andrecastro's picture
End of training
79633c9
raw
history blame
8.8 kB
{
"best_metric": 0.9959183673469387,
"best_model_checkpoint": "swin-tiny-patch4-window7-224-finetuned-eurosat/checkpoint-413",
"epoch": 2.9927360774818403,
"eval_steps": 500,
"global_step": 618,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.05,
"learning_rate": 8.064516129032258e-06,
"loss": 0.6891,
"step": 10
},
{
"epoch": 0.1,
"learning_rate": 1.6129032258064517e-05,
"loss": 0.6032,
"step": 20
},
{
"epoch": 0.15,
"learning_rate": 2.4193548387096777e-05,
"loss": 0.452,
"step": 30
},
{
"epoch": 0.19,
"learning_rate": 3.2258064516129034e-05,
"loss": 0.3304,
"step": 40
},
{
"epoch": 0.24,
"learning_rate": 4.032258064516129e-05,
"loss": 0.2486,
"step": 50
},
{
"epoch": 0.29,
"learning_rate": 4.8387096774193554e-05,
"loss": 0.1912,
"step": 60
},
{
"epoch": 0.34,
"learning_rate": 4.9280575539568345e-05,
"loss": 0.1454,
"step": 70
},
{
"epoch": 0.39,
"learning_rate": 4.838129496402878e-05,
"loss": 0.1398,
"step": 80
},
{
"epoch": 0.44,
"learning_rate": 4.748201438848921e-05,
"loss": 0.1292,
"step": 90
},
{
"epoch": 0.48,
"learning_rate": 4.658273381294964e-05,
"loss": 0.1324,
"step": 100
},
{
"epoch": 0.53,
"learning_rate": 4.5683453237410076e-05,
"loss": 0.077,
"step": 110
},
{
"epoch": 0.58,
"learning_rate": 4.47841726618705e-05,
"loss": 0.0785,
"step": 120
},
{
"epoch": 0.63,
"learning_rate": 4.3884892086330935e-05,
"loss": 0.1228,
"step": 130
},
{
"epoch": 0.68,
"learning_rate": 4.298561151079137e-05,
"loss": 0.054,
"step": 140
},
{
"epoch": 0.73,
"learning_rate": 4.20863309352518e-05,
"loss": 0.1124,
"step": 150
},
{
"epoch": 0.77,
"learning_rate": 4.1187050359712234e-05,
"loss": 0.0875,
"step": 160
},
{
"epoch": 0.82,
"learning_rate": 4.0287769784172666e-05,
"loss": 0.1152,
"step": 170
},
{
"epoch": 0.87,
"learning_rate": 3.938848920863309e-05,
"loss": 0.107,
"step": 180
},
{
"epoch": 0.92,
"learning_rate": 3.8489208633093525e-05,
"loss": 0.0887,
"step": 190
},
{
"epoch": 0.97,
"learning_rate": 3.758992805755396e-05,
"loss": 0.0826,
"step": 200
},
{
"epoch": 1.0,
"eval_accuracy": 0.9918367346938776,
"eval_loss": 0.03725149855017662,
"eval_runtime": 12.9921,
"eval_samples_per_second": 56.573,
"eval_steps_per_second": 7.081,
"step": 206
},
{
"epoch": 1.02,
"learning_rate": 3.669064748201439e-05,
"loss": 0.0654,
"step": 210
},
{
"epoch": 1.07,
"learning_rate": 3.5791366906474824e-05,
"loss": 0.0904,
"step": 220
},
{
"epoch": 1.11,
"learning_rate": 3.489208633093525e-05,
"loss": 0.0879,
"step": 230
},
{
"epoch": 1.16,
"learning_rate": 3.399280575539568e-05,
"loss": 0.0731,
"step": 240
},
{
"epoch": 1.21,
"learning_rate": 3.3093525179856116e-05,
"loss": 0.0715,
"step": 250
},
{
"epoch": 1.26,
"learning_rate": 3.219424460431655e-05,
"loss": 0.0513,
"step": 260
},
{
"epoch": 1.31,
"learning_rate": 3.129496402877698e-05,
"loss": 0.0458,
"step": 270
},
{
"epoch": 1.36,
"learning_rate": 3.039568345323741e-05,
"loss": 0.0795,
"step": 280
},
{
"epoch": 1.4,
"learning_rate": 2.9496402877697844e-05,
"loss": 0.0743,
"step": 290
},
{
"epoch": 1.45,
"learning_rate": 2.8597122302158273e-05,
"loss": 0.0943,
"step": 300
},
{
"epoch": 1.5,
"learning_rate": 2.7697841726618706e-05,
"loss": 0.0961,
"step": 310
},
{
"epoch": 1.55,
"learning_rate": 2.679856115107914e-05,
"loss": 0.0387,
"step": 320
},
{
"epoch": 1.6,
"learning_rate": 2.589928057553957e-05,
"loss": 0.0843,
"step": 330
},
{
"epoch": 1.65,
"learning_rate": 2.5e-05,
"loss": 0.0816,
"step": 340
},
{
"epoch": 1.69,
"learning_rate": 2.4100719424460434e-05,
"loss": 0.0263,
"step": 350
},
{
"epoch": 1.74,
"learning_rate": 2.3201438848920864e-05,
"loss": 0.0794,
"step": 360
},
{
"epoch": 1.79,
"learning_rate": 2.2302158273381296e-05,
"loss": 0.0321,
"step": 370
},
{
"epoch": 1.84,
"learning_rate": 2.140287769784173e-05,
"loss": 0.0087,
"step": 380
},
{
"epoch": 1.89,
"learning_rate": 2.050359712230216e-05,
"loss": 0.0621,
"step": 390
},
{
"epoch": 1.94,
"learning_rate": 1.960431654676259e-05,
"loss": 0.0501,
"step": 400
},
{
"epoch": 1.99,
"learning_rate": 1.8705035971223024e-05,
"loss": 0.0658,
"step": 410
},
{
"epoch": 2.0,
"eval_accuracy": 0.9959183673469387,
"eval_loss": 0.02241336554288864,
"eval_runtime": 14.7711,
"eval_samples_per_second": 49.759,
"eval_steps_per_second": 6.228,
"step": 413
},
{
"epoch": 2.03,
"learning_rate": 1.7805755395683454e-05,
"loss": 0.0063,
"step": 420
},
{
"epoch": 2.08,
"learning_rate": 1.6906474820143887e-05,
"loss": 0.0207,
"step": 430
},
{
"epoch": 2.13,
"learning_rate": 1.6007194244604316e-05,
"loss": 0.0129,
"step": 440
},
{
"epoch": 2.18,
"learning_rate": 1.5107913669064749e-05,
"loss": 0.0116,
"step": 450
},
{
"epoch": 2.23,
"learning_rate": 1.420863309352518e-05,
"loss": 0.0049,
"step": 460
},
{
"epoch": 2.28,
"learning_rate": 1.3309352517985613e-05,
"loss": 0.0311,
"step": 470
},
{
"epoch": 2.32,
"learning_rate": 1.2410071942446044e-05,
"loss": 0.037,
"step": 480
},
{
"epoch": 2.37,
"learning_rate": 1.1510791366906475e-05,
"loss": 0.0572,
"step": 490
},
{
"epoch": 2.42,
"learning_rate": 1.0611510791366908e-05,
"loss": 0.0038,
"step": 500
},
{
"epoch": 2.47,
"learning_rate": 9.71223021582734e-06,
"loss": 0.0551,
"step": 510
},
{
"epoch": 2.52,
"learning_rate": 8.812949640287769e-06,
"loss": 0.1015,
"step": 520
},
{
"epoch": 2.57,
"learning_rate": 7.913669064748202e-06,
"loss": 0.0154,
"step": 530
},
{
"epoch": 2.62,
"learning_rate": 7.014388489208633e-06,
"loss": 0.0531,
"step": 540
},
{
"epoch": 2.66,
"learning_rate": 6.115107913669065e-06,
"loss": 0.0558,
"step": 550
},
{
"epoch": 2.71,
"learning_rate": 5.215827338129497e-06,
"loss": 0.035,
"step": 560
},
{
"epoch": 2.76,
"learning_rate": 4.316546762589929e-06,
"loss": 0.0335,
"step": 570
},
{
"epoch": 2.81,
"learning_rate": 3.41726618705036e-06,
"loss": 0.0457,
"step": 580
},
{
"epoch": 2.86,
"learning_rate": 2.5179856115107916e-06,
"loss": 0.016,
"step": 590
},
{
"epoch": 2.91,
"learning_rate": 1.618705035971223e-06,
"loss": 0.0124,
"step": 600
},
{
"epoch": 2.95,
"learning_rate": 7.194244604316547e-07,
"loss": 0.012,
"step": 610
},
{
"epoch": 2.99,
"eval_accuracy": 0.9931972789115646,
"eval_loss": 0.027184385806322098,
"eval_runtime": 13.1253,
"eval_samples_per_second": 55.999,
"eval_steps_per_second": 7.009,
"step": 618
},
{
"epoch": 2.99,
"step": 618,
"total_flos": 4.9150232620631654e+17,
"train_loss": 0.0970435019064104,
"train_runtime": 528.4446,
"train_samples_per_second": 37.508,
"train_steps_per_second": 1.169
}
],
"logging_steps": 10,
"max_steps": 618,
"num_train_epochs": 3,
"save_steps": 500,
"total_flos": 4.9150232620631654e+17,
"trial_name": null,
"trial_params": null
}