segram.nlp.pipeline.merger module

class segram.nlp.pipeline.merger.Merger(vocab: ~spacy.vocab.Vocab, name: str = 'attribute_ruler', *, validate: bool = False, scorer: ~typing.Callable | None = <function attribute_ruler_score>)[source]

Bases: Annotator

Merger class for merging standard multitokens found in a given language (such as phrasal verbs in English) as well as multitoken entities.

__call__(doc: Doc) → Doc[source]: Retokenize document and merge multitokens.