Hyperparticle/one-pixel-attack-keras
Keras implementation of "One pixel attack for fooling deep neural networks" using differential evolution on Cifar10 and ImageNet
Jupyter NotebookMIT
Issues
- 1
你们是用的是什么环境啊
#30 opened by Ghostraveler - 0
Experiment on your own dataset
#29 opened by wmrenr - 5
Not able to reproduce Google Collab notebook
#27 opened by martinehman - 0
About the function `perturb_image`, should it return one image or multi image depending the parameter `xs`
#28 opened by DAJINZI01 - 0
Demonstrate attack on vision transformers
#26 opened by binbbaz - 3
- 1
About dataset
#24 opened by lvhuanhuan123 - 2
- 1
- 1
How does the targeted attack work?
#20 opened by TudouJack - 2
The argument targeted
#19 opened by TudouJack - 3
About image processing
#18 opened by tuji-sjp - 1
Bug on attack_all function
#16 opened by falcaopetri - 2
Passing img instead of id
#15 opened by pniedzwiedzinski - 3
Getting a lot of adversarial images
#14 opened by moisbeug22 - 9
Replicate results seen in arXiv paper
#5 opened by Hyperparticle - 7
How to use this code to attack VGG16?
#11 opened by tuji-sjp - 2
- 3
- 1
rerunning attack on the same images
#10 opened by p-shah - 2
How to use gpu in the model ?
#9 opened by lith0613 - 7
- 0
Try new Differential Evolution techniques seen in "Attacking Convolutional Neural Network using Differential Evolution"
#7 opened by Hyperparticle - 4
Download ImageNet images automatically
#3 opened by Hyperparticle - 4
SyntaxError: invalid syntax
#6 opened by ghghgh0001 - 0
Demonstrate attack on ImageNet
#4 opened by Hyperparticle