hoangdeeptry's picture
Training in progress, step 1000
840c4b5
{
"best_metric": 207.2559468643806,
"best_model_checkpoint": "./whisper-vietnamese-2/checkpoint-1000",
"epoch": 5.319148936170213,
"global_step": 1000,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.13,
"learning_rate": 4.6000000000000004e-07,
"loss": 2.9537,
"step": 25
},
{
"epoch": 0.27,
"learning_rate": 9.600000000000001e-07,
"loss": 2.092,
"step": 50
},
{
"epoch": 0.4,
"learning_rate": 1.46e-06,
"loss": 1.332,
"step": 75
},
{
"epoch": 0.53,
"learning_rate": 1.9600000000000003e-06,
"loss": 0.9112,
"step": 100
},
{
"epoch": 0.66,
"learning_rate": 2.46e-06,
"loss": 0.7232,
"step": 125
},
{
"epoch": 0.8,
"learning_rate": 2.96e-06,
"loss": 0.6644,
"step": 150
},
{
"epoch": 0.93,
"learning_rate": 3.46e-06,
"loss": 0.5903,
"step": 175
},
{
"epoch": 1.06,
"learning_rate": 3.96e-06,
"loss": 0.5687,
"step": 200
},
{
"epoch": 1.2,
"learning_rate": 4.4600000000000005e-06,
"loss": 0.4747,
"step": 225
},
{
"epoch": 1.33,
"learning_rate": 4.960000000000001e-06,
"loss": 0.4551,
"step": 250
},
{
"epoch": 1.46,
"learning_rate": 5.460000000000001e-06,
"loss": 0.4318,
"step": 275
},
{
"epoch": 1.6,
"learning_rate": 5.9600000000000005e-06,
"loss": 0.3781,
"step": 300
},
{
"epoch": 1.73,
"learning_rate": 6.460000000000001e-06,
"loss": 0.2374,
"step": 325
},
{
"epoch": 1.86,
"learning_rate": 6.96e-06,
"loss": 0.2031,
"step": 350
},
{
"epoch": 1.99,
"learning_rate": 7.4600000000000006e-06,
"loss": 0.1761,
"step": 375
},
{
"epoch": 2.13,
"learning_rate": 7.960000000000002e-06,
"loss": 0.1332,
"step": 400
},
{
"epoch": 2.26,
"learning_rate": 8.46e-06,
"loss": 0.1072,
"step": 425
},
{
"epoch": 2.39,
"learning_rate": 8.96e-06,
"loss": 0.1034,
"step": 450
},
{
"epoch": 2.53,
"learning_rate": 9.460000000000001e-06,
"loss": 0.1021,
"step": 475
},
{
"epoch": 2.66,
"learning_rate": 9.960000000000001e-06,
"loss": 0.093,
"step": 500
},
{
"epoch": 2.79,
"learning_rate": 9.934285714285715e-06,
"loss": 0.0883,
"step": 525
},
{
"epoch": 2.93,
"learning_rate": 9.862857142857144e-06,
"loss": 0.0991,
"step": 550
},
{
"epoch": 3.06,
"learning_rate": 9.791428571428571e-06,
"loss": 0.0689,
"step": 575
},
{
"epoch": 3.19,
"learning_rate": 9.72e-06,
"loss": 0.0418,
"step": 600
},
{
"epoch": 3.32,
"learning_rate": 9.648571428571429e-06,
"loss": 0.0428,
"step": 625
},
{
"epoch": 3.46,
"learning_rate": 9.577142857142858e-06,
"loss": 0.0337,
"step": 650
},
{
"epoch": 3.59,
"learning_rate": 9.505714285714287e-06,
"loss": 0.0329,
"step": 675
},
{
"epoch": 3.72,
"learning_rate": 9.434285714285714e-06,
"loss": 0.0353,
"step": 700
},
{
"epoch": 3.86,
"learning_rate": 9.362857142857143e-06,
"loss": 0.0378,
"step": 725
},
{
"epoch": 3.99,
"learning_rate": 9.291428571428572e-06,
"loss": 0.0336,
"step": 750
},
{
"epoch": 4.12,
"learning_rate": 9.220000000000002e-06,
"loss": 0.0178,
"step": 775
},
{
"epoch": 4.26,
"learning_rate": 9.148571428571429e-06,
"loss": 0.013,
"step": 800
},
{
"epoch": 4.39,
"learning_rate": 9.077142857142858e-06,
"loss": 0.0143,
"step": 825
},
{
"epoch": 4.52,
"learning_rate": 9.005714285714287e-06,
"loss": 0.0186,
"step": 850
},
{
"epoch": 4.65,
"learning_rate": 8.934285714285716e-06,
"loss": 0.0149,
"step": 875
},
{
"epoch": 4.79,
"learning_rate": 8.862857142857143e-06,
"loss": 0.0116,
"step": 900
},
{
"epoch": 4.92,
"learning_rate": 8.791428571428572e-06,
"loss": 0.0166,
"step": 925
},
{
"epoch": 5.05,
"learning_rate": 8.720000000000001e-06,
"loss": 0.0096,
"step": 950
},
{
"epoch": 5.19,
"learning_rate": 8.64857142857143e-06,
"loss": 0.0073,
"step": 975
},
{
"epoch": 5.32,
"learning_rate": 8.577142857142858e-06,
"loss": 0.0059,
"step": 1000
},
{
"epoch": 5.32,
"eval_cer": 160.7235524597301,
"eval_loss": 0.4744850695133209,
"eval_runtime": 925.951,
"eval_samples_per_second": 0.926,
"eval_steps_per_second": 0.117,
"eval_wer": 207.2559468643806,
"step": 1000
}
],
"max_steps": 4000,
"num_train_epochs": 22,
"total_flos": 4.5971654270976e+18,
"trial_name": null,
"trial_params": null
}