trainer_state.json

{
  "best_metric": 0.8946666666666667,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_vit-mae-base/model_idx_0960/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": 4.850126564054637e-05,
      "loss": 1.762,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.7941333333333334,
      "eval_loss": 0.7764983177185059,
      "eval_runtime": 6.457,
      "eval_samples_per_second": 580.763,
      "eval_steps_per_second": 9.137,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 3.712557554244995,
      "learning_rate": 4.416794553834106e-05,
      "loss": 0.5034,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8349333333333333,
      "eval_loss": 0.5390514731407166,
      "eval_runtime": 5.508,
      "eval_samples_per_second": 680.822,
      "eval_steps_per_second": 10.712,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 1.0650310516357422,
      "learning_rate": 3.7522688315464166e-05,
      "loss": 0.2587,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8584,
      "eval_loss": 0.46237435936927795,
      "eval_runtime": 5.3388,
      "eval_samples_per_second": 702.41,
      "eval_steps_per_second": 11.051,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": Infinity,
      "learning_rate": 2.9392810892864702e-05,
      "loss": 0.1405,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8738666666666667,
      "eval_loss": 0.43893587589263916,
      "eval_runtime": 5.2634,
      "eval_samples_per_second": 712.471,
      "eval_steps_per_second": 11.21,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 24.664640426635742,
      "learning_rate": 2.0710421090307246e-05,
      "loss": 0.0674,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8829333333333333,
      "eval_loss": 0.4201619327068329,
      "eval_runtime": 5.3825,
      "eval_samples_per_second": 696.707,
      "eval_steps_per_second": 10.962,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 0.12595321238040924,
      "learning_rate": 1.2545417811701016e-05,
      "loss": 0.0287,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8866666666666667,
      "eval_loss": 0.41986656188964844,
      "eval_runtime": 5.3162,
      "eval_samples_per_second": 705.387,
      "eval_steps_per_second": 11.098,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": 0.1869022250175476,
      "learning_rate": 5.882620955060414e-06,
      "loss": 0.0125,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.8925333333333333,
      "eval_loss": 0.4146372377872467,
      "eval_runtime": 5.5082,
      "eval_samples_per_second": 680.803,
      "eval_steps_per_second": 10.711,
      "step": 2331
    },
    {
      "epoch": 8.0,
      "grad_norm": 0.09499373286962509,
      "learning_rate": 1.5256621537052152e-06,
      "loss": 0.0069,
      "step": 2664
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.8946666666666667,
      "eval_loss": 0.4153720438480377,
      "eval_runtime": 5.4204,
      "eval_samples_per_second": 691.827,
      "eval_steps_per_second": 10.885,
      "step": 2664
    }
  ],
  "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": 1.317930564980736e+19,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}