Hi, I want to ask for a question: I’m deploying a Rasa chatbot on a server, and I want to put it on production. The problem is that, say, I want to update my Rasa-core model because I’ve added a new intent or slot. Then, I have to retrain, shut down my server and run it again, but when it’s off I’m without chatbot to answer to the clients. Is there any way to handle this? I was thinking in a online model updating or in two parallel servers, where first one of them it’s updated (and the queries are sent to the other one) and then the second one is updated, but in this case I would need to share the tracker between the two servers.
Could anyone help me please?