Random input - intent classified with high confidence

On providing some random input, nlu is returning greet intent for the below example with high confidence,

Config.yml

Configuration for Rasa NLU.

https://rasa.com/docs/rasa/nlu/components/

language: en pipeline:

  • name: WhitespaceTokenizer
  • name: RegexFeaturizer
  • name: LexicalSyntacticFeaturizer
  • name: CountVectorsFeaturizer
  • name: CountVectorsFeaturizer analyzer: “char_wb” min_ngram: 1 max_ngram: 4
  • name: DIETClassifier epochs: 100
  • name: EntitySynonymMapper
  • name: ResponseSelector epochs: 100

Configuration for Rasa Core.

https://rasa.com/docs/rasa/core/policies/

policies:

  • name: TEDPolicy max_history: 5 epochs: 100
  • name: AugmentedMemoizationPolicy
  • name: MappingPolicy
  • name: FallbackPolicy nlu_threshold: 0.7 core_threshold: 0.7 fallback_action_name: “action_default_fallback”

nlu:

intent:greet

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

Input provided: english --> intent identified is greet with confidence as .93 Rasa version - Rasa 1.10.16 Setup is running in - conda environment.

Please guide to fix this issue.

@MohanSampagaon Delete all model which you’ve trained and retrain your model and use. i hope your problem will be fixed.

Thanks Yash, I did try deleting all the models and retrained again. Still the same.

Did you use default fallback action or custom? @MohanSampagaon

I have used default fallback

i have the same problem. I’am using Rasa 1.10.2 any solution please ???