whisper-large-v2-kk / trainer_state.json
anuragshas's picture
End of training
b2a398a
raw
history blame
5.67 kB
{
"best_metric": null,
"best_model_checkpoint": null,
"epoch": 39.025,
"global_step": 1000,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.03,
"learning_rate": 6.712113404111031e-06,
"loss": 0.4926,
"step": 25
},
{
"epoch": 1.02,
"learning_rate": 8.360489289678585e-06,
"loss": 0.1973,
"step": 50
},
{
"epoch": 2.02,
"learning_rate": 9.28666248215634e-06,
"loss": 0.0815,
"step": 75
},
{
"epoch": 3.02,
"learning_rate": 9.933858671331224e-06,
"loss": 0.0412,
"step": 100
},
{
"epoch": 4.03,
"learning_rate": 9.766666666666667e-06,
"loss": 0.0264,
"step": 125
},
{
"epoch": 5.03,
"learning_rate": 9.48888888888889e-06,
"loss": 0.0193,
"step": 150
},
{
"epoch": 6.03,
"learning_rate": 9.222222222222224e-06,
"loss": 0.0178,
"step": 175
},
{
"epoch": 7.03,
"learning_rate": 8.944444444444446e-06,
"loss": 0.0163,
"step": 200
},
{
"epoch": 8.03,
"learning_rate": 8.666666666666668e-06,
"loss": 0.0088,
"step": 225
},
{
"epoch": 9.03,
"learning_rate": 8.38888888888889e-06,
"loss": 0.0058,
"step": 250
},
{
"epoch": 10.03,
"learning_rate": 8.111111111111112e-06,
"loss": 0.0065,
"step": 275
},
{
"epoch": 11.03,
"learning_rate": 7.833333333333333e-06,
"loss": 0.0062,
"step": 300
},
{
"epoch": 12.03,
"learning_rate": 7.555555555555556e-06,
"loss": 0.0049,
"step": 325
},
{
"epoch": 13.03,
"learning_rate": 7.277777777777778e-06,
"loss": 0.0035,
"step": 350
},
{
"epoch": 14.03,
"learning_rate": 7e-06,
"loss": 0.0023,
"step": 375
},
{
"epoch": 15.03,
"learning_rate": 6.7222222222222235e-06,
"loss": 0.0026,
"step": 400
},
{
"epoch": 16.02,
"learning_rate": 6.444444444444445e-06,
"loss": 0.0027,
"step": 425
},
{
"epoch": 17.02,
"learning_rate": 6.166666666666667e-06,
"loss": 0.0025,
"step": 450
},
{
"epoch": 18.02,
"learning_rate": 5.88888888888889e-06,
"loss": 0.0014,
"step": 475
},
{
"epoch": 19.02,
"learning_rate": 5.611111111111112e-06,
"loss": 0.0025,
"step": 500
},
{
"epoch": 20.02,
"learning_rate": 5.333333333333334e-06,
"loss": 0.0009,
"step": 525
},
{
"epoch": 21.02,
"learning_rate": 5.0555555555555555e-06,
"loss": 0.0005,
"step": 550
},
{
"epoch": 22.02,
"learning_rate": 4.777777777777778e-06,
"loss": 0.0004,
"step": 575
},
{
"epoch": 23.02,
"learning_rate": 4.5e-06,
"loss": 0.0004,
"step": 600
},
{
"epoch": 24.02,
"learning_rate": 4.222222222222223e-06,
"loss": 0.0004,
"step": 625
},
{
"epoch": 25.02,
"learning_rate": 3.944444444444445e-06,
"loss": 0.0003,
"step": 650
},
{
"epoch": 26.02,
"learning_rate": 3.6666666666666666e-06,
"loss": 0.0003,
"step": 675
},
{
"epoch": 27.02,
"learning_rate": 3.3888888888888893e-06,
"loss": 0.0003,
"step": 700
},
{
"epoch": 28.02,
"learning_rate": 3.1111111111111116e-06,
"loss": 0.0003,
"step": 725
},
{
"epoch": 29.02,
"learning_rate": 2.8333333333333335e-06,
"loss": 0.0003,
"step": 750
},
{
"epoch": 30.02,
"learning_rate": 2.5555555555555557e-06,
"loss": 0.0003,
"step": 775
},
{
"epoch": 31.02,
"learning_rate": 2.277777777777778e-06,
"loss": 0.0003,
"step": 800
},
{
"epoch": 32.02,
"learning_rate": 2.0000000000000003e-06,
"loss": 0.0003,
"step": 825
},
{
"epoch": 33.02,
"learning_rate": 1.7222222222222224e-06,
"loss": 0.0003,
"step": 850
},
{
"epoch": 34.02,
"learning_rate": 1.4444444444444445e-06,
"loss": 0.0002,
"step": 875
},
{
"epoch": 35.02,
"learning_rate": 1.1666666666666668e-06,
"loss": 0.0002,
"step": 900
},
{
"epoch": 36.02,
"learning_rate": 8.88888888888889e-07,
"loss": 0.0002,
"step": 925
},
{
"epoch": 37.02,
"learning_rate": 6.111111111111112e-07,
"loss": 0.0002,
"step": 950
},
{
"epoch": 38.02,
"learning_rate": 3.3333333333333335e-07,
"loss": 0.0002,
"step": 975
},
{
"epoch": 39.02,
"learning_rate": 5.555555555555556e-08,
"loss": 0.0002,
"step": 1000
},
{
"epoch": 39.02,
"eval_loss": 0.57470703125,
"eval_runtime": 141.1189,
"eval_samples_per_second": 2.657,
"eval_steps_per_second": 0.17,
"eval_wer": 35.95004460303301,
"step": 1000
},
{
"epoch": 39.02,
"step": 1000,
"total_flos": 6.743191013696733e+19,
"train_loss": 0.023719379663467408,
"train_runtime": 7941.5914,
"train_samples_per_second": 4.029,
"train_steps_per_second": 0.126
}
],
"max_steps": 1000,
"num_train_epochs": 9223372036854775807,
"total_flos": 6.743191013696733e+19,
"trial_name": null,
"trial_params": null
}