EigenDecomposition<T>
Models & Types · AiDotNet.DecompositionMethods.MatrixDecomposition
Performs eigenvalue decomposition of a matrix, breaking it down into its eigenvalues and eigenvectors.
Inheritance: MatrixDecompositionBase<T> → EigenDecomposition<T>
For Beginners
Eigenvalue decomposition finds special directions (eigenvectors) in which a matrix acts like simple scaling. When you multiply the matrix by an eigenvector, you get the same vector back but scaled by a number (the eigenvalue). Think of it like finding the “natural directions” of a transformation.
How It Works
Eigenvalue decomposition is a way to factorize a square matrix into a set of eigenvectors and eigenvalues. This is useful in many applications including principal component analysis, vibration analysis, and solving systems of differential equations.
Real-world applications:
- Principal Component Analysis (PCA) for data dimensionality reduction
- Vibration analysis in mechanical engineering
- Google’s PageRank algorithm
Constructors
| Constructor | Summary |
|---|---|
EigenDecomposition(Matrix<T> matrix, EigenAlgorithmType algorithm) |
Creates a new eigenvalue decomposition for the specified matrix. |
Properties
| Property | Summary |
|---|---|
Matrix<T> EigenVectors { get; set; } |
Gets the eigenvectors of the decomposed matrix. |
Vector<T> EigenValues { get; set; } |
Gets the eigenvalues of the decomposed matrix. |
Methods
| Method | Summary |
|---|---|
Matrix<T> Invert() |
Computes the inverse of the original matrix using the eigenvalue decomposition. |
Vector<T> Solve(Vector<T> b) |
Solves a system of linear equations Ax = b using the eigenvalue decomposition. |