Hands-On Ensemble Learning with Python

Implementation

For this example, we will use 1500 instances as the train set, and the remaining 297 as the test set. We will generate 10 bootstrap samples, and consequently 10 decision-tree models. We will then combine the base predictions using hard voting:

1. We load the libraries and data as shown in the following example:

# --- SECTION 1 ---# Libraries and data loadingfrom sklearn.datasets import load_digitsfrom sklearn.tree import DecisionTreeClassifierfrom sklearn import metricsimport numpy as npdigits = load_digits()train_size = 1500train_x, train_y = digits.data[:train_size], digits.target[:train_size]test_x, test_y = digits.data[train_size:], digits.target[train_size:]

2. We then create our bootstrap samples and train the corresponding ...