Workshops

Dates: Jan 3-5, 2025.

Venue: Department of Statistics, Banaras Hindu University, Varanasi, India.

Topic: Boosting R Code Performance via C++ Integration within RStudio

Pre-requisites: R beginner, familiarity with R objects, R functions. No prior knowledge in C++ (CPP) is required. Participants are encouraged to bring their own laptop (optional) with RStudio installed; along with an established internet connection (while attending the session; optional).

Abstract: Despite being the most popular coding platform among statisticians, R suffers from several drawbacks that hinder its full potential, particularly for computationally demanding algorithms. These drawbacks stem from R's poor performance in essential BLAS operations (e.g., matrix inversion, SVD) and its inefficiency in handling nested for-loops, making it significantly slower compared to Python,  MATLAB, Julia, and a few other coding platforms. To address this gap, the computationally demanding operations of a function can be coded in C++ within the RStudio platform and subsequently called from the original master R file. This approach significantly reduces computation time, enhancing speed by up to 10–50 folds.