trainer_state.json

{
  "best_metric": 0.8725333333333334,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/google_vit-base-patch16-224/model_idx_0089/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": 10.149615287780762,
      "learning_rate": 0.0003,
      "loss": 0.6245,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8426666666666667,
      "eval_loss": 0.5129416584968567,
      "eval_runtime": 11.0823,
      "eval_samples_per_second": 338.377,
      "eval_steps_per_second": 5.324,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 15.719572067260742,
      "learning_rate": 0.0003,
      "loss": 0.3368,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8554666666666667,
      "eval_loss": 0.47429004311561584,
      "eval_runtime": 11.1239,
      "eval_samples_per_second": 337.113,
      "eval_steps_per_second": 5.304,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 3.3151206970214844,
      "learning_rate": 0.0003,
      "loss": 0.2452,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8674666666666667,
      "eval_loss": 0.44912728667259216,
      "eval_runtime": 11.3237,
      "eval_samples_per_second": 331.164,
      "eval_steps_per_second": 5.21,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.06811171770095825,
      "learning_rate": 0.0003,
      "loss": 0.1955,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8648,
      "eval_loss": 0.491460382938385,
      "eval_runtime": 11.3257,
      "eval_samples_per_second": 331.105,
      "eval_steps_per_second": 5.209,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": Infinity,
      "learning_rate": 0.0003,
      "loss": 0.1715,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8658666666666667,
      "eval_loss": 0.4994823932647705,
      "eval_runtime": 12.0497,
      "eval_samples_per_second": 311.212,
      "eval_steps_per_second": 4.896,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 0.25336286425590515,
      "learning_rate": 0.0003,
      "loss": 0.1463,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8725333333333334,
      "eval_loss": 0.4674731194972992,
      "eval_runtime": 11.6146,
      "eval_samples_per_second": 322.87,
      "eval_steps_per_second": 5.08,
      "step": 1998
    }
  ],
  "logging_steps": 500,
  "max_steps": 2664,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 8,
  "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
}