B2_1000_1e-5_hp-mehrdad / trainer_state.json
lnxdx's picture
End of training
517e83c
{
"best_metric": 0.31366822429906543,
"best_model_checkpoint": "/content/drive/MyDrive/wav2vec2-large-xlsr-persian-asr-shemo_sm95/checkpoint-1000",
"epoch": 6.25,
"eval_steps": 100,
"global_step": 1000,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.62,
"learning_rate": 2.0000000000000003e-06,
"loss": 0.7624,
"step": 100
},
{
"epoch": 0.62,
"eval_loss": 0.6707654595375061,
"eval_runtime": 34.4464,
"eval_samples_per_second": 8.245,
"eval_steps_per_second": 1.045,
"eval_wer": 0.3235981308411215,
"step": 100
},
{
"epoch": 1.25,
"learning_rate": 3.980000000000001e-06,
"loss": 0.78,
"step": 200
},
{
"epoch": 1.25,
"eval_loss": 0.6667919754981995,
"eval_runtime": 33.8693,
"eval_samples_per_second": 8.385,
"eval_steps_per_second": 1.063,
"eval_wer": 0.3244742990654206,
"step": 200
},
{
"epoch": 1.88,
"learning_rate": 5.98e-06,
"loss": 0.7856,
"step": 300
},
{
"epoch": 1.88,
"eval_loss": 0.6599859595298767,
"eval_runtime": 34.4905,
"eval_samples_per_second": 8.234,
"eval_steps_per_second": 1.044,
"eval_wer": 0.32739485981308414,
"step": 300
},
{
"epoch": 2.5,
"learning_rate": 7.980000000000002e-06,
"loss": 0.7239,
"step": 400
},
{
"epoch": 2.5,
"eval_loss": 0.6671751141548157,
"eval_runtime": 33.9261,
"eval_samples_per_second": 8.371,
"eval_steps_per_second": 1.061,
"eval_wer": 0.32330607476635514,
"step": 400
},
{
"epoch": 3.12,
"learning_rate": 9.960000000000001e-06,
"loss": 0.7311,
"step": 500
},
{
"epoch": 3.12,
"eval_loss": 0.6747905611991882,
"eval_runtime": 34.5601,
"eval_samples_per_second": 8.218,
"eval_steps_per_second": 1.042,
"eval_wer": 0.3142523364485981,
"step": 500
},
{
"epoch": 3.75,
"learning_rate": 8.040000000000001e-06,
"loss": 0.7408,
"step": 600
},
{
"epoch": 3.75,
"eval_loss": 0.651818037033081,
"eval_runtime": 33.8044,
"eval_samples_per_second": 8.401,
"eval_steps_per_second": 1.065,
"eval_wer": 0.3247663551401869,
"step": 600
},
{
"epoch": 4.38,
"learning_rate": 6.040000000000001e-06,
"loss": 0.713,
"step": 700
},
{
"epoch": 4.38,
"eval_loss": 0.6586844325065613,
"eval_runtime": 34.0097,
"eval_samples_per_second": 8.351,
"eval_steps_per_second": 1.059,
"eval_wer": 0.3177570093457944,
"step": 700
},
{
"epoch": 5.0,
"learning_rate": 4.04e-06,
"loss": 0.7068,
"step": 800
},
{
"epoch": 5.0,
"eval_loss": 0.659990131855011,
"eval_runtime": 33.9629,
"eval_samples_per_second": 8.362,
"eval_steps_per_second": 1.06,
"eval_wer": 0.3171728971962617,
"step": 800
},
{
"epoch": 5.62,
"learning_rate": 2.04e-06,
"loss": 0.6938,
"step": 900
},
{
"epoch": 5.62,
"eval_loss": 0.6598456501960754,
"eval_runtime": 34.354,
"eval_samples_per_second": 8.267,
"eval_steps_per_second": 1.048,
"eval_wer": 0.3157126168224299,
"step": 900
},
{
"epoch": 6.25,
"learning_rate": 4e-08,
"loss": 0.6809,
"step": 1000
},
{
"epoch": 6.25,
"eval_loss": 0.6591429114341736,
"eval_runtime": 33.6087,
"eval_samples_per_second": 8.45,
"eval_steps_per_second": 1.071,
"eval_wer": 0.31366822429906543,
"step": 1000
},
{
"epoch": 6.25,
"step": 1000,
"total_flos": 4.798362622332561e+18,
"train_loss": 0.7318317260742188,
"train_runtime": 4802.1983,
"train_samples_per_second": 3.332,
"train_steps_per_second": 0.208
}
],
"logging_steps": 100,
"max_steps": 1000,
"num_train_epochs": 7,
"save_steps": 100,
"total_flos": 4.798362622332561e+18,
"trial_name": null,
"trial_params": null
}