rkkumar16
commented
6 years ago
**Rasa Core version**: 0.11
**Python version**: 3.6
**Operating system** (windows, osx, ...): windows 7
**Issue**:
Using code from documentation I tried to handle a channel from user input. When I train the NLU then I get proper intent and entity recognition but when I try same with rasa_core it does not recognize the entities and intents are also not identified. As a result no slot is filled.
```
def run(serve_forever=True):
input_channel = SimpleWebBot()
if serve_forever:
agent.handle_channels(
[input_channel], 5001
)
return agent
if __name__ == "__main__":
run()
```
**Content of domain file** (if used & relevant):
```yaml
slots:
user_mail:
type: text
bot_action:
type: text
cancel_type:
type: text
cancel_from:
type: text
cancel_upto:
type: text
entities:
- user_mail
- bot_action
- cancel_type
- cancel_from
- cancel_upto
- htype
intents:
- start_bot
- get_mail
- trip_cancel
- trip_type
- get_from_date
- get_to_date
actions:
- utter_greet
- utter_ask_email
- utter_bot_action
- utter_cancel_options
- utter_cancel_from
- utter_cancel_upto
- utter_cancel_result
templates:
utter_greet:
- text: "Hi, I'm Helpbot. How may I help you?"
utter_ask_email:
- text: "Can you please enter your email ID"
utter_bot_action:
- text: "Hi {user_mail}! How may I help you on trip
cancellation. Pls opt one of the below options"
buttons:
- title: "Trip cancellation"
payload: "Trip cancellation"
- title: "Log a feedback"
payload: "Log a feedback"
utter_cancel_options:
- text: "Hi {user_mail}! How may I help you on
trip cancellation. Please select one of the below options"
buttons:
- title: "cancel pick"
payload: "cancel pick"
- title: "cancel drop"
payload: "cancel drop"
utter_cancel_from:
- text: "OK. We will help you the cancellation. Pls mention from date
in DD/MM/YYY format. For single day cancellation from and to date will be
same"
utter_cancel_upto:
- text: "OK. Pls mention upto date in DD/MM/YYY format. For single day
cancellation from and to date will be same"
utter_cancel_result:
- text: "Hi, {user_mail}! Your trip has been
cancelled for the requested date."
```
stories.md
* start_bot
- utter_ask_email
* get_mail{"user_mail": "xyz@gmail.com"}
- utter_bot_action
* trip_cancel{"bot_action": "Trip Cancel"}
- utter_cancel_options
* trip_type{"cancel_type": "pick up"}
- utter_cancel_from
* get_from_date{"cancel_from": "27/11/2018"}
- utter_cancel_upto
* get_to_date{"cancel_upto": "27/11/2018"}
- utter_cancel_result
- action_restart
**nlu.md**
## intent:start_bot
- Hi
- Hi
- Hii
- Hii
- awake
- Hey
- Hey
- Hello
- Hello
## intent:get_mail
- My email is [abc@gmail.com](user_mail)
- My email is [saket.kumar@yahoo.com](user_mail)
- My email is [testmail@remail.com](user_mail)
- My email is [dummy@email.in](user_mail)
- My email is [m@yahoo.in](user_mail)
- My email is [what@dex.com](user_mail)
- My email is [testmail@testmail.com](user_mail)
- My email is [Myname@in.com](user_mail)
- My email is [My@gmail.com](user_mail)
- My email is [My@mail.com](user_mail)
- My email is [My@dummy.in](user_mail)
- My email is [My@whyme.com](user_mail)
- [rahul.kumar@yahoo.com](user_mail) is my email address
- [user.name@gmail.com](user_mail) is my email address
## intent:trip_type
- [cancel pick](cancel_type)
- [cancel pick](cancel_type)
- [cancel pick](cancel_type)
- [cancel both](cancel_type)
- [cancel both](cancel_type)
- [cancel both](cancel_type)
- [cancel drop](cancel_type)
- [cancel drop](cancel_type)
- [cancel drop](cancel_type)
## intent:bot_action
- [Trip Cancellation](action_type)
- [Trip Cancellation](action_type)
- [Log a feedback](action_type)
- [Log a feedback](action_type)
## intent:need_help
- Need help
- Need to contact
- Help required
- How can I get help
## intent:get_from_date
- from [27/11/2018](cancel_from)
- From [7/11/2018](cancel_from)
- from [2/11/2018](cancel_from)
- From [27/1/2018](cancel_from)
- From [27/11/2016](cancel_from)
- From [27/11/2012](cancel_from)
- from [27/11/2011](cancel_from)
- From [27/11/2010](cancel_from)
- From [27/11/2015](cancel_from)
- From [27/11/2014](cancel_from)
## intent:get_to_date
- to [28/11/2018](cancel_upto)
- upto [28/11/2018](cancel_upto)
- Till [28/11/2018](cancel_upto)
- Uptill [28](cancel_upto)
- Next [28](cancel_upto)
- coming [Monday](cancel_upto)
## regex:email
- My email is .+
**Logs for the code are**
```
The default 'Loader' for 'load(stream)' without further arguments can be unsafe.
Use 'load(stream, Loader=ruamel.yaml.Loader)' explicitly if that is OK.
Alternatively include the following in your code:
import warnings
warnings.simplefilter('ignore', ruamel.yaml.error.UnsafeLoaderWarning)
In most other cases you should consider using 'safe_load(stream)'
data = yaml.load(stream)
Using TensorFlow backend.
2018-08-29 13:31:14 DEBUG rasa_core.tracker_store - Creating a new tracker for id '233'.
2018-08-29 13:31:14 DEBUG rasa_core.processor - Received user message 'Hi' with intent '{'name': 'Hi', 'confidence': 1.0}' and entities '[]'
2018-08-29 13:31:14 DEBUG rasa_core.processor - Logged UserUtterance - tracker now has 2 events
2018-08-29 13:31:14 DEBUG rasa_core.processor - Current slot values:
bot_action: None
cancel_from: None
cancel_type: None
cancel_upto: None
user_mail: None
2018-08-29 13:31:14 DEBUG rasa_core.policies.memoization - Current tracker state [{}, {'prev_action_listen': 1.0, 'intent_Hi': 1.0}]
2018-08-29 13:31:14 DEBUG rasa_core.policies.memoization - There is no memorised next action
2018-08-29 13:31:14 DEBUG rasa_core.featurizers - Feature 'intent_Hi' (value: '1.0') could not be found in feature map. Make sure you added all intents and entities to the domain
2018-08-29 13:31:14 DEBUG rasa_core.policies.ensemble - Predicted next action using policy_1_KerasPolicy
2018-08-29 13:31:14 DEBUG rasa_core.processor - Predicted next action 'action_listen' with prob 0.99.
2018-08-29 13:31:14 DEBUG rasa_core.processor - Action 'action_listen' ended with events '[]'
192.168.54.44 - - [2018-08-29 13:31:14] "POST /chat/ HTTP/1.1" 200 110 0.091000
```
Please look into the issue and suggest some solution for this.
akelad