/FedDWA

[IJCAI2023] "FedDWA: personalized federated learning with dynamic weight adjustment." by Jiahao Liu, Jiang Wu, Jinyu Chen, Miao Hu, Yipeng Zhou, Di Wu

Primary LanguagePython

FedDWA: Personalized Federated Learning with Dynamic Weight Adjustment

This repository is for the paper:

FedDWA: Personalized Federated Learning with Dynamic Weight Adjustmnet
Jiahao Liu, Jiang Wu, Jinyu Chen, Miao Hu, Yipeng Zhou, Di Wu.
IJCAI 2023

Setup

pip install -r requirements.txt

Training

You can run the codes in script.sh to reproduce the results using CIFAR10 and CIFAR100 datasets in the original paper. While for CINIC-10 and Tiny-ImageNet datasets, you may need to download these datasets by yourself. You can download these datasets from the following websites:

https://kaggle.com/datasets/mengcius/cinic10
http://cs231n.stanford.edu/tiny-imagenet-200.zip