Hi Andrew,

> I think that it is sufficient just to remove perturbation and repeat the
> iteration like in other cases (iteration limit, time limit, etc.),

I agree, this is a better approach - the solver is faster with this
change, requiring less iterations than with my patches.
On a related note, in order to build a version with perturbation
disabled, the attached trivial patch is required.

Best Regards,

Chris Matrakidis

Attachment: spydual.patch
Description: Binary data

_______________________________________________
Help-glpk mailing list
[email protected]
https://lists.gnu.org/mailman/listinfo/help-glpk

Reply via email to