trainer_state.json

{
  "best_metric": 0.6160714285714286,
  "best_model_checkpoint": "convnext-tiny-224-jvadlamudi2/checkpoint-24",
  "epoch": 3.0,
  "global_step": 24,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "eval_accuracy": 0.5535714285714286,
      "eval_loss": 0.6731433272361755,
      "eval_runtime": 0.8011,
      "eval_samples_per_second": 139.814,
      "eval_steps_per_second": 4.993,
      "step": 8
    },
    {
      "epoch": 1.25,
      "learning_rate": 3.3333333333333335e-05,
      "loss": 0.6901,
      "step": 10
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.5982142857142857,
      "eval_loss": 0.6700941920280457,
      "eval_runtime": 0.9327,
      "eval_samples_per_second": 120.086,
      "eval_steps_per_second": 4.289,
      "step": 16
    },
    {
      "epoch": 2.5,
      "learning_rate": 9.523809523809523e-06,
      "loss": 0.6819,
      "step": 20
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.6160714285714286,
      "eval_loss": 0.6697055697441101,
      "eval_runtime": 0.846,
      "eval_samples_per_second": 132.395,
      "eval_steps_per_second": 4.728,
      "step": 24
    },
    {
      "epoch": 3.0,
      "step": 24,
      "total_flos": 7.553368114429133e+16,
      "train_loss": 0.6847228904565176,
      "train_runtime": 88.414,
      "train_samples_per_second": 33.999,
      "train_steps_per_second": 0.271
    }
  ],
  "max_steps": 24,
  "num_train_epochs": 3,
  "total_flos": 7.553368114429133e+16,
  "trial_name": null,
  "trial_params": null
}