naver/sqlova

How to get pytorch_model_uncased for chinese?

Opened this issue · 2 comments

zyc1310517843 commented

I wanted to train Chinese-based datasets, but I found that I lacked the corresponding "pytorch_model_uncased_L-24_H-1024_A-16" file. How can I get it?

IgorMunizS commented

@zyc1310517843 you should use multilingual bert.
Btw, had you succeed in training with other language?

youssefmellah commented

Hi @zyc1310517843 @IgorMunizS

Can you tell me please how to train with another Language Model instead of BERT ? (Like mBERT or RoBERTa) ?

Thank you so much