trainer_state.json

{
  "best_metric": 0.5442666666666667,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_vit-mae-base/model_idx_0603/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": 0.0004287859287859288,
      "loss": 3.1461,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.232,
      "eval_loss": 2.831552267074585,
      "eval_runtime": 5.2521,
      "eval_samples_per_second": 713.999,
      "eval_steps_per_second": 11.234,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": Infinity,
      "learning_rate": 0.00035757185757185754,
      "loss": 2.5387,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.3429333333333333,
      "eval_loss": 2.3665196895599365,
      "eval_runtime": 5.7067,
      "eval_samples_per_second": 657.118,
      "eval_steps_per_second": 10.339,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 11.231011390686035,
      "learning_rate": 0.00028614328614328614,
      "loss": 2.1037,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.4109333333333333,
      "eval_loss": 2.0938985347747803,
      "eval_runtime": 6.1569,
      "eval_samples_per_second": 609.077,
      "eval_steps_per_second": 9.583,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 9.619653701782227,
      "learning_rate": 0.0002147147147147147,
      "loss": 1.7175,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.46986666666666665,
      "eval_loss": 1.8772075176239014,
      "eval_runtime": 6.2577,
      "eval_samples_per_second": 599.257,
      "eval_steps_per_second": 9.428,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 17.803062438964844,
      "learning_rate": 0.0001432861432861433,
      "loss": 1.2834,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.5098666666666667,
      "eval_loss": 1.7589306831359863,
      "eval_runtime": 5.2205,
      "eval_samples_per_second": 718.328,
      "eval_steps_per_second": 11.302,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 10.03609561920166,
      "learning_rate": 7.185757185757186e-05,
      "loss": 0.7666,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.5296,
      "eval_loss": 1.750725269317627,
      "eval_runtime": 5.2201,
      "eval_samples_per_second": 718.373,
      "eval_steps_per_second": 11.302,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": 6.372138023376465,
      "learning_rate": 4.2900042900042903e-07,
      "loss": 0.2899,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.5442666666666667,
      "eval_loss": 1.7569172382354736,
      "eval_runtime": 5.4353,
      "eval_samples_per_second": 689.933,
      "eval_steps_per_second": 10.855,
      "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
}