Chinese_models_for_SpaCy

Why we included this project

For anyone doing Chinese text processing in the SpaCy ecosystem, this project was the first practical source of pretrained Chinese models for the library, covering tokenization, part-of-speech tagging, dependency parsing, and named entity recognition. The models ship as installable binary packages, so you can add them to a SpaCy pipeline and work with Chinese documents using the same API you already know from English. The repository also documents the full training workflow and the OntoNotes 5.0 corpus it was built on, which helps if you want to retrain or adapt the models yourself. The official SpaCy Chinese models now exist and were developed with reference to this project, so the repo is mostly in maintenance mode, but it still works as a clear, self-contained reference for setting up Chinese NLP with SpaCy.