ML-privacy

Privacy risk assesment tool on Deep learning models

How to use

    cd src/

    python attack.py

Available attack: MIA (Memebership Inference)

Default attacker: EnsembleAttacker

Target model: DenseNet121

Dataset: Cifar10 (default) and Cifar100

To enable GPU:

    pip install torch==1.12.1+cu113 torchvision==0.13.1+cu113 -f https://download.pytorch.org/whl/cu113/torch_stable.html