Perform lemmatization using spaCy with optimized pipeline settings.
Disables unnecessary components (NER, parser) for faster processing.
Usage
spacy_lemmatize(x, batch_size = 100, model = "en_core_web_sm")
Arguments
- x
Character vector of texts OR a quanteda tokens object.
- batch_size
Integer; batch size for processing (default: 100).
- model
Character; spaCy model to use (default: "en_core_web_sm").
Value
A data frame with columns: doc_id, token_id, token, lemma.
Details
This function disables NER, entity_ruler, and parser components to speed up
lemmatization. Use this for lemmas without other annotations.
