trainer_state.json

{
  "best_metric": 0.8970666666666667,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_vit-mae-base/model_idx_0905/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": 8.659485808097889e-05,
      "loss": 1.8134,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.7472,
      "eval_loss": 0.8433942794799805,
      "eval_runtime": 6.2628,
      "eval_samples_per_second": 598.772,
      "eval_steps_per_second": 9.421,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 0.28875094652175903,
      "learning_rate": 7.685730736804761e-05,
      "loss": 0.6089,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8221333333333334,
      "eval_loss": 0.5808557271957397,
      "eval_runtime": 6.0032,
      "eval_samples_per_second": 624.666,
      "eval_steps_per_second": 9.828,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 7.927828788757324,
      "learning_rate": 6.226977039554153e-05,
      "loss": 0.368,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8448,
      "eval_loss": 0.5122900605201721,
      "eval_runtime": 6.2485,
      "eval_samples_per_second": 600.142,
      "eval_steps_per_second": 9.442,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 1.1187453269958496,
      "learning_rate": 4.505306743117272e-05,
      "loss": 0.2269,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8584,
      "eval_loss": 0.46336236596107483,
      "eval_runtime": 6.1755,
      "eval_samples_per_second": 607.237,
      "eval_steps_per_second": 9.554,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 0.21356363594532013,
      "learning_rate": 2.7828285431465317e-05,
      "loss": 0.1349,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8856,
      "eval_loss": 0.40472227334976196,
      "eval_runtime": 6.5693,
      "eval_samples_per_second": 570.833,
      "eval_steps_per_second": 8.981,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": Infinity,
      "learning_rate": 1.3255331978432106e-05,
      "loss": 0.0752,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8914666666666666,
      "eval_loss": 0.39476585388183594,
      "eval_runtime": 6.6109,
      "eval_samples_per_second": 567.246,
      "eval_steps_per_second": 8.925,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": 0.7325019836425781,
      "learning_rate": 3.4661526970736e-06,
      "loss": 0.0367,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.8917333333333334,
      "eval_loss": 0.39315691590309143,
      "eval_runtime": 6.05,
      "eval_samples_per_second": 619.831,
      "eval_steps_per_second": 9.752,
      "step": 2331
    },
    {
      "epoch": 8.0,
      "grad_norm": 9.370495796203613,
      "learning_rate": 1.2516232227643975e-10,
      "loss": 0.0259,
      "step": 2664
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.8970666666666667,
      "eval_loss": 0.38344863057136536,
      "eval_runtime": 6.2835,
      "eval_samples_per_second": 596.806,
      "eval_steps_per_second": 9.39,
      "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
}