|
{ |
|
"best_metric": 0.9738372093023255, |
|
"best_model_checkpoint": "swin-tiny-patch4-window7-224-blank_img/checkpoint-360", |
|
"epoch": 4.948453608247423, |
|
"eval_steps": 500, |
|
"global_step": 360, |
|
"is_hyper_param_search": false, |
|
"is_local_process_zero": true, |
|
"is_world_process_zero": true, |
|
"log_history": [ |
|
{ |
|
"epoch": 0.07, |
|
"learning_rate": 6.944444444444445e-06, |
|
"loss": 0.0881, |
|
"step": 5 |
|
}, |
|
{ |
|
"epoch": 0.14, |
|
"learning_rate": 1.388888888888889e-05, |
|
"loss": 0.0624, |
|
"step": 10 |
|
}, |
|
{ |
|
"epoch": 0.21, |
|
"learning_rate": 2.0833333333333336e-05, |
|
"loss": 0.063, |
|
"step": 15 |
|
}, |
|
{ |
|
"epoch": 0.27, |
|
"learning_rate": 2.777777777777778e-05, |
|
"loss": 0.0901, |
|
"step": 20 |
|
}, |
|
{ |
|
"epoch": 0.34, |
|
"learning_rate": 3.472222222222222e-05, |
|
"loss": 0.0872, |
|
"step": 25 |
|
}, |
|
{ |
|
"epoch": 0.41, |
|
"learning_rate": 4.166666666666667e-05, |
|
"loss": 0.1016, |
|
"step": 30 |
|
}, |
|
{ |
|
"epoch": 0.48, |
|
"learning_rate": 4.8611111111111115e-05, |
|
"loss": 0.0905, |
|
"step": 35 |
|
}, |
|
{ |
|
"epoch": 0.55, |
|
"learning_rate": 4.938271604938271e-05, |
|
"loss": 0.1495, |
|
"step": 40 |
|
}, |
|
{ |
|
"epoch": 0.62, |
|
"learning_rate": 4.8611111111111115e-05, |
|
"loss": 0.0949, |
|
"step": 45 |
|
}, |
|
{ |
|
"epoch": 0.69, |
|
"learning_rate": 4.783950617283951e-05, |
|
"loss": 0.1178, |
|
"step": 50 |
|
}, |
|
{ |
|
"epoch": 0.76, |
|
"learning_rate": 4.70679012345679e-05, |
|
"loss": 0.08, |
|
"step": 55 |
|
}, |
|
{ |
|
"epoch": 0.82, |
|
"learning_rate": 4.62962962962963e-05, |
|
"loss": 0.0919, |
|
"step": 60 |
|
}, |
|
{ |
|
"epoch": 0.89, |
|
"learning_rate": 4.5524691358024696e-05, |
|
"loss": 0.0773, |
|
"step": 65 |
|
}, |
|
{ |
|
"epoch": 0.96, |
|
"learning_rate": 4.4753086419753084e-05, |
|
"loss": 0.0502, |
|
"step": 70 |
|
}, |
|
{ |
|
"epoch": 0.99, |
|
"eval_accuracy": 0.9651162790697675, |
|
"eval_loss": 0.12999847531318665, |
|
"eval_runtime": 6.6038, |
|
"eval_samples_per_second": 156.273, |
|
"eval_steps_per_second": 4.997, |
|
"step": 72 |
|
}, |
|
{ |
|
"epoch": 1.03, |
|
"learning_rate": 4.3981481481481486e-05, |
|
"loss": 0.0838, |
|
"step": 75 |
|
}, |
|
{ |
|
"epoch": 1.1, |
|
"learning_rate": 4.3209876543209875e-05, |
|
"loss": 0.067, |
|
"step": 80 |
|
}, |
|
{ |
|
"epoch": 1.17, |
|
"learning_rate": 4.243827160493827e-05, |
|
"loss": 0.0932, |
|
"step": 85 |
|
}, |
|
{ |
|
"epoch": 1.24, |
|
"learning_rate": 4.166666666666667e-05, |
|
"loss": 0.1037, |
|
"step": 90 |
|
}, |
|
{ |
|
"epoch": 1.31, |
|
"learning_rate": 4.089506172839506e-05, |
|
"loss": 0.1057, |
|
"step": 95 |
|
}, |
|
{ |
|
"epoch": 1.37, |
|
"learning_rate": 4.012345679012346e-05, |
|
"loss": 0.0774, |
|
"step": 100 |
|
}, |
|
{ |
|
"epoch": 1.44, |
|
"learning_rate": 3.935185185185186e-05, |
|
"loss": 0.1107, |
|
"step": 105 |
|
}, |
|
{ |
|
"epoch": 1.51, |
|
"learning_rate": 3.8580246913580246e-05, |
|
"loss": 0.1109, |
|
"step": 110 |
|
}, |
|
{ |
|
"epoch": 1.58, |
|
"learning_rate": 3.780864197530865e-05, |
|
"loss": 0.1078, |
|
"step": 115 |
|
}, |
|
{ |
|
"epoch": 1.65, |
|
"learning_rate": 3.7037037037037037e-05, |
|
"loss": 0.0935, |
|
"step": 120 |
|
}, |
|
{ |
|
"epoch": 1.72, |
|
"learning_rate": 3.626543209876543e-05, |
|
"loss": 0.0741, |
|
"step": 125 |
|
}, |
|
{ |
|
"epoch": 1.79, |
|
"learning_rate": 3.5493827160493834e-05, |
|
"loss": 0.1031, |
|
"step": 130 |
|
}, |
|
{ |
|
"epoch": 1.86, |
|
"learning_rate": 3.472222222222222e-05, |
|
"loss": 0.0638, |
|
"step": 135 |
|
}, |
|
{ |
|
"epoch": 1.92, |
|
"learning_rate": 3.395061728395062e-05, |
|
"loss": 0.0915, |
|
"step": 140 |
|
}, |
|
{ |
|
"epoch": 1.99, |
|
"learning_rate": 3.317901234567901e-05, |
|
"loss": 0.1107, |
|
"step": 145 |
|
}, |
|
{ |
|
"epoch": 1.99, |
|
"eval_accuracy": 0.9728682170542635, |
|
"eval_loss": 0.10234300047159195, |
|
"eval_runtime": 7.16, |
|
"eval_samples_per_second": 144.135, |
|
"eval_steps_per_second": 4.609, |
|
"step": 145 |
|
}, |
|
{ |
|
"epoch": 2.06, |
|
"learning_rate": 3.240740740740741e-05, |
|
"loss": 0.0784, |
|
"step": 150 |
|
}, |
|
{ |
|
"epoch": 2.13, |
|
"learning_rate": 3.16358024691358e-05, |
|
"loss": 0.0672, |
|
"step": 155 |
|
}, |
|
{ |
|
"epoch": 2.2, |
|
"learning_rate": 3.08641975308642e-05, |
|
"loss": 0.0598, |
|
"step": 160 |
|
}, |
|
{ |
|
"epoch": 2.27, |
|
"learning_rate": 3.0092592592592593e-05, |
|
"loss": 0.0641, |
|
"step": 165 |
|
}, |
|
{ |
|
"epoch": 2.34, |
|
"learning_rate": 2.9320987654320992e-05, |
|
"loss": 0.0805, |
|
"step": 170 |
|
}, |
|
{ |
|
"epoch": 2.41, |
|
"learning_rate": 2.8549382716049384e-05, |
|
"loss": 0.0864, |
|
"step": 175 |
|
}, |
|
{ |
|
"epoch": 2.47, |
|
"learning_rate": 2.777777777777778e-05, |
|
"loss": 0.0954, |
|
"step": 180 |
|
}, |
|
{ |
|
"epoch": 2.54, |
|
"learning_rate": 2.700617283950617e-05, |
|
"loss": 0.0717, |
|
"step": 185 |
|
}, |
|
{ |
|
"epoch": 2.61, |
|
"learning_rate": 2.623456790123457e-05, |
|
"loss": 0.0634, |
|
"step": 190 |
|
}, |
|
{ |
|
"epoch": 2.68, |
|
"learning_rate": 2.5462962962962965e-05, |
|
"loss": 0.1082, |
|
"step": 195 |
|
}, |
|
{ |
|
"epoch": 2.75, |
|
"learning_rate": 2.4691358024691357e-05, |
|
"loss": 0.0748, |
|
"step": 200 |
|
}, |
|
{ |
|
"epoch": 2.82, |
|
"learning_rate": 2.3919753086419755e-05, |
|
"loss": 0.1221, |
|
"step": 205 |
|
}, |
|
{ |
|
"epoch": 2.89, |
|
"learning_rate": 2.314814814814815e-05, |
|
"loss": 0.0757, |
|
"step": 210 |
|
}, |
|
{ |
|
"epoch": 2.96, |
|
"learning_rate": 2.2376543209876542e-05, |
|
"loss": 0.0917, |
|
"step": 215 |
|
}, |
|
{ |
|
"epoch": 3.0, |
|
"eval_accuracy": 0.9651162790697675, |
|
"eval_loss": 0.12774048745632172, |
|
"eval_runtime": 6.6558, |
|
"eval_samples_per_second": 155.053, |
|
"eval_steps_per_second": 4.958, |
|
"step": 218 |
|
}, |
|
{ |
|
"epoch": 3.02, |
|
"learning_rate": 2.1604938271604937e-05, |
|
"loss": 0.0553, |
|
"step": 220 |
|
}, |
|
{ |
|
"epoch": 3.09, |
|
"learning_rate": 2.0833333333333336e-05, |
|
"loss": 0.0942, |
|
"step": 225 |
|
}, |
|
{ |
|
"epoch": 3.16, |
|
"learning_rate": 2.006172839506173e-05, |
|
"loss": 0.0834, |
|
"step": 230 |
|
}, |
|
{ |
|
"epoch": 3.23, |
|
"learning_rate": 1.9290123456790123e-05, |
|
"loss": 0.057, |
|
"step": 235 |
|
}, |
|
{ |
|
"epoch": 3.3, |
|
"learning_rate": 1.8518518518518518e-05, |
|
"loss": 0.0773, |
|
"step": 240 |
|
}, |
|
{ |
|
"epoch": 3.37, |
|
"learning_rate": 1.7746913580246917e-05, |
|
"loss": 0.0617, |
|
"step": 245 |
|
}, |
|
{ |
|
"epoch": 3.44, |
|
"learning_rate": 1.697530864197531e-05, |
|
"loss": 0.0785, |
|
"step": 250 |
|
}, |
|
{ |
|
"epoch": 3.51, |
|
"learning_rate": 1.6203703703703704e-05, |
|
"loss": 0.057, |
|
"step": 255 |
|
}, |
|
{ |
|
"epoch": 3.57, |
|
"learning_rate": 1.54320987654321e-05, |
|
"loss": 0.0746, |
|
"step": 260 |
|
}, |
|
{ |
|
"epoch": 3.64, |
|
"learning_rate": 1.4660493827160496e-05, |
|
"loss": 0.1117, |
|
"step": 265 |
|
}, |
|
{ |
|
"epoch": 3.71, |
|
"learning_rate": 1.388888888888889e-05, |
|
"loss": 0.0923, |
|
"step": 270 |
|
}, |
|
{ |
|
"epoch": 3.78, |
|
"learning_rate": 1.3117283950617285e-05, |
|
"loss": 0.0599, |
|
"step": 275 |
|
}, |
|
{ |
|
"epoch": 3.85, |
|
"learning_rate": 1.2345679012345678e-05, |
|
"loss": 0.0895, |
|
"step": 280 |
|
}, |
|
{ |
|
"epoch": 3.92, |
|
"learning_rate": 1.1574074074074075e-05, |
|
"loss": 0.0933, |
|
"step": 285 |
|
}, |
|
{ |
|
"epoch": 3.99, |
|
"learning_rate": 1.0802469135802469e-05, |
|
"loss": 0.1022, |
|
"step": 290 |
|
}, |
|
{ |
|
"epoch": 4.0, |
|
"eval_accuracy": 0.9718992248062015, |
|
"eval_loss": 0.12579788267612457, |
|
"eval_runtime": 7.0589, |
|
"eval_samples_per_second": 146.199, |
|
"eval_steps_per_second": 4.675, |
|
"step": 291 |
|
}, |
|
{ |
|
"epoch": 4.05, |
|
"learning_rate": 1.0030864197530866e-05, |
|
"loss": 0.0889, |
|
"step": 295 |
|
}, |
|
{ |
|
"epoch": 4.12, |
|
"learning_rate": 9.259259259259259e-06, |
|
"loss": 0.088, |
|
"step": 300 |
|
}, |
|
{ |
|
"epoch": 4.19, |
|
"learning_rate": 8.487654320987654e-06, |
|
"loss": 0.0679, |
|
"step": 305 |
|
}, |
|
{ |
|
"epoch": 4.26, |
|
"learning_rate": 7.71604938271605e-06, |
|
"loss": 0.0632, |
|
"step": 310 |
|
}, |
|
{ |
|
"epoch": 4.33, |
|
"learning_rate": 6.944444444444445e-06, |
|
"loss": 0.095, |
|
"step": 315 |
|
}, |
|
{ |
|
"epoch": 4.4, |
|
"learning_rate": 6.172839506172839e-06, |
|
"loss": 0.084, |
|
"step": 320 |
|
}, |
|
{ |
|
"epoch": 4.47, |
|
"learning_rate": 5.401234567901234e-06, |
|
"loss": 0.07, |
|
"step": 325 |
|
}, |
|
{ |
|
"epoch": 4.54, |
|
"learning_rate": 4.6296296296296296e-06, |
|
"loss": 0.0692, |
|
"step": 330 |
|
}, |
|
{ |
|
"epoch": 4.6, |
|
"learning_rate": 3.858024691358025e-06, |
|
"loss": 0.0575, |
|
"step": 335 |
|
}, |
|
{ |
|
"epoch": 4.67, |
|
"learning_rate": 3.0864197530864196e-06, |
|
"loss": 0.0625, |
|
"step": 340 |
|
}, |
|
{ |
|
"epoch": 4.74, |
|
"learning_rate": 2.3148148148148148e-06, |
|
"loss": 0.0715, |
|
"step": 345 |
|
}, |
|
{ |
|
"epoch": 4.81, |
|
"learning_rate": 1.5432098765432098e-06, |
|
"loss": 0.0651, |
|
"step": 350 |
|
}, |
|
{ |
|
"epoch": 4.88, |
|
"learning_rate": 7.716049382716049e-07, |
|
"loss": 0.1432, |
|
"step": 355 |
|
}, |
|
{ |
|
"epoch": 4.95, |
|
"learning_rate": 0.0, |
|
"loss": 0.0888, |
|
"step": 360 |
|
}, |
|
{ |
|
"epoch": 4.95, |
|
"eval_accuracy": 0.9738372093023255, |
|
"eval_loss": 0.10162452608346939, |
|
"eval_runtime": 6.9642, |
|
"eval_samples_per_second": 148.186, |
|
"eval_steps_per_second": 4.739, |
|
"step": 360 |
|
}, |
|
{ |
|
"epoch": 4.95, |
|
"step": 360, |
|
"total_flos": 1.1422818298339983e+18, |
|
"train_loss": 0.08449313590923944, |
|
"train_runtime": 710.4711, |
|
"train_samples_per_second": 65.316, |
|
"train_steps_per_second": 0.507 |
|
} |
|
], |
|
"logging_steps": 5, |
|
"max_steps": 360, |
|
"num_train_epochs": 5, |
|
"save_steps": 500, |
|
"total_flos": 1.1422818298339983e+18, |
|
"trial_name": null, |
|
"trial_params": null |
|
} |
|
|