elsayedissa's picture
Training in progress, step 1000
a11d12e
raw
history blame
5.36 kB
{
"best_metric": 0.78878255063432,
"best_model_checkpoint": "/storage/elsayedissa/whisper-large-v2-japanese-24h/checkpoint-1000",
"epoch": 7.633587786259542,
"global_step": 1000,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.19,
"learning_rate": 4.6000000000000004e-07,
"loss": 0.3048,
"step": 25
},
{
"epoch": 0.38,
"learning_rate": 9.600000000000001e-07,
"loss": 0.2896,
"step": 50
},
{
"epoch": 0.57,
"learning_rate": 1.46e-06,
"loss": 0.2433,
"step": 75
},
{
"epoch": 0.76,
"learning_rate": 1.9600000000000003e-06,
"loss": 0.2303,
"step": 100
},
{
"epoch": 0.95,
"learning_rate": 2.4400000000000004e-06,
"loss": 0.2237,
"step": 125
},
{
"epoch": 1.15,
"learning_rate": 2.9400000000000002e-06,
"loss": 0.1888,
"step": 150
},
{
"epoch": 1.34,
"learning_rate": 3.44e-06,
"loss": 0.1636,
"step": 175
},
{
"epoch": 1.53,
"learning_rate": 3.94e-06,
"loss": 0.1607,
"step": 200
},
{
"epoch": 1.72,
"learning_rate": 4.440000000000001e-06,
"loss": 0.1674,
"step": 225
},
{
"epoch": 1.91,
"learning_rate": 4.94e-06,
"loss": 0.1656,
"step": 250
},
{
"epoch": 2.1,
"learning_rate": 5.4400000000000004e-06,
"loss": 0.1375,
"step": 275
},
{
"epoch": 2.29,
"learning_rate": 5.94e-06,
"loss": 0.0927,
"step": 300
},
{
"epoch": 2.48,
"learning_rate": 6.440000000000001e-06,
"loss": 0.1047,
"step": 325
},
{
"epoch": 2.67,
"learning_rate": 6.9400000000000005e-06,
"loss": 0.1162,
"step": 350
},
{
"epoch": 2.86,
"learning_rate": 7.420000000000001e-06,
"loss": 0.1093,
"step": 375
},
{
"epoch": 3.05,
"learning_rate": 7.92e-06,
"loss": 0.0929,
"step": 400
},
{
"epoch": 3.24,
"learning_rate": 8.42e-06,
"loss": 0.0587,
"step": 425
},
{
"epoch": 3.44,
"learning_rate": 8.920000000000001e-06,
"loss": 0.0672,
"step": 450
},
{
"epoch": 3.63,
"learning_rate": 9.42e-06,
"loss": 0.0695,
"step": 475
},
{
"epoch": 3.82,
"learning_rate": 9.920000000000002e-06,
"loss": 0.0706,
"step": 500
},
{
"epoch": 4.01,
"learning_rate": 9.953333333333333e-06,
"loss": 0.0652,
"step": 525
},
{
"epoch": 4.2,
"learning_rate": 9.89777777777778e-06,
"loss": 0.0411,
"step": 550
},
{
"epoch": 4.39,
"learning_rate": 9.842222222222223e-06,
"loss": 0.043,
"step": 575
},
{
"epoch": 4.58,
"learning_rate": 9.786666666666667e-06,
"loss": 0.0422,
"step": 600
},
{
"epoch": 4.77,
"learning_rate": 9.731111111111113e-06,
"loss": 0.0424,
"step": 625
},
{
"epoch": 4.96,
"learning_rate": 9.675555555555555e-06,
"loss": 0.0447,
"step": 650
},
{
"epoch": 5.15,
"learning_rate": 9.620000000000001e-06,
"loss": 0.0294,
"step": 675
},
{
"epoch": 5.34,
"learning_rate": 9.564444444444445e-06,
"loss": 0.0252,
"step": 700
},
{
"epoch": 5.53,
"learning_rate": 9.508888888888889e-06,
"loss": 0.0272,
"step": 725
},
{
"epoch": 5.73,
"learning_rate": 9.453333333333335e-06,
"loss": 0.027,
"step": 750
},
{
"epoch": 5.92,
"learning_rate": 9.397777777777779e-06,
"loss": 0.0273,
"step": 775
},
{
"epoch": 6.11,
"learning_rate": 9.342222222222223e-06,
"loss": 0.0195,
"step": 800
},
{
"epoch": 6.3,
"learning_rate": 9.286666666666667e-06,
"loss": 0.0184,
"step": 825
},
{
"epoch": 6.49,
"learning_rate": 9.231111111111111e-06,
"loss": 0.0156,
"step": 850
},
{
"epoch": 6.68,
"learning_rate": 9.175555555555557e-06,
"loss": 0.0176,
"step": 875
},
{
"epoch": 6.87,
"learning_rate": 9.12e-06,
"loss": 0.0181,
"step": 900
},
{
"epoch": 7.06,
"learning_rate": 9.064444444444447e-06,
"loss": 0.0138,
"step": 925
},
{
"epoch": 7.25,
"learning_rate": 9.008888888888889e-06,
"loss": 0.0103,
"step": 950
},
{
"epoch": 7.44,
"learning_rate": 8.953333333333335e-06,
"loss": 0.0116,
"step": 975
},
{
"epoch": 7.63,
"learning_rate": 8.897777777777779e-06,
"loss": 0.0111,
"step": 1000
},
{
"epoch": 7.63,
"eval_loss": 0.320997953414917,
"eval_runtime": 1728.016,
"eval_samples_per_second": 2.595,
"eval_steps_per_second": 0.163,
"eval_wer": 0.78878255063432,
"step": 1000
}
],
"max_steps": 5000,
"num_train_epochs": 39,
"total_flos": 1.05489752168448e+20,
"trial_name": null,
"trial_params": null
}