[Coin-discuss] using CLP barrier with OSI

[John J Forrest]

Renaud,

I have modified OsiClpSolverInterface in trunk version so that you can 
pass in options which will be used in initialSolve.  These options are 
from ClpSolve and so are more general than just barrier, but I thought it 
could be useful.  So you can play around using stand-alone "clp" to see 
which options work and then transfer code to your driver.  There is a 
sample program "barrier.cpp" in Cbc/examples (I know that it is nothing to 
do with Cbc, but there is no Osi/examples directory). 

However the standard barrier in Clp is not good if you do not use a 
third-party ordering and factorization code.  One from the University of 
Florida is suitable.  If I get time next week I will try and make it 
easier to use with the svn version of Coin - It was working in the old cvs 
version.  This what you have to do - add to your configure

ADD_CXXFLAGS="-DUFL_BARRIER -I(directory with amd.h) -L(directory with 
libamd.a) -lamd

You will get mild warnings that -lamd is being ignored for compilation.

Then play with clp using -chol UFL

Barrier.cpp will also work but you have to add an extra -lamd in Makefile 
(before -lm).  As I said I will try and make prettier next week.

You might also try the dualCuts.cpp example in Clp/examples which might 
work well with a problem of the shape you describe.  If you then want to 
convert to Osi the function calls would be very similar.

John Forrest



"Renaud Lepere" <renaud.lepere at alma.fr> 
Sent by: coin-discuss-bounces at list.coin-or.org
10/16/2006 06:15 AM
Please respond to
Discussions about open source software for Operations Research 
<coin-discuss at list.coin-or.org>


To
<coin-discuss at list.coin-or.org>
cc

Subject
[Coin-discuss] using CLP barrier with OSI






I am using linear programing and COIN-OR (osi and clp) to optimize
nesting problems. I have quite a lot of variables (up to 1000) 
representing positions of nested parts and approximatly the same
order of magnitude of constaints (up to 5000). Those constraints
corresponds to the non-overlapping condition between nested parts.

I would like to solve those problems as fast as possible. I would like 
to know if a barrier algorihtm can speed up my computation time ; and
if it is possible to use the barrier algorithm included in CLP with OSI
?

Thanks

Renaud

_______________________________________________
Coin-discuss mailing list
Coin-discuss at list.coin-or.org
http://list.coin-or.org/mailman/listinfo/coin-discuss

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://list.coin-or.org/pipermail/coin-discuss/attachments/20061020/f46b8366/attachment.html>