Cannot Train or Talk to bot through Rasa-X local machine

[Deprecated] Rasa X Community Edition
1

Hello There, I have two conda venv with specification

rasa                    2.8.25
rasa-sdk                2.8.4
rasa-x                  1.0.1

and

rasa                         3.1.0
rasa-sdk                     3.1.1
rasa-x                       1.1.0

running locally on my machine i cannot talk to bot via rasa-x web interface, I can train the bot and models are generated. but i always get error 500 interal server error

3.9fcf71a1.chunk.js:2          GET http://localhost:5002/api/projects/default/models?limit=50&offset=0 500 (Internal Server Error)

and also sometimes i notice this in the debug output

It's possible that no model has been trained, or that the requested tag hasn't been assigned.
Checking the model discovery from model_for_tag

can you please help me.

2

ok thanks for the gentleman who did fix my original post. After fiddling with too many version of rasa, rasa-x and checking Compatibility Matrix.

I have used nik202 Steps:

Now i was able to launch rasa x and talk to the bot but please give it sometime when you are loading and tagging models as active , it’s slow in here

venv config running everything in local machine Rasa Version : 2.8.1 Minimum Compatible Version: 2.8.0 Rasa SDK Version : 2.8.1 Rasa X Version : 0.39.3 Python Version : 3.8.13 Operating System : Windows-10-10.0.19043-SP0 Python Path : E:\anaconda3\envs\rasa2.8.1\python.exe

3

@abdelrahman.fathy great my post helped you, good luck for your project.

1 Like
4

Dear nik202, Have you tried anything with rasa version 3.0.0 or higher, so that you can recommend anything like your older post?

5

@abdelrahman.fathy personally not, but I highly recommend using rasa 3.x with botfront widget as it will give you live chatbot experience and rasa x are good for testing, etc and you cant’ use custom actions in rasa x this version else you need server deployment using docker, etc.