Skip to content
This repository has been archived by the owner on Jun 9, 2020. It is now read-only.

Latest commit

 

History

History
36 lines (23 loc) · 1.32 KB

README.md

File metadata and controls

36 lines (23 loc) · 1.32 KB
Raw

gpu-stan-paper-materials

Replication scripts, measurement data, visualization scripts, and installation instructions for the paper "GPU-based Parallel Computation Support for Stan".

The replication scripts are split in two parts: the C++ function-level measurement scripts for the Cholesky decomposition function and the end-to-end R scripts that run Stan models with generated input data.

Cholesky decomposition C++ measurement scripts

Link

Prerequisites to run:

  • C++ toolchain (make and a C++ compiler)
  • GPU driver with OpenCL support
  • A copy of the Stan Math library of version 3.1.1 or newer

These scripts run three time measurements on the C++ level for the Cholesky decomposition function:

  • function evaluation when the input is a matrix of doubles
  • function evaluation when the input is a matrix of stan::math::var
  • gradient evaluation

End-to-end tests for Stan models

Link

Prerequisites to run:

  • a R enviroment and a C++ toolchain
  • GPU driver with OpenCL support
  • cmdstanr R package

These scripts run three time measurements on the C++ level for the Cholesky decomposition function:

  • function evaluation when the input is a matrix of doubles
  • function evaluation when the input is a matrix of stan::math::var
  • gradient evaluation