mansee's picture
End of training
75fd85c
raw
history blame
7.91 kB
{
"best_metric": 0.9930313588850174,
"best_model_checkpoint": "swin-tiny-patch4-window7-224-spa_saloon_classification-spa-saloon/checkpoint-121",
"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": 0.25,
"learning_rate": 1.25e-05,
"loss": 0.3756,
"step": 5
},
{
"epoch": 0.49,
"learning_rate": 2.5e-05,
"loss": 0.3312,
"step": 10
},
{
"epoch": 0.74,
"learning_rate": 3.7500000000000003e-05,
"loss": 0.3086,
"step": 15
},
{
"epoch": 0.99,
"learning_rate": 5e-05,
"loss": 0.2637,
"step": 20
},
{
"epoch": 0.99,
"eval_accuracy": 0.9442508710801394,
"eval_loss": 0.1274152249097824,
"eval_runtime": 2.2323,
"eval_samples_per_second": 128.565,
"eval_steps_per_second": 4.032,
"step": 20
},
{
"epoch": 1.23,
"learning_rate": 4.8611111111111115e-05,
"loss": 0.2486,
"step": 25
},
{
"epoch": 1.48,
"learning_rate": 4.722222222222222e-05,
"loss": 0.2594,
"step": 30
},
{
"epoch": 1.73,
"learning_rate": 4.5833333333333334e-05,
"loss": 0.2133,
"step": 35
},
{
"epoch": 1.98,
"learning_rate": 4.4444444444444447e-05,
"loss": 0.2582,
"step": 40
},
{
"epoch": 1.98,
"eval_accuracy": 0.975609756097561,
"eval_loss": 0.09369026869535446,
"eval_runtime": 1.916,
"eval_samples_per_second": 149.792,
"eval_steps_per_second": 4.697,
"step": 40
},
{
"epoch": 2.22,
"learning_rate": 4.305555555555556e-05,
"loss": 0.1924,
"step": 45
},
{
"epoch": 2.47,
"learning_rate": 4.166666666666667e-05,
"loss": 0.1749,
"step": 50
},
{
"epoch": 2.72,
"learning_rate": 4.027777777777778e-05,
"loss": 0.1551,
"step": 55
},
{
"epoch": 2.96,
"learning_rate": 3.888888888888889e-05,
"loss": 0.161,
"step": 60
},
{
"epoch": 2.96,
"eval_accuracy": 0.9581881533101045,
"eval_loss": 0.09241745620965958,
"eval_runtime": 2.1953,
"eval_samples_per_second": 130.734,
"eval_steps_per_second": 4.1,
"step": 60
},
{
"epoch": 3.21,
"learning_rate": 3.7500000000000003e-05,
"loss": 0.2067,
"step": 65
},
{
"epoch": 3.46,
"learning_rate": 3.611111111111111e-05,
"loss": 0.1642,
"step": 70
},
{
"epoch": 3.7,
"learning_rate": 3.472222222222222e-05,
"loss": 0.1413,
"step": 75
},
{
"epoch": 3.95,
"learning_rate": 3.3333333333333335e-05,
"loss": 0.1535,
"step": 80
},
{
"epoch": 4.0,
"eval_accuracy": 0.9860627177700348,
"eval_loss": 0.061225663870573044,
"eval_runtime": 1.959,
"eval_samples_per_second": 146.501,
"eval_steps_per_second": 4.594,
"step": 81
},
{
"epoch": 4.2,
"learning_rate": 3.194444444444444e-05,
"loss": 0.1393,
"step": 85
},
{
"epoch": 4.44,
"learning_rate": 3.055555555555556e-05,
"loss": 0.1377,
"step": 90
},
{
"epoch": 4.69,
"learning_rate": 2.916666666666667e-05,
"loss": 0.1266,
"step": 95
},
{
"epoch": 4.94,
"learning_rate": 2.777777777777778e-05,
"loss": 0.1347,
"step": 100
},
{
"epoch": 4.99,
"eval_accuracy": 0.9790940766550522,
"eval_loss": 0.05363203585147858,
"eval_runtime": 2.5164,
"eval_samples_per_second": 114.052,
"eval_steps_per_second": 3.577,
"step": 101
},
{
"epoch": 5.19,
"learning_rate": 2.6388888888888892e-05,
"loss": 0.1041,
"step": 105
},
{
"epoch": 5.43,
"learning_rate": 2.5e-05,
"loss": 0.0922,
"step": 110
},
{
"epoch": 5.68,
"learning_rate": 2.361111111111111e-05,
"loss": 0.1152,
"step": 115
},
{
"epoch": 5.93,
"learning_rate": 2.2222222222222223e-05,
"loss": 0.1155,
"step": 120
},
{
"epoch": 5.98,
"eval_accuracy": 0.9930313588850174,
"eval_loss": 0.040804799646139145,
"eval_runtime": 1.9704,
"eval_samples_per_second": 145.658,
"eval_steps_per_second": 4.568,
"step": 121
},
{
"epoch": 6.17,
"learning_rate": 2.0833333333333336e-05,
"loss": 0.1115,
"step": 125
},
{
"epoch": 6.42,
"learning_rate": 1.9444444444444445e-05,
"loss": 0.0837,
"step": 130
},
{
"epoch": 6.67,
"learning_rate": 1.8055555555555555e-05,
"loss": 0.1156,
"step": 135
},
{
"epoch": 6.91,
"learning_rate": 1.6666666666666667e-05,
"loss": 0.1306,
"step": 140
},
{
"epoch": 6.96,
"eval_accuracy": 0.9930313588850174,
"eval_loss": 0.04172113910317421,
"eval_runtime": 2.3226,
"eval_samples_per_second": 123.57,
"eval_steps_per_second": 3.875,
"step": 141
},
{
"epoch": 7.16,
"learning_rate": 1.527777777777778e-05,
"loss": 0.0813,
"step": 145
},
{
"epoch": 7.41,
"learning_rate": 1.388888888888889e-05,
"loss": 0.1014,
"step": 150
},
{
"epoch": 7.65,
"learning_rate": 1.25e-05,
"loss": 0.1172,
"step": 155
},
{
"epoch": 7.9,
"learning_rate": 1.1111111111111112e-05,
"loss": 0.1017,
"step": 160
},
{
"epoch": 8.0,
"eval_accuracy": 0.9895470383275261,
"eval_loss": 0.038007255643606186,
"eval_runtime": 1.9669,
"eval_samples_per_second": 145.914,
"eval_steps_per_second": 4.576,
"step": 162
},
{
"epoch": 8.15,
"learning_rate": 9.722222222222223e-06,
"loss": 0.0897,
"step": 165
},
{
"epoch": 8.4,
"learning_rate": 8.333333333333334e-06,
"loss": 0.0959,
"step": 170
},
{
"epoch": 8.64,
"learning_rate": 6.944444444444445e-06,
"loss": 0.0867,
"step": 175
},
{
"epoch": 8.89,
"learning_rate": 5.555555555555556e-06,
"loss": 0.0859,
"step": 180
},
{
"epoch": 8.99,
"eval_accuracy": 0.9895470383275261,
"eval_loss": 0.041699331253767014,
"eval_runtime": 2.3485,
"eval_samples_per_second": 122.208,
"eval_steps_per_second": 3.832,
"step": 182
},
{
"epoch": 9.14,
"learning_rate": 4.166666666666667e-06,
"loss": 0.0991,
"step": 185
},
{
"epoch": 9.38,
"learning_rate": 2.777777777777778e-06,
"loss": 0.091,
"step": 190
},
{
"epoch": 9.63,
"learning_rate": 1.388888888888889e-06,
"loss": 0.103,
"step": 195
},
{
"epoch": 9.88,
"learning_rate": 0.0,
"loss": 0.0897,
"step": 200
},
{
"epoch": 9.88,
"eval_accuracy": 0.9895470383275261,
"eval_loss": 0.03927117958664894,
"eval_runtime": 2.2228,
"eval_samples_per_second": 129.119,
"eval_steps_per_second": 4.049,
"step": 200
},
{
"epoch": 9.88,
"step": 200,
"total_flos": 6.325810471729889e+17,
"train_loss": 0.15392300799489023,
"train_runtime": 432.1174,
"train_samples_per_second": 59.59,
"train_steps_per_second": 0.463
}
],
"logging_steps": 5,
"max_steps": 200,
"num_train_epochs": 10,
"save_steps": 500,
"total_flos": 6.325810471729889e+17,
"trial_name": null,
"trial_params": null
}