Version 0.11.1

BoundaryAttack less verbose

Version 0.11.0

BoundaryAttack added: a new, powerful decision-based attack that does not rely on gradients or probabilities

Version 0.10.0

Performance improvements and minor changes. Easier to write fast attacks.

Version 0.9.6

several bugfixes

Version 0.9.5

Fixes a bug in ContrastAttack when the lower bound is non-zero (thanks to @whg1g15)

Version 0.9.3

MXNet model bugfixes
improved support for other Keras backends (in particular Theano; in addition to TensorFlow)

Version 0.9.2

Keras model improvements

Version 0.9.1

MXNet model improvements

Version 0.9.0

Preprocessing has been changed and fixed.

• Preprocessing can now be specified in all models
• Preprocessing must now be specified as subtraction and division only
• Gradients are now correctly backpropagated through the preprocessing
• Gradients are now tested numerically

Version 0.8.1

fix for the Readme on PyPI