polejowska's picture
End of training
28995c2
{
"best_metric": 0.9851851851851852,
"best_model_checkpoint": "swin-tiny-patch4-window7-224-eurosat/checkpoint-588",
"epoch": 4.99492385786802,
"global_step": 735,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.07,
"learning_rate": 6.7567567567567575e-06,
"loss": 2.2699,
"step": 10
},
{
"epoch": 0.14,
"learning_rate": 1.3513513513513515e-05,
"loss": 2.0753,
"step": 20
},
{
"epoch": 0.2,
"learning_rate": 2.0270270270270273e-05,
"loss": 1.6254,
"step": 30
},
{
"epoch": 0.27,
"learning_rate": 2.702702702702703e-05,
"loss": 0.9696,
"step": 40
},
{
"epoch": 0.34,
"learning_rate": 3.3783783783783784e-05,
"loss": 0.529,
"step": 50
},
{
"epoch": 0.41,
"learning_rate": 4.0540540540540545e-05,
"loss": 0.358,
"step": 60
},
{
"epoch": 0.47,
"learning_rate": 4.72972972972973e-05,
"loss": 0.24,
"step": 70
},
{
"epoch": 0.54,
"learning_rate": 4.9546142208774585e-05,
"loss": 0.2678,
"step": 80
},
{
"epoch": 0.61,
"learning_rate": 4.878971255673222e-05,
"loss": 0.2537,
"step": 90
},
{
"epoch": 0.68,
"learning_rate": 4.803328290468986e-05,
"loss": 0.1921,
"step": 100
},
{
"epoch": 0.74,
"learning_rate": 4.7276853252647505e-05,
"loss": 0.1691,
"step": 110
},
{
"epoch": 0.81,
"learning_rate": 4.652042360060514e-05,
"loss": 0.1716,
"step": 120
},
{
"epoch": 0.88,
"learning_rate": 4.576399394856278e-05,
"loss": 0.1516,
"step": 130
},
{
"epoch": 0.95,
"learning_rate": 4.5007564296520425e-05,
"loss": 0.1547,
"step": 140
},
{
"epoch": 0.99,
"eval_accuracy": 0.9711111111111111,
"eval_loss": 0.09559287130832672,
"eval_runtime": 38.4532,
"eval_samples_per_second": 105.323,
"eval_steps_per_second": 3.303,
"step": 147
},
{
"epoch": 1.02,
"learning_rate": 4.425113464447806e-05,
"loss": 0.1434,
"step": 150
},
{
"epoch": 1.09,
"learning_rate": 4.34947049924357e-05,
"loss": 0.0951,
"step": 160
},
{
"epoch": 1.16,
"learning_rate": 4.2738275340393345e-05,
"loss": 0.1042,
"step": 170
},
{
"epoch": 1.22,
"learning_rate": 4.198184568835098e-05,
"loss": 0.0972,
"step": 180
},
{
"epoch": 1.29,
"learning_rate": 4.122541603630862e-05,
"loss": 0.0714,
"step": 190
},
{
"epoch": 1.36,
"learning_rate": 4.0468986384266265e-05,
"loss": 0.1095,
"step": 200
},
{
"epoch": 1.43,
"learning_rate": 3.97125567322239e-05,
"loss": 0.0779,
"step": 210
},
{
"epoch": 1.49,
"learning_rate": 3.895612708018154e-05,
"loss": 0.0869,
"step": 220
},
{
"epoch": 1.56,
"learning_rate": 3.8199697428139184e-05,
"loss": 0.0936,
"step": 230
},
{
"epoch": 1.63,
"learning_rate": 3.744326777609682e-05,
"loss": 0.092,
"step": 240
},
{
"epoch": 1.7,
"learning_rate": 3.668683812405446e-05,
"loss": 0.1006,
"step": 250
},
{
"epoch": 1.76,
"learning_rate": 3.5930408472012104e-05,
"loss": 0.0909,
"step": 260
},
{
"epoch": 1.83,
"learning_rate": 3.517397881996974e-05,
"loss": 0.0705,
"step": 270
},
{
"epoch": 1.9,
"learning_rate": 3.441754916792738e-05,
"loss": 0.0935,
"step": 280
},
{
"epoch": 1.97,
"learning_rate": 3.3661119515885024e-05,
"loss": 0.0707,
"step": 290
},
{
"epoch": 1.99,
"eval_accuracy": 0.9733333333333334,
"eval_loss": 0.07587520033121109,
"eval_runtime": 25.8049,
"eval_samples_per_second": 156.947,
"eval_steps_per_second": 4.922,
"step": 294
},
{
"epoch": 2.04,
"learning_rate": 3.290468986384266e-05,
"loss": 0.0665,
"step": 300
},
{
"epoch": 2.11,
"learning_rate": 3.21482602118003e-05,
"loss": 0.0556,
"step": 310
},
{
"epoch": 2.18,
"learning_rate": 3.1391830559757944e-05,
"loss": 0.0517,
"step": 320
},
{
"epoch": 2.24,
"learning_rate": 3.063540090771558e-05,
"loss": 0.0705,
"step": 330
},
{
"epoch": 2.31,
"learning_rate": 2.9878971255673223e-05,
"loss": 0.071,
"step": 340
},
{
"epoch": 2.38,
"learning_rate": 2.9122541603630864e-05,
"loss": 0.0466,
"step": 350
},
{
"epoch": 2.45,
"learning_rate": 2.83661119515885e-05,
"loss": 0.0351,
"step": 360
},
{
"epoch": 2.51,
"learning_rate": 2.7609682299546143e-05,
"loss": 0.0698,
"step": 370
},
{
"epoch": 2.58,
"learning_rate": 2.6853252647503784e-05,
"loss": 0.0423,
"step": 380
},
{
"epoch": 2.65,
"learning_rate": 2.609682299546142e-05,
"loss": 0.0475,
"step": 390
},
{
"epoch": 2.72,
"learning_rate": 2.5340393343419063e-05,
"loss": 0.0491,
"step": 400
},
{
"epoch": 2.79,
"learning_rate": 2.45839636913767e-05,
"loss": 0.0403,
"step": 410
},
{
"epoch": 2.85,
"learning_rate": 2.382753403933434e-05,
"loss": 0.0569,
"step": 420
},
{
"epoch": 2.92,
"learning_rate": 2.3071104387291982e-05,
"loss": 0.0509,
"step": 430
},
{
"epoch": 2.99,
"learning_rate": 2.231467473524962e-05,
"loss": 0.0537,
"step": 440
},
{
"epoch": 2.99,
"eval_accuracy": 0.9767901234567902,
"eval_loss": 0.06802858412265778,
"eval_runtime": 35.3363,
"eval_samples_per_second": 114.613,
"eval_steps_per_second": 3.594,
"step": 441
},
{
"epoch": 3.06,
"learning_rate": 2.155824508320726e-05,
"loss": 0.0502,
"step": 450
},
{
"epoch": 3.13,
"learning_rate": 2.0801815431164902e-05,
"loss": 0.0452,
"step": 460
},
{
"epoch": 3.2,
"learning_rate": 2.004538577912254e-05,
"loss": 0.0352,
"step": 470
},
{
"epoch": 3.26,
"learning_rate": 1.928895612708018e-05,
"loss": 0.0385,
"step": 480
},
{
"epoch": 3.33,
"learning_rate": 1.8532526475037822e-05,
"loss": 0.0441,
"step": 490
},
{
"epoch": 3.4,
"learning_rate": 1.777609682299546e-05,
"loss": 0.0394,
"step": 500
},
{
"epoch": 3.47,
"learning_rate": 1.70196671709531e-05,
"loss": 0.0351,
"step": 510
},
{
"epoch": 3.53,
"learning_rate": 1.6263237518910742e-05,
"loss": 0.0251,
"step": 520
},
{
"epoch": 3.6,
"learning_rate": 1.550680786686838e-05,
"loss": 0.044,
"step": 530
},
{
"epoch": 3.67,
"learning_rate": 1.4750378214826023e-05,
"loss": 0.0246,
"step": 540
},
{
"epoch": 3.74,
"learning_rate": 1.3993948562783662e-05,
"loss": 0.0252,
"step": 550
},
{
"epoch": 3.81,
"learning_rate": 1.3237518910741301e-05,
"loss": 0.0305,
"step": 560
},
{
"epoch": 3.87,
"learning_rate": 1.248108925869894e-05,
"loss": 0.0219,
"step": 570
},
{
"epoch": 3.94,
"learning_rate": 1.172465960665658e-05,
"loss": 0.0302,
"step": 580
},
{
"epoch": 3.99,
"eval_accuracy": 0.9851851851851852,
"eval_loss": 0.044658709317445755,
"eval_runtime": 25.6563,
"eval_samples_per_second": 157.856,
"eval_steps_per_second": 4.95,
"step": 588
},
{
"epoch": 4.01,
"learning_rate": 1.0968229954614221e-05,
"loss": 0.0154,
"step": 590
},
{
"epoch": 4.08,
"learning_rate": 1.021180030257186e-05,
"loss": 0.0225,
"step": 600
},
{
"epoch": 4.15,
"learning_rate": 9.4553706505295e-06,
"loss": 0.0181,
"step": 610
},
{
"epoch": 4.22,
"learning_rate": 8.698940998487141e-06,
"loss": 0.0199,
"step": 620
},
{
"epoch": 4.28,
"learning_rate": 7.94251134644478e-06,
"loss": 0.0254,
"step": 630
},
{
"epoch": 4.35,
"learning_rate": 7.186081694402421e-06,
"loss": 0.0244,
"step": 640
},
{
"epoch": 4.42,
"learning_rate": 6.429652042360061e-06,
"loss": 0.0181,
"step": 650
},
{
"epoch": 4.49,
"learning_rate": 5.6732223903177e-06,
"loss": 0.0154,
"step": 660
},
{
"epoch": 4.55,
"learning_rate": 4.9167927382753406e-06,
"loss": 0.0164,
"step": 670
},
{
"epoch": 4.62,
"learning_rate": 4.16036308623298e-06,
"loss": 0.0111,
"step": 680
},
{
"epoch": 4.69,
"learning_rate": 3.40393343419062e-06,
"loss": 0.0199,
"step": 690
},
{
"epoch": 4.76,
"learning_rate": 2.64750378214826e-06,
"loss": 0.0191,
"step": 700
},
{
"epoch": 4.83,
"learning_rate": 1.8910741301059002e-06,
"loss": 0.0171,
"step": 710
},
{
"epoch": 4.89,
"learning_rate": 1.13464447806354e-06,
"loss": 0.0164,
"step": 720
},
{
"epoch": 4.96,
"learning_rate": 3.7821482602118004e-07,
"loss": 0.0225,
"step": 730
},
{
"epoch": 4.99,
"eval_accuracy": 0.9837037037037037,
"eval_loss": 0.048928551375865936,
"eval_runtime": 25.2637,
"eval_samples_per_second": 160.309,
"eval_steps_per_second": 5.027,
"step": 735
},
{
"epoch": 4.99,
"step": 735,
"total_flos": 2.347327658721411e+18,
"train_loss": 0.1689021515602968,
"train_runtime": 1731.9704,
"train_samples_per_second": 54.562,
"train_steps_per_second": 0.424
}
],
"max_steps": 735,
"num_train_epochs": 5,
"total_flos": 2.347327658721411e+18,
"trial_name": null,
"trial_params": null
}