Implementation of Ordered Dropout (OD) for different type of NN layers in PyTorch.
The repo contains:
- layers: Linear, CNN, LSTM,
- simple example to showcase OD: example.ipynb.
pip install -r requirements
jupyter notebook
If you find this repo useful, please cite the paper that introduced notion of Ordered Dropout:
@article{horvath2021fjord,
title={Fjord: Fair and accurate federated learning under heterogeneous targets with ordered dropout},
author={Horvath, Samuel and Laskaridis, Stefanos and Almeida, Mario and Leontiadis, Ilias and Venieris, Stylianos and Lane, Nicholas},
journal={Advances in Neural Information Processing Systems},
volume={34},
pages={12876--12889},
year={2021}
}