andrecastro's picture
End of training
21de218
raw
history blame
7.21 kB
{
"best_metric": 0.7149341806627326,
"best_model_checkpoint": "swin-tiny-patch4-window7-224-finetuned-eurosat/checkpoint-480",
"epoch": 2.9860031104199067,
"eval_steps": 500,
"global_step": 480,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.06,
"learning_rate": 1.0416666666666668e-05,
"loss": 0.6441,
"step": 10
},
{
"epoch": 0.12,
"learning_rate": 2.0833333333333336e-05,
"loss": 0.5084,
"step": 20
},
{
"epoch": 0.19,
"learning_rate": 3.125e-05,
"loss": 0.3659,
"step": 30
},
{
"epoch": 0.25,
"learning_rate": 4.166666666666667e-05,
"loss": 0.2162,
"step": 40
},
{
"epoch": 0.31,
"learning_rate": 4.976851851851852e-05,
"loss": 0.1387,
"step": 50
},
{
"epoch": 0.37,
"learning_rate": 4.8611111111111115e-05,
"loss": 0.1187,
"step": 60
},
{
"epoch": 0.44,
"learning_rate": 4.745370370370371e-05,
"loss": 0.1003,
"step": 70
},
{
"epoch": 0.5,
"learning_rate": 4.62962962962963e-05,
"loss": 0.104,
"step": 80
},
{
"epoch": 0.56,
"learning_rate": 4.5138888888888894e-05,
"loss": 0.1448,
"step": 90
},
{
"epoch": 0.62,
"learning_rate": 4.3981481481481486e-05,
"loss": 0.0851,
"step": 100
},
{
"epoch": 0.68,
"learning_rate": 4.282407407407408e-05,
"loss": 0.0664,
"step": 110
},
{
"epoch": 0.75,
"learning_rate": 4.166666666666667e-05,
"loss": 0.0775,
"step": 120
},
{
"epoch": 0.81,
"learning_rate": 4.0509259259259265e-05,
"loss": 0.0996,
"step": 130
},
{
"epoch": 0.87,
"learning_rate": 3.935185185185186e-05,
"loss": 0.0929,
"step": 140
},
{
"epoch": 0.93,
"learning_rate": 3.8194444444444444e-05,
"loss": 0.0983,
"step": 150
},
{
"epoch": 1.0,
"learning_rate": 3.7037037037037037e-05,
"loss": 0.1091,
"step": 160
},
{
"epoch": 1.0,
"eval_accuracy": 0.634135270086246,
"eval_loss": 1.134523868560791,
"eval_runtime": 48.9231,
"eval_samples_per_second": 45.03,
"eval_steps_per_second": 5.642,
"step": 160
},
{
"epoch": 1.06,
"learning_rate": 3.587962962962963e-05,
"loss": 0.1446,
"step": 170
},
{
"epoch": 1.12,
"learning_rate": 3.472222222222222e-05,
"loss": 0.0648,
"step": 180
},
{
"epoch": 1.18,
"learning_rate": 3.3564814814814815e-05,
"loss": 0.0704,
"step": 190
},
{
"epoch": 1.24,
"learning_rate": 3.240740740740741e-05,
"loss": 0.1206,
"step": 200
},
{
"epoch": 1.31,
"learning_rate": 3.125e-05,
"loss": 0.0428,
"step": 210
},
{
"epoch": 1.37,
"learning_rate": 3.0092592592592593e-05,
"loss": 0.0271,
"step": 220
},
{
"epoch": 1.43,
"learning_rate": 2.8935185185185186e-05,
"loss": 0.0786,
"step": 230
},
{
"epoch": 1.49,
"learning_rate": 2.777777777777778e-05,
"loss": 0.0233,
"step": 240
},
{
"epoch": 1.56,
"learning_rate": 2.6620370370370372e-05,
"loss": 0.01,
"step": 250
},
{
"epoch": 1.62,
"learning_rate": 2.5462962962962965e-05,
"loss": 0.0297,
"step": 260
},
{
"epoch": 1.68,
"learning_rate": 2.4305555555555558e-05,
"loss": 0.0598,
"step": 270
},
{
"epoch": 1.74,
"learning_rate": 2.314814814814815e-05,
"loss": 0.0714,
"step": 280
},
{
"epoch": 1.8,
"learning_rate": 2.1990740740740743e-05,
"loss": 0.1358,
"step": 290
},
{
"epoch": 1.87,
"learning_rate": 2.0833333333333336e-05,
"loss": 0.0164,
"step": 300
},
{
"epoch": 1.93,
"learning_rate": 1.967592592592593e-05,
"loss": 0.0072,
"step": 310
},
{
"epoch": 1.99,
"learning_rate": 1.8518518518518518e-05,
"loss": 0.0426,
"step": 320
},
{
"epoch": 2.0,
"eval_accuracy": 0.6627326373127553,
"eval_loss": 1.377668857574463,
"eval_runtime": 47.5961,
"eval_samples_per_second": 46.285,
"eval_steps_per_second": 5.799,
"step": 321
},
{
"epoch": 2.05,
"learning_rate": 1.736111111111111e-05,
"loss": 0.0439,
"step": 330
},
{
"epoch": 2.12,
"learning_rate": 1.6203703703703704e-05,
"loss": 0.0253,
"step": 340
},
{
"epoch": 2.18,
"learning_rate": 1.5046296296296297e-05,
"loss": 0.0197,
"step": 350
},
{
"epoch": 2.24,
"learning_rate": 1.388888888888889e-05,
"loss": 0.0112,
"step": 360
},
{
"epoch": 2.3,
"learning_rate": 1.2731481481481482e-05,
"loss": 0.0445,
"step": 370
},
{
"epoch": 2.36,
"learning_rate": 1.1574074074074075e-05,
"loss": 0.0338,
"step": 380
},
{
"epoch": 2.43,
"learning_rate": 1.0416666666666668e-05,
"loss": 0.0344,
"step": 390
},
{
"epoch": 2.49,
"learning_rate": 9.259259259259259e-06,
"loss": 0.0226,
"step": 400
},
{
"epoch": 2.55,
"learning_rate": 8.101851851851852e-06,
"loss": 0.0748,
"step": 410
},
{
"epoch": 2.61,
"learning_rate": 6.944444444444445e-06,
"loss": 0.0099,
"step": 420
},
{
"epoch": 2.67,
"learning_rate": 5.787037037037038e-06,
"loss": 0.0558,
"step": 430
},
{
"epoch": 2.74,
"learning_rate": 4.6296296296296296e-06,
"loss": 0.0243,
"step": 440
},
{
"epoch": 2.8,
"learning_rate": 3.4722222222222224e-06,
"loss": 0.0239,
"step": 450
},
{
"epoch": 2.86,
"learning_rate": 2.3148148148148148e-06,
"loss": 0.0645,
"step": 460
},
{
"epoch": 2.92,
"learning_rate": 1.1574074074074074e-06,
"loss": 0.0547,
"step": 470
},
{
"epoch": 2.99,
"learning_rate": 0.0,
"loss": 0.0151,
"step": 480
},
{
"epoch": 2.99,
"eval_accuracy": 0.7149341806627326,
"eval_loss": 1.1311845779418945,
"eval_runtime": 51.1784,
"eval_samples_per_second": 43.045,
"eval_steps_per_second": 5.393,
"step": 480
},
{
"epoch": 2.99,
"step": 480,
"total_flos": 3.815394309328896e+17,
"train_loss": 0.09319638077480098,
"train_runtime": 491.2433,
"train_samples_per_second": 31.384,
"train_steps_per_second": 0.977
}
],
"logging_steps": 10,
"max_steps": 480,
"num_train_epochs": 3,
"save_steps": 500,
"total_flos": 3.815394309328896e+17,
"trial_name": null,
"trial_params": null
}