YisenWang/symmetric_cross_entropy_for_noisy_labels

Issues

• tabular data/ noisy instances

  #11 opened 2 years ago by nazaretl
  0

• Asymmetric Noise-

  #10 opened 3 years ago by jaideep11061982
  4

• Is the accuracy reported in tables 1, 2, and 3 from the best epoch or the last epoch?

  #8 opened 4 years ago by guixianjin
  2

• About alpha and beta

  #7 opened 4 years ago by Zhengtq
  1

• Maybe there are something wrongs with your code.

  #9 opened 4 years ago by ningkp
  0

• Multi-label scenario

  #6 opened 4 years ago by aditya1709
  1

• How to regulate the alpha and the beta of SCEloss?

  #5 opened 4 years ago by firelgh
  4

• Is symmetric_cross_entropy suitable for binary classification issue?

  #4 opened 4 years ago by useric
  1

• Why Results reported in SL and D2L are different?

  #3 opened 4 years ago by XinshaoAmosWang
  3

• How do you create asymmetric label noise on CIFAR-100?

  #2 opened 4 years ago by XinshaoAmosWang
  3

• Matrix p = ?, when training on cifar-100 using Forward and Backward

  #1 opened 4 years ago by zhangchbin
  3