trainer_state.json

{
  "best_metric": 0.9056,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_vit-mae-base/model_idx_0273/checkpoints/checkpoint-2331",
  "epoch": 7.0,
  "eval_steps": 500,
  "global_step": 2331,
  "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": 9.507764069616767e-05,
      "loss": 1.549,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.7904,
      "eval_loss": 0.7246555089950562,
      "eval_runtime": 11.5615,
      "eval_samples_per_second": 324.353,
      "eval_steps_per_second": 5.103,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": Infinity,
      "learning_rate": 8.127974762931168e-05,
      "loss": 0.5387,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8410666666666666,
      "eval_loss": 0.5558333396911621,
      "eval_runtime": 11.3233,
      "eval_samples_per_second": 331.176,
      "eval_steps_per_second": 5.211,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 0.13733869791030884,
      "learning_rate": 6.125740150220697e-05,
      "loss": 0.319,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.868,
      "eval_loss": 0.4362860321998596,
      "eval_runtime": 11.3271,
      "eval_samples_per_second": 331.065,
      "eval_steps_per_second": 5.209,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 32.674686431884766,
      "learning_rate": 3.900538894462112e-05,
      "loss": 0.1854,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8808,
      "eval_loss": 0.3910314738750458,
      "eval_runtime": 11.8463,
      "eval_samples_per_second": 316.555,
      "eval_steps_per_second": 4.98,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 1.0494736433029175,
      "learning_rate": 1.893099394660879e-05,
      "loss": 0.0932,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8930666666666667,
      "eval_loss": 0.3572295606136322,
      "eval_runtime": 12.0623,
      "eval_samples_per_second": 310.887,
      "eval_steps_per_second": 4.891,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 0.07632418721914291,
      "learning_rate": 5.0101966338242764e-06,
      "loss": 0.0433,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.9005333333333333,
      "eval_loss": 0.3487182557582855,
      "eval_runtime": 11.2337,
      "eval_samples_per_second": 333.816,
      "eval_steps_per_second": 5.252,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": 0.15126509964466095,
      "learning_rate": 1.8164143966159507e-10,
      "loss": 0.0255,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.9056,
      "eval_loss": 0.3378746807575226,
      "eval_runtime": 11.6297,
      "eval_samples_per_second": 322.451,
      "eval_steps_per_second": 5.073,
      "step": 2331
    }
  ],
  "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": true
      },
      "attributes": {}
    }
  },
  "total_flos": 1.153189244358144e+19,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}