Daniel Oberhoff wrote:

Hi,
This is not directly linked to octave, but I assume that there are
enough people on this who know about such stuff:
Is there an efficient way to compute the eigenvalues of a matrix
given it's dot product with an aribtrary vector?
The problem is that the matrix M is nxn with n=O(10,000), which
simply blows memory, but there is a relatively quick way to get those
dot products M*v. The sparsity of this matrix is at about 50%, on top
of this it is symmetric, and probably positive definite.
An octave solution (or pointers) would be nice of course ;).
Daniel
_______________________________________________
Help-octave mailing list
address@hidden
https://www.cae.wisc.edu/mailman/listinfo/help-octave

You want the eigs function from octave-forge. The basic method of eigs
is based purely on matrix vector products..
D.
--

`David Bateman
``address@hidden
`Motorola Labs - Paris +33 1 69 35 48 04 (Ph)
Parc Les Algorithmes, Commune de St Aubin +33 6 72 01 06 33 (Mob)
91193 Gif-Sur-Yvette FRANCE +33 1 69 35 77 01 (Fax)

`The information contained in this communication has been classified
``as:
`
[x] General Business Information
[ ] Motorola Internal Use Only
[ ] Motorola Confidential Proprietary