Token Classification
GLiNER
PyTorch
multilingual

Update gliner_config.json

#1

polyglot does not work -> make app crash when trying to load model:
GLiNER.from_pretrained("thegenerativegeneration/gliner-finetune-small-v2.5")
ValueError: polyglot is not implemented, choose between 'whitespace', 'spacy', 'jieba', 'hanlp' and 'mecab'

Hey Ivan
Sorry for the problem.
I have been using polyglot in my own fork https://github.com/urchade/GLiNER/compare/main...thegenerativegeneration:GLiNER:main because I wanted a splitter that works for languages that do not use whitespace.

However, I am not sure it is the best solution either.

thegenerativegeneration changed pull request status to closed

Sign up or log in to comment