Introduction

OPT++ is an object-oriented nonlinear optimization package. It solves optimization problems of the form

in which the user specifies the function f (and, when available, its first and second analytical derivatives) and the functions h and g. OPT++ provides many different solution algorithms including:

1. various Newton methods;
2. a parallel Newton method;
3. a nonlinear conjugate gradient method
4. a parallel direct search method
5. a nonlinear interior point method.

Users

OPT++ can be used to solve general nonlinear constrained optimization problems as defined above. The image in the left panel comes from an energy minimization problem for the protein T162, from CASP5 (click on the icon to display the full image in a new window). The initial configuration was created with ProteinShop and the minimization was computed with OPT++/LBFGS.

Documentation

• OPT++: An Object-Oriented Class Library for Nonlinear Optimization 
• OPT++  website

• Related Tools:

• APPSPACK 
• DAKOTA

Developers

OPT++ was developed at the the Computational Sciences and Mathematics Research Department at Sandia National Laboratories. The original author was Juan Meza with substantial contributions by Patty Hough and Pamela Williams