Gaussian Processes on Graphs and Lattices in 'Stan'

Gaussian processes are flexible distributions to model functional data. Whilst theoretically appealing, they are computationally cumbersome except for small datasets. This package implements two methods for scaling Gaussian process inference in 'Stan'. First, a sparse approximation of the likelihood that is generally applicable and, second, an exact method for regularly spaced data modeled by stationary kernels using fast Fourier methods. Utility functions are provided to compile and fit 'Stan' models using the 'cmdstanr' interface. References: Hoffmann and Onnela (2025) .

gptoolsStan

gptoolsStan is a minimal package to publish Stan code for efficient Gaussian process inference. The package can be used with the cmdstanr interface for Stan in R.

Getting Started

1. Install cmdstanr if you haven't already (see here for details).
2. Install this package by running install.packages("gptoolsStan").
3. Compile your first model.

library(cmdstanr)
library(gptoolsStan)

model <- cmdstan_model(
  stan_file="path/to/your/model.stan",
  include_paths=gptools_include_path(),
)

For an end-to-end example, see this vignette. More comprehensive documentation, including many examples, is available although using the cmdstanpy interface for Python.