search

Hyperparticle / one-pixel-attack-keras

Keras implementation of "One pixel attack for fooling deep neural networks" using differential evolution on Cifar10 and ImageNet
https://arxiv.org/abs/1710.08864
MIT License
1.21k stars 213 forks source link

how can i calculate the value of success rate for target attack? #12

Closed ghost closed 5 years ago

ghost commented 5 years ago

Hello, could you please tell me about how to calculate success rate for LeNet in case of targeted attack? is there any formulas? thanks!

Hyperparticle commented 5 years ago

The success rate is simply the images that successfully fooled the network divided by the total number of attack images: success_images / total_images. The calculation is performed by attack_stats() inside helper.py.

Does that answer your question?

ghost commented 5 years ago

Thanks sir...

On Fri, Apr 12, 2019, 21:11 Dan Kondratyuk <notifications@github.com wrote:

The success rate is simply the images that successfully fooled the network divided by the total number of attack images: success_images / total_images. The calculation is performed by attack_stats() inside helper.py.

— You are receiving this because you modified the open/close state. Reply to this email directly, view it on GitHub https://github.com/Hyperparticle/one-pixel-attack-keras/issues/12#issuecomment-482622344, or mute the thread https://github.com/notifications/unsubscribe-auth/ApFr6dPCTtkEPkqzcSIKywQimnPJds3gks5vgKkcgaJpZM4cmLv7 .