Suppose I have three intents, intent_A, intent_B and intent_C, and for each of these intents there is a corresponding entity, entity_A, entity_B and entity_C. I would like to extract all these entities and assign them to slots. For that purpose I am using FormAction. I have only included the following story (as the starting point) which I expected to work:
story_happy
- intent_A
- my_form
- form{“name”: “my_form”}
- form{“name”: null}
- utter_slots_values
However, the problem is that when I train the nlu and core model and run them the form is not activated by giving an input of intent_A which contains an entity_A (although both the intent and the entity are detected correctly). Here is the debug output:
2019-04-08 17:57:57 DEBUG rasa_core.policies.memoization - Current tracker state [None, None, None, {}, {‘entity_entity_A’: 1.0, ‘intent_intent_A’: 1.0, ‘prev_action_listen’: 1.0}]
2019-04-08 17:57:57 DEBUG rasa_core.policies.memoization - There is no memorised next action
2019-04-08 17:57:57 DEBUG rasa_core.policies.form_policy - There is no active form
2019-04-08 17:57:57 DEBUG rasa_core.policies.ensemble - Predicted next action using policy_1_FallbackPolicy
2019-04-08 17:57:57 DEBUG rasa_core.processor - Predicted next action ‘action_default_fallback’ with prob 1.00.
2019-04-08 17:57:57 DEBUG rasa_core.processor - Action ‘action_default_fallback’ ended with events ‘[‘UserUtteranceReverted()’]’
2019-04-08 17:57:57 DEBUG rasa_core.processor - Predicted next action ‘action_listen’ with prob 1.00.
2019-04-08 17:57:57 DEBUG rasa_core.processor - Action ‘action_listen’ ended with events ‘’
I don’t know what I am doing wrong? I guess there is something wrong with the story, but I can’t find what it is. It’s similar to the restaurant bot example in Rasa repository:
happy path
- request_restaurant
- restaurant_form
- form{“name”: “restaurant_form”}
- form{“name”: null}