trainer_state.json

{
  "best_metric": 0.6866666666666666,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_vit-mae-base/model_idx_0856/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": Infinity,
      "learning_rate": 0.00047538820348083834,
      "loss": 3.041,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.3144,
      "eval_loss": 2.565349578857422,
      "eval_runtime": 5.8332,
      "eval_samples_per_second": 642.872,
      "eval_steps_per_second": 10.115,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": Infinity,
      "learning_rate": 0.00040639873814655834,
      "loss": 2.1572,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.4688,
      "eval_loss": 1.9179697036743164,
      "eval_runtime": 6.3476,
      "eval_samples_per_second": 590.777,
      "eval_steps_per_second": 9.295,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 16.21799087524414,
      "learning_rate": 0.00030628700751103485,
      "loss": 1.5368,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.5629333333333333,
      "eval_loss": 1.5169215202331543,
      "eval_runtime": 5.9114,
      "eval_samples_per_second": 634.368,
      "eval_steps_per_second": 9.981,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": Infinity,
      "learning_rate": 0.000195355683908512,
      "loss": 1.0273,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.6309333333333333,
      "eval_loss": 1.293600082397461,
      "eval_runtime": 5.8262,
      "eval_samples_per_second": 643.639,
      "eval_steps_per_second": 10.127,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 9.02326774597168,
      "learning_rate": 9.491910332815573e-05,
      "loss": 0.5093,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.6616,
      "eval_loss": 1.2050461769104004,
      "eval_runtime": 5.891,
      "eval_samples_per_second": 636.566,
      "eval_steps_per_second": 10.015,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 0.6700507998466492,
      "learning_rate": 2.519819627604078e-05,
      "loss": 0.1534,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.6866666666666666,
      "eval_loss": 1.199965000152588,
      "eval_runtime": 6.3353,
      "eval_samples_per_second": 591.919,
      "eval_steps_per_second": 9.313,
      "step": 1998
    }
  ],
  "logging_steps": 500,
  "max_steps": 2331,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 7,
  "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
}