Awesome
Compression via Gradient Flow Preservation
This repository is the official implementation of Winning the Lottery Ahead of Time: Efficient Early Network Pruning published at ICML 2022.
Setup
- Install virtualenv
pip3 install virtualenv
- Create environment
virtualenv -p python3 ~/virtualenvs/EarlyCroP
- Activate environment
source ~/virtualenvs/EarlyCroP/bin/activate
- Install requirements:
pip install -r requirements.txt
- If you mean to run the 'Tiny-Imagenet' dataset: download and unpack in
/gitignored/data/, then replace CIFAR10 with TINYIMAGENET below to run. Additional datasets can be added in a similar way (Imagewoof, imagenette, etc.)
Image Classification
To reproduce Image Classification results refer to Image Classification
NLP
To reproduce NLP results on the PSMM network refer to the folder NLP
Cite
Please cite our paper if you use our code in your own work:
@InProceedings{earlycrop,
title = {Winning the Lottery Ahead of Time: Efficient Early Network Pruning},
author = {Rachwan, John and Z{\"u}gner, Daniel and Charpentier, Bertrand and Geisler, Simon and Ayle, Morgane and G{\"u}nnemann, Stephan},
booktitle = {Proceedings of the 39th International Conference on Machine Learning},
year = {2022},
volume = {162},
series = {Proceedings of Machine Learning Research},
publisher = {PMLR},
}