BridgeStan – efficient, in-memory access to the methods of a Stan model

BridgeStan is a library (with bindings available in Python, Julia, R, and Rust) that grants access to the methods of a Stan model, including log densities, gradients, Hessians, and constraining and unconstraining transforms.

This allows researchers and library authors to develop inference algorithms in higher-level languages and apply them to arbitrary Stan models, such as those in posteriordb.

Contents:

• Getting Started
  • Requirement: C++ toolchain
  • Downloading BridgeStan
  • Installing an Interface
  • Optional: Customizing BridgeStan
• Language Interfaces
  • Python Interface
  • Julia Interface
  • R Interface
  • Rust Interface
  • C API
• How It Works
  • Foreign Function Interface Overview
  • Developer Information
  • Testing
  • Documentation
  • Stan Implementation Details