trainer_state.json

{
  "best_metric": 0.8538666666666667,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_vit-mae-base/model_idx_0574/checkpoints/checkpoint-1665",
  "epoch": 5.0,
  "eval_steps": 500,
  "global_step": 1665,
  "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": 1.7611,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.7394666666666667,
      "eval_loss": 0.8836338520050049,
      "eval_runtime": 12.062,
      "eval_samples_per_second": 310.894,
      "eval_steps_per_second": 4.891,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 1.3581434488296509,
      "learning_rate": 9e-05,
      "loss": 0.5576,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8162666666666667,
      "eval_loss": 0.6368408799171448,
      "eval_runtime": 12.4786,
      "eval_samples_per_second": 300.515,
      "eval_steps_per_second": 4.728,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 0.7468198537826538,
      "learning_rate": 9e-05,
      "loss": 0.3268,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8464,
      "eval_loss": 0.5222966074943542,
      "eval_runtime": 12.2351,
      "eval_samples_per_second": 306.494,
      "eval_steps_per_second": 4.822,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.08254683017730713,
      "learning_rate": 9e-05,
      "loss": 0.2139,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8432,
      "eval_loss": 0.5487236976623535,
      "eval_runtime": 12.6177,
      "eval_samples_per_second": 297.203,
      "eval_steps_per_second": 4.676,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 0.9445380568504333,
      "learning_rate": 9e-05,
      "loss": 0.1501,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8538666666666667,
      "eval_loss": 0.5307081341743469,
      "eval_runtime": 12.6762,
      "eval_samples_per_second": 295.829,
      "eval_steps_per_second": 4.654,
      "step": 1665
    }
  ],
  "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": false
      },
      "attributes": {}
    }
  },
  "total_flos": 8.2370660311296e+18,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}