junnyu / roformer_v2_chinese_char_large

Other483Updated May 11, 2022

Downloads

48

Likes

3

Last updated

May 11, 2022

License

Not available

def forward(self, features, kwargs): x = features[:, 0, :] # take <s token (equiv. to [CLS]) x = self.dropout(x) x = self.dense(x) x = ACT2FNself.config.hiddenact # 这里是relu x = self.dropout(x) x = self.outproj(x) return x

Tasks & library

  • fill mask
  • transformers

Tags

  • transformers
  • pytorch
  • roformer
  • fill-mask
  • roformer-v2
  • tf2.0

Technical specifications

ParametersNot available
Librarytransformers
Languageszh
LicenseNot available