Skip to content

Based on Pytorch, the Adversarial Attack algorithm DeepFool, targeting the Mnist data set and ResNet18 network

Notifications You must be signed in to change notification settings

Dawn-David/DeepFool_MNIST

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

2 Commits
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

基于Mnist,DeepFool攻击ResNet18

文件说明

  • data:Mnist数据集
  • images:对抗样本,9957张,剩余43张图像网络本身无法正确分类,为原图像
  • model:ResNet18的训练函数,和预训练参数文件
  • logs:训练日志

代码文档说明

  • model.py:ResNet18网络定义
  • utils.py:其他函数
  • deepfool_fashion.py:DeepFool核心算法,当使用其他数据集时,代码中的图像预处理需要微调
  • Resnet-Deepfool.ipynd:主函数
  • tran.py:模型训练函数。

环境配置

  • python 3.7
  • pytorch 1.4.0
  • cuda 10.1

About

Based on Pytorch, the Adversarial Attack algorithm DeepFool, targeting the Mnist data set and ResNet18 network

Topics

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published