Sean Massung

Yes, it is. :)

Thanks! I will merge a PR that fixes this.

:+1:

Thanks for the feedback! For the Naive Bayes catch, yes you are definitely right. I forgot to pull _p(y)_ out of that statement somehow. I've fixed it and will release...

Might be a way to do this with existing filters?

Can this be created using shallow parsing with the CRF?

Once we implement this, we should have some way of saving the tokenization setup to ensure queries using the LM are tokenized the same way.

See if we can store `context -> (id, prob), (id, prob), ...` in the forward index format.

Thanks! We have another docker image in our contrib/ directory, but I don't see why you couldn't add yours as well. I would just distinguish it from the other one...

(also see #6 )