Dan Ofer

How can I extract the vectors (for individual words)? I.E, given a new text corpus, I want to transform each individual word (given its context) into the vector space representation,...

Quora released a dataset of question pairs. (Albeit in CSV format). It would be a nice example to show the distance similarity metrics and NER also. (And to get traffic,...

I've definelty seen the spacy post (and it's sequel). I've been trying to use textacy on the challenge, though it's tricky given the format and how feature engineering works here..

This issue persists when trying to get text features with Ranking models. (And it does not appear anywhere in the documentation)

In the above, how would we get the topk predictions per user? i.e for the last item in a sequence, I would want the top k predictions, not just the...

Any update on this? :)

+1 - It's mentioned but seems to be missing from the repo. (I was looking for it due to an apparent bug with the sequence evaluation functions, and wanted to...

Has anyone actually seen an improvement on real data using it? i.e is the lightFM implementation possibly broken entirely?

Same issue, latest version of mids and sklearn, discrete (0/1) target. Input features are a mix of continous and discrete, after scaling and nan imputation. Happens when using any of...

Is this the same speed benchmark as the front poage for xgboost or lightgbm (r packages)?