[SCIP] Question about long time solving using SoPlex with GMP

[Leon Eifler]

Hello Feiran,

I guess it is probably the same problem as the last time you asked the 
same question. You relaxed the floating point tolerances to 1e-3. Is 
there a specific reason for this? I suggest you try again with default 
values for fpopttol and fpfeastol.

Best,
Leon

On 23.08.19 23:54, Feiran Li wrote:
>
> Hi,
>
> I am a new user for soplex4.0.0 and GMP. I was trying to solve an LP 
> file (see google drive 
> https://drive.google.com/open?id=10GjuxsKvz6gcIQ61HXNyfBFVmEDviHCY ) 
> with coefficients up to 15 significant digits by using the code below:
>
> /tools/build/bin/soplex -s0 -g5 -f1e-10 -o1e-10 -x -q -c --readmode=1 
> --solvemode=2 --int:checkmode=2 --real:fpfeastol=1e-3 
> --real:fpopttol=1e-3  test > test.lp.out
>
> Currently this code takes more than 2hour and generate a huge out file 
> during the calculation. I attached an out file here as *test.lp.out*. 
> Could you help me to see where is the problem and what should I do to 
> fast the progress but keep the precision higher than 10^-11 . Many thanks!
>
> Best,
>
> Feiran
>
>
> _______________________________________________
> 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/20190826/ee356554/attachment.html>