trainer_state.json

{
  "best_metric": 0.8568,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_vit-mae-base/model_idx_0641/checkpoints/checkpoint-2331",
  "epoch": 7.0,
  "eval_steps": 500,
  "global_step": 2331,
  "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": 0.00025727155727155724,
      "loss": 2.3618,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.6042666666666666,
      "eval_loss": 1.3193330764770508,
      "eval_runtime": 13.8737,
      "eval_samples_per_second": 270.296,
      "eval_steps_per_second": 4.253,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": Infinity,
      "learning_rate": 0.0002145431145431145,
      "loss": 0.9494,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.7493333333333333,
      "eval_loss": 0.8402338624000549,
      "eval_runtime": 14.4574,
      "eval_samples_per_second": 259.383,
      "eval_steps_per_second": 4.081,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": Infinity,
      "learning_rate": 0.0001718146718146718,
      "loss": 0.5599,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.7928,
      "eval_loss": 0.6813160181045532,
      "eval_runtime": 13.524,
      "eval_samples_per_second": 277.285,
      "eval_steps_per_second": 4.363,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 58.57990646362305,
      "learning_rate": 0.00012895752895752895,
      "loss": 0.3364,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.808,
      "eval_loss": 0.6438312530517578,
      "eval_runtime": 13.3019,
      "eval_samples_per_second": 281.914,
      "eval_steps_per_second": 4.435,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 4.359346866607666,
      "learning_rate": 8.61003861003861e-05,
      "loss": 0.1825,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8336,
      "eval_loss": 0.5756457448005676,
      "eval_runtime": 13.7463,
      "eval_samples_per_second": 272.802,
      "eval_steps_per_second": 4.292,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 0.06497708708047867,
      "learning_rate": 4.324324324324324e-05,
      "loss": 0.0889,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8429333333333333,
      "eval_loss": 0.5477652549743652,
      "eval_runtime": 14.5913,
      "eval_samples_per_second": 257.002,
      "eval_steps_per_second": 4.043,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": 0.3372875452041626,
      "learning_rate": 3.861003861003861e-07,
      "loss": 0.0341,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.8568,
      "eval_loss": 0.5335781574249268,
      "eval_runtime": 13.8419,
      "eval_samples_per_second": 270.917,
      "eval_steps_per_second": 4.262,
      "step": 2331
    }
  ],
  "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": true
      },
      "attributes": {}
    }
  },
  "total_flos": 1.153189244358144e+19,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}