anirudhmu's picture
End of training
820af68
{
"best_metric": 0.9719298245614035,
"best_model_checkpoint": "swin-tiny-patch4-window7-224-finetuned-soccer-binary2/checkpoint-40",
"epoch": 9.876543209876543,
"eval_steps": 500,
"global_step": 200,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.49,
"learning_rate": 2.5e-05,
"loss": 0.9588,
"step": 10
},
{
"epoch": 0.99,
"learning_rate": 5e-05,
"loss": 0.4085,
"step": 20
},
{
"epoch": 0.99,
"eval_accuracy": 0.9543859649122807,
"eval_loss": 0.17403419315814972,
"eval_runtime": 63.6356,
"eval_samples_per_second": 4.479,
"eval_steps_per_second": 0.141,
"step": 20
},
{
"epoch": 1.48,
"learning_rate": 4.722222222222222e-05,
"loss": 0.1662,
"step": 30
},
{
"epoch": 1.98,
"learning_rate": 4.4444444444444447e-05,
"loss": 0.1281,
"step": 40
},
{
"epoch": 1.98,
"eval_accuracy": 0.9719298245614035,
"eval_loss": 0.10784568637609482,
"eval_runtime": 2.5004,
"eval_samples_per_second": 113.981,
"eval_steps_per_second": 3.599,
"step": 40
},
{
"epoch": 2.47,
"learning_rate": 4.166666666666667e-05,
"loss": 0.1387,
"step": 50
},
{
"epoch": 2.96,
"learning_rate": 3.888888888888889e-05,
"loss": 0.108,
"step": 60
},
{
"epoch": 2.96,
"eval_accuracy": 0.968421052631579,
"eval_loss": 0.0977824479341507,
"eval_runtime": 2.5139,
"eval_samples_per_second": 113.371,
"eval_steps_per_second": 3.58,
"step": 60
},
{
"epoch": 3.46,
"learning_rate": 3.611111111111111e-05,
"loss": 0.1012,
"step": 70
},
{
"epoch": 3.95,
"learning_rate": 3.3333333333333335e-05,
"loss": 0.1077,
"step": 80
},
{
"epoch": 4.0,
"eval_accuracy": 0.968421052631579,
"eval_loss": 0.1005701944231987,
"eval_runtime": 2.2572,
"eval_samples_per_second": 126.265,
"eval_steps_per_second": 3.987,
"step": 81
},
{
"epoch": 4.44,
"learning_rate": 3.055555555555556e-05,
"loss": 0.0921,
"step": 90
},
{
"epoch": 4.94,
"learning_rate": 2.777777777777778e-05,
"loss": 0.0916,
"step": 100
},
{
"epoch": 4.99,
"eval_accuracy": 0.9649122807017544,
"eval_loss": 0.09544265270233154,
"eval_runtime": 2.253,
"eval_samples_per_second": 126.497,
"eval_steps_per_second": 3.995,
"step": 101
},
{
"epoch": 5.43,
"learning_rate": 2.5e-05,
"loss": 0.1179,
"step": 110
},
{
"epoch": 5.93,
"learning_rate": 2.2222222222222223e-05,
"loss": 0.0824,
"step": 120
},
{
"epoch": 5.98,
"eval_accuracy": 0.968421052631579,
"eval_loss": 0.093499556183815,
"eval_runtime": 2.3169,
"eval_samples_per_second": 123.011,
"eval_steps_per_second": 3.885,
"step": 121
},
{
"epoch": 6.42,
"learning_rate": 1.9444444444444445e-05,
"loss": 0.0937,
"step": 130
},
{
"epoch": 6.91,
"learning_rate": 1.6666666666666667e-05,
"loss": 0.0859,
"step": 140
},
{
"epoch": 6.96,
"eval_accuracy": 0.968421052631579,
"eval_loss": 0.09752921760082245,
"eval_runtime": 2.523,
"eval_samples_per_second": 112.962,
"eval_steps_per_second": 3.567,
"step": 141
},
{
"epoch": 7.41,
"learning_rate": 1.388888888888889e-05,
"loss": 0.0909,
"step": 150
},
{
"epoch": 7.9,
"learning_rate": 1.1111111111111112e-05,
"loss": 0.0927,
"step": 160
},
{
"epoch": 8.0,
"eval_accuracy": 0.968421052631579,
"eval_loss": 0.09489437192678452,
"eval_runtime": 2.2777,
"eval_samples_per_second": 125.124,
"eval_steps_per_second": 3.951,
"step": 162
},
{
"epoch": 8.4,
"learning_rate": 8.333333333333334e-06,
"loss": 0.0937,
"step": 170
},
{
"epoch": 8.89,
"learning_rate": 5.555555555555556e-06,
"loss": 0.0836,
"step": 180
},
{
"epoch": 8.99,
"eval_accuracy": 0.968421052631579,
"eval_loss": 0.09275341033935547,
"eval_runtime": 2.6022,
"eval_samples_per_second": 109.523,
"eval_steps_per_second": 3.459,
"step": 182
},
{
"epoch": 9.38,
"learning_rate": 2.777777777777778e-06,
"loss": 0.0947,
"step": 190
},
{
"epoch": 9.88,
"learning_rate": 0.0,
"loss": 0.0958,
"step": 200
},
{
"epoch": 9.88,
"eval_accuracy": 0.968421052631579,
"eval_loss": 0.09396716207265854,
"eval_runtime": 2.3823,
"eval_samples_per_second": 119.633,
"eval_steps_per_second": 3.778,
"step": 200
},
{
"epoch": 9.88,
"step": 200,
"total_flos": 6.300672139168727e+17,
"train_loss": 0.16160290569067,
"train_runtime": 1090.2893,
"train_samples_per_second": 23.517,
"train_steps_per_second": 0.183
}
],
"logging_steps": 10,
"max_steps": 200,
"num_train_epochs": 10,
"save_steps": 500,
"total_flos": 6.300672139168727e+17,
"trial_name": null,
"trial_params": null
}