trainer_state.json

{
  "best_metric": 0.9330666666666667,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/google_vit-base-patch16-224/model_idx_1001/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": 0.39385589957237244,
      "learning_rate": 0.0001,
      "loss": 0.5953,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.9272,
      "eval_loss": 0.25444772839546204,
      "eval_runtime": 3.9395,
      "eval_samples_per_second": 951.894,
      "eval_steps_per_second": 14.976,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 0.049566857516765594,
      "learning_rate": 0.0001,
      "loss": 0.138,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.9216,
      "eval_loss": 0.2617899477481842,
      "eval_runtime": 3.8696,
      "eval_samples_per_second": 969.087,
      "eval_steps_per_second": 15.247,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 0.0463557131588459,
      "learning_rate": 0.0001,
      "loss": 0.0768,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.9330666666666667,
      "eval_loss": 0.2634746730327606,
      "eval_runtime": 3.9117,
      "eval_samples_per_second": 958.655,
      "eval_steps_per_second": 15.083,
      "step": 999
    }
  ],
  "logging_steps": 500,
  "max_steps": 2997,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 9,
  "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": 4.94223961867776e+18,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}