stanfordnlp / stanza

Stanford NLP Python library for tokenization, sentence segmentation, NER, and parsing of many human languages
https://stanfordnlp.github.io/stanza/
Other
7.22k stars 887 forks source link

FileNotFoundError for Old French lemmatizer #157

Closed kylepjohnson closed 4 years ago

kylepjohnson commented 4 years ago

Describe the bug When calling the default Old French NLP models (stanfordnlp.Pipeline(lang="fro")), then following error arises:

FileNotFoundError: [Errno 2] No such file or directory: '/Users/kyle.p.johnson/stanfordnlp_resources/fro_srcmf_models/fro_srcmf_lemmatizer.pt'

To Reproduce

See gist here for commands and full traceback: https://gist.github.com/kylepjohnson/d40215b380be4b050b5cc1ceac09e369

Expected behavior Expect the Pipeline object to become instantiated.

Environment (please complete the following information):

Additional context I've tried several other languages (Ancient Greek grc and Latin lat) and the Pipeline works fine.

J38 commented 4 years ago

We don't seem to officially support fro, though I can see a models download file for it. You should just set lemma_use_identity to True in your application.

kylepjohnson commented 4 years ago

Thank you, I will try this out and come back for help, if needed.