{
  "best_metric": 0.6448598130841121,
  "best_model_checkpoint": "swinv2-tiny-patch4-window8-256-RH\\checkpoint-208",
  "epoch": 40.0,
  "eval_steps": 500,
  "global_step": 320,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "eval_accuracy": 0.411214953271028,
      "eval_loss": 4.565852165222168,
      "eval_runtime": 1.9881,
      "eval_samples_per_second": 53.821,
      "eval_steps_per_second": 3.521,
      "step": 8
    },
    {
      "epoch": 1.25,
      "learning_rate": 4.6874999999999994e-05,
      "loss": 4.5175,
      "step": 10
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.411214953271028,
      "eval_loss": 3.636173963546753,
      "eval_runtime": 1.951,
      "eval_samples_per_second": 54.845,
      "eval_steps_per_second": 3.588,
      "step": 16
    },
    {
      "epoch": 2.5,
      "learning_rate": 9.374999999999999e-05,
      "loss": 3.9284,
      "step": 20
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.411214953271028,
      "eval_loss": 1.6019313335418701,
      "eval_runtime": 1.8634,
      "eval_samples_per_second": 57.421,
      "eval_steps_per_second": 3.757,
      "step": 24
    },
    {
      "epoch": 3.75,
      "learning_rate": 0.000140625,
      "loss": 1.6086,
      "step": 30
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.411214953271028,
      "eval_loss": 0.7110145092010498,
      "eval_runtime": 1.8614,
      "eval_samples_per_second": 57.482,
      "eval_steps_per_second": 3.761,
      "step": 32
    },
    {
      "epoch": 5.0,
      "learning_rate": 0.00014583333333333332,
      "loss": 0.7392,
      "step": 40
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.5887850467289719,
      "eval_loss": 0.6824622750282288,
      "eval_runtime": 1.8634,
      "eval_samples_per_second": 57.42,
      "eval_steps_per_second": 3.756,
      "step": 40
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.5887850467289719,
      "eval_loss": 0.679547905921936,
      "eval_runtime": 1.9335,
      "eval_samples_per_second": 55.341,
      "eval_steps_per_second": 3.62,
      "step": 48
    },
    {
      "epoch": 6.25,
      "learning_rate": 0.000140625,
      "loss": 0.7073,
      "step": 50
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.5887850467289719,
      "eval_loss": 0.6814281940460205,
      "eval_runtime": 1.8789,
      "eval_samples_per_second": 56.947,
      "eval_steps_per_second": 3.725,
      "step": 56
    },
    {
      "epoch": 7.5,
      "learning_rate": 0.00013541666666666666,
      "loss": 0.6956,
      "step": 60
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.5887850467289719,
      "eval_loss": 0.7060601711273193,
      "eval_runtime": 1.8394,
      "eval_samples_per_second": 58.17,
      "eval_steps_per_second": 3.805,
      "step": 64
    },
    {
      "epoch": 8.75,
      "learning_rate": 0.00013020833333333333,
      "loss": 0.6898,
      "step": 70
    },
    {
      "epoch": 9.0,
      "eval_accuracy": 0.5887850467289719,
      "eval_loss": 0.7014247179031372,
      "eval_runtime": 1.9775,
      "eval_samples_per_second": 54.11,
      "eval_steps_per_second": 3.54,
      "step": 72
    },
    {
      "epoch": 10.0,
      "learning_rate": 0.000125,
      "loss": 0.7026,
      "step": 80
    },
    {
      "epoch": 10.0,
      "eval_accuracy": 0.411214953271028,
      "eval_loss": 0.7213616967201233,
      "eval_runtime": 1.881,
      "eval_samples_per_second": 56.886,
      "eval_steps_per_second": 3.721,
      "step": 80
    },
    {
      "epoch": 11.0,
      "eval_accuracy": 0.5887850467289719,
      "eval_loss": 0.7185712456703186,
      "eval_runtime": 1.9299,
      "eval_samples_per_second": 55.442,
      "eval_steps_per_second": 3.627,
      "step": 88
    },
    {
      "epoch": 11.25,
      "learning_rate": 0.00011979166666666666,
      "loss": 0.7696,
      "step": 90
    },
    {
      "epoch": 12.0,
      "eval_accuracy": 0.5887850467289719,
      "eval_loss": 0.6837121844291687,
      "eval_runtime": 1.8429,
      "eval_samples_per_second": 58.06,
      "eval_steps_per_second": 3.798,
      "step": 96
    },
    {
      "epoch": 12.5,
      "learning_rate": 0.00011458333333333332,
      "loss": 0.6909,
      "step": 100
    },
    {
      "epoch": 13.0,
      "eval_accuracy": 0.5887850467289719,
      "eval_loss": 0.6822527647018433,
      "eval_runtime": 1.8714,
      "eval_samples_per_second": 57.175,
      "eval_steps_per_second": 3.74,
      "step": 104
    },
    {
      "epoch": 13.75,
      "learning_rate": 0.00010937499999999999,
      "loss": 0.6799,
      "step": 110
    },
    {
      "epoch": 14.0,
      "eval_accuracy": 0.5887850467289719,
      "eval_loss": 0.6780915260314941,
      "eval_runtime": 1.8835,
      "eval_samples_per_second": 56.811,
      "eval_steps_per_second": 3.717,
      "step": 112
    },
    {
      "epoch": 15.0,
      "learning_rate": 0.00010416666666666666,
      "loss": 0.6782,
      "step": 120
    },
    {
      "epoch": 15.0,
      "eval_accuracy": 0.5887850467289719,
      "eval_loss": 0.6938412189483643,
      "eval_runtime": 2.0385,
      "eval_samples_per_second": 52.489,
      "eval_steps_per_second": 3.434,
      "step": 120
    },
    {
      "epoch": 16.0,
      "eval_accuracy": 0.5887850467289719,
      "eval_loss": 0.6766122579574585,
      "eval_runtime": 2.019,
      "eval_samples_per_second": 52.997,
      "eval_steps_per_second": 3.467,
      "step": 128
    },
    {
      "epoch": 16.25,
      "learning_rate": 9.895833333333332e-05,
      "loss": 0.6952,
      "step": 130
    },
    {
      "epoch": 17.0,
      "eval_accuracy": 0.5887850467289719,
      "eval_loss": 0.7123367190361023,
      "eval_runtime": 1.8724,
      "eval_samples_per_second": 57.145,
      "eval_steps_per_second": 3.738,
      "step": 136
    },
    {
      "epoch": 17.5,
      "learning_rate": 9.374999999999999e-05,
      "loss": 0.6875,
      "step": 140
    },
    {
      "epoch": 18.0,
      "eval_accuracy": 0.5607476635514018,
      "eval_loss": 0.6890613436698914,
      "eval_runtime": 1.9095,
      "eval_samples_per_second": 56.037,
      "eval_steps_per_second": 3.666,
      "step": 144
    },
    {
      "epoch": 18.75,
      "learning_rate": 8.854166666666666e-05,
      "loss": 0.6919,
      "step": 150
    },
    {
      "epoch": 19.0,
      "eval_accuracy": 0.5887850467289719,
      "eval_loss": 0.7076395153999329,
      "eval_runtime": 1.941,
      "eval_samples_per_second": 55.127,
      "eval_steps_per_second": 3.606,
      "step": 152
    },
    {
      "epoch": 20.0,
      "learning_rate": 8.333333333333333e-05,
      "loss": 0.6751,
      "step": 160
    },
    {
      "epoch": 20.0,
      "eval_accuracy": 0.4953271028037383,
      "eval_loss": 0.7010804414749146,
      "eval_runtime": 1.8779,
      "eval_samples_per_second": 56.977,
      "eval_steps_per_second": 3.727,
      "step": 160
    },
    {
      "epoch": 21.0,
      "eval_accuracy": 0.5887850467289719,
      "eval_loss": 0.6961632966995239,
      "eval_runtime": 1.8549,
      "eval_samples_per_second": 57.684,
      "eval_steps_per_second": 3.774,
      "step": 168
    },
    {
      "epoch": 21.25,
      "learning_rate": 7.8125e-05,
      "loss": 0.689,
      "step": 170
    },
    {
      "epoch": 22.0,
      "eval_accuracy": 0.5700934579439252,
      "eval_loss": 0.6856827139854431,
      "eval_runtime": 1.8844,
      "eval_samples_per_second": 56.781,
      "eval_steps_per_second": 3.715,
      "step": 176
    },
    {
      "epoch": 22.5,
      "learning_rate": 7.291666666666666e-05,
      "loss": 0.6826,
      "step": 180
    },
    {
      "epoch": 23.0,
      "eval_accuracy": 0.5887850467289719,
      "eval_loss": 0.6935387253761292,
      "eval_runtime": 1.9324,
      "eval_samples_per_second": 55.37,
      "eval_steps_per_second": 3.622,
      "step": 184
    },
    {
      "epoch": 23.75,
      "learning_rate": 6.770833333333333e-05,
      "loss": 0.6841,
      "step": 190
    },
    {
      "epoch": 24.0,
      "eval_accuracy": 0.5887850467289719,
      "eval_loss": 0.7218518853187561,
      "eval_runtime": 1.9105,
      "eval_samples_per_second": 56.008,
      "eval_steps_per_second": 3.664,
      "step": 192
    },
    {
      "epoch": 25.0,
      "learning_rate": 6.25e-05,
      "loss": 0.6657,
      "step": 200
    },
    {
      "epoch": 25.0,
      "eval_accuracy": 0.5887850467289719,
      "eval_loss": 0.6609508395195007,
      "eval_runtime": 1.8645,
      "eval_samples_per_second": 57.39,
      "eval_steps_per_second": 3.754,
      "step": 200
    },
    {
      "epoch": 26.0,
      "eval_accuracy": 0.6448598130841121,
      "eval_loss": 0.6681405305862427,
      "eval_runtime": 1.8574,
      "eval_samples_per_second": 57.607,
      "eval_steps_per_second": 3.769,
      "step": 208
    },
    {
      "epoch": 26.25,
      "learning_rate": 5.729166666666666e-05,
      "loss": 0.6524,
      "step": 210
    },
    {
      "epoch": 27.0,
      "eval_accuracy": 0.5887850467289719,
      "eval_loss": 0.7224901914596558,
      "eval_runtime": 1.881,
      "eval_samples_per_second": 56.886,
      "eval_steps_per_second": 3.722,
      "step": 216
    },
    {
      "epoch": 27.5,
      "learning_rate": 5.208333333333333e-05,
      "loss": 0.6567,
      "step": 220
    },
    {
      "epoch": 28.0,
      "eval_accuracy": 0.5887850467289719,
      "eval_loss": 0.7117226719856262,
      "eval_runtime": 1.9114,
      "eval_samples_per_second": 55.979,
      "eval_steps_per_second": 3.662,
      "step": 224
    },
    {
      "epoch": 28.75,
      "learning_rate": 4.6874999999999994e-05,
      "loss": 0.6402,
      "step": 230
    },
    {
      "epoch": 29.0,
      "eval_accuracy": 0.6261682242990654,
      "eval_loss": 0.6999447345733643,
      "eval_runtime": 1.913,
      "eval_samples_per_second": 55.934,
      "eval_steps_per_second": 3.659,
      "step": 232
    },
    {
      "epoch": 30.0,
      "learning_rate": 4.1666666666666665e-05,
      "loss": 0.66,
      "step": 240
    },
    {
      "epoch": 30.0,
      "eval_accuracy": 0.6074766355140186,
      "eval_loss": 0.6799367070198059,
      "eval_runtime": 1.8619,
      "eval_samples_per_second": 57.467,
      "eval_steps_per_second": 3.76,
      "step": 240
    },
    {
      "epoch": 31.0,
      "eval_accuracy": 0.6074766355140186,
      "eval_loss": 0.6676725745201111,
      "eval_runtime": 1.8574,
      "eval_samples_per_second": 57.606,
      "eval_steps_per_second": 3.769,
      "step": 248
    },
    {
      "epoch": 31.25,
      "learning_rate": 3.645833333333333e-05,
      "loss": 0.6469,
      "step": 250
    },
    {
      "epoch": 32.0,
      "eval_accuracy": 0.5981308411214953,
      "eval_loss": 0.6734670996665955,
      "eval_runtime": 1.8664,
      "eval_samples_per_second": 57.329,
      "eval_steps_per_second": 3.75,
      "step": 256
    },
    {
      "epoch": 32.5,
      "learning_rate": 3.125e-05,
      "loss": 0.6355,
      "step": 260
    },
    {
      "epoch": 33.0,
      "eval_accuracy": 0.616822429906542,
      "eval_loss": 0.6852976083755493,
      "eval_runtime": 2.0311,
      "eval_samples_per_second": 52.681,
      "eval_steps_per_second": 3.446,
      "step": 264
    },
    {
      "epoch": 33.75,
      "learning_rate": 2.6041666666666665e-05,
      "loss": 0.6245,
      "step": 270
    },
    {
      "epoch": 34.0,
      "eval_accuracy": 0.6261682242990654,
      "eval_loss": 0.7008370161056519,
      "eval_runtime": 1.8744,
      "eval_samples_per_second": 57.084,
      "eval_steps_per_second": 3.734,
      "step": 272
    },
    {
      "epoch": 35.0,
      "learning_rate": 2.0833333333333333e-05,
      "loss": 0.6306,
      "step": 280
    },
    {
      "epoch": 35.0,
      "eval_accuracy": 0.5981308411214953,
      "eval_loss": 0.6990097761154175,
      "eval_runtime": 1.9194,
      "eval_samples_per_second": 55.745,
      "eval_steps_per_second": 3.647,
      "step": 280
    },
    {
      "epoch": 36.0,
      "eval_accuracy": 0.6355140186915887,
      "eval_loss": 0.6980569958686829,
      "eval_runtime": 1.9115,
      "eval_samples_per_second": 55.978,
      "eval_steps_per_second": 3.662,
      "step": 288
    },
    {
      "epoch": 36.25,
      "learning_rate": 1.5625e-05,
      "loss": 0.6208,
      "step": 290
    },
    {
      "epoch": 37.0,
      "eval_accuracy": 0.6261682242990654,
      "eval_loss": 0.7103211879730225,
      "eval_runtime": 1.9169,
      "eval_samples_per_second": 55.818,
      "eval_steps_per_second": 3.652,
      "step": 296
    },
    {
      "epoch": 37.5,
      "learning_rate": 1.0416666666666666e-05,
      "loss": 0.6339,
      "step": 300
    },
    {
      "epoch": 38.0,
      "eval_accuracy": 0.6355140186915887,
      "eval_loss": 0.7050479054450989,
      "eval_runtime": 1.9345,
      "eval_samples_per_second": 55.312,
      "eval_steps_per_second": 3.619,
      "step": 304
    },
    {
      "epoch": 38.75,
      "learning_rate": 5.208333333333333e-06,
      "loss": 0.5959,
      "step": 310
    },
    {
      "epoch": 39.0,
      "eval_accuracy": 0.6355140186915887,
      "eval_loss": 0.698934018611908,
      "eval_runtime": 1.8634,
      "eval_samples_per_second": 57.421,
      "eval_steps_per_second": 3.757,
      "step": 312
    },
    {
      "epoch": 40.0,
      "learning_rate": 0.0,
      "loss": 0.6059,
      "step": 320
    },
    {
      "epoch": 40.0,
      "eval_accuracy": 0.6355140186915887,
      "eval_loss": 0.6989688277244568,
      "eval_runtime": 1.8409,
      "eval_samples_per_second": 58.123,
      "eval_steps_per_second": 3.802,
      "step": 320
    },
    {
      "epoch": 40.0,
      "step": 320,
      "total_flos": 6.480838238876467e+17,
      "train_loss": 0.9213116973638534,
      "train_runtime": 581.2908,
      "train_samples_per_second": 34.269,
      "train_steps_per_second": 0.55
    }
  ],
  "logging_steps": 10,
  "max_steps": 320,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 40,
  "save_steps": 500,
  "total_flos": 6.480838238876467e+17,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}