Note

Go to the end to download the full example code or to run this example in your browser via JupyterLite or Binder.

Nearest Centroid Classification#

Sample usage of the Nearest Centroid Classifier with different shrink thresholds. It will plot the decision boundaries for each class.

# Authors: The scikit-learn developers
# SPDX-License-Identifier: BSD-3-Clause

import matplotlib.pyplot as plt
import numpy as np
from matplotlib.colors import ListedColormap

from sklearn import datasets
from sklearn.inspection import DecisionBoundaryDisplay
from sklearn.neighbors import NearestCentroid

# import some data to play with
iris = datasets.load_iris()
# we only take the first two features. We could avoid this ugly
# slicing by using a two-dim dataset
X = iris.data[:, :2]
y = iris.target

for shrinkage in [None, 0.2]:
    # we create an instance of Nearest Centroid Classifier and fit the data.
    clf = NearestCentroid(shrink_threshold=shrinkage)
    clf.fit(X, y)
    y_pred = clf.predict(X)
    acc = np.mean(y == y_pred)

    _, ax = plt.subplots()
    disp = DecisionBoundaryDisplay.from_estimator(
        clf, X, ax=ax, response_method="predict", alpha=0.5
    )

    # Plot also the training points
    cmap = ListedColormap(disp.multiclass_colors_)
    plt.scatter(X[:, 0], X[:, 1], c=y, cmap=cmap, edgecolor="k", s=20)
    plt.title(
        f"3-Class classification (shrink_threshold={shrinkage})\nAccuracy: {acc:.2f}"
    )
    plt.axis("tight")

plt.show()