This is an R interface to the CUDA software development kit (SDK ) and allows us to query and manipulate GPUs with R code. We can invoke code on the GPU and transfer data to and from the GPU and CPU. It is similar to Simon Urbanek's OpenCL package. The idea is to be able execute code on the GPU from within R. This offers a high-level programming interface to use GPUs within R. It provides a full set of bindings to the CUDA SDK so that others can experiment with different features of the SDK such as pinned memory, streams, events, and explore how to leverage these for greater performance. We have longer term goals which will allow us to compile (simple) R code into GPU kernels and also compile code that refers to the CUDA routines that work on the host. This is similar to libNVVM.
Package details |
|
---|---|
Author | Duncan Temple Lang, Paul Baines |
Maintainer | Duncan Temple Lang <duncan@r-project.org> |
License | BSD_3_clause+file LICENSE |
Version | 0.5-0 |
Package repository | View on GitHub |
Installation |
Install the latest version of this package by entering the following in R:
|
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.