Hey, I just test rasa with a minimal and default example on the same laptop. Chatting over shell or a UI is rather slow, because of the action server I think. So without using any advanced stuff or model you get a large latency by just using the server.
In v1 of rasa before the action server this was not the case (if I remember correctly).
Is there a way to avoid this or just not using the action server and using just the plain actions.py?
Example. just printing out text from a text file sentence per sentence shows that each sentences in my ase needs several seconds just because of the http requests?
How are you handle this as this is really unpracticable…?