| { | |
| "best_metric": 0.9328228831291199, | |
| "best_model_checkpoint": "wav2vec2-base-swbd-turn-eos-long_short_utt_removed_3percent/checkpoint-3000", | |
| "epoch": 19.525801952580196, | |
| "global_step": 14000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.7, | |
| "learning_rate": 4.9800000000000004e-05, | |
| "loss": 3.6678, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.39, | |
| "learning_rate": 9.970000000000001e-05, | |
| "loss": 1.8593, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.39, | |
| "eval_loss": 1.3255794048309326, | |
| "eval_runtime": 303.7268, | |
| "eval_samples_per_second": 4.488, | |
| "eval_steps_per_second": 0.142, | |
| "eval_wer": 0.5794877337079866, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 2.09, | |
| "learning_rate": 9.627436281859072e-05, | |
| "loss": 0.9996, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 2.79, | |
| "learning_rate": 9.253373313343329e-05, | |
| "loss": 0.7576, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 2.79, | |
| "eval_loss": 1.0275607109069824, | |
| "eval_runtime": 383.8668, | |
| "eval_samples_per_second": 3.551, | |
| "eval_steps_per_second": 0.112, | |
| "eval_wer": 0.4334810331784286, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 3.49, | |
| "learning_rate": 8.87856071964018e-05, | |
| "loss": 0.6464, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 4.18, | |
| "learning_rate": 8.503748125937032e-05, | |
| "loss": 0.556, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 4.18, | |
| "eval_loss": 0.9328228831291199, | |
| "eval_runtime": 295.659, | |
| "eval_samples_per_second": 4.61, | |
| "eval_steps_per_second": 0.145, | |
| "eval_wer": 0.3811376490507583, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 4.88, | |
| "learning_rate": 8.128935532233884e-05, | |
| "loss": 0.4897, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 5.58, | |
| "learning_rate": 7.75487256371814e-05, | |
| "loss": 0.4301, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 5.58, | |
| "eval_loss": 0.9558691382408142, | |
| "eval_runtime": 391.7361, | |
| "eval_samples_per_second": 3.479, | |
| "eval_steps_per_second": 0.11, | |
| "eval_wer": 0.3691775640332865, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 6.28, | |
| "learning_rate": 7.380059970014994e-05, | |
| "loss": 0.3938, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 6.97, | |
| "learning_rate": 7.005247376311844e-05, | |
| "loss": 0.3649, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 6.97, | |
| "eval_loss": 1.070428729057312, | |
| "eval_runtime": 346.5542, | |
| "eval_samples_per_second": 3.933, | |
| "eval_steps_per_second": 0.124, | |
| "eval_wer": 0.38769408119889043, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 7.67, | |
| "learning_rate": 6.630434782608695e-05, | |
| "loss": 0.3179, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 8.37, | |
| "learning_rate": 6.255622188905547e-05, | |
| "loss": 0.3064, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 8.37, | |
| "eval_loss": 1.0000288486480713, | |
| "eval_runtime": 352.1995, | |
| "eval_samples_per_second": 3.87, | |
| "eval_steps_per_second": 0.122, | |
| "eval_wer": 0.33495442919413526, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 9.07, | |
| "learning_rate": 5.8815592203898054e-05, | |
| "loss": 0.2794, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 9.76, | |
| "learning_rate": 5.5067466266866574e-05, | |
| "loss": 0.259, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 9.76, | |
| "eval_loss": 1.0542327165603638, | |
| "eval_runtime": 347.0396, | |
| "eval_samples_per_second": 3.928, | |
| "eval_steps_per_second": 0.124, | |
| "eval_wer": 0.3227421737094276, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 10.46, | |
| "learning_rate": 5.131934032983509e-05, | |
| "loss": 0.2359, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 11.16, | |
| "learning_rate": 4.7571214392803595e-05, | |
| "loss": 0.2288, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 11.16, | |
| "eval_loss": 1.0856573581695557, | |
| "eval_runtime": 411.1438, | |
| "eval_samples_per_second": 3.315, | |
| "eval_steps_per_second": 0.105, | |
| "eval_wer": 0.3141683778234086, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 11.85, | |
| "learning_rate": 4.3823088455772116e-05, | |
| "loss": 0.2083, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 12.55, | |
| "learning_rate": 4.007496251874063e-05, | |
| "loss": 0.1915, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 12.55, | |
| "eval_loss": 1.0417132377624512, | |
| "eval_runtime": 346.3488, | |
| "eval_samples_per_second": 3.935, | |
| "eval_steps_per_second": 0.124, | |
| "eval_wer": 0.3155012788645124, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 13.25, | |
| "learning_rate": 3.6326836581709144e-05, | |
| "loss": 0.1805, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 13.95, | |
| "learning_rate": 3.2578710644677665e-05, | |
| "loss": 0.1706, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 13.95, | |
| "eval_loss": 1.1116188764572144, | |
| "eval_runtime": 345.9351, | |
| "eval_samples_per_second": 3.94, | |
| "eval_steps_per_second": 0.124, | |
| "eval_wer": 0.31132245397888975, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 14.64, | |
| "learning_rate": 2.883058470764618e-05, | |
| "loss": 0.1577, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 15.34, | |
| "learning_rate": 2.5089955022488758e-05, | |
| "loss": 0.1444, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 15.34, | |
| "eval_loss": 1.156752586364746, | |
| "eval_runtime": 344.3787, | |
| "eval_samples_per_second": 3.958, | |
| "eval_steps_per_second": 0.125, | |
| "eval_wer": 0.30908894412622934, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 16.04, | |
| "learning_rate": 2.1349325337331336e-05, | |
| "loss": 0.1396, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 16.74, | |
| "learning_rate": 1.760119940029985e-05, | |
| "loss": 0.1247, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 16.74, | |
| "eval_loss": 1.1672117710113525, | |
| "eval_runtime": 361.7016, | |
| "eval_samples_per_second": 3.768, | |
| "eval_steps_per_second": 0.119, | |
| "eval_wer": 0.30195612233870095, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 17.43, | |
| "learning_rate": 1.3853073463268368e-05, | |
| "loss": 0.1239, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 18.13, | |
| "learning_rate": 1.0104947526236883e-05, | |
| "loss": 0.116, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 18.13, | |
| "eval_loss": 1.1757261753082275, | |
| "eval_runtime": 358.9537, | |
| "eval_samples_per_second": 3.797, | |
| "eval_steps_per_second": 0.12, | |
| "eval_wer": 0.30591880110955005, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 18.83, | |
| "learning_rate": 6.36431784107946e-06, | |
| "loss": 0.1093, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 19.53, | |
| "learning_rate": 2.6161919040479763e-06, | |
| "loss": 0.1057, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 19.53, | |
| "eval_loss": 1.1997692584991455, | |
| "eval_runtime": 297.9761, | |
| "eval_samples_per_second": 4.574, | |
| "eval_steps_per_second": 0.144, | |
| "eval_wer": 0.30609892287186136, | |
| "step": 14000 | |
| } | |
| ], | |
| "max_steps": 14340, | |
| "num_train_epochs": 20, | |
| "total_flos": 2.5513525084170887e+19, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |