[SciPy-user] most negative eigenvalue

dmitrey openopt@ukr....
Tue Aug 28 09:07:10 CDT 2007

yes, first of all I'm interested in tools that are provided by numpy (I 
see numpy.linalg have some routines for eigenvalues).
Matthieu and I are trying to avoid any additional openopt dependences 
except numpy, moreover, because of single routine from dozens of the 
same ones (I mean Golgfeld step here).
I don't think the disadvantages of numpy vs others will be significant 
in the case involved, so I decided to use eigvalsh from numpy.linalg.


Alan G Isaac wrote:
> I know you mean, what is the best SciPy way,
> but you may be interested in the PETSc routines.
> There is some minimum eigenvalue stuff in there.
> (Some of these may also be useful to openopt;
> note that the license is NOT GPL:
> http://www-unix.mcs.anl.gov/petsc/petsc-as/documentation/copyright.html)
> Cheers,
> Alan Isaac
> _______________________________________________
> SciPy-user mailing list
> SciPy-user@scipy.org
> http://projects.scipy.org/mailman/listinfo/scipy-user

More information about the SciPy-user mailing list