Moving on to more complex ensembles, we will utilize stacking to combine basic regressors more efficiently. Using StackingRegressor from Chapter 4, Stacking, we will try to combine the same algorithms as we did with voting. First, we modify the predict function of our ensemble (to allow for single-instance prediction) as follows:

 # Generates the predictions
 def predict(self, x_data):

# Create the predictions matrix
 predictions = np.zeros((len(x_data), len(self.base_learners)))

names = list(self.base_learners.keys())

# For each base learner
 for i in range(len(self.base_learners)):
 name = names[i]
 learner = self.base_learners[name]

# Store the predictions in a column
 preds = learner.predict(x_data)
 predictions[:,i] = preds

# Take the row-average
 predictions = np.mean(predictions, axis=1)
 return predictions

Again, we modify the code to use the stacking regressor, as follows:

base_learners = [[SVR(), LinearRegression(), KNeighborsRegressor()],
                [LinearRegression()]]
lr = StackingRegressor(base_learners)

In this setup, the ensemble yields a model with an MSE of 16.17 and a Sharpe value of 0.21.