The goal of armacmp is to create a DSL to formulate linear algebra code in R that is compiled to C++ using the Armadillo Template Library. It also offers an mathematical optimization that uses RcppEnsmallen to optimize functions in C++.

The scope of the package is linear algebra and Armadillo. It is not meant to evolve into a general purpose R to C++ transpiler.

It has three main functions:

  • compile compiles an R function to C++ and makes that function again avaliable in your R session.
  • translate translates an R function to C++ and returns the code as text.
  • compile_optimization_problem uses RcppEnsmallen and the functions above to compile continuous mathematical optimizations problems to C++.

This is currently an experimental prototype with most certainly bugs or unexpected behaviour. However I would be happy for any type of feedback, alpha testers, feature requests and potential use cases.

Potential use cases:

  • Speed up your code :)
  • Quickly estimate Rcpp speedup gain for linear algebra code
  • Learn how R linear algebra code can be expressed in C++ using translate and use the code as a starting point for further development.
  • Mathematical optimization with optimize

Installation

remotes::install_github("dirkschumacher/armacmp")

Caveats and limitations

  • speed: R is already really fast when it comes to linear algebra operations. So simply compiling your code to C++ might not give you a significant and relevant speed boost. The best way to check is to measure it yourself and see for your specific use-case, if compiling your code to C++ justifies the additional complexity.
  • NAs: there is currently no NA handling. In fact everything is assumed to be double (if you use matrices/vectors).
  • numerical stability: Note that your C++ code might produce different results in certain situations. Always validate before you use it for important applications.

Example

You can compile R like code to C++. Not all R functions are supported.

library(armacmp)

Takes a matrix and returns its transpose.

Or a slightly larger example using QR decomposition

API

armacmp always compiles functions. Every function needs to have a return statement with an optional type argument.

my_fun <- compile(function(X, y = type_colvec())) {
  return(X %*% y, type = type_colvec())
}

A lot of linear algebra functions/operators are defined as well some control flow (for loops and if/else). Please take a look at the function reference article for more details what can be expressed.

Optimization of arbitrary and differentiable functions using ensmallen

The package now also supports optimization of functions using RcppEnsmallen. Find out more at ensmallen.org.

All code is compiled to C++. During the optimization there is no context switch back to R.

Arbitrary function

Here we minimize 2 * norm(x)^2 using simulated annealing.

Optimizers:

  • Simulated Annealing through optimizer_SA
  • Conventional Neural Evolution optimizer_CNE

Differentiable functions

Here solve a linear regression problem using L-BFGS.

Optimizers:

  • L-BFGS through optimizer_L_BFGS
  • Gradient Descent through optimizer_GradientDescent

When does armacmp improve performance?

It really depends on the use-case and your code. In general Armadillo can combine linear algebra operations. For example the addition of 4 matrices A + B + C + D can be done in a single for loop. Armadillo can detect that and generates efficient code.

So whenever you combine many different operations, armacmp might be helpful in speeding things up.

We gather some examples on the wiki to further explore if compiling linear algebra code to C++ actually makes sense for pure speed reasons.

Contribute

armacmp is experimental and has a volatile codebase. The best way to contribute is to write issues/report bugs/propose features and test the package with your specific use-case.

Code of conduct

Please note that the ‘armacmp’ project is released with a Contributor Code of Conduct. By contributing to this project, you agree to abide by its terms.

References

  • Conrad Sanderson and Ryan Curtin. Armadillo: a template-based C++ library for linear algebra. Journal of Open Source Software, Vol. 1, pp. 26, 2016.
  • S. Bhardwaj, R. Curtin, M. Edel, Y. Mentekidis, C. Sanderson. ensmallen: a flexible C++ library for efficient function optimization. Workshop on Systems for ML and Open Source Software at NIPS 2018.