A collection of numerical analysis programs, written in GNU Guile.
This folder contains a collection of scalar minimization routines that are commonly found in textbooks. If you just wanted to find a local minimum efficiently, I think that the best one is Brent-Jarratt method (brent.scm).
This folder contains (part of an) implementation of a genetic optimiser. Check back later and I may have completed it. Right now I am not sure that I like the structure of the code, I had translated it from some Python code that I wrote years ago.
This folder contains a collection of scalar root finding routines that are commonly taught in courses on numerical methods. The best one is from the Boost library, called Toms748. I have not implemented it yet. Of the algorithms that remain, the Brent-Dekker (brent.scm) algorithm is the best. Most of the algorithms break up a real interval in order to find the root, and are not useful for finding complex roots. The Muller method can apparently be used for finding complex roots (muller.scm).