trainer_state.json

{
  "best_metric": 0.532,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_vit-mae-base/model_idx_0233/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": 0.00047538820348083834,
      "loss": 3.2948,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.20746666666666666,
      "eval_loss": 2.990375518798828,
      "eval_runtime": 13.201,
      "eval_samples_per_second": 284.069,
      "eval_steps_per_second": 4.469,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 8.396687507629395,
      "learning_rate": 0.00040613573610928476,
      "loss": 2.7461,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.30106666666666665,
      "eval_loss": 2.60339093208313,
      "eval_runtime": 13.3496,
      "eval_samples_per_second": 280.907,
      "eval_steps_per_second": 4.42,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": Infinity,
      "learning_rate": 0.00030628700751103485,
      "loss": 2.2986,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.38453333333333334,
      "eval_loss": 2.2876806259155273,
      "eval_runtime": 13.1194,
      "eval_samples_per_second": 285.836,
      "eval_steps_per_second": 4.497,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": Infinity,
      "learning_rate": 0.000195355683908512,
      "loss": 1.861,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.45386666666666664,
      "eval_loss": 1.9824025630950928,
      "eval_runtime": 13.1827,
      "eval_samples_per_second": 284.463,
      "eval_steps_per_second": 4.476,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 8.16332721710205,
      "learning_rate": 9.491910332815573e-05,
      "loss": 1.3994,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.5024,
      "eval_loss": 1.8293713331222534,
      "eval_runtime": 13.1242,
      "eval_samples_per_second": 285.732,
      "eval_steps_per_second": 4.496,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 18.197389602661133,
      "learning_rate": 2.519819627604078e-05,
      "loss": 0.9489,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.5304,
      "eval_loss": 1.7412528991699219,
      "eval_runtime": 13.1276,
      "eval_samples_per_second": 285.658,
      "eval_steps_per_second": 4.494,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": 16.149574279785156,
      "learning_rate": 2.0434646496314548e-09,
      "loss": 0.6784,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.532,
      "eval_loss": 1.720076084136963,
      "eval_runtime": 13.2106,
      "eval_samples_per_second": 283.863,
      "eval_steps_per_second": 4.466,
      "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
}