Henrieke's picture
End of training
5183382
raw
history blame
9.51 kB
{
"best_metric": 0.9801980198019802,
"best_model_checkpoint": "swin-tiny-patch4-window7-224-finetuned-eurosat\\checkpoint-453",
"epoch": 2.990077177508269,
"eval_steps": 500,
"global_step": 678,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.04,
"learning_rate": 7.3529411764705884e-06,
"loss": 1.8817,
"step": 10
},
{
"epoch": 0.09,
"learning_rate": 1.4705882352941177e-05,
"loss": 1.7748,
"step": 20
},
{
"epoch": 0.13,
"learning_rate": 2.2058823529411766e-05,
"loss": 1.6352,
"step": 30
},
{
"epoch": 0.18,
"learning_rate": 2.9411764705882354e-05,
"loss": 1.2838,
"step": 40
},
{
"epoch": 0.22,
"learning_rate": 3.6764705882352945e-05,
"loss": 0.8832,
"step": 50
},
{
"epoch": 0.26,
"learning_rate": 4.411764705882353e-05,
"loss": 0.5259,
"step": 60
},
{
"epoch": 0.31,
"learning_rate": 4.9836065573770496e-05,
"loss": 0.4059,
"step": 70
},
{
"epoch": 0.35,
"learning_rate": 4.9016393442622957e-05,
"loss": 0.4344,
"step": 80
},
{
"epoch": 0.4,
"learning_rate": 4.819672131147541e-05,
"loss": 0.2773,
"step": 90
},
{
"epoch": 0.44,
"learning_rate": 4.737704918032787e-05,
"loss": 0.3712,
"step": 100
},
{
"epoch": 0.49,
"learning_rate": 4.655737704918033e-05,
"loss": 0.3117,
"step": 110
},
{
"epoch": 0.53,
"learning_rate": 4.5737704918032786e-05,
"loss": 0.2357,
"step": 120
},
{
"epoch": 0.57,
"learning_rate": 4.491803278688525e-05,
"loss": 0.3511,
"step": 130
},
{
"epoch": 0.62,
"learning_rate": 4.409836065573771e-05,
"loss": 0.2619,
"step": 140
},
{
"epoch": 0.66,
"learning_rate": 4.327868852459017e-05,
"loss": 0.2678,
"step": 150
},
{
"epoch": 0.71,
"learning_rate": 4.245901639344262e-05,
"loss": 0.3344,
"step": 160
},
{
"epoch": 0.75,
"learning_rate": 4.163934426229508e-05,
"loss": 0.2304,
"step": 170
},
{
"epoch": 0.79,
"learning_rate": 4.081967213114754e-05,
"loss": 0.3894,
"step": 180
},
{
"epoch": 0.84,
"learning_rate": 4e-05,
"loss": 0.2455,
"step": 190
},
{
"epoch": 0.88,
"learning_rate": 3.9180327868852464e-05,
"loss": 0.202,
"step": 200
},
{
"epoch": 0.93,
"learning_rate": 3.836065573770492e-05,
"loss": 0.2385,
"step": 210
},
{
"epoch": 0.97,
"learning_rate": 3.754098360655738e-05,
"loss": 0.1774,
"step": 220
},
{
"epoch": 1.0,
"eval_accuracy": 0.9702970297029703,
"eval_loss": 0.11055441945791245,
"eval_runtime": 18.9759,
"eval_samples_per_second": 21.29,
"eval_steps_per_second": 5.323,
"step": 226
},
{
"epoch": 1.01,
"learning_rate": 3.672131147540984e-05,
"loss": 0.1992,
"step": 230
},
{
"epoch": 1.06,
"learning_rate": 3.590163934426229e-05,
"loss": 0.1551,
"step": 240
},
{
"epoch": 1.1,
"learning_rate": 3.508196721311476e-05,
"loss": 0.1366,
"step": 250
},
{
"epoch": 1.15,
"learning_rate": 3.4262295081967214e-05,
"loss": 0.2028,
"step": 260
},
{
"epoch": 1.19,
"learning_rate": 3.3442622950819675e-05,
"loss": 0.2502,
"step": 270
},
{
"epoch": 1.23,
"learning_rate": 3.2622950819672136e-05,
"loss": 0.1706,
"step": 280
},
{
"epoch": 1.28,
"learning_rate": 3.180327868852459e-05,
"loss": 0.2043,
"step": 290
},
{
"epoch": 1.32,
"learning_rate": 3.098360655737705e-05,
"loss": 0.1227,
"step": 300
},
{
"epoch": 1.37,
"learning_rate": 3.016393442622951e-05,
"loss": 0.183,
"step": 310
},
{
"epoch": 1.41,
"learning_rate": 2.934426229508197e-05,
"loss": 0.1389,
"step": 320
},
{
"epoch": 1.46,
"learning_rate": 2.852459016393443e-05,
"loss": 0.2355,
"step": 330
},
{
"epoch": 1.5,
"learning_rate": 2.7704918032786886e-05,
"loss": 0.1368,
"step": 340
},
{
"epoch": 1.54,
"learning_rate": 2.6885245901639343e-05,
"loss": 0.1441,
"step": 350
},
{
"epoch": 1.59,
"learning_rate": 2.6065573770491804e-05,
"loss": 0.23,
"step": 360
},
{
"epoch": 1.63,
"learning_rate": 2.5245901639344264e-05,
"loss": 0.2492,
"step": 370
},
{
"epoch": 1.68,
"learning_rate": 2.442622950819672e-05,
"loss": 0.0825,
"step": 380
},
{
"epoch": 1.72,
"learning_rate": 2.3606557377049182e-05,
"loss": 0.1994,
"step": 390
},
{
"epoch": 1.76,
"learning_rate": 2.278688524590164e-05,
"loss": 0.157,
"step": 400
},
{
"epoch": 1.81,
"learning_rate": 2.1967213114754097e-05,
"loss": 0.3219,
"step": 410
},
{
"epoch": 1.85,
"learning_rate": 2.114754098360656e-05,
"loss": 0.2,
"step": 420
},
{
"epoch": 1.9,
"learning_rate": 2.0327868852459018e-05,
"loss": 0.2441,
"step": 430
},
{
"epoch": 1.94,
"learning_rate": 1.9508196721311475e-05,
"loss": 0.2106,
"step": 440
},
{
"epoch": 1.98,
"learning_rate": 1.8688524590163936e-05,
"loss": 0.1347,
"step": 450
},
{
"epoch": 2.0,
"eval_accuracy": 0.9801980198019802,
"eval_loss": 0.07785981893539429,
"eval_runtime": 39.5393,
"eval_samples_per_second": 10.218,
"eval_steps_per_second": 2.554,
"step": 453
},
{
"epoch": 2.03,
"learning_rate": 1.7868852459016393e-05,
"loss": 0.0916,
"step": 460
},
{
"epoch": 2.07,
"learning_rate": 1.7049180327868854e-05,
"loss": 0.1534,
"step": 470
},
{
"epoch": 2.12,
"learning_rate": 1.6229508196721314e-05,
"loss": 0.07,
"step": 480
},
{
"epoch": 2.16,
"learning_rate": 1.5409836065573772e-05,
"loss": 0.0953,
"step": 490
},
{
"epoch": 2.21,
"learning_rate": 1.4590163934426229e-05,
"loss": 0.1831,
"step": 500
},
{
"epoch": 2.25,
"learning_rate": 1.377049180327869e-05,
"loss": 0.1728,
"step": 510
},
{
"epoch": 2.29,
"learning_rate": 1.2950819672131149e-05,
"loss": 0.1736,
"step": 520
},
{
"epoch": 2.34,
"learning_rate": 1.2131147540983608e-05,
"loss": 0.0832,
"step": 530
},
{
"epoch": 2.38,
"learning_rate": 1.1311475409836065e-05,
"loss": 0.2344,
"step": 540
},
{
"epoch": 2.43,
"learning_rate": 1.0491803278688525e-05,
"loss": 0.1703,
"step": 550
},
{
"epoch": 2.47,
"learning_rate": 9.672131147540984e-06,
"loss": 0.0834,
"step": 560
},
{
"epoch": 2.51,
"learning_rate": 8.852459016393443e-06,
"loss": 0.1182,
"step": 570
},
{
"epoch": 2.56,
"learning_rate": 8.032786885245902e-06,
"loss": 0.1647,
"step": 580
},
{
"epoch": 2.6,
"learning_rate": 7.213114754098361e-06,
"loss": 0.0524,
"step": 590
},
{
"epoch": 2.65,
"learning_rate": 6.393442622950819e-06,
"loss": 0.1996,
"step": 600
},
{
"epoch": 2.69,
"learning_rate": 5.573770491803279e-06,
"loss": 0.1121,
"step": 610
},
{
"epoch": 2.73,
"learning_rate": 4.754098360655738e-06,
"loss": 0.1367,
"step": 620
},
{
"epoch": 2.78,
"learning_rate": 3.934426229508197e-06,
"loss": 0.1491,
"step": 630
},
{
"epoch": 2.82,
"learning_rate": 3.114754098360656e-06,
"loss": 0.1707,
"step": 640
},
{
"epoch": 2.87,
"learning_rate": 2.295081967213115e-06,
"loss": 0.141,
"step": 650
},
{
"epoch": 2.91,
"learning_rate": 1.4754098360655739e-06,
"loss": 0.0584,
"step": 660
},
{
"epoch": 2.95,
"learning_rate": 6.557377049180328e-07,
"loss": 0.1468,
"step": 670
},
{
"epoch": 2.99,
"eval_accuracy": 0.9801980198019802,
"eval_loss": 0.07111147046089172,
"eval_runtime": 42.1047,
"eval_samples_per_second": 9.595,
"eval_steps_per_second": 2.399,
"step": 678
},
{
"epoch": 2.99,
"step": 678,
"total_flos": 2.696181882987479e+17,
"train_loss": 0.29536557975596034,
"train_runtime": 1530.5351,
"train_samples_per_second": 7.109,
"train_steps_per_second": 0.443
}
],
"logging_steps": 10,
"max_steps": 678,
"num_train_epochs": 3,
"save_steps": 500,
"total_flos": 2.696181882987479e+17,
"trial_name": null,
"trial_params": null
}