trainer_state.json

{
  "best_metric": 0.41173333333333334,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_dino-vitb16/model_idx_0932/checkpoints/checkpoint-2331",
  "epoch": 7.0,
  "eval_steps": 500,
  "global_step": 2331,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": Infinity,
      "learning_rate": 0.0003,
      "loss": 3.2493,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.24373333333333333,
      "eval_loss": 2.7670705318450928,
      "eval_runtime": 11.3087,
      "eval_samples_per_second": 331.602,
      "eval_steps_per_second": 5.217,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 35.53753662109375,
      "learning_rate": 0.0003,
      "loss": 2.6674,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.2688,
      "eval_loss": 2.623633861541748,
      "eval_runtime": 11.2783,
      "eval_samples_per_second": 332.496,
      "eval_steps_per_second": 5.231,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 19.894624710083008,
      "learning_rate": 0.0003,
      "loss": 2.4333,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.32453333333333334,
      "eval_loss": 2.420166254043579,
      "eval_runtime": 10.638,
      "eval_samples_per_second": 352.511,
      "eval_steps_per_second": 5.546,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 35.323062896728516,
      "learning_rate": 0.0003,
      "loss": 2.2536,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.36933333333333335,
      "eval_loss": 2.3188657760620117,
      "eval_runtime": 10.9263,
      "eval_samples_per_second": 343.209,
      "eval_steps_per_second": 5.4,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 14.66744613647461,
      "learning_rate": 0.0003,
      "loss": 2.0793,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.40186666666666665,
      "eval_loss": 2.1680188179016113,
      "eval_runtime": 10.5019,
      "eval_samples_per_second": 357.077,
      "eval_steps_per_second": 5.618,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 16.40332794189453,
      "learning_rate": 0.0003,
      "loss": 1.9283,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.38906666666666667,
      "eval_loss": 2.1872315406799316,
      "eval_runtime": 10.683,
      "eval_samples_per_second": 351.025,
      "eval_steps_per_second": 5.523,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": 23.722400665283203,
      "learning_rate": 0.0003,
      "loss": 1.7799,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.41173333333333334,
      "eval_loss": 2.0766520500183105,
      "eval_runtime": 10.6754,
      "eval_samples_per_second": 351.275,
      "eval_steps_per_second": 5.527,
      "step": 2331
    }
  ],
  "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": true
      },
      "attributes": {}
    }
  },
  "total_flos": 1.153189244358144e+19,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}