aboSamoor / pycld2

Apache License 2.0
165 stars 63 forks source link

Updated deprecated language code for Hebrew from 'iw' to 'he'. #47

Open dudio92 opened 2 years ago

dudio92 commented 2 years ago

Hi,

When using the model, lang detection for Hebrew returns language code "iw".

According to ISO 639-1, this language code is deprecated and now the standard is "he".

langdetect, and fasttext correctly returns 'he' for the same input.

Reference: http://www.mathguide.de/info/tools/languagecode.html

I see that some issue was already opened here - https://github.com/aboSamoor/polyglot/issues/64