Rcpp

Why C++

One word - performance

Low level compiled language ( vs. R is interpreted)
Static type system ( vs. R is dynamic)
Rigorous language definition / standard (vs. R is ad hoc)
Extensive ecosystem of code / libraries
Standard Template Library (STL)

R and C++ - The old way

SEXP convolve2(SEXP a, SEXP b) 
{
  SEXP ab;

  PROTECT(a = AS_NUMERIC(a));
  PROTECT(b = AS_NUMERIC(b));
  int na = LENGTH(a);
  int nb = LENGTH(b);
  int nab = na + nb - 1;
  PROTECT(ab = NEW_NUMERIC(nab));
  
  double *xa = NUMERIC_POINTER(a); 
  double *xb = NUMERIC_POINTER(b);
  double *xab = NUMERIC_POINTER(ab);
  
  for(int i = 0; i < nab; i++) 
      xab[i] = 0.0;

  for(int i = 0; i < na; i++)
      for(int j = 0; j < nb; j++) 
          xab[i + j] += xa[i] * xb[j];
  
  UNPROTECT(3);
  return(ab);
}

R and C++ - The Rcpp way

using namespace Rcpp;

// [[Rcpp::export]]
NumericVector convolveCpp(NumericVector a, NumericVector b) 
{
    int na = a.size(), nb = b.size();
    int nab = na + nb - 1;
    
    NumericVector xab(nab);
    for (int i = 0; i < na; i++)
        for (int j = 0; j < nb; j++)
            xab[i + j] += a[i] * b[j];
    
    return xab;
}

What does Rcpp do?

Provides a much more elegant (less painful) interface between R and C++ replacing .C and .Call

Native C++ representations of all R objects (vectors, matrices, lists, S4, etc.)
- Straight forward access to object attributes
- Simple translation from and to R (as and wrap)
High level support for compiling / linking / wrapper creation
- evalCpp, cppFunction, sourceCpp
Rcpp Sugar brings some of R’s functional tools into C++

Examples

Simple expressions

Simple function (Fibonacci)

Performance

Rcpp and Armadillo

One of the reasons to use Rcpp is to have easier access to the C++ library ecosystem.

One example that is relevant to our discussion from last time is RcppArmadillo, which provides access to the Armadillo C++ library for linear algebra (a more expressive / higher level / R-like way of using BLAS / LAPACK in C++).

Example

rmvnorm