We are pleased to announce the release of RxInfer.jl v4.0.0, which introduces significant enhancements to our probabilistic programming framework.
A bit of a background
RxInfer.jl is a Julia package designed for efficient and scalable Bayesian inference using reactive message passing on factor graphs. It enables automatic transformation of probabilistic models into sequences of local computations, facilitating real-time processing of streaming data and handling large-scale models with numerous latent variables.
Highlighted New Features
- Introducing a new approach to analyze the performance of RxInfer inference routines, with optional sharing capabilities to assist in debugging and support.
- A built-in hook is now available for monitoring inference performance metrics.
- Users can now disable error hints permanently using Preferences.jl, offering a customizable development experience.
- As usually we found many bugs but we also introduced new bugs for you to find them
And many more!
Enhanced Documentation
In tandem with this release, we’ve overhauled our documentation to improve accessibility and user experience. We:
- Transitioned from complex GitHub-hosted URLs to custom domain with more readable links
- Enhanced documentation structure for better search engine visibility, making it easier to find relevant information
Explore the updated documentation at docs.rxinfer.ml.
Enhanced Examples
In addition, explore a wide range of practical examples demonstrating RxInfer.jl’s capabilities in probabilistic programming and reactive message passing at examples.rxinfer.ml. These examples cover various topics, from basic models like Bayesian Linear Regression and Coin Toss simulations to advanced applications such as Nonlinear Sensor Fusion and Active Inference in control systems. Each example provides detailed explanations and code to facilitate understanding and practical application. Whether you’re a beginner or an experienced practitioner, these resources are designed to help you effectively utilize RxInfer.jl in your projects.
We encourage you to update to v4.0.0 and take advantage of these new features and improvements. As always, your feedback is invaluable to us. Please share your thoughts and experiences on this thread or open an issue on our GitHub repository.
Thank you for your continued support and contributions to the RxInfer community.