Location via proxy:   [ UP ]  
[Report a bug]   [Manage cookies]                
Skip to content

[AISTATS 2024] Pathwise Explanation of ReLU Neural Networks

License

Notifications You must be signed in to change notification settings

Jo-won/PathwiseExplanation

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

3 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Pathwise Explanation of ReLU Neural Networks

Official Pytorch Implementation of Pathwise Explanation of ReLU Neural Networks


⏩ Installation

  • Create a conda environment (If you want, you can use a docker)
conda create -n pathwise_explanation python=3.8
conda activate pathwise_explanation
  • Intall Package (Maybe other packages are needed to implement this code, so you just install them by pip, like "pip install [PACKAGE]" since their version is not important.)
conda install pytorch==1.11.0 torchvision==0.12.0 torchaudio==0.11.0 cudatoolkit=10.2 -c pytorch
conda install numpy==1.21.6
conda install matplotlib==3.5.1
conda install captum==0.6.0
conda install scikit-learn==1.3.0
pip install opencv-python==3.4.8.29
pip install opencv-contrib-python==3.4.8.29
pip install captum==0.6.0

⏩ Data Preparation

Getting the Dataset

├── PathwiseExplanation
   ├── main.py
   ├── ...
   └── ILSVRC2012_selected_224/

Download pretrained weights

├── PathwiseExplanation
   ├── data
      └── ILSVRC2012_selected_224/

⏩ Impletementation

  • You can simply use the below line for all implementations of the proposed method according to depth and width.
bash run_imgSN_all.sh

👍 References

We referenced the repos below for implementing other methods or metrics.

About

[AISTATS 2024] Pathwise Explanation of ReLU Neural Networks

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages