Hi All,
I'm doing a Machine Learning project for a paper at Auckland University
(NZ).
We require a Bayesian/standard back-propagation neural network.
Rather than re-inventing the wheel, and doing what has already been done
better by others, we would like to use other peoples source code if
possible.
The project is purely for academic research, and credit will be given to the
origin author(s) (though the research isn't likely to result in a published
document).
Our requirements are fairly basic: we don't anticipate using more than 7000
traing examples - but ARE interested in convergence time - so the code
doesn't need to run fast, but should converge quickly (if you see what I
mean).
We're working in Java/C/C++.
If you have any suggestions please contact me directly, and I will post
summary to the group. mailto:nick@1site.co.nz (or, at Uni.
mailto:nhyn001@ec.auckland.ac.nz)
=)
Cheers,
Nick Hynes.
This archive was generated by hypermail 2b29 : Sun May 13 2001 - 21:28:43 PDT