udapi.block.ud.lemmatize module

Block to add missing lemmas in cases where it seems obvious what the lemma should be.

class udapi.block.ud.lemmatize.Lemmatize(zones='all', if_empty_tree='process', **kwargs)

Bases: Block

process_node(node)

Some treebanks lack lemmas for some or all words. Occasionally we may be able to guess that the lemma is identical to the word form. This block will then fill out the lemma.

For some parts of speech, we can only say that the form is the lemma if we have morphological features that will confirm it is the right form.