A Python toolbox to create adversarial examples that fool neural networks in PyTorch, TensorFlow, and JAX
MIT License
Bot releases are hidden (Show)
Published by zimmerrol over 2 years ago
Published by zimmerrol over 2 years ago
Published by jonasrauber over 3 years ago
Serious bug fix for SaltAndPepperAttack
(thanks @maurapintor @zangobot)
Published by jonasrauber over 3 years ago
Published by jonasrauber about 4 years ago
Foolbox release for Zenodo
Published by jonasrauber about 4 years ago
Published by jonasrauber about 4 years ago
Bug fix
Published by jonasrauber about 4 years ago
HopSkipJump
attack to v3data_format
JAXModel
now supports data_format
EADAttack
bug fixesGenAttack
bug fixesPublished by jonasrauber over 4 years ago
Fixed the version number
Published by jonasrauber over 4 years ago
Fixes a bug in the BrendelBethgeAttack
and updated Numba to silence warnings.
Published by jonasrauber over 4 years ago
Fixes a bug in the BrendelBethgeAttack
(thanks @AidanKelley)
Published by jonasrauber over 4 years ago
py.typed
(file was missing in MANIFEST)data_format
(thanks @zimmerrol)Published by jonasrauber over 4 years ago
Foolbox 3 aka Foolbox Native has been rewritten from scratch with performance in mind. All code is running natively in PyTorch, TensorFlow and JAX, and all attacks have been rewritten with real batch support.
Published by jonasrauber over 4 years ago
foolbox.gradient_estimators
Published by jonasrauber over 4 years ago
Foolbox 3 aka Foolbox Native has been rewritten from scratch with performance in mind. All code is running natively in PyTorch, TensorFlow and JAX, and all attacks have been rewritten with real batch support.
Warning: This is a pre-release beta version. Expect breaking changes.
Published by jonasrauber over 4 years ago
randomgen
versionPublished by jonasrauber almost 5 years ago
EnsembleAveragedModel
(thanks to @zimmerrol)foolbox.utils.flatten
foolbox.utils.atleast_kd
foolbox.utils.accuracy
PyTorchModel
now always warns if model is in train mode, not just onceModelWithEstimatedGradients
Published by jonasrauber almost 5 years ago
foolbox.ext
namespacePublished by jonasrauber almost 5 years ago
foolbox.models.JAXModel
class to support JAX models (https://github.com/google/jax)preprocessing
argument of models now supports a flip_axis
key to support common preprocessing operations like RGB to BGR in a nice way. This builds on the ability to pass dicts to preprocessing
introduced in Foolbox 2.0.LocalSearchAttack
(thanks to @duoergun0729)foolbox.utils.samples
now warns if samples are repeatedfoolbox.utils.sampels
now uses PNGs instead of JPGs (except for ImageNet)