{
  "best_metric": 0.9139414802065404,
  "best_model_checkpoint": "swin-tiny-patch4-window7-224-finetuned-azure-poc-img-classification/checkpoint-123",
  "epoch": 5.0,
  "eval_steps": 500,
  "global_step": 205,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.24390243902439024,
      "grad_norm": 69.68915557861328,
      "learning_rate": 2.380952380952381e-05,
      "loss": 1.5545,
      "step": 10
    },
    {
      "epoch": 0.4878048780487805,
      "grad_norm": 46.480247497558594,
      "learning_rate": 4.761904761904762e-05,
      "loss": 0.8946,
      "step": 20
    },
    {
      "epoch": 0.7317073170731707,
      "grad_norm": 28.4664363861084,
      "learning_rate": 4.7554347826086956e-05,
      "loss": 0.5093,
      "step": 30
    },
    {
      "epoch": 0.975609756097561,
      "grad_norm": 26.042448043823242,
      "learning_rate": 4.483695652173913e-05,
      "loss": 0.4227,
      "step": 40
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8898450946643718,
      "eval_loss": 0.3245888352394104,
      "eval_runtime": 28.5684,
      "eval_samples_per_second": 20.337,
      "eval_steps_per_second": 0.665,
      "step": 41
    },
    {
      "epoch": 1.2195121951219512,
      "grad_norm": 30.95467758178711,
      "learning_rate": 4.2119565217391304e-05,
      "loss": 0.3611,
      "step": 50
    },
    {
      "epoch": 1.4634146341463414,
      "grad_norm": 15.45040225982666,
      "learning_rate": 3.940217391304348e-05,
      "loss": 0.3267,
      "step": 60
    },
    {
      "epoch": 1.7073170731707317,
      "grad_norm": 17.348590850830078,
      "learning_rate": 3.668478260869566e-05,
      "loss": 0.3091,
      "step": 70
    },
    {
      "epoch": 1.951219512195122,
      "grad_norm": 17.025691986083984,
      "learning_rate": 3.3967391304347826e-05,
      "loss": 0.3169,
      "step": 80
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8950086058519794,
      "eval_loss": 0.27360019087791443,
      "eval_runtime": 25.8613,
      "eval_samples_per_second": 22.466,
      "eval_steps_per_second": 0.735,
      "step": 82
    },
    {
      "epoch": 2.1951219512195124,
      "grad_norm": 14.608966827392578,
      "learning_rate": 3.125e-05,
      "loss": 0.2459,
      "step": 90
    },
    {
      "epoch": 2.4390243902439024,
      "grad_norm": 15.097740173339844,
      "learning_rate": 2.8532608695652175e-05,
      "loss": 0.2455,
      "step": 100
    },
    {
      "epoch": 2.682926829268293,
      "grad_norm": 16.881877899169922,
      "learning_rate": 2.5815217391304346e-05,
      "loss": 0.2439,
      "step": 110
    },
    {
      "epoch": 2.926829268292683,
      "grad_norm": 17.772117614746094,
      "learning_rate": 2.3097826086956523e-05,
      "loss": 0.2632,
      "step": 120
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.9139414802065404,
      "eval_loss": 0.2521795928478241,
      "eval_runtime": 27.019,
      "eval_samples_per_second": 21.503,
      "eval_steps_per_second": 0.703,
      "step": 123
    },
    {
      "epoch": 3.1707317073170733,
      "grad_norm": 16.634872436523438,
      "learning_rate": 2.0380434782608694e-05,
      "loss": 0.2085,
      "step": 130
    },
    {
      "epoch": 3.4146341463414633,
      "grad_norm": 25.425128936767578,
      "learning_rate": 1.766304347826087e-05,
      "loss": 0.2293,
      "step": 140
    },
    {
      "epoch": 3.658536585365854,
      "grad_norm": 9.363799095153809,
      "learning_rate": 1.4945652173913044e-05,
      "loss": 0.2041,
      "step": 150
    },
    {
      "epoch": 3.902439024390244,
      "grad_norm": 8.764758110046387,
      "learning_rate": 1.2228260869565218e-05,
      "loss": 0.2034,
      "step": 160
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8984509466437177,
      "eval_loss": 0.24487625062465668,
      "eval_runtime": 25.7575,
      "eval_samples_per_second": 22.557,
      "eval_steps_per_second": 0.738,
      "step": 164
    },
    {
      "epoch": 4.146341463414634,
      "grad_norm": 11.34645938873291,
      "learning_rate": 9.510869565217392e-06,
      "loss": 0.224,
      "step": 170
    },
    {
      "epoch": 4.390243902439025,
      "grad_norm": 16.645652770996094,
      "learning_rate": 6.7934782608695655e-06,
      "loss": 0.1933,
      "step": 180
    },
    {
      "epoch": 4.634146341463414,
      "grad_norm": 12.6370267868042,
      "learning_rate": 4.07608695652174e-06,
      "loss": 0.1932,
      "step": 190
    },
    {
      "epoch": 4.878048780487805,
      "grad_norm": 20.207782745361328,
      "learning_rate": 1.3586956521739131e-06,
      "loss": 0.1984,
      "step": 200
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8984509466437177,
      "eval_loss": 0.24636875092983246,
      "eval_runtime": 27.2859,
      "eval_samples_per_second": 21.293,
      "eval_steps_per_second": 0.696,
      "step": 205
    },
    {
      "epoch": 5.0,
      "step": 205,
      "total_flos": 6.487956947146752e+17,
      "train_loss": 0.36339920555672994,
      "train_runtime": 1344.0908,
      "train_samples_per_second": 19.418,
      "train_steps_per_second": 0.153
    }
  ],
  "logging_steps": 10,
  "max_steps": 205,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 5,
  "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": 6.487956947146752e+17,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": null
}