Skip to main content
Version: 0.93

operators.spacy.VerbPhraseFeaturizer

class operators.spacy.VerbPhraseFeaturizer(field, target_field=None, min_length=2, **spacy_span_kwargs)

A Featurizer that yields all verb phrases according to a simple part-of-speech verb match.

Used by the Seq Tagging application template as one of the base candidate generators for producing embeddings from.

Parameters:

  • field (str) – The field to parse into verb phrases.

  • target_field (Optional[str], default: None) – The field in which to store the extracted verb phrase candidates.

  • min_length (int, default: 2) – The minimum number of tokens for extracted verb phrases. Used to avoid single token phrases. Default: 2.

  • spacy_span_kwargs (Any) – kwargs to pass to the spaCy model.