{
  "best_metric": 0.8947939262472885,
  "best_model_checkpoint": "swin-tiny-patch4-window7-224-MM_Classification/checkpoint-160",
  "epoch": 9.846153846153847,
  "eval_steps": 500,
  "global_step": 160,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.6153846153846154,
      "grad_norm": 2.122436285018921,
      "learning_rate": 3.125e-05,
      "loss": 1.0041,
      "step": 10
    },
    {
      "epoch": 0.9846153846153847,
      "eval_accuracy": 0.7082429501084598,
      "eval_loss": 0.6398540735244751,
      "eval_runtime": 126.0652,
      "eval_samples_per_second": 7.314,
      "eval_steps_per_second": 0.063,
      "step": 16
    },
    {
      "epoch": 1.2307692307692308,
      "grad_norm": 3.6463255882263184,
      "learning_rate": 4.8611111111111115e-05,
      "loss": 0.6378,
      "step": 20
    },
    {
      "epoch": 1.8461538461538463,
      "grad_norm": 4.6528730392456055,
      "learning_rate": 4.5138888888888894e-05,
      "loss": 0.4441,
      "step": 30
    },
    {
      "epoch": 1.9692307692307693,
      "eval_accuracy": 0.868763557483731,
      "eval_loss": 0.3671414256095886,
      "eval_runtime": 130.7883,
      "eval_samples_per_second": 7.05,
      "eval_steps_per_second": 0.061,
      "step": 32
    },
    {
      "epoch": 2.4615384615384617,
      "grad_norm": 4.03159236907959,
      "learning_rate": 4.166666666666667e-05,
      "loss": 0.3563,
      "step": 40
    },
    {
      "epoch": 2.953846153846154,
      "eval_accuracy": 0.868763557483731,
      "eval_loss": 0.3453761339187622,
      "eval_runtime": 128.7386,
      "eval_samples_per_second": 7.162,
      "eval_steps_per_second": 0.062,
      "step": 48
    },
    {
      "epoch": 3.076923076923077,
      "grad_norm": 3.524843454360962,
      "learning_rate": 3.8194444444444444e-05,
      "loss": 0.3533,
      "step": 50
    },
    {
      "epoch": 3.6923076923076925,
      "grad_norm": 2.8140647411346436,
      "learning_rate": 3.472222222222222e-05,
      "loss": 0.3071,
      "step": 60
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.886117136659436,
      "eval_loss": 0.30997157096862793,
      "eval_runtime": 128.8853,
      "eval_samples_per_second": 7.154,
      "eval_steps_per_second": 0.062,
      "step": 65
    },
    {
      "epoch": 4.3076923076923075,
      "grad_norm": 2.523513078689575,
      "learning_rate": 3.125e-05,
      "loss": 0.3079,
      "step": 70
    },
    {
      "epoch": 4.923076923076923,
      "grad_norm": 2.9005069732666016,
      "learning_rate": 2.777777777777778e-05,
      "loss": 0.2933,
      "step": 80
    },
    {
      "epoch": 4.984615384615385,
      "eval_accuracy": 0.8893709327548807,
      "eval_loss": 0.289992094039917,
      "eval_runtime": 126.7574,
      "eval_samples_per_second": 7.274,
      "eval_steps_per_second": 0.063,
      "step": 81
    },
    {
      "epoch": 5.538461538461538,
      "grad_norm": 1.9166383743286133,
      "learning_rate": 2.4305555555555558e-05,
      "loss": 0.2841,
      "step": 90
    },
    {
      "epoch": 5.969230769230769,
      "eval_accuracy": 0.8828633405639913,
      "eval_loss": 0.29174771904945374,
      "eval_runtime": 130.4143,
      "eval_samples_per_second": 7.07,
      "eval_steps_per_second": 0.061,
      "step": 97
    },
    {
      "epoch": 6.153846153846154,
      "grad_norm": 2.0499794483184814,
      "learning_rate": 2.0833333333333336e-05,
      "loss": 0.2653,
      "step": 100
    },
    {
      "epoch": 6.769230769230769,
      "grad_norm": 2.789874792098999,
      "learning_rate": 1.736111111111111e-05,
      "loss": 0.2715,
      "step": 110
    },
    {
      "epoch": 6.953846153846154,
      "eval_accuracy": 0.8893709327548807,
      "eval_loss": 0.2846406400203705,
      "eval_runtime": 124.3311,
      "eval_samples_per_second": 7.416,
      "eval_steps_per_second": 0.064,
      "step": 113
    },
    {
      "epoch": 7.384615384615385,
      "grad_norm": 2.3379404544830322,
      "learning_rate": 1.388888888888889e-05,
      "loss": 0.2726,
      "step": 120
    },
    {
      "epoch": 8.0,
      "grad_norm": 2.90551495552063,
      "learning_rate": 1.0416666666666668e-05,
      "loss": 0.2564,
      "step": 130
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.8926247288503254,
      "eval_loss": 0.28346362709999084,
      "eval_runtime": 128.5623,
      "eval_samples_per_second": 7.172,
      "eval_steps_per_second": 0.062,
      "step": 130
    },
    {
      "epoch": 8.615384615384615,
      "grad_norm": 2.5919594764709473,
      "learning_rate": 6.944444444444445e-06,
      "loss": 0.2639,
      "step": 140
    },
    {
      "epoch": 8.984615384615385,
      "eval_accuracy": 0.8926247288503254,
      "eval_loss": 0.27986812591552734,
      "eval_runtime": 134.096,
      "eval_samples_per_second": 6.876,
      "eval_steps_per_second": 0.06,
      "step": 146
    },
    {
      "epoch": 9.23076923076923,
      "grad_norm": 2.698216438293457,
      "learning_rate": 3.4722222222222224e-06,
      "loss": 0.264,
      "step": 150
    },
    {
      "epoch": 9.846153846153847,
      "grad_norm": 3.030851125717163,
      "learning_rate": 0.0,
      "loss": 0.2505,
      "step": 160
    },
    {
      "epoch": 9.846153846153847,
      "eval_accuracy": 0.8947939262472885,
      "eval_loss": 0.2758063077926636,
      "eval_runtime": 136.222,
      "eval_samples_per_second": 6.768,
      "eval_steps_per_second": 0.059,
      "step": 160
    },
    {
      "epoch": 9.846153846153847,
      "step": 160,
      "total_flos": 2.0293244994235208e+18,
      "train_loss": 0.3645105704665184,
      "train_runtime": 24595.022,
      "train_samples_per_second": 3.37,
      "train_steps_per_second": 0.007
    }
  ],
  "logging_steps": 10,
  "max_steps": 160,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 10,
  "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": 2.0293244994235208e+18,
  "train_batch_size": 128,
  "trial_name": null,
  "trial_params": null
}