CLUEbenchmark/FewCLUE

Issues

• 使用RoBERTa-large无法复现P-Tuning结果

  #16 opened 2 years ago by PineappleWill
  13

• chid任务

  #23 opened 2 years ago by githubjixin
  0

• PET和Ptuning有pytorch版本的代码吗？

  #22 opened 2 years ago by Faith-Uchiha
  0

• README中的实验结果是如何计算的？

  #21 opened 2 years ago by Armorhtk
  0

• python run_gpt.py -t chid -z 这个命令zero shot 为什么会进入样本训练逻辑？什么时候predict操作？

  #20 opened 2 years ago by ThetaRgo
  0

• baseline/models_keras/ptuning_origin

  #19 opened 3 years ago by Torres9999
  1

• CSLDCP数据量和描述中不一致

  #18 opened 3 years ago by PineappleWill
  0

• 人工标注eprstmt 得分86.59，数据集质量堪忧。。

  #17 opened 3 years ago by lindabigwheel
  3

• EFL模型无法复现

  #12 opened 3 years ago by sunyilgdx
  3

• 训练或者验证阶段能否使用test_public数据

  #15 opened 3 years ago by wmathor
  2

• train_1.json集和dev_0.json集应该一一对应吗

  #14 opened 3 years ago by xsthunder
  1

• P-Tuning baseline 代码缺失

  #13 opened 3 years ago by PineappleWill
  1

• FewCLUE LM-BFF can't run in prompt-demo mode

  #11 opened 3 years ago by o0mahan0o
  1

• 可以更新一下加群二维码吗？

  #3 opened 4 years ago by Riroaki
  1