|
{ |
|
"best_metric": 0.9732785200411099, |
|
"best_model_checkpoint": "swinv2-large-patch4-window12to16-192to256-22kto1k-ft-finetuned-eurosat-50/checkpoint-549", |
|
"epoch": 10.0, |
|
"global_step": 610, |
|
"is_hyper_param_search": false, |
|
"is_local_process_zero": true, |
|
"is_world_process_zero": true, |
|
"log_history": [ |
|
{ |
|
"epoch": 0.16, |
|
"learning_rate": 3.278688524590164e-07, |
|
"loss": 2.1255, |
|
"step": 10 |
|
}, |
|
{ |
|
"epoch": 0.33, |
|
"learning_rate": 6.557377049180328e-07, |
|
"loss": 2.045, |
|
"step": 20 |
|
}, |
|
{ |
|
"epoch": 0.49, |
|
"learning_rate": 9.836065573770493e-07, |
|
"loss": 2.0287, |
|
"step": 30 |
|
}, |
|
{ |
|
"epoch": 0.66, |
|
"learning_rate": 1.3114754098360657e-06, |
|
"loss": 2.0054, |
|
"step": 40 |
|
}, |
|
{ |
|
"epoch": 0.82, |
|
"learning_rate": 1.6393442622950819e-06, |
|
"loss": 1.9418, |
|
"step": 50 |
|
}, |
|
{ |
|
"epoch": 0.98, |
|
"learning_rate": 1.9672131147540985e-06, |
|
"loss": 1.9035, |
|
"step": 60 |
|
}, |
|
{ |
|
"epoch": 1.0, |
|
"eval_accuracy": 0.27132579650565264, |
|
"eval_loss": 1.8945504426956177, |
|
"eval_runtime": 12.371, |
|
"eval_samples_per_second": 78.652, |
|
"eval_steps_per_second": 4.931, |
|
"step": 61 |
|
}, |
|
{ |
|
"epoch": 1.15, |
|
"learning_rate": 2.295081967213115e-06, |
|
"loss": 1.8717, |
|
"step": 70 |
|
}, |
|
{ |
|
"epoch": 1.31, |
|
"learning_rate": 2.6229508196721314e-06, |
|
"loss": 1.8202, |
|
"step": 80 |
|
}, |
|
{ |
|
"epoch": 1.48, |
|
"learning_rate": 2.9508196721311478e-06, |
|
"loss": 1.7631, |
|
"step": 90 |
|
}, |
|
{ |
|
"epoch": 1.64, |
|
"learning_rate": 3.2786885245901638e-06, |
|
"loss": 1.7051, |
|
"step": 100 |
|
}, |
|
{ |
|
"epoch": 1.8, |
|
"learning_rate": 3.6065573770491806e-06, |
|
"loss": 1.6242, |
|
"step": 110 |
|
}, |
|
{ |
|
"epoch": 1.97, |
|
"learning_rate": 3.934426229508197e-06, |
|
"loss": 1.4731, |
|
"step": 120 |
|
}, |
|
{ |
|
"epoch": 2.0, |
|
"eval_accuracy": 0.5560123329907503, |
|
"eval_loss": 1.2931360006332397, |
|
"eval_runtime": 12.3906, |
|
"eval_samples_per_second": 78.527, |
|
"eval_steps_per_second": 4.923, |
|
"step": 122 |
|
}, |
|
{ |
|
"epoch": 2.13, |
|
"learning_rate": 4.2622950819672135e-06, |
|
"loss": 1.3791, |
|
"step": 130 |
|
}, |
|
{ |
|
"epoch": 2.3, |
|
"learning_rate": 4.59016393442623e-06, |
|
"loss": 1.2578, |
|
"step": 140 |
|
}, |
|
{ |
|
"epoch": 2.46, |
|
"learning_rate": 4.918032786885246e-06, |
|
"loss": 1.1893, |
|
"step": 150 |
|
}, |
|
{ |
|
"epoch": 2.62, |
|
"learning_rate": 5.245901639344263e-06, |
|
"loss": 1.037, |
|
"step": 160 |
|
}, |
|
{ |
|
"epoch": 2.79, |
|
"learning_rate": 5.573770491803278e-06, |
|
"loss": 1.0718, |
|
"step": 170 |
|
}, |
|
{ |
|
"epoch": 2.95, |
|
"learning_rate": 5.9016393442622956e-06, |
|
"loss": 0.9549, |
|
"step": 180 |
|
}, |
|
{ |
|
"epoch": 3.0, |
|
"eval_accuracy": 0.6998972250770812, |
|
"eval_loss": 0.7530122399330139, |
|
"eval_runtime": 12.2864, |
|
"eval_samples_per_second": 79.193, |
|
"eval_steps_per_second": 4.965, |
|
"step": 183 |
|
}, |
|
{ |
|
"epoch": 3.11, |
|
"learning_rate": 6.229508196721312e-06, |
|
"loss": 1.0102, |
|
"step": 190 |
|
}, |
|
{ |
|
"epoch": 3.28, |
|
"learning_rate": 6.5573770491803276e-06, |
|
"loss": 0.8888, |
|
"step": 200 |
|
}, |
|
{ |
|
"epoch": 3.44, |
|
"learning_rate": 6.885245901639345e-06, |
|
"loss": 0.8389, |
|
"step": 210 |
|
}, |
|
{ |
|
"epoch": 3.61, |
|
"learning_rate": 7.213114754098361e-06, |
|
"loss": 0.8072, |
|
"step": 220 |
|
}, |
|
{ |
|
"epoch": 3.77, |
|
"learning_rate": 7.540983606557377e-06, |
|
"loss": 0.748, |
|
"step": 230 |
|
}, |
|
{ |
|
"epoch": 3.93, |
|
"learning_rate": 7.868852459016394e-06, |
|
"loss": 0.7375, |
|
"step": 240 |
|
}, |
|
{ |
|
"epoch": 4.0, |
|
"eval_accuracy": 0.8129496402877698, |
|
"eval_loss": 0.4988514482975006, |
|
"eval_runtime": 12.1672, |
|
"eval_samples_per_second": 79.969, |
|
"eval_steps_per_second": 5.013, |
|
"step": 244 |
|
}, |
|
{ |
|
"epoch": 4.1, |
|
"learning_rate": 8.19672131147541e-06, |
|
"loss": 0.7001, |
|
"step": 250 |
|
}, |
|
{ |
|
"epoch": 4.26, |
|
"learning_rate": 8.524590163934427e-06, |
|
"loss": 0.6759, |
|
"step": 260 |
|
}, |
|
{ |
|
"epoch": 4.43, |
|
"learning_rate": 8.852459016393443e-06, |
|
"loss": 0.6793, |
|
"step": 270 |
|
}, |
|
{ |
|
"epoch": 4.59, |
|
"learning_rate": 9.18032786885246e-06, |
|
"loss": 0.6572, |
|
"step": 280 |
|
}, |
|
{ |
|
"epoch": 4.75, |
|
"learning_rate": 9.508196721311476e-06, |
|
"loss": 0.6086, |
|
"step": 290 |
|
}, |
|
{ |
|
"epoch": 4.92, |
|
"learning_rate": 9.836065573770493e-06, |
|
"loss": 0.615, |
|
"step": 300 |
|
}, |
|
{ |
|
"epoch": 5.0, |
|
"eval_accuracy": 0.8746145940390545, |
|
"eval_loss": 0.3544515371322632, |
|
"eval_runtime": 12.2633, |
|
"eval_samples_per_second": 79.343, |
|
"eval_steps_per_second": 4.974, |
|
"step": 305 |
|
}, |
|
{ |
|
"epoch": 5.08, |
|
"learning_rate": 9.836065573770493e-06, |
|
"loss": 0.5495, |
|
"step": 310 |
|
}, |
|
{ |
|
"epoch": 5.25, |
|
"learning_rate": 9.508196721311476e-06, |
|
"loss": 0.5625, |
|
"step": 320 |
|
}, |
|
{ |
|
"epoch": 5.41, |
|
"learning_rate": 9.18032786885246e-06, |
|
"loss": 0.5274, |
|
"step": 330 |
|
}, |
|
{ |
|
"epoch": 5.57, |
|
"learning_rate": 8.852459016393443e-06, |
|
"loss": 0.4914, |
|
"step": 340 |
|
}, |
|
{ |
|
"epoch": 5.74, |
|
"learning_rate": 8.524590163934427e-06, |
|
"loss": 0.3984, |
|
"step": 350 |
|
}, |
|
{ |
|
"epoch": 5.9, |
|
"learning_rate": 8.19672131147541e-06, |
|
"loss": 0.4751, |
|
"step": 360 |
|
}, |
|
{ |
|
"epoch": 6.0, |
|
"eval_accuracy": 0.9167523124357657, |
|
"eval_loss": 0.2399262934923172, |
|
"eval_runtime": 12.1531, |
|
"eval_samples_per_second": 80.062, |
|
"eval_steps_per_second": 5.019, |
|
"step": 366 |
|
}, |
|
{ |
|
"epoch": 6.07, |
|
"learning_rate": 7.868852459016394e-06, |
|
"loss": 0.4427, |
|
"step": 370 |
|
}, |
|
{ |
|
"epoch": 6.23, |
|
"learning_rate": 7.540983606557377e-06, |
|
"loss": 0.3884, |
|
"step": 380 |
|
}, |
|
{ |
|
"epoch": 6.39, |
|
"learning_rate": 7.213114754098361e-06, |
|
"loss": 0.3789, |
|
"step": 390 |
|
}, |
|
{ |
|
"epoch": 6.56, |
|
"learning_rate": 6.885245901639345e-06, |
|
"loss": 0.4063, |
|
"step": 400 |
|
}, |
|
{ |
|
"epoch": 6.72, |
|
"learning_rate": 6.5573770491803276e-06, |
|
"loss": 0.3784, |
|
"step": 410 |
|
}, |
|
{ |
|
"epoch": 6.89, |
|
"learning_rate": 6.229508196721312e-06, |
|
"loss": 0.3778, |
|
"step": 420 |
|
}, |
|
{ |
|
"epoch": 7.0, |
|
"eval_accuracy": 0.9558067831449126, |
|
"eval_loss": 0.16275165975093842, |
|
"eval_runtime": 12.3058, |
|
"eval_samples_per_second": 79.068, |
|
"eval_steps_per_second": 4.957, |
|
"step": 427 |
|
}, |
|
{ |
|
"epoch": 7.05, |
|
"learning_rate": 5.9016393442622956e-06, |
|
"loss": 0.3681, |
|
"step": 430 |
|
}, |
|
{ |
|
"epoch": 7.21, |
|
"learning_rate": 5.573770491803278e-06, |
|
"loss": 0.3397, |
|
"step": 440 |
|
}, |
|
{ |
|
"epoch": 7.38, |
|
"learning_rate": 5.245901639344263e-06, |
|
"loss": 0.3429, |
|
"step": 450 |
|
}, |
|
{ |
|
"epoch": 7.54, |
|
"learning_rate": 4.918032786885246e-06, |
|
"loss": 0.3384, |
|
"step": 460 |
|
}, |
|
{ |
|
"epoch": 7.7, |
|
"learning_rate": 4.59016393442623e-06, |
|
"loss": 0.3018, |
|
"step": 470 |
|
}, |
|
{ |
|
"epoch": 7.87, |
|
"learning_rate": 4.2622950819672135e-06, |
|
"loss": 0.3054, |
|
"step": 480 |
|
}, |
|
{ |
|
"epoch": 8.0, |
|
"eval_accuracy": 0.9619732785200411, |
|
"eval_loss": 0.12017930299043655, |
|
"eval_runtime": 12.2634, |
|
"eval_samples_per_second": 79.342, |
|
"eval_steps_per_second": 4.974, |
|
"step": 488 |
|
}, |
|
{ |
|
"epoch": 8.03, |
|
"learning_rate": 3.934426229508197e-06, |
|
"loss": 0.3419, |
|
"step": 490 |
|
}, |
|
{ |
|
"epoch": 8.2, |
|
"learning_rate": 3.6065573770491806e-06, |
|
"loss": 0.3126, |
|
"step": 500 |
|
}, |
|
{ |
|
"epoch": 8.36, |
|
"learning_rate": 3.2786885245901638e-06, |
|
"loss": 0.2932, |
|
"step": 510 |
|
}, |
|
{ |
|
"epoch": 8.52, |
|
"learning_rate": 2.9508196721311478e-06, |
|
"loss": 0.2949, |
|
"step": 520 |
|
}, |
|
{ |
|
"epoch": 8.69, |
|
"learning_rate": 2.6229508196721314e-06, |
|
"loss": 0.327, |
|
"step": 530 |
|
}, |
|
{ |
|
"epoch": 8.85, |
|
"learning_rate": 2.295081967213115e-06, |
|
"loss": 0.2787, |
|
"step": 540 |
|
}, |
|
{ |
|
"epoch": 9.0, |
|
"eval_accuracy": 0.9732785200411099, |
|
"eval_loss": 0.0988311693072319, |
|
"eval_runtime": 12.0955, |
|
"eval_samples_per_second": 80.443, |
|
"eval_steps_per_second": 5.043, |
|
"step": 549 |
|
}, |
|
{ |
|
"epoch": 9.02, |
|
"learning_rate": 1.9672131147540985e-06, |
|
"loss": 0.2774, |
|
"step": 550 |
|
}, |
|
{ |
|
"epoch": 9.18, |
|
"learning_rate": 1.6393442622950819e-06, |
|
"loss": 0.2333, |
|
"step": 560 |
|
}, |
|
{ |
|
"epoch": 9.34, |
|
"learning_rate": 1.3114754098360657e-06, |
|
"loss": 0.2573, |
|
"step": 570 |
|
}, |
|
{ |
|
"epoch": 9.51, |
|
"learning_rate": 9.836065573770493e-07, |
|
"loss": 0.2715, |
|
"step": 580 |
|
}, |
|
{ |
|
"epoch": 9.67, |
|
"learning_rate": 6.557377049180328e-07, |
|
"loss": 0.2168, |
|
"step": 590 |
|
}, |
|
{ |
|
"epoch": 9.84, |
|
"learning_rate": 3.278688524590164e-07, |
|
"loss": 0.3328, |
|
"step": 600 |
|
}, |
|
{ |
|
"epoch": 10.0, |
|
"learning_rate": 0.0, |
|
"loss": 0.253, |
|
"step": 610 |
|
}, |
|
{ |
|
"epoch": 10.0, |
|
"eval_accuracy": 0.9722507708119219, |
|
"eval_loss": 0.09661855548620224, |
|
"eval_runtime": 12.2414, |
|
"eval_samples_per_second": 79.484, |
|
"eval_steps_per_second": 4.983, |
|
"step": 610 |
|
}, |
|
{ |
|
"epoch": 10.0, |
|
"step": 610, |
|
"total_flos": 8.960328656655483e+18, |
|
"train_loss": 0.807003553578111, |
|
"train_runtime": 6787.6894, |
|
"train_samples_per_second": 5.732, |
|
"train_steps_per_second": 0.09 |
|
} |
|
], |
|
"max_steps": 610, |
|
"num_train_epochs": 10, |
|
"total_flos": 8.960328656655483e+18, |
|
"trial_name": null, |
|
"trial_params": null |
|
} |
|
|