PyBrain: Machine Learning Toolkit for Python

pybrain.org

53 pointsdurin424 comments

Wow, that looks good. Just what I need in fact, I can stop writing my own backprop FFN using NumPy now and use this, which means I can get on to the more interesting stuff quicker! Thanks durin42 for posting this!

Now I just need to find a decent RBF implementation and I'm set :p

If you're writing back propogation algorithms you need to check out "automatic differentiation" which gives it to you effectively for free.

http://searchyc.com/%2522automatic+differentiation%2522

Thanks, I saw the article on that a few days ago here and it looks really interesting.

Is it really right that I could just use that to automatically recreate the entire backprop algorithm? Is it likely to be reasonable in its performance?

Pass. I've read the papers, I've seen the demo, but I haven't implemented it myself. I also read it here, but I've seen it before. It looks like it should work, but YMMV.

As always, you might need to spend some time understanding what it does and how it works before relying on it.

Reply on news.ycombinator.com