Scilab Home page | Wiki | Bug tracker | Forge | Mailing list archives | ATOMS | File exchange
Please login or create an account
Change language to: English - Français - Português - 日本語 -
Справка Scilab >> Linear Algebra > Eigenvalue and Singular Value > sva

sva

singular value approximation

Syntax

[U,s,V]=sva(A,k)
[U,s,V]=sva(A,tol)

Arguments

A

real or complex matrix

k

integer

tol

nonnegative real number

Description

Singular value approximation.

[U,S,V]=sva(A,k) with k an integer >=1, returns U,S and V such that B=U*S*V' is the best L2 approximation of A with rank(B)=k.

[U,S,V]=sva(A,tol) with tol a real number, returns U,S and V such that B=U*S*V' such that L2-norm of A-B is at most tol.

Examples

A=rand(5,4)*rand(4,5);
[U,s,V]=sva(A,2);
B=U*s*V';
svd(A)
svd(B)
clean(svd(A-B))

See also

  • svd — singular value decomposition
Scilab Enterprises
Copyright (c) 2011-2017 (Scilab Enterprises)
Copyright (c) 1989-2012 (INRIA)
Copyright (c) 1989-2007 (ENPC)
with contributors
Last updated:
Tue Feb 14 15:13:21 CET 2017