laiguokun/Funnel-Transformer

Issues

• Funnel Transformer with Roberta Tokenizer.

  #13 opened 3 years ago by Vignesh-Sampat
  0

• Pretrained models link error

  #11 opened 3 years ago by a2108073
  1

• Train From Scratch

  #12 opened 3 years ago by shivanisrivarshini
  0

• Question related to pretraining with residual connection

  #10 opened 4 years ago by hxu38691
  0

• Deep VAEs as a research direction for the successor of XLnet

  #9 opened 4 years ago by LifeIsStrange
  0

• An Issue about the position embedding code

  #8 opened 4 years ago by hscspring
  0

• Several Issue About the Paper

  #7 opened 4 years ago by hscspring
  2

• Pretraining Issues

  #6 opened 4 years ago by nemani
  7

• Issue when trying to create TFrecords

  #5 opened 4 years ago by nemani
  4

• Would you like to provide a efficiency pipeline to preprocess large corpus?

  #2 opened 4 years ago by RyanHuangNLP
  4

• Do you apply wnli trick ? If so, can you open the code ?

  #4 opened 4 years ago by richarddwang
  0

• How long does it take to do pre-training?

  #3 opened 4 years ago by shizhediao
  1

• Creation of pretraining data

  #1 opened 4 years ago by stefan-it
  2