[R] A combinatorial optimization problem: finding the best permutation of a complex vector

Gottlieb, Neil ngottlieb at marinercapital.com
Wed Nov 18 16:11:23 CET 2009


Ravi:


Cplex use to be an independent firm. Was acquired by Ilog which was acquired by IBM.
It has a nice callable API which is written in C++. It is a very good product.

The other optimizer you might want to look at that performs fast is FortMP which might
provide source code also. I last re-call it was written in Fortran
And they had some kind of academic pricing.

http://www.optirisk-systems.com/default.asp

Cplex and FortMp performance was very close both are good. I have
used both as standalone program and via the API calls.

Another less expensive Optimizer is Loqo from Princeton.

http://www.princeton.edu/~rvdb/loqo/LOQO.html

The optimizer to select depends on the model trying to solve.
I suggest you look at Optimization Software Guide by Jorge J. Moré and Stephen J. Wright
Available from SIAM:
http://www.ec-securehost.com/SIAM/FR14.html

Hope this detail helps!

Neil Gottlieb


-----Original Message-----
From: r-help-bounces at r-project.org [mailto:r-help-bounces at r-project.org] On Behalf Of Ravi Varadhan
Sent: Wednesday, November 18, 2009 9:39 AM
To: 'Erwin Kalvelagen'; r-help at stat.math.ethz.ch
Subject: Re: [R] A combinatorial optimization problem: finding the best permutation of a complex vector

Hi Erwin,

Thank you for the information about Cplex.  It seems quite impressive.  Is
it a proprietary software?  I saw that there is a Matlab interface to it.
Is there an R interface?


Thanks,
Ravi.

----------------------------------------------------------------------------
-------

Ravi Varadhan, Ph.D.

Assistant Professor, The Center on Aging and Health

Division of Geriatric Medicine and Gerontology 

Johns Hopkins University

Ph: (410) 502-2619

Fax: (410) 614-9625

Email: rvaradhan at jhmi.edu

Webpage:
http://www.jhsph.edu/agingandhealth/People/Faculty_personal_pages/Varadhan.h
tml



----------------------------------------------------------------------------
--------


-----Original Message-----
From: r-help-bounces at r-project.org [mailto:r-help-bounces at r-project.org] On
Behalf Of Erwin Kalvelagen
Sent: Wednesday, November 18, 2009 12:20 AM
To: r-help at stat.math.ethz.ch
Subject: Re: [R] A combinatorial optimization problem: finding the best
permutation of a complex vector

Ravi Varadhan <rvaradhan <at> jhmi.edu> writes:
> 
> 
> When I increased N = 1000, the time was about 1400 seconds!  
> 

Not sure of this is important for you: This can be solved much faster. A
good 
solver can solve the n=1000 problem in less than 2 seconds. The Cplex
network 
code shows:

Network - Optimal:  Objective =   1.6173194067e+003
Network time =    1.58 sec.  Iterations = 209126 (102313)

Even solved as an LP this takes about 150 seconds.

(The solutions are the same as reported by solve_LSAP).


----------------------------------------------------------------
Erwin Kalvelagen
Amsterdam Optimization Modeling Group
erwin at amsterdamoptimization.com
http://amsterdamoptimization.com

______________________________________________
R-help at r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.

______________________________________________
R-help at r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.




More information about the R-help mailing list