Parallel training functionality

[gunthercox]

Using python's multiprocessing, large amounts of training data can be applied to a ChatterBot instance in parallel as long as the database being used with that chat bot supports concurrent writes.

The goal would to be to create a new training class that will split a large file into multiple files, and then train the chat bot using multiple parallel processes.

[vkosuri]

Some links that might be useful https://pythonhosted.org/joblib/why.html, https://github.com/joblib/joblib

Spacy has done things using joblib https://spacy.io/usage/processing-pipelines#multithreading