[Scipy-tickets] [SciPy] #231: A wrapper for ARPACK

SciPy scipy-tickets at scipy.net
Tue Oct 10 14:02:43 CDT 2006


#231: A wrapper for ARPACK
--------------------------+-------------------------------------------------
 Reporter:  nils          |        Owner:  somebody
     Type:  enhancement   |       Status:  new     
 Priority:  normal        |    Milestone:          
Component:  scipy.sparse  |      Version:          
 Severity:  normal        |   Resolution:          
 Keywords:                |  
--------------------------+-------------------------------------------------
Comment (by aric):

 I didn't spend too much time trying to track down the error/issue with
 <s,d>neupd.
 I agree that the behavior is different depending on whether
 ritz/eigenvectors are
 computed or not.  I get segfaults occasionally but only when I am
 requesting eigenvectors.

 One obvious thing to do would be to make sure the FORTRAN examples work
 correctly.
 I'm mildly allergic to FORTRAN so I didn't attempt that.  <s,d>neupd do
 also call
 plenty of LAPACK routines and I'm using my system (Linux,OSX/fink) LAPACK
 libraries
 instead of the parts provided with ARPACK.  In fact when I compile in the
 LAPACK and
 BLAS routines supplied it doesn't work at all (hang).

 I wasn't sure what to do with the solver.  The superlu seemed reasonable.

 Did I take the same approach you did to the interface?
 Could we merge the two efforts and try to get this into the sandbox or
 somewhere where we could work on it?
 I also left a message at scipy-dev:
 http://article.gmane.org/gmane.comp.python.scientific.devel/5166

-- 
Ticket URL: <http://projects.scipy.org/scipy/scipy/ticket/231#comment:3>
SciPy <http://www.scipy.org/>
SciPy is open-source software for mathematics, science, and engineering.


More information about the Scipy-tickets mailing list