whisper-small-gom / trainer_state.json
thak123's picture
Upload 9 files
fdb0b26
{
"best_metric": 99.95633187772926,
"best_model_checkpoint": "./whisper-small-gom/checkpoint-1000",
"epoch": 1.41643059490085,
"global_step": 1000,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.04,
"learning_rate": 4.6000000000000004e-07,
"loss": 2.481,
"step": 25
},
{
"epoch": 0.07,
"learning_rate": 9.600000000000001e-07,
"loss": 2.0967,
"step": 50
},
{
"epoch": 0.11,
"learning_rate": 1.46e-06,
"loss": 1.8738,
"step": 75
},
{
"epoch": 0.14,
"learning_rate": 1.9600000000000003e-06,
"loss": 1.5778,
"step": 100
},
{
"epoch": 0.18,
"learning_rate": 2.46e-06,
"loss": 1.5275,
"step": 125
},
{
"epoch": 0.21,
"learning_rate": 2.96e-06,
"loss": 1.4266,
"step": 150
},
{
"epoch": 0.25,
"learning_rate": 3.46e-06,
"loss": 1.3163,
"step": 175
},
{
"epoch": 0.28,
"learning_rate": 3.96e-06,
"loss": 1.3454,
"step": 200
},
{
"epoch": 0.32,
"learning_rate": 4.4600000000000005e-06,
"loss": 1.248,
"step": 225
},
{
"epoch": 0.35,
"learning_rate": 4.960000000000001e-06,
"loss": 1.2251,
"step": 250
},
{
"epoch": 0.39,
"learning_rate": 5.460000000000001e-06,
"loss": 1.2044,
"step": 275
},
{
"epoch": 0.42,
"learning_rate": 5.9600000000000005e-06,
"loss": 1.1433,
"step": 300
},
{
"epoch": 0.46,
"learning_rate": 6.460000000000001e-06,
"loss": 1.1663,
"step": 325
},
{
"epoch": 0.5,
"learning_rate": 6.96e-06,
"loss": 1.1463,
"step": 350
},
{
"epoch": 0.53,
"learning_rate": 7.4600000000000006e-06,
"loss": 1.0479,
"step": 375
},
{
"epoch": 0.57,
"learning_rate": 7.960000000000002e-06,
"loss": 1.0869,
"step": 400
},
{
"epoch": 0.6,
"learning_rate": 8.46e-06,
"loss": 1.0555,
"step": 425
},
{
"epoch": 0.64,
"learning_rate": 8.96e-06,
"loss": 0.9438,
"step": 450
},
{
"epoch": 0.67,
"learning_rate": 9.460000000000001e-06,
"loss": 0.9683,
"step": 475
},
{
"epoch": 0.71,
"learning_rate": 9.960000000000001e-06,
"loss": 0.9701,
"step": 500
},
{
"epoch": 0.74,
"learning_rate": 9.934285714285715e-06,
"loss": 0.9346,
"step": 525
},
{
"epoch": 0.78,
"learning_rate": 9.862857142857144e-06,
"loss": 0.959,
"step": 550
},
{
"epoch": 0.81,
"learning_rate": 9.791428571428571e-06,
"loss": 0.8805,
"step": 575
},
{
"epoch": 0.85,
"learning_rate": 9.72e-06,
"loss": 0.8644,
"step": 600
},
{
"epoch": 0.89,
"learning_rate": 9.648571428571429e-06,
"loss": 0.8948,
"step": 625
},
{
"epoch": 0.92,
"learning_rate": 9.577142857142858e-06,
"loss": 0.9455,
"step": 650
},
{
"epoch": 0.96,
"learning_rate": 9.505714285714287e-06,
"loss": 0.8696,
"step": 675
},
{
"epoch": 0.99,
"learning_rate": 9.434285714285714e-06,
"loss": 0.8829,
"step": 700
},
{
"epoch": 1.03,
"learning_rate": 9.362857142857143e-06,
"loss": 0.7547,
"step": 725
},
{
"epoch": 1.06,
"learning_rate": 9.291428571428572e-06,
"loss": 0.815,
"step": 750
},
{
"epoch": 1.1,
"learning_rate": 9.220000000000002e-06,
"loss": 0.8324,
"step": 775
},
{
"epoch": 1.13,
"learning_rate": 9.148571428571429e-06,
"loss": 0.7684,
"step": 800
},
{
"epoch": 1.17,
"learning_rate": 9.077142857142858e-06,
"loss": 0.7822,
"step": 825
},
{
"epoch": 1.2,
"learning_rate": 9.005714285714287e-06,
"loss": 0.7616,
"step": 850
},
{
"epoch": 1.24,
"learning_rate": 8.934285714285716e-06,
"loss": 0.8268,
"step": 875
},
{
"epoch": 1.27,
"learning_rate": 8.862857142857143e-06,
"loss": 0.8051,
"step": 900
},
{
"epoch": 1.31,
"learning_rate": 8.791428571428572e-06,
"loss": 0.804,
"step": 925
},
{
"epoch": 1.35,
"learning_rate": 8.720000000000001e-06,
"loss": 0.7387,
"step": 950
},
{
"epoch": 1.38,
"learning_rate": 8.64857142857143e-06,
"loss": 0.7993,
"step": 975
},
{
"epoch": 1.42,
"learning_rate": 8.577142857142858e-06,
"loss": 0.8099,
"step": 1000
},
{
"epoch": 1.42,
"eval_loss": 0.8143458366394043,
"eval_runtime": 1609.901,
"eval_samples_per_second": 1.238,
"eval_steps_per_second": 0.155,
"eval_wer": 99.95633187772926,
"step": 1000
}
],
"max_steps": 4000,
"num_train_epochs": 6,
"total_flos": 4.6159234781184e+18,
"trial_name": null,
"trial_params": null
}