CRISP: Hybrid Structured Sparsity for Class-aware Model Pruning

Introduction

In this repository you will find a pytorch implementation of CRISP for three models.

Getting Started

When using anaconda virtual environment all you need to do is run the following command and conda will install everything for you. See environment.yml:

conda env create --file environment.yml
conda activate crisp-env

To reproduce the results on the ResNet-50 benchmark you just need to run the following code:

chmod +x run_resnet_imagenet.sh
./run_resnet_imagenet.sh

Feel free to change the model and dataset type in the script.