DuaneNielsen/DeepInfomaxPytorch

Issues

• some potential bugs w.r.t BN layers

  #18 opened 2 years ago by JingyuYang1997
  1

• Matching representations to a prior distribution is wrong?

  #16 opened 3 years ago by kosta-jo
  3

• How would I apply this to non-image (1-dimensional) data?

  #17 opened 3 years ago by JAEarly
  2

• Why this code don't apply adversarial training?

  #13 opened 4 years ago by GuardSkill
  1

• model_path = Path(r'c:\data\deepinfomax\models\run1\encoder' + str(epoch))

  #15 opened 3 years ago by zhaoji-1024
  0

• FileNotFoundError: [Errno 2] No such file or directory: 'c:\\data\\deepinfomax\\models\\run5/encoder860.wgt'

  #5 opened 5 years ago by haohuazhu
  3

• How was the result with full loss used?

  #14 opened 4 years ago by YounkHo
  0

• epoch restart?

  #12 opened 4 years ago by yujiapingff
  1

• Why does prior distribution have no encoder loss？

  #6 opened 5 years ago by HaopengZhang96
  10

• The loss value is negative

  #11 opened 4 years ago by tianlili1
  1

• RuntimeError: one of the variables needed for gradient computation has been modified by an inplace operation

  #10 opened 5 years ago by liunanbo
  1

• loss score become negative infinity and nan

  #9 opened 5 years ago by liunanbo
  5

• Why do you fine-tune the encoder when training a classifier?

  #8 opened 5 years ago by universome
  1

• Why the experimental results are inferior to those reported in paper?

  #7 opened 5 years ago by wangxu-scu
  2

• The prior is wrong

  #1 opened 6 years ago by Abraham0912541
  12

• Questions about loss functions

  #4 opened 6 years ago by TulsiJain
  2

• Question about the loss function

  #3 opened 6 years ago by JaeDukSeo
  3

• Great Code!

  #2 opened 6 years ago by JaeDukSeo
  2