[Coin-lpsolver] CLP Presolve

[John J Forrest]

Matt,

I have fixed on main branch (and on development branch where it is 
CoinPresolve....).  An array was not long enough for this odd shaped 
problem - 5833 rows and only 23 columns and 437 elements.

As to your questions:

1) CoinPresolve does not produce a proof of infeasibility - it just 
declares the problem infeasible.  Presumably it could produce a proof.  I 
suspect that the problem could be proved to be infeasible by presolve only 
in first few iterations.

2)  Which codes provide multiple rays?  If others do and people think it 
is a good idea then I can look into it.

John Forrest




Matthew Galati <magh at lehigh.edu>
Sent by: coin-lpsolver-admin at www-124.southbury.usf.ibm.com
07/28/2003 10:56 PM
 
        To:     coin-lpsolver at www-124.southbury.usf.ibm.com
        cc: 
        Subject:        [Coin-lpsolver] CLP Presolve


Hi CLP,

I have a series of LPs to solve which are really just feasibility 
problems (null objective) - it is solved in the context of a column 
generation routine - so what I need out of the LP solver are the 
dualRays to create new columns. Attached, I have an example that seems 
to work fine using dualSimplex. However, if I turn on presolve, it 
crashes inside resolve when calling the presolve routine.

I have attached an example MPS file. The following is how I invoke. If I 
comment out the setHintParam, it runs fine.

Two associated questions:
1.) Does/can presolve produce a proof of infeasibility (dual 
unboundeness)? Or even if it gets through resolve, will getDualRays fail?
2.) Is there a way to return more than 1 dual ray at a time? It is 
possible that many rays exist that prove dual unboundness. Does CLP 
provide a way to grab several of these? Currently 'getDualRays' is 
hard-coded to return just 1.

Thanks,
Matt G

 

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://list.coin-or.org/pipermail/clp/attachments/20030729/521aca69/attachment.html>