Geek Answers Handbook

Can we choose which decision tree algorithm to use in sklearn?

My question is, is it possible to choose a decision tree algorithm for use in sklearn?

The sklearn user guide mentions an optimized version of the CART algorithm.

Can we switch to other algorithms like C4.5?

+7
scikit-learn decision-tree
source share
2 answers

Not. See documentation

scikit-learn uses an optimised version of the CART algorithm.
+5
source share

But there is a params criterion that we can choose to use "gini" or "entropy":

 clf = tree.DecisionTreeClassifier(criterion="entropy")

criterion: string, optional (default = "gini") Function for measuring split quality. Supported criteria: "gini" for Gini impurities and "entropy" for information.

see Documents

0
source share

More articles:


All Articles