whisper-tiny-it-12 / checkpoint-1000 /trainer_state.json
GIanlucaRub's picture
Copied from Google Drive
158c81a
{
"best_metric": 48.98779811425402,
"best_model_checkpoint": "./whisper-tiny-it-12/checkpoint-1000",
"epoch": 0.3818251240931653,
"global_step": 1000,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.01,
"learning_rate": 4.0000000000000003e-07,
"loss": 3.7311,
"step": 25
},
{
"epoch": 0.02,
"learning_rate": 9.000000000000001e-07,
"loss": 3.2401,
"step": 50
},
{
"epoch": 0.03,
"learning_rate": 1.4000000000000001e-06,
"loss": 2.4666,
"step": 75
},
{
"epoch": 0.04,
"learning_rate": 1.9000000000000002e-06,
"loss": 1.7924,
"step": 100
},
{
"epoch": 0.05,
"learning_rate": 2.4000000000000003e-06,
"loss": 1.2235,
"step": 125
},
{
"epoch": 0.06,
"learning_rate": 2.9e-06,
"loss": 0.9162,
"step": 150
},
{
"epoch": 0.07,
"learning_rate": 3.4000000000000005e-06,
"loss": 0.8499,
"step": 175
},
{
"epoch": 0.08,
"learning_rate": 3.900000000000001e-06,
"loss": 0.8388,
"step": 200
},
{
"epoch": 0.09,
"learning_rate": 4.4e-06,
"loss": 0.7834,
"step": 225
},
{
"epoch": 0.1,
"learning_rate": 4.9000000000000005e-06,
"loss": 0.7666,
"step": 250
},
{
"epoch": 0.11,
"learning_rate": 5.400000000000001e-06,
"loss": 0.7621,
"step": 275
},
{
"epoch": 0.11,
"learning_rate": 5.9e-06,
"loss": 0.7067,
"step": 300
},
{
"epoch": 0.12,
"learning_rate": 6.4000000000000006e-06,
"loss": 0.745,
"step": 325
},
{
"epoch": 0.13,
"learning_rate": 6.9e-06,
"loss": 0.6709,
"step": 350
},
{
"epoch": 0.14,
"learning_rate": 7.4e-06,
"loss": 0.6614,
"step": 375
},
{
"epoch": 0.15,
"learning_rate": 7.9e-06,
"loss": 0.6952,
"step": 400
},
{
"epoch": 0.16,
"learning_rate": 8.400000000000001e-06,
"loss": 0.6711,
"step": 425
},
{
"epoch": 0.17,
"learning_rate": 8.900000000000001e-06,
"loss": 0.674,
"step": 450
},
{
"epoch": 0.18,
"learning_rate": 9.4e-06,
"loss": 0.631,
"step": 475
},
{
"epoch": 0.19,
"learning_rate": 9.9e-06,
"loss": 0.6322,
"step": 500
},
{
"epoch": 0.2,
"learning_rate": 9.942857142857145e-06,
"loss": 0.6427,
"step": 525
},
{
"epoch": 0.21,
"learning_rate": 9.871428571428572e-06,
"loss": 0.6419,
"step": 550
},
{
"epoch": 0.22,
"learning_rate": 9.800000000000001e-06,
"loss": 0.6484,
"step": 575
},
{
"epoch": 0.23,
"learning_rate": 9.72857142857143e-06,
"loss": 0.6228,
"step": 600
},
{
"epoch": 0.24,
"learning_rate": 9.657142857142859e-06,
"loss": 0.6652,
"step": 625
},
{
"epoch": 0.25,
"learning_rate": 9.585714285714286e-06,
"loss": 0.6331,
"step": 650
},
{
"epoch": 0.26,
"learning_rate": 9.514285714285715e-06,
"loss": 0.6281,
"step": 675
},
{
"epoch": 0.27,
"learning_rate": 9.442857142857144e-06,
"loss": 0.6164,
"step": 700
},
{
"epoch": 0.28,
"learning_rate": 9.371428571428572e-06,
"loss": 0.5798,
"step": 725
},
{
"epoch": 0.29,
"learning_rate": 9.3e-06,
"loss": 0.6095,
"step": 750
},
{
"epoch": 0.3,
"learning_rate": 9.22857142857143e-06,
"loss": 0.5941,
"step": 775
},
{
"epoch": 0.31,
"learning_rate": 9.157142857142857e-06,
"loss": 0.5803,
"step": 800
},
{
"epoch": 0.32,
"learning_rate": 9.085714285714286e-06,
"loss": 0.5709,
"step": 825
},
{
"epoch": 0.32,
"learning_rate": 9.014285714285715e-06,
"loss": 0.606,
"step": 850
},
{
"epoch": 0.33,
"learning_rate": 8.942857142857142e-06,
"loss": 0.5946,
"step": 875
},
{
"epoch": 0.34,
"learning_rate": 8.871428571428571e-06,
"loss": 0.5791,
"step": 900
},
{
"epoch": 0.35,
"learning_rate": 8.8e-06,
"loss": 0.6073,
"step": 925
},
{
"epoch": 0.36,
"learning_rate": 8.72857142857143e-06,
"loss": 0.5667,
"step": 950
},
{
"epoch": 0.37,
"learning_rate": 8.657142857142858e-06,
"loss": 0.5819,
"step": 975
},
{
"epoch": 0.38,
"learning_rate": 8.585714285714286e-06,
"loss": 0.5844,
"step": 1000
},
{
"epoch": 0.38,
"eval_loss": 0.8011854887008667,
"eval_runtime": 332.5485,
"eval_samples_per_second": 4.511,
"eval_steps_per_second": 0.565,
"eval_wer": 48.98779811425402,
"step": 1000
}
],
"max_steps": 4000,
"num_train_epochs": 2,
"total_flos": 3.9390216192e+17,
"trial_name": null,
"trial_params": null
}