Skip to content

Latest commit

 

History

History
16 lines (7 loc) · 1.25 KB

README.md

File metadata and controls

16 lines (7 loc) · 1.25 KB

pdphmc

Piecewise deterministic processes using Hamiltonian dynamics

This package has been discontinued - the current development of numerical generalized randomized HMC is at https://github.com/torekleppe/pdmphmc, which constitute an almost complete rewrite of the methology

This is the development version of the R-package pdphmc, which is used in the paper "Connecting the Dots: Towards Continuous Time Hamiltonian Monte Carlo" by Tore Selland Kleppe https://arxiv.org/abs/2005.01285

See also https://github.com/torekleppe/PDPHMCpaperCode for the model- and data files, and the specific version of the package used in the above paper.

The R-package has currently been tested on mac (using the clang compiler and R 4.x), linux (using native gcc and R 4.x) and windows 10 (R 4.x and using the mingw64 compiler that ships with Rtools4.0). In general, it requires that you have a working installation of R>=4.0.0 and a working installation of the R-package rstan.

Installation of the package is most easily done using the devtools function install_github(), e.g.: devtools::install_github("https://github.com/torekleppe/pdphmc")