👋 Introducing the Rasa Playground

Building contextual assistants with Rasa Forms

Check out our new tutorial on Rasa Forms! Learn how to use them to handle all the happy paths with just one training story! Have you tried Rasa Forms already? Share you experience with us in a thread below!

post by Juste on Jan 10, 2019

I had created a video tutorial based on the @Juste Article which shows how to set up the bot and make it running:

Code Snippet

* request_restaurant
    - restaurant_form
    - form{"name": "restaurant_form"}
    - form{"name": null}
    - utter_slots_values

post by srikarplus on Jan 18, 2019

Hi @Juste,

In the below code snippet

• request_restaurant - restaurant_form - form{"name": "restaurant_form"} - form{"name": null} - utter_slots_values

Could you please explain the use of form{"name": "restaurant_form"} and also form{"name": null}?

Thanks.

post by akshay27 on Jan 22, 2019

Hi JiteshGaikwad can I integrate rasa with whatsapp? If yes then please provide solution.

post by JiteshGaikwad on Jan 22, 2019

I haven’t integrated the rasa bot with Whatsapp yet, but what I know you can do so using Twilio. for example you can relate with this : Twilio Blog

post by rasa5 on Feb 5, 2019

Hi @JiteshGaikwad and @Juste

I am getting this error when I train for rasa_core: Can not access action ‘restaurant_form’, as that name is not a registered action for this domain

Do I need to add “restaurant_form” under -actions as well?

post by Juste on Feb 5, 2019

Hello @rasa5. I couldn’t reproduce the error. Any chance you made some changes to the domain? Also, are you getting the following error when running make train-core?

post by rasa5 on Feb 6, 2019

Hi @Juste @juste_petr

“make” command doesn’t run on my cmd, so I ran 2 commands:

1. for train-nlu (which worked perfectly and the NLU model was trained)

   python -m rasa_nlu.train -c nlu_tensorflow.yml --fixed_model_name current --data data/nlu_data.md -o models --project nlu --verbose
   

   2. train-core(which displayed this error):

   python -m rasa_core.train -s data/stories.md -d domain.yml -o models/dialogue --debug
   

   Also, I am using rasa-core 0.10.4 rasa-core-sdk 0.12.1 rasa-nlu 0.14.3

post by Juste on Feb 6, 2019

Ok, so you are using an old version of Rasa Core. This new Form Action is a new addition to the stack so I would suggest you upgrading your Rasa Core to at least 0.12.0

post by rasa5 on Feb 11, 2019

Hi @Juste

I installed the latest version and implemented form action in my own code. I have one required slot and the bot asks for it, but after receiving the response my bot doesn’t execute submit(). Instead it tries to find a suitable intent for the response.

post by Juste on Feb 14, 2019

Hi @rasa5. Can you share how your submit function looks like? Also, if your submit function triggers an utterance, are you sure you have it included in your domain?

post by nsmanoj on Feb 21, 2019

docker run -p 8000:8000 rasa/duckling

How do I get this up in windows?

post by somlaychong on Feb 24, 2019

Could you please explain the use of form{"name": "restaurant_form"} and also form{"name": null}mantigames. Thank you

post by Juste on Feb 25, 2019

@somlaychong @srikarplus Apologies for a late reply! form{"name":"restaurant_form"} means that the form has been activated and that now it’s time for the FormPolicy to jump in. Once the form is activated, the code of the ActionForm will be used to check which required slots are populated and which ones to ask for next. form{"name":"null"} means the deactivation of the form and allows other policies in the configuration to take over the prediction of what happens in the dialogue next. The form can be deactivated in a happy path (when all slots are populated) or during the deviations from it (chitchat, etc).

post by rasa5 on Feb 28, 2019

Hi @Juste @juste_petr

You have created a separate intent as chitchat and in stories, you have defined different paths- happy, unhappy, stop and continue path etc.

However, in the paths that involve chitchat intent, I wish to give users the flexibility that they can talk about any other intent but I have about 10 different intents!

So, how do I produce stories that replace chitchat intent with the other 10 intents?

post by tobiwalter on Mar 3, 2019

Hi @Juste

When I train the model using the keras policy I get a bunch of warnings:

“Feature ‘active_form_restaurant_form’ could not be found in feature map.”

Why is that and is it hurting performance?

post by Serge on Mar 17, 2019

Hi @Juste. Thank you for your post. We went through your post as well We have a FormAction that has a postal code as an only slot in the domain file:

Domain File Configuration

forms:
  - store_search_form

slots:
  postal_code:
    type: unfeaturized
  stores:
    type: unfeaturized

post by Serge on Mar 17, 2019

And we have the convention-based utterance that FormAction needs to prompt the user for the only value it needs as per the required_slots method:

Required Slots Method

- utter_ask_postal_code:
  - text: "Please enter the Postal Code you wish to use to search for a Bell Store"
- utter_postal_code_slot_value:
  - text: "I am going to run a Google API search for The Source near:{postal_code}"

post by Serge on Apr 26, 2019

Hello @Serge. Apologies for taking so long to respond. Do you have an intent for providing info (in our Rasa examples we usually call it inform intent where the user simply provides a piece of information. I think the underlying issue is related to the NLU data, because a fallback action gets triggered where an actual intent should be predicted.

post by Juste on Apr 26, 2019

Yes, you can use the action_restart especially if you want all slots to be reset at the end of the conversation.

post by jackpotandy on Apr 8, 2020

Is it possible to create an input/submit window with Rasa Forms?

Many thanks for your response!

post by Juste on Apr 14, 2020

Hi @jackpotandy. Yes, you can use the action_restart especially if you want all slots to be reset at the end of the conversation.