clustering on fasttext embeddings

[ar...@capiot.com]

Hi, trying to use fasttext for document classification. How do we convert the fasttext embeddings for document in a corpus to a pandas dataframe so as to apply some clustering algorithm on documents

[Matan Safriel]

Dude, they are just vectors :-)) if you choose fasttext to generate document embeddings and not some other algorithm/tool for document embeddings, I guess generate your vectors through the `print-sentence-vectors` option of fasttext, and just transform them into dataframe/s as you would transform any other data for a dataframe. Hopefully you won't memory overflow. Hope this helps!