I have one that could show a complete implementation of an algorithm and that would be useful. But it needs serious work before being published on the web site, like writing more comments, removing my "special" optimizations, etc...

Oh nothing amazing in the algorithm: it's an implementation of a classical Newton-Raphson algorithm for fitting a logistic model in IR^n with a maximum likelihood.
But it uses noalias, tranposition, multiplication, square and non square matrices, inversion (which can be done through other librairies with the bindings), etc... With a few examples on how to use it, it can make a nice tutorial overall. I can even provide the web pages. Maths behind it are very simple: you draw a line tangent to a smooth curve, you follow it, and you start again until convergence :-D Well, it's more or less the idea.

Will work on it and be back with a full doc on that as soon as possible if that's ok for the documentation.

David

On Fri, Mar 5, 2010 at 00:10, Gunter Winkler <guwi17@gmx.de> wrote:
Hello,

has anyone some simple (but non-trivial) examples which can help to
improve the documentation:

https://svn.boost.org/trac/boost/ticket/3974

Some examples for bindings or external libs are welcome, too.

quote from task:

"... it [uBLAS doc] doesn't make a single reference regarding basic
functionality such as factorization, let alone how to solve a simple
system of linear equations, which happens to be the main purpose of
using such a component."

mfg
Gunter

_______________________________________________
ublas mailing list
ublas@lists.boost.org
http://lists.boost.org/mailman/listinfo.cgi/ublas
Sent to: david.bellot@gmail.com



--
David Bellot, PhD
david.bellot@gmail.com
http://david.bellot.free.fr