trainer_state.json

{
  "best_metric": 0.9229333333333334,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/google_vit-base-patch16-224/model_idx_0688/checkpoints/checkpoint-2664",
  "epoch": 8.0,
  "eval_steps": 500,
  "global_step": 2664,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 0.09430141001939774,
      "learning_rate": 0.0004375,
      "loss": 0.8219,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.7722666666666667,
      "eval_loss": 0.8184120059013367,
      "eval_runtime": 13.5064,
      "eval_samples_per_second": 277.645,
      "eval_steps_per_second": 4.368,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": Infinity,
      "learning_rate": 0.0003751876876876877,
      "loss": 0.4268,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8402666666666667,
      "eval_loss": 0.5332589149475098,
      "eval_runtime": 13.4099,
      "eval_samples_per_second": 279.644,
      "eval_steps_per_second": 4.4,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 21.40583038330078,
      "learning_rate": 0.0003126876876876877,
      "loss": 0.2679,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8618666666666667,
      "eval_loss": 0.48754042387008667,
      "eval_runtime": 13.3423,
      "eval_samples_per_second": 281.061,
      "eval_steps_per_second": 4.422,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.008856230415403843,
      "learning_rate": 0.0002501876876876877,
      "loss": 0.1668,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8728,
      "eval_loss": 0.4770253896713257,
      "eval_runtime": 13.4212,
      "eval_samples_per_second": 279.409,
      "eval_steps_per_second": 4.396,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 0.06272929906845093,
      "learning_rate": 0.00018768768768768769,
      "loss": 0.1022,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8856,
      "eval_loss": 0.43388810753822327,
      "eval_runtime": 13.3939,
      "eval_samples_per_second": 279.978,
      "eval_steps_per_second": 4.405,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 0.018051713705062866,
      "learning_rate": 0.00012518768768768768,
      "loss": 0.0475,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.9061333333333333,
      "eval_loss": 0.3840394616127014,
      "eval_runtime": 13.4304,
      "eval_samples_per_second": 279.218,
      "eval_steps_per_second": 4.393,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": 0.003932427614927292,
      "learning_rate": 6.26876876876877e-05,
      "loss": 0.0154,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.9162666666666667,
      "eval_loss": 0.3513264060020447,
      "eval_runtime": 13.566,
      "eval_samples_per_second": 276.427,
      "eval_steps_per_second": 4.349,
      "step": 2331
    },
    {
      "epoch": 8.0,
      "grad_norm": 19.282190322875977,
      "learning_rate": 1.876876876876877e-07,
      "loss": 0.0088,
      "step": 2664
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.9229333333333334,
      "eval_loss": 0.3362186849117279,
      "eval_runtime": 13.7802,
      "eval_samples_per_second": 272.129,
      "eval_steps_per_second": 4.281,
      "step": 2664
    }
  ],
  "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": true
      },
      "attributes": {}
    }
  },
  "total_flos": 1.317930564980736e+19,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}