NLopt
From AbInitio
(Difference between revisions)
| Revision as of 22:13, 1 September 2008 (edit) Stevenj (Talk | contribs) ← Previous diff |
Revision as of 22:18, 1 September 2008 (edit) Stevenj (Talk | contribs) Next diff → |
||
| Line 1: | Line 1: | ||
| - | '''NLopt''' is a free/open-source nonlinear optimization library that we have been using internally, and plan to release by the end of 2008. It provides a common interface for a large number of different optimization algorithms, both by other authors (where available as free software) and original implementations by us, and is designed to make it easy to switch between different algorithms to see which ones work best for your problem. It includes both local and global optimization algorithms, for functions either with or without derivatives supplied, and supports unconstrained, box-constrained, and nonlinear-inequality constrained problems. | + | '''NLopt''' is a free/open-source nonlinear optimization library that we have been using internally, and plan to release by the end of 2008. It provides a common interface for a number of different optimization algorithms, both by other authors (where available as free software) and original implementations by us, and is designed to make it easy to switch between different algorithms to see which ones work best for your problem. It includes both local and global optimization algorithms, for functions either with or without derivatives supplied, and supports unconstrained, box-constrained, and nonlinear-inequality constrained problems. |
| [[User:Stevenj|—Steven G. Johnson]] 18:13, 1 September 2008 (EDT) | [[User:Stevenj|—Steven G. Johnson]] 18:13, 1 September 2008 (EDT) | ||
Revision as of 22:18, 1 September 2008
NLopt is a free/open-source nonlinear optimization library that we have been using internally, and plan to release by the end of 2008. It provides a common interface for a number of different optimization algorithms, both by other authors (where available as free software) and original implementations by us, and is designed to make it easy to switch between different algorithms to see which ones work best for your problem. It includes both local and global optimization algorithms, for functions either with or without derivatives supplied, and supports unconstrained, box-constrained, and nonlinear-inequality constrained problems.
—Steven G. Johnson 18:13, 1 September 2008 (EDT)
