Principle:Interpretml Interpret Decision Tree Classification
Metadata
| Field | Value |
|---|---|
| Sources | Interpretml_Interpret |
| Domains | Machine_Learning, Interpretability |
| Updated | 2026-02-07 |
Overview
Shallow decision trees provide inherently interpretable classification and regression models through axis-aligned splits with interactive tree visualizations.
Description
ClassificationTree and RegressionTree wrap scikit-learn's DecisionTreeClassifier and DecisionTreeRegressor with a default maximum depth of 3, producing compact trees that are human-readable. They conform to the InterpretML ExplainerMixin API, providing both global explanations (full tree structure with feature importance) and local explanations (per-instance decision paths with feature contributions).
Usage
Use shallow decision trees when interpretability is paramount and the relationship between features and target can be adequately captured by a small number of axis-aligned splits. The default depth of 3 ensures the tree remains visually interpretable.