The package is a R wrapper for a temporal progression model based on a generalized logistic curve with patient and study random effects on progression baseline and progression rate. The model also supports an arbitrary number of covariates linked with baseline and rate with a linear term and an additional 2-parameter Bateman-type (equlibriation and elimination rate) placebo term. The package uses Stan (cmdstan) for the back-end estimation. It is tailored specifically to be used on C-Path disease data. The modifications to stan include explicit derivation of the model derivative implemented in C++ and OpenCL to achieve maximum performance.
Make sure that you have R package devtools installed. You can install this package by executing the following command in R:
devtools::install_github("bstatcomp/project-cpath")If you have a GPU available in your system and the drivers properly installed, the package will automatically detect the GPU and try to use to maximum effect.
The package currently supports Windows and Ubuntu OS.