For most users, the recommended installation is the released version of randomForestSGT from CRAN.

Recommended: Install from CRAN unless a package maintainer has specifically asked you to test unreleased source code.

Install from CRAN

Open R or RStudio and run:

install.packages("randomForestSGT")

R installs randomForestSGT together with any required CRAN dependencies that are not already installed. For a normal CRAN installation, the dependencies do not need to be installed separately.

After installation, load the package and check the installed version:

R version and system requirements

The exact minimum R version is recorded in the package metadata and may change from one release to another. The current requirement is always shown in the Depends field on the randomForestSGT CRAN page.

You can check the version of R currently running with:

Begin by running the CRAN installation command above. If R reports that randomForestSGT is not available for your version of R, compare your R version with the current requirement on CRAN. Update R if necessary, restart R or RStudio, and run the installation command again.

randomForestSGT and some of its dependencies contain compiled code. CRAN normally provides precompiled binary packages for supported Windows and macOS versions. When a binary is available, most users on those systems do not need to install a compiler. Linux users, and anyone installing from source, need the standard development tools for their platform.

Updating the package

To install the newest CRAN release, rerun the same command:

install.packages("randomForestSGT")

R will replace an older installed version with the current CRAN version.

Development source code

The public GitHub repository is intended for development and testing. Most users should install the CRAN release. Install from GitHub only when a package maintainer has asked you to test a specific unreleased version and has provided any additional dependency instructions needed for that version.

The source code is available at github.com/kogalur/randomForestSGT.

Troubleshooting

“Package ‘randomForestSGT’ is not available for this version of R”

Check the current Depends entry on the randomForestSGT CRAN page. If your version of R is older than the version listed there, update R, restart R or RStudio, and run the CRAN installation command again.

A required dependency is missing or out of date

A normal CRAN installation resolves required dependencies automatically. First restart R or RStudio and rerun:

install.packages("randomForestSGT")

If the error identifies a particular dependency, include the complete error message when requesting help rather than trying to guess which development version is needed.

Compilation fails or the installation ends with “non-zero exit status”

First confirm that you are using the CRAN installation command and a currently supported version of R. On Windows and macOS, use the CRAN binary when one is available. A source installation requires the appropriate compiler and development tools for the operating system.

R reports a locked library, permission error, or package already in use

Restart R or RStudio and try again. If the error identifies a temporary 00LOCK directory left by an interrupted installation, remove that directory from the affected R library only after all R sessions using that library have been closed.

Reporting an installation problem

When requesting help, include the complete installation error and the output of:

Package-specific problems can be reported through the randomForestSGT issue tracker.



Cite this vignette as
H. Ishwaran, M. Lu, and U. B. Kogalur. 2025. “randomForestSGT: installing randomForestSGT vignette.” http://www.randomforestsgt.org/articles/installation.html.

@misc{LuGettingStarted,
    author = "Hemant Ishwaran and Min Lu and Udaya B. Kogalur",
    title = {{randomForestSGT}: installing {randomForestSGT} vignette},
    year = {2025},
    url = {http://www.randomforestsgt.org/articles/installation.html},
    howpublished = "\url{http://www.randomforestsgt.org/articles/installation.html}",
    note = "[accessed date]"
}