Could you please add some models for Ukrainian language?

Open dchaplinsky opened this issue 2 years ago • 4 comments

Currently I know a few: https://huggingface.co/youscan/ukr-roberta-base https://huggingface.co/dbmdz/electra-base-ukrainian-cased-discriminator https://huggingface.co/lang-uk/electra-base-ukrainian-cased-discriminator https://huggingface.co/lang-uk/electra-base-ukrainian-dbmdz-vocab-cased-discriminator

Feb 20 '23 23:02 dchaplinsky

We need it to measure the performance of our GPT-2 model for the paper.

Feb 20 '23 23:02 dchaplinsky

I can see that WMT22 includes Ukrainian language: https://www.statmt.org/wmt22/translation-task.html

Feb 20 '23 23:02 dchaplinsky

hi @dchaplinsky , if you don't use the rescale_with_baseline feature, you should be able to load these models directly

Feb 21 '23 00:02 Tiiiger

Thanks for the information, I will try. Still is there a chance to make ukrainian models a first class citizens in the lib?

Feb 21 '23 07:02 dchaplinsky