|
{ |
|
"best_metric": 0.9846153846153847, |
|
"best_model_checkpoint": "swin-tiny-patch4-window7-224-finetuned_ASL_Isolated_Swin_dataset2/checkpoint-800", |
|
"epoch": 10.0, |
|
"eval_steps": 100, |
|
"global_step": 920, |
|
"is_hyper_param_search": false, |
|
"is_local_process_zero": true, |
|
"is_world_process_zero": true, |
|
"log_history": [ |
|
{ |
|
"epoch": 0.11, |
|
"learning_rate": 0.00019782608695652175, |
|
"loss": 3.3609, |
|
"step": 10 |
|
}, |
|
{ |
|
"epoch": 0.22, |
|
"learning_rate": 0.0001956521739130435, |
|
"loss": 3.1386, |
|
"step": 20 |
|
}, |
|
{ |
|
"epoch": 0.33, |
|
"learning_rate": 0.00019347826086956523, |
|
"loss": 2.9847, |
|
"step": 30 |
|
}, |
|
{ |
|
"epoch": 0.43, |
|
"learning_rate": 0.00019130434782608697, |
|
"loss": 2.7624, |
|
"step": 40 |
|
}, |
|
{ |
|
"epoch": 0.54, |
|
"learning_rate": 0.0001891304347826087, |
|
"loss": 2.2708, |
|
"step": 50 |
|
}, |
|
{ |
|
"epoch": 0.65, |
|
"learning_rate": 0.00018717391304347826, |
|
"loss": 2.1984, |
|
"step": 60 |
|
}, |
|
{ |
|
"epoch": 0.76, |
|
"learning_rate": 0.00018500000000000002, |
|
"loss": 1.8882, |
|
"step": 70 |
|
}, |
|
{ |
|
"epoch": 0.87, |
|
"learning_rate": 0.00018282608695652174, |
|
"loss": 1.6971, |
|
"step": 80 |
|
}, |
|
{ |
|
"epoch": 0.98, |
|
"learning_rate": 0.00018065217391304348, |
|
"loss": 1.4803, |
|
"step": 90 |
|
}, |
|
{ |
|
"epoch": 1.09, |
|
"learning_rate": 0.00017847826086956522, |
|
"loss": 1.39, |
|
"step": 100 |
|
}, |
|
{ |
|
"epoch": 1.09, |
|
"eval_accuracy": 0.6346153846153846, |
|
"eval_loss": 1.182677984237671, |
|
"eval_runtime": 2.4526, |
|
"eval_samples_per_second": 106.009, |
|
"eval_steps_per_second": 13.455, |
|
"step": 100 |
|
}, |
|
{ |
|
"epoch": 1.2, |
|
"learning_rate": 0.00017630434782608695, |
|
"loss": 1.2313, |
|
"step": 110 |
|
}, |
|
{ |
|
"epoch": 1.3, |
|
"learning_rate": 0.00017413043478260872, |
|
"loss": 1.2511, |
|
"step": 120 |
|
}, |
|
{ |
|
"epoch": 1.41, |
|
"learning_rate": 0.00017217391304347827, |
|
"loss": 1.1545, |
|
"step": 130 |
|
}, |
|
{ |
|
"epoch": 1.52, |
|
"learning_rate": 0.00017, |
|
"loss": 1.0298, |
|
"step": 140 |
|
}, |
|
{ |
|
"epoch": 1.63, |
|
"learning_rate": 0.00016782608695652175, |
|
"loss": 1.0783, |
|
"step": 150 |
|
}, |
|
{ |
|
"epoch": 1.74, |
|
"learning_rate": 0.0001656521739130435, |
|
"loss": 1.0336, |
|
"step": 160 |
|
}, |
|
{ |
|
"epoch": 1.85, |
|
"learning_rate": 0.00016347826086956523, |
|
"loss": 1.0377, |
|
"step": 170 |
|
}, |
|
{ |
|
"epoch": 1.96, |
|
"learning_rate": 0.00016130434782608697, |
|
"loss": 1.1622, |
|
"step": 180 |
|
}, |
|
{ |
|
"epoch": 2.07, |
|
"learning_rate": 0.00015913043478260868, |
|
"loss": 0.9864, |
|
"step": 190 |
|
}, |
|
{ |
|
"epoch": 2.17, |
|
"learning_rate": 0.00015695652173913045, |
|
"loss": 0.8972, |
|
"step": 200 |
|
}, |
|
{ |
|
"epoch": 2.17, |
|
"eval_accuracy": 0.7807692307692308, |
|
"eval_loss": 0.6287302374839783, |
|
"eval_runtime": 2.6916, |
|
"eval_samples_per_second": 96.596, |
|
"eval_steps_per_second": 12.26, |
|
"step": 200 |
|
}, |
|
{ |
|
"epoch": 2.28, |
|
"learning_rate": 0.0001547826086956522, |
|
"loss": 0.7377, |
|
"step": 210 |
|
}, |
|
{ |
|
"epoch": 2.39, |
|
"learning_rate": 0.00015260869565217393, |
|
"loss": 0.8899, |
|
"step": 220 |
|
}, |
|
{ |
|
"epoch": 2.5, |
|
"learning_rate": 0.00015043478260869567, |
|
"loss": 0.6429, |
|
"step": 230 |
|
}, |
|
{ |
|
"epoch": 2.61, |
|
"learning_rate": 0.0001482608695652174, |
|
"loss": 0.8942, |
|
"step": 240 |
|
}, |
|
{ |
|
"epoch": 2.72, |
|
"learning_rate": 0.00014608695652173914, |
|
"loss": 0.8716, |
|
"step": 250 |
|
}, |
|
{ |
|
"epoch": 2.83, |
|
"learning_rate": 0.00014391304347826086, |
|
"loss": 0.8024, |
|
"step": 260 |
|
}, |
|
{ |
|
"epoch": 2.93, |
|
"learning_rate": 0.00014173913043478262, |
|
"loss": 0.5412, |
|
"step": 270 |
|
}, |
|
{ |
|
"epoch": 3.04, |
|
"learning_rate": 0.00013956521739130434, |
|
"loss": 0.7387, |
|
"step": 280 |
|
}, |
|
{ |
|
"epoch": 3.15, |
|
"learning_rate": 0.0001373913043478261, |
|
"loss": 0.5791, |
|
"step": 290 |
|
}, |
|
{ |
|
"epoch": 3.26, |
|
"learning_rate": 0.00013521739130434784, |
|
"loss": 0.4884, |
|
"step": 300 |
|
}, |
|
{ |
|
"epoch": 3.26, |
|
"eval_accuracy": 0.8961538461538462, |
|
"eval_loss": 0.29269811511039734, |
|
"eval_runtime": 1.9876, |
|
"eval_samples_per_second": 130.808, |
|
"eval_steps_per_second": 16.603, |
|
"step": 300 |
|
}, |
|
{ |
|
"epoch": 3.37, |
|
"learning_rate": 0.00013304347826086958, |
|
"loss": 0.6485, |
|
"step": 310 |
|
}, |
|
{ |
|
"epoch": 3.48, |
|
"learning_rate": 0.00013086956521739132, |
|
"loss": 0.7623, |
|
"step": 320 |
|
}, |
|
{ |
|
"epoch": 3.59, |
|
"learning_rate": 0.00012869565217391303, |
|
"loss": 0.6887, |
|
"step": 330 |
|
}, |
|
{ |
|
"epoch": 3.7, |
|
"learning_rate": 0.0001265217391304348, |
|
"loss": 0.5794, |
|
"step": 340 |
|
}, |
|
{ |
|
"epoch": 3.8, |
|
"learning_rate": 0.0001243478260869565, |
|
"loss": 0.6654, |
|
"step": 350 |
|
}, |
|
{ |
|
"epoch": 3.91, |
|
"learning_rate": 0.00012217391304347828, |
|
"loss": 0.5027, |
|
"step": 360 |
|
}, |
|
{ |
|
"epoch": 4.02, |
|
"learning_rate": 0.00012, |
|
"loss": 0.4336, |
|
"step": 370 |
|
}, |
|
{ |
|
"epoch": 4.13, |
|
"learning_rate": 0.00011782608695652176, |
|
"loss": 0.5295, |
|
"step": 380 |
|
}, |
|
{ |
|
"epoch": 4.24, |
|
"learning_rate": 0.00011565217391304348, |
|
"loss": 0.6389, |
|
"step": 390 |
|
}, |
|
{ |
|
"epoch": 4.35, |
|
"learning_rate": 0.00011347826086956523, |
|
"loss": 0.4179, |
|
"step": 400 |
|
}, |
|
{ |
|
"epoch": 4.35, |
|
"eval_accuracy": 0.9423076923076923, |
|
"eval_loss": 0.1795085072517395, |
|
"eval_runtime": 2.6974, |
|
"eval_samples_per_second": 96.391, |
|
"eval_steps_per_second": 12.234, |
|
"step": 400 |
|
}, |
|
{ |
|
"epoch": 4.46, |
|
"learning_rate": 0.00011152173913043477, |
|
"loss": 0.5516, |
|
"step": 410 |
|
}, |
|
{ |
|
"epoch": 4.57, |
|
"learning_rate": 0.00010934782608695653, |
|
"loss": 0.4264, |
|
"step": 420 |
|
}, |
|
{ |
|
"epoch": 4.67, |
|
"learning_rate": 0.00010717391304347825, |
|
"loss": 0.4827, |
|
"step": 430 |
|
}, |
|
{ |
|
"epoch": 4.78, |
|
"learning_rate": 0.000105, |
|
"loss": 0.4434, |
|
"step": 440 |
|
}, |
|
{ |
|
"epoch": 4.89, |
|
"learning_rate": 0.00010282608695652174, |
|
"loss": 0.4812, |
|
"step": 450 |
|
}, |
|
{ |
|
"epoch": 5.0, |
|
"learning_rate": 0.0001006521739130435, |
|
"loss": 0.5592, |
|
"step": 460 |
|
}, |
|
{ |
|
"epoch": 5.11, |
|
"learning_rate": 9.847826086956522e-05, |
|
"loss": 0.4754, |
|
"step": 470 |
|
}, |
|
{ |
|
"epoch": 5.22, |
|
"learning_rate": 9.630434782608696e-05, |
|
"loss": 0.3776, |
|
"step": 480 |
|
}, |
|
{ |
|
"epoch": 5.33, |
|
"learning_rate": 9.41304347826087e-05, |
|
"loss": 0.4084, |
|
"step": 490 |
|
}, |
|
{ |
|
"epoch": 5.43, |
|
"learning_rate": 9.195652173913044e-05, |
|
"loss": 0.4169, |
|
"step": 500 |
|
}, |
|
{ |
|
"epoch": 5.43, |
|
"eval_accuracy": 0.95, |
|
"eval_loss": 0.1563507616519928, |
|
"eval_runtime": 1.8046, |
|
"eval_samples_per_second": 144.078, |
|
"eval_steps_per_second": 18.287, |
|
"step": 500 |
|
}, |
|
{ |
|
"epoch": 5.54, |
|
"learning_rate": 8.978260869565218e-05, |
|
"loss": 0.371, |
|
"step": 510 |
|
}, |
|
{ |
|
"epoch": 5.65, |
|
"learning_rate": 8.760869565217392e-05, |
|
"loss": 0.4745, |
|
"step": 520 |
|
}, |
|
{ |
|
"epoch": 5.76, |
|
"learning_rate": 8.543478260869566e-05, |
|
"loss": 0.4209, |
|
"step": 530 |
|
}, |
|
{ |
|
"epoch": 5.87, |
|
"learning_rate": 8.32608695652174e-05, |
|
"loss": 0.4645, |
|
"step": 540 |
|
}, |
|
{ |
|
"epoch": 5.98, |
|
"learning_rate": 8.108695652173914e-05, |
|
"loss": 0.4322, |
|
"step": 550 |
|
}, |
|
{ |
|
"epoch": 6.09, |
|
"learning_rate": 7.891304347826088e-05, |
|
"loss": 0.489, |
|
"step": 560 |
|
}, |
|
{ |
|
"epoch": 6.2, |
|
"learning_rate": 7.673913043478262e-05, |
|
"loss": 0.4531, |
|
"step": 570 |
|
}, |
|
{ |
|
"epoch": 6.3, |
|
"learning_rate": 7.456521739130435e-05, |
|
"loss": 0.2837, |
|
"step": 580 |
|
}, |
|
{ |
|
"epoch": 6.41, |
|
"learning_rate": 7.23913043478261e-05, |
|
"loss": 0.37, |
|
"step": 590 |
|
}, |
|
{ |
|
"epoch": 6.52, |
|
"learning_rate": 7.021739130434783e-05, |
|
"loss": 0.3427, |
|
"step": 600 |
|
}, |
|
{ |
|
"epoch": 6.52, |
|
"eval_accuracy": 0.95, |
|
"eval_loss": 0.14258505403995514, |
|
"eval_runtime": 1.9248, |
|
"eval_samples_per_second": 135.078, |
|
"eval_steps_per_second": 17.145, |
|
"step": 600 |
|
}, |
|
{ |
|
"epoch": 6.63, |
|
"learning_rate": 6.804347826086956e-05, |
|
"loss": 0.3713, |
|
"step": 610 |
|
}, |
|
{ |
|
"epoch": 6.74, |
|
"learning_rate": 6.58695652173913e-05, |
|
"loss": 0.2211, |
|
"step": 620 |
|
}, |
|
{ |
|
"epoch": 6.85, |
|
"learning_rate": 6.369565217391305e-05, |
|
"loss": 0.2769, |
|
"step": 630 |
|
}, |
|
{ |
|
"epoch": 6.96, |
|
"learning_rate": 6.152173913043479e-05, |
|
"loss": 0.2778, |
|
"step": 640 |
|
}, |
|
{ |
|
"epoch": 7.07, |
|
"learning_rate": 5.934782608695652e-05, |
|
"loss": 0.3071, |
|
"step": 650 |
|
}, |
|
{ |
|
"epoch": 7.17, |
|
"learning_rate": 5.717391304347827e-05, |
|
"loss": 0.2988, |
|
"step": 660 |
|
}, |
|
{ |
|
"epoch": 7.28, |
|
"learning_rate": 5.500000000000001e-05, |
|
"loss": 0.2773, |
|
"step": 670 |
|
}, |
|
{ |
|
"epoch": 7.39, |
|
"learning_rate": 5.2826086956521735e-05, |
|
"loss": 0.2628, |
|
"step": 680 |
|
}, |
|
{ |
|
"epoch": 7.5, |
|
"learning_rate": 5.0652173913043474e-05, |
|
"loss": 0.4372, |
|
"step": 690 |
|
}, |
|
{ |
|
"epoch": 7.61, |
|
"learning_rate": 4.847826086956522e-05, |
|
"loss": 0.2939, |
|
"step": 700 |
|
}, |
|
{ |
|
"epoch": 7.61, |
|
"eval_accuracy": 0.9730769230769231, |
|
"eval_loss": 0.11737523972988129, |
|
"eval_runtime": 1.9862, |
|
"eval_samples_per_second": 130.905, |
|
"eval_steps_per_second": 16.615, |
|
"step": 700 |
|
}, |
|
{ |
|
"epoch": 7.72, |
|
"learning_rate": 4.630434782608696e-05, |
|
"loss": 0.3101, |
|
"step": 710 |
|
}, |
|
{ |
|
"epoch": 7.83, |
|
"learning_rate": 4.41304347826087e-05, |
|
"loss": 0.1584, |
|
"step": 720 |
|
}, |
|
{ |
|
"epoch": 7.93, |
|
"learning_rate": 4.195652173913044e-05, |
|
"loss": 0.3143, |
|
"step": 730 |
|
}, |
|
{ |
|
"epoch": 8.04, |
|
"learning_rate": 3.978260869565217e-05, |
|
"loss": 0.3705, |
|
"step": 740 |
|
}, |
|
{ |
|
"epoch": 8.15, |
|
"learning_rate": 3.7608695652173917e-05, |
|
"loss": 0.3003, |
|
"step": 750 |
|
}, |
|
{ |
|
"epoch": 8.26, |
|
"learning_rate": 3.5434782608695656e-05, |
|
"loss": 0.3864, |
|
"step": 760 |
|
}, |
|
{ |
|
"epoch": 8.37, |
|
"learning_rate": 3.3260869565217395e-05, |
|
"loss": 0.1961, |
|
"step": 770 |
|
}, |
|
{ |
|
"epoch": 8.48, |
|
"learning_rate": 3.108695652173913e-05, |
|
"loss": 0.2859, |
|
"step": 780 |
|
}, |
|
{ |
|
"epoch": 8.59, |
|
"learning_rate": 2.891304347826087e-05, |
|
"loss": 0.3521, |
|
"step": 790 |
|
}, |
|
{ |
|
"epoch": 8.7, |
|
"learning_rate": 2.673913043478261e-05, |
|
"loss": 0.1605, |
|
"step": 800 |
|
}, |
|
{ |
|
"epoch": 8.7, |
|
"eval_accuracy": 0.9846153846153847, |
|
"eval_loss": 0.06404715776443481, |
|
"eval_runtime": 1.8828, |
|
"eval_samples_per_second": 138.091, |
|
"eval_steps_per_second": 17.527, |
|
"step": 800 |
|
}, |
|
{ |
|
"epoch": 8.8, |
|
"learning_rate": 2.456521739130435e-05, |
|
"loss": 0.1828, |
|
"step": 810 |
|
}, |
|
{ |
|
"epoch": 8.91, |
|
"learning_rate": 2.239130434782609e-05, |
|
"loss": 0.1448, |
|
"step": 820 |
|
}, |
|
{ |
|
"epoch": 9.02, |
|
"learning_rate": 2.0217391304347828e-05, |
|
"loss": 0.2442, |
|
"step": 830 |
|
}, |
|
{ |
|
"epoch": 9.13, |
|
"learning_rate": 1.8043478260869567e-05, |
|
"loss": 0.2182, |
|
"step": 840 |
|
}, |
|
{ |
|
"epoch": 9.24, |
|
"learning_rate": 1.5869565217391306e-05, |
|
"loss": 0.1857, |
|
"step": 850 |
|
}, |
|
{ |
|
"epoch": 9.35, |
|
"learning_rate": 1.3695652173913042e-05, |
|
"loss": 0.123, |
|
"step": 860 |
|
}, |
|
{ |
|
"epoch": 9.46, |
|
"learning_rate": 1.1521739130434783e-05, |
|
"loss": 0.1286, |
|
"step": 870 |
|
}, |
|
{ |
|
"epoch": 9.57, |
|
"learning_rate": 9.347826086956523e-06, |
|
"loss": 0.3302, |
|
"step": 880 |
|
}, |
|
{ |
|
"epoch": 9.67, |
|
"learning_rate": 7.173913043478261e-06, |
|
"loss": 0.3513, |
|
"step": 890 |
|
}, |
|
{ |
|
"epoch": 9.78, |
|
"learning_rate": 5e-06, |
|
"loss": 0.1865, |
|
"step": 900 |
|
}, |
|
{ |
|
"epoch": 9.78, |
|
"eval_accuracy": 0.9807692307692307, |
|
"eval_loss": 0.07018034905195236, |
|
"eval_runtime": 2.3303, |
|
"eval_samples_per_second": 111.572, |
|
"eval_steps_per_second": 14.161, |
|
"step": 900 |
|
}, |
|
{ |
|
"epoch": 9.89, |
|
"learning_rate": 2.8260869565217393e-06, |
|
"loss": 0.1316, |
|
"step": 910 |
|
}, |
|
{ |
|
"epoch": 10.0, |
|
"learning_rate": 6.521739130434782e-07, |
|
"loss": 0.1979, |
|
"step": 920 |
|
}, |
|
{ |
|
"epoch": 10.0, |
|
"step": 920, |
|
"total_flos": 3.651306173035315e+17, |
|
"train_loss": 0.6966753153697304, |
|
"train_runtime": 217.2288, |
|
"train_samples_per_second": 67.579, |
|
"train_steps_per_second": 4.235 |
|
} |
|
], |
|
"logging_steps": 10, |
|
"max_steps": 920, |
|
"num_train_epochs": 10, |
|
"save_steps": 100, |
|
"total_flos": 3.651306173035315e+17, |
|
"trial_name": null, |
|
"trial_params": null |
|
} |
|
|