facebookresearch / LASER

Language-Agnostic SEntence Representations

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

An error initializing English pipeline (on the MLH-dev branch)

avidale opened this issue · comments

When I run the following code

from laser_encoders import LaserEncoderPipeline
laser2 = LaserEncoderPipeline(lang="eng")

I get an error:

KeyError                                  Traceback (most recent call last)
File ~/dev/LASER/laser_encoders/download_models.py:73, in LaserModelDownloader.get_language_code(self, language_list, lang)
     72 try:
---> 73     lang_3_4 = language_list[lang]
     74     if isinstance(lang_3_4, tuple):

KeyError: 'eng'

Expected behavior: loading the tokenizer and the model for LASER-2 (which is the default choice for English and 100-ish other languages).