Kernel Ridge Regression using 'RcppArmadillo'
Print method for fitted Kernel Ridge Regression models
Summary method for fitted Kernel Ridge Regression models
Expose tunable parameters for "krr_reg"
Compute low-rank approximations(Nyström, Pivoted Cholesky, RFF)
Coef method for fitted Kernel Ridge Regression models
Compute Model Error for Kernel Ridge Regression Models
Compute Model Error (Generic)
Kernel Ridge Regression using the RcppArmadillo Package
Fit kernel ridge regression using exact or approximate methods
Kernel Ridge Regression
Kernel matrix construction for given datasets
Param method for fitted Kernel Ridge Regression models
Extract/print hyperparameters of fitted models
Plot method for fitted Kernel Ridge Regression (KRR) models
Predict responses for new data using fitted KRR model
Print method for approximated kernel matrices
Print method for kernel matrices
Provides core computational operations in C++ via 'RcppArmadillo', enabling faster performance than pure R, improved numerical stability, and parallel execution with OpenMP where available. On systems without OpenMP support, the package automatically falls back to single-threaded execution with no user configuration required. For efficient model selection, it integrates with 'CVST' to provide sequential-testing cross-validation that identifies competitive hyperparameters without exhaustive grid search. The package offers a unified interface for exact kernel ridge regression and three scalable approximations—Nyström, Pivoted Cholesky, and Random Fourier Features—allowing analyses with substantially larger sample sizes than are feasible with exact KRR. It also integrates with the 'tidymodels' ecosystem via the 'parsnip' model specification 'krr_reg', and the S3 method tunable.krr_reg(). To understand the theoretical background, one can refer to Wainwright (2019) <doi:10.1017/9781108627771>.
Useful links