trainer_state.json

{
  "best_metric": 0.9117333333333333,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/google_vit-base-patch16-224/model_idx_0324/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": 5.812752723693848,
      "learning_rate": 0.0001,
      "loss": 0.6665,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.9024,
      "eval_loss": 0.3203156292438507,
      "eval_runtime": 7.3235,
      "eval_samples_per_second": 512.051,
      "eval_steps_per_second": 8.056,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 0.18169958889484406,
      "learning_rate": 0.0001,
      "loss": 0.1771,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.9098666666666667,
      "eval_loss": 0.29889172315597534,
      "eval_runtime": 6.5456,
      "eval_samples_per_second": 572.907,
      "eval_steps_per_second": 9.014,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 0.632624626159668,
      "learning_rate": 0.0001,
      "loss": 0.1007,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.9117333333333333,
      "eval_loss": 0.3099280595779419,
      "eval_runtime": 6.4574,
      "eval_samples_per_second": 580.732,
      "eval_steps_per_second": 9.137,
      "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
}