Add numberbatch word embeddings. Fix #9

piskvorky / gensim-data

Data repository for pretrained NLP models and NLP corpora.

https://rare-technologies.com/new-api-for-pretrained-nlp-models-and-datasets-in-gensim/

GNU Lesser General Public License v2.1

965 stars 128 forks source link

Add numberbatch word embeddings. Fix #9 #10

Closed markroxor closed 6 years ago

markroxor commented 6 years ago

@menshikh-iv please tell me

[x] from where can I generate the checksum.
[x] Should I raise another PR in the gensim repo to include the init.py file associated with the model?
[x] Is the description correct or you want some amendments?

Associated issue - #9

markroxor commented 6 years ago

Also note that the model generates embeddings in the following format -

/c/en/absolute_value -0.0847 -0.1316 -0.0800 -0.0708 -0.2514 -0.1687 -...
/c/en/absolute_zero 0.0056 -0.0051 0.0332 -0.1525 -0.0955 -0.0902 0.07...
/c/en/absoluteless 0.2740 0.0718 0.1548 0.1118 -0.1669 -0.0216 -0.0508...
/c/en/absolutely 0.0065 -0.1813 0.0335 0.0991 -0.1123 0.0060 -0.0009 0...
/c/en/absolutely_convergent 0.3752 0.1087 -0.1299 -0.0796 -0.2753 -0.1...

Where to document this?

menshikh-iv commented 6 years ago

Thanks @markroxor