trainer_state.json

{
  "best_metric": 0.928,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_vit-mae-base/model_idx_0829/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": Infinity,
      "learning_rate": 2.886495269365963e-05,
      "loss": 2.0825,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.832,
      "eval_loss": 0.8521221280097961,
      "eval_runtime": 10.4812,
      "eval_samples_per_second": 357.783,
      "eval_steps_per_second": 5.629,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": Infinity,
      "learning_rate": 2.5631588426322576e-05,
      "loss": 0.563,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8813333333333333,
      "eval_loss": 0.4725189507007599,
      "eval_runtime": 11.2908,
      "eval_samples_per_second": 332.13,
      "eval_steps_per_second": 5.226,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 4.604273796081543,
      "learning_rate": 2.0772920772568672e-05,
      "loss": 0.284,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8997333333333334,
      "eval_loss": 0.37761715054512024,
      "eval_runtime": 11.7329,
      "eval_samples_per_second": 319.614,
      "eval_steps_per_second": 5.029,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.7053096294403076,
      "learning_rate": 1.5035378262848308e-05,
      "loss": 0.1615,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.9152,
      "eval_loss": 0.31736618280410767,
      "eval_runtime": 10.5366,
      "eval_samples_per_second": 355.903,
      "eval_steps_per_second": 5.6,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 1.6922423839569092,
      "learning_rate": 9.292449733314041e-06,
      "loss": 0.0916,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.9208,
      "eval_loss": 0.29007476568222046,
      "eval_runtime": 10.4675,
      "eval_samples_per_second": 358.252,
      "eval_steps_per_second": 5.637,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 0.41555315256118774,
      "learning_rate": 4.418443992810701e-06,
      "loss": 0.0503,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.9205333333333333,
      "eval_loss": 0.2815112769603729,
      "eval_runtime": 11.6171,
      "eval_samples_per_second": 322.799,
      "eval_steps_per_second": 5.079,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": 0.39773860573768616,
      "learning_rate": 1.1553842323578667e-06,
      "loss": 0.0326,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.9256,
      "eval_loss": 0.27255481481552124,
      "eval_runtime": 10.5903,
      "eval_samples_per_second": 354.098,
      "eval_steps_per_second": 5.571,
      "step": 2331
    },
    {
      "epoch": 8.0,
      "grad_norm": 0.6393812894821167,
      "learning_rate": 4.172077409214658e-11,
      "loss": 0.0256,
      "step": 2664
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.928,
      "eval_loss": 0.26778700947761536,
      "eval_runtime": 10.3715,
      "eval_samples_per_second": 361.568,
      "eval_steps_per_second": 5.689,
      "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
}