trainer_state.json

{
  "best_metric": 0.9512,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/google_vit-base-patch16-224/model_idx_0174/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": 3e-05,
      "loss": 1.1541,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.9397333333333333,
      "eval_loss": 0.26778319478034973,
      "eval_runtime": 7.2661,
      "eval_samples_per_second": 516.099,
      "eval_steps_per_second": 8.12,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 0.2729506492614746,
      "learning_rate": 3e-05,
      "loss": 0.1483,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.9416,
      "eval_loss": 0.21597477793693542,
      "eval_runtime": 6.6025,
      "eval_samples_per_second": 567.967,
      "eval_steps_per_second": 8.936,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 2.2993080615997314,
      "learning_rate": 3e-05,
      "loss": 0.0547,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.9405333333333333,
      "eval_loss": 0.2101166844367981,
      "eval_runtime": 7.8158,
      "eval_samples_per_second": 479.799,
      "eval_steps_per_second": 7.549,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.2175711691379547,
      "learning_rate": 3e-05,
      "loss": 0.0192,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.9482666666666667,
      "eval_loss": 0.21670879423618317,
      "eval_runtime": 6.4989,
      "eval_samples_per_second": 577.02,
      "eval_steps_per_second": 9.078,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 0.023446686565876007,
      "learning_rate": 3e-05,
      "loss": 0.0088,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.9506666666666667,
      "eval_loss": 0.21623298525810242,
      "eval_runtime": 7.7238,
      "eval_samples_per_second": 485.512,
      "eval_steps_per_second": 7.639,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 0.022313987836241722,
      "learning_rate": 3e-05,
      "loss": 0.0029,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.9504,
      "eval_loss": 0.2268170267343521,
      "eval_runtime": 7.1298,
      "eval_samples_per_second": 525.962,
      "eval_steps_per_second": 8.275,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": 0.020740941166877747,
      "learning_rate": 3e-05,
      "loss": 0.0029,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.9512,
      "eval_loss": 0.22798499464988708,
      "eval_runtime": 8.0137,
      "eval_samples_per_second": 467.949,
      "eval_steps_per_second": 7.362,
      "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
}