TyMill/EdgeFL: Initial Release

Summary EdgeFL is an experimentation toolkit for federated learning (FL) in edge environments. This release provides a compact but extensible framework to run end-to-end FL simulations, generate synthetic client datasets, and evaluate collaborative training behavior across heterogeneous nodes. The project includes: Synthetic data generation utilities for edge sensing scenarios. A shared model abstraction (BaseModel) with reference implementations. Federated training primitives (ClientNode, FederatedServer) for round-based simulation. Multi-agent orchestration support (Agent, AgentCoordinator) to model adaptive participation and coordination dynamics. Key contributions in this release Initial public release of EdgeFL. Baseline FL workflow for training and validation across multiple clients. Modular package structure designed for research prototyping and educational use. Documentation and runnable examples to accelerate onboarding. Intended use This software is intended for: Research and teaching in federated learning and decentralized AI. Rapid prototyping of edge FL workflows. Reproducible simulation studies on participation, fairness, and communication-aware training. License Apache License 2.0 Project links Repository: https://github.com/TyMill/EdgeFL Documentation: https://https://tymill.github.io/EdgeFL/ Citation note If you use this software in academic work, please cite the Zenodo record corresponding to this release. With passion dr Tymoteusz Miller