Thank you for your interest in the Barkley Reference Architecture.
This repository is a research demonstrator that accompanies the Barkley framework paper. It is the minimal, runnable computational backbone for individual-referenced behavioral intelligence, built on fully synthetic data. It is not an actively developed community product, and it is not diagnostic.
Because of that scope:
- Questions, corrections, and discussion are welcome. Open an issue, or email labs@getbarkley.com. Conceptual questions about the framework, the methods, or the synthetic experiments are especially welcome.
- Small fixes (documentation, typos, reproducibility, obvious bugs) are welcome as pull requests. Please keep them focused and describe the motivation.
- Larger changes (new estimators, new pipeline stages, scope expansions) are best raised as an issue first, so fit can be discussed before you invest effort. This repository deliberately stays minimal and illustrative; not every good idea belongs here rather than in the broader (separate, non-public) Barkley system.
- Contributions are accepted under the repository LICENSE and the terms in NOTICE.md. Contributions must use synthetic data only - never real animal data of any kind.
(c) 2026 Elodie Aishwarya P. Remoissenet - Barkley AI - labs@getbarkley.com