KNeighborsClassifier

#include <Skigen/Neighbors>

template <typename Scalar = double>
class Skigen::KNeighborsClassifier(n_neighbors=5)

Classifier implementing the k-nearest neighbors vote.

Uses brute-force computation of distances. For each query point, finds the n_neighbors closest training points and predicts by majority vote.

Mirrors sklearn.neighbors.KNeighborsClassifier.

---

Parameters:

• n_neighbors : int, default=5 Number of neighbors to use (int, default 5).

---

Attributes:

• is_fitted : bool Whether the estimator has been fitted.

• n_neighbors : int Number of neighbors.

---

Methods

fit(X, y)

Fit the k-nearest neighbors classifier from the training set.

Stores the training data for later queries.

Parameters:

• X : MatrixType Training data of shape (n_samples, n_features).

• y : IndexVector Target values of shape (n_samples,) with integer class labels.

Returns:

• result : KNeighborsClassifier Reference to the fitted estimator (*this).

Throws:

• std::invalid_argument — if n_samples < n_neighbors or X and y have inconsistent lengths.

---

predict(X)

Predict class labels for the provided data.

Parameters:

• X : MatrixType Test samples of shape (n_samples, n_features).

Returns:

• result : IndexVector Integer vector of predicted class labels (n_samples,).

Throws:

• std::runtime_error — if the model has not been fitted.
• std::invalid_argument — if feature count doesn't match training data.

---

score(X, y)

Return the mean accuracy on the given test data and labels.

Parameters:

• X : MatrixType Test samples of shape (n_samples, n_features).

• y : IndexVector True class labels of shape (n_samples,).

Returns:

• result : Scalar Mean accuracy.

---

Example

// Compare different k values
std::cout << "=== KNN: varying k ===\n";
for (int k : {1, 3, 5, 7, 11}) {
    Skigen::KNeighborsClassifier<double> knn(k);
    knn.fit(split.X_train, split.y_train);
    auto pred = knn.predict(split.X_test);
    std::cout << "  k=" << std::setw(2) << k
              << "  accuracy=" << Skigen::Metrics::accuracy_score(split.y_test, pred)
              << "  F1=" << Skigen::Metrics::f1_score(split.y_test, pred) << "\n";
}

---