pyINLA is a Python interface for the Integrated Nested Laplace Approximation (INLA) method, built to bring the power of Bayesian inference into modern Python-based workflows. It bridges INLA’s efficiency with Python’s versatility — enabling scalable, reproducible modeling in a wide range of applications.
Highlights
📘 Handbook Coming Fall 2025
A full handbook with examples, applications, and best practices is in development and will be released in Fall 2025. It will help both new users and advanced practitioners take full advantage of pyINLA.
🤝 Looking for Collaborations
- Researchers with potential applications in spatial modeling, time-series, epidemiology, and more
- Developers interested in extending or testing the Python interface
- Feedback and benchmarking across domains and computing environments
💡 Applications
We're especially interested in real-world uses in fields such as environmental science, geostatistics, health data, and any domain requiring fast Bayesian inference on large models.
📬 Stay Connected
If you’d like to receive updates about the project or the handbook release, or if you're interested in applying pyINLA or contributing, feel free to contact me at: