#@teakit/matrix

A TypeScript port of ml-matrix (v6.12.2): a comprehensive, class-based matrix manipulation and computation library with feature parity (element-wise ops, linear algebra, decompositions, views, wrappers, statistics, special matrices). Zero runtime dependencies.

Canonical import (default also works):

import { Matrix } from "@teakit/matrix";
// or: import Matrix from "@teakit/matrix";

const a = new Matrix([
  [1, 2],
  [3, 4],
]);

API names and semantics match ml-matrix. It is class-based and mutation-first: instance methods mutate this and return it (chainable); static methods (e.g. Matrix.add(a, b)) return a new matrix. Views and wrappers are zero-copy subclasses of AbstractMatrix (writing through them mutates the underlying data). new Matrix(...) takes a 2D array, a (rows, columns) pair, or another matrix — not a 1D array (use rowVector / columnVector / from1DArray). Errors are native Error / TypeError / RangeError (no stable codes).

For behavior details, open the matching reference page under teakit-matrix/references/.

#Table of Contents

• Core concept — shape, construction, get/set, mutation model, conversion.
• Operations — element-wise arithmetic, bitwise, Math.*.
• Linear algebra — multiply, transpose, determinant, inverse, solve, pseudo-inverse.
• Decompositions — SVD, EVD, LU, QR, Cholesky, NIPALS.
• Views & wrappers — zero-copy views and array wrappers.
• Statistics — sum/mean/variance, min/max, center/scale, covariance/correlation.
• Special matrices — SymmetricMatrix, DistanceMatrix.
• Errors — what throws and when.