Hi, Could you please list down the exact steps to train MPLBERT on a new language? How to preprocess? What the tokens_map.pkl should look like? Any other training configurations that needs to be taken care of?
· Sign up or log in to comment
