{
  "best_metric": 0.9302325581395349,
  "best_model_checkpoint": "vit-base-patch16-224-dmae-va-da-40D/checkpoint-91",
  "epoch": 36.92307692307692,
  "eval_steps": 500,
  "global_step": 120,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.92,
      "eval_accuracy": 0.4186046511627907,
      "eval_loss": 1.2900550365447998,
      "eval_runtime": 1.1372,
      "eval_samples_per_second": 37.812,
      "eval_steps_per_second": 1.759,
      "step": 3
    },
    {
      "epoch": 1.85,
      "eval_accuracy": 0.4418604651162791,
      "eval_loss": 1.2313674688339233,
      "eval_runtime": 0.5544,
      "eval_samples_per_second": 77.555,
      "eval_steps_per_second": 3.607,
      "step": 6
    },
    {
      "epoch": 2.77,
      "eval_accuracy": 0.46511627906976744,
      "eval_loss": 1.1530039310455322,
      "eval_runtime": 0.558,
      "eval_samples_per_second": 77.054,
      "eval_steps_per_second": 3.584,
      "step": 9
    },
    {
      "epoch": 3.69,
      "learning_rate": 5e-05,
      "loss": 1.2976,
      "step": 12
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.6046511627906976,
      "eval_loss": 0.9851991534233093,
      "eval_runtime": 0.5772,
      "eval_samples_per_second": 74.501,
      "eval_steps_per_second": 3.465,
      "step": 13
    },
    {
      "epoch": 4.92,
      "eval_accuracy": 0.7674418604651163,
      "eval_loss": 0.8450184464454651,
      "eval_runtime": 0.6419,
      "eval_samples_per_second": 66.992,
      "eval_steps_per_second": 3.116,
      "step": 16
    },
    {
      "epoch": 5.85,
      "eval_accuracy": 0.6511627906976745,
      "eval_loss": 0.8366963863372803,
      "eval_runtime": 0.6054,
      "eval_samples_per_second": 71.03,
      "eval_steps_per_second": 3.304,
      "step": 19
    },
    {
      "epoch": 6.77,
      "eval_accuracy": 0.7209302325581395,
      "eval_loss": 0.7544969916343689,
      "eval_runtime": 0.6137,
      "eval_samples_per_second": 70.066,
      "eval_steps_per_second": 3.259,
      "step": 22
    },
    {
      "epoch": 7.38,
      "learning_rate": 4.4444444444444447e-05,
      "loss": 0.8294,
      "step": 24
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.7906976744186046,
      "eval_loss": 0.6711422204971313,
      "eval_runtime": 0.6274,
      "eval_samples_per_second": 68.534,
      "eval_steps_per_second": 3.188,
      "step": 26
    },
    {
      "epoch": 8.92,
      "eval_accuracy": 0.7209302325581395,
      "eval_loss": 0.6739276051521301,
      "eval_runtime": 0.6711,
      "eval_samples_per_second": 64.073,
      "eval_steps_per_second": 2.98,
      "step": 29
    },
    {
      "epoch": 9.85,
      "eval_accuracy": 0.7441860465116279,
      "eval_loss": 0.6009616851806641,
      "eval_runtime": 0.6278,
      "eval_samples_per_second": 68.489,
      "eval_steps_per_second": 3.186,
      "step": 32
    },
    {
      "epoch": 10.77,
      "eval_accuracy": 0.7441860465116279,
      "eval_loss": 0.5369246602058411,
      "eval_runtime": 0.6174,
      "eval_samples_per_second": 69.646,
      "eval_steps_per_second": 3.239,
      "step": 35
    },
    {
      "epoch": 11.08,
      "learning_rate": 3.888888888888889e-05,
      "loss": 0.4293,
      "step": 36
    },
    {
      "epoch": 12.0,
      "eval_accuracy": 0.7906976744186046,
      "eval_loss": 0.5271600484848022,
      "eval_runtime": 0.6878,
      "eval_samples_per_second": 62.521,
      "eval_steps_per_second": 2.908,
      "step": 39
    },
    {
      "epoch": 12.92,
      "eval_accuracy": 0.7441860465116279,
      "eval_loss": 0.5216893553733826,
      "eval_runtime": 0.6118,
      "eval_samples_per_second": 70.282,
      "eval_steps_per_second": 3.269,
      "step": 42
    },
    {
      "epoch": 13.85,
      "eval_accuracy": 0.7674418604651163,
      "eval_loss": 0.4843863546848297,
      "eval_runtime": 0.605,
      "eval_samples_per_second": 71.078,
      "eval_steps_per_second": 3.306,
      "step": 45
    },
    {
      "epoch": 14.77,
      "learning_rate": 3.3333333333333335e-05,
      "loss": 0.2695,
      "step": 48
    },
    {
      "epoch": 14.77,
      "eval_accuracy": 0.7906976744186046,
      "eval_loss": 0.4947708547115326,
      "eval_runtime": 0.664,
      "eval_samples_per_second": 64.758,
      "eval_steps_per_second": 3.012,
      "step": 48
    },
    {
      "epoch": 16.0,
      "eval_accuracy": 0.7674418604651163,
      "eval_loss": 0.4775972068309784,
      "eval_runtime": 0.6155,
      "eval_samples_per_second": 69.858,
      "eval_steps_per_second": 3.249,
      "step": 52
    },
    {
      "epoch": 16.92,
      "eval_accuracy": 0.7906976744186046,
      "eval_loss": 0.44103220105171204,
      "eval_runtime": 0.6148,
      "eval_samples_per_second": 69.936,
      "eval_steps_per_second": 3.253,
      "step": 55
    },
    {
      "epoch": 17.85,
      "eval_accuracy": 0.7441860465116279,
      "eval_loss": 0.48707905411720276,
      "eval_runtime": 0.6184,
      "eval_samples_per_second": 69.53,
      "eval_steps_per_second": 3.234,
      "step": 58
    },
    {
      "epoch": 18.46,
      "learning_rate": 2.777777777777778e-05,
      "loss": 0.1905,
      "step": 60
    },
    {
      "epoch": 18.77,
      "eval_accuracy": 0.7906976744186046,
      "eval_loss": 0.4374682903289795,
      "eval_runtime": 0.6741,
      "eval_samples_per_second": 63.79,
      "eval_steps_per_second": 2.967,
      "step": 61
    },
    {
      "epoch": 20.0,
      "eval_accuracy": 0.813953488372093,
      "eval_loss": 0.45782390236854553,
      "eval_runtime": 0.6053,
      "eval_samples_per_second": 71.04,
      "eval_steps_per_second": 3.304,
      "step": 65
    },
    {
      "epoch": 20.92,
      "eval_accuracy": 0.813953488372093,
      "eval_loss": 0.4955892264842987,
      "eval_runtime": 0.6057,
      "eval_samples_per_second": 70.987,
      "eval_steps_per_second": 3.302,
      "step": 68
    },
    {
      "epoch": 21.85,
      "eval_accuracy": 0.813953488372093,
      "eval_loss": 0.4500349164009094,
      "eval_runtime": 0.6339,
      "eval_samples_per_second": 67.831,
      "eval_steps_per_second": 3.155,
      "step": 71
    },
    {
      "epoch": 22.15,
      "learning_rate": 2.2222222222222223e-05,
      "loss": 0.135,
      "step": 72
    },
    {
      "epoch": 22.77,
      "eval_accuracy": 0.8604651162790697,
      "eval_loss": 0.40709012746810913,
      "eval_runtime": 0.6806,
      "eval_samples_per_second": 63.179,
      "eval_steps_per_second": 2.939,
      "step": 74
    },
    {
      "epoch": 24.0,
      "eval_accuracy": 0.8604651162790697,
      "eval_loss": 0.41575565934181213,
      "eval_runtime": 0.6102,
      "eval_samples_per_second": 70.471,
      "eval_steps_per_second": 3.278,
      "step": 78
    },
    {
      "epoch": 24.92,
      "eval_accuracy": 0.8372093023255814,
      "eval_loss": 0.43799346685409546,
      "eval_runtime": 0.6137,
      "eval_samples_per_second": 70.068,
      "eval_steps_per_second": 3.259,
      "step": 81
    },
    {
      "epoch": 25.85,
      "learning_rate": 1.6666666666666667e-05,
      "loss": 0.1485,
      "step": 84
    },
    {
      "epoch": 25.85,
      "eval_accuracy": 0.813953488372093,
      "eval_loss": 0.4280536472797394,
      "eval_runtime": 0.6787,
      "eval_samples_per_second": 63.361,
      "eval_steps_per_second": 2.947,
      "step": 84
    },
    {
      "epoch": 26.77,
      "eval_accuracy": 0.8837209302325582,
      "eval_loss": 0.3776519298553467,
      "eval_runtime": 0.6159,
      "eval_samples_per_second": 69.822,
      "eval_steps_per_second": 3.248,
      "step": 87
    },
    {
      "epoch": 28.0,
      "eval_accuracy": 0.9302325581395349,
      "eval_loss": 0.340448260307312,
      "eval_runtime": 0.6251,
      "eval_samples_per_second": 68.79,
      "eval_steps_per_second": 3.2,
      "step": 91
    },
    {
      "epoch": 28.92,
      "eval_accuracy": 0.9069767441860465,
      "eval_loss": 0.3580568730831146,
      "eval_runtime": 0.6149,
      "eval_samples_per_second": 69.927,
      "eval_steps_per_second": 3.252,
      "step": 94
    },
    {
      "epoch": 29.54,
      "learning_rate": 1.1111111111111112e-05,
      "loss": 0.1001,
      "step": 96
    },
    {
      "epoch": 29.85,
      "eval_accuracy": 0.8604651162790697,
      "eval_loss": 0.38068118691444397,
      "eval_runtime": 0.6756,
      "eval_samples_per_second": 63.646,
      "eval_steps_per_second": 2.96,
      "step": 97
    },
    {
      "epoch": 30.77,
      "eval_accuracy": 0.8837209302325582,
      "eval_loss": 0.3700192868709564,
      "eval_runtime": 0.6122,
      "eval_samples_per_second": 70.236,
      "eval_steps_per_second": 3.267,
      "step": 100
    },
    {
      "epoch": 32.0,
      "eval_accuracy": 0.8837209302325582,
      "eval_loss": 0.37302762269973755,
      "eval_runtime": 0.6248,
      "eval_samples_per_second": 68.822,
      "eval_steps_per_second": 3.201,
      "step": 104
    },
    {
      "epoch": 32.92,
      "eval_accuracy": 0.8837209302325582,
      "eval_loss": 0.38681066036224365,
      "eval_runtime": 0.6675,
      "eval_samples_per_second": 64.42,
      "eval_steps_per_second": 2.996,
      "step": 107
    },
    {
      "epoch": 33.23,
      "learning_rate": 5.555555555555556e-06,
      "loss": 0.0797,
      "step": 108
    },
    {
      "epoch": 33.85,
      "eval_accuracy": 0.8604651162790697,
      "eval_loss": 0.38833436369895935,
      "eval_runtime": 0.6127,
      "eval_samples_per_second": 70.184,
      "eval_steps_per_second": 3.264,
      "step": 110
    },
    {
      "epoch": 34.77,
      "eval_accuracy": 0.8372093023255814,
      "eval_loss": 0.39326179027557373,
      "eval_runtime": 0.6145,
      "eval_samples_per_second": 69.979,
      "eval_steps_per_second": 3.255,
      "step": 113
    },
    {
      "epoch": 36.0,
      "eval_accuracy": 0.8372093023255814,
      "eval_loss": 0.39980578422546387,
      "eval_runtime": 0.6129,
      "eval_samples_per_second": 70.154,
      "eval_steps_per_second": 3.263,
      "step": 117
    },
    {
      "epoch": 36.92,
      "learning_rate": 0.0,
      "loss": 0.0991,
      "step": 120
    },
    {
      "epoch": 36.92,
      "eval_accuracy": 0.8372093023255814,
      "eval_loss": 0.40144917368888855,
      "eval_runtime": 0.6828,
      "eval_samples_per_second": 62.975,
      "eval_steps_per_second": 2.929,
      "step": 120
    },
    {
      "epoch": 36.92,
      "step": 120,
      "total_flos": 1.1038156860172861e+18,
      "train_loss": 0.3578883737325668,
      "train_runtime": 762.5104,
      "train_samples_per_second": 20.196,
      "train_steps_per_second": 0.157
    }
  ],
  "logging_steps": 12,
  "max_steps": 120,
  "num_train_epochs": 40,
  "save_steps": 500,
  "total_flos": 1.1038156860172861e+18,
  "trial_name": null,
  "trial_params": null
}