trainer_state.json

{
  "best_metric": 0.8221333333333334,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/google_vit-base-patch16-224/model_idx_0856/checkpoints/checkpoint-1998",
  "epoch": 6.0,
  "eval_steps": 500,
  "global_step": 1998,
  "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": 0.0005,
      "loss": 0.8583,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8109333333333333,
      "eval_loss": 0.6328275203704834,
      "eval_runtime": 11.2175,
      "eval_samples_per_second": 334.3,
      "eval_steps_per_second": 5.26,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 4.7194743156433105,
      "learning_rate": 0.0005,
      "loss": 0.5054,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8096,
      "eval_loss": 0.6719980239868164,
      "eval_runtime": 10.821,
      "eval_samples_per_second": 346.547,
      "eval_steps_per_second": 5.452,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 17.18718147277832,
      "learning_rate": 0.0005,
      "loss": 0.4214,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8082666666666667,
      "eval_loss": 0.6545153260231018,
      "eval_runtime": 11.6534,
      "eval_samples_per_second": 321.795,
      "eval_steps_per_second": 5.063,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 1.9554195404052734,
      "learning_rate": 0.0005,
      "loss": 0.3439,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8018666666666666,
      "eval_loss": 0.7012310028076172,
      "eval_runtime": 10.8423,
      "eval_samples_per_second": 345.868,
      "eval_steps_per_second": 5.442,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 0.05399654433131218,
      "learning_rate": 0.0005,
      "loss": 0.2953,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8170666666666667,
      "eval_loss": 0.6724411249160767,
      "eval_runtime": 10.8492,
      "eval_samples_per_second": 345.646,
      "eval_steps_per_second": 5.438,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 7.107763290405273,
      "learning_rate": 0.0005,
      "loss": 0.2684,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8221333333333334,
      "eval_loss": 0.678329348564148,
      "eval_runtime": 11.4226,
      "eval_samples_per_second": 328.297,
      "eval_steps_per_second": 5.165,
      "step": 1998
    }
  ],
  "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": false
      },
      "attributes": {}
    }
  },
  "total_flos": 9.88447923735552e+18,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}