| { | |
| "best_metric": 41.53041112622491, | |
| "best_model_checkpoint": "./whisper-small-Persian/checkpoint-2500", | |
| "epoch": 7.786936236391913, | |
| "global_step": 2500, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.08, | |
| "learning_rate": 4.4e-08, | |
| "loss": 1.4232, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "learning_rate": 9.4e-08, | |
| "loss": 1.4027, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.23, | |
| "learning_rate": 1.44e-07, | |
| "loss": 1.3731, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "learning_rate": 1.94e-07, | |
| "loss": 1.2969, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.39, | |
| "learning_rate": 2.4399999999999996e-07, | |
| "loss": 1.1964, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 0.47, | |
| "learning_rate": 2.9399999999999996e-07, | |
| "loss": 1.0995, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.54, | |
| "learning_rate": 3.4399999999999996e-07, | |
| "loss": 1.0068, | |
| "step": 175 | |
| }, | |
| { | |
| "epoch": 0.62, | |
| "learning_rate": 3.94e-07, | |
| "loss": 0.8949, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.7, | |
| "learning_rate": 4.44e-07, | |
| "loss": 0.8521, | |
| "step": 225 | |
| }, | |
| { | |
| "epoch": 0.78, | |
| "learning_rate": 4.94e-07, | |
| "loss": 0.7694, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.86, | |
| "learning_rate": 5.44e-07, | |
| "loss": 0.7196, | |
| "step": 275 | |
| }, | |
| { | |
| "epoch": 0.93, | |
| "learning_rate": 5.939999999999999e-07, | |
| "loss": 0.6902, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 1.01, | |
| "learning_rate": 6.44e-07, | |
| "loss": 0.6748, | |
| "step": 325 | |
| }, | |
| { | |
| "epoch": 1.09, | |
| "learning_rate": 6.939999999999999e-07, | |
| "loss": 0.6247, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 1.17, | |
| "learning_rate": 7.44e-07, | |
| "loss": 0.63, | |
| "step": 375 | |
| }, | |
| { | |
| "epoch": 1.25, | |
| "learning_rate": 7.94e-07, | |
| "loss": 0.5899, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 1.32, | |
| "learning_rate": 8.439999999999999e-07, | |
| "loss": 0.5628, | |
| "step": 425 | |
| }, | |
| { | |
| "epoch": 1.4, | |
| "learning_rate": 8.939999999999999e-07, | |
| "loss": 0.5634, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 1.48, | |
| "learning_rate": 9.439999999999999e-07, | |
| "loss": 0.5549, | |
| "step": 475 | |
| }, | |
| { | |
| "epoch": 1.56, | |
| "learning_rate": 9.94e-07, | |
| "loss": 0.5533, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.56, | |
| "eval_loss": 0.7044046521186829, | |
| "eval_runtime": 1265.1903, | |
| "eval_samples_per_second": 8.132, | |
| "eval_steps_per_second": 0.127, | |
| "eval_wer": 54.54987091580379, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.63, | |
| "learning_rate": 9.95111111111111e-07, | |
| "loss": 0.5329, | |
| "step": 525 | |
| }, | |
| { | |
| "epoch": 1.71, | |
| "learning_rate": 9.895555555555554e-07, | |
| "loss": 0.5116, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 1.79, | |
| "learning_rate": 9.84e-07, | |
| "loss": 0.511, | |
| "step": 575 | |
| }, | |
| { | |
| "epoch": 1.87, | |
| "learning_rate": 9.784444444444444e-07, | |
| "loss": 0.4963, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 1.95, | |
| "learning_rate": 9.728888888888888e-07, | |
| "loss": 0.5177, | |
| "step": 625 | |
| }, | |
| { | |
| "epoch": 2.02, | |
| "learning_rate": 9.673333333333332e-07, | |
| "loss": 0.4947, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 2.1, | |
| "learning_rate": 9.617777777777776e-07, | |
| "loss": 0.4529, | |
| "step": 675 | |
| }, | |
| { | |
| "epoch": 2.18, | |
| "learning_rate": 9.562222222222223e-07, | |
| "loss": 0.4543, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 2.26, | |
| "learning_rate": 9.506666666666667e-07, | |
| "loss": 0.4478, | |
| "step": 725 | |
| }, | |
| { | |
| "epoch": 2.34, | |
| "learning_rate": 9.451111111111111e-07, | |
| "loss": 0.4421, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 2.41, | |
| "learning_rate": 9.395555555555556e-07, | |
| "loss": 0.4465, | |
| "step": 775 | |
| }, | |
| { | |
| "epoch": 2.49, | |
| "learning_rate": 9.34e-07, | |
| "loss": 0.4164, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 2.57, | |
| "learning_rate": 9.284444444444444e-07, | |
| "loss": 0.4291, | |
| "step": 825 | |
| }, | |
| { | |
| "epoch": 2.65, | |
| "learning_rate": 9.228888888888888e-07, | |
| "loss": 0.4149, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 2.72, | |
| "learning_rate": 9.173333333333333e-07, | |
| "loss": 0.4242, | |
| "step": 875 | |
| }, | |
| { | |
| "epoch": 2.8, | |
| "learning_rate": 9.117777777777778e-07, | |
| "loss": 0.4329, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 2.88, | |
| "learning_rate": 9.062222222222222e-07, | |
| "loss": 0.4165, | |
| "step": 925 | |
| }, | |
| { | |
| "epoch": 2.96, | |
| "learning_rate": 9.006666666666666e-07, | |
| "loss": 0.4166, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 3.04, | |
| "learning_rate": 8.95111111111111e-07, | |
| "loss": 0.409, | |
| "step": 975 | |
| }, | |
| { | |
| "epoch": 3.12, | |
| "learning_rate": 8.895555555555555e-07, | |
| "loss": 0.3951, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 3.12, | |
| "eval_loss": 0.5893104076385498, | |
| "eval_runtime": 1326.5175, | |
| "eval_samples_per_second": 7.756, | |
| "eval_steps_per_second": 0.121, | |
| "eval_wer": 47.521028231962916, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 3.19, | |
| "learning_rate": 8.839999999999999e-07, | |
| "loss": 0.3808, | |
| "step": 1025 | |
| }, | |
| { | |
| "epoch": 3.27, | |
| "learning_rate": 8.784444444444444e-07, | |
| "loss": 0.3786, | |
| "step": 1050 | |
| }, | |
| { | |
| "epoch": 3.35, | |
| "learning_rate": 8.728888888888889e-07, | |
| "loss": 0.3811, | |
| "step": 1075 | |
| }, | |
| { | |
| "epoch": 3.43, | |
| "learning_rate": 8.673333333333332e-07, | |
| "loss": 0.3644, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 3.5, | |
| "learning_rate": 8.617777777777777e-07, | |
| "loss": 0.3737, | |
| "step": 1125 | |
| }, | |
| { | |
| "epoch": 3.58, | |
| "learning_rate": 8.562222222222222e-07, | |
| "loss": 0.3698, | |
| "step": 1150 | |
| }, | |
| { | |
| "epoch": 3.66, | |
| "learning_rate": 8.506666666666667e-07, | |
| "loss": 0.3601, | |
| "step": 1175 | |
| }, | |
| { | |
| "epoch": 3.74, | |
| "learning_rate": 8.451111111111111e-07, | |
| "loss": 0.3675, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 3.81, | |
| "learning_rate": 8.395555555555556e-07, | |
| "loss": 0.3588, | |
| "step": 1225 | |
| }, | |
| { | |
| "epoch": 3.89, | |
| "learning_rate": 8.34e-07, | |
| "loss": 0.3541, | |
| "step": 1250 | |
| }, | |
| { | |
| "epoch": 3.97, | |
| "learning_rate": 8.284444444444444e-07, | |
| "loss": 0.3537, | |
| "step": 1275 | |
| }, | |
| { | |
| "epoch": 4.05, | |
| "learning_rate": 8.228888888888889e-07, | |
| "loss": 0.3666, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 4.13, | |
| "learning_rate": 8.173333333333333e-07, | |
| "loss": 0.3242, | |
| "step": 1325 | |
| }, | |
| { | |
| "epoch": 4.21, | |
| "learning_rate": 8.117777777777778e-07, | |
| "loss": 0.3212, | |
| "step": 1350 | |
| }, | |
| { | |
| "epoch": 4.28, | |
| "learning_rate": 8.062222222222221e-07, | |
| "loss": 0.3374, | |
| "step": 1375 | |
| }, | |
| { | |
| "epoch": 4.36, | |
| "learning_rate": 8.006666666666666e-07, | |
| "loss": 0.3259, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 4.44, | |
| "learning_rate": 7.95111111111111e-07, | |
| "loss": 0.3361, | |
| "step": 1425 | |
| }, | |
| { | |
| "epoch": 4.52, | |
| "learning_rate": 7.895555555555555e-07, | |
| "loss": 0.3355, | |
| "step": 1450 | |
| }, | |
| { | |
| "epoch": 4.59, | |
| "learning_rate": 7.84e-07, | |
| "loss": 0.3254, | |
| "step": 1475 | |
| }, | |
| { | |
| "epoch": 4.67, | |
| "learning_rate": 7.784444444444444e-07, | |
| "loss": 0.3296, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 4.67, | |
| "eval_loss": 0.5428555011749268, | |
| "eval_runtime": 1299.6224, | |
| "eval_samples_per_second": 7.916, | |
| "eval_steps_per_second": 0.124, | |
| "eval_wer": 44.64508786053355, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 4.75, | |
| "learning_rate": 7.728888888888888e-07, | |
| "loss": 0.3229, | |
| "step": 1525 | |
| }, | |
| { | |
| "epoch": 4.83, | |
| "learning_rate": 7.673333333333332e-07, | |
| "loss": 0.3372, | |
| "step": 1550 | |
| }, | |
| { | |
| "epoch": 4.91, | |
| "learning_rate": 7.617777777777778e-07, | |
| "loss": 0.3297, | |
| "step": 1575 | |
| }, | |
| { | |
| "epoch": 4.98, | |
| "learning_rate": 7.562222222222222e-07, | |
| "loss": 0.3239, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 5.06, | |
| "learning_rate": 7.506666666666667e-07, | |
| "loss": 0.3113, | |
| "step": 1625 | |
| }, | |
| { | |
| "epoch": 5.14, | |
| "learning_rate": 7.451111111111111e-07, | |
| "loss": 0.3142, | |
| "step": 1650 | |
| }, | |
| { | |
| "epoch": 5.22, | |
| "learning_rate": 7.395555555555555e-07, | |
| "loss": 0.303, | |
| "step": 1675 | |
| }, | |
| { | |
| "epoch": 5.3, | |
| "learning_rate": 7.34e-07, | |
| "loss": 0.2973, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 5.37, | |
| "learning_rate": 7.284444444444444e-07, | |
| "loss": 0.2998, | |
| "step": 1725 | |
| }, | |
| { | |
| "epoch": 5.45, | |
| "learning_rate": 7.228888888888889e-07, | |
| "loss": 0.2962, | |
| "step": 1750 | |
| }, | |
| { | |
| "epoch": 5.53, | |
| "learning_rate": 7.173333333333333e-07, | |
| "loss": 0.3007, | |
| "step": 1775 | |
| }, | |
| { | |
| "epoch": 5.61, | |
| "learning_rate": 7.117777777777777e-07, | |
| "loss": 0.2906, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 5.68, | |
| "learning_rate": 7.062222222222222e-07, | |
| "loss": 0.295, | |
| "step": 1825 | |
| }, | |
| { | |
| "epoch": 5.76, | |
| "learning_rate": 7.006666666666666e-07, | |
| "loss": 0.3083, | |
| "step": 1850 | |
| }, | |
| { | |
| "epoch": 5.84, | |
| "learning_rate": 6.951111111111111e-07, | |
| "loss": 0.2869, | |
| "step": 1875 | |
| }, | |
| { | |
| "epoch": 5.92, | |
| "learning_rate": 6.895555555555555e-07, | |
| "loss": 0.2936, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "learning_rate": 6.84e-07, | |
| "loss": 0.2982, | |
| "step": 1925 | |
| }, | |
| { | |
| "epoch": 6.07, | |
| "learning_rate": 6.784444444444443e-07, | |
| "loss": 0.2885, | |
| "step": 1950 | |
| }, | |
| { | |
| "epoch": 6.15, | |
| "learning_rate": 6.728888888888888e-07, | |
| "loss": 0.2719, | |
| "step": 1975 | |
| }, | |
| { | |
| "epoch": 6.23, | |
| "learning_rate": 6.673333333333334e-07, | |
| "loss": 0.2662, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 6.23, | |
| "eval_loss": 0.5223153829574585, | |
| "eval_runtime": 1347.3336, | |
| "eval_samples_per_second": 7.636, | |
| "eval_steps_per_second": 0.119, | |
| "eval_wer": 42.664408849901456, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 6.31, | |
| "learning_rate": 6.617777777777778e-07, | |
| "loss": 0.2739, | |
| "step": 2025 | |
| }, | |
| { | |
| "epoch": 6.39, | |
| "learning_rate": 6.562222222222223e-07, | |
| "loss": 0.2694, | |
| "step": 2050 | |
| }, | |
| { | |
| "epoch": 6.46, | |
| "learning_rate": 6.506666666666666e-07, | |
| "loss": 0.2678, | |
| "step": 2075 | |
| }, | |
| { | |
| "epoch": 6.54, | |
| "learning_rate": 6.451111111111111e-07, | |
| "loss": 0.2716, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 6.62, | |
| "learning_rate": 6.395555555555555e-07, | |
| "loss": 0.27, | |
| "step": 2125 | |
| }, | |
| { | |
| "epoch": 6.7, | |
| "learning_rate": 6.34e-07, | |
| "loss": 0.2715, | |
| "step": 2150 | |
| }, | |
| { | |
| "epoch": 6.77, | |
| "learning_rate": 6.284444444444445e-07, | |
| "loss": 0.2705, | |
| "step": 2175 | |
| }, | |
| { | |
| "epoch": 6.85, | |
| "learning_rate": 6.228888888888889e-07, | |
| "loss": 0.2771, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 6.93, | |
| "learning_rate": 6.173333333333333e-07, | |
| "loss": 0.2814, | |
| "step": 2225 | |
| }, | |
| { | |
| "epoch": 7.01, | |
| "learning_rate": 6.117777777777777e-07, | |
| "loss": 0.302, | |
| "step": 2250 | |
| }, | |
| { | |
| "epoch": 7.09, | |
| "learning_rate": 6.062222222222222e-07, | |
| "loss": 0.2619, | |
| "step": 2275 | |
| }, | |
| { | |
| "epoch": 7.16, | |
| "learning_rate": 6.006666666666666e-07, | |
| "loss": 0.2561, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 7.24, | |
| "learning_rate": 5.951111111111111e-07, | |
| "loss": 0.2559, | |
| "step": 2325 | |
| }, | |
| { | |
| "epoch": 7.32, | |
| "learning_rate": 5.895555555555555e-07, | |
| "loss": 0.2524, | |
| "step": 2350 | |
| }, | |
| { | |
| "epoch": 7.4, | |
| "learning_rate": 5.839999999999999e-07, | |
| "loss": 0.2583, | |
| "step": 2375 | |
| }, | |
| { | |
| "epoch": 7.48, | |
| "learning_rate": 5.784444444444444e-07, | |
| "loss": 0.2434, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 7.55, | |
| "learning_rate": 5.728888888888888e-07, | |
| "loss": 0.2546, | |
| "step": 2425 | |
| }, | |
| { | |
| "epoch": 7.63, | |
| "learning_rate": 5.673333333333334e-07, | |
| "loss": 0.2492, | |
| "step": 2450 | |
| }, | |
| { | |
| "epoch": 7.71, | |
| "learning_rate": 5.617777777777778e-07, | |
| "loss": 0.2544, | |
| "step": 2475 | |
| }, | |
| { | |
| "epoch": 7.79, | |
| "learning_rate": 5.562222222222222e-07, | |
| "loss": 0.2535, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 7.79, | |
| "eval_loss": 0.5044606924057007, | |
| "eval_runtime": 1321.646, | |
| "eval_samples_per_second": 7.784, | |
| "eval_steps_per_second": 0.122, | |
| "eval_wer": 41.53041112622491, | |
| "step": 2500 | |
| } | |
| ], | |
| "max_steps": 5000, | |
| "num_train_epochs": 16, | |
| "total_flos": 4.623830749703412e+19, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |