7 episodes in the playlist and Rasa finally bothers telling us how to actually try and test everything they have been telling us about. Glad this video exists though.
If assistant is always replying with 'None' instead of Name, try using slots: name: type: text influence_conversation: true mappings: - type: from_entity entity: name intent: greet
@Rasa, What if I don't want to show the /moo_great in response when user click on great then it should be show as only Great not /mood_great what should I do to show that type of repsonse ?
In my case, the example at 6:28 didn't work for me. I ran "rasa data validate" and its said that the slot was missing a "mappings" key. I fix it doing: Slots: name: type: any mappings: - type: from_entity entity: name Anyway, I have no idea of what I'm doing haha. Just sharing in case you find the same problem.
@Rasa, i tried to use the name variable in my response to but the slot is not setting and I always get None. do I have to do anything else like setting pipeline ? or provide some example data ?
Hey there, I would like to ask, how I can make the chatbot give a different greeting answer depending on whether the user gives his name or not. I don't want the chatbot to reply like "Hey None" if the user doesn't give his name.
11:16 why did "Did that help you?" show up too? I thought each turn, 1 user question will only prompt 1 response? This looks like both utter_cheer_up and utter_did_that_help got triggered. Whats the logic to what response gets triggered?
you can trigger multiple actions (including responses) in a row if that's modelled in the training data or a rule. action_listen is what tells the bot to wait for a response and added automatically during training. you don't need to do anything manually but you can see where it shows up using --debug. -- Rachael
The unique thing is that you make this topic understandable for everybody. Many thanks.
7 episodes in the playlist and Rasa finally bothers telling us how to actually try and test everything they have been telling us about. Glad this video exists though.
There's something called "documentation" :P
If assistant is always replying with 'None' instead of Name, try using
slots:
name:
type: text
influence_conversation: true
mappings:
- type: from_entity
entity: name
intent: greet
i simply change the response texts ..and trained it its always giving the default answers.Can you help?
Thanks for awesome video
@Rasa, What if I don't want to show the /moo_great in response when user click on great then it should be show as only Great not /mood_great what should I do to show that type of repsonse ?
In my case, the example at 6:28 didn't work for me.
I ran "rasa data validate" and its said that the slot was missing a "mappings" key.
I fix it doing:
Slots:
name:
type: any
mappings:
- type: from_entity
entity: name
Anyway, I have no idea of what I'm doing haha. Just sharing in case you find the same problem.
I was also stuck there, thanks it worked also for me
thank you 🤍
Thank you so much
@Rasa, i tried to use the name variable in my response to but the slot is not setting and I always get None. do I have to do anything else like setting pipeline ? or provide some example data ?
i did everything you are saying but still my chatbot giving same answer of default chatbot.
Does the feature "Share your assistant with real testers" come with Rasa Open Source too, or it's just available for Rasa X?
You can can absolutely share with Rasa Open Source, it's just simpler with Rasa X. (Plus then you can look at & tag your user conversations easier)
@@RasaHQ thank you!
Hey there, I would like to ask, how I can make the chatbot give a different greeting answer depending on whether the user gives his name or not. I don't want the chatbot to reply like "Hey None" if the user doesn't give his name.
(Vincent here)
You may want to check out the conditional response feature: rasa.com/blog/rasa-conditional-response-variations/
@@RasaHQ I solved it by creating a custom action but thank you for this. Seems cool! Thank you for these helpful videos guys.
11:16 why did "Did that help you?" show up too? I thought each turn, 1 user question will only prompt 1 response? This looks like both utter_cheer_up and utter_did_that_help got triggered. Whats the logic to what response gets triggered?
you can trigger multiple actions (including responses) in a row if that's modelled in the training data or a rule. action_listen is what tells the bot to wait for a response and added automatically during training. you don't need to do anything manually but you can see where it shows up using --debug. -- Rachael
give the rasa x back !!!
6:47
Can I know your name😊❤
HELP
Why when I type another name (different than the name defined in square brackets in nlu.yml), the assistant call me None?
Maybe needs more examples to learn properly