Please use 'Bert' related tokenizer classes and 'Nezha' related model classes

NEZHA: Neural Contextualized Representation for Chinese Language Understanding Junqiu Wei, Xiaozhe Ren, Xiaoguang Li, Wenyong Huang, Yi Liao, Yasheng Wang, Jiashu Lin, Xin Jiang, Xiao Chen and Qun Liu.

The original checkpoints can be found here

Example Usage

from transformers import BertTokenizer, NezhaModel
tokenizer = BertTokenizer.from_pretrained("sijunhe/nezha-base-wwm")
model = NezhaModel.from_pretrained("sijunhe/nezha-base-wwm")
text = "我爱北京天安门"
encoded_input = tokenizer(text, return_tensors='pt')
output = model(**encoded_input)