{
  "best_metric": 0.903125,
  "best_model_checkpoint": "swinv2-tiny-patch4-window8-256-finalterm/checkpoint-120",
  "epoch": 20.0,
  "eval_steps": 500,
  "global_step": 200,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 4.020592212677002,
      "learning_rate": 2.5e-05,
      "loss": 1.3728,
      "step": 10
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.515625,
      "eval_loss": 1.2643654346466064,
      "eval_runtime": 2.8905,
      "eval_samples_per_second": 110.709,
      "eval_steps_per_second": 3.46,
      "step": 10
    },
    {
      "epoch": 2.0,
      "grad_norm": 5.058677673339844,
      "learning_rate": 5e-05,
      "loss": 1.1308,
      "step": 20
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.625,
      "eval_loss": 0.8816311955451965,
      "eval_runtime": 2.7866,
      "eval_samples_per_second": 114.835,
      "eval_steps_per_second": 3.589,
      "step": 20
    },
    {
      "epoch": 3.0,
      "grad_norm": 7.673556327819824,
      "learning_rate": 4.722222222222222e-05,
      "loss": 0.8721,
      "step": 30
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.70625,
      "eval_loss": 0.6829319000244141,
      "eval_runtime": 2.7596,
      "eval_samples_per_second": 115.959,
      "eval_steps_per_second": 3.624,
      "step": 30
    },
    {
      "epoch": 4.0,
      "grad_norm": 4.715519905090332,
      "learning_rate": 4.4444444444444447e-05,
      "loss": 0.6919,
      "step": 40
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.80625,
      "eval_loss": 0.5298391580581665,
      "eval_runtime": 2.8001,
      "eval_samples_per_second": 114.281,
      "eval_steps_per_second": 3.571,
      "step": 40
    },
    {
      "epoch": 5.0,
      "grad_norm": 7.957090377807617,
      "learning_rate": 4.166666666666667e-05,
      "loss": 0.5876,
      "step": 50
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.86875,
      "eval_loss": 0.4100002348423004,
      "eval_runtime": 2.7806,
      "eval_samples_per_second": 115.082,
      "eval_steps_per_second": 3.596,
      "step": 50
    },
    {
      "epoch": 6.0,
      "grad_norm": 6.996800899505615,
      "learning_rate": 3.888888888888889e-05,
      "loss": 0.5504,
      "step": 60
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.853125,
      "eval_loss": 0.4152528643608093,
      "eval_runtime": 2.9983,
      "eval_samples_per_second": 106.727,
      "eval_steps_per_second": 3.335,
      "step": 60
    },
    {
      "epoch": 7.0,
      "grad_norm": 4.525465965270996,
      "learning_rate": 3.611111111111111e-05,
      "loss": 0.459,
      "step": 70
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.859375,
      "eval_loss": 0.3827503025531769,
      "eval_runtime": 2.7918,
      "eval_samples_per_second": 114.62,
      "eval_steps_per_second": 3.582,
      "step": 70
    },
    {
      "epoch": 8.0,
      "grad_norm": 4.351122856140137,
      "learning_rate": 3.3333333333333335e-05,
      "loss": 0.4501,
      "step": 80
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.8625,
      "eval_loss": 0.39407286047935486,
      "eval_runtime": 2.7588,
      "eval_samples_per_second": 115.992,
      "eval_steps_per_second": 3.625,
      "step": 80
    },
    {
      "epoch": 9.0,
      "grad_norm": 5.307958602905273,
      "learning_rate": 3.055555555555556e-05,
      "loss": 0.4312,
      "step": 90
    },
    {
      "epoch": 9.0,
      "eval_accuracy": 0.871875,
      "eval_loss": 0.36500272154808044,
      "eval_runtime": 2.7271,
      "eval_samples_per_second": 117.34,
      "eval_steps_per_second": 3.667,
      "step": 90
    },
    {
      "epoch": 10.0,
      "grad_norm": 5.813844203948975,
      "learning_rate": 2.777777777777778e-05,
      "loss": 0.4119,
      "step": 100
    },
    {
      "epoch": 10.0,
      "eval_accuracy": 0.875,
      "eval_loss": 0.3515123128890991,
      "eval_runtime": 2.736,
      "eval_samples_per_second": 116.961,
      "eval_steps_per_second": 3.655,
      "step": 100
    },
    {
      "epoch": 11.0,
      "grad_norm": 8.393081665039062,
      "learning_rate": 2.5e-05,
      "loss": 0.4014,
      "step": 110
    },
    {
      "epoch": 11.0,
      "eval_accuracy": 0.896875,
      "eval_loss": 0.3110010325908661,
      "eval_runtime": 2.7536,
      "eval_samples_per_second": 116.212,
      "eval_steps_per_second": 3.632,
      "step": 110
    },
    {
      "epoch": 12.0,
      "grad_norm": 5.969035625457764,
      "learning_rate": 2.2222222222222223e-05,
      "loss": 0.3896,
      "step": 120
    },
    {
      "epoch": 12.0,
      "eval_accuracy": 0.903125,
      "eval_loss": 0.3030492663383484,
      "eval_runtime": 2.79,
      "eval_samples_per_second": 114.693,
      "eval_steps_per_second": 3.584,
      "step": 120
    },
    {
      "epoch": 13.0,
      "grad_norm": 4.165198802947998,
      "learning_rate": 1.9444444444444445e-05,
      "loss": 0.3822,
      "step": 130
    },
    {
      "epoch": 13.0,
      "eval_accuracy": 0.88125,
      "eval_loss": 0.34730494022369385,
      "eval_runtime": 2.7802,
      "eval_samples_per_second": 115.099,
      "eval_steps_per_second": 3.597,
      "step": 130
    },
    {
      "epoch": 14.0,
      "grad_norm": 5.396005153656006,
      "learning_rate": 1.6666666666666667e-05,
      "loss": 0.3985,
      "step": 140
    },
    {
      "epoch": 14.0,
      "eval_accuracy": 0.8875,
      "eval_loss": 0.32879379391670227,
      "eval_runtime": 2.7703,
      "eval_samples_per_second": 115.511,
      "eval_steps_per_second": 3.61,
      "step": 140
    },
    {
      "epoch": 15.0,
      "grad_norm": 5.696004390716553,
      "learning_rate": 1.388888888888889e-05,
      "loss": 0.3826,
      "step": 150
    },
    {
      "epoch": 15.0,
      "eval_accuracy": 0.9,
      "eval_loss": 0.2924533486366272,
      "eval_runtime": 2.7303,
      "eval_samples_per_second": 117.204,
      "eval_steps_per_second": 3.663,
      "step": 150
    },
    {
      "epoch": 16.0,
      "grad_norm": 7.218973636627197,
      "learning_rate": 1.1111111111111112e-05,
      "loss": 0.3716,
      "step": 160
    },
    {
      "epoch": 16.0,
      "eval_accuracy": 0.875,
      "eval_loss": 0.36188262701034546,
      "eval_runtime": 2.726,
      "eval_samples_per_second": 117.39,
      "eval_steps_per_second": 3.668,
      "step": 160
    },
    {
      "epoch": 17.0,
      "grad_norm": 7.728688716888428,
      "learning_rate": 8.333333333333334e-06,
      "loss": 0.365,
      "step": 170
    },
    {
      "epoch": 17.0,
      "eval_accuracy": 0.9,
      "eval_loss": 0.2941049039363861,
      "eval_runtime": 2.7206,
      "eval_samples_per_second": 117.621,
      "eval_steps_per_second": 3.676,
      "step": 170
    },
    {
      "epoch": 18.0,
      "grad_norm": 5.3787126541137695,
      "learning_rate": 5.555555555555556e-06,
      "loss": 0.3379,
      "step": 180
    },
    {
      "epoch": 18.0,
      "eval_accuracy": 0.884375,
      "eval_loss": 0.32390105724334717,
      "eval_runtime": 2.7433,
      "eval_samples_per_second": 116.648,
      "eval_steps_per_second": 3.645,
      "step": 180
    },
    {
      "epoch": 19.0,
      "grad_norm": 7.723823070526123,
      "learning_rate": 2.777777777777778e-06,
      "loss": 0.3365,
      "step": 190
    },
    {
      "epoch": 19.0,
      "eval_accuracy": 0.890625,
      "eval_loss": 0.32600170373916626,
      "eval_runtime": 2.7308,
      "eval_samples_per_second": 117.184,
      "eval_steps_per_second": 3.662,
      "step": 190
    },
    {
      "epoch": 20.0,
      "grad_norm": 3.916741132736206,
      "learning_rate": 0.0,
      "loss": 0.3429,
      "step": 200
    },
    {
      "epoch": 20.0,
      "eval_accuracy": 0.89375,
      "eval_loss": 0.3095899522304535,
      "eval_runtime": 2.7374,
      "eval_samples_per_second": 116.898,
      "eval_steps_per_second": 3.653,
      "step": 200
    },
    {
      "epoch": 20.0,
      "step": 200,
      "total_flos": 8.32925255860224e+17,
      "train_loss": 0.5333085978031158,
      "train_runtime": 482.5391,
      "train_samples_per_second": 53.053,
      "train_steps_per_second": 0.414
    }
  ],
  "logging_steps": 10,
  "max_steps": 200,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 20,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 8.32925255860224e+17,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": null
}