OPTPACK

Unconstrained optimization and nonlinear constrained optimization with special software to handle bound constraints, linear equality constraints, and general nonlinear constraints

Unconstrained optimization is performed using the conjugate gradient algorithm. Constrained optimization is performed using a new scheme that combines multiplier methods with preconditioning and linearization techniques to accelerate convergence.

The software is written in double precision Fortran. The code is documented by internal comments. Research reports providing the theoretical basis for the algorithms are available on request. User feedback is much appreciated.

Need more info?

See the OPTPACK home page.

The conjugate gradient software can also obtained from netlib (send cg from napack). If you have trouble linking to the OPTPACK home page, you can obtain the constrained optimization software from

Professor William W. Hager
Department of Mathematics 
University of Florida 
Gainesville, FL 32611 
Phone: (904) 392-0286 
hager@math.ufl.edu

References:

W. W. Hager, Analysis and implementation of a dual algorithm for constrained optimization, Research Report, Department of Mathematics, University of Florida, Gainesville, Florida, 1990, to appear in J. Optim. Theory Appl. 79 (1993).

[ Optimization Software Guide | OTC Home Page | NEOS Server | NEOS Guide ]