trainer_state.json

{
  "best_metric": 0.8968,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_vit-mae-base/model_idx_0770/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": 5.604204204204204e-05,
      "loss": 1.5446,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.82,
      "eval_loss": 0.6313930153846741,
      "eval_runtime": 7.2849,
      "eval_samples_per_second": 514.762,
      "eval_steps_per_second": 8.099,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 2.698061227798462,
      "learning_rate": 4.2042042042042036e-05,
      "loss": 0.3974,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8706666666666667,
      "eval_loss": 0.43352577090263367,
      "eval_runtime": 6.8209,
      "eval_samples_per_second": 549.778,
      "eval_steps_per_second": 8.65,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 1.542324185371399,
      "learning_rate": 2.8042042042042038e-05,
      "loss": 0.1851,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8837333333333334,
      "eval_loss": 0.37009739875793457,
      "eval_runtime": 6.934,
      "eval_samples_per_second": 540.816,
      "eval_steps_per_second": 8.509,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.4199316203594208,
      "learning_rate": 1.404204204204204e-05,
      "loss": 0.0898,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8952,
      "eval_loss": 0.3607785403728485,
      "eval_runtime": 7.216,
      "eval_samples_per_second": 519.679,
      "eval_steps_per_second": 8.176,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 0.5445083975791931,
      "learning_rate": 4.204204204204204e-08,
      "loss": 0.0368,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8968,
      "eval_loss": 0.35065704584121704,
      "eval_runtime": 6.7661,
      "eval_samples_per_second": 554.235,
      "eval_steps_per_second": 8.72,
      "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
}