There are no reviews yet. Be the first to send feedback to the community and the maintainers!
eagerpy
PyTorch, TensorFlow, JAX and NumPy — all of them natively using the same codelinear-region-attack
A powerful white-box adversarial attack that exploits knowledge about the geometry of neural networks to find minimal adversarial perturbations without doing gradient descentuniformly-sampling-nd-ball
Efficiently sampling vectors from the n-sphere and n-ballarxiv-bear-app
My personal reference management system for arXiv papers using Bear.appfoolbox-native-tutorial
OSXNotifier.jl
Julia package to send notifications to the OS X Notification Centerfoolbox-native
Foolbox Native brings native performance to Foolboxtensorflow-imagenet
Jupyter notebook templates for training or fine-tuning on ImageNet using TensorFlow.foolbox-tensorflow-keras-applications
The pretrained TensorFlow Keras models with a Foolbox Zoo compatible interfacec2s-docker
Dockerfile for the c2s toolbox https://github.com/lucastheis/c2sanalysis-by-synthesis
Analysis by Synthesis — reimplementedattax
Attax: adversarial attacks using JAXlockfile
lockfile provides a minimalistic and modern implementation of a simple file-based lock mechanism for PythonDiscreteEntropyEstimators.jl
Discrete Entropy Estimators implemented in Julianorm
Norm is a tiny command-line utility to compute the norm of the difference between two imagesclipping-aware-rescaling
Calculates eta such that norm(clip(x + eta * delta, a, b) - x) == eps.cifar10-fast-reimplemented
A simplified reimplementation of the original fast CIFAR10 training codec2s-ipython-docker
Dockerfile for the c2s toolbox with IPython support https://github.com/lucastheis/c2splotspikes.py
A minimal python function for beautiful spike plottingLove Open Source and this site? Check out how you can help us