| { | |
| "best_metric": 0.3309297263622284, | |
| "best_model_checkpoint": "nrshoudi/hubert_arabic_mdd/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": 5.6351, | |
| "step": 1637 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": 0.7704485058784485, | |
| "eval_per": 0.3110336379911429, | |
| "eval_runtime": 117.9251, | |
| "eval_samples_per_second": 6.894, | |
| "eval_steps_per_second": 3.451, | |
| "eval_wer": 0.3277028603424945, | |
| "step": 1637 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "learning_rate": 0.0001, | |
| "loss": 0.3646, | |
| "step": 3274 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 0.433285653591156, | |
| "eval_per": 0.05403750117780081, | |
| "eval_runtime": 117.5906, | |
| "eval_samples_per_second": 6.914, | |
| "eval_steps_per_second": 3.461, | |
| "eval_wer": 0.06728570761980308, | |
| "step": 3274 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "learning_rate": 9.444444444444444e-05, | |
| "loss": 0.1587, | |
| "step": 4911 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_loss": 0.3875606060028076, | |
| "eval_per": 0.05804202393291247, | |
| "eval_runtime": 117.765, | |
| "eval_samples_per_second": 6.904, | |
| "eval_steps_per_second": 3.456, | |
| "eval_wer": 0.07101861788997246, | |
| "step": 4911 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "learning_rate": 8.888888888888889e-05, | |
| "loss": 0.1105, | |
| "step": 6548 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_loss": 0.3607988655567169, | |
| "eval_per": 0.05578064637708471, | |
| "eval_runtime": 117.9127, | |
| "eval_samples_per_second": 6.895, | |
| "eval_steps_per_second": 3.452, | |
| "eval_wer": 0.06695907797116327, | |
| "step": 6548 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "learning_rate": 8.333333333333334e-05, | |
| "loss": 0.0878, | |
| "step": 8185 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_loss": 0.3808358907699585, | |
| "eval_per": 0.05618109865259587, | |
| "eval_runtime": 117.5542, | |
| "eval_samples_per_second": 6.916, | |
| "eval_steps_per_second": 3.462, | |
| "eval_wer": 0.06775232140357426, | |
| "step": 8185 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "learning_rate": 7.777777777777778e-05, | |
| "loss": 0.0738, | |
| "step": 9822 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_loss": 0.3576147258281708, | |
| "eval_per": 0.05177612362197305, | |
| "eval_runtime": 118.5636, | |
| "eval_samples_per_second": 6.857, | |
| "eval_steps_per_second": 3.433, | |
| "eval_wer": 0.06392608837665065, | |
| "step": 9822 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "learning_rate": 7.222222222222222e-05, | |
| "loss": 0.0594, | |
| "step": 11459 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_loss": 0.3309297263622284, | |
| "eval_per": 0.050810326957504946, | |
| "eval_runtime": 117.7279, | |
| "eval_samples_per_second": 6.906, | |
| "eval_steps_per_second": 3.457, | |
| "eval_wer": 0.06173300359292613, | |
| "step": 11459 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "learning_rate": 6.666666666666667e-05, | |
| "loss": 0.056, | |
| "step": 13096 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_loss": 0.3569846451282501, | |
| "eval_per": 0.046711580137567134, | |
| "eval_runtime": 118.1175, | |
| "eval_samples_per_second": 6.883, | |
| "eval_steps_per_second": 3.446, | |
| "eval_wer": 0.058700013998413514, | |
| "step": 13096 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "learning_rate": 6.111111111111112e-05, | |
| "loss": 0.0478, | |
| "step": 14733 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_loss": 0.34502002596855164, | |
| "eval_per": 0.04727692452652407, | |
| "eval_runtime": 117.9103, | |
| "eval_samples_per_second": 6.895, | |
| "eval_steps_per_second": 3.452, | |
| "eval_wer": 0.05832672297139657, | |
| "step": 14733 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "learning_rate": 5.555555555555556e-05, | |
| "loss": 0.0388, | |
| "step": 16370 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_loss": 0.3993436396121979, | |
| "eval_per": 0.04803071704513333, | |
| "eval_runtime": 117.7116, | |
| "eval_samples_per_second": 6.907, | |
| "eval_steps_per_second": 3.458, | |
| "eval_wer": 0.05907330502543045, | |
| "step": 16370 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "learning_rate": 5e-05, | |
| "loss": 0.0378, | |
| "step": 18007 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_loss": 0.4171537160873413, | |
| "eval_per": 0.0437435220955432, | |
| "eval_runtime": 118.1365, | |
| "eval_samples_per_second": 6.882, | |
| "eval_steps_per_second": 3.445, | |
| "eval_wer": 0.05501376510662125, | |
| "step": 18007 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "learning_rate": 4.4444444444444447e-05, | |
| "loss": 0.0319, | |
| "step": 19644 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_loss": 0.3625492751598358, | |
| "eval_per": 0.04131725242626967, | |
| "eval_runtime": 118.0928, | |
| "eval_samples_per_second": 6.884, | |
| "eval_steps_per_second": 3.446, | |
| "eval_wer": 0.051794129998600155, | |
| "step": 19644 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "learning_rate": 3.888888888888889e-05, | |
| "loss": 0.0309, | |
| "step": 21281 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "eval_loss": 0.3794006109237671, | |
| "eval_per": 0.03919721096768115, | |
| "eval_runtime": 118.3172, | |
| "eval_samples_per_second": 6.871, | |
| "eval_steps_per_second": 3.44, | |
| "eval_wer": 0.05053427278241799, | |
| "step": 21281 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "learning_rate": 3.3333333333333335e-05, | |
| "loss": 0.0257, | |
| "step": 22918 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_loss": 0.3896752893924713, | |
| "eval_per": 0.04263638933383586, | |
| "eval_runtime": 118.7663, | |
| "eval_samples_per_second": 6.845, | |
| "eval_steps_per_second": 3.427, | |
| "eval_wer": 0.05342727824179926, | |
| "step": 22918 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "learning_rate": 2.777777777777778e-05, | |
| "loss": 0.0202, | |
| "step": 24555 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "eval_loss": 0.40769127011299133, | |
| "eval_per": 0.04042212381042118, | |
| "eval_runtime": 118.2316, | |
| "eval_samples_per_second": 6.876, | |
| "eval_steps_per_second": 3.442, | |
| "eval_wer": 0.05123419345807475, | |
| "step": 24555 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "learning_rate": 2.2222222222222223e-05, | |
| "loss": 0.0197, | |
| "step": 26192 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_loss": 0.38464316725730896, | |
| "eval_per": 0.041128804296617354, | |
| "eval_runtime": 118.9714, | |
| "eval_samples_per_second": 6.834, | |
| "eval_steps_per_second": 3.421, | |
| "eval_wer": 0.05076757967430358, | |
| "step": 26192 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "learning_rate": 1.6666666666666667e-05, | |
| "loss": 0.016, | |
| "step": 27829 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "eval_loss": 0.3807629942893982, | |
| "eval_per": 0.03910298690285499, | |
| "eval_runtime": 118.4593, | |
| "eval_samples_per_second": 6.863, | |
| "eval_steps_per_second": 3.436, | |
| "eval_wer": 0.04913443143110448, | |
| "step": 27829 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "learning_rate": 1.1111111111111112e-05, | |
| "loss": 0.0157, | |
| "step": 29466 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "eval_loss": 0.4060150980949402, | |
| "eval_per": 0.03726561763874493, | |
| "eval_runtime": 118.2175, | |
| "eval_samples_per_second": 6.877, | |
| "eval_steps_per_second": 3.443, | |
| "eval_wer": 0.047127992160888435, | |
| "step": 29466 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "learning_rate": 5.555555555555556e-06, | |
| "loss": 0.0131, | |
| "step": 31103 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "eval_loss": 0.3925394117832184, | |
| "eval_per": 0.03611137284462452, | |
| "eval_runtime": 118.443, | |
| "eval_samples_per_second": 6.864, | |
| "eval_steps_per_second": 3.436, | |
| "eval_wer": 0.04652139424198591, | |
| "step": 31103 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "learning_rate": 0.0, | |
| "loss": 0.0115, | |
| "step": 32740 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_loss": 0.39911746978759766, | |
| "eval_per": 0.036252708941863755, | |
| "eval_runtime": 118.88, | |
| "eval_samples_per_second": 6.839, | |
| "eval_steps_per_second": 3.424, | |
| "eval_wer": 0.04656805562036303, | |
| "step": 32740 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "step": 32740, | |
| "total_flos": 1.0473975577487264e+19, | |
| "train_loss": 0.3427419745201153, | |
| "train_runtime": 19585.0411, | |
| "train_samples_per_second": 3.342, | |
| "train_steps_per_second": 1.672 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 32740, | |
| "num_train_epochs": 20, | |
| "save_steps": 500, | |
| "total_flos": 1.0473975577487264e+19, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |