Bag of, not words, but tricks!
This project contains a couple of "tricks" on tokens. It's a collection of tricks for sparse data that might be trained on a stream of data too.
While exploring these tricks was super fun, I do feel like there are plenty of better alternatives than the ideas I explore here. In the end, TfIDF + LogReg can be "fine" for a bunch of tasks that don't require embeddings.
And for embeddings ... there's embetter.
So I archived this repo. Bit of a shame, because I really liked the name of this package.