{
  "best_metric": 0.8387096774193549,
  "best_model_checkpoint": "beit-base-patch16-224-OT-2\\checkpoint-148",
  "epoch": 36.36363636363637,
  "eval_steps": 500,
  "global_step": 200,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.91,
      "eval_accuracy": 0.08064516129032258,
      "eval_loss": 1.853175163269043,
      "eval_runtime": 2.2425,
      "eval_samples_per_second": 27.647,
      "eval_steps_per_second": 1.784,
      "step": 5
    },
    {
      "epoch": 1.82,
      "learning_rate": 1.75e-05,
      "loss": 1.7494,
      "step": 10
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.08064516129032258,
      "eval_loss": 1.7818024158477783,
      "eval_runtime": 2.2481,
      "eval_samples_per_second": 27.579,
      "eval_steps_per_second": 1.779,
      "step": 11
    },
    {
      "epoch": 2.91,
      "eval_accuracy": 0.08064516129032258,
      "eval_loss": 1.6613073348999023,
      "eval_runtime": 2.2572,
      "eval_samples_per_second": 27.468,
      "eval_steps_per_second": 1.772,
      "step": 16
    },
    {
      "epoch": 3.64,
      "learning_rate": 3.5e-05,
      "loss": 1.6235,
      "step": 20
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.08064516129032258,
      "eval_loss": 1.4651364088058472,
      "eval_runtime": 2.2621,
      "eval_samples_per_second": 27.409,
      "eval_steps_per_second": 1.768,
      "step": 22
    },
    {
      "epoch": 4.91,
      "eval_accuracy": 0.08064516129032258,
      "eval_loss": 1.3293113708496094,
      "eval_runtime": 2.221,
      "eval_samples_per_second": 27.915,
      "eval_steps_per_second": 1.801,
      "step": 27
    },
    {
      "epoch": 5.45,
      "learning_rate": 3.3055555555555553e-05,
      "loss": 1.3836,
      "step": 30
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.5161290322580645,
      "eval_loss": 1.2034481763839722,
      "eval_runtime": 2.214,
      "eval_samples_per_second": 28.003,
      "eval_steps_per_second": 1.807,
      "step": 33
    },
    {
      "epoch": 6.91,
      "eval_accuracy": 0.3709677419354839,
      "eval_loss": 1.1748452186584473,
      "eval_runtime": 2.2906,
      "eval_samples_per_second": 27.068,
      "eval_steps_per_second": 1.746,
      "step": 38
    },
    {
      "epoch": 7.27,
      "learning_rate": 3.111111111111111e-05,
      "loss": 1.2192,
      "step": 40
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.46774193548387094,
      "eval_loss": 1.0814906358718872,
      "eval_runtime": 2.3561,
      "eval_samples_per_second": 26.315,
      "eval_steps_per_second": 1.698,
      "step": 44
    },
    {
      "epoch": 8.91,
      "eval_accuracy": 0.5,
      "eval_loss": 1.0237908363342285,
      "eval_runtime": 2.2371,
      "eval_samples_per_second": 27.715,
      "eval_steps_per_second": 1.788,
      "step": 49
    },
    {
      "epoch": 9.09,
      "learning_rate": 2.9166666666666666e-05,
      "loss": 1.093,
      "step": 50
    },
    {
      "epoch": 10.0,
      "eval_accuracy": 0.45161290322580644,
      "eval_loss": 1.0225210189819336,
      "eval_runtime": 2.1705,
      "eval_samples_per_second": 28.565,
      "eval_steps_per_second": 1.843,
      "step": 55
    },
    {
      "epoch": 10.91,
      "learning_rate": 2.722222222222222e-05,
      "loss": 0.9938,
      "step": 60
    },
    {
      "epoch": 10.91,
      "eval_accuracy": 0.6451612903225806,
      "eval_loss": 0.9649767279624939,
      "eval_runtime": 2.178,
      "eval_samples_per_second": 28.466,
      "eval_steps_per_second": 1.837,
      "step": 60
    },
    {
      "epoch": 12.0,
      "eval_accuracy": 0.6935483870967742,
      "eval_loss": 0.9313719868659973,
      "eval_runtime": 2.2561,
      "eval_samples_per_second": 27.482,
      "eval_steps_per_second": 1.773,
      "step": 66
    },
    {
      "epoch": 12.73,
      "learning_rate": 2.5277777777777776e-05,
      "loss": 0.9235,
      "step": 70
    },
    {
      "epoch": 12.91,
      "eval_accuracy": 0.6451612903225806,
      "eval_loss": 0.9490375518798828,
      "eval_runtime": 2.2295,
      "eval_samples_per_second": 27.809,
      "eval_steps_per_second": 1.794,
      "step": 71
    },
    {
      "epoch": 14.0,
      "eval_accuracy": 0.7258064516129032,
      "eval_loss": 0.8233911395072937,
      "eval_runtime": 2.1996,
      "eval_samples_per_second": 28.187,
      "eval_steps_per_second": 1.819,
      "step": 77
    },
    {
      "epoch": 14.55,
      "learning_rate": 2.333333333333333e-05,
      "loss": 0.8258,
      "step": 80
    },
    {
      "epoch": 14.91,
      "eval_accuracy": 0.7258064516129032,
      "eval_loss": 0.8159334659576416,
      "eval_runtime": 2.2264,
      "eval_samples_per_second": 27.848,
      "eval_steps_per_second": 1.797,
      "step": 82
    },
    {
      "epoch": 16.0,
      "eval_accuracy": 0.7419354838709677,
      "eval_loss": 0.7513789534568787,
      "eval_runtime": 2.242,
      "eval_samples_per_second": 27.653,
      "eval_steps_per_second": 1.784,
      "step": 88
    },
    {
      "epoch": 16.36,
      "learning_rate": 2.138888888888889e-05,
      "loss": 0.716,
      "step": 90
    },
    {
      "epoch": 16.91,
      "eval_accuracy": 0.7419354838709677,
      "eval_loss": 0.7468788027763367,
      "eval_runtime": 2.2824,
      "eval_samples_per_second": 27.165,
      "eval_steps_per_second": 1.753,
      "step": 93
    },
    {
      "epoch": 18.0,
      "eval_accuracy": 0.7903225806451613,
      "eval_loss": 0.6733937859535217,
      "eval_runtime": 2.2143,
      "eval_samples_per_second": 28.0,
      "eval_steps_per_second": 1.806,
      "step": 99
    },
    {
      "epoch": 18.18,
      "learning_rate": 1.9444444444444442e-05,
      "loss": 0.6026,
      "step": 100
    },
    {
      "epoch": 18.91,
      "eval_accuracy": 0.7580645161290323,
      "eval_loss": 0.6926217079162598,
      "eval_runtime": 2.2096,
      "eval_samples_per_second": 28.06,
      "eval_steps_per_second": 1.81,
      "step": 104
    },
    {
      "epoch": 20.0,
      "learning_rate": 1.75e-05,
      "loss": 0.5725,
      "step": 110
    },
    {
      "epoch": 20.0,
      "eval_accuracy": 0.7258064516129032,
      "eval_loss": 0.7951985001564026,
      "eval_runtime": 2.1841,
      "eval_samples_per_second": 28.388,
      "eval_steps_per_second": 1.831,
      "step": 110
    },
    {
      "epoch": 20.91,
      "eval_accuracy": 0.7741935483870968,
      "eval_loss": 0.6283877491950989,
      "eval_runtime": 2.2365,
      "eval_samples_per_second": 27.721,
      "eval_steps_per_second": 1.788,
      "step": 115
    },
    {
      "epoch": 21.82,
      "learning_rate": 1.5555555555555555e-05,
      "loss": 0.554,
      "step": 120
    },
    {
      "epoch": 22.0,
      "eval_accuracy": 0.7741935483870968,
      "eval_loss": 0.6316818594932556,
      "eval_runtime": 2.1575,
      "eval_samples_per_second": 28.737,
      "eval_steps_per_second": 1.854,
      "step": 121
    },
    {
      "epoch": 22.91,
      "eval_accuracy": 0.7419354838709677,
      "eval_loss": 0.6361408233642578,
      "eval_runtime": 2.2285,
      "eval_samples_per_second": 27.821,
      "eval_steps_per_second": 1.795,
      "step": 126
    },
    {
      "epoch": 23.64,
      "learning_rate": 1.361111111111111e-05,
      "loss": 0.5162,
      "step": 130
    },
    {
      "epoch": 24.0,
      "eval_accuracy": 0.8225806451612904,
      "eval_loss": 0.5500665903091431,
      "eval_runtime": 2.2304,
      "eval_samples_per_second": 27.798,
      "eval_steps_per_second": 1.793,
      "step": 132
    },
    {
      "epoch": 24.91,
      "eval_accuracy": 0.7580645161290323,
      "eval_loss": 0.6277778744697571,
      "eval_runtime": 2.2482,
      "eval_samples_per_second": 27.577,
      "eval_steps_per_second": 1.779,
      "step": 137
    },
    {
      "epoch": 25.45,
      "learning_rate": 1.1666666666666665e-05,
      "loss": 0.4768,
      "step": 140
    },
    {
      "epoch": 26.0,
      "eval_accuracy": 0.7903225806451613,
      "eval_loss": 0.5867522358894348,
      "eval_runtime": 2.211,
      "eval_samples_per_second": 28.041,
      "eval_steps_per_second": 1.809,
      "step": 143
    },
    {
      "epoch": 26.91,
      "eval_accuracy": 0.8387096774193549,
      "eval_loss": 0.5047271251678467,
      "eval_runtime": 2.199,
      "eval_samples_per_second": 28.194,
      "eval_steps_per_second": 1.819,
      "step": 148
    },
    {
      "epoch": 27.27,
      "learning_rate": 9.722222222222221e-06,
      "loss": 0.4488,
      "step": 150
    },
    {
      "epoch": 28.0,
      "eval_accuracy": 0.7903225806451613,
      "eval_loss": 0.5263858437538147,
      "eval_runtime": 2.226,
      "eval_samples_per_second": 27.852,
      "eval_steps_per_second": 1.797,
      "step": 154
    },
    {
      "epoch": 28.91,
      "eval_accuracy": 0.8387096774193549,
      "eval_loss": 0.49419379234313965,
      "eval_runtime": 2.2036,
      "eval_samples_per_second": 28.136,
      "eval_steps_per_second": 1.815,
      "step": 159
    },
    {
      "epoch": 29.09,
      "learning_rate": 7.777777777777777e-06,
      "loss": 0.4281,
      "step": 160
    },
    {
      "epoch": 30.0,
      "eval_accuracy": 0.8387096774193549,
      "eval_loss": 0.5126697421073914,
      "eval_runtime": 2.1663,
      "eval_samples_per_second": 28.62,
      "eval_steps_per_second": 1.846,
      "step": 165
    },
    {
      "epoch": 30.91,
      "learning_rate": 5.833333333333332e-06,
      "loss": 0.4126,
      "step": 170
    },
    {
      "epoch": 30.91,
      "eval_accuracy": 0.8387096774193549,
      "eval_loss": 0.5026701092720032,
      "eval_runtime": 2.1795,
      "eval_samples_per_second": 28.447,
      "eval_steps_per_second": 1.835,
      "step": 170
    },
    {
      "epoch": 32.0,
      "eval_accuracy": 0.7741935483870968,
      "eval_loss": 0.5387425422668457,
      "eval_runtime": 2.5256,
      "eval_samples_per_second": 24.549,
      "eval_steps_per_second": 1.584,
      "step": 176
    },
    {
      "epoch": 32.73,
      "learning_rate": 3.888888888888889e-06,
      "loss": 0.4326,
      "step": 180
    },
    {
      "epoch": 32.91,
      "eval_accuracy": 0.7903225806451613,
      "eval_loss": 0.5250905156135559,
      "eval_runtime": 2.2035,
      "eval_samples_per_second": 28.137,
      "eval_steps_per_second": 1.815,
      "step": 181
    },
    {
      "epoch": 34.0,
      "eval_accuracy": 0.8064516129032258,
      "eval_loss": 0.5091389417648315,
      "eval_runtime": 2.3211,
      "eval_samples_per_second": 26.712,
      "eval_steps_per_second": 1.723,
      "step": 187
    },
    {
      "epoch": 34.55,
      "learning_rate": 1.9444444444444444e-06,
      "loss": 0.3765,
      "step": 190
    },
    {
      "epoch": 34.91,
      "eval_accuracy": 0.8064516129032258,
      "eval_loss": 0.5141739845275879,
      "eval_runtime": 2.2025,
      "eval_samples_per_second": 28.15,
      "eval_steps_per_second": 1.816,
      "step": 192
    },
    {
      "epoch": 36.0,
      "eval_accuracy": 0.7903225806451613,
      "eval_loss": 0.514202356338501,
      "eval_runtime": 2.1798,
      "eval_samples_per_second": 28.443,
      "eval_steps_per_second": 1.835,
      "step": 198
    },
    {
      "epoch": 36.36,
      "learning_rate": 0.0,
      "loss": 0.3913,
      "step": 200
    },
    {
      "epoch": 36.36,
      "eval_accuracy": 0.7903225806451613,
      "eval_loss": 0.5144444108009338,
      "eval_runtime": 2.196,
      "eval_samples_per_second": 28.233,
      "eval_steps_per_second": 1.821,
      "step": 200
    },
    {
      "epoch": 36.36,
      "step": 200,
      "total_flos": 9.859141601338982e+17,
      "train_loss": 0.786993111371994,
      "train_runtime": 718.789,
      "train_samples_per_second": 19.477,
      "train_steps_per_second": 0.278
    }
  ],
  "logging_steps": 10,
  "max_steps": 200,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 40,
  "save_steps": 500,
  "total_flos": 9.859141601338982e+17,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}