xgboost_erro.py

error_model= xgb.XGBRegressor()
model_errors = np.square(np.subtract(y_pred, y_test.reshape(1, -1)[0]))
parameters = {'learning_rate': [0.0001, 0.001, 0.01, 0.1, 0.2, 0.3, 0.5, 0.9],
                  'n_estimators': [100, 200, 300, 400, 500],
                  'max_depth': [3,4,6],
                  'min_child_weight': [1, 2, 3]
                  }
grid_xgb = RandomizedSearchCV(error_model, parameters, cv=5, n_jobs=-1)
grid_xgb = grid_xgb.fit(X_test, model_errors)
error_model = grid_xgb.best_estimator_

print("O {} calibrado tem um score de {:.4f} no conjunto de teste.".format(xgb_reg.__class__.__name__,
                                                                           xgb_reg.score(X_test, y_test)))
print("Estes são os melhores parâmetros: {}".format(grid_xgb.best_params_))