This repository hosts examples, how-tos and tutorials for FEDn and STACKn.
The examples in this repository are all maintained by Scaleout and/or core developers in the community, and can be expected to be regularly updated with each new release of the software. Please read the README in each respective example for more information.
- mnist-keras: Getting started with FEDn (MNIST, Keras version)
- mnist-pytorch: Getting started with FEDn (MNIST, Pytorch version)
- imdb-keras: FEDn sentiment analysis, IMDB dataset.
- how-tos: Various tips-and tricks when using FEDn and STACKn.
Below we maintain a list of public examples and projects by external users and partners. If you have an example that you want to include, talk to a core developer in Discord and/or send a PR updating this README. To be included, please make sure to have a clear contact person listed for the example.
Federated learning/FEDn
- FedBird - Federated object detection for Baltic Seabirds FEDn project by AI Sweden and Zenseact.
- FedLM - Federatad Electra. FEDn NLP project by the Swedish Royal Library.
- Human activity recognition with a Keras CNN based on the casa dataset (cross-device)
- Fraud detection with a Keras auto-encoder (ANN encoder)
- VGG16 trained on cifar-10 with a PyTorch client (cross-silo)
- Sentiment analysis with a PyTorch CNN trained on the IMDB dataset (cross-silo)