Why my bot handle one conversation?

hajoura · July 23, 2019, 5:14pm

this my config file :

   language: en
      pipeline: supervised_embeddings
   policies:
    - name: FallbackPolicy
    - name: MemoizationPolicy
    - name: FormPolicy
    - name: MappingPolicy

I would to use form actions but not working at all one conversation

amn41 · July 25, 2019, 9:10am

Hi @hajoura - unfortunately I cannot help you with just the info you’ve provided.

what are you trying to achieve?
what commands are you running?
what is the expected output, and what do you get instead?

hajoura · July 27, 2019, 12:14pm

I would like to create hotel bot .the chatbot go to database and find the equivalent ticket for user and in my actions I defined two slots one for departure and destination but after one conversation the bot start to listen without any answers back

jnadaud · July 28, 2019, 1:30pm

I have exactly the same issue using forms, slots and mapping, it works once and then only listen without doing anything. I tried to run the formbot example and I have the same behavior. I tried chatting using rasa x or rasa shell, and I have only one conversation, after it just stop working. I have the latest rasa 1.1.8 with latest keras/tensorflow with GPU.

Here is a conversation, the one that works and then just listening:

Bot loaded. Type a message and press enter (use '/stop' to exit):
Your input ->  Hi
Hello! I am restaurant search assistant! How can I help?
Your input ->  I'm looking for a restaurant
what cuisine?
Your input ->  Italian
how many people?
Your input ->  3
do you want to seat outside?
Your input ->  No
please provide additional preferences
Your input ->
please give your feedback on your experience so far
Your input ->  Good
All done!
I am going to run a restaurant search using the following parameters:
 - cuisine: Italian
 - num_people: 3
 - outdoor_seating: False
 - preferences:
 - feedback: Good
Your input ->  Hi
Your input ->  I'm looking for a restaurant
Your input ->

PS: I have removed all the annoying duckling warnings: 2019-07-28 09:23:13 WARNING root - Could not parse timestamp 7e77404016504dc39cedfe0d34770da7. Instead current UTC time will be passed to duckling. Error: invalid literal for int() with base 10: ‘7e77404016504dc39cedfe0d34770da7’

amn41 · July 30, 2019, 2:12pm

hi - have you tried doing this in rasa interactive mode and providing feedback when the bot makes a mistake? It could be that you just don’t have enough training data.

if you run rasa shell --debug you should be able to see some output showing what dialogue policy is being used and what it’s predicting

jnadaud · July 31, 2019, 10:20pm

Here is the shell in debug mode starting from the end of the first conversation, trying to reinitiate a new one using the standard formbot example:

All done!
I am going to run a restaurant search using the following parameters:
 - cuisine: Chinese
 - num_people: 3
 - outdoor_seating: False
 - preferences: None
 - feedback: Hi
Your input ->  Thanks
2019-07-31 18:16:42 DEBUG    rasa.core.agent  - Created a new lock for conversation 'default'
2019-07-31 18:16:42 DEBUG    rasa.core.tracker_store  - Recreating tracker for id 'default'
2019-07-31 18:16:42 WARNING  root  - Could not parse timestamp 856650db29d646bea416e4f17cd86426. Instead current UTC time will be passed to duckling. Error: invalid literal for int() with base 10: '856650db29d646bea416e4f17cd86426'
2019-07-31 18:16:42 DEBUG    rasa.core.processor  - Received user message 'Thanks' with intent '{'name': 'thankyou', 'confidence': 0.9594088792800903}' and entities '[]'
2019-07-31 18:16:42 DEBUG    rasa.core.processor  - Logged UserUtterance - tracker now has 35 events
2019-07-31 18:16:42 DEBUG    rasa.core.policies.memoization  - Current tracker state [{'prev_utter_greet': 1.0, 'intent_greet': 1.0}, {'prev_action_listen': 1.0, 'intent_request_restaurant': 1.0}, {'prev_restaurant_form': 1.0, 'intent_request_restaurant': 1.0}, {'prev_utter_slots_values': 1.0, 'intent_request_restaurant': 1.0}, {'prev_action_listen': 1.0, 'intent_thankyou': 1.0}]
2019-07-31 18:16:42 DEBUG    rasa.core.policies.memoization  - There is a memorised next action '11'
2019-07-31 18:16:42 DEBUG    rasa.core.policies.form_policy  - There is no active form
2019-07-31 18:16:42 DEBUG    rasa.core.policies.ensemble  - Predicted next action using policy_0_MemoizationPolicy
2019-07-31 18:16:42 DEBUG    rasa.core.processor  - Predicted next action 'utter_noworries' with confidence 1.00.
2019-07-31 18:16:42 DEBUG    rasa.core.processor  - Action 'utter_noworries' ended with events '['BotUttered(text: you are welcome :), data: {"elements": null, "quick_replies": null, "buttons": null, "attachment": null, "image": null, "custom": null}, metadata: {})']'
2019-07-31 18:16:42 DEBUG    rasa.core.policies.memoization  - Current tracker state [{'prev_action_listen': 1.0, 'intent_request_restaurant': 1.0}, {'prev_restaurant_form': 1.0, 'intent_request_restaurant': 1.0}, {'prev_utter_slots_values': 1.0, 'intent_request_restaurant': 1.0}, {'prev_action_listen': 1.0, 'intent_thankyou': 1.0}, {'prev_utter_noworries': 1.0, 'intent_thankyou': 1.0}]
2019-07-31 18:16:42 DEBUG    rasa.core.policies.memoization  - There is a memorised next action '0'
2019-07-31 18:16:42 DEBUG    rasa.core.policies.form_policy  - There is no active form
2019-07-31 18:16:42 DEBUG    rasa.core.policies.mapping_policy  - There is no mapped action for the predicted intent, 'thankyou'.
2019-07-31 18:16:42 DEBUG    rasa.core.policies.ensemble  - Predicted next action using policy_0_MemoizationPolicy
2019-07-31 18:16:42 DEBUG    rasa.core.processor  - Predicted next action 'action_listen' with confidence 1.00.
2019-07-31 18:16:42 DEBUG    rasa.core.processor  - Action 'action_listen' ended with events '[]'
2019-07-31 18:16:42 DEBUG    rasa.core.agent  - Deleted lock for conversation 'default' (unused)
you are welcome :)
Your input ->  Hello
2019-07-31 18:16:47 DEBUG    rasa.core.agent  - Created a new lock for conversation 'default'
2019-07-31 18:16:47 DEBUG    rasa.core.tracker_store  - Recreating tracker for id 'default'
2019-07-31 18:16:47 WARNING  root  - Could not parse timestamp af7eaca426ed46e9874540674c25de47. Instead current UTC time will be passed to duckling. Error: invalid literal for int() with base 10: 'af7eaca426ed46e9874540674c25de47'
2019-07-31 18:16:47 DEBUG    rasa.core.processor  - Received user message 'Hello' with intent '{'name': 'greet', 'confidence': 0.9759040474891663}' and entities '[]'
2019-07-31 18:16:47 DEBUG    rasa.core.processor  - Logged UserUtterance - tracker now has 39 events
2019-07-31 18:16:47 DEBUG    rasa.core.policies.memoization  - Current tracker state [{'prev_restaurant_form': 1.0, 'intent_request_restaurant': 1.0}, {'prev_utter_slots_values': 1.0, 'intent_request_restaurant': 1.0}, {'prev_action_listen': 1.0, 'intent_thankyou': 1.0}, {'prev_utter_noworries': 1.0, 'intent_thankyou': 1.0}, {'prev_action_listen': 1.0, 'intent_greet': 1.0}]
2019-07-31 18:16:47 DEBUG    rasa.core.policies.memoization  - There is no memorised next action
2019-07-31 18:16:47 DEBUG    rasa.core.policies.form_policy  - There is no active form
2019-07-31 18:16:47 DEBUG    rasa.core.policies.ensemble  - Predicted next action using policy_1_FormPolicy
2019-07-31 18:16:47 DEBUG    rasa.core.processor  - Predicted next action 'action_listen' with confidence 0.00.
2019-07-31 18:16:47 DEBUG    rasa.core.processor  - Action 'action_listen' ended with events '[]'
2019-07-31 18:16:47 DEBUG    rasa.core.agent  - Deleted lock for conversation 'default' (unused)
Your input ->  I'd like to book a restaurant please
2019-07-31 18:17:09 DEBUG    rasa.core.agent  - Created a new lock for conversation 'default'
2019-07-31 18:17:09 DEBUG    rasa.core.tracker_store  - Recreating tracker for id 'default'
2019-07-31 18:17:09 WARNING  root  - Could not parse timestamp 4d9de499edc44a8a84e8fccaed6ade4c. Instead current UTC time will be passed to duckling. Error: invalid literal for int() with base 10: '4d9de499edc44a8a84e8fccaed6ade4c'
2019-07-31 18:17:09 DEBUG    rasa.core.processor  - Received user message 'I'd like to book a restaurant please' with intent '{'name': 'request_restaurant', 'confidence': 0.9860775470733643}' and entities '[]'
2019-07-31 18:17:09 DEBUG    rasa.core.processor  - Logged UserUtterance - tracker now has 41 events
2019-07-31 18:17:09 DEBUG    rasa.core.policies.memoization  - Current tracker state [{'prev_utter_slots_values': 1.0, 'intent_request_restaurant': 1.0}, {'prev_action_listen': 1.0, 'intent_thankyou': 1.0}, {'prev_utter_noworries': 1.0, 'intent_thankyou': 1.0}, {'prev_action_listen': 1.0, 'intent_greet': 1.0}, {'prev_action_listen': 1.0, 'intent_request_restaurant': 1.0}]
2019-07-31 18:17:09 DEBUG    rasa.core.policies.memoization  - There is no memorised next action
2019-07-31 18:17:09 DEBUG    rasa.core.policies.form_policy  - There is no active form
2019-07-31 18:17:09 DEBUG    rasa.core.policies.ensemble  - Predicted next action using policy_1_FormPolicy
2019-07-31 18:17:09 DEBUG    rasa.core.processor  - Predicted next action 'action_listen' with confidence 0.00.
2019-07-31 18:17:09 DEBUG    rasa.core.processor  - Action 'action_listen' ended with events '[]'
2019-07-31 18:17:09 DEBUG    rasa.core.agent  - Deleted lock for conversation 'default' (unused)
Your input ->

hajoura · August 1, 2019, 11:05am

@jnadaud thank you for your replay but the problem still not work I have no idea to solve it ?

hajoura · August 1, 2019, 11:11am

@amn41 he i have writing more training data as those these my date : ## intent:deny

no
no new selection
no thanks
no thank you
uh no
breath no
do you have something else
no this does not work for me

intent:goodbye

bye
goodbye
see you around
see you later
bye
bye
bye
bye
bye
bye

intent:greet

hey
hello
hi
good morning
good evening
hey there
hey
hey
hey
hello
hello
hey
hey
hey
hey

intent:inform

I am looking for expensive hotels
I want cheep hotel
i want expensive hotel
i need normal hotel
hotel in center
find for me hotel in center town
show all expensive hotels
i need normal hotel
i would like hotel with five
search for me hotel with two
i need hotel with middle
what hotel in north
hotel in east with three
look for me hotel in north with five
book for me hotel in west
i need higher hotel
i need two stars in hotel
book for me north of town
looking for hotel east of town
north of town
east of town
anywhere but in center
four stars
one stars
i look for two
my feedback is good
my feedback is great
it was terrible
i consider it success
you are awful
looking for a hotel with 5 stars
find me a hotel with 4 people
Can I get a hotel with 3
i need hotel in north of town
i want five stars in north of town
i want 3 stars
book for me 5 stars
the rating for the hotel is 3
i want hotel in north with two stars
normal price
expensive price
middle price
low price
i need hotel in north of town

intent:request_hotels

i am searching for hotels
i am looking for hotels
i want hotels
hotel,please
i need hotel
book me hotel
what hotel you propose to me
propose me hotel
give me hotel
what hotel do you suggest to me
i want hotel
book for me hotel in town
find for me the closer hotel
looking for hotel
hotel please

intent:stop

ok then you cant help me
that was shit, you’re not helping
you can’t help me
you can’t help me with what i need
i guess you can’t help me then
ok i guess you can’t help me
that’s not what i want
ok, but that doesnt help me
this is leading to nothing
this conversation is not really helpful
you cannot help me with what I want
I think you cant help me
hm i don’t think you can do what i want
stop
stop go back
do you get anything?
and you call yourself bot company? pff
and that’s it?
nothing else?

intent:thankyou

um thank you good bye
okay cool uh good bye thank you
okay thank you good bye
you rock
and thats all thank you and good bye
thank you and good bye
sorry about my mistakes thank you good bye
noise thank you good bye
thank you goodbye noise
okay thank you goodbye
uh thank you good bye
thank you goodbye
thank you goodbye noise thank you goodbye
breath thank you goodbye
thank you
okay thank you
thanks goodbye
ah thank you goodbye
thank you noise
thank you good bye
breath thank you very much goodbye
thanks
noise thank you goodbye
unintelligible thank you goodbye
uh okay thank you good bye
thank you bye
um okay thank you good bye

synonym:1

one

synonym:2

two

synonym:3

three

synonym:4

four

synonym:5

five and also these my config file

language: en pipeline: supervised_embeddings

policies:

name: MemoizationPolicy
name: KerasPolicy
name: MappingPolicy
name: FormPolicy

and my domain :

%YAML 1.1

actions:

utter_ask_continue
utter_chitchat
utter_goodbye
utter_greet
utter_noworries
utter_slots_values entities:
area
feedback
price
stars forms:
hotel_form intents:
goodbye
greet
inform
chitchat
request_hotels
thankyou
request_hotel: use_entities: []
deny
affirm slots: area: auto_fill: false type: unfeaturized feedback: auto_fill: false

type: unfeaturized

preferences:

auto_fill: false

type: unfeaturized requested_slot: auto_fill: false type: unfeaturized stars: auto_fill: false type: unfeaturized price: auto_fill: false type: unfeaturized templates: forms:
- text: hotel_form utter_ask_area:
- text: " give me the area" utter_ask_continue:
- text: “do you want to continue?” utter_ask_feedback:
- text: " please give your feedback on your experience so far" utter_ask_preferences:
- text: " please provide additional preferences" utter_ask_price:
  - text: " do you want normal price ?" utter_ask_stars:
- text: " how many stars?" utter_chitchat:
- text: “chitchat” utter_goodbye:
- text: “Bye” utter_greet:
- text: “Hey! I am hotel chatbot.How can i help you?” utter_noworries:
- text: “you are welcome :)” utter_slots_values:
- text: "I am going to run a restaurant search using the following parameters: - area: {area} - stars: {stars} - price: {price} - preferences: {preferences}
  - feedback: {feedback}" utter_submit:
- text: "All done/n I am going to run a restaurant search using the following parameters:
  - area: {area} - stars: {stars} - price: {price} - preferences: {preferences}
  - feedback: {feedback}" utter_wrong_area:
- text: “wrong area” utter_wrong_price:
- text: “wrong price” utter_wrong_stars:
- text: “wrong stars”

hajoura · August 1, 2019, 11:18am

2019-08-01 14:16:55 DEBUG rasa.core.policies.memoization - There is no memorised next action

2019-08-01 14:16:55 DEBUG rasa.core.policies.form_policy - There is an active form ‘hotel_form’

2019-08-01 14:16:55 DEBUG rasa.core.policies.ensemble - Predicted next action using policy_1_KerasPolicy

2019-08-01 14:16:55 DEBUG rasa.core.processor - Predicted next action ‘action_listen’ with confidence 0.21.

2019-08-01 14:16:55 DEBUG rasa.core.processor - Action ‘action_listen’ ended with events ‘[]’

2019-08-01 14:16:55 DEBUG rasa.core.agent - Deleted lock for conversation ‘default’ (unused)

Your input -> hey

tion_listen’: 1.0}, {‘intent_greet’: 1.0, ‘active_form_hotel_form’: 1.0, ‘prev_action_listen’: 1.0}, {‘intent_greet’: 1.0, ‘active_form_hotel_form’: 1.0, ‘prev_action_listen’: 1.0}, {‘intent_greet’: 1.0, ‘active_form_hotel_form’: 1.0, ‘prev_action_listen’: 1.0}]

2019-08-01 14:16:58 DEBUG rasa.core.policies.memoization - There is no memorised next action

2019-08-01 14:16:58 DEBUG rasa.core.policies.form_policy - There is an active form ‘hotel_form’

2019-08-01 14:16:58 DEBUG rasa.core.policies.ensemble - Predicted next action using policy_1_KerasPolicy

2019-08-01 14:16:58 DEBUG rasa.core.processor - Predicted next action ‘action_listen’ with confidence 0.18.

2019-08-01 14:16:58 DEBUG rasa.core.processor - Action ‘action_listen’ ended with events ‘[]’

2019-08-01 14:16:58 DEBUG rasa.core.agent - Deleted lock for conversation ‘default’ (unused)

Your input ->.
how to solve that issue ?

amn41 · August 7, 2019, 10:54am

it’s very hard to see from this partial debug log, but my guess is your problem is that there is still an active form

hajoura · August 7, 2019, 11:24am

@amn41 hi , how can I solve that issue as I told one conversation and stop ?? I did not clone the form action should I clone before I used it ?

Topic		Replies	Views
WHY MY BOT HANDLE ONE CONVERSATION WITH FROM ACTION? Rasa Open Source	1	565	August 12, 2019
One intent, one story .Rasa not response any messsage Rasa Open Source	6	1272	July 2, 2019
Keep responding default action Rasa Open Source	5	1672	October 14, 2020
Rasa Conversations are not responding (intermittently) Rasa Open Source	4	664	March 24, 2021
Help needed! Issue with dererministic policies and payloads Rasa Open Source	4	812	August 13, 2019