trainer_state.json

{
  "best_metric": 0.9664039039039038,
  "best_model_checkpoint": "fine-tune-DeiT-gtsrb\\checkpoint-1665",
  "epoch": 5.0,
  "global_step": 1665,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8986486486486487,
      "eval_loss": 0.3261682987213135,
      "eval_runtime": 78.9998,
      "eval_samples_per_second": 67.443,
      "eval_steps_per_second": 1.063,
      "step": 333
    },
    {
      "epoch": 1.5,
      "learning_rate": 0.00013993993993993996,
      "loss": 0.4929,
      "step": 500
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.9283033033033034,
      "eval_loss": 0.21911460161209106,
      "eval_runtime": 72.9813,
      "eval_samples_per_second": 73.005,
      "eval_steps_per_second": 1.151,
      "step": 666
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.946509009009009,
      "eval_loss": 0.16831828653812408,
      "eval_runtime": 31.3847,
      "eval_samples_per_second": 169.765,
      "eval_steps_per_second": 2.676,
      "step": 999
    },
    {
      "epoch": 3.0,
      "learning_rate": 7.987987987987988e-05,
      "loss": 0.2085,
      "step": 1000
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.9513888888888888,
      "eval_loss": 0.1475089192390442,
      "eval_runtime": 79.8191,
      "eval_samples_per_second": 66.751,
      "eval_steps_per_second": 1.052,
      "step": 1332
    },
    {
      "epoch": 4.5,
      "learning_rate": 1.981981981981982e-05,
      "loss": 0.1434,
      "step": 1500
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.9664039039039038,
      "eval_loss": 0.11026506125926971,
      "eval_runtime": 77.8498,
      "eval_samples_per_second": 68.439,
      "eval_steps_per_second": 1.079,
      "step": 1665
    }
  ],
  "max_steps": 1665,
  "num_train_epochs": 5,
  "total_flos": 8.260580814172324e+18,
  "trial_name": null,
  "trial_params": null
}