trainer_state.json

{
  "best_metric": 0.015230490826070309,
  "best_model_checkpoint": "./vit-airplanes/checkpoint-100",
  "epoch": 4.0,
  "global_step": 168,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.24,
      "learning_rate": 0.0001880952380952381,
      "loss": 0.3543,
      "step": 10
    },
    {
      "epoch": 0.48,
      "learning_rate": 0.0001761904761904762,
      "loss": 0.1221,
      "step": 20
    },
    {
      "epoch": 0.71,
      "learning_rate": 0.00016428571428571428,
      "loss": 0.0587,
      "step": 30
    },
    {
      "epoch": 0.95,
      "learning_rate": 0.00015238095238095237,
      "loss": 0.0441,
      "step": 40
    },
    {
      "epoch": 1.19,
      "learning_rate": 0.00014047619047619049,
      "loss": 0.0281,
      "step": 50
    },
    {
      "epoch": 1.43,
      "learning_rate": 0.00012857142857142858,
      "loss": 0.0302,
      "step": 60
    },
    {
      "epoch": 1.67,
      "learning_rate": 0.00011666666666666668,
      "loss": 0.0417,
      "step": 70
    },
    {
      "epoch": 1.9,
      "learning_rate": 0.00010476190476190477,
      "loss": 0.0192,
      "step": 80
    },
    {
      "epoch": 2.14,
      "learning_rate": 9.285714285714286e-05,
      "loss": 0.0129,
      "step": 90
    },
    {
      "epoch": 2.38,
      "learning_rate": 8.095238095238096e-05,
      "loss": 0.0165,
      "step": 100
    },
    {
      "epoch": 2.38,
      "eval_accuracy": 1.0,
      "eval_loss": 0.015230490826070309,
      "eval_runtime": 56.286,
      "eval_samples_per_second": 11.85,
      "eval_steps_per_second": 1.492,
      "step": 100
    },
    {
      "epoch": 2.62,
      "learning_rate": 6.904761904761905e-05,
      "loss": 0.0184,
      "step": 110
    },
    {
      "epoch": 2.86,
      "learning_rate": 5.714285714285714e-05,
      "loss": 0.0117,
      "step": 120
    },
    {
      "epoch": 3.1,
      "learning_rate": 4.523809523809524e-05,
      "loss": 0.0128,
      "step": 130
    },
    {
      "epoch": 3.33,
      "learning_rate": 3.3333333333333335e-05,
      "loss": 0.0132,
      "step": 140
    },
    {
      "epoch": 3.57,
      "learning_rate": 2.1428571428571428e-05,
      "loss": 0.0099,
      "step": 150
    },
    {
      "epoch": 3.81,
      "learning_rate": 9.523809523809523e-06,
      "loss": 0.0082,
      "step": 160
    },
    {
      "epoch": 4.0,
      "step": 168,
      "total_flos": 2.067504813142917e+17,
      "train_loss": 0.04814572525875909,
      "train_runtime": 540.2683,
      "train_samples_per_second": 4.938,
      "train_steps_per_second": 0.311
    }
  ],
  "max_steps": 168,
  "num_train_epochs": 4,
  "total_flos": 2.067504813142917e+17,
  "trial_name": null,
  "trial_params": null
}