| { | |
| "best_metric": 26.367876079171644, | |
| "best_model_checkpoint": "./whisper-large-persian/checkpoint-1000", | |
| "epoch": 0.7776049766718507, | |
| "global_step": 1000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.02, | |
| "learning_rate": 6.712113404111031e-06, | |
| "loss": 0.5237, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.04, | |
| "learning_rate": 8.360489289678585e-06, | |
| "loss": 0.3836, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "learning_rate": 9.28666248215634e-06, | |
| "loss": 0.338, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "learning_rate": 9.933858671331224e-06, | |
| "loss": 0.3656, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "learning_rate": 9.766666666666667e-06, | |
| "loss": 0.3353, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "learning_rate": 9.48888888888889e-06, | |
| "loss": 0.3375, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "learning_rate": 9.211111111111111e-06, | |
| "loss": 0.3143, | |
| "step": 175 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "learning_rate": 8.933333333333333e-06, | |
| "loss": 0.313, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.17, | |
| "learning_rate": 8.655555555555557e-06, | |
| "loss": 0.2926, | |
| "step": 225 | |
| }, | |
| { | |
| "epoch": 0.19, | |
| "learning_rate": 8.377777777777779e-06, | |
| "loss": 0.2948, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.19, | |
| "eval_loss": 0.42578125, | |
| "eval_runtime": 7096.3806, | |
| "eval_samples_per_second": 1.45, | |
| "eval_steps_per_second": 0.091, | |
| "eval_wer": 35.602254115426256, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.21, | |
| "learning_rate": 8.1e-06, | |
| "loss": 0.2774, | |
| "step": 275 | |
| }, | |
| { | |
| "epoch": 0.23, | |
| "learning_rate": 7.822222222222224e-06, | |
| "loss": 0.2789, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "learning_rate": 7.555555555555556e-06, | |
| "loss": 0.2671, | |
| "step": 325 | |
| }, | |
| { | |
| "epoch": 0.27, | |
| "learning_rate": 7.277777777777778e-06, | |
| "loss": 0.2649, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "learning_rate": 7e-06, | |
| "loss": 0.2644, | |
| "step": 375 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "learning_rate": 6.7222222222222235e-06, | |
| "loss": 0.2589, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.33, | |
| "learning_rate": 6.444444444444445e-06, | |
| "loss": 0.2433, | |
| "step": 425 | |
| }, | |
| { | |
| "epoch": 0.35, | |
| "learning_rate": 6.166666666666667e-06, | |
| "loss": 0.2338, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 0.37, | |
| "learning_rate": 5.88888888888889e-06, | |
| "loss": 0.2461, | |
| "step": 475 | |
| }, | |
| { | |
| "epoch": 0.39, | |
| "learning_rate": 5.611111111111112e-06, | |
| "loss": 0.2443, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.39, | |
| "eval_loss": 0.364990234375, | |
| "eval_runtime": 5639.1292, | |
| "eval_samples_per_second": 1.824, | |
| "eval_steps_per_second": 0.114, | |
| "eval_wer": 30.97465508147572, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.41, | |
| "learning_rate": 5.333333333333334e-06, | |
| "loss": 0.2462, | |
| "step": 525 | |
| }, | |
| { | |
| "epoch": 0.43, | |
| "learning_rate": 5.0555555555555555e-06, | |
| "loss": 0.2143, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 0.45, | |
| "learning_rate": 4.7888888888888894e-06, | |
| "loss": 0.235, | |
| "step": 575 | |
| }, | |
| { | |
| "epoch": 0.47, | |
| "learning_rate": 4.511111111111111e-06, | |
| "loss": 0.211, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.49, | |
| "learning_rate": 4.233333333333334e-06, | |
| "loss": 0.2492, | |
| "step": 625 | |
| }, | |
| { | |
| "epoch": 0.51, | |
| "learning_rate": 3.955555555555556e-06, | |
| "loss": 0.2031, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 0.52, | |
| "learning_rate": 3.6777777777777778e-06, | |
| "loss": 0.2229, | |
| "step": 675 | |
| }, | |
| { | |
| "epoch": 0.54, | |
| "learning_rate": 3.4000000000000005e-06, | |
| "loss": 0.1983, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 0.56, | |
| "learning_rate": 3.1222222222222228e-06, | |
| "loss": 0.1866, | |
| "step": 725 | |
| }, | |
| { | |
| "epoch": 0.58, | |
| "learning_rate": 2.8444444444444446e-06, | |
| "loss": 0.1956, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 0.58, | |
| "eval_loss": 0.32275390625, | |
| "eval_runtime": 5514.2809, | |
| "eval_samples_per_second": 1.866, | |
| "eval_steps_per_second": 0.117, | |
| "eval_wer": 28.019598589789858, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "learning_rate": 2.566666666666667e-06, | |
| "loss": 0.1779, | |
| "step": 775 | |
| }, | |
| { | |
| "epoch": 0.62, | |
| "learning_rate": 2.2888888888888892e-06, | |
| "loss": 0.1763, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 0.64, | |
| "learning_rate": 2.011111111111111e-06, | |
| "loss": 0.1974, | |
| "step": 825 | |
| }, | |
| { | |
| "epoch": 0.66, | |
| "learning_rate": 1.7333333333333336e-06, | |
| "loss": 0.1934, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 0.68, | |
| "learning_rate": 1.4555555555555557e-06, | |
| "loss": 0.1955, | |
| "step": 875 | |
| }, | |
| { | |
| "epoch": 0.7, | |
| "learning_rate": 1.1777777777777778e-06, | |
| "loss": 0.1927, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 0.72, | |
| "learning_rate": 9.000000000000001e-07, | |
| "loss": 0.1865, | |
| "step": 925 | |
| }, | |
| { | |
| "epoch": 0.74, | |
| "learning_rate": 6.222222222222223e-07, | |
| "loss": 0.1678, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 0.76, | |
| "learning_rate": 3.444444444444445e-07, | |
| "loss": 0.1611, | |
| "step": 975 | |
| }, | |
| { | |
| "epoch": 0.78, | |
| "learning_rate": 6.666666666666668e-08, | |
| "loss": 0.1715, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.78, | |
| "eval_loss": 0.3046875, | |
| "eval_runtime": 5582.029, | |
| "eval_samples_per_second": 1.843, | |
| "eval_steps_per_second": 0.115, | |
| "eval_wer": 26.367876079171644, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.78, | |
| "step": 1000, | |
| "total_flos": 3.397073665165099e+19, | |
| "train_loss": 0.2514926452636719, | |
| "train_runtime": 31909.1593, | |
| "train_samples_per_second": 0.501, | |
| "train_steps_per_second": 0.031 | |
| } | |
| ], | |
| "max_steps": 1000, | |
| "num_train_epochs": 1, | |
| "total_flos": 3.397073665165099e+19, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |