Here are my plans for a major refactoring. Since I last wrote the repo, I have learned a lot more design patterns and best practices which I want to incorporate here.
Make everything a class/object (currently everything runs a standalone script)
Remove dependencies on saved pickles. While it is good to be able to save it, pickle can be unreliable and I unnecessarily slows down the processing.
Adding gensim-data support. Downloading glove vectors and the brown corpus can be tedious. All this is done by gensim-data and can be used out of the box.
Here are my plans for a major refactoring. Since I last wrote the repo, I have learned a lot more design patterns and best practices which I want to incorporate here.