Hands-On Data Science for Marketing

Interpreting decision trees

Now that we have trained a decision tree model, we need to extract the insights from the model. In this section, we are going to use a package called graphviz. You can install this package by using the following command in your Terminal:

conda install python-graphviz

Once you have installed this package correctly, you should be able to import the package as follows:

import graphviz

Now that we have set up our environment with the new package, graphviz, let's take a look at the following code to see how we can visualize the trained decision tree:

dot_data = tree.export_graphviz(    dt_model,     feature_names=features,     class_names=['0', '1'],     filled=True,     rounded=True,     special_characters=True) graph = graphviz.Source(dot_data) ...