[Insight-developers] FEM Optimization

[Luis Ibanez]

Hi Stephen,


Thanks a lot for finding a fix for this intrincated problem.


I just committed the the fixed dsrc2c.c file that you sent me,
and restored the original FEM files that depend on it.


The FEM Image Registration test is now running in 42 seconds
in zion.kitware.


This test should be clean in tomorrow's dashboard


  Thanks,


      Luis



---------------------------------------------------------------------------------------------------
On 11/7/06, Stephen R. Aylward <Stephen.Aylward at kitware.com> wrote:
>
>
> The speed and copyright issues with FEM optimization in the
> itkFEMRegistrationFilterTest seem to be resolved.
>
> I'm sending the changes to Luis to submit to cvs.
>
> The changes were made to dsrc2c.c in Insight/Code/Numerics/FEM
> - Removed zbrent - copyright issue
> - Removed eqrt1s - copyright issue
> - Removed eigvss_ - required zbrent
> - Changed call in eigvns_ to eqrt1s with call to v3p_netlib_tqlrat
> - Modified chgcon_ to always call eigvns_
>     - Before would call eigvss_ if the matrix used symmetric storage
>        but we've hard-coded itkFEMLinearsystemWrapperItpack
>        to always use nonsymmetric storage
>
> The optimizer now runs in a matter of seconds.
>
> Stephen
>
> --
> =============================================================
> Stephen R. Aylward, Ph.D.
> Chief Medical Scientist
> Kitware, Inc. - Chapel Hill Office
> http://www.kitware.com
> Phone: (518)371-3971 x300
> _______________________________________________
> Insight-developers mailing list
> Insight-developers at itk.org
> http://www.itk.org/mailman/listinfo/insight-developers
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://www.itk.org/mailman/private/insight-developers/attachments/20061107/743ea329/attachment.html