xls-r-300m-sv-robust / trainer_state.json
marinone94's picture
End of training
dfda964
raw
history blame
2.16 kB
{
"best_metric": null,
"best_model_checkpoint": null,
"epoch": 0.029006526468455404,
"global_step": 10,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.0,
"learning_rate": 3.75e-06,
"loss": 13.605,
"step": 1
},
{
"epoch": 0.01,
"learning_rate": 7.5e-06,
"loss": 11.0063,
"step": 2
},
{
"epoch": 0.01,
"learning_rate": 1.1249999999999999e-05,
"loss": 11.6693,
"step": 3
},
{
"epoch": 0.01,
"learning_rate": 1.5e-05,
"loss": 13.432,
"step": 4
},
{
"epoch": 0.01,
"learning_rate": 1.875e-05,
"loss": 10.3458,
"step": 5
},
{
"epoch": 0.01,
"eval_loss": 15.09913444519043,
"eval_runtime": 139.8551,
"eval_samples_per_second": 33.034,
"eval_steps_per_second": 4.133,
"eval_wer": 1.0,
"step": 5
},
{
"epoch": 0.02,
"learning_rate": 2.2499999999999998e-05,
"loss": 15.2451,
"step": 6
},
{
"epoch": 0.02,
"learning_rate": 2.625e-05,
"loss": 10.0481,
"step": 7
},
{
"epoch": 0.02,
"learning_rate": 3e-05,
"loss": 12.3838,
"step": 8
},
{
"epoch": 0.03,
"learning_rate": 3.3749999999999994e-05,
"loss": 11.9858,
"step": 9
},
{
"epoch": 0.03,
"learning_rate": 3.75e-05,
"loss": 11.9029,
"step": 10
},
{
"epoch": 0.03,
"eval_loss": 14.213573455810547,
"eval_runtime": 160.2552,
"eval_samples_per_second": 28.829,
"eval_steps_per_second": 3.607,
"eval_wer": 1.0,
"step": 10
},
{
"epoch": 0.03,
"step": 10,
"total_flos": 4.405923604988928e+16,
"train_loss": 12.162414741516113,
"train_runtime": 400.3916,
"train_samples_per_second": 0.799,
"train_steps_per_second": 0.025
}
],
"max_steps": 10,
"num_train_epochs": 1,
"total_flos": 4.405923604988928e+16,
"trial_name": null,
"trial_params": null
}