whisper-large-v2-ne / trainer_state.json
anuragshas's picture
End of training
b1a1522
raw
history blame
5.58 kB
{
"best_metric": null,
"best_model_checkpoint": null,
"epoch": 999.001,
"global_step": 1000,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 24.0,
"learning_rate": 6.505149978319905e-06,
"loss": 0.2468,
"step": 25
},
{
"epoch": 49.0,
"learning_rate": 8.266062568876717e-06,
"loss": 0.0002,
"step": 50
},
{
"epoch": 74.0,
"learning_rate": 9.225490200071284e-06,
"loss": 0.0001,
"step": 75
},
{
"epoch": 99.0,
"learning_rate": 9.888618026444238e-06,
"loss": 0.0,
"step": 100
},
{
"epoch": 124.0,
"learning_rate": 9.78888888888889e-06,
"loss": 0.0,
"step": 125
},
{
"epoch": 149.0,
"learning_rate": 9.511111111111112e-06,
"loss": 0.0,
"step": 150
},
{
"epoch": 174.0,
"learning_rate": 9.233333333333334e-06,
"loss": 0.0,
"step": 175
},
{
"epoch": 199.0,
"learning_rate": 8.955555555555555e-06,
"loss": 0.0,
"step": 200
},
{
"epoch": 224.0,
"learning_rate": 8.677777777777779e-06,
"loss": 0.0,
"step": 225
},
{
"epoch": 249.0,
"learning_rate": 8.400000000000001e-06,
"loss": 0.0,
"step": 250
},
{
"epoch": 274.0,
"learning_rate": 8.122222222222223e-06,
"loss": 0.0,
"step": 275
},
{
"epoch": 299.0,
"learning_rate": 7.844444444444446e-06,
"loss": 0.0,
"step": 300
},
{
"epoch": 324.0,
"learning_rate": 7.566666666666667e-06,
"loss": 0.0,
"step": 325
},
{
"epoch": 349.0,
"learning_rate": 7.28888888888889e-06,
"loss": 0.0,
"step": 350
},
{
"epoch": 374.0,
"learning_rate": 7.011111111111112e-06,
"loss": 0.0,
"step": 375
},
{
"epoch": 399.0,
"learning_rate": 6.733333333333334e-06,
"loss": 0.0,
"step": 400
},
{
"epoch": 424.0,
"learning_rate": 6.455555555555556e-06,
"loss": 0.0,
"step": 425
},
{
"epoch": 449.0,
"learning_rate": 6.177777777777778e-06,
"loss": 0.0,
"step": 450
},
{
"epoch": 474.0,
"learning_rate": 5.9e-06,
"loss": 0.0,
"step": 475
},
{
"epoch": 499.0,
"learning_rate": 5.622222222222222e-06,
"loss": 0.0,
"step": 500
},
{
"epoch": 524.0,
"learning_rate": 5.344444444444446e-06,
"loss": 0.0,
"step": 525
},
{
"epoch": 549.0,
"learning_rate": 5.0666666666666676e-06,
"loss": 0.0,
"step": 550
},
{
"epoch": 574.0,
"learning_rate": 4.7888888888888894e-06,
"loss": 0.0,
"step": 575
},
{
"epoch": 599.0,
"learning_rate": 4.511111111111111e-06,
"loss": 0.0,
"step": 600
},
{
"epoch": 624.0,
"learning_rate": 4.233333333333334e-06,
"loss": 0.0,
"step": 625
},
{
"epoch": 649.0,
"learning_rate": 3.955555555555556e-06,
"loss": 0.0,
"step": 650
},
{
"epoch": 674.0,
"learning_rate": 3.6777777777777778e-06,
"loss": 0.0,
"step": 675
},
{
"epoch": 699.0,
"learning_rate": 3.4000000000000005e-06,
"loss": 0.0,
"step": 700
},
{
"epoch": 724.0,
"learning_rate": 3.1222222222222228e-06,
"loss": 0.0,
"step": 725
},
{
"epoch": 749.0,
"learning_rate": 2.8444444444444446e-06,
"loss": 0.0,
"step": 750
},
{
"epoch": 774.0,
"learning_rate": 2.566666666666667e-06,
"loss": 0.0,
"step": 775
},
{
"epoch": 799.0,
"learning_rate": 2.2888888888888892e-06,
"loss": 0.0,
"step": 800
},
{
"epoch": 824.0,
"learning_rate": 2.011111111111111e-06,
"loss": 0.0,
"step": 825
},
{
"epoch": 849.0,
"learning_rate": 1.7333333333333336e-06,
"loss": 0.0,
"step": 850
},
{
"epoch": 874.0,
"learning_rate": 1.4555555555555557e-06,
"loss": 0.0,
"step": 875
},
{
"epoch": 899.0,
"learning_rate": 1.1777777777777778e-06,
"loss": 0.0,
"step": 900
},
{
"epoch": 924.0,
"learning_rate": 9.000000000000001e-07,
"loss": 0.0,
"step": 925
},
{
"epoch": 949.0,
"learning_rate": 6.222222222222223e-07,
"loss": 0.0,
"step": 950
},
{
"epoch": 974.0,
"learning_rate": 3.444444444444445e-07,
"loss": 0.0,
"step": 975
},
{
"epoch": 999.0,
"learning_rate": 6.666666666666668e-08,
"loss": 0.0,
"step": 1000
},
{
"epoch": 999.0,
"eval_loss": 1.572265625,
"eval_runtime": 3.6566,
"eval_samples_per_second": 0.547,
"eval_steps_per_second": 0.273,
"eval_wer": 56.09756097560976,
"step": 1000
},
{
"epoch": 999.0,
"step": 1000,
"total_flos": 1.4862197628694692e+19,
"train_loss": 0.006189345955848694,
"train_runtime": 3609.1373,
"train_samples_per_second": 8.866,
"train_steps_per_second": 0.277
}
],
"max_steps": 1000,
"num_train_epochs": 9223372036854775807,
"total_flos": 1.4862197628694692e+19,
"trial_name": null,
"trial_params": null
}