| { | |
| "best_metric": 0.30562952160835266, | |
| "best_model_checkpoint": "nrshoudi/hubert-large-ls960-ft-V2/checkpoint-11459", | |
| "epoch": 20.0, | |
| "eval_steps": 500, | |
| "global_step": 32740, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "learning_rate": 5e-05, | |
| "loss": 5.7632, | |
| "step": 1637 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": 0.827274739742279, | |
| "eval_per": 0.37939319702251956, | |
| "eval_runtime": 121.0895, | |
| "eval_samples_per_second": 6.714, | |
| "eval_steps_per_second": 3.361, | |
| "eval_wer": 0.38756940880033597, | |
| "step": 1637 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "learning_rate": 0.0001, | |
| "loss": 0.374, | |
| "step": 3274 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 0.44683241844177246, | |
| "eval_per": 0.04930274192028644, | |
| "eval_runtime": 121.2244, | |
| "eval_samples_per_second": 6.707, | |
| "eval_steps_per_second": 3.357, | |
| "eval_wer": 0.06345947459287947, | |
| "step": 3274 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "learning_rate": 9.444444444444444e-05, | |
| "loss": 0.1657, | |
| "step": 4911 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_loss": 0.36616575717926025, | |
| "eval_per": 0.045957787618957885, | |
| "eval_runtime": 121.3137, | |
| "eval_samples_per_second": 6.702, | |
| "eval_steps_per_second": 3.355, | |
| "eval_wer": 0.059586580187578744, | |
| "step": 4911 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "learning_rate": 8.888888888888889e-05, | |
| "loss": 0.1179, | |
| "step": 6548 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_loss": 0.38089635968208313, | |
| "eval_per": 0.040068783567323095, | |
| "eval_runtime": 121.6978, | |
| "eval_samples_per_second": 6.68, | |
| "eval_steps_per_second": 3.344, | |
| "eval_wer": 0.05184079137697727, | |
| "step": 6548 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "learning_rate": 8.333333333333334e-05, | |
| "loss": 0.0922, | |
| "step": 8185 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_loss": 0.42326560616493225, | |
| "eval_per": 0.038608310562517664, | |
| "eval_runtime": 120.8486, | |
| "eval_samples_per_second": 6.727, | |
| "eval_steps_per_second": 3.368, | |
| "eval_wer": 0.05193411413373151, | |
| "step": 8185 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "learning_rate": 7.777777777777778e-05, | |
| "loss": 0.0763, | |
| "step": 9822 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_loss": 0.38515937328338623, | |
| "eval_per": 0.03900876283802883, | |
| "eval_runtime": 122.1839, | |
| "eval_samples_per_second": 6.654, | |
| "eval_steps_per_second": 3.331, | |
| "eval_wer": 0.05170080724184593, | |
| "step": 9822 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "learning_rate": 7.222222222222222e-05, | |
| "loss": 0.0649, | |
| "step": 11459 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_loss": 0.30562952160835266, | |
| "eval_per": 0.03771318194666918, | |
| "eval_runtime": 122.2792, | |
| "eval_samples_per_second": 6.649, | |
| "eval_steps_per_second": 3.328, | |
| "eval_wer": 0.0486678176473333, | |
| "step": 11459 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "learning_rate": 6.666666666666667e-05, | |
| "loss": 0.056, | |
| "step": 13096 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_loss": 0.3265678882598877, | |
| "eval_per": 0.03531046829360219, | |
| "eval_runtime": 122.0342, | |
| "eval_samples_per_second": 6.662, | |
| "eval_steps_per_second": 3.335, | |
| "eval_wer": 0.046008119079837616, | |
| "step": 13096 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "learning_rate": 6.111111111111112e-05, | |
| "loss": 0.0467, | |
| "step": 14733 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_loss": 0.36078736186027527, | |
| "eval_per": 0.034603787807406014, | |
| "eval_runtime": 122.8523, | |
| "eval_samples_per_second": 6.618, | |
| "eval_steps_per_second": 3.313, | |
| "eval_wer": 0.045774812187952034, | |
| "step": 14733 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "learning_rate": 5.555555555555556e-05, | |
| "loss": 0.04, | |
| "step": 16370 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_loss": 0.3399020731449127, | |
| "eval_per": 0.0354989164232545, | |
| "eval_runtime": 122.172, | |
| "eval_samples_per_second": 6.655, | |
| "eval_steps_per_second": 3.331, | |
| "eval_wer": 0.046054780458214734, | |
| "step": 16370 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "learning_rate": 5e-05, | |
| "loss": 0.0356, | |
| "step": 18007 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_loss": 0.3549004793167114, | |
| "eval_per": 0.03406199943465561, | |
| "eval_runtime": 122.2968, | |
| "eval_samples_per_second": 6.648, | |
| "eval_steps_per_second": 3.328, | |
| "eval_wer": 0.044141663944752926, | |
| "step": 18007 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "learning_rate": 4.4444444444444447e-05, | |
| "loss": 0.0336, | |
| "step": 19644 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_loss": 0.3391400873661041, | |
| "eval_per": 0.0332139828512202, | |
| "eval_runtime": 121.4522, | |
| "eval_samples_per_second": 6.694, | |
| "eval_steps_per_second": 3.351, | |
| "eval_wer": 0.043021790863702114, | |
| "step": 19644 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "learning_rate": 3.888888888888889e-05, | |
| "loss": 0.0338, | |
| "step": 21281 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "eval_loss": 0.396840363740921, | |
| "eval_per": 0.0383256383680392, | |
| "eval_runtime": 121.446, | |
| "eval_samples_per_second": 6.694, | |
| "eval_steps_per_second": 3.351, | |
| "eval_wer": 0.04824786524193925, | |
| "step": 21281 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "learning_rate": 3.3333333333333335e-05, | |
| "loss": 0.0278, | |
| "step": 22918 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_loss": 0.3698951303958893, | |
| "eval_per": 0.03759540186563649, | |
| "eval_runtime": 121.8935, | |
| "eval_samples_per_second": 6.67, | |
| "eval_steps_per_second": 3.339, | |
| "eval_wer": 0.04675470113387149, | |
| "step": 22918 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "learning_rate": 2.777777777777778e-05, | |
| "loss": 0.0224, | |
| "step": 24555 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "eval_loss": 0.36799514293670654, | |
| "eval_per": 0.036511825120135684, | |
| "eval_runtime": 121.7292, | |
| "eval_samples_per_second": 6.679, | |
| "eval_steps_per_second": 3.343, | |
| "eval_wer": 0.04582147356632915, | |
| "step": 24555 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "learning_rate": 2.2222222222222223e-05, | |
| "loss": 0.0225, | |
| "step": 26192 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_loss": 0.37129172682762146, | |
| "eval_per": 0.03917365495147461, | |
| "eval_runtime": 121.6732, | |
| "eval_samples_per_second": 6.682, | |
| "eval_steps_per_second": 3.345, | |
| "eval_wer": 0.04768792870141384, | |
| "step": 26192 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "learning_rate": 1.6666666666666667e-05, | |
| "loss": 0.0189, | |
| "step": 27829 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "eval_loss": 0.3680973947048187, | |
| "eval_per": 0.035027796099123716, | |
| "eval_runtime": 121.6009, | |
| "eval_samples_per_second": 6.686, | |
| "eval_steps_per_second": 3.347, | |
| "eval_wer": 0.044188325323130044, | |
| "step": 27829 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "learning_rate": 1.1111111111111112e-05, | |
| "loss": 0.0185, | |
| "step": 29466 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "eval_loss": 0.38294902443885803, | |
| "eval_per": 0.03523980024498257, | |
| "eval_runtime": 121.5661, | |
| "eval_samples_per_second": 6.688, | |
| "eval_steps_per_second": 3.348, | |
| "eval_wer": 0.04442163221501563, | |
| "step": 29466 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "learning_rate": 5.555555555555556e-06, | |
| "loss": 0.018, | |
| "step": 31103 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "eval_loss": 0.38662299513816833, | |
| "eval_per": 0.03542824837463488, | |
| "eval_runtime": 121.986, | |
| "eval_samples_per_second": 6.665, | |
| "eval_steps_per_second": 3.336, | |
| "eval_wer": 0.04498156875554104, | |
| "step": 31103 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "learning_rate": 0.0, | |
| "loss": 0.0169, | |
| "step": 32740 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_loss": 0.37597647309303284, | |
| "eval_per": 0.03542824837463488, | |
| "eval_runtime": 122.2942, | |
| "eval_samples_per_second": 6.648, | |
| "eval_steps_per_second": 3.328, | |
| "eval_wer": 0.04446829359339275, | |
| "step": 32740 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "step": 32740, | |
| "total_flos": 1.0473975577487264e+19, | |
| "train_loss": 0.3522421885789278, | |
| "train_runtime": 20020.8901, | |
| "train_samples_per_second": 3.27, | |
| "train_steps_per_second": 1.635 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 32740, | |
| "num_train_epochs": 20, | |
| "save_steps": 500, | |
| "total_flos": 1.0473975577487264e+19, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |