Skip to content
/ lgpr2 Public

R package for approximate GP modeling of longitudinal data

License

Unknown, MIT licenses found

Licenses found

Unknown
LICENSE
MIT
LICENSE.md
Notifications You must be signed in to change notification settings

jtimonen/lgpr2

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

74 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

lgpr2

Differences compared to the original lgpr package

  • lgpr2 uses CmdstanR as backend, whereas lgpr uses rstan
  • lgpr2 generates and compiles new Stan code each time a model is specified, whereas lgpr uses precompiled Stan code
  • lgpr2 uses approximate GPs (basis function implementation) whereas lgpr uses exact ones (covariance matrix implementation)
  • lgpr2 is not on CRAN
  • lgpr2 is a lot faster and scales to bigger data
  • lgpr2 has less features (for example, currently only supports Gaussian observation model)
  • Stan code generated by lgpr2 is more readable and can be easily customized
  • lgpr2 contains the projection predictive method additional model reduction technique

Requirements

Installation

devtools::install_github('jtimonen/lgpr2')

Reference

Juho Timonen & Harri Lähdesmäki, Scalable mixed-domain Gaussian process modeling and model reduction for longitudinal data, arXiv (2024)

About

R package for approximate GP modeling of longitudinal data

Topics

Resources

License

Unknown, MIT licenses found

Licenses found

Unknown
LICENSE
MIT
LICENSE.md

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published