Neighbors.KDTree

constructor(
    X: number[][] = [],
    leafSize: number = 40,
    metric: Distance.IDistanceType = 'euclidiean',
    p: number = 2
)

Parameters

• X (number[][]): data used to build the tree. You can also call fit later.
• leafSize (number): maximum samples per leaf. Default is 40.
• metric (Distance.IDistanceType): distance metric used for queries. Default 'euclidiean'.
• p (number): order of the norm when using Minkowski distance. Default 2.

Algorithm

KD-tree recursively splits points by dimension. Each internal node stores a split dimension and value and points to left and right subtrees. During search the tree is pruned using bounding boxes to efficiently locate nearest neighbors.

query(X: number[][], k: number = 1) returns distances and indices of nearest neighbors.

queryRadius(X: number[][], r: number, returnDistance = false) finds neighbors within given radius.

Example

const tree = new KDTree(X, 2);
const result = tree.query(X.slice(0, 1), 3);

Practical guide: KDTree in JavaScript and TypeScript

KDTree speeds up nearest-neighbor searches for low-to-medium dimensional numeric feature spaces.

When to use KDTree

• You need many repeated neighbor lookups with Euclidean-like distances.
• Dataset dimensionality is not too high for kd-tree pruning to remain effective.
• You want faster KNN-style operations in JS services or browser apps.

Implementation workflow

1. Index feature vectors with KDTree construction.
2. Run neighbor queries and capture distances/indices.
3. Tune leaf/query parameters to balance speed and accuracy.

JavaScript deployment notes

• Prefer feature scaling for distance-based and gradient-based algorithms to improve stability.
• In browser apps, run heavy training in Web Workers to keep UI interactions smooth.
• Keep a simple baseline from the same module as a fallback model for comparison.

Search intents this page targets

• KDTree JavaScript
• KDTree TypeScript
• KDTree browser machine learning
• @kanaries/ml KDTree