search

stanfordnlp / stanza

Stanford NLP Python library for tokenization, sentence segmentation, NER, and parsing of many human languages
https://stanfordnlp.github.io/stanza/
Other
7.14k stars 880 forks source link

Lemma goeswith #1346

Closed AngledLuffa closed 4 months ago

AngledLuffa commented 4 months ago

Eliminate goeswith phrases from the lemmatizer training data. Doesn't do anything to the lemmatizer in the case of eval data (dev or test sets, and more importantly, Pipelines)

Addresses https://github.com/stanfordnlp/stanza/issues/1345 once the models are retrained

AngledLuffa commented 4 months ago

Had to update the previous caseless example to better match the new models. Should be good to go now. The new model is available in 1.7.0 already