SanghunYun/UDA_pytorch

Issues

• apply softmax temperature controlling to augumentation example?

  #9 opened 5 years ago by matricer
  5

• The training stops without triggering errors

  #15 opened 3 years ago by ipietri
  10

• larger_than_threshold

  #22 opened 2 years ago by LiuChen19960902
  0

• Why we have the check global_step > 4999?

  #21 opened 3 years ago by kgarg8
  0

• Why unsup_loss is divided by unsup_loss_mask not total?

  #20 opened 3 years ago by kgarg8
  0

• Why sup_iter is not repeated when uda_mode is "False"?

  #18 opened 3 years ago by Liu-Jingyao
  1

• Does anyone know how total step is set, how much is appropriate, and what is the relationship between total step and epoch?

  #19 opened 3 years ago by hahah181292
  1

• Has anyone run successfully with their own raw dataset??

  #17 opened 3 years ago by hahah181292
  0

• how to process new dataset?

  #8 opened 5 years ago by pengpenglove
  1

• how port the M to some other clasiffication model like textCNN etc

  #16 opened 3 years ago by SeekPoint
  0

• TSA(Training Signal Annealing) or Confidence-based masking

  #14 opened 4 years ago by CheungZeeCn
  0

• Can the project run on tensorflow 2.0? Because I need to use transformers (it needs to run on tensorflow 2.0)

  #13 opened 4 years ago by kognweizheng
  0

• Are the quoted accuracies correct in the README?

  #12 opened 4 years ago by lewtun
  2

• kl divergence uses ori_prob instead of ori_log_prob

  #11 opened 4 years ago by stellaywu
  1

• model.eval() missing?

  #10 opened 5 years ago by matricer
  0

• Application of dropout when calculating the consistency loss

  #7 opened 5 years ago by shamanez
  0

• Where is the entropy minimization loss ?

  #6 opened 5 years ago by shamanez
  0

• what are the "start" and "end" parameters used when calculating the TSA?

  #5 opened 5 years ago by shamanez
  2

• Loading Fine-Tuned Pytorch BERT model.

  #4 opened 5 years ago by nvshrao
  1

• calculating the threshold when applying TSA

  #1 opened 5 years ago by shamanez
  1

• explain how unsup data pre-process works

  #3 opened 5 years ago by huangjinsuzhou
  1

• could you provide the example of raw text file format for train, test, unsup?

  #2 opened 5 years ago by huangjinsuzhou
  6