Using Rasa Core with my own nlu model

Hi everyone,

I’m trying to use Rasa Core only. The reasons are simples, I already have a model for intent prediction stored in server (mlflow) so I don’t want to retrain it. I would like to know if there is a clean way to use Rasa core only to handle the dialog with the polycies and other What I’m doing for now is predict the messages with my model and send the prediction with the HTTP API, like this :


With a json containing the predicted sentence, its intent and entities

But then if I want to generate an answer from the bot I have to use trigger_intent

But I can’t trigger any actionsForm with this solution

Is there a better way to use my own pretrained nlu service with the Rasa core ? If there is no other way, how can I trigger the actions from HTTP API ?

Thanks for reading,


My bad I did some mystakes it is possible so no problem anymore

@aloizel ,Can you please let me know how you solved this issue