Re: SVD require 'k' singular values/vectors only

From: spasmous (spasmous_at_yahoo.com)
Date: 01/25/05 

Date: 25 Jan 2005 01:37:03 -0800

Steven Lord wrote:
>
> SVDS converts that matrix of size [m n] into a sparse matrix of size
[(m+n),
> (m+n)] and uses EIGS to compute the eigenvalues of that matrix. If
your
> original matrix is full, this will probably be a Big Sparse matrix.
You
> might want to try using the economy SVD using "SVD(A, 0)" [see HELP
SVD for
> a full description of what that syntax does] and pull the required K
> singular values out of that -- that may be quicker.
>

Thanks Steven - I know svds() forms [A I;I A'] (or something like that)
and also that svd(A,0) is faster than the full svd. I'm actually asking
whether there is an svd algorithm (other than svds :) that pulls out
the singular values/vectors in descending order rather than getting
them all and sorting afterwards.

Relevant Pages

  • Meschach - SVD on a sparse matrix
    ... I'm trying use meschach, compiling some programs, etc. because I need ... compute svd on a sparse matrix. ... There are, of course, sparse matrix packages that are ...
    (sci.math.num-analysis)
  • Re: Meschach - SVD on a sparse matrix
    ... I'm trying use meschach, compiling some programs, etc. because I need ... compute svd on a sparse matrix. ... There are, of course, sparse matrix packages that are ...
    (sci.math.num-analysis)
  • Re: Meschach - SVD on a sparse matrix
    ... compute svd on a sparse matrix. ... There are, of course, sparse matrix packages that are ... matrices are self-contained data structures, ...
    (sci.math.num-analysis)
  • Large & sparse SVD
    ... I need to compute the SVD of a large and sparse matrix: ... and 20 columns with 10% non-zero elements. ... memory everytime I tried. ...
    (sci.math.num-analysis)
  • Re: SVD for a large sparse matrix
    ... The matrix dimension is of the size 17640 x 254016. ... When i use the matlab's SVD function its gives "out of memory" ... as it tries to convert the sparse matrix into a full one. ... Then i tried using SVDS which is svd for sparse matrices, ...
    (comp.soft-sys.matlab)