
| "eval_wer": 114.51398135818908, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 3.64, | |
| "eval_loss": 0.36537379026412964, | |
| "eval_runtime": 137.4183, | |
| "eval_samples_per_second": 0.146, | |
| "eval_steps_per_second": 0.022, | |
| "eval_wer": 107.0572569906791, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 3.79, | |
| "learning_rate": 3.846153846153847e-06, | |
| "loss": 0.0953, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 3.94, | |
| "eval_loss": 0.35574963688850403, | |
| "eval_runtime": 129.8707, | |
| "eval_samples_per_second": 0.154, | |
| "eval_steps_per_second": 0.023, | |
| "eval_wer": 85.88548601864181, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 4.24, | |
| "eval_loss": 0.3518906533718109, | |
| "eval_runtime": 132.4888, | |
| "eval_samples_per_second": 0.151, | |
| "eval_steps_per_second": 0.023, | |
| "eval_wer": 54.32756324900133, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 4.55, | |
| "learning_rate": 2.564102564102564e-06, | |
| "loss": 0.0589, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 4.55, | |
| "eval_loss": 0.36332806944847107, | |
| "eval_runtime": 134.4235, | |
| "eval_samples_per_second": 0.149, | |
| "eval_steps_per_second": 0.022, | |
| "eval_wer": 43.67509986684421, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 4.85, | |
| "eval_loss": 0.3542396128177643, | |
| "eval_runtime": 181.4645, | |
| "eval_samples_per_second": 0.11, | |
| "eval_steps_per_second": 0.017, | |
| "eval_wer": 43.54194407456725, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 5.15, | |
| "eval_loss": 0.35989412665367126, | |
| "eval_runtime": 113.4241, | |
| "eval_samples_per_second": 0.176, | |
| "eval_steps_per_second": 0.026, | |
| "eval_wer": 28.62849533954727, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 5.3, | |
| "learning_rate": 1.282051282051282e-06, | |
| "loss": 0.0373, | |
| "step": 175 | |
| }, | |
| { | |
| "epoch": 5.45, | |
| "eval_loss": 0.3669775724411011, | |
| "eval_runtime": 115.4134, | |
| "eval_samples_per_second": 0.173, | |
| "eval_steps_per_second": 0.026, | |
| "eval_wer": 28.62849533954727, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 5.76, | |
| "eval_loss": 0.3680533766746521, | |
| "eval_runtime": 114.6467, | |
| "eval_samples_per_second": 0.174, | |
| "eval_steps_per_second": 0.026, | |
| "eval_wer": 32.35685752330226, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 6.06, | |
| "learning_rate": 0.0, | |
| "loss": 0.0271, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 6.06, | |
| "eval_loss": 0.3673397898674011, | |
| "eval_runtime": 189.6433, | |
| "eval_samples_per_second": 0.105, | |
| "eval_steps_per_second": 0.016, | |
| "eval_wer": 32.35685752330226, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 6.06, | |
| "step": 200, | |
| "total_flos": 4.5481059090432e+17, | |
| "train_loss": 0.29501178592443467, | |
| "train_runtime": 14912.6939, | |
| "train_samples_per_second": 0.107, | |
| "train_steps_per_second": 0.013 | |
| } | |
| ], | |
| "max_steps": 200, | |
| "num_train_epochs": 7, | |
| "total_flos": 4.5481059090432e+17, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |