I have a defined user story that follows this template : -
## story1
* intent1{'entity':'value1'}
- action_1
- utter_was_that_useful
* affirm
- utter_goodbye
- action_reset_slot
* deny
- utter_take_action
## story2
* intent2{'entity':'value2'}
- action_2
- utter_was_that_useful
* affirm
- utter_goodbye
- action_reset_slot
* deny
- utter_take_action
I have entity labeled training examples for all the intents in the nlu.md file, and am using Memoization/Keras policy with max history set to 2 and augmentation factor set to 0.
However, the core consistently triggers the wrong action after correctly identifying the intent, and completely abandons the story in some cases ( triggering goodbye after the deny intent, despite that chain not being a part of any story). I suspect that has something to do with the way I have defined the story. Can someone please help me figure this out? Thanks
I have consulted this solution, but it made no difference in the behavior of my bot.