Artificial Neural Network example at trapexit.org - any successes?

[Michael Turner]

http://www.trapexit.org/Erlang_and_Neural_Networks

The code isn't complete, and when I fill in the blanks as best I can --
well, I realized upon running it there doesn't seem to be any guarantee
that the network is finished building itself before you do any
operations on it.

I've written the original blog author for help, but no reply yet.

Does anyone have the complete code?  Has anyone reproduced the results
shown in the trapexit.org article?  Searching around, I find some
comments on smaller details (like whether the author implemented
dot-product efficiently or idiomatically) but nobody talking about
getting the whole thing working, successfully or not.

An article that contained code that actually worked would, I think,
redound somewhat more to the credit of Erlang.  I liked looking at the
pictures, though.

-michael turner