marco-rudolph/differnet

Issues

• How can I locate the defects on the test image?

  #46 opened 10 months ago by Algabeno
  1

• Is it normal for the train loss to often be negative?

  #45 opened a year ago by supersuperdong
  1

• Question about using FrEIA instead of custom NF codes.

  #43 opened 2 years ago by AndSonder
  1

• Train loss might be negative during training, what does this situation mean?

  #44 opened 2 years ago by WenPX
  0

• train loss

  #42 opened 2 years ago by emiyashiroz
  1

• Why is the loss function negative

  #40 opened 2 years ago by TaoTXiXi
  2

• Test after training

  #41 opened 2 years ago by hechengbo-H
  1

• error when I run command

  #31 opened 3 years ago by valadad
  1

• Can this work process one channel pics?

  #36 opened 3 years ago
  1

• Qualitative results between CS-Flow and DifferNet on MTD dataset

  #37 opened 3 years ago by sungwool
  1

• The work can be used in finding small exceptional points in images?

  #35 opened 3 years ago by guyiyifeurach
  5

• How to improve accuracy of the localisation of anomalies at gradient map?

  #34 opened 3 years ago by volkov-maxim
  1

• How to train a model for several classes?

  #33 opened 3 years ago by volkov-maxim
  2

• AttributeError: 'ImageFolderMultiTransform' object has no attribute '_find_classes'

  #32 opened 3 years ago by volkov-maxim
  4

• size issue in anomaly score

  #30 opened 3 years ago by jxzhangjhu
  1

• Error in running dummy dataset

  #29 opened 3 years ago by OzanErcan6
  2

• How to train anomaly data set to improve accuracy ?

  #28 opened 3 years ago by rfuk0503
  2

• when set grad_map_viz to be True, I got a mistake: rotate() got an unexpected keyword argument 'reshape'

  #27 opened 3 years ago by imkelt
  6

• How to test?

  #25 opened 3 years ago by EnverUsta
  1

• Get a gradient map from scored image

  #24 opened 3 years ago by aditya-pratama96
  1

• Few-shots

  #26 opened 3 years ago by renqiqi
  1

• Magnatice tile dataset train and test part

  #23 opened 3 years ago by AmmarKamoona
  2

• Is it normal that anomaly_score is super large

  #3 opened 4 years ago by lihyin
  4

• About the Resnet training

  #20 opened 4 years ago by renqiqi
  4

• UserWarning

  #16 opened 4 years ago by jeff-ai-deep-learning
  1

• Test code

  #15 opened 4 years ago by nxm19830306
  3

• How to make predictions

  #21 opened 4 years ago by simonefinelli
  1

• How is θ given

  #19 opened 4 years ago by renqiqi
  2

• Questions about the test dataset and AUROC

  #18 opened 4 years ago by renqiqi
  0

• test_loss is super large, is that normal？

  #17 opened 4 years ago by renqiqi
  2

• Why the train loss is negative ?

  #14 opened 4 years ago by nxm19830306
  1

• Histogram - Figure 6

  #13 opened 4 years ago by bnascimento
  2

• Reproducibility of results

  #12 opened 4 years ago by d3adc0c0
  6

• Possible typo

  #8 opened 4 years ago by omergra
  1

• Classification Results

  #11 opened 4 years ago by msergenergen
  3

• the values of train loss are becoming negative

  #10 opened 4 years ago by QingL0218
  2

• How to judge whether the prediction result is OK or ng

  #7 opened 4 years ago by machine52vision
  1

• How to do the test model with an image?

  #6 opened 4 years ago by nguyenanhtuan1008
  2

• Should the test data contain rotation tranform when calculating anomaly_score?

  #4 opened 4 years ago by jinzishuai
  4

• test on the good sample of dummy_dataset

  #2 opened 4 years ago by lihyin
  1

• Inference performance

  #1 opened 4 years ago by AlgirdasKartavicius
  1