Skip to content
/ psp Public

Implements an n-dimensional parameter space partitioning algorithm for evaluating the global behaviour of formal computational models as described by Pitt, Kim, Navarro and Myung (2006)

License

Notifications You must be signed in to change notification settings

lenarddome/psp

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

R-CMD-check GitHub release (latest SemVer including pre-releases) CRAN/METACRAN CRAN/METACRAN

Parameter Space Partitioning MCMC for Global Model Evaluation (Pitt, Kim, Navarro & Myung, 2006).

CRAN page of psp

To cite package ‘psp’ in publications use:

Lenard Dome and Andy Wills (2021). psp: Parameter Space Partitioning MCMC for Global Model Evaluation. R package version 0.1. https://CRAN.R-project.org/package=psp

A BibTeX entry for LaTeX users is

  @Manual{,
    title = {psp: Parameter Space Partitioning MCMC for Global Model Evaluation},
    author = {Lenard Dome and {Andy Wills}},
    year = {2023},
    note = {R package version 1.0.2},
    url = {https://CRAN.R-project.org/package=psp},
  }

Disclaimer

This package only recieves maintenance updates. It is not actively developed. If you would like to contribute, please contact me (@lenarddome).

Install

For the stable version

install.packages("psp")

For the developmental version:

devtools::install_github("lenarddome/psp")

About psp

Philosophy

A big influence on this implementation is an instantiation of the Open Models Initiative, catlearn. Watch the talk of Andy Wills: “The OpenModels project” from Open Research Working Group (ORWG) virtual meeting 08/09/20.

The project's architecture is also influenced by DEoptim. DEoptim implements a Differential Evolutionary Optimization algorithm formodel-fitting.

We are completely open-source and free. Anyone can contribute. If you would like to raise an issue or contribute code, use Github, message or email me (@lenarddome).

Code Style

We use Google’s R Style Guide with some extra caveats:

  • Do not use roxygen. Write your documentation from scratch as an Rdocumentation file. It is desirable to avoid writing test units for converting your comments to documentation.
  • We have a 100 characters per line policy. You are still encouraged to keep line lengths to 80 characters, but sometimes going over 80 can be better than the alternative.
  • Use TODO, FIXME, HACK, BUG, WARNING, NOTE comments for code that is temporary, a short-term solution, or good-enough but not perfect.
    • TODO comments must adhere to the following format: ## TODO: change this to use matrix operations.
  • Use spaces!
    • Good practice: if (makeup == TRUE) list(variable = runif(100), id = rep(letters, 5)[1:100]).
    • Bad practice: if(makeup==TRUE) list(variable=runif(100),id=rep(letters,5)[1:100]).
  • Do not number your files. File naming conventions are pretty straightforward:
    • Files are prefixed by their relation to groups of functions. For example, the prefix psp means that the file contains function relevant for the parameter space partitioning routines.
    • Files use camelCase.
    • File names must not contain spaces.
  • Apply the same conventions to C++ code as well. While we do not militantly adhere to this, study Google's C++ Style Guide.

About

Implements an n-dimensional parameter space partitioning algorithm for evaluating the global behaviour of formal computational models as described by Pitt, Kim, Navarro and Myung (2006)

Topics

Resources

License

Stars

Watchers

Forks

Packages

No packages published