Harry24k/adversarial-attacks-pytorch

Issues

• [QUESTION]If My model return a dict,how can I produce adv sample？

  #201 opened 6 days ago by LYAFJJ
  0

• [QUESTION] question about normalization

  #200 opened 23 days ago by kydxh
  0

• [FEATURE] Compatibility with pytorch 2.5

  #199 opened a month ago by moghadas76
  1

• [QUESTION]

  #183 opened 8 months ago by ShuchunXu
  2

• [QUESTION]The attack success rate of CW in the least-likely mode is 0, and the BIM in the same situation is above 90

  #198 opened 3 months ago by wang1528186571
  0

• When I tested CIFAR-10 and CIFAR-100, I wanted to use the standard version of Resnet50. The robust version was added to the robustbench library.

  #197 opened 3 months ago by wang1528186571
  1

• Adversarial attacks for YOLOv8 Detection task

  #175 opened 10 months ago by BilAlHomsi
  2

• [QUESTION] AutoAttack iterations

  #196 opened 4 months ago by brsunagad
  0

• Errors when downloading models from model zoom

  #195 opened 5 months ago by JAMES-YI
  1

• requests package version

  #194 opened 5 months ago by jungcheolseung
  0

• [QUESTION] using torchattacks on HuggingFace transformer models

  #193 opened 6 months ago by bakachan19
  2

• set_normalization_used

  #191 opened 6 months ago by talrub
  4

• [BUG] JSMA massive gpu memory consumption

  #187 opened 6 months ago by Dontoronto
  9

• Implementation of SuperDeepFool attack

  #190 opened 6 months ago by alirezaabdollahpour
  1

• Using Saved model for generating adversarial samples.

  #189 opened 6 months ago by chamanbanolia
  1

• Difference in robust accuracies for Autoattack

  #188 opened 6 months ago by iamsh4shank
  1

• [QUESTION] I have got confused by the code description

  #186 opened 7 months ago by EthanChu7
  3

• [QUESTION]BIM and PGD have the same output

  #185 opened 7 months ago by pleaseWind
  6

• [BUG] Bugs in the f function of cw.py

  #184 opened 7 months ago by EthanChu7
  2

• [BUG] The implementation of eps-ball in MI-FGSM

  #182 opened 8 months ago by ZhongliangGuo
  1

• The model loading in the demo seems to have become inactive[BUG]

  #166 opened a year ago by wan3333
  11

• FileNotFoundError: [Errno 2] No such file or directory: './data/imagenet_class_index.json'

  #181 opened 9 months ago by Qucl457
  2

• [BUG]The one-pixel attack is not working as expected. It seems not work?

  #180 opened 9 months ago by finger-monkey
  1

• [BUG] FAB is not working

  #179 opened 9 months ago by mmajewsk
  3

• [BUG] Applying attacks on 2 instances of the same model gives too far away results

  #178 opened 9 months ago by talrub
  1

• torchattacks.FGSM(model, eps=0) perturbs image

  #172 opened a year ago by michellerosehb
  5

• FileNotFoundError: [WinError 3] The system cannot find the path specified: './data/imagenet'

  #177 opened 10 months ago by tryingmybest99
  1

• [QUESTION]How to use Square or Pixle to attack?

  #176 opened 10 months ago by allanxia
  3

• [BUG] Getting same image as input

  #155 opened a year ago by Moebits
  6

• One of the differentiated Tensors appears to not have been used in the graph. Set allow_unused=True if this is the desired behavior.

  #170 opened a year ago by rishiyanth
  4

• Adversarial Training

  #169 opened a year ago by COD1995
  2

• [QUESTION] Low success rate for the Carlini-Wagner attack.

  #171 opened a year ago by Adversarian
  7

• [QUESTION] Binary search for Carlini-Wagner

  #167 opened a year ago by Adversarian
  4

• [BUG] no function "set_training_mode" but "set_model_training_mode" instead

  #159 opened a year ago by skadiscarlet
  3

• one pixel numpy, tensor incompatiblity

  #162 opened a year ago by hosnahoseini
  1

• [QUESTION] What is the difference between autoattack.py in this repository and autoattack.py in the auto-attack repository?

  #164 opened a year ago by seopp
  1

• [QUESTION] Can be used for segmentation tasks?

  #141 opened a year ago by MarlonGarcia
  3

• [BUG] miss of res for L1 in torchattack.fab.py

  #163 opened a year ago by JieHarry-Hu
  1

• [BUG] PGDL2 very low attack success rate

  #142 opened a year ago by rikonaka
  8

• [BUG] Incorrect random start of PGD L2

  #161 opened a year ago by Framartin
  2

• [QUESTION] About check_validity function

  #158 opened a year ago by WWWWWLI
  2

• [BUG]when i use pip to install the new version, it uinstall my cuda and torch(1.12.0) and install the cpu version(1.13.0)

  #157 opened a year ago by biuboomc
  2

• [QUESTION]The 'inf' in 'deepfool.py'

  #156 opened a year ago by songty21110133
  2

• [QUESTION] Using this on blip captioner

  #154 opened 2 years ago by Moebits
  5

• [BUG]if use pip install torchattacks, the DeepFool is wrong

  #153 opened 2 years ago by kld20092009
  3

• [QUESTION]autoattack in binary classification

  #152 opened 2 years ago by xarryon
  0

• [FEATURE] Target attack of PI-FGSM

  #146 opened 2 years ago by CHENBIN99
  4

• TorchAttack for Quantised pytorch model

  #150 opened 2 years ago by mohanrajroboticist
  7

• [QUESTION] Why the input must have a range of [0, 1]?

  #145 opened 2 years ago by youyinnn
  4

• [FEATURE] Consider adding PI-FGSM attack

  #143 opened 2 years ago by CHENBIN99
  2