I have a large database of semi-structured legal documents that I have already labeled using regex and basic ML, but now I am considering adding all of those labels and more complicated new labels into a Prodigy/spaCy model.
Does anyone have any experience using spaCy/Prodigy on semi-structured documents, where there are few sentences?
These are the documents I am wanting to further classify - https://alexschwab.com/table.php