epca
is an R package for comprehending any data matrix that contains
low-rank and sparse underlying signals of interest. The package
currently features two key tools:
sca
for sparse principal component analysis.sma
for sparse matrix approximation, a two-way data analysis for simultaneously row and column dimensionality reductions.
You can install the released version of epca from CRAN with:
install.packages("epca")
or the development version from GitHub with:
# install.packages("devtools")
devtools::install_github("fchen365/epca")
The usage of sca
and sma
is straightforward. For example, to find
k
sparse PCs of a data matrix X
:
sca(X, k)
Similarly, we can find a rank-k
sparse matrix decomposition by
sma(X, k)
For more examples, please see the vignette:
vignette("epca")
If you encounter a clear bug, please file an issue with a minimal reproducible example on GitHub.
Chen F and Rohe K, โA New Basis for Sparse PCA.โ (arXiv)