Songyue-Guo/FedGR

DASFAA2023 FedGR Code Repository. Federated learning for double unbalance settings (sample quantities imbalance for different classes in client and label or class imbalance for different client cross-client)

33
/ 100
Emerging

This is a federated learning framework designed for machine learning researchers and practitioners who build models using data distributed across multiple devices or organizations. It addresses common challenges where data imbalances exist both within each client's dataset (e.g., more images of cats than dogs) and across different clients (e.g., some clients have very few examples of a specific class). You input distributed datasets with these double imbalance characteristics, and it outputs improved, more accurate federated learning models.

No commits in the last 6 months.

Use this if you are developing federated learning models and frequently encounter situations where your training data is unevenly distributed across clients and also imbalanced in terms of class representation within each client's dataset.

Not ideal if your machine learning problem does not involve federated learning, or if your distributed data is well-balanced across clients and within each client's specific classes.

federated-learning distributed-machine-learning imbalanced-data-handling model-training privacy-preserving-ai
Stale 6m No Package No Dependents
Maintenance 0 / 25
Adoption 7 / 25
Maturity 16 / 25
Community 10 / 25

How are scores calculated?

Stars

41

Forks

4

Language

Python

License

Apache-2.0

Last pushed

Feb 26, 2023

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/ml-frameworks/Songyue-Guo/FedGR"

Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.