I am working on lemmatizer for Brazilian portugese (pt_br). I know spacy provides lemmatizer for pt_pt. Can this library be used for pt_br as well or there will be significant difference between pt_pt and pt_br.
The current pt
version in spaCy is actually supposed to cover both. See for instance this PR https://github.com/explosion/spaCy/pull/2790 which specifically caters for the currency symbol from Brazil, and this PR https://github.com/explosion/spaCy/pull/2307 where both versions of both spelling variants of numbers are specifically taken into account.