trainer_state.json

{
  "best_metric": 0.5898666666666667,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_vit-mae-base/model_idx_0026/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": 2.9939,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.28986666666666666,
      "eval_loss": 2.5390334129333496,
      "eval_runtime": 7.3055,
      "eval_samples_per_second": 513.314,
      "eval_steps_per_second": 8.076,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": Infinity,
      "learning_rate": 0.00040639873814655834,
      "loss": 2.3008,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.3648,
      "eval_loss": 2.200368642807007,
      "eval_runtime": 6.6464,
      "eval_samples_per_second": 564.219,
      "eval_steps_per_second": 8.877,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 3.0007660388946533,
      "learning_rate": 0.00030628700751103485,
      "loss": 1.887,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.4514666666666667,
      "eval_loss": 1.8945167064666748,
      "eval_runtime": 6.7098,
      "eval_samples_per_second": 558.883,
      "eval_steps_per_second": 8.793,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 19.042726516723633,
      "learning_rate": 0.00019502694472310563,
      "loss": 1.4876,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.5128,
      "eval_loss": 1.6634453535079956,
      "eval_runtime": 6.8208,
      "eval_samples_per_second": 549.788,
      "eval_steps_per_second": 8.65,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 6.574552536010742,
      "learning_rate": 9.465496973304394e-05,
      "loss": 1.0335,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.5501333333333334,
      "eval_loss": 1.5553016662597656,
      "eval_runtime": 6.6461,
      "eval_samples_per_second": 564.237,
      "eval_steps_per_second": 8.877,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 13.29838752746582,
      "learning_rate": 2.505098316912138e-05,
      "loss": 0.5712,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.5850666666666666,
      "eval_loss": 1.4898344278335571,
      "eval_runtime": 6.6383,
      "eval_samples_per_second": 564.905,
      "eval_steps_per_second": 8.888,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": 4.348320960998535,
      "learning_rate": 9.082071983079754e-10,
      "loss": 0.3112,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.5898666666666667,
      "eval_loss": 1.476991891860962,
      "eval_runtime": 6.9751,
      "eval_samples_per_second": 537.627,
      "eval_steps_per_second": 8.459,
      "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
}