| { | |
| "best_metric": 15.558516353633904, | |
| "best_model_checkpoint": "/scratch/project_465000370/asr_children/models/whisper-large-et-clinic/checkpoint-1500", | |
| "epoch": 16.12806403201601, | |
| "global_step": 2000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.2, | |
| "learning_rate": 1.2000000000000002e-06, | |
| "loss": 0.5006, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "learning_rate": 2.4000000000000003e-06, | |
| "loss": 0.2673, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "learning_rate": 3.6000000000000003e-06, | |
| "loss": 0.2076, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "learning_rate": 4.85e-06, | |
| "loss": 0.2199, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 1.01, | |
| "learning_rate": 6.1e-06, | |
| "loss": 0.2007, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 1.21, | |
| "learning_rate": 7.350000000000001e-06, | |
| "loss": 0.1242, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 1.41, | |
| "learning_rate": 8.6e-06, | |
| "loss": 0.1211, | |
| "step": 175 | |
| }, | |
| { | |
| "epoch": 1.61, | |
| "learning_rate": 9.85e-06, | |
| "loss": 0.1206, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 1.81, | |
| "learning_rate": 9.877777777777778e-06, | |
| "loss": 0.1183, | |
| "step": 225 | |
| }, | |
| { | |
| "epoch": 2.02, | |
| "learning_rate": 9.73888888888889e-06, | |
| "loss": 0.1292, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 2.22, | |
| "learning_rate": 9.605555555555556e-06, | |
| "loss": 0.049, | |
| "step": 275 | |
| }, | |
| { | |
| "epoch": 2.42, | |
| "learning_rate": 9.466666666666667e-06, | |
| "loss": 0.0546, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 2.62, | |
| "learning_rate": 9.32777777777778e-06, | |
| "loss": 0.0605, | |
| "step": 325 | |
| }, | |
| { | |
| "epoch": 2.82, | |
| "learning_rate": 9.188888888888889e-06, | |
| "loss": 0.0569, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 3.02, | |
| "learning_rate": 9.050000000000001e-06, | |
| "loss": 0.0469, | |
| "step": 375 | |
| }, | |
| { | |
| "epoch": 3.22, | |
| "learning_rate": 8.91111111111111e-06, | |
| "loss": 0.0267, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 3.42, | |
| "learning_rate": 8.772222222222223e-06, | |
| "loss": 0.0254, | |
| "step": 425 | |
| }, | |
| { | |
| "epoch": 3.62, | |
| "learning_rate": 8.633333333333334e-06, | |
| "loss": 0.0272, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 3.82, | |
| "learning_rate": 8.494444444444445e-06, | |
| "loss": 0.0274, | |
| "step": 475 | |
| }, | |
| { | |
| "epoch": 4.03, | |
| "learning_rate": 8.355555555555556e-06, | |
| "loss": 0.0302, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 4.03, | |
| "eval_loss": 0.2971087098121643, | |
| "eval_runtime": 2340.0934, | |
| "eval_samples_per_second": 0.55, | |
| "eval_steps_per_second": 0.55, | |
| "eval_wer": 16.289164911677485, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 4.23, | |
| "learning_rate": 8.216666666666667e-06, | |
| "loss": 0.0143, | |
| "step": 525 | |
| }, | |
| { | |
| "epoch": 4.43, | |
| "learning_rate": 8.077777777777778e-06, | |
| "loss": 0.0142, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 4.63, | |
| "learning_rate": 7.938888888888889e-06, | |
| "loss": 0.0159, | |
| "step": 575 | |
| }, | |
| { | |
| "epoch": 4.83, | |
| "learning_rate": 7.800000000000002e-06, | |
| "loss": 0.0144, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 5.04, | |
| "learning_rate": 7.66111111111111e-06, | |
| "loss": 0.0137, | |
| "step": 625 | |
| }, | |
| { | |
| "epoch": 5.24, | |
| "learning_rate": 7.5222222222222226e-06, | |
| "loss": 0.0094, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 5.44, | |
| "learning_rate": 7.3833333333333335e-06, | |
| "loss": 0.0103, | |
| "step": 675 | |
| }, | |
| { | |
| "epoch": 5.64, | |
| "learning_rate": 7.244444444444445e-06, | |
| "loss": 0.0095, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 5.84, | |
| "learning_rate": 7.105555555555556e-06, | |
| "loss": 0.009, | |
| "step": 725 | |
| }, | |
| { | |
| "epoch": 6.05, | |
| "learning_rate": 6.966666666666667e-06, | |
| "loss": 0.0088, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 6.25, | |
| "learning_rate": 6.827777777777779e-06, | |
| "loss": 0.0065, | |
| "step": 775 | |
| }, | |
| { | |
| "epoch": 6.45, | |
| "learning_rate": 6.688888888888889e-06, | |
| "loss": 0.0055, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 6.65, | |
| "learning_rate": 6.550000000000001e-06, | |
| "loss": 0.0074, | |
| "step": 825 | |
| }, | |
| { | |
| "epoch": 6.85, | |
| "learning_rate": 6.411111111111111e-06, | |
| "loss": 0.0049, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 7.06, | |
| "learning_rate": 6.272222222222223e-06, | |
| "loss": 0.0052, | |
| "step": 875 | |
| }, | |
| { | |
| "epoch": 7.26, | |
| "learning_rate": 6.133333333333334e-06, | |
| "loss": 0.0037, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 7.46, | |
| "learning_rate": 5.9944444444444446e-06, | |
| "loss": 0.0034, | |
| "step": 925 | |
| }, | |
| { | |
| "epoch": 7.66, | |
| "learning_rate": 5.855555555555556e-06, | |
| "loss": 0.0032, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 7.86, | |
| "learning_rate": 5.716666666666667e-06, | |
| "loss": 0.0035, | |
| "step": 975 | |
| }, | |
| { | |
| "epoch": 8.06, | |
| "learning_rate": 5.577777777777778e-06, | |
| "loss": 0.0042, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 8.06, | |
| "eval_loss": 0.3406260907649994, | |
| "eval_runtime": 2343.2491, | |
| "eval_samples_per_second": 0.549, | |
| "eval_steps_per_second": 0.549, | |
| "eval_wer": 15.855073709545708, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 8.26, | |
| "learning_rate": 5.438888888888889e-06, | |
| "loss": 0.0023, | |
| "step": 1025 | |
| }, | |
| { | |
| "epoch": 8.46, | |
| "learning_rate": 5.300000000000001e-06, | |
| "loss": 0.0036, | |
| "step": 1050 | |
| }, | |
| { | |
| "epoch": 8.66, | |
| "learning_rate": 5.161111111111111e-06, | |
| "loss": 0.0027, | |
| "step": 1075 | |
| }, | |
| { | |
| "epoch": 8.86, | |
| "learning_rate": 5.022222222222223e-06, | |
| "loss": 0.0027, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 9.07, | |
| "learning_rate": 4.883333333333334e-06, | |
| "loss": 0.002, | |
| "step": 1125 | |
| }, | |
| { | |
| "epoch": 9.27, | |
| "learning_rate": 4.744444444444445e-06, | |
| "loss": 0.0022, | |
| "step": 1150 | |
| }, | |
| { | |
| "epoch": 9.47, | |
| "learning_rate": 4.605555555555556e-06, | |
| "loss": 0.002, | |
| "step": 1175 | |
| }, | |
| { | |
| "epoch": 9.67, | |
| "learning_rate": 4.4666666666666665e-06, | |
| "loss": 0.0019, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 9.87, | |
| "learning_rate": 4.327777777777778e-06, | |
| "loss": 0.0019, | |
| "step": 1225 | |
| }, | |
| { | |
| "epoch": 10.08, | |
| "learning_rate": 4.188888888888889e-06, | |
| "loss": 0.0017, | |
| "step": 1250 | |
| }, | |
| { | |
| "epoch": 10.28, | |
| "learning_rate": 4.05e-06, | |
| "loss": 0.0013, | |
| "step": 1275 | |
| }, | |
| { | |
| "epoch": 10.48, | |
| "learning_rate": 3.911111111111112e-06, | |
| "loss": 0.0026, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 10.68, | |
| "learning_rate": 3.7722222222222225e-06, | |
| "loss": 0.0018, | |
| "step": 1325 | |
| }, | |
| { | |
| "epoch": 10.88, | |
| "learning_rate": 3.633333333333334e-06, | |
| "loss": 0.0024, | |
| "step": 1350 | |
| }, | |
| { | |
| "epoch": 11.09, | |
| "learning_rate": 3.4944444444444448e-06, | |
| "loss": 0.0015, | |
| "step": 1375 | |
| }, | |
| { | |
| "epoch": 11.29, | |
| "learning_rate": 3.3555555555555557e-06, | |
| "loss": 0.0012, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 11.49, | |
| "learning_rate": 3.2166666666666666e-06, | |
| "loss": 0.0012, | |
| "step": 1425 | |
| }, | |
| { | |
| "epoch": 11.69, | |
| "learning_rate": 3.077777777777778e-06, | |
| "loss": 0.0017, | |
| "step": 1450 | |
| }, | |
| { | |
| "epoch": 11.89, | |
| "learning_rate": 2.938888888888889e-06, | |
| "loss": 0.0019, | |
| "step": 1475 | |
| }, | |
| { | |
| "epoch": 12.1, | |
| "learning_rate": 2.8000000000000003e-06, | |
| "loss": 0.0017, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 12.1, | |
| "eval_loss": 0.37141528725624084, | |
| "eval_runtime": 2358.6091, | |
| "eval_samples_per_second": 0.545, | |
| "eval_steps_per_second": 0.545, | |
| "eval_wer": 15.558516353633904, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 12.3, | |
| "learning_rate": 2.6611111111111117e-06, | |
| "loss": 0.0012, | |
| "step": 1525 | |
| }, | |
| { | |
| "epoch": 12.5, | |
| "learning_rate": 2.5222222222222226e-06, | |
| "loss": 0.0016, | |
| "step": 1550 | |
| }, | |
| { | |
| "epoch": 12.7, | |
| "learning_rate": 2.3833333333333335e-06, | |
| "loss": 0.0016, | |
| "step": 1575 | |
| }, | |
| { | |
| "epoch": 12.9, | |
| "learning_rate": 2.2444444444444445e-06, | |
| "loss": 0.0011, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 13.1, | |
| "learning_rate": 2.105555555555556e-06, | |
| "loss": 0.0012, | |
| "step": 1625 | |
| }, | |
| { | |
| "epoch": 13.3, | |
| "learning_rate": 1.9666666666666668e-06, | |
| "loss": 0.0011, | |
| "step": 1650 | |
| }, | |
| { | |
| "epoch": 13.5, | |
| "learning_rate": 1.8277777777777781e-06, | |
| "loss": 0.0014, | |
| "step": 1675 | |
| }, | |
| { | |
| "epoch": 13.7, | |
| "learning_rate": 1.688888888888889e-06, | |
| "loss": 0.0011, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 13.9, | |
| "learning_rate": 1.5500000000000002e-06, | |
| "loss": 0.0014, | |
| "step": 1725 | |
| }, | |
| { | |
| "epoch": 14.11, | |
| "learning_rate": 1.4111111111111111e-06, | |
| "loss": 0.0013, | |
| "step": 1750 | |
| }, | |
| { | |
| "epoch": 14.31, | |
| "learning_rate": 1.2722222222222223e-06, | |
| "loss": 0.0012, | |
| "step": 1775 | |
| }, | |
| { | |
| "epoch": 14.51, | |
| "learning_rate": 1.1333333333333334e-06, | |
| "loss": 0.001, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 14.71, | |
| "learning_rate": 9.944444444444446e-07, | |
| "loss": 0.001, | |
| "step": 1825 | |
| }, | |
| { | |
| "epoch": 14.91, | |
| "learning_rate": 8.555555555555556e-07, | |
| "loss": 0.0012, | |
| "step": 1850 | |
| }, | |
| { | |
| "epoch": 15.12, | |
| "learning_rate": 7.166666666666668e-07, | |
| "loss": 0.0011, | |
| "step": 1875 | |
| }, | |
| { | |
| "epoch": 15.32, | |
| "learning_rate": 5.777777777777778e-07, | |
| "loss": 0.0011, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 15.52, | |
| "learning_rate": 4.388888888888889e-07, | |
| "loss": 0.0009, | |
| "step": 1925 | |
| }, | |
| { | |
| "epoch": 15.72, | |
| "learning_rate": 3.0000000000000004e-07, | |
| "loss": 0.0012, | |
| "step": 1950 | |
| }, | |
| { | |
| "epoch": 15.92, | |
| "learning_rate": 1.6111111111111113e-07, | |
| "loss": 0.0012, | |
| "step": 1975 | |
| }, | |
| { | |
| "epoch": 16.13, | |
| "learning_rate": 2.2222222222222224e-08, | |
| "loss": 0.0009, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 16.13, | |
| "eval_loss": 0.393410325050354, | |
| "eval_runtime": 2357.7409, | |
| "eval_samples_per_second": 0.545, | |
| "eval_steps_per_second": 0.545, | |
| "eval_wer": 15.644475007521383, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 16.13, | |
| "step": 2000, | |
| "total_flos": 1.36902067421184e+20, | |
| "train_loss": 0.03306677715945989, | |
| "train_runtime": 38266.3528, | |
| "train_samples_per_second": 1.672, | |
| "train_steps_per_second": 0.052 | |
| } | |
| ], | |
| "max_steps": 2000, | |
| "num_train_epochs": 17, | |
| "total_flos": 1.36902067421184e+20, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |