This repository contains a reimplementation for paper "Simultaneously Learning Architectures and Features of Deep Neural Networks, ICANN 2019".
git clone https://github.com/vi2enne/Neural-Network-Pruning
cd Neural-Network-Pruning
Currently, the code supports Python 3
- numpy
- PyTorch (>=1.1.0)
- pandas
- thop
- sklearn
Download audio spectrograms data from ISMIR2018 Tutorial from https://github.com/slychief/ismir2018_tutorial
python train.py -sr --filter-percent 0.5
If you use this code, please cite:
@inproceedings{wang2019simultaneously,
title={Simultaneously learning architectures and features of deep neural networks},
author={Wang, Tinghuai and Fan, Lixin and Wang, Huiling},
booktitle={International Conference on Artificial Neural Networks},
pages={275--287},
year={2019},
organization={Springer}
}