opacus
Explore Opacus, a library designed for seamless integration of differential privacy into PyTorch models. Providing robust privacy measures and minimal impact on performance, Opacus allows for real-time privacy budget tracking. Benefit from interactive tutorials, extensive examples, and guides, including a version 1.0 migration guide. Installation is easy via pip, conda, or directly from the source.
