Skip to content

Latest commit

 

History

History

adversarial-attacks-defenses

Folders and files

NameName
Last commit message
Last commit date

parent directory

..
README.md
README.md
 
 
adversarial-tutorial.ipynb
adversarial-tutorial.ipynb
 
 
adversarial-tutorial.py
adversarial-tutorial.py
 
 

README.md

Implementing Adversarial Attacks and Defenses in Keras & Tensorflow 2.0

From 90% Accuracy to 20%

Available here: https://medium.com/analytics-vidhya/implementing-adversarial-attacks-and-defenses-in-keras-tensorflow-2-0-cab6120c5715

Requirements

  • Tested in Python 3.7.7 (pretty much any version of Python 3 should work)
  • tensorflow==2.1.0
  • matplotlib==3.2.0

The code will likely still work with other versions of these packages, but that hasn't been tested. If you have a capable GPU, you can substitute tensorflow with tensorflow-gpu.

Conda Instructions

conda create -n adversarial-attacks-defenses -y python=3.7.7
conda activate adversarial-attacks-defenses
conda install -y tensorflow==2.1.0
conda install -y -c conda-forge matplotlib==3.2.0