commons-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Dimitri Pourbaix <>
Subject Re: Math Release 2.1 SVD
Date Fri, 26 Mar 2010 21:43:52 GMT

> The 2.1 API docs for the Singular Value Decomposition say:
> The size p depends on the chosen algorithm:
> for full SVD, p is n,
> for compact SVD, p is the rank r of the matrix (i. e. the number of
> positive singular values),
> for truncated SVD p is min(r, t) where t is user-specified.
> but I don't see any method or constructor to specify whether to do the
> full, compact or truncated SVD.  Am I missing something, or is the code
> missing something?

The doc is not synchronized with the source, I apologize (but the
javadoc should be).  The code offers only one version which is none of
the above!  From the Java file:

  The Singular Value Decomposition of matrix A is a set of three
  matrices: U, &Sigma; and V such that
  A = U &times; &Sigma; &times; V<sup>T</sup>. Let A be  a m &times;
  matrix, then U is a m &times; p orthogonal matrix, &Sigma; is a
  p &times; p diagonal matrix with positive or null elements, V is a
  p &times; n orthogonal matrix (hence V<sup>T</sup> is also orthogonal)
  where p=min(m,n).

Dimitri Pourbaix                         *
Institut d'Astronomie et d'Astrophysique *      Don't worry, be happy
CP 226, office 2.N4.211, building NO     *         and CARPE DIEM.
Universite Libre de Bruxelles            *
Boulevard du Triomphe                    *      Tel : +32-2-650.35.71
  B-1050 Bruxelles                        *      Fax : +32-2-650.42.26     *

To unsubscribe, e-mail:
For additional commands, e-mail:

View raw message