{
  "best_metric": 1.0,
  "best_model_checkpoint": "vit-base-patch16-224-Trial008-YEL_STEM2/checkpoint-10",
  "epoch": 33.333333333333336,
  "global_step": 50,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.67,
      "learning_rate": 1e-05,
      "loss": 0.7598,
      "step": 1
    },
    {
      "epoch": 0.67,
      "eval_accuracy": 0.6470588235294118,
      "eval_loss": 0.6967736482620239,
      "eval_runtime": 0.1155,
      "eval_samples_per_second": 147.167,
      "eval_steps_per_second": 8.657,
      "step": 1
    },
    {
      "epoch": 1.33,
      "learning_rate": 2e-05,
      "loss": 0.7926,
      "step": 2
    },
    {
      "epoch": 2.0,
      "learning_rate": 3e-05,
      "loss": 0.7294,
      "step": 3
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.47058823529411764,
      "eval_loss": 0.766238808631897,
      "eval_runtime": 0.1178,
      "eval_samples_per_second": 144.35,
      "eval_steps_per_second": 8.491,
      "step": 3
    },
    {
      "epoch": 2.67,
      "learning_rate": 4e-05,
      "loss": 0.6662,
      "step": 4
    },
    {
      "epoch": 2.67,
      "eval_accuracy": 0.5882352941176471,
      "eval_loss": 0.7196066379547119,
      "eval_runtime": 0.1128,
      "eval_samples_per_second": 150.728,
      "eval_steps_per_second": 8.866,
      "step": 4
    },
    {
      "epoch": 3.33,
      "learning_rate": 5e-05,
      "loss": 0.6872,
      "step": 5
    },
    {
      "epoch": 4.0,
      "learning_rate": 4.888888888888889e-05,
      "loss": 0.5662,
      "step": 6
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8235294117647058,
      "eval_loss": 0.3941333293914795,
      "eval_runtime": 0.1124,
      "eval_samples_per_second": 151.225,
      "eval_steps_per_second": 8.896,
      "step": 6
    },
    {
      "epoch": 4.67,
      "learning_rate": 4.7777777777777784e-05,
      "loss": 0.4781,
      "step": 7
    },
    {
      "epoch": 4.67,
      "eval_accuracy": 0.8235294117647058,
      "eval_loss": 0.3458152413368225,
      "eval_runtime": 0.1128,
      "eval_samples_per_second": 150.724,
      "eval_steps_per_second": 8.866,
      "step": 7
    },
    {
      "epoch": 5.33,
      "learning_rate": 4.666666666666667e-05,
      "loss": 0.3621,
      "step": 8
    },
    {
      "epoch": 6.0,
      "learning_rate": 4.555555555555556e-05,
      "loss": 0.3259,
      "step": 9
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.9411764705882353,
      "eval_loss": 0.16989070177078247,
      "eval_runtime": 0.1116,
      "eval_samples_per_second": 152.266,
      "eval_steps_per_second": 8.957,
      "step": 9
    },
    {
      "epoch": 6.67,
      "learning_rate": 4.4444444444444447e-05,
      "loss": 0.2903,
      "step": 10
    },
    {
      "epoch": 6.67,
      "eval_accuracy": 1.0,
      "eval_loss": 0.10236643999814987,
      "eval_runtime": 0.112,
      "eval_samples_per_second": 151.781,
      "eval_steps_per_second": 8.928,
      "step": 10
    },
    {
      "epoch": 7.33,
      "learning_rate": 4.3333333333333334e-05,
      "loss": 0.2539,
      "step": 11
    },
    {
      "epoch": 8.0,
      "learning_rate": 4.222222222222222e-05,
      "loss": 0.2206,
      "step": 12
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 1.0,
      "eval_loss": 0.07876813411712646,
      "eval_runtime": 0.1154,
      "eval_samples_per_second": 147.279,
      "eval_steps_per_second": 8.663,
      "step": 12
    },
    {
      "epoch": 8.67,
      "learning_rate": 4.111111111111111e-05,
      "loss": 0.3215,
      "step": 13
    },
    {
      "epoch": 8.67,
      "eval_accuracy": 1.0,
      "eval_loss": 0.04137047752737999,
      "eval_runtime": 0.1141,
      "eval_samples_per_second": 148.944,
      "eval_steps_per_second": 8.761,
      "step": 13
    },
    {
      "epoch": 9.33,
      "learning_rate": 4e-05,
      "loss": 0.1603,
      "step": 14
    },
    {
      "epoch": 10.0,
      "learning_rate": 3.888888888888889e-05,
      "loss": 0.1741,
      "step": 15
    },
    {
      "epoch": 10.0,
      "eval_accuracy": 1.0,
      "eval_loss": 0.021776309236884117,
      "eval_runtime": 0.1108,
      "eval_samples_per_second": 153.382,
      "eval_steps_per_second": 9.022,
      "step": 15
    },
    {
      "epoch": 10.67,
      "learning_rate": 3.777777777777778e-05,
      "loss": 0.2222,
      "step": 16
    },
    {
      "epoch": 10.67,
      "eval_accuracy": 1.0,
      "eval_loss": 0.020698698237538338,
      "eval_runtime": 0.1126,
      "eval_samples_per_second": 151.006,
      "eval_steps_per_second": 8.883,
      "step": 16
    },
    {
      "epoch": 11.33,
      "learning_rate": 3.6666666666666666e-05,
      "loss": 0.228,
      "step": 17
    },
    {
      "epoch": 12.0,
      "learning_rate": 3.555555555555556e-05,
      "loss": 0.1534,
      "step": 18
    },
    {
      "epoch": 12.0,
      "eval_accuracy": 1.0,
      "eval_loss": 0.012763559818267822,
      "eval_runtime": 0.1133,
      "eval_samples_per_second": 150.062,
      "eval_steps_per_second": 8.827,
      "step": 18
    },
    {
      "epoch": 12.67,
      "learning_rate": 3.444444444444445e-05,
      "loss": 0.273,
      "step": 19
    },
    {
      "epoch": 12.67,
      "eval_accuracy": 1.0,
      "eval_loss": 0.010324841365218163,
      "eval_runtime": 0.1154,
      "eval_samples_per_second": 147.298,
      "eval_steps_per_second": 8.665,
      "step": 19
    },
    {
      "epoch": 13.33,
      "learning_rate": 3.3333333333333335e-05,
      "loss": 0.3743,
      "step": 20
    },
    {
      "epoch": 14.0,
      "learning_rate": 3.222222222222223e-05,
      "loss": 0.2021,
      "step": 21
    },
    {
      "epoch": 14.0,
      "eval_accuracy": 1.0,
      "eval_loss": 0.007954830303788185,
      "eval_runtime": 0.1144,
      "eval_samples_per_second": 148.646,
      "eval_steps_per_second": 8.744,
      "step": 21
    },
    {
      "epoch": 14.67,
      "learning_rate": 3.111111111111111e-05,
      "loss": 0.2193,
      "step": 22
    },
    {
      "epoch": 14.67,
      "eval_accuracy": 1.0,
      "eval_loss": 0.009969166480004787,
      "eval_runtime": 0.113,
      "eval_samples_per_second": 150.48,
      "eval_steps_per_second": 8.852,
      "step": 22
    },
    {
      "epoch": 15.33,
      "learning_rate": 3e-05,
      "loss": 0.1586,
      "step": 23
    },
    {
      "epoch": 16.0,
      "learning_rate": 2.8888888888888888e-05,
      "loss": 0.2132,
      "step": 24
    },
    {
      "epoch": 16.0,
      "eval_accuracy": 1.0,
      "eval_loss": 0.02467266283929348,
      "eval_runtime": 0.114,
      "eval_samples_per_second": 149.139,
      "eval_steps_per_second": 8.773,
      "step": 24
    },
    {
      "epoch": 16.67,
      "learning_rate": 2.777777777777778e-05,
      "loss": 0.2163,
      "step": 25
    },
    {
      "epoch": 16.67,
      "eval_accuracy": 1.0,
      "eval_loss": 0.02664419636130333,
      "eval_runtime": 0.1146,
      "eval_samples_per_second": 148.38,
      "eval_steps_per_second": 8.728,
      "step": 25
    },
    {
      "epoch": 17.33,
      "learning_rate": 2.6666666666666667e-05,
      "loss": 0.2017,
      "step": 26
    },
    {
      "epoch": 18.0,
      "learning_rate": 2.5555555555555554e-05,
      "loss": 0.1626,
      "step": 27
    },
    {
      "epoch": 18.0,
      "eval_accuracy": 1.0,
      "eval_loss": 0.01014714501798153,
      "eval_runtime": 0.1128,
      "eval_samples_per_second": 150.755,
      "eval_steps_per_second": 8.868,
      "step": 27
    },
    {
      "epoch": 18.67,
      "learning_rate": 2.4444444444444445e-05,
      "loss": 0.2492,
      "step": 28
    },
    {
      "epoch": 18.67,
      "eval_accuracy": 1.0,
      "eval_loss": 0.005902642849832773,
      "eval_runtime": 0.1149,
      "eval_samples_per_second": 147.968,
      "eval_steps_per_second": 8.704,
      "step": 28
    },
    {
      "epoch": 19.33,
      "learning_rate": 2.3333333333333336e-05,
      "loss": 0.1315,
      "step": 29
    },
    {
      "epoch": 20.0,
      "learning_rate": 2.2222222222222223e-05,
      "loss": 0.1308,
      "step": 30
    },
    {
      "epoch": 20.0,
      "eval_accuracy": 1.0,
      "eval_loss": 0.005620644427835941,
      "eval_runtime": 0.1147,
      "eval_samples_per_second": 148.149,
      "eval_steps_per_second": 8.715,
      "step": 30
    },
    {
      "epoch": 20.67,
      "learning_rate": 2.111111111111111e-05,
      "loss": 0.2144,
      "step": 31
    },
    {
      "epoch": 20.67,
      "eval_accuracy": 1.0,
      "eval_loss": 0.006035304628312588,
      "eval_runtime": 0.1154,
      "eval_samples_per_second": 147.274,
      "eval_steps_per_second": 8.663,
      "step": 31
    },
    {
      "epoch": 21.33,
      "learning_rate": 2e-05,
      "loss": 0.1608,
      "step": 32
    },
    {
      "epoch": 22.0,
      "learning_rate": 1.888888888888889e-05,
      "loss": 0.1389,
      "step": 33
    },
    {
      "epoch": 22.0,
      "eval_accuracy": 1.0,
      "eval_loss": 0.004421359859406948,
      "eval_runtime": 0.1131,
      "eval_samples_per_second": 150.363,
      "eval_steps_per_second": 8.845,
      "step": 33
    },
    {
      "epoch": 22.67,
      "learning_rate": 1.777777777777778e-05,
      "loss": 0.2548,
      "step": 34
    },
    {
      "epoch": 22.67,
      "eval_accuracy": 1.0,
      "eval_loss": 0.004027603659778833,
      "eval_runtime": 0.113,
      "eval_samples_per_second": 150.382,
      "eval_steps_per_second": 8.846,
      "step": 34
    },
    {
      "epoch": 23.33,
      "learning_rate": 1.6666666666666667e-05,
      "loss": 0.1916,
      "step": 35
    },
    {
      "epoch": 24.0,
      "learning_rate": 1.5555555555555555e-05,
      "loss": 0.1324,
      "step": 36
    },
    {
      "epoch": 24.0,
      "eval_accuracy": 1.0,
      "eval_loss": 0.0037342249415814877,
      "eval_runtime": 0.1133,
      "eval_samples_per_second": 150.098,
      "eval_steps_per_second": 8.829,
      "step": 36
    },
    {
      "epoch": 24.67,
      "learning_rate": 1.4444444444444444e-05,
      "loss": 0.1958,
      "step": 37
    },
    {
      "epoch": 24.67,
      "eval_accuracy": 1.0,
      "eval_loss": 0.0036163795739412308,
      "eval_runtime": 0.1138,
      "eval_samples_per_second": 149.366,
      "eval_steps_per_second": 8.786,
      "step": 37
    },
    {
      "epoch": 25.33,
      "learning_rate": 1.3333333333333333e-05,
      "loss": 0.1812,
      "step": 38
    },
    {
      "epoch": 26.0,
      "learning_rate": 1.2222222222222222e-05,
      "loss": 0.2476,
      "step": 39
    },
    {
      "epoch": 26.0,
      "eval_accuracy": 1.0,
      "eval_loss": 0.0034641919191926718,
      "eval_runtime": 0.1123,
      "eval_samples_per_second": 151.377,
      "eval_steps_per_second": 8.905,
      "step": 39
    },
    {
      "epoch": 26.67,
      "learning_rate": 1.1111111111111112e-05,
      "loss": 0.1439,
      "step": 40
    },
    {
      "epoch": 26.67,
      "eval_accuracy": 1.0,
      "eval_loss": 0.0033218893222510815,
      "eval_runtime": 0.116,
      "eval_samples_per_second": 146.604,
      "eval_steps_per_second": 8.624,
      "step": 40
    },
    {
      "epoch": 27.33,
      "learning_rate": 1e-05,
      "loss": 0.1375,
      "step": 41
    },
    {
      "epoch": 28.0,
      "learning_rate": 8.88888888888889e-06,
      "loss": 0.1202,
      "step": 42
    },
    {
      "epoch": 28.0,
      "eval_accuracy": 1.0,
      "eval_loss": 0.002967243082821369,
      "eval_runtime": 0.1135,
      "eval_samples_per_second": 149.812,
      "eval_steps_per_second": 8.812,
      "step": 42
    },
    {
      "epoch": 28.67,
      "learning_rate": 7.777777777777777e-06,
      "loss": 0.1368,
      "step": 43
    },
    {
      "epoch": 28.67,
      "eval_accuracy": 1.0,
      "eval_loss": 0.002836533123627305,
      "eval_runtime": 0.1133,
      "eval_samples_per_second": 150.077,
      "eval_steps_per_second": 8.828,
      "step": 43
    },
    {
      "epoch": 29.33,
      "learning_rate": 6.666666666666667e-06,
      "loss": 0.403,
      "step": 44
    },
    {
      "epoch": 30.0,
      "learning_rate": 5.555555555555556e-06,
      "loss": 0.1016,
      "step": 45
    },
    {
      "epoch": 30.0,
      "eval_accuracy": 1.0,
      "eval_loss": 0.0027249290142208338,
      "eval_runtime": 0.1143,
      "eval_samples_per_second": 148.728,
      "eval_steps_per_second": 8.749,
      "step": 45
    },
    {
      "epoch": 30.67,
      "learning_rate": 4.444444444444445e-06,
      "loss": 0.1282,
      "step": 46
    },
    {
      "epoch": 30.67,
      "eval_accuracy": 1.0,
      "eval_loss": 0.0026905094273388386,
      "eval_runtime": 0.1127,
      "eval_samples_per_second": 150.9,
      "eval_steps_per_second": 8.876,
      "step": 46
    },
    {
      "epoch": 31.33,
      "learning_rate": 3.3333333333333333e-06,
      "loss": 0.0778,
      "step": 47
    },
    {
      "epoch": 32.0,
      "learning_rate": 2.2222222222222225e-06,
      "loss": 0.1128,
      "step": 48
    },
    {
      "epoch": 32.0,
      "eval_accuracy": 1.0,
      "eval_loss": 0.0026230851653963327,
      "eval_runtime": 0.1139,
      "eval_samples_per_second": 149.284,
      "eval_steps_per_second": 8.781,
      "step": 48
    },
    {
      "epoch": 32.67,
      "learning_rate": 1.1111111111111112e-06,
      "loss": 0.2366,
      "step": 49
    },
    {
      "epoch": 32.67,
      "eval_accuracy": 1.0,
      "eval_loss": 0.0026192485820502043,
      "eval_runtime": 0.1122,
      "eval_samples_per_second": 151.562,
      "eval_steps_per_second": 8.915,
      "step": 49
    },
    {
      "epoch": 33.33,
      "learning_rate": 0.0,
      "loss": 0.1727,
      "step": 50
    },
    {
      "epoch": 33.33,
      "eval_accuracy": 1.0,
      "eval_loss": 0.002621703315526247,
      "eval_runtime": 0.1147,
      "eval_samples_per_second": 148.149,
      "eval_steps_per_second": 8.715,
      "step": 50
    },
    {
      "epoch": 33.33,
      "step": 50,
      "total_flos": 3.9079210362638746e+17,
      "train_loss": 0.26825646698474886,
      "train_runtime": 260.0145,
      "train_samples_per_second": 29.037,
      "train_steps_per_second": 0.192
    }
  ],
  "max_steps": 50,
  "num_train_epochs": 50,
  "total_flos": 3.9079210362638746e+17,
  "trial_name": null,
  "trial_params": null
}