Python Feature Engineering Cookbook

There's more...

We can perform decision tree discretization with scikit-learn within a grid search to find the optimal parameters to determine the most predictive decision tree. To do so, we first do the imports, as in step 1 of the main recipe, and we add one additional import:

1. Let's import the grid search from scikit-learn:

from sklearn.model_selection import GridSearchCV

Next, let's load the dataset and divide it into train and test sets, as in step 2 and step 3 of the main recipe, and now let's set up a parameter search grid.

2. Let's set up a dictionary with the parameters we would like to test, in this case, four different tree depths:

param_grid = {'max_depth': [1,2,3,4]}

3. Let's now set up the decision tree inside a grid search with ...