Friday, June 1, 2012

OOQP


OOQP is an object-oriented C++ package, based on a primal-dual interior-point method, for solving convex quadratic programming problems (QPs). It contains code that can be used "out of the box" to solve a variety of structured QPs, including general sparse QPs, QPs arising from support vector machines, Huber regression problems, and QPs with bound constraints.
OOQP also can be used as a framework can be used to design efficient solvers for new classes of structured QPs. Its design allows for easy substitution of the linear algebra modules, allowing different standard linear algebra packages to be tried.

Updated 3/9/2008

Authors

Links

  • PCx: Interior-point code for linear programming.
  • Online documentation for HCL: The Hilbert Class Library (Mark Gockenbach and Bill Symes).
  • Overview of rSQP++: ReducedSpaceSQP (postscript) (Roscoe Bartlett). An object-oriented package implementing SQP methods for nonlinear programming.
  • PETSc: Portable, Extensible Toolkit for Scientific Computation (Satish Balay, Lois Curfman McInnes, Bill Gropp, Barry Smith). Tools for linear algebra and PDE calculations on advanced computer architectures.
  • HSL Archive, which contains the MA27 and MA57 codes that are used in different implementations of OOQP to solve the symmetric indefinite systems that arise at each interior-point iteration.

Downloads

If you would like a copy of OOQP, go to the OOQP Request Form and fill it out. This will give you access to the OOQP download page. Having downloaded the tarball, consult the file INSTALL in the OOQP directory. The contents of this file are also listed here.
The file OOQP/doc/index.html in your unpacked distribution contains links to comprehensive information on installation of the various instantiations of OOQP. Point your browser to this file by typing an entry of the following form in the navigator URL window: 
file:/complete-path-to-directory/OOQP/doc/index.html

Updates

See the log of changes for the various releases.

Auxiliary Software

If you wish to use OOQP's solvers for general sparse QP, you will need to obtain a copy of MA27 from the HSL Archive before proceeding. Complete instructions are given in the INSTALL file for OOQP. If you wish to install the Matlab interface to OOQP's solvers, you may run into compatibility problems with Mex, Matlab, and your operating system. We have prepared precompiled Mex files for operating systems such as Solaris and Linux, which you can obtain by mailing us .
(Before these files will be sent to you, you are required to download MA27 from the HSL Archive site; OOQP's authors will be notified when you have done so.)

Copyright

OOQP is copyrighted to the University of Chicago. Here is the Copyright Statement.

Documentation

Acknowledgments

Additional coding and maintenance for releases 0.99.10 through 0.99.21 was performed by Vanitha Suresh. We acknowledge the contributions of various users who have reported bugs.

Support

Development of OOQP was supported by
  • The Department of Energy, Office of Advanced Scientific Computing Research
  • The National Science Foundation Grants CDA-9726385 and ACI-0082065, and 0196485. Any opinions, findings, and conclusions or recommendations expressed in this material are those of the authors and do not necessarily reflect the views of the National Science Foundation.
Posted by at

No comments:

Post a Comment

Subscribe to: Post Comments (Atom)