[Coin-symphony] Three questions

[Menal Guzelsoy]

Hi Francois,

1- Yes, It should be safe to increase the value of SYM_INFINITY. However, I am
not sure if it would cause a problem if assigned a value greater than 1e20
since the lp solvers SYMPHONY uses (at least CPLEX) assume this number to be
inf.

2- The answer is 'not yet' but that is one of the options we definitely 
consider
to implement for the future releases of SYMPHONY.

3- Did you try to set the verbosity level to -1?

Hope this helps.

Menal.

Quoting francois dionne <brainstorm at videotron.ca>:

> Dear mailing list,
>
> I have Symphony, downloaded using "CVS checkout" on 2005-08-17, and I 
> have a few questions.
>
> 1. I think that in SYMPHONY/include/BB_constants.h, SYM_INFINITY may 
> be too small. In one of my mixed integer problems, all of the 
> variables are much smaller. However, some of the largest coefficients 
> in the constraint matrix are bigger, around 192*2^26 but no more than 
> that. They are in the columns of continuous variables. Is it safe and 
> useful to increase the value of SYM_INFINITY ?
>
> 2. If I tell Symphony to keep a warm start, solve it and then tighten 
> the integer variables's upper and lower bounds, can Symphony use the 
> warm start information for a warm re-solve?
>
> 3. Finally, I appreciate it when Symphony tells me that a solution 
> has been found or that there is none, but is there a way to tell it 
> not to dump the whole solution vector on the screen? I think I use 
> the default verbosity, which would be 0 .
>
> That will be all. Thank you in advance and have a nice day.
>
>
>
> _______________________________________________
> Coin-symphony mailing list
> Coin-symphony at list.coin-or.org
> http://list.coin-or.org/mailman/listinfo/coin-symphony
>



----------------------------------------------------------------
This message was sent using IMP, the Internet Messaging Program.