trainer_state.json

{
  "best_metric": 0.8514666666666667,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_vit-mae-base/model_idx_0751/checkpoints/checkpoint-1998",
  "epoch": 6.0,
  "eval_steps": 500,
  "global_step": 1998,
  "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": 9e-05,
      "loss": 1.5694,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.7733333333333333,
      "eval_loss": 0.750067949295044,
      "eval_runtime": 5.8002,
      "eval_samples_per_second": 646.533,
      "eval_steps_per_second": 10.172,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 10.435206413269043,
      "learning_rate": 9e-05,
      "loss": 0.5335,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8338666666666666,
      "eval_loss": 0.5442525744438171,
      "eval_runtime": 6.6859,
      "eval_samples_per_second": 560.881,
      "eval_steps_per_second": 8.825,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": Infinity,
      "learning_rate": 9e-05,
      "loss": 0.3261,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8309333333333333,
      "eval_loss": 0.562221109867096,
      "eval_runtime": 6.1933,
      "eval_samples_per_second": 605.494,
      "eval_steps_per_second": 9.526,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 1.791517734527588,
      "learning_rate": 9e-05,
      "loss": 0.224,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8405333333333334,
      "eval_loss": 0.5483905076980591,
      "eval_runtime": 6.3134,
      "eval_samples_per_second": 593.977,
      "eval_steps_per_second": 9.345,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 0.4625927805900574,
      "learning_rate": 9e-05,
      "loss": 0.1685,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8272,
      "eval_loss": 0.5907855033874512,
      "eval_runtime": 5.7882,
      "eval_samples_per_second": 647.87,
      "eval_steps_per_second": 10.193,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 0.039571862667798996,
      "learning_rate": 9e-05,
      "loss": 0.1268,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8514666666666667,
      "eval_loss": 0.5376471281051636,
      "eval_runtime": 5.6418,
      "eval_samples_per_second": 664.681,
      "eval_steps_per_second": 10.458,
      "step": 1998
    }
  ],
  "logging_steps": 500,
  "max_steps": 2997,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 9,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 9.88447923735552e+18,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}