navradio's picture
Training in progress, epoch 0
33c9ae5
raw
history blame
14.2 kB
{
"best_metric": 0.8001030396702731,
"best_model_checkpoint": "swin-tiny-patch4-window7-224-finetuned-eurosat/checkpoint-310",
"epoch": 29.565217391304348,
"global_step": 510,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.58,
"learning_rate": 9.803921568627451e-06,
"loss": 0.6794,
"step": 10
},
{
"epoch": 0.99,
"eval_accuracy": 0.7047913446676971,
"eval_loss": 0.5848240256309509,
"eval_runtime": 9.3525,
"eval_samples_per_second": 207.538,
"eval_steps_per_second": 0.855,
"step": 17
},
{
"epoch": 1.16,
"learning_rate": 1.9607843137254903e-05,
"loss": 0.6308,
"step": 20
},
{
"epoch": 1.74,
"learning_rate": 2.9411764705882354e-05,
"loss": 0.5835,
"step": 30
},
{
"epoch": 1.97,
"eval_accuracy": 0.7429160226687275,
"eval_loss": 0.5206928253173828,
"eval_runtime": 9.455,
"eval_samples_per_second": 205.288,
"eval_steps_per_second": 0.846,
"step": 34
},
{
"epoch": 2.32,
"learning_rate": 3.9215686274509805e-05,
"loss": 0.5523,
"step": 40
},
{
"epoch": 2.9,
"learning_rate": 4.901960784313725e-05,
"loss": 0.5257,
"step": 50
},
{
"epoch": 2.96,
"eval_accuracy": 0.7640391550747038,
"eval_loss": 0.4912124574184418,
"eval_runtime": 9.367,
"eval_samples_per_second": 207.218,
"eval_steps_per_second": 0.854,
"step": 51
},
{
"epoch": 3.48,
"learning_rate": 4.901960784313725e-05,
"loss": 0.5128,
"step": 60
},
{
"epoch": 4.0,
"eval_accuracy": 0.7743431221020093,
"eval_loss": 0.46713554859161377,
"eval_runtime": 9.3663,
"eval_samples_per_second": 207.232,
"eval_steps_per_second": 0.854,
"step": 69
},
{
"epoch": 4.06,
"learning_rate": 4.793028322440087e-05,
"loss": 0.5028,
"step": 70
},
{
"epoch": 4.64,
"learning_rate": 4.684095860566449e-05,
"loss": 0.4963,
"step": 80
},
{
"epoch": 4.99,
"eval_accuracy": 0.7599175682637815,
"eval_loss": 0.479315847158432,
"eval_runtime": 9.3717,
"eval_samples_per_second": 207.114,
"eval_steps_per_second": 0.854,
"step": 86
},
{
"epoch": 5.22,
"learning_rate": 4.5751633986928104e-05,
"loss": 0.4965,
"step": 90
},
{
"epoch": 5.8,
"learning_rate": 4.466230936819172e-05,
"loss": 0.5005,
"step": 100
},
{
"epoch": 5.97,
"eval_accuracy": 0.7650695517774343,
"eval_loss": 0.485215961933136,
"eval_runtime": 9.3698,
"eval_samples_per_second": 207.154,
"eval_steps_per_second": 0.854,
"step": 103
},
{
"epoch": 6.38,
"learning_rate": 4.357298474945534e-05,
"loss": 0.4963,
"step": 110
},
{
"epoch": 6.96,
"learning_rate": 4.2483660130718954e-05,
"loss": 0.4748,
"step": 120
},
{
"epoch": 6.96,
"eval_accuracy": 0.768160741885626,
"eval_loss": 0.4841272234916687,
"eval_runtime": 9.371,
"eval_samples_per_second": 207.128,
"eval_steps_per_second": 0.854,
"step": 120
},
{
"epoch": 7.54,
"learning_rate": 4.1394335511982573e-05,
"loss": 0.4732,
"step": 130
},
{
"epoch": 8.0,
"eval_accuracy": 0.7748583204533745,
"eval_loss": 0.4430771768093109,
"eval_runtime": 9.3845,
"eval_samples_per_second": 206.83,
"eval_steps_per_second": 0.852,
"step": 138
},
{
"epoch": 8.12,
"learning_rate": 4.0305010893246186e-05,
"loss": 0.4842,
"step": 140
},
{
"epoch": 8.7,
"learning_rate": 3.9215686274509805e-05,
"loss": 0.4614,
"step": 150
},
{
"epoch": 8.99,
"eval_accuracy": 0.7856774858320453,
"eval_loss": 0.4429071843624115,
"eval_runtime": 9.3915,
"eval_samples_per_second": 206.677,
"eval_steps_per_second": 0.852,
"step": 155
},
{
"epoch": 9.28,
"learning_rate": 3.8126361655773424e-05,
"loss": 0.4645,
"step": 160
},
{
"epoch": 9.86,
"learning_rate": 3.7037037037037037e-05,
"loss": 0.4647,
"step": 170
},
{
"epoch": 9.97,
"eval_accuracy": 0.7717671303451829,
"eval_loss": 0.4715192914009094,
"eval_runtime": 9.3758,
"eval_samples_per_second": 207.023,
"eval_steps_per_second": 0.853,
"step": 172
},
{
"epoch": 10.43,
"learning_rate": 3.5947712418300656e-05,
"loss": 0.4601,
"step": 180
},
{
"epoch": 10.96,
"eval_accuracy": 0.7753735188047398,
"eval_loss": 0.4722177982330322,
"eval_runtime": 9.3838,
"eval_samples_per_second": 206.847,
"eval_steps_per_second": 0.853,
"step": 189
},
{
"epoch": 11.01,
"learning_rate": 3.4858387799564275e-05,
"loss": 0.4496,
"step": 190
},
{
"epoch": 11.59,
"learning_rate": 3.376906318082789e-05,
"loss": 0.4626,
"step": 200
},
{
"epoch": 12.0,
"eval_accuracy": 0.7872230808861411,
"eval_loss": 0.44434165954589844,
"eval_runtime": 9.3667,
"eval_samples_per_second": 207.223,
"eval_steps_per_second": 0.854,
"step": 207
},
{
"epoch": 12.17,
"learning_rate": 3.2679738562091506e-05,
"loss": 0.4447,
"step": 210
},
{
"epoch": 12.75,
"learning_rate": 3.1590413943355126e-05,
"loss": 0.4526,
"step": 220
},
{
"epoch": 12.99,
"eval_accuracy": 0.7908294693456981,
"eval_loss": 0.4368865489959717,
"eval_runtime": 9.3607,
"eval_samples_per_second": 207.356,
"eval_steps_per_second": 0.855,
"step": 224
},
{
"epoch": 13.33,
"learning_rate": 3.0501089324618738e-05,
"loss": 0.447,
"step": 230
},
{
"epoch": 13.91,
"learning_rate": 2.9411764705882354e-05,
"loss": 0.4494,
"step": 240
},
{
"epoch": 13.97,
"eval_accuracy": 0.7970118495620814,
"eval_loss": 0.42481139302253723,
"eval_runtime": 9.3639,
"eval_samples_per_second": 207.286,
"eval_steps_per_second": 0.854,
"step": 241
},
{
"epoch": 14.49,
"learning_rate": 2.832244008714597e-05,
"loss": 0.458,
"step": 250
},
{
"epoch": 14.96,
"eval_accuracy": 0.7867078825347759,
"eval_loss": 0.4402971863746643,
"eval_runtime": 9.4177,
"eval_samples_per_second": 206.101,
"eval_steps_per_second": 0.849,
"step": 258
},
{
"epoch": 15.07,
"learning_rate": 2.7233115468409585e-05,
"loss": 0.4501,
"step": 260
},
{
"epoch": 15.65,
"learning_rate": 2.6143790849673204e-05,
"loss": 0.4387,
"step": 270
},
{
"epoch": 16.0,
"eval_accuracy": 0.7964966512107161,
"eval_loss": 0.4251194894313812,
"eval_runtime": 9.3671,
"eval_samples_per_second": 207.214,
"eval_steps_per_second": 0.854,
"step": 276
},
{
"epoch": 16.23,
"learning_rate": 2.5054466230936817e-05,
"loss": 0.4448,
"step": 280
},
{
"epoch": 16.81,
"learning_rate": 2.3965141612200436e-05,
"loss": 0.4314,
"step": 290
},
{
"epoch": 16.99,
"eval_accuracy": 0.7980422462648119,
"eval_loss": 0.4256560206413269,
"eval_runtime": 9.3785,
"eval_samples_per_second": 206.963,
"eval_steps_per_second": 0.853,
"step": 293
},
{
"epoch": 17.39,
"learning_rate": 2.2875816993464052e-05,
"loss": 0.4354,
"step": 300
},
{
"epoch": 17.97,
"learning_rate": 2.178649237472767e-05,
"loss": 0.432,
"step": 310
},
{
"epoch": 17.97,
"eval_accuracy": 0.8001030396702731,
"eval_loss": 0.42466819286346436,
"eval_runtime": 9.3749,
"eval_samples_per_second": 207.041,
"eval_steps_per_second": 0.853,
"step": 310
},
{
"epoch": 18.55,
"learning_rate": 2.0697167755991287e-05,
"loss": 0.4372,
"step": 320
},
{
"epoch": 18.96,
"eval_accuracy": 0.7856774858320453,
"eval_loss": 0.44541600346565247,
"eval_runtime": 9.3667,
"eval_samples_per_second": 207.222,
"eval_steps_per_second": 0.854,
"step": 327
},
{
"epoch": 19.13,
"learning_rate": 1.9607843137254903e-05,
"loss": 0.4298,
"step": 330
},
{
"epoch": 19.71,
"learning_rate": 1.8518518518518518e-05,
"loss": 0.434,
"step": 340
},
{
"epoch": 20.0,
"eval_accuracy": 0.7903142709943328,
"eval_loss": 0.43263283371925354,
"eval_runtime": 9.367,
"eval_samples_per_second": 207.217,
"eval_steps_per_second": 0.854,
"step": 345
},
{
"epoch": 20.29,
"learning_rate": 1.7429193899782137e-05,
"loss": 0.4257,
"step": 350
},
{
"epoch": 20.87,
"learning_rate": 1.6339869281045753e-05,
"loss": 0.4313,
"step": 360
},
{
"epoch": 20.99,
"eval_accuracy": 0.7959814528593508,
"eval_loss": 0.4253856837749481,
"eval_runtime": 9.3553,
"eval_samples_per_second": 207.475,
"eval_steps_per_second": 0.855,
"step": 362
},
{
"epoch": 21.45,
"learning_rate": 1.5250544662309369e-05,
"loss": 0.421,
"step": 370
},
{
"epoch": 21.97,
"eval_accuracy": 0.7970118495620814,
"eval_loss": 0.42772090435028076,
"eval_runtime": 9.3638,
"eval_samples_per_second": 207.288,
"eval_steps_per_second": 0.854,
"step": 379
},
{
"epoch": 22.03,
"learning_rate": 1.4161220043572985e-05,
"loss": 0.4207,
"step": 380
},
{
"epoch": 22.61,
"learning_rate": 1.3071895424836602e-05,
"loss": 0.4272,
"step": 390
},
{
"epoch": 22.96,
"eval_accuracy": 0.7980422462648119,
"eval_loss": 0.4280939996242523,
"eval_runtime": 9.3579,
"eval_samples_per_second": 207.419,
"eval_steps_per_second": 0.855,
"step": 396
},
{
"epoch": 23.19,
"learning_rate": 1.1982570806100218e-05,
"loss": 0.4185,
"step": 400
},
{
"epoch": 23.77,
"learning_rate": 1.0893246187363835e-05,
"loss": 0.4245,
"step": 410
},
{
"epoch": 24.0,
"eval_accuracy": 0.7959814528593508,
"eval_loss": 0.4282556176185608,
"eval_runtime": 9.3525,
"eval_samples_per_second": 207.539,
"eval_steps_per_second": 0.855,
"step": 414
},
{
"epoch": 24.35,
"learning_rate": 9.803921568627451e-06,
"loss": 0.4179,
"step": 420
},
{
"epoch": 24.93,
"learning_rate": 8.714596949891069e-06,
"loss": 0.4207,
"step": 430
},
{
"epoch": 24.99,
"eval_accuracy": 0.7892838742916023,
"eval_loss": 0.4409063458442688,
"eval_runtime": 9.3603,
"eval_samples_per_second": 207.365,
"eval_steps_per_second": 0.855,
"step": 431
},
{
"epoch": 25.51,
"learning_rate": 7.6252723311546845e-06,
"loss": 0.4134,
"step": 440
},
{
"epoch": 25.97,
"eval_accuracy": 0.7964966512107161,
"eval_loss": 0.43111562728881836,
"eval_runtime": 9.3664,
"eval_samples_per_second": 207.229,
"eval_steps_per_second": 0.854,
"step": 448
},
{
"epoch": 26.09,
"learning_rate": 6.535947712418301e-06,
"loss": 0.4216,
"step": 450
},
{
"epoch": 26.67,
"learning_rate": 5.446623093681918e-06,
"loss": 0.4157,
"step": 460
},
{
"epoch": 26.96,
"eval_accuracy": 0.794435857805255,
"eval_loss": 0.43559595942497253,
"eval_runtime": 9.3485,
"eval_samples_per_second": 207.628,
"eval_steps_per_second": 0.856,
"step": 465
},
{
"epoch": 27.25,
"learning_rate": 4.357298474945534e-06,
"loss": 0.4235,
"step": 470
},
{
"epoch": 27.83,
"learning_rate": 3.2679738562091506e-06,
"loss": 0.4207,
"step": 480
},
{
"epoch": 28.0,
"eval_accuracy": 0.7980422462648119,
"eval_loss": 0.4301479756832123,
"eval_runtime": 9.3663,
"eval_samples_per_second": 207.233,
"eval_steps_per_second": 0.854,
"step": 483
},
{
"epoch": 28.41,
"learning_rate": 2.178649237472767e-06,
"loss": 0.4128,
"step": 490
},
{
"epoch": 28.99,
"learning_rate": 1.0893246187363836e-06,
"loss": 0.4104,
"step": 500
},
{
"epoch": 28.99,
"eval_accuracy": 0.7954662545079856,
"eval_loss": 0.4339882433414459,
"eval_runtime": 9.3665,
"eval_samples_per_second": 207.228,
"eval_steps_per_second": 0.854,
"step": 500
},
{
"epoch": 29.57,
"learning_rate": 0.0,
"loss": 0.4187,
"step": 510
},
{
"epoch": 29.57,
"eval_accuracy": 0.7954662545079856,
"eval_loss": 0.4336837828159332,
"eval_runtime": 9.3744,
"eval_samples_per_second": 207.053,
"eval_steps_per_second": 0.853,
"step": 510
},
{
"epoch": 29.57,
"step": 510,
"total_flos": 1.2835906128174883e+19,
"train_loss": 0.46080853518317727,
"train_runtime": 3932.5078,
"train_samples_per_second": 133.22,
"train_steps_per_second": 0.13
}
],
"max_steps": 510,
"num_train_epochs": 30,
"total_flos": 1.2835906128174883e+19,
"trial_name": null,
"trial_params": null
}