Course Content

# Ensemble Learning

1. Basic Principles of Building Ensemble Models

Ensemble Learning

## Challenge: Solving Task Using Stacking Regressor

**Stacking Regressor** is a stacking ensemble learning model used to solve regression tasks. The principle of work of this model is similar to the Stacking Classifier: the only difference is that we use **regression algorithms** as the base and meta-models of the ensemble.

We can use the `StackingRegressor`

class from the `sklearn`

library to implement this model in Python.

# Task

The `make_friedman1`

dataset is a synthetic dataset frequently used for regression tasks in machine learning. This dataset is widely used in regression tutorials and experimentation because it's simple, yet it can be customized with different noise levels and feature dimensions to simulate various regression scenarios.

Your task is to solve the regression task on the Friedman dataset using Stacking Regressor:

- Provide split on train and test subsets of the training set: the proportion of the dataset to be included in the test split must be
`0.2`

. - Use Decision Tree Regressor with
`max_depth`

equals`3`

as one of the base models. - Create a Stacking Regressor model using base models and meta model.
- Fit Stacking Regressor model on the training data.

Everything was clear?