trainer_state.json

{
  "best_metric": 0.8709333333333333,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_vit-mae-base/model_idx_0945/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": 7e-05,
      "loss": 1.5764,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8157333333333333,
      "eval_loss": 0.6605719923973083,
      "eval_runtime": 7.0492,
      "eval_samples_per_second": 531.977,
      "eval_steps_per_second": 8.37,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 20.512264251708984,
      "learning_rate": 7e-05,
      "loss": 0.5116,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8330666666666666,
      "eval_loss": 0.5511621236801147,
      "eval_runtime": 7.2681,
      "eval_samples_per_second": 515.953,
      "eval_steps_per_second": 8.118,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 3.595431327819824,
      "learning_rate": 7e-05,
      "loss": 0.3107,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8642666666666666,
      "eval_loss": 0.46801769733428955,
      "eval_runtime": 7.2536,
      "eval_samples_per_second": 516.988,
      "eval_steps_per_second": 8.134,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.6758065819740295,
      "learning_rate": 7e-05,
      "loss": 0.2127,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8629333333333333,
      "eval_loss": 0.48042017221450806,
      "eval_runtime": 6.9868,
      "eval_samples_per_second": 536.723,
      "eval_steps_per_second": 8.444,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 1.6564340591430664,
      "learning_rate": 7e-05,
      "loss": 0.1482,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8666666666666667,
      "eval_loss": 0.4771919250488281,
      "eval_runtime": 6.9082,
      "eval_samples_per_second": 542.83,
      "eval_steps_per_second": 8.541,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 3.857462167739868,
      "learning_rate": 7e-05,
      "loss": 0.1168,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8669333333333333,
      "eval_loss": 0.5012549757957458,
      "eval_runtime": 7.9015,
      "eval_samples_per_second": 474.592,
      "eval_steps_per_second": 7.467,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": Infinity,
      "learning_rate": 7e-05,
      "loss": 0.0972,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.8709333333333333,
      "eval_loss": 0.47507956624031067,
      "eval_runtime": 7.5761,
      "eval_samples_per_second": 494.976,
      "eval_steps_per_second": 7.788,
      "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
}