[Scip] Overly motivated presolver?

[Tobias Achterberg]

Okay. This means we have to dig deeper into the problem. If we are fast enough, then the 
bug fix could still make it into the new release.

Just to summarize: if you do

constraints/linear/maxprerounds = 0
presolving/dualfix/maxrounds = 0
presolving/probing/maxrounds = 0
presolving/implics/maxrounds = 0
presolving/inttobinary/maxrounds = 0
presolving/trivial/maxrounds = 0

the problem is gone, but if you do

constraints/linear/maxprerounds = 0
presolving/dualfix/maxrounds = 0
presolving/probing/maxrounds = 0
presolving/implics/maxrounds = 0
presolving/inttobinary/maxrounds = 0

(without the "trivial"), the problem is still there. You also mentioned that disabling 
"trivial" but enabling "probing" instead also produces the bug, but let's concentrate on 
"trivial" for now since it is much simpler than "probing".


So, what you should do is to use the above settings without the "trivial" deactivation, 
and add a

#define SCIP_DEBUG

to the very beginning of <your scip>/src/scip/presol_trivial.c.

You should also use SCIP in debug mode ("make OPT=dbg"). Maybe, an assert will be 
triggered that tells us what goes wrong.

Then, send the output to Timo and me (but not to the list, since our joint debugging 
session may not be that interesting for everyone).


Tobias

Julian Lamberty wrote:
> Sadly it does not solve the problem. When the probing or/and the trivial 
> presolver are activated, the variables disappear from the constraints.
> 
> Regards,
> Julian
> 
> Timo Berthold schrieb:
>> Hi.
>>
>> I attached a "very-close-to-next-release" version of SCIP and a 
>> updated version of SoPlex (which you need if you want to compile the 
>> new SCIP version with SoPlex as LP solver).
>> Please let me know if this fixes your problem.
>> And please don't spread around this inofficial version.
>>
>> Best regards, Timo
>>
>> Am Dienstag 23 September 2008 16:49:44 schrieb Julian Lamberty:
>>>> Do I understand it correctly, that if you disable all the presolvers I
>>>>
>>>> mentioned:
>>>>  >> constraints/linear/maxprerounds = 0
>>>>  >> presolving/dualfix/maxrounds = 0
>>>>  >> presolving/probing/maxrounds = 0
>>>>  >> presolving/implics/maxrounds = 0
>>>>  >> presolving/inttobinary/maxrounds = 0
>>>>  >> presolving/trivial/maxrounds = 0
>>>>
>>>> then it works fine, but if you reenable only the "trivial" presolver,
>>>> then the issue appears?
>>>>
>>>>
>>>> Timo just told me that your question raised some bell in his mind. So,
>>>> it may very well be that we have already fixed your bug in our current
>>>> development version. SCIP 1.1.0 will be released very soon, but if you
>>>> like, Timo can send you a prerelease immediately such that you can 
>>>> check
>>>> whether this solves the issue.
>>>>
>>>> Note, however, that SCIP 1.1.0 will have some API changes, so you will
>>>> also have to update your code (usually, it means that some additional
>>>> parameters are needed for some methods; just read the CHANGELOG).
>>>>
>>>>
>>>> Tobias
>>> It works fine until I reenable either the "trivial" or the "probing"
>>> presolver.
>>>
>>> I will check the changelog and talk to Moritz again concerning the new
>>> SCIP version. Would be great if this issue would be solved with the new
>>> version.
>>>
>>> Julian
>>
> 
> 
> ------------------------------------------------------------------------
> 
> _______________________________________________
> Scip mailing list
> Scip at zib.de
> http://listserv.zib.de/mailman/listinfo/scip