[Scip] Bad dual bound

[Stefan Vigerske]

Hi,

I get a gap of 161.04% after 100 nodes, still big though.

I think I wrote this twice already, but here again: In order to help, we 
will need more information, like which SCIP version you are using with 
which external components (e.g., LP and NLP solvers), on which platform, 
using which input format.

One possibility for having a larger gap than I do is that you run 
without Ipopt, but that's just speculation.

Stefan

On 04/04/2013 12:36 PM, Liffey1986 at gmx.de wrote:
> Hi all,
> I tried some instances from the MINLPLib and got a result I can't understand.
> For the instance fac2 I got a pretty good upper bound after 2 hours compared to
> the best known value from the MINLPLib. But the dual bound is very bad, so I got
> a Gap of 15455.65 %.
> Does anyone know how this can happen?
> Best, Ina
>
>
>
> _______________________________________________
> Scip mailing list
> Scip at zib.de
> http://listserv.zib.de/mailman/listinfo/scip
>