{
  "best_metric": 0.8915401301518439,
  "best_model_checkpoint": "swin-tiny-patch4-window7-224-MM_Classification/checkpoint-97",
  "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": 2.865189790725708,
      "learning_rate": 3.125e-05,
      "loss": 1.0635,
      "step": 10
    },
    {
      "epoch": 0.9846153846153847,
      "eval_accuracy": 0.6724511930585684,
      "eval_loss": 0.752371609210968,
      "eval_runtime": 123.4184,
      "eval_samples_per_second": 7.471,
      "eval_steps_per_second": 0.065,
      "step": 16
    },
    {
      "epoch": 1.2307692307692308,
      "grad_norm": 2.9992876052856445,
      "learning_rate": 4.8611111111111115e-05,
      "loss": 0.689,
      "step": 20
    },
    {
      "epoch": 1.8461538461538463,
      "grad_norm": 3.645339250564575,
      "learning_rate": 4.5138888888888894e-05,
      "loss": 0.4571,
      "step": 30
    },
    {
      "epoch": 1.9692307692307693,
      "eval_accuracy": 0.8741865509761388,
      "eval_loss": 0.3691757023334503,
      "eval_runtime": 86.3686,
      "eval_samples_per_second": 10.675,
      "eval_steps_per_second": 0.093,
      "step": 32
    },
    {
      "epoch": 2.4615384615384617,
      "grad_norm": 2.6993319988250732,
      "learning_rate": 4.166666666666667e-05,
      "loss": 0.3819,
      "step": 40
    },
    {
      "epoch": 2.953846153846154,
      "eval_accuracy": 0.868763557483731,
      "eval_loss": 0.34995290637016296,
      "eval_runtime": 86.4691,
      "eval_samples_per_second": 10.663,
      "eval_steps_per_second": 0.093,
      "step": 48
    },
    {
      "epoch": 3.076923076923077,
      "grad_norm": 2.87962007522583,
      "learning_rate": 3.8194444444444444e-05,
      "loss": 0.3314,
      "step": 50
    },
    {
      "epoch": 3.6923076923076925,
      "grad_norm": 1.9204134941101074,
      "learning_rate": 3.472222222222222e-05,
      "loss": 0.3278,
      "step": 60
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8796095444685467,
      "eval_loss": 0.3158332407474518,
      "eval_runtime": 86.6345,
      "eval_samples_per_second": 10.642,
      "eval_steps_per_second": 0.092,
      "step": 65
    },
    {
      "epoch": 4.3076923076923075,
      "grad_norm": 2.5817983150482178,
      "learning_rate": 3.125e-05,
      "loss": 0.3014,
      "step": 70
    },
    {
      "epoch": 4.923076923076923,
      "grad_norm": 4.615570068359375,
      "learning_rate": 2.777777777777778e-05,
      "loss": 0.2941,
      "step": 80
    },
    {
      "epoch": 4.984615384615385,
      "eval_accuracy": 0.8882863340563991,
      "eval_loss": 0.28860825300216675,
      "eval_runtime": 86.3289,
      "eval_samples_per_second": 10.68,
      "eval_steps_per_second": 0.093,
      "step": 81
    },
    {
      "epoch": 5.538461538461538,
      "grad_norm": 2.7304208278656006,
      "learning_rate": 2.4305555555555558e-05,
      "loss": 0.2912,
      "step": 90
    },
    {
      "epoch": 5.969230769230769,
      "eval_accuracy": 0.8915401301518439,
      "eval_loss": 0.2895461618900299,
      "eval_runtime": 86.653,
      "eval_samples_per_second": 10.64,
      "eval_steps_per_second": 0.092,
      "step": 97
    },
    {
      "epoch": 6.153846153846154,
      "grad_norm": 3.1544618606567383,
      "learning_rate": 2.0833333333333336e-05,
      "loss": 0.2859,
      "step": 100
    },
    {
      "epoch": 6.769230769230769,
      "grad_norm": 3.098551034927368,
      "learning_rate": 1.736111111111111e-05,
      "loss": 0.2575,
      "step": 110
    },
    {
      "epoch": 6.953846153846154,
      "eval_accuracy": 0.8839479392624728,
      "eval_loss": 0.2800944447517395,
      "eval_runtime": 86.242,
      "eval_samples_per_second": 10.691,
      "eval_steps_per_second": 0.093,
      "step": 113
    },
    {
      "epoch": 7.384615384615385,
      "grad_norm": 2.2395167350769043,
      "learning_rate": 1.388888888888889e-05,
      "loss": 0.272,
      "step": 120
    },
    {
      "epoch": 8.0,
      "grad_norm": 6.406205177307129,
      "learning_rate": 1.0416666666666668e-05,
      "loss": 0.2604,
      "step": 130
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.886117136659436,
      "eval_loss": 0.28469106554985046,
      "eval_runtime": 86.3518,
      "eval_samples_per_second": 10.677,
      "eval_steps_per_second": 0.093,
      "step": 130
    },
    {
      "epoch": 8.615384615384615,
      "grad_norm": 3.8146533966064453,
      "learning_rate": 6.944444444444445e-06,
      "loss": 0.2519,
      "step": 140
    },
    {
      "epoch": 8.984615384615385,
      "eval_accuracy": 0.8872017353579176,
      "eval_loss": 0.2804240584373474,
      "eval_runtime": 86.6439,
      "eval_samples_per_second": 10.641,
      "eval_steps_per_second": 0.092,
      "step": 146
    },
    {
      "epoch": 9.23076923076923,
      "grad_norm": 2.1230437755584717,
      "learning_rate": 3.4722222222222224e-06,
      "loss": 0.2614,
      "step": 150
    },
    {
      "epoch": 9.846153846153847,
      "grad_norm": 1.9836598634719849,
      "learning_rate": 0.0,
      "loss": 0.2592,
      "step": 160
    },
    {
      "epoch": 9.846153846153847,
      "eval_accuracy": 0.8872017353579176,
      "eval_loss": 0.27946537733078003,
      "eval_runtime": 86.4591,
      "eval_samples_per_second": 10.664,
      "eval_steps_per_second": 0.093,
      "step": 160
    },
    {
      "epoch": 9.846153846153847,
      "step": 160,
      "total_flos": 2.0293244994235208e+18,
      "train_loss": 0.3741051137447357,
      "train_runtime": 7958.3559,
      "train_samples_per_second": 10.415,
      "train_steps_per_second": 0.02
    }
  ],
  "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
}