holisticai.explainability.metrics.weighted_average_depth

holisticai.explainability.metrics.weighted_average_depth(tree)

Weighted Average Depth calculates the average depth of a tree considering the number of samples that pass through each cut.

Parameters

tree: Tree

The tree to calculate the weighted average depth of.

Returns

float

The weighted average depth value

Examples

>>> from sklearn.datasets import load_iris
>>> from sklearn.tree import DecisionTreeClassifier
>>> from holisticai.explainability.metrics import weighted_average_depth
>>> X, y = load_iris(return_X_y=True)
>>> clf = DecisionTreeClassifier()
>>> clf.fit(X, y)
>>> weighted_average_depth(clf.tree_)

Reference

Laber, E., Murtinho, L., & Oliveira, F. (2023). Shallow decision trees for explainable k-means clustering. Pattern Recognition, 137, 109239.