trainer_state.json

{
  "best_metric": 0.848,
  "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0574/checkpoints/checkpoint-666",
  "epoch": 2.0,
  "eval_steps": 500,
  "global_step": 666,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 43.23154067993164,
      "learning_rate": 0.0005,
      "loss": 1.3825,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8133333333333334,
      "eval_loss": 0.6039751172065735,
      "eval_runtime": 13.5703,
      "eval_samples_per_second": 276.338,
      "eval_steps_per_second": 4.348,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 13.650593757629395,
      "learning_rate": 0.0005,
      "loss": 0.5047,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.848,
      "eval_loss": 0.46983763575553894,
      "eval_runtime": 13.3902,
      "eval_samples_per_second": 280.056,
      "eval_steps_per_second": 4.406,
      "step": 666
    }
  ],
  "logging_steps": 500,
  "max_steps": 666,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 2,
  "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": 1.6352858479104e+18,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}