Home
Reading
Searching
Subscribe
Sponsors
Statistics
Posting
Contact
Spam
Lists
Links
About
Hosting
Filtering
Features Download
Marketing
Archives
FAQ
Blog
 
Gmane
From: Andrew MacLean <andrew.maclean <at> gmail.com>
Subject: Re: Incorrect values from scipy.sparse.linalg.lobpcg using large matrices
Newsgroups: gmane.comp.python.scientific.user
Date: Tuesday 7th June 2011 22:18:47 UTC (over 5 years ago)
Version 0.7, so that might be possible. I will look into trying this
in Scipy 0.9. However, I have also tried this using the most up to
date Matlab version of lobpcg (available from
http://www.mathworks.com/matlabcentral/fileexchange/),
and that also
gave values that were about 1000 times too large.

On Jun 7, 11:50 am, Pauli Virtanen  wrote:
> Mon, 06 Jun 2011 15:07:48 -0700, Andrew MacLean wrote:
>
> > I have been using lobpcg (scipy.sparse.linalg.lobpcg) to solve
symmetric
> > generalized eigenvalue problems with large, sparse stiffness and mass
> > matrices, say 'A' and 'B'. The problem is of the form Av = λBV.
>
> Which version of Scipy? In Scipy 0.9, some bugs in lobpcg that appeared
> on certain platforms were fixed. If you are using Scipy < 0.9, it's
> possible you are hitting these.
>
> --
> Pauli Virtanen
>
> _______________________________________________
> SciPy-User mailing list
> SciPy-U...@scipy.orghttp://mail.scipy.org/mailman/listinfo/scipy-user
_______________________________________________
SciPy-User mailing list
[email protected]
http://mail.scipy.org/mailman/listinfo/scipy-user
 
CD: 17ms