mansee's picture
End of training
173e935
raw
history blame
7.91 kB
{
"best_metric": 0.9825783972125436,
"best_model_checkpoint": "swin-tiny-patch4-window7-224-spa_saloon_classification-spa-saloon/checkpoint-101",
"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": 0.3456,
"step": 5
},
{
"epoch": 0.49,
"learning_rate": 2.5e-05,
"loss": 0.2937,
"step": 10
},
{
"epoch": 0.74,
"learning_rate": 3.7500000000000003e-05,
"loss": 0.3306,
"step": 15
},
{
"epoch": 0.99,
"learning_rate": 5e-05,
"loss": 0.2585,
"step": 20
},
{
"epoch": 0.99,
"eval_accuracy": 0.9407665505226481,
"eval_loss": 0.16160520911216736,
"eval_runtime": 1.8112,
"eval_samples_per_second": 158.456,
"eval_steps_per_second": 4.969,
"step": 20
},
{
"epoch": 1.23,
"learning_rate": 4.8611111111111115e-05,
"loss": 0.2229,
"step": 25
},
{
"epoch": 1.48,
"learning_rate": 4.722222222222222e-05,
"loss": 0.2354,
"step": 30
},
{
"epoch": 1.73,
"learning_rate": 4.5833333333333334e-05,
"loss": 0.1735,
"step": 35
},
{
"epoch": 1.98,
"learning_rate": 4.4444444444444447e-05,
"loss": 0.2042,
"step": 40
},
{
"epoch": 1.98,
"eval_accuracy": 0.9337979094076655,
"eval_loss": 0.2162116914987564,
"eval_runtime": 1.8419,
"eval_samples_per_second": 155.818,
"eval_steps_per_second": 4.886,
"step": 40
},
{
"epoch": 2.22,
"learning_rate": 4.305555555555556e-05,
"loss": 0.2087,
"step": 45
},
{
"epoch": 2.47,
"learning_rate": 4.166666666666667e-05,
"loss": 0.1561,
"step": 50
},
{
"epoch": 2.72,
"learning_rate": 4.027777777777778e-05,
"loss": 0.1463,
"step": 55
},
{
"epoch": 2.96,
"learning_rate": 3.888888888888889e-05,
"loss": 0.1464,
"step": 60
},
{
"epoch": 2.96,
"eval_accuracy": 0.9721254355400697,
"eval_loss": 0.10012032091617584,
"eval_runtime": 1.9543,
"eval_samples_per_second": 146.855,
"eval_steps_per_second": 4.605,
"step": 60
},
{
"epoch": 3.21,
"learning_rate": 3.7500000000000003e-05,
"loss": 0.1456,
"step": 65
},
{
"epoch": 3.46,
"learning_rate": 3.611111111111111e-05,
"loss": 0.1674,
"step": 70
},
{
"epoch": 3.7,
"learning_rate": 3.472222222222222e-05,
"loss": 0.1502,
"step": 75
},
{
"epoch": 3.95,
"learning_rate": 3.3333333333333335e-05,
"loss": 0.1621,
"step": 80
},
{
"epoch": 4.0,
"eval_accuracy": 0.9790940766550522,
"eval_loss": 0.09151919186115265,
"eval_runtime": 1.8539,
"eval_samples_per_second": 154.81,
"eval_steps_per_second": 4.855,
"step": 81
},
{
"epoch": 4.2,
"learning_rate": 3.194444444444444e-05,
"loss": 0.1445,
"step": 85
},
{
"epoch": 4.44,
"learning_rate": 3.055555555555556e-05,
"loss": 0.1356,
"step": 90
},
{
"epoch": 4.69,
"learning_rate": 2.916666666666667e-05,
"loss": 0.1197,
"step": 95
},
{
"epoch": 4.94,
"learning_rate": 2.777777777777778e-05,
"loss": 0.1469,
"step": 100
},
{
"epoch": 4.99,
"eval_accuracy": 0.9825783972125436,
"eval_loss": 0.07965687662363052,
"eval_runtime": 2.379,
"eval_samples_per_second": 120.638,
"eval_steps_per_second": 3.783,
"step": 101
},
{
"epoch": 5.19,
"learning_rate": 2.6388888888888892e-05,
"loss": 0.1085,
"step": 105
},
{
"epoch": 5.43,
"learning_rate": 2.5e-05,
"loss": 0.1257,
"step": 110
},
{
"epoch": 5.68,
"learning_rate": 2.361111111111111e-05,
"loss": 0.1157,
"step": 115
},
{
"epoch": 5.93,
"learning_rate": 2.2222222222222223e-05,
"loss": 0.1272,
"step": 120
},
{
"epoch": 5.98,
"eval_accuracy": 0.975609756097561,
"eval_loss": 0.07534805685281754,
"eval_runtime": 1.8754,
"eval_samples_per_second": 153.031,
"eval_steps_per_second": 4.799,
"step": 121
},
{
"epoch": 6.17,
"learning_rate": 2.0833333333333336e-05,
"loss": 0.093,
"step": 125
},
{
"epoch": 6.42,
"learning_rate": 1.9444444444444445e-05,
"loss": 0.0924,
"step": 130
},
{
"epoch": 6.67,
"learning_rate": 1.8055555555555555e-05,
"loss": 0.1556,
"step": 135
},
{
"epoch": 6.91,
"learning_rate": 1.6666666666666667e-05,
"loss": 0.0985,
"step": 140
},
{
"epoch": 6.96,
"eval_accuracy": 0.9790940766550522,
"eval_loss": 0.08600831031799316,
"eval_runtime": 2.1983,
"eval_samples_per_second": 130.553,
"eval_steps_per_second": 4.094,
"step": 141
},
{
"epoch": 7.16,
"learning_rate": 1.527777777777778e-05,
"loss": 0.1133,
"step": 145
},
{
"epoch": 7.41,
"learning_rate": 1.388888888888889e-05,
"loss": 0.0928,
"step": 150
},
{
"epoch": 7.65,
"learning_rate": 1.25e-05,
"loss": 0.0923,
"step": 155
},
{
"epoch": 7.9,
"learning_rate": 1.1111111111111112e-05,
"loss": 0.1013,
"step": 160
},
{
"epoch": 8.0,
"eval_accuracy": 0.9651567944250871,
"eval_loss": 0.11782138049602509,
"eval_runtime": 1.8695,
"eval_samples_per_second": 153.521,
"eval_steps_per_second": 4.814,
"step": 162
},
{
"epoch": 8.15,
"learning_rate": 9.722222222222223e-06,
"loss": 0.1086,
"step": 165
},
{
"epoch": 8.4,
"learning_rate": 8.333333333333334e-06,
"loss": 0.1161,
"step": 170
},
{
"epoch": 8.64,
"learning_rate": 6.944444444444445e-06,
"loss": 0.0612,
"step": 175
},
{
"epoch": 8.89,
"learning_rate": 5.555555555555556e-06,
"loss": 0.111,
"step": 180
},
{
"epoch": 8.99,
"eval_accuracy": 0.9651567944250871,
"eval_loss": 0.10362739115953445,
"eval_runtime": 2.1015,
"eval_samples_per_second": 136.567,
"eval_steps_per_second": 4.283,
"step": 182
},
{
"epoch": 9.14,
"learning_rate": 4.166666666666667e-06,
"loss": 0.1008,
"step": 185
},
{
"epoch": 9.38,
"learning_rate": 2.777777777777778e-06,
"loss": 0.1,
"step": 190
},
{
"epoch": 9.63,
"learning_rate": 1.388888888888889e-06,
"loss": 0.0694,
"step": 195
},
{
"epoch": 9.88,
"learning_rate": 0.0,
"loss": 0.0737,
"step": 200
},
{
"epoch": 9.88,
"eval_accuracy": 0.9686411149825784,
"eval_loss": 0.09821867197751999,
"eval_runtime": 1.893,
"eval_samples_per_second": 151.615,
"eval_steps_per_second": 4.754,
"step": 200
},
{
"epoch": 9.88,
"step": 200,
"total_flos": 6.325810471729889e+17,
"train_loss": 0.1487744191288948,
"train_runtime": 421.2195,
"train_samples_per_second": 61.132,
"train_steps_per_second": 0.475
}
],
"logging_steps": 5,
"max_steps": 200,
"num_train_epochs": 10,
"save_steps": 500,
"total_flos": 6.325810471729889e+17,
"trial_name": null,
"trial_params": null
}