Problem in reading stories while creating my first bot

[apurv (Apurv Thapa)]

Hi, RASA newbie here. After going through the official playlist and rock paper scissor bot. I decided to create my own bot. It’s a pretty simple apartment hunting bot which asks the users about the state and number of bedrooms, then depending on the data provided the bot will return the apartment from my own excel file. The excel file have three columns : name, state and number of bedrooms. So the name that the bot returns will be based on state and number of bedrooms. Since the problem statement in clear I will share some screen shots and explain the problem I am facing.

These are my domain files

My End Points

my stories

My story graph

My chat log

So this is where the problem is happening. According to the story after ‘hi’(intent: greet), action: utter_greet should trigger but it is not happening. Similarly the next steps of the story are also not happening. I have been stuck here for weeks now. Please help.

Also since at the moment I am not processing any data, (I just want to make my story work) do I have to run my action server? for the stories to run.

[siriusraja (Raja Duraisingam)]

Hi @apurv

Can you share your training data - nlu.yml ?

On a different note, to read data from excel you might need to write custom action to extract the name.

[apurv (Apurv Thapa)]

version: “3.1” nlu:

• intent: greet examples: |

  • hey
  • hello
  • hi
  • hello there
  • good morning
  • good evening
  • moin
  • hey there
  • let’s go
  • hey dude
  • goodmorning
  • goodevening
  • good afternoon

• intent: goodbye examples: |

  • cu
  • good by
  • cee you later
  • good night
  • bye
  • goodbye
  • have a nice day
  • see you around
  • bye bye
  • see you later

• intent: affirm examples: |

  • yes
  • y
  • indeed
  • of course
  • that sounds good
  • correct

• intent: deny examples: |

  • no
  • n
  • never
  • I don’t think so
  • don’t like that
  • no way
  • not really

• intent: full_search_apartment examples: |

  • I am looking for 3bhk in delhi.
  • I am looking for 3bhk in mumbai.
  • I am looking for 3bhk in dehradun.

• intent: looking_for examples: |

  • I am searching for an apartment.
  • apartment.
  • looking for a place.
  • finding a place.
  • a place to live.
  • apartment

• intent: bot_challenge examples: |

  • are you a bot?
  • are you a human?
  • am I talking to a bot?
  • am I talking to a human?

• intent: location examples: |

  • delhi
  • mumbai
  • dehradun

• intent: room examples: |

  • 1
  • 2
  • 3
  • 4
  • 5
  • 6

[apurv (Apurv Thapa)]

hi @siriusraja , my stories are working fine now. I re-installed rasa on a new virtual environment and changed the config files to default values. Which seems to work for now. Thanks for replying though.

[siriusraja (Raja Duraisingam)]

Glad that its working now

[apurv (Apurv Thapa)]

Hi @siriusraja , sorry to bother you but can you help me with a problem that i have been facing with actions.py

This being my action file

this being my chat log

as you can see by action.py is running till this chuck of code

meaning it is able to capture entities and slots.

But for every value of ‘location’ and ‘bedroom’ it is returning ‘Sorry, we couldn’t find an address that matches your criteria.’

This is how my excel file looks

[siriusraja (Raja Duraisingam)]

@apurv

I suspect some issue on the code where you read the excel file and apply the filter condition on the dataframe. Can you print and see if those lines are extracting the data from excel.

Hint:

1. Check if the datatype of num_bedrooms is int or same as what is read from the excel.
2. Strip any trailing whitespace in the values you compare.

[Janeer (Jane Wilson)]

If the stories are poorly structured, inconsistent or irrelevant to the chatbot’s intended purpose, it may struggle to provide accurate and helpful responses. Additionally, a lack of diversity in the training data could also limit the chatbot’s ability to understand and interpret different types of user input, leading to inaccuracies and misinterpretations. Regards, Jane, manager of productivity system (description)

[apurv (Apurv Thapa)]

Thanks this works now.

[apurv (Apurv Thapa)]

Thanks, as you mentioned the problem was for wrong data type.