Implementation:DistrictDataLabs Yellowbrick DecisionBoundariesVisualizer

Knowledge Sources	DistrictDataLabs_Yellowbrick
Domains	Classification, Visualization
Last Updated	2026-02-08 05:00 GMT

Overview

Concrete tool for visualizing decision boundaries of classifiers in bivariate feature space, provided by the Yellowbrick contrib module.

Description

The DecisionBoundariesVisualizer renders the decision regions of a trained classifier as a colored mesh over a 2D feature grid. It overlays scatter points colored by class label and supports automatic feature column selection from DataFrames, structured arrays, and numpy arrays. The visualizer creates a meshgrid from the selected feature pair and uses the classifier's predictions to color the background.

Usage

Import this visualizer when you need to inspect how a classifier partitions bivariate feature space. It is particularly useful for understanding decision boundaries of models like SVM, KNN, and decision trees on two selected features.

Code Reference

Source Location

• Repository: DistrictDataLabs_Yellowbrick
• File: yellowbrick/contrib/classifier/boundaries.py
• Lines: 1-523

Signature

class DecisionBoundariesVisualizer(ClassificationScoreVisualizer):
    def __init__(
        self,
        estimator,
        ax=None,
        x_name=None,
        y_name=None,
        features=None,
        classes=None,
        show_scatter=True,
        step_size=0.0025,
        markers=None,
        pcolormesh_alpha=0.8,
        scatter_alpha=1.0,
        encoder=None,
        is_fitted="auto",
        force_model=False,
        **kwargs,
    ):
        """Bivariate decision boundary visualizer for classifiers."""

def decisionviz(
    estimator, X, y, ax=None, x_name=None, y_name=None,
    features=None, classes=None, show_scatter=True, step_size=0.0025,
    markers=None, pcolormesh_alpha=0.8, scatter_alpha=1.0,
    encoder=None, is_fitted="auto", force_model=False, **kwargs,
):
    """Quick method for one-off decision boundary visualization."""

# Alias
DecisionViz = DecisionBoundariesVisualizer

Import

from yellowbrick.contrib.classifier import DecisionBoundariesVisualizer, DecisionViz, decisionviz

I/O Contract

Inputs

Name	Type	Required	Description
estimator	sklearn classifier	Yes	Fitted or unfitted classifier
X	array-like (n, 2)	Yes	Bivariate feature data
y	array-like	Yes	Target class labels
features	list of str	No	Feature names for axis labels
classes	list of str	No	Human-readable class names
step_size	float	No	Resolution of decision boundary mesh (default: 0.0025)

Outputs

Name	Type	Description
ax	matplotlib.Axes	Axes with decision boundary mesh and scatter plot

Usage Examples

from sklearn.svm import SVC
from sklearn.datasets import make_moons
from yellowbrick.contrib.classifier import DecisionViz

X, y = make_moons(n_samples=200, noise=0.2)

viz = DecisionViz(SVC(), features=["Feature 1", "Feature 2"], classes=["A", "B"])
viz.fit(X, y)
viz.show()

# Quick method
from yellowbrick.contrib.classifier import decisionviz

decisionviz(SVC(), X, y, features=["x1", "x2"], classes=["A", "B"])

Related Pages

• Environment:DistrictDataLabs_Yellowbrick_Python_Scikit_Learn_Environment