def cat_hyp(depth, bagging_temperature): # Function to optimize depth and bagging temperature
  params = {"iterations": 100,
            "learning_rate": 0.05,
            "eval_metric": "R2",
            "verbose": False} # Default Parameters
  params[ "depth"] = int(round(depth))
  params["bagging_temperature"] = bagging_temperature

  cat_feat = [] # Categorical features list
  cv_dataset = cgb.Pool(data=X,
                  label=y,
                  cat_features=cat_feat)

  scores = cgb.cv(cv_dataset,
              params,
              fold_count=3)
  return np.max(scores['test-R2-mean'])  # Return maximum R-Squared value