trainer_state.json

{
  "best_metric": 0.9306666666666666,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/google_vit-base-patch16-224/model_idx_0051/checkpoints/checkpoint-666",
  "epoch": 2.0,
  "eval_steps": 500,
  "global_step": 666,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": NaN,
      "learning_rate": 9e-05,
      "loss": 0.6519,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.9192,
      "eval_loss": 0.2829256057739258,
      "eval_runtime": 12.7139,
      "eval_samples_per_second": 294.953,
      "eval_steps_per_second": 4.641,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 0.04599888622760773,
      "learning_rate": 9e-05,
      "loss": 0.1362,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.9306666666666666,
      "eval_loss": 0.24789191782474518,
      "eval_runtime": 13.6245,
      "eval_samples_per_second": 275.239,
      "eval_steps_per_second": 4.33,
      "step": 666
    }
  ],
  "logging_steps": 500,
  "max_steps": 2331,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 7,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 3.29482641245184e+18,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}