rcppmlpackexamples: Example of using mlpack via its C++ API from R
About
mlpack has supported header-only compilation since the 4.0.0 release. It also contains an R package that has been part of CRAN for years. Its focus is on access from R at the R prompt: underlying mlpack functions are properly wrapped.
Sometimes, however, we would like to access the C++ API of mlpack from the C++ side of an R package. This package provides a working demonstration of how to do this.
Scope
Currently, the package contains two examples covering
- linear regression (with optional regularization)
- random forest classification of a multinomial response
- loan default prediction using a decision tree
Installation
The package is on CRAN and can be installed via the usual
> install.packages("rcppmlpackexamples")The r-univere
page has source and binary builds from the repository version which
may differ from the CRAN version and shows how to install them. Lastly,
one can always rely on remotes::install_github():
> remotes::install_github("eddelbuettel/rcppmlpack-examples")Miscellaneous
We use the rcpp prefix to not conflate this repo with
the existing mlpack
examples repo.
Author
Dirk Eddelbuettel for this R package
The mlpack authors for all of mlpack
License
GPL (>= 2) for this package
MIT for mlpack