metadata

language:
  - zh
thumbnail: >-
  https://user-images.githubusercontent.com/9592150/97142000-cad08e00-179a-11eb-88df-aff9221482d8.png
tags:
  - chinese
  - classical chinese
  - literary chinese
  - ancient chinese
  - bert
  - pytorch
license: apache-2.0
pipeline_tag: token-classification
widget:
  - text: >-
      及秦始皇，灭先代典籍，焚书坑儒，天下学士逃难解散，我先人用藏其家书于屋壁。汉室龙兴，开设学校，旁求儒雅，以阐大猷。济南伏生，年过九十，失其本经，口以传授，裁二十馀篇，以其上古之书，谓之尚书。百篇之义，世莫得闻。

Guwen NER

A Classical Chinese Named Entity Recognizer.

Note: There are some problems with decoding using the default sequence classification model. Use the CRF model to achieve the best results. CRF related code please refer to Guwen Models.