Incorporate a chinese tokenizer into rasa packages

Dear sir/madam,

I would like to use a chinese tokenizer and incorporate into rasa_packages. May i know is there any way to incorporate into rasa_config file so I can train a rasa_nlu model?

Yes, you have to make sure it returns the correct format for Rasa Core. Actually this one i think allows loading a chinese tokenizer: rasa_nlu/jieba_tokenizer.py at master · RasaHQ/rasa_nlu · GitHub You then just have to specify it in your config file

thank you so much @akelad.