0

I try to train my bot based on rasa_nlu. Below is my config file and i have problems because entity like "next month" is recognized by ner_spacy to be something else than time data. I want this type of entity to be recognized only by duckling module.
Thanks

language: "en" project: "nav-os" pipeline: - name: "nlp_spacy" model: "en" - name: "ner_spacy" - name: "tokenizer_spacy" - name: "intent_entity_featurizer_regex" - name: "intent_featurizer_spacy" - name: "ner_crf" - name: "ner_synonyms" - name: "intent_classifier_sklearn" - name: "ner_duckling" dimensions: - "time"

1 Answers1

0

You could exclude the dimension for spaCy by defining the ones you want to include like it is described in the documentation.

Means you could configure the spacy_ner component like the following to only extract PERSON entities (just as example).

pipeline:
- name: "SpacyEntityExtractor"
  # dimensions to extract
  dimensions: ["PERSON"]
Tobias
  • 1,880
  • 11
  • 17