When creating a new rasa assistant with rasa init I get the interface below. What command can I run from the command line to get back to this interface after I exit? Sometimes I’d rather just use the cmd interface rather than running rasa x and using the browser.
? Do you want to speak to the trained assistant on the command line? 🤖 Yes
2020-11-03 10:01:30 INFO root - Connecting to channel 'cmdline' which was specified by the '--connector' argument. Any other channels will be ignored. To connect to all given channels, omit the '--connector' argument.
2020-11-03 10:01:30 INFO root - Starting Rasa server on http://localhost:5005
2020-11-03 10:01:32 INFO root - Rasa server is up and running.
Bot loaded. Type a message and press enter (use '/stop' to exit):
Your input -> hi
Hey! How are you?
Your input -> good
Great, carry on!
Your input -> I am sad
Great, carry on!
Your input -> hi
Hey! How are you?
Your input -> bad
Here is something to cheer you up:
Image: https://i.imgur.com/nGF1K8f.jpg
Did that help you?
Your input -> /stop
2020-11-03 10:02:38 INFO root - Killing Sanic server now.
Thanks for the response @tyd! When I do rasa shell here is what I see:
(base) jmlockl2@C02ZM744MD6M remote_gig % source activate rasa
(rasa) jmlockl2@C02ZM744MD6M remote_gig % rasa shell
NLU model loaded. Type a message and press enter to parse it.
Next message:
hi
{
"intent": {
"name": "greet",
"confidence": 0.9994694590568542
},
"entities": [],
"intent_ranking": [
{
"name": "greet",
"confidence": 0.9994694590568542
},
{
...
I want the chat/response with the assistant, not the intents/entities output. I do see a note in the docs that says:
If you start the shell with an NLU-only model, rasa shell will output the intents and entities predicted for any message you enter.
@tyd Ah! Thank you…I had not made the connection between how the shell behaves and rasa train vs. rasa train nlu. I’ll do some more reading up on that so I better understand it.
When I did rasa train then ran rasa shell I was able to converse with the assistant.