[erlang-questions] Machine Learning

[Samuel]

>> What would be the best direction or tools to start with?
>
>
> I would start by writing a tensorflow program as something you can talk to
> via a port, should I start today. Tensorflow networks are configured and
> trained in Python, but can be run everywhere you can call into C. The nice
> idea is that you split network tinkering from actually using the NN you are
> generating.

I am not aware of existing ML or linear algrebra libraries in erlang
that can be used to quick start an ML project, but just piping into
tensorflow (or any other existing library/framework) isn't really
doing ML with erlang, is it? You can as well just use tensorflow
directly.

Unless the problem to solve is orchestrating a big, distributed system
that has some ML components in it; then doing the orchestration in
Erlang that way might be a good idea.

It still could be an interesting project to create a ML framework in
erlang itself, a quick google reveals some previous works on that, but
I haven't gone through any of them, so I'll let the Theepan do the
research :)