HessenbergDecomposition<T>

Models & Types · AiDotNet.DecompositionMethods.MatrixDecomposition

Implements Hessenberg decomposition, which transforms a matrix into a form that is almost triangular.

Inheritance: MatrixDecompositionBase<T> → HessenbergDecomposition<T>

For Beginners

Think of Hessenberg decomposition as a way to simplify a matrix by making most elements below the diagonal equal to zero, which makes further calculations much faster. It’s like organizing a messy room by putting everything in its place - the matrix becomes easier to work with even though it’s not completely triangular.

How It Works

A Hessenberg matrix has zeros below the first subdiagonal, making it easier to work with for many numerical algorithms. This decomposition is often used as a preprocessing step for eigenvalue calculations and solving linear systems.

Real-world applications:

  • Preprocessing for eigenvalue computation
  • Accelerating iterative methods for solving linear systems
  • Control theory and system analysis

Constructors

Constructor Summary
HessenbergDecomposition(Matrix<T> matrix, HessenbergAlgorithmType algorithm) Initializes a new instance of the Hessenberg decomposition for the specified matrix.

Properties

Property Summary
Matrix<T> HessenbergMatrix { get; set; } Gets the resulting Hessenberg matrix after decomposition.
Matrix<T> OrthogonalMatrix { get; set; } Gets the orthogonal transformation matrix Q from the decomposition.

Methods

Method Summary
Matrix<T> Invert() Calculates the inverse of the original matrix using the Hessenberg decomposition.
Vector<T> Solve(Vector<T> b) Solves a linear system Ax = b using the Hessenberg decomposition.