Cleanup and documentation

[adelespinasse]

This code is hard to follow. A high-level explanation in the README.md would be a big help, as would more docstrings, and I think there's a lot of dead code (I noticed duplicate code among several Python source files in the top-level directory, for example).

Maybe this repo wasn't really intended as a clear demonstration of an email classifier, but this blog post implied that it is.

Here's a more detailed write-up that covers the modeling aspects as well.
https://docs.google.com/document/d/1km1tkU-KyHderCeR-qCWV1uRd_D3JwoC1s4ZAIw2kXQ/edit

Hope this helps.

[adelespinasse]

Thanks... that document is private. I just requested access, but if it's the best writeup, maybe it should be world-readable?

[moyinho20]

Hi! I just requested for access to this document. Please grant :)

[adelespinasse]

I was never given access either. I think this project is dead, and the document was probably never intended for the public.