trainer_state.json

{
  "best_metric": 0.868,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_vit-mae-base/model_idx_0728/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": 1.6115,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8218666666666666,
      "eval_loss": 0.6447762250900269,
      "eval_runtime": 10.649,
      "eval_samples_per_second": 352.145,
      "eval_steps_per_second": 5.54,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 2.7947280406951904,
      "learning_rate": 9e-05,
      "loss": 0.4468,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8509333333333333,
      "eval_loss": 0.5274326205253601,
      "eval_runtime": 10.4375,
      "eval_samples_per_second": 359.281,
      "eval_steps_per_second": 5.653,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 0.28197550773620605,
      "learning_rate": 9e-05,
      "loss": 0.2361,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.852,
      "eval_loss": 0.5307477116584778,
      "eval_runtime": 10.5468,
      "eval_samples_per_second": 355.56,
      "eval_steps_per_second": 5.594,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.8218750953674316,
      "learning_rate": 9e-05,
      "loss": 0.1453,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8576,
      "eval_loss": 0.5152024030685425,
      "eval_runtime": 10.3627,
      "eval_samples_per_second": 361.875,
      "eval_steps_per_second": 5.694,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": Infinity,
      "learning_rate": 9e-05,
      "loss": 0.1014,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.868,
      "eval_loss": 0.5012599229812622,
      "eval_runtime": 10.5912,
      "eval_samples_per_second": 354.068,
      "eval_steps_per_second": 5.571,
      "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
}