trainer_state.json

{
  "best_metric": 0.4304,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_vit-mae-base/model_idx_0658/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": 0.0005,
      "loss": 3.1129,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.23413333333333333,
      "eval_loss": 2.78165864944458,
      "eval_runtime": 5.8561,
      "eval_samples_per_second": 640.361,
      "eval_steps_per_second": 10.075,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": Infinity,
      "learning_rate": 0.0005,
      "loss": 2.5349,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.3253333333333333,
      "eval_loss": 2.4504144191741943,
      "eval_runtime": 5.8502,
      "eval_samples_per_second": 641.001,
      "eval_steps_per_second": 10.085,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 10.048211097717285,
      "learning_rate": 0.0005,
      "loss": 2.1705,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.39093333333333335,
      "eval_loss": 2.162705183029175,
      "eval_runtime": 5.7873,
      "eval_samples_per_second": 647.969,
      "eval_steps_per_second": 10.195,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 15.149518966674805,
      "learning_rate": 0.0005,
      "loss": 1.9114,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.42186666666666667,
      "eval_loss": 2.0907955169677734,
      "eval_runtime": 6.106,
      "eval_samples_per_second": 614.148,
      "eval_steps_per_second": 9.663,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 14.534804344177246,
      "learning_rate": 0.0005,
      "loss": 1.6987,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.4304,
      "eval_loss": 1.9864001274108887,
      "eval_runtime": 5.8218,
      "eval_samples_per_second": 644.126,
      "eval_steps_per_second": 10.134,
      "step": 1665
    }
  ],
  "logging_steps": 500,
  "max_steps": 1665,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 5,
  "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": 8.2370660311296e+18,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}