This release of ART 1.7.1 provides updates to ART 1.7.
Added
- Added wrapper
Mp3CompressionPyTorch
forMp3Compression
to make it compatible with PyTorch-specific attack implementations. (#1210) - Added new install option
non-framework
tosetup.py
to install all non-framework dependencies of ART. (#1209) - Added wrapper
VideoCompressionPyTorch
forVideoCompression
to make it compatible with PyTorch-specific attack implementations. (#1210)
Changed
- Changed
Mp3Compression
to add back reapplication of normalization to the compressed result. (#1210) - Changed
KerasClassifier.fit
to use batching provided by the methodfit
of the Keras model. (#1182)
Removed
[None]
Fixed
- Fixed bug of not passing user-provided device type, and instead always using default
gpu
, to standardisation preprocessor in allPyTorchEstimator
by using user-provided device type. (#1223) - Fixed bug in method
BaseEstimator.fit_generator
for fitting generators in cases where preprocessing is defined to not apply preprocessing twice. (#1219) - Fixed bug in
ImperceptibleASRPyTorch
to prevent NaN loss value for batch size larger than 1 by removing unnecessary zero-padding. (#1198) - Fixed two bugs in
OverTheAirFlickeringPyTorch
by making sure that the regularization norms are computed over the whole batch of perturbations, rather than per sample's perturbation and second that the "roll" operations are performed over the batch samples, rather than over the frames. (#1192) - Fixed bug in
SpectralSignatureDefense
, that lead to rejections of all clean images, by correctly indexing the label data. (#1189) - Fixed bug of accidentally removed checks for
apply_fit
andapply_predict
properties of framework-independentPreprocessor
tools inPyTorchEstimator
andTensorFlowV2Estimator
. With the bug thePreprocessor
tools were always applied in methodsfit
andpredict
independent of the values ofapply_fit
andapply_predict
. (#1181) - Fixed bug in
MembershipInferenceBlackBoxRemove.infer
by removing unnecessary shuffling of the test data. (#1173) - Fixed bug in
PixelAttack
andThresholdAttack
by casting input data to correct dtype. (#1175)