RcppSparse

• Documentation
• Example
• Install: install.packages("RcppSparse")

RcppSparse provides a seamless Rcpp object class for R sparse matrix objects. The RcppSparse::Matrix class can directly import Matrix::dgCMatrix-class objects from R without any copying by simply using base Rcpp types (IntegerVector and NumericVector).

The result is a constant, by-reference, zero-copy Rcpp sparse matrix class. In contrast, RcppArmadillo and RcppEigen are deep copies.

Install RcppSparse from CRAN with install.packages("RcppSparse") and then load the header library in your C++ file:

//[[Rcpp::depends(RcppSparse)]]
#include RcppSparse.h // this line before ever including Rcpp.h

Now you're set to roll! Here's a simple program to compute column sums of a sparse matrix:

//[[Rcpp::export]]
Rcpp::NumericVector columnSums(RcppSparse::Matrix& A) {
    Rcpp::NumericVector sums(A.cols());
    for (size_t col = 0; col < A.cols(); ++col)
        for (RcppSparse::Matrix::InnerIterator it(A, col); it; ++it)
            sums(col) += it.value();
    return sums;
}

To use this function from the R end, just make sure you pass a dgCMatrix into the function:

library(Matrix)
A <- rsparsematrix(nrow = 10, ncol = 10, density = 0.1)
class(A) # this is a "dgCMatrix"
columnSums(A)

The idea of a zero-copy sparse matrix class is discussed on the Rcpp gallery.