Here is the rasa core output:
2021-01-05 16:21:55 DEBUG rasa.nlu.classifiers.fallback_classifier - NLU confidence 0.2250533103942871 for intent 'greet' is lower than NLU threshold 0.50.
2021-01-05 16:21:55 DEBUG rasa.core.processor - Received user message 'hi' with intent '{'name': 'nlu_fallback', 'confidence': 1.0}' and entities '[]'
2021-01-05 16:21:55 DEBUG rasa.core.processor - Logged UserUtterance - tracker now has 4 events.
2021-01-05 16:21:55 DEBUG rasa.core.policies.memoization - Current tracker state [{}, {'user': {'intent': 'nlu_fallback'}, 'prev_action': {'action_name': 'action_listen'}}]
2021-01-05 16:21:55 DEBUG rasa.core.policies.memoization - There is no memorised next action
2021-01-05 16:21:55 DEBUG rasa.core.policies.rule_policy - Current tracker state: [{}, {'user': {'intent': 'nlu_fallback'}, 'prev_action': {'action_name': 'action_listen'}}]
2021-01-05 16:21:55 DEBUG rasa.core.policies.rule_policy - There is a rule for the next action 'action_default_fallback'.
2021-01-05 16:21:55 DEBUG rasa.core.policies.ensemble - Predicted next action using policy_2_RulePolicy
2021-01-05 16:21:55 DEBUG rasa.core.processor - Predicted next action 'action_default_fallback' with confidence 1.00.
I have already given enough training data to rasa, as in my nlu.yml
:
version: "2.0"
nlu:
- intent: greet
examples: |
- hey
- hello
- hi
- good morning
- good evening
- hey there
- yo
- Hi
- intent: goodbye
examples: |
- bye
- goodbye
- see you around
- see you later
- bye!
- intent: affirm
examples: |
- yes
- indeed
- of course
- that sounds good
- correct
- perfect
- good
- great
- amazing
- wonderful
- intent: deny
examples: |
- no
- never
- I don't think so
- don't like that
- no way
- not really
- sad
- very sad
- unhappy
- bad
- very bad
- awful
- terrible
- not very good
- extremely sad
- so sad
- intent: bot_challenge
examples: |
- are you a bot?
- are you a human?
- am I talking to a bot?
- am I talking to a human?
Here is my config.yml
:
language: en
pipeline:
- name: WhitespaceTokenizer
case_sensitive: false
- name: RegexFeaturizer
- name: CRFEntityExtractor
- name: DucklingHTTPExtractor
url: http://localhost:8000
dimensions:
- time
- number
- amount-of-money
- distance
locale: EN
timezone: UTC
timeout: 3
- name: EntitySynonymMapper
- name: CountVectorsFeaturizer
- name: CountVectorsFeaturizer
analyzer: char_wb
min_ngram: 1
max_ngram: 4
lowercase: true
- name: DIETClassifier
epochs: 10
- name: FallbackClassifier
threshold: 0.5
ambiguity_threshold: 0.1
policies:
- name: MemoizationPolicy
max_history: 5
- name: TEDPolicy
epochs: 10
- name: RulePolicy
core_fallback_threshold: 0.4
core_fallback_action_name: action_default_fallback
Any idea why my bot cannot understand ‘hi’ as intent ‘greet’? Thanks.