|
{ |
|
"best_metric": 0.0693531029534856, |
|
"best_model_checkpoint": "./wav2vec2-xls-r-300m-ar/checkpoint-3000", |
|
"epoch": 26.548672566371682, |
|
"global_step": 3000, |
|
"is_hyper_param_search": false, |
|
"is_local_process_zero": true, |
|
"is_world_process_zero": true, |
|
"log_history": [ |
|
{ |
|
"epoch": 0.22, |
|
"learning_rate": 1.4999999999999999e-05, |
|
"loss": 9.1248, |
|
"step": 25 |
|
}, |
|
{ |
|
"epoch": 0.44, |
|
"learning_rate": 2.9999999999999997e-05, |
|
"loss": 6.6739, |
|
"step": 50 |
|
}, |
|
{ |
|
"epoch": 0.66, |
|
"learning_rate": 4.4999999999999996e-05, |
|
"loss": 4.632, |
|
"step": 75 |
|
}, |
|
{ |
|
"epoch": 0.88, |
|
"learning_rate": 5.9999999999999995e-05, |
|
"loss": 3.917, |
|
"step": 100 |
|
}, |
|
{ |
|
"epoch": 1.11, |
|
"learning_rate": 7.5e-05, |
|
"loss": 3.3449, |
|
"step": 125 |
|
}, |
|
{ |
|
"epoch": 1.33, |
|
"learning_rate": 8.999999999999999e-05, |
|
"loss": 2.8486, |
|
"step": 150 |
|
}, |
|
{ |
|
"epoch": 1.55, |
|
"learning_rate": 0.00010499999999999999, |
|
"loss": 2.5717, |
|
"step": 175 |
|
}, |
|
{ |
|
"epoch": 1.77, |
|
"learning_rate": 0.00011999999999999999, |
|
"loss": 2.3746, |
|
"step": 200 |
|
}, |
|
{ |
|
"epoch": 1.99, |
|
"learning_rate": 0.000135, |
|
"loss": 2.2744, |
|
"step": 225 |
|
}, |
|
{ |
|
"epoch": 2.21, |
|
"learning_rate": 0.00015, |
|
"loss": 2.0759, |
|
"step": 250 |
|
}, |
|
{ |
|
"epoch": 2.43, |
|
"learning_rate": 0.000165, |
|
"loss": 1.7813, |
|
"step": 275 |
|
}, |
|
{ |
|
"epoch": 2.65, |
|
"learning_rate": 0.00017999999999999998, |
|
"loss": 1.2002, |
|
"step": 300 |
|
}, |
|
{ |
|
"epoch": 2.88, |
|
"learning_rate": 0.000195, |
|
"loss": 0.7684, |
|
"step": 325 |
|
}, |
|
{ |
|
"epoch": 3.1, |
|
"learning_rate": 0.00020999999999999998, |
|
"loss": 0.527, |
|
"step": 350 |
|
}, |
|
{ |
|
"epoch": 3.32, |
|
"learning_rate": 0.000225, |
|
"loss": 0.3734, |
|
"step": 375 |
|
}, |
|
{ |
|
"epoch": 3.54, |
|
"learning_rate": 0.00023999999999999998, |
|
"loss": 0.3184, |
|
"step": 400 |
|
}, |
|
{ |
|
"epoch": 3.76, |
|
"learning_rate": 0.00025499999999999996, |
|
"loss": 0.2822, |
|
"step": 425 |
|
}, |
|
{ |
|
"epoch": 3.98, |
|
"learning_rate": 0.00027, |
|
"loss": 0.2074, |
|
"step": 450 |
|
}, |
|
{ |
|
"epoch": 4.2, |
|
"learning_rate": 0.000285, |
|
"loss": 0.1985, |
|
"step": 475 |
|
}, |
|
{ |
|
"epoch": 4.42, |
|
"learning_rate": 0.0003, |
|
"loss": 0.1917, |
|
"step": 500 |
|
}, |
|
{ |
|
"epoch": 4.65, |
|
"learning_rate": 0.00029921052631578946, |
|
"loss": 0.1853, |
|
"step": 525 |
|
}, |
|
{ |
|
"epoch": 4.87, |
|
"learning_rate": 0.00029842105263157894, |
|
"loss": 0.1828, |
|
"step": 550 |
|
}, |
|
{ |
|
"epoch": 5.09, |
|
"learning_rate": 0.00029763157894736837, |
|
"loss": 0.1715, |
|
"step": 575 |
|
}, |
|
{ |
|
"epoch": 5.31, |
|
"learning_rate": 0.00029684210526315785, |
|
"loss": 0.1457, |
|
"step": 600 |
|
}, |
|
{ |
|
"epoch": 5.53, |
|
"learning_rate": 0.00029605263157894733, |
|
"loss": 0.1247, |
|
"step": 625 |
|
}, |
|
{ |
|
"epoch": 5.75, |
|
"learning_rate": 0.0002952631578947368, |
|
"loss": 0.1266, |
|
"step": 650 |
|
}, |
|
{ |
|
"epoch": 5.97, |
|
"learning_rate": 0.0002944736842105263, |
|
"loss": 0.117, |
|
"step": 675 |
|
}, |
|
{ |
|
"epoch": 6.19, |
|
"learning_rate": 0.0002936842105263158, |
|
"loss": 0.117, |
|
"step": 700 |
|
}, |
|
{ |
|
"epoch": 6.42, |
|
"learning_rate": 0.00029289473684210527, |
|
"loss": 0.125, |
|
"step": 725 |
|
}, |
|
{ |
|
"epoch": 6.64, |
|
"learning_rate": 0.0002921052631578947, |
|
"loss": 0.1084, |
|
"step": 750 |
|
}, |
|
{ |
|
"epoch": 6.86, |
|
"learning_rate": 0.0002913157894736842, |
|
"loss": 0.1178, |
|
"step": 775 |
|
}, |
|
{ |
|
"epoch": 7.08, |
|
"learning_rate": 0.00029052631578947366, |
|
"loss": 0.1193, |
|
"step": 800 |
|
}, |
|
{ |
|
"epoch": 7.3, |
|
"learning_rate": 0.00028973684210526314, |
|
"loss": 0.0925, |
|
"step": 825 |
|
}, |
|
{ |
|
"epoch": 7.52, |
|
"learning_rate": 0.00028894736842105263, |
|
"loss": 0.085, |
|
"step": 850 |
|
}, |
|
{ |
|
"epoch": 7.74, |
|
"learning_rate": 0.00028815789473684206, |
|
"loss": 0.0865, |
|
"step": 875 |
|
}, |
|
{ |
|
"epoch": 7.96, |
|
"learning_rate": 0.00028736842105263154, |
|
"loss": 0.0811, |
|
"step": 900 |
|
}, |
|
{ |
|
"epoch": 8.19, |
|
"learning_rate": 0.000286578947368421, |
|
"loss": 0.0835, |
|
"step": 925 |
|
}, |
|
{ |
|
"epoch": 8.41, |
|
"learning_rate": 0.0002857894736842105, |
|
"loss": 0.0886, |
|
"step": 950 |
|
}, |
|
{ |
|
"epoch": 8.63, |
|
"learning_rate": 0.000285, |
|
"loss": 0.0991, |
|
"step": 975 |
|
}, |
|
{ |
|
"epoch": 8.85, |
|
"learning_rate": 0.0002842105263157894, |
|
"loss": 0.0793, |
|
"step": 1000 |
|
}, |
|
{ |
|
"epoch": 8.85, |
|
"eval_loss": 0.16262491047382355, |
|
"eval_runtime": 11.6481, |
|
"eval_samples_per_second": 8.585, |
|
"eval_steps_per_second": 1.116, |
|
"eval_wer": 0.07856032524213799, |
|
"step": 1000 |
|
}, |
|
{ |
|
"epoch": 9.07, |
|
"learning_rate": 0.00028342105263157895, |
|
"loss": 0.0863, |
|
"step": 1025 |
|
}, |
|
{ |
|
"epoch": 9.29, |
|
"learning_rate": 0.0002826315789473684, |
|
"loss": 0.0638, |
|
"step": 1050 |
|
}, |
|
{ |
|
"epoch": 9.51, |
|
"learning_rate": 0.00028184210526315787, |
|
"loss": 0.0604, |
|
"step": 1075 |
|
}, |
|
{ |
|
"epoch": 9.73, |
|
"learning_rate": 0.00028105263157894735, |
|
"loss": 0.0621, |
|
"step": 1100 |
|
}, |
|
{ |
|
"epoch": 9.96, |
|
"learning_rate": 0.00028026315789473683, |
|
"loss": 0.0707, |
|
"step": 1125 |
|
}, |
|
{ |
|
"epoch": 10.18, |
|
"learning_rate": 0.0002794736842105263, |
|
"loss": 0.0799, |
|
"step": 1150 |
|
}, |
|
{ |
|
"epoch": 10.4, |
|
"learning_rate": 0.00027868421052631574, |
|
"loss": 0.0694, |
|
"step": 1175 |
|
}, |
|
{ |
|
"epoch": 10.62, |
|
"learning_rate": 0.00027789473684210523, |
|
"loss": 0.0655, |
|
"step": 1200 |
|
}, |
|
{ |
|
"epoch": 10.84, |
|
"learning_rate": 0.0002771052631578947, |
|
"loss": 0.0776, |
|
"step": 1225 |
|
}, |
|
{ |
|
"epoch": 11.06, |
|
"learning_rate": 0.0002763157894736842, |
|
"loss": 0.0582, |
|
"step": 1250 |
|
}, |
|
{ |
|
"epoch": 11.28, |
|
"learning_rate": 0.0002755263157894737, |
|
"loss": 0.0547, |
|
"step": 1275 |
|
}, |
|
{ |
|
"epoch": 11.5, |
|
"learning_rate": 0.0002747368421052631, |
|
"loss": 0.0597, |
|
"step": 1300 |
|
}, |
|
{ |
|
"epoch": 11.73, |
|
"learning_rate": 0.0002739473684210526, |
|
"loss": 0.0492, |
|
"step": 1325 |
|
}, |
|
{ |
|
"epoch": 11.95, |
|
"learning_rate": 0.00027315789473684207, |
|
"loss": 0.0699, |
|
"step": 1350 |
|
}, |
|
{ |
|
"epoch": 12.17, |
|
"learning_rate": 0.00027236842105263155, |
|
"loss": 0.0594, |
|
"step": 1375 |
|
}, |
|
{ |
|
"epoch": 12.39, |
|
"learning_rate": 0.00027157894736842104, |
|
"loss": 0.0541, |
|
"step": 1400 |
|
}, |
|
{ |
|
"epoch": 12.61, |
|
"learning_rate": 0.0002707894736842105, |
|
"loss": 0.0522, |
|
"step": 1425 |
|
}, |
|
{ |
|
"epoch": 12.83, |
|
"learning_rate": 0.00027, |
|
"loss": 0.0645, |
|
"step": 1450 |
|
}, |
|
{ |
|
"epoch": 13.05, |
|
"learning_rate": 0.00026921052631578943, |
|
"loss": 0.0583, |
|
"step": 1475 |
|
}, |
|
{ |
|
"epoch": 13.27, |
|
"learning_rate": 0.0002684210526315789, |
|
"loss": 0.0443, |
|
"step": 1500 |
|
}, |
|
{ |
|
"epoch": 13.5, |
|
"learning_rate": 0.0002676315789473684, |
|
"loss": 0.0548, |
|
"step": 1525 |
|
}, |
|
{ |
|
"epoch": 13.72, |
|
"learning_rate": 0.0002668421052631579, |
|
"loss": 0.0484, |
|
"step": 1550 |
|
}, |
|
{ |
|
"epoch": 13.94, |
|
"learning_rate": 0.00026605263157894736, |
|
"loss": 0.0775, |
|
"step": 1575 |
|
}, |
|
{ |
|
"epoch": 14.16, |
|
"learning_rate": 0.0002652631578947368, |
|
"loss": 0.0554, |
|
"step": 1600 |
|
}, |
|
{ |
|
"epoch": 14.38, |
|
"learning_rate": 0.0002644736842105263, |
|
"loss": 0.0456, |
|
"step": 1625 |
|
}, |
|
{ |
|
"epoch": 14.6, |
|
"learning_rate": 0.00026368421052631576, |
|
"loss": 0.0563, |
|
"step": 1650 |
|
}, |
|
{ |
|
"epoch": 14.82, |
|
"learning_rate": 0.00026289473684210524, |
|
"loss": 0.0493, |
|
"step": 1675 |
|
}, |
|
{ |
|
"epoch": 15.04, |
|
"learning_rate": 0.0002621052631578947, |
|
"loss": 0.0452, |
|
"step": 1700 |
|
}, |
|
{ |
|
"epoch": 15.27, |
|
"learning_rate": 0.0002613157894736842, |
|
"loss": 0.0389, |
|
"step": 1725 |
|
}, |
|
{ |
|
"epoch": 15.49, |
|
"learning_rate": 0.0002605263157894737, |
|
"loss": 0.0436, |
|
"step": 1750 |
|
}, |
|
{ |
|
"epoch": 15.71, |
|
"learning_rate": 0.0002597368421052631, |
|
"loss": 0.0407, |
|
"step": 1775 |
|
}, |
|
{ |
|
"epoch": 15.93, |
|
"learning_rate": 0.0002589473684210526, |
|
"loss": 0.0448, |
|
"step": 1800 |
|
}, |
|
{ |
|
"epoch": 16.15, |
|
"learning_rate": 0.0002581578947368421, |
|
"loss": 0.0457, |
|
"step": 1825 |
|
}, |
|
{ |
|
"epoch": 16.37, |
|
"learning_rate": 0.00025736842105263157, |
|
"loss": 0.0461, |
|
"step": 1850 |
|
}, |
|
{ |
|
"epoch": 16.59, |
|
"learning_rate": 0.00025657894736842105, |
|
"loss": 0.0452, |
|
"step": 1875 |
|
}, |
|
{ |
|
"epoch": 16.81, |
|
"learning_rate": 0.0002557894736842105, |
|
"loss": 0.0484, |
|
"step": 1900 |
|
}, |
|
{ |
|
"epoch": 17.04, |
|
"learning_rate": 0.00025499999999999996, |
|
"loss": 0.0376, |
|
"step": 1925 |
|
}, |
|
{ |
|
"epoch": 17.26, |
|
"learning_rate": 0.00025421052631578945, |
|
"loss": 0.0409, |
|
"step": 1950 |
|
}, |
|
{ |
|
"epoch": 17.48, |
|
"learning_rate": 0.00025342105263157893, |
|
"loss": 0.0315, |
|
"step": 1975 |
|
}, |
|
{ |
|
"epoch": 17.7, |
|
"learning_rate": 0.00025263157894736836, |
|
"loss": 0.0396, |
|
"step": 2000 |
|
}, |
|
{ |
|
"epoch": 17.7, |
|
"eval_loss": 0.21985749900341034, |
|
"eval_runtime": 11.0577, |
|
"eval_samples_per_second": 9.043, |
|
"eval_steps_per_second": 1.176, |
|
"eval_wer": 0.08071266292000479, |
|
"step": 2000 |
|
}, |
|
{ |
|
"epoch": 17.92, |
|
"learning_rate": 0.00025184210526315784, |
|
"loss": 0.0406, |
|
"step": 2025 |
|
}, |
|
{ |
|
"epoch": 18.14, |
|
"learning_rate": 0.0002510526315789474, |
|
"loss": 0.0322, |
|
"step": 2050 |
|
}, |
|
{ |
|
"epoch": 18.36, |
|
"learning_rate": 0.0002502631578947368, |
|
"loss": 0.0393, |
|
"step": 2075 |
|
}, |
|
{ |
|
"epoch": 18.58, |
|
"learning_rate": 0.0002494736842105263, |
|
"loss": 0.0438, |
|
"step": 2100 |
|
}, |
|
{ |
|
"epoch": 18.81, |
|
"learning_rate": 0.0002486842105263158, |
|
"loss": 0.0372, |
|
"step": 2125 |
|
}, |
|
{ |
|
"epoch": 19.03, |
|
"learning_rate": 0.00024789473684210526, |
|
"loss": 0.0335, |
|
"step": 2150 |
|
}, |
|
{ |
|
"epoch": 19.25, |
|
"learning_rate": 0.00024710526315789474, |
|
"loss": 0.0327, |
|
"step": 2175 |
|
}, |
|
{ |
|
"epoch": 19.47, |
|
"learning_rate": 0.00024631578947368417, |
|
"loss": 0.0307, |
|
"step": 2200 |
|
}, |
|
{ |
|
"epoch": 19.69, |
|
"learning_rate": 0.00024552631578947365, |
|
"loss": 0.0374, |
|
"step": 2225 |
|
}, |
|
{ |
|
"epoch": 19.91, |
|
"learning_rate": 0.00024473684210526314, |
|
"loss": 0.035, |
|
"step": 2250 |
|
}, |
|
{ |
|
"epoch": 20.13, |
|
"learning_rate": 0.0002439473684210526, |
|
"loss": 0.0336, |
|
"step": 2275 |
|
}, |
|
{ |
|
"epoch": 20.35, |
|
"learning_rate": 0.00024315789473684207, |
|
"loss": 0.0426, |
|
"step": 2300 |
|
}, |
|
{ |
|
"epoch": 20.58, |
|
"learning_rate": 0.00024236842105263156, |
|
"loss": 0.0375, |
|
"step": 2325 |
|
}, |
|
{ |
|
"epoch": 20.8, |
|
"learning_rate": 0.000241578947368421, |
|
"loss": 0.0352, |
|
"step": 2350 |
|
}, |
|
{ |
|
"epoch": 21.02, |
|
"learning_rate": 0.00024078947368421052, |
|
"loss": 0.0319, |
|
"step": 2375 |
|
}, |
|
{ |
|
"epoch": 21.24, |
|
"learning_rate": 0.00023999999999999998, |
|
"loss": 0.0339, |
|
"step": 2400 |
|
}, |
|
{ |
|
"epoch": 21.46, |
|
"learning_rate": 0.00023921052631578946, |
|
"loss": 0.0335, |
|
"step": 2425 |
|
}, |
|
{ |
|
"epoch": 21.68, |
|
"learning_rate": 0.00023842105263157895, |
|
"loss": 0.0329, |
|
"step": 2450 |
|
}, |
|
{ |
|
"epoch": 21.9, |
|
"learning_rate": 0.0002376315789473684, |
|
"loss": 0.0373, |
|
"step": 2475 |
|
}, |
|
{ |
|
"epoch": 22.12, |
|
"learning_rate": 0.00023684210526315788, |
|
"loss": 0.0371, |
|
"step": 2500 |
|
}, |
|
{ |
|
"epoch": 22.35, |
|
"learning_rate": 0.00023605263157894734, |
|
"loss": 0.0273, |
|
"step": 2525 |
|
}, |
|
{ |
|
"epoch": 22.57, |
|
"learning_rate": 0.00023526315789473682, |
|
"loss": 0.0361, |
|
"step": 2550 |
|
}, |
|
{ |
|
"epoch": 22.79, |
|
"learning_rate": 0.00023447368421052628, |
|
"loss": 0.0367, |
|
"step": 2575 |
|
}, |
|
{ |
|
"epoch": 23.01, |
|
"learning_rate": 0.00023368421052631576, |
|
"loss": 0.0374, |
|
"step": 2600 |
|
}, |
|
{ |
|
"epoch": 23.23, |
|
"learning_rate": 0.00023289473684210524, |
|
"loss": 0.0264, |
|
"step": 2625 |
|
}, |
|
{ |
|
"epoch": 23.45, |
|
"learning_rate": 0.0002321052631578947, |
|
"loss": 0.0331, |
|
"step": 2650 |
|
}, |
|
{ |
|
"epoch": 23.67, |
|
"learning_rate": 0.00023131578947368418, |
|
"loss": 0.0407, |
|
"step": 2675 |
|
}, |
|
{ |
|
"epoch": 23.89, |
|
"learning_rate": 0.00023052631578947364, |
|
"loss": 0.031, |
|
"step": 2700 |
|
}, |
|
{ |
|
"epoch": 24.12, |
|
"learning_rate": 0.00022973684210526315, |
|
"loss": 0.0294, |
|
"step": 2725 |
|
}, |
|
{ |
|
"epoch": 24.34, |
|
"learning_rate": 0.00022894736842105263, |
|
"loss": 0.0345, |
|
"step": 2750 |
|
}, |
|
{ |
|
"epoch": 24.56, |
|
"learning_rate": 0.0002281578947368421, |
|
"loss": 0.027, |
|
"step": 2775 |
|
}, |
|
{ |
|
"epoch": 24.78, |
|
"learning_rate": 0.00022736842105263157, |
|
"loss": 0.0241, |
|
"step": 2800 |
|
}, |
|
{ |
|
"epoch": 25.0, |
|
"learning_rate": 0.00022657894736842103, |
|
"loss": 0.0304, |
|
"step": 2825 |
|
}, |
|
{ |
|
"epoch": 25.22, |
|
"learning_rate": 0.0002257894736842105, |
|
"loss": 0.026, |
|
"step": 2850 |
|
}, |
|
{ |
|
"epoch": 25.44, |
|
"learning_rate": 0.000225, |
|
"loss": 0.0325, |
|
"step": 2875 |
|
}, |
|
{ |
|
"epoch": 25.66, |
|
"learning_rate": 0.00022421052631578945, |
|
"loss": 0.0374, |
|
"step": 2900 |
|
}, |
|
{ |
|
"epoch": 25.88, |
|
"learning_rate": 0.00022342105263157893, |
|
"loss": 0.036, |
|
"step": 2925 |
|
}, |
|
{ |
|
"epoch": 26.11, |
|
"learning_rate": 0.0002226315789473684, |
|
"loss": 0.0339, |
|
"step": 2950 |
|
}, |
|
{ |
|
"epoch": 26.33, |
|
"learning_rate": 0.00022184210526315787, |
|
"loss": 0.0312, |
|
"step": 2975 |
|
}, |
|
{ |
|
"epoch": 26.55, |
|
"learning_rate": 0.00022105263157894733, |
|
"loss": 0.0285, |
|
"step": 3000 |
|
}, |
|
{ |
|
"epoch": 26.55, |
|
"eval_loss": 0.22891539335250854, |
|
"eval_runtime": 10.8646, |
|
"eval_samples_per_second": 9.204, |
|
"eval_steps_per_second": 1.197, |
|
"eval_wer": 0.0693531029534856, |
|
"step": 3000 |
|
} |
|
], |
|
"max_steps": 10000, |
|
"num_train_epochs": 89, |
|
"total_flos": 1.1850399847250571e+19, |
|
"trial_name": null, |
|
"trial_params": null |
|
} |
|
|