trainer_state.json

{
  "best_metric": 0.9373333333333334,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/google_vit-base-patch16-224/model_idx_0278/checkpoints/checkpoint-999",
  "epoch": 3.0,
  "eval_steps": 500,
  "global_step": 999,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 0.26252812147140503,
      "learning_rate": 7e-05,
      "loss": 0.6897,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.9328,
      "eval_loss": 0.24310742318630219,
      "eval_runtime": 8.5853,
      "eval_samples_per_second": 436.792,
      "eval_steps_per_second": 6.872,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": Infinity,
      "learning_rate": 7e-05,
      "loss": 0.1347,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.928,
      "eval_loss": 0.2531828284263611,
      "eval_runtime": 7.5085,
      "eval_samples_per_second": 499.433,
      "eval_steps_per_second": 7.858,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 0.0662384033203125,
      "learning_rate": 7e-05,
      "loss": 0.0749,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.9373333333333334,
      "eval_loss": 0.2220987230539322,
      "eval_runtime": 7.4702,
      "eval_samples_per_second": 501.992,
      "eval_steps_per_second": 7.898,
      "step": 999
    }
  ],
  "logging_steps": 500,
  "max_steps": 999,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 4.94223961867776e+18,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}