bert-base-han-chinese

CKIP BERT Base Han Chinese

Pretrained model on Ancient Chinese language using a masked language modeling (MLM) objective.

Homepage

• ckiplab/han-transformers

Training Datasets

The copyright of the datasets belongs to the Institute of Linguistics, Academia Sinica.

• 中央研究院上古漢語標記語料庫
• 中央研究院中古漢語語料庫
• 中央研究院近代漢語語料庫
• 中央研究院現代漢語語料庫

Contributors

• Chin-Tung Lin at CKIP

Usage

• Using our model in your script

  from transformers import (
    AutoTokenizer,
    AutoModel,
  )
  
  tokenizer = AutoTokenizer.from_pretrained("ckiplab/bert-base-han-chinese")
  model = AutoModel.from_pretrained("ckiplab/bert-base-han-chinese")
  

• Using our model for inference

  >>> from transformers import pipeline
  >>> unmasker = pipeline('fill-mask', model='ckiplab/bert-base-han-chinese')
  >>> unmasker("黎[MASK]於變時雍。")
  
  [{'sequence': '黎 民 於 變 時 雍 。',
  'score': 0.14885780215263367,
  'token': 3696,
  'token_str': '民'},
  {'sequence': '黎 庶 於 變 時 雍 。',
  'score': 0.0859643816947937,
  'token': 2433,
  'token_str': '庶'},
  {'sequence': '黎 氏 於 變 時 雍 。',
  'score': 0.027848130092024803,
  'token': 3694,
  'token_str': '氏'},
  {'sequence': '黎 人 於 變 時 雍 。',
  'score': 0.023678112775087357,
  'token': 782,
  'token_str': '人'},
  {'sequence': '黎 生 於 變 時 雍 。',
  'score': 0.018718384206295013,
  'token': 4495,
  'token_str': '生'}]