search

aboSamoor / polyglot

Multilingual text (NLP) processing toolkit
http://polyglot-nlp.com
Other
2.3k stars 337 forks source link

Transliteration error #150

Open matanox opened 6 years ago

matanox commented 6 years ago

Code like this:

blob = u"""גאבאפנטין"""
text = Text(blob)

for x in text.transliterate("en"):
  print(x)

ValueError: Package 'transliteration2.iw' not found in index

If we change the input string to one in e.g. Arabic, a slightly different error message:

ValueError: Package 'transliteration2.sd' not found in index

but there seems to be no such packages to download, or it is unclear how to obtain them. Other transliteration2 packages download just fine and do facilitate transliteration from English to the downloaded target language. Is transliteration to English supported?

matanox commented 6 years ago

Well, based on the quoted research article, I think the answer may be no. On the other hand, the nature of the errors may hint that they are coming from incorrect source language identification.

bhx98 commented 2 years ago

https://polyglot.readthedocs.io/en/latest/Download.html download this model for me makes it work!

marksworn commented 1 year ago

I downloaded the .he file and manually changed the filenames to .iw - worked with no issue