|
{ |
|
"best_metric": 0.8493150684931506, |
|
"best_model_checkpoint": "swin-base-patch4-window7-224-finetuned-piid/checkpoint-348", |
|
"epoch": 19.51219512195122, |
|
"eval_steps": 500, |
|
"global_step": 400, |
|
"is_hyper_param_search": false, |
|
"is_local_process_zero": true, |
|
"is_world_process_zero": true, |
|
"log_history": [ |
|
{ |
|
"epoch": 0.49, |
|
"learning_rate": 1.25e-05, |
|
"loss": 1.4202, |
|
"step": 10 |
|
}, |
|
{ |
|
"epoch": 0.98, |
|
"learning_rate": 2.5e-05, |
|
"loss": 1.1815, |
|
"step": 20 |
|
}, |
|
{ |
|
"epoch": 0.98, |
|
"eval_accuracy": 0.5251141552511416, |
|
"eval_loss": 1.0441278219223022, |
|
"eval_runtime": 2.969, |
|
"eval_samples_per_second": 73.762, |
|
"eval_steps_per_second": 9.431, |
|
"step": 20 |
|
}, |
|
{ |
|
"epoch": 1.46, |
|
"learning_rate": 3.7500000000000003e-05, |
|
"loss": 0.8542, |
|
"step": 30 |
|
}, |
|
{ |
|
"epoch": 1.95, |
|
"learning_rate": 5e-05, |
|
"loss": 0.6548, |
|
"step": 40 |
|
}, |
|
{ |
|
"epoch": 2.0, |
|
"eval_accuracy": 0.639269406392694, |
|
"eval_loss": 0.8150126934051514, |
|
"eval_runtime": 2.9105, |
|
"eval_samples_per_second": 75.245, |
|
"eval_steps_per_second": 9.62, |
|
"step": 41 |
|
}, |
|
{ |
|
"epoch": 2.44, |
|
"learning_rate": 4.8611111111111115e-05, |
|
"loss": 0.51, |
|
"step": 50 |
|
}, |
|
{ |
|
"epoch": 2.93, |
|
"learning_rate": 4.722222222222222e-05, |
|
"loss": 0.6083, |
|
"step": 60 |
|
}, |
|
{ |
|
"epoch": 2.98, |
|
"eval_accuracy": 0.6986301369863014, |
|
"eval_loss": 0.6395346522331238, |
|
"eval_runtime": 2.9853, |
|
"eval_samples_per_second": 73.36, |
|
"eval_steps_per_second": 9.379, |
|
"step": 61 |
|
}, |
|
{ |
|
"epoch": 3.41, |
|
"learning_rate": 4.5833333333333334e-05, |
|
"loss": 0.4977, |
|
"step": 70 |
|
}, |
|
{ |
|
"epoch": 3.9, |
|
"learning_rate": 4.4444444444444447e-05, |
|
"loss": 0.4925, |
|
"step": 80 |
|
}, |
|
{ |
|
"epoch": 4.0, |
|
"eval_accuracy": 0.680365296803653, |
|
"eval_loss": 0.6272767782211304, |
|
"eval_runtime": 3.0039, |
|
"eval_samples_per_second": 72.904, |
|
"eval_steps_per_second": 9.321, |
|
"step": 82 |
|
}, |
|
{ |
|
"epoch": 4.39, |
|
"learning_rate": 4.305555555555556e-05, |
|
"loss": 0.5078, |
|
"step": 90 |
|
}, |
|
{ |
|
"epoch": 4.88, |
|
"learning_rate": 4.166666666666667e-05, |
|
"loss": 0.4448, |
|
"step": 100 |
|
}, |
|
{ |
|
"epoch": 4.98, |
|
"eval_accuracy": 0.817351598173516, |
|
"eval_loss": 0.48121878504753113, |
|
"eval_runtime": 3.0486, |
|
"eval_samples_per_second": 71.836, |
|
"eval_steps_per_second": 9.185, |
|
"step": 102 |
|
}, |
|
{ |
|
"epoch": 5.37, |
|
"learning_rate": 4.027777777777778e-05, |
|
"loss": 0.4302, |
|
"step": 110 |
|
}, |
|
{ |
|
"epoch": 5.85, |
|
"learning_rate": 3.888888888888889e-05, |
|
"loss": 0.3387, |
|
"step": 120 |
|
}, |
|
{ |
|
"epoch": 6.0, |
|
"eval_accuracy": 0.7945205479452054, |
|
"eval_loss": 0.586761474609375, |
|
"eval_runtime": 3.0271, |
|
"eval_samples_per_second": 72.346, |
|
"eval_steps_per_second": 9.25, |
|
"step": 123 |
|
}, |
|
{ |
|
"epoch": 6.34, |
|
"learning_rate": 3.7500000000000003e-05, |
|
"loss": 0.3207, |
|
"step": 130 |
|
}, |
|
{ |
|
"epoch": 6.83, |
|
"learning_rate": 3.611111111111111e-05, |
|
"loss": 0.2622, |
|
"step": 140 |
|
}, |
|
{ |
|
"epoch": 6.98, |
|
"eval_accuracy": 0.726027397260274, |
|
"eval_loss": 0.7868249416351318, |
|
"eval_runtime": 3.0111, |
|
"eval_samples_per_second": 72.731, |
|
"eval_steps_per_second": 9.299, |
|
"step": 143 |
|
}, |
|
{ |
|
"epoch": 7.32, |
|
"learning_rate": 3.472222222222222e-05, |
|
"loss": 0.3422, |
|
"step": 150 |
|
}, |
|
{ |
|
"epoch": 7.8, |
|
"learning_rate": 3.3333333333333335e-05, |
|
"loss": 0.2656, |
|
"step": 160 |
|
}, |
|
{ |
|
"epoch": 8.0, |
|
"eval_accuracy": 0.8127853881278538, |
|
"eval_loss": 0.4432077705860138, |
|
"eval_runtime": 3.1504, |
|
"eval_samples_per_second": 69.514, |
|
"eval_steps_per_second": 8.888, |
|
"step": 164 |
|
}, |
|
{ |
|
"epoch": 8.29, |
|
"learning_rate": 3.194444444444444e-05, |
|
"loss": 0.2049, |
|
"step": 170 |
|
}, |
|
{ |
|
"epoch": 8.78, |
|
"learning_rate": 3.055555555555556e-05, |
|
"loss": 0.2259, |
|
"step": 180 |
|
}, |
|
{ |
|
"epoch": 8.98, |
|
"eval_accuracy": 0.7488584474885844, |
|
"eval_loss": 0.6553168892860413, |
|
"eval_runtime": 3.0029, |
|
"eval_samples_per_second": 72.929, |
|
"eval_steps_per_second": 9.324, |
|
"step": 184 |
|
}, |
|
{ |
|
"epoch": 9.27, |
|
"learning_rate": 2.916666666666667e-05, |
|
"loss": 0.239, |
|
"step": 190 |
|
}, |
|
{ |
|
"epoch": 9.76, |
|
"learning_rate": 2.777777777777778e-05, |
|
"loss": 0.1997, |
|
"step": 200 |
|
}, |
|
{ |
|
"epoch": 10.0, |
|
"eval_accuracy": 0.7853881278538812, |
|
"eval_loss": 0.5142815709114075, |
|
"eval_runtime": 3.172, |
|
"eval_samples_per_second": 69.041, |
|
"eval_steps_per_second": 8.827, |
|
"step": 205 |
|
}, |
|
{ |
|
"epoch": 10.24, |
|
"learning_rate": 2.6388888888888892e-05, |
|
"loss": 0.1559, |
|
"step": 210 |
|
}, |
|
{ |
|
"epoch": 10.73, |
|
"learning_rate": 2.5e-05, |
|
"loss": 0.1892, |
|
"step": 220 |
|
}, |
|
{ |
|
"epoch": 10.98, |
|
"eval_accuracy": 0.7945205479452054, |
|
"eval_loss": 0.5656914114952087, |
|
"eval_runtime": 3.0489, |
|
"eval_samples_per_second": 71.829, |
|
"eval_steps_per_second": 9.184, |
|
"step": 225 |
|
}, |
|
{ |
|
"epoch": 11.22, |
|
"learning_rate": 2.361111111111111e-05, |
|
"loss": 0.1942, |
|
"step": 230 |
|
}, |
|
{ |
|
"epoch": 11.71, |
|
"learning_rate": 2.2222222222222223e-05, |
|
"loss": 0.1522, |
|
"step": 240 |
|
}, |
|
{ |
|
"epoch": 12.0, |
|
"eval_accuracy": 0.7579908675799086, |
|
"eval_loss": 0.7339126467704773, |
|
"eval_runtime": 3.1764, |
|
"eval_samples_per_second": 68.946, |
|
"eval_steps_per_second": 8.815, |
|
"step": 246 |
|
}, |
|
{ |
|
"epoch": 12.2, |
|
"learning_rate": 2.0833333333333336e-05, |
|
"loss": 0.1607, |
|
"step": 250 |
|
}, |
|
{ |
|
"epoch": 12.68, |
|
"learning_rate": 1.9444444444444445e-05, |
|
"loss": 0.1309, |
|
"step": 260 |
|
}, |
|
{ |
|
"epoch": 12.98, |
|
"eval_accuracy": 0.817351598173516, |
|
"eval_loss": 0.6064341068267822, |
|
"eval_runtime": 3.034, |
|
"eval_samples_per_second": 72.182, |
|
"eval_steps_per_second": 9.229, |
|
"step": 266 |
|
}, |
|
{ |
|
"epoch": 13.17, |
|
"learning_rate": 1.8055555555555555e-05, |
|
"loss": 0.1803, |
|
"step": 270 |
|
}, |
|
{ |
|
"epoch": 13.66, |
|
"learning_rate": 1.6666666666666667e-05, |
|
"loss": 0.1482, |
|
"step": 280 |
|
}, |
|
{ |
|
"epoch": 14.0, |
|
"eval_accuracy": 0.8127853881278538, |
|
"eval_loss": 0.5874872207641602, |
|
"eval_runtime": 3.1935, |
|
"eval_samples_per_second": 68.576, |
|
"eval_steps_per_second": 8.768, |
|
"step": 287 |
|
}, |
|
{ |
|
"epoch": 14.15, |
|
"learning_rate": 1.527777777777778e-05, |
|
"loss": 0.1641, |
|
"step": 290 |
|
}, |
|
{ |
|
"epoch": 14.63, |
|
"learning_rate": 1.388888888888889e-05, |
|
"loss": 0.1459, |
|
"step": 300 |
|
}, |
|
{ |
|
"epoch": 14.98, |
|
"eval_accuracy": 0.7899543378995434, |
|
"eval_loss": 0.6442859768867493, |
|
"eval_runtime": 3.0922, |
|
"eval_samples_per_second": 70.824, |
|
"eval_steps_per_second": 9.055, |
|
"step": 307 |
|
}, |
|
{ |
|
"epoch": 15.12, |
|
"learning_rate": 1.25e-05, |
|
"loss": 0.136, |
|
"step": 310 |
|
}, |
|
{ |
|
"epoch": 15.61, |
|
"learning_rate": 1.1111111111111112e-05, |
|
"loss": 0.1224, |
|
"step": 320 |
|
}, |
|
{ |
|
"epoch": 16.0, |
|
"eval_accuracy": 0.8036529680365296, |
|
"eval_loss": 0.652053713798523, |
|
"eval_runtime": 3.1086, |
|
"eval_samples_per_second": 70.45, |
|
"eval_steps_per_second": 9.007, |
|
"step": 328 |
|
}, |
|
{ |
|
"epoch": 16.1, |
|
"learning_rate": 9.722222222222223e-06, |
|
"loss": 0.1393, |
|
"step": 330 |
|
}, |
|
{ |
|
"epoch": 16.59, |
|
"learning_rate": 8.333333333333334e-06, |
|
"loss": 0.0533, |
|
"step": 340 |
|
}, |
|
{ |
|
"epoch": 16.98, |
|
"eval_accuracy": 0.8493150684931506, |
|
"eval_loss": 0.5914698243141174, |
|
"eval_runtime": 3.1723, |
|
"eval_samples_per_second": 69.034, |
|
"eval_steps_per_second": 8.826, |
|
"step": 348 |
|
}, |
|
{ |
|
"epoch": 17.07, |
|
"learning_rate": 6.944444444444445e-06, |
|
"loss": 0.1186, |
|
"step": 350 |
|
}, |
|
{ |
|
"epoch": 17.56, |
|
"learning_rate": 5.555555555555556e-06, |
|
"loss": 0.1133, |
|
"step": 360 |
|
}, |
|
{ |
|
"epoch": 18.0, |
|
"eval_accuracy": 0.8264840182648402, |
|
"eval_loss": 0.6151906251907349, |
|
"eval_runtime": 3.0754, |
|
"eval_samples_per_second": 71.21, |
|
"eval_steps_per_second": 9.104, |
|
"step": 369 |
|
}, |
|
{ |
|
"epoch": 18.05, |
|
"learning_rate": 4.166666666666667e-06, |
|
"loss": 0.1631, |
|
"step": 370 |
|
}, |
|
{ |
|
"epoch": 18.54, |
|
"learning_rate": 2.777777777777778e-06, |
|
"loss": 0.0923, |
|
"step": 380 |
|
}, |
|
{ |
|
"epoch": 18.98, |
|
"eval_accuracy": 0.7853881278538812, |
|
"eval_loss": 0.6819412112236023, |
|
"eval_runtime": 3.2565, |
|
"eval_samples_per_second": 67.25, |
|
"eval_steps_per_second": 8.598, |
|
"step": 389 |
|
}, |
|
{ |
|
"epoch": 19.02, |
|
"learning_rate": 1.388888888888889e-06, |
|
"loss": 0.1299, |
|
"step": 390 |
|
}, |
|
{ |
|
"epoch": 19.51, |
|
"learning_rate": 0.0, |
|
"loss": 0.086, |
|
"step": 400 |
|
}, |
|
{ |
|
"epoch": 19.51, |
|
"eval_accuracy": 0.8127853881278538, |
|
"eval_loss": 0.6629763245582581, |
|
"eval_runtime": 3.0498, |
|
"eval_samples_per_second": 71.808, |
|
"eval_steps_per_second": 9.181, |
|
"step": 400 |
|
}, |
|
{ |
|
"epoch": 19.51, |
|
"step": 400, |
|
"total_flos": 9.98381008961151e+17, |
|
"train_loss": 0.31942258909344673, |
|
"train_runtime": 678.6178, |
|
"train_samples_per_second": 19.245, |
|
"train_steps_per_second": 0.589 |
|
} |
|
], |
|
"logging_steps": 10, |
|
"max_steps": 400, |
|
"num_train_epochs": 20, |
|
"save_steps": 500, |
|
"total_flos": 9.98381008961151e+17, |
|
"trial_name": null, |
|
"trial_params": null |
|
} |
|
|