trainer_state.json

{
  "best_metric": 0.8848,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_vit-mae-base/model_idx_0908/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": 5e-05,
      "loss": 1.8823,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.7789333333333334,
      "eval_loss": 0.8475160002708435,
      "eval_runtime": 7.6137,
      "eval_samples_per_second": 492.535,
      "eval_steps_per_second": 7.749,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 2.1861202716827393,
      "learning_rate": 5e-05,
      "loss": 0.5891,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8578666666666667,
      "eval_loss": 0.5199984908103943,
      "eval_runtime": 7.7756,
      "eval_samples_per_second": 482.277,
      "eval_steps_per_second": 7.588,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 2.0400218963623047,
      "learning_rate": 5e-05,
      "loss": 0.35,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8624,
      "eval_loss": 0.5003975629806519,
      "eval_runtime": 7.5553,
      "eval_samples_per_second": 496.337,
      "eval_steps_per_second": 7.809,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 1.0186737775802612,
      "learning_rate": 5e-05,
      "loss": 0.2264,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8754666666666666,
      "eval_loss": 0.4601824879646301,
      "eval_runtime": 7.3256,
      "eval_samples_per_second": 511.904,
      "eval_steps_per_second": 8.054,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 0.6033766269683838,
      "learning_rate": 5e-05,
      "loss": 0.1594,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8746666666666667,
      "eval_loss": 0.45001721382141113,
      "eval_runtime": 7.2229,
      "eval_samples_per_second": 519.18,
      "eval_steps_per_second": 8.168,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 1.7927721738815308,
      "learning_rate": 5e-05,
      "loss": 0.1124,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8824,
      "eval_loss": 0.44116437435150146,
      "eval_runtime": 8.2154,
      "eval_samples_per_second": 456.459,
      "eval_steps_per_second": 7.182,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": 2.942488193511963,
      "learning_rate": 5e-05,
      "loss": 0.0876,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.8848,
      "eval_loss": 0.43750691413879395,
      "eval_runtime": 7.3652,
      "eval_samples_per_second": 509.154,
      "eval_steps_per_second": 8.011,
      "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
}