Sry, excuse me.I would like to inquire about how to save model weights. In the newsgroup training, is it necessary to train 10 categories first and then proceed with incremental training? Or do you add ten categories to the existing BERT based on datasets_random?
Sry, excuse me.I would like to inquire about how to save model weights. In the newsgroup training, is it necessary to train 10 categories first and then proceed with incremental training? Or do you add ten categories to the existing BERT based on datasets_random?