trainer_state.json

{
  "best_metric": 0.8962666666666667,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_vit-mae-base/model_idx_0142/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": 8.145560752888351e-05,
      "loss": 1.8963,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.7669333333333334,
      "eval_loss": 0.8458038568496704,
      "eval_runtime": 6.9978,
      "eval_samples_per_second": 535.882,
      "eval_steps_per_second": 8.431,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 5.685427188873291,
      "learning_rate": 5.898649216615971e-05,
      "loss": 0.5308,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8362666666666667,
      "eval_loss": 0.5525839924812317,
      "eval_runtime": 7.4858,
      "eval_samples_per_second": 500.946,
      "eval_steps_per_second": 7.882,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 0.3780677914619446,
      "learning_rate": 3.117501217934705e-05,
      "loss": 0.2474,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8762666666666666,
      "eval_loss": 0.41846010088920593,
      "eval_runtime": 7.2707,
      "eval_samples_per_second": 515.772,
      "eval_steps_per_second": 8.115,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.20403502881526947,
      "learning_rate": 8.644207645970472e-06,
      "loss": 0.1038,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8877333333333334,
      "eval_loss": 0.37733352184295654,
      "eval_runtime": 7.4732,
      "eval_samples_per_second": 501.79,
      "eval_steps_per_second": 7.895,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 2.890033721923828,
      "learning_rate": 8.010389962842535e-11,
      "loss": 0.0477,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8962666666666667,
      "eval_loss": 0.35839173197746277,
      "eval_runtime": 7.3384,
      "eval_samples_per_second": 511.007,
      "eval_steps_per_second": 8.04,
      "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
}