trainer_state.json

{
  "best_metric": 0.75,
  "best_model_checkpoint": "convnext-tiny-224-jvadlamudi2/checkpoint-21",
  "epoch": 3.0,
  "global_step": 21,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "eval_accuracy": 0.6964285714285714,
      "eval_loss": 0.6321507692337036,
      "eval_runtime": 0.9364,
      "eval_samples_per_second": 119.604,
      "eval_steps_per_second": 4.272,
      "step": 7
    },
    {
      "epoch": 1.43,
      "learning_rate": 3.055555555555556e-05,
      "loss": 0.6503,
      "step": 10
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.7142857142857143,
      "eval_loss": 0.6245036721229553,
      "eval_runtime": 0.788,
      "eval_samples_per_second": 142.126,
      "eval_steps_per_second": 5.076,
      "step": 14
    },
    {
      "epoch": 2.86,
      "learning_rate": 2.777777777777778e-06,
      "loss": 0.6372,
      "step": 20
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.75,
      "eval_loss": 0.6208862662315369,
      "eval_runtime": 0.7892,
      "eval_samples_per_second": 141.913,
      "eval_steps_per_second": 5.068,
      "step": 21
    },
    {
      "epoch": 3.0,
      "step": 21,
      "total_flos": 6.716617754447462e+16,
      "train_loss": 0.642929474512736,
      "train_runtime": 92.7694,
      "train_samples_per_second": 28.813,
      "train_steps_per_second": 0.226
    }
  ],
  "max_steps": 21,
  "num_train_epochs": 3,
  "total_flos": 6.716617754447462e+16,
  "trial_name": null,
  "trial_params": null
}