文件名称:adversarial-attacks-pytorch:PyTorch对抗攻击的实现
文件大小:10.09MB
文件格式:ZIP
更新时间:2024-06-01 04:09:33
deep-learning pytorch adversarial-attacks Python
对抗攻击PyTorch 是一个PyTorch库,其中包含对抗性攻击以生成对抗性示例。 干净的图像 对抗形象 目录 推荐的地点和配套 用法 :clipboard: 依存关系 火炬== 1.4.0 Python== 3.6 :hammer: 安装 pip install torchattacks或 git clone https://github.com/Harry24k/adversairal-attacks-pytorch import torchattacks atk = torchattacks . PGD ( model , eps = 8 / 255 , alpha = 2 / 255 , steps = 4 ) adversarial_images = atk ( images , labels ) :warning: 预防措施 在用于攻击之前,应使用transform [to.Tensor()]将所有图像缩放为
【文件预览】:
adversarial-attacks-pytorch-master
----demos()
--------Adversairal Training (MNIST).ipynb(7KB)
--------checkpoints()
--------White Box Attack (ImageNet).ipynb(2.97MB)
--------Applications of MultiAttack (CIFAR10).ipynb(6KB)
--------utils.py(1KB)
--------Attack Mode (ImageNet).ipynb(1.03MB)
--------Difference between eval mode and training mode (CIFAR10).ipynb(3KB)
--------models.py(3KB)
--------Black Box Attack (CIFAR10).ipynb(90KB)
--------data()
--------Performance Comparison (CIFAR10).ipynb(26KB)
--------Model with Multiple Outputs.ipynb(12KB)
----pic()
--------rfgsm.png(212KB)
--------adv_kor.png(588KB)
--------stepll.png(143KB)
--------fgsm.png(129KB)
--------bim.png(144KB)
--------pgd.png(201KB)
--------clean.png(99KB)
--------rpgd.png(256KB)
--------deepfool.png(116KB)
--------cw.png(114KB)
----requirements.txt(31B)
----update_records.md(8KB)
----LICENSE(1KB)
----README.md(21KB)
----torchattacks()
--------attacks()
--------__init__.py(773B)
--------attack.py(8KB)
----README_KOR.md(21KB)
----docs()
--------source()
--------make.bat(780B)
--------attacks.rst(2KB)
--------attack.rst(433B)
--------conf.py(5KB)
--------index.rst(351B)
--------Makefile(609B)
----.gitignore(286B)
----contributions.md(5KB)