[Coin-discuss] problem with initialsolve

[Matthew Saltzman]

On Wed, 19 Apr 2006, John J Forrest wrote:

> Matt,
>
> We accept BV and UI.  Not sure we can accept SC as no OsiSolverInterface
> definition.  I would like to be able to accept S1 and S2 sets if we decide
> to extend to SC.

I didn't think there were any OSIs that accepted SC yet, but I spotted a 
string in a unitTest someplace that referred to it.  Couldn't find its 
origin, but I later realized it must be in the linked solver library.

>
> May get round to putting in LI if noone else does first.

OK Thanks.

>
> John
>
>
>
> Matthew Saltzman <mjs at ces.clemson.edu>
> Sent by: coin-discuss-bounces at list.coin-or.org
> 04/19/2006 08:47 AM
> Please respond to
> Discussions about open source software for Operations Research
> <coin-discuss at list.coin-or.org>
>
>
> To
> Discussions about open source software for Operations Research
> <coin-discuss at list.coin-or.org>
> cc
>
> Subject
> Re: [Coin-discuss] problem with initialsolve
>
>
>
>
>
>
> On Wed, 19 Apr 2006, John J Forrest wrote:
>
>> Anureet,
>>
>> Cplex seems to accept a non standard MPS format bound.  There are LI
> bounds
>> which I have never heard of (they obviously allow a lower bound of minus
>> infinity).  The Mps standard is MI and when I changed them to that then
> the
>> problem was feasible.
>>
>> CoinMpsIo should have given error messages on unrecognized bound types -
> I
>> think I have corrected that.
>
> These are part of CPLEX's and other solvers' integer extensions--an
> alternative to the MARKER format.
>
> LI is CPLEX-speak for a variable that must be integer and has the
> specified lower bound.  UI is integer with specified upper found, BV is
> binary, SC is semi-continuous with given upper bound (presumably the lower
>
> bound must be specified with an LB).
>
> I think we probably want to accept these extensions. There are other CPLEX
>
> extensions that may be more or less common that we may want to consider
> supporting in the future.
>
>
>                                 Matt
>
>>
>> John Forrest
>>
>>
>>
>>             Anureet Saxena
>>             <anureet at yahoo.co
>>             m> To
>>             Sent by:                  coin-discuss at list.coin-or.org,
>>             coin-discuss-boun         anureet at cmu.edu
>>             ces at list.coin-or. cc
>>             org
>> Subject
>>                                       [Coin-discuss] problem with
>>             04/11/06 12:50 PM         initialsolve
>>
>>
>>             Please respond to
>>             Discussions about
>>                open source
>>               software for
>>                Operations
>>                 Research
>>             <coin-discuss at lis
>>              t.coin-or.org>
>>
>>
>>
>>
>>
>>
>> hi,
>>
>> I am trying to solve a linear program by
>> OsiClpSolverInterface::initialSolve() which is known to be feasible and
>> bounded (cplex solves it to optimality). However, OsiClp reports the
> linear
>> program to be primal infeasible. The .mps file for the linear program
> and a
>> short code for verification can be found at,
>>
>> http://www.andrew.cmu.edu/user/anureets/COIN_tmp/workspace_my.tgz .
>>
>> It would be very helpful if someone could suggest what is going wrong
> and a
>> possible way to fix it.
>>
>> regards,
>> Anureet Saxena
>>
>>
>>
>>
>>
>>
>>
>> Anureet Saxena
>> anureet at cmu.edu
>> anureet at yahoo.com
>>
>> --This mail was delivered to you on 100% recyclable electrons.
>>
>>
>> Yahoo! Messenger with Voice. Make PC-to-Phone Calls to the US (and 30+
>> countries) for 2¢/min or less.
>> _______________________________________________
>> Coin-discuss mailing list
>> Coin-discuss at list.coin-or.org
>> http://list.coin-or.org/mailman/listinfo/coin-discuss
>>
>>
>>
>>
>>
>> _______________________________________________
>> Coin-discuss mailing list
>> Coin-discuss at list.coin-or.org
>> http://list.coin-or.org/mailman/listinfo/coin-discuss
>>
>
>

-- 
 		Matthew Saltzman

Clemson University Math Sciences
mjs AT clemson DOT edu
http://www.math.clemson.edu/~mjs