An R package for regularized Principal Component Analysis via Variational Bayes methods.
Davide Vidotto [email protected]
bayespca performs Bayesian estimation of weight vectors in PCA.
To achieve regularization, the method allows specifying fixed precisions
in the prior distributions of the weights; alternatively, it is possible
to implement Gamma priors on such parameters. The method allows
for variable selection through Automatic Relevance Determination.
Check the vignettes and package documentation for further details.
vbpcafor model estimationvbpca_controlfor settings of control parametersis.vbpcafor testing the classplotheatmapfor plotting the precision and weights matrices;plothpdifor plotting high probability density intervals
devtools::install_github("davidevdt/bayespca")
0.3.0
R (>= 3.3.3)
GPL-2