A unified framework of perturbation and gradient-based attribution methods for Deep Neural Networks interpretability. DeepExplain also includes support for Shapley Values sampling. (ICLR 2018)
This pull request is to enable installing DeepExplain via pip in default mode (i.e. not editable):
pip install git+https://github.com/marcoancona/DeepExplain.git
Before, this would result in the error ModuleNotFoundError: No module named 'deepexplain.tensorflow' as the packages are not all configured in setup.py
This pull request is to enable installing DeepExplain via pip in default mode (i.e. not editable):
pip install git+https://github.com/marcoancona/DeepExplain.git
Before, this would result in the error
ModuleNotFoundError: No module named 'deepexplain.tensorflow'
as the packages are not all configured in setup.py