trainer_state.json

{
  "best_metric": 0.9424,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/google_vit-base-patch16-224/model_idx_0830/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": 24.010496139526367,
      "learning_rate": 5e-05,
      "loss": 0.8816,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.9352,
      "eval_loss": 0.2582147419452667,
      "eval_runtime": 13.1738,
      "eval_samples_per_second": 284.655,
      "eval_steps_per_second": 4.479,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 0.211179718375206,
      "learning_rate": 5e-05,
      "loss": 0.1544,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.9298666666666666,
      "eval_loss": 0.24806633591651917,
      "eval_runtime": 13.2007,
      "eval_samples_per_second": 284.075,
      "eval_steps_per_second": 4.469,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 7.392267227172852,
      "learning_rate": 5e-05,
      "loss": 0.0788,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.9301333333333334,
      "eval_loss": 0.2626202702522278,
      "eval_runtime": 13.0024,
      "eval_samples_per_second": 288.408,
      "eval_steps_per_second": 4.538,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": Infinity,
      "learning_rate": 5e-05,
      "loss": 0.0534,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.9336,
      "eval_loss": 0.25609391927719116,
      "eval_runtime": 13.3482,
      "eval_samples_per_second": 280.938,
      "eval_steps_per_second": 4.42,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 4.595235347747803,
      "learning_rate": 5e-05,
      "loss": 0.0291,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.9354666666666667,
      "eval_loss": 0.28477755188941956,
      "eval_runtime": 13.3908,
      "eval_samples_per_second": 280.044,
      "eval_steps_per_second": 4.406,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 0.2725362479686737,
      "learning_rate": 5e-05,
      "loss": 0.0259,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.9341333333333334,
      "eval_loss": 0.2824524939060211,
      "eval_runtime": 13.2512,
      "eval_samples_per_second": 282.993,
      "eval_steps_per_second": 4.452,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": 0.009900455363094807,
      "learning_rate": 5e-05,
      "loss": 0.0191,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.9424,
      "eval_loss": 0.25854119658470154,
      "eval_runtime": 13.1461,
      "eval_samples_per_second": 285.255,
      "eval_steps_per_second": 4.488,
      "step": 2331
    }
  ],
  "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": 1.153189244358144e+19,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}