[SCIP] get initial feasible solution

[Gregor Hendel]

Dear Natalia,

the battery of primal SCIP heuristics consists of many heuristics from 
different categories that should usually find feasible solutions 
quickly. If you can construct a good initial solution with your 
problem-specific knowledge, you can create a solution and set solution 
values after the construction of the problem yourself.
All API methods to interact with primal solutions are grouped together 
here (http://scip.zib.de/doc/html/group__PublicSolutionMethods.php) in 
our doxygen documentation.

Writing your own primal heuristic plugin is explained in the How To 
Section (http://scip.zib.de/doc/html/HEUR.php) about Primal Heuristics. 
This is slightly more involved, but may be beneficial if you need to 
consider additional information like the LP relaxation solution for the 
assignment. Some of the coding examples implement their problem specific 
heuristics.
Please refer to those as well as the c-files of the default primal 
heuristics (src/scip/heur_*.c) for reference.

Best,
Gregor




Am 24.04.2017 um 01:50 schrieb Natalia Perina:
> Hi SCIP users,
>
> What is the best way to just get an initial feasible solution for a 
> given scip instance?
> I have seen several heuristics, but am on clear on how to implement it 
> to get a feasible solution before solving for optimization.
>
>
> Thanks,
> Natalia
>
> -- 
> Natalia Perina
> Princeton University l 2017
> nperina at princeton.edu <mailto:nperina at princeton.edu>
> 202 360 9536
>
>
> _______________________________________________
> Scip mailing list
> Scip at zib.de
> https://listserv.zib.de/mailman/listinfo/scip

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://listserv.zib.de/pipermail/scip/attachments/20170424/c57c0a02/attachment.html>