[Symphony] warm_solve getting stuck?

Menal Guzelsoy megb at lehigh.edu
Thu May 6 00:39:25 EDT 2010


Ross,

Could you send me the test file and let me know which version you are using?

Menal.

Menal Guzelsoy
1610 4620455



On Thu, May 6, 2010 at 12:22 AM, Ross Anderson <rpanderson at gmail.com> wrote:
> Hi all,
>
> I'm attempting a warm start on a large MIP. I'm having trouble
> creating a minimal example, so for now I'll keep this general and hope
> there's an easy fix.
> The original solution works fine. I call sym_solve(env); with options
> keep_warm_start=TRUE
> node_limit=8000
> do_reduced_cost_fixing=0
>
> and I get a solution.
> Then I reset some params
>
> sym_set_int_param(env, "do_reduced_cost_fixing", 0); //this one is
> repeated so that I don't receive SYMPHONY can not resolve for the obj
> coeff change when reduced cost fixing is on
> sym_set_int_param(env, "node_limit", -1);
>
> I reset some obj coeffs,
> sym_set_obj_coeff(env, 142, -1);
>
> and resolve
> sym_warm_solve(env);
>
> Screen output starts repeatedly saying
> done: 10639 left: 2 ub: -886.16 lb: -886.15 gap: 0.00 time: 820
> ....The time increments each line, but "done", "left", "ub", "lb", and
> "gap" don't change. This happens until I kill the process.
>
> If I instead begin the problem with the 142'th variable objective
> coefficient set to -1 (ie. without the warm start), everything works
> fine.
>
>
> Why is this getting stuck?
>
> Thank you
>
> _______________________________________________
> Symphony mailing list
> Symphony at list.coin-or.org
> http://list.coin-or.org/mailman/listinfo/symphony
>




More information about the Symphony mailing list