pinei/edgyr-pi

edgyR-pi: Docker Images for Raspberry Pi R Developers

edgyR-pi: Docker Images for Raspberry Pi R Developers

Introduction

The edgyR project was designed by M. Edward (Ed) Borasky [email protected] to make R a first-class language for edge computing developers (arm64/aarch64 architecture). It is forged to run in NVIDIA Jetson Nano cluster and applied to digital audio synthesis.

This edgyR-pi project is a minimalist port for Raspberry Pi developers.

How is edgyR pronounced?

There are (at least) five options:

1. ED-grrr

2. ED-gyre

3. EDGE-er

4. Edgier

5. Edgy R

It is not pronounced like any brand of peanut butter. Otherwise, it's your choice!

Getting set up

1. Buy a Raspberry Pi 4.

   • Raspberry Pi 4 has a quad-core ARMv8 64-bit CPU.

   • 4 GB RAM: This is the minimum recommended environment for edgyR. A 64 GB microSD card should be fine for docker use.

2. Set up Docker runtime

3. Just about everything you'd ever want is available via docker run or sudo apt install in Linux, and install.packages or devtools::install in R.

   If you want more - arm64 binaries downloaded from upstream projects or built from upstream source for

   • pandoc 2.11.2,

   • R 4.0.4 with tidyverse, R Markdown, ShinyⓇ, and reticulate,

   • RStudioⓇ Server 1.4,

   • libnode-dev for the V8 R package,

   • conda-forge,

   all on a Docker image, then read on!

4. Set up a swap memory (we will need for the building to avoid Out of Memory error)

   • As an Ubuntu user, I followed this tutorial from somewhere in Digital Ocean community.

   # if the system has any configured swap
   $ sudo swapon --show
   
   # see if there is no active swap
   $ free -h
   
   # check available space
   $ df -h
   
   # creates a file for the swap memory (just used 1 GB, didn't check the recommendations)
   $ sudo fallocate -l 1G /swapfile
   
   # check the current amount of space that was resserved
   $ ls -lh /swapfile
   
   # make accessible to root
   $ sudo chmod 600 /swapfile
   
   # mark the file as swap space
   $ sudo mkswap /swapfile
   
   # enable the swap file
   $ sudo swapon /swapfile
   
   # verify that the swap is available
   $ sudo swapon --show
   
   # also check with the free utility
   $ free -h
   

Licensing

Because the edgyR Docker image redistributes RStudio Server in binary form, it uses the same license as RStudio - the GNU Affero General Public License. This essentially means I have to make all the source code I've modified to anyone who wants it.

For edgyR, that comprises two places:

• the code in this repository, mostly Dockerfiles and bash scripts, and

• some directories on the internal builder images, which reside in the Docker Hub pinei organization. s

Contributing

Contributing to the project is really quite simple:

1. Read the code of conduct at https://github.com/pinei/edgyR-pi/blob/master/CODE_OF_CONDUCT.md.

2. Everything starts with an issue. See Always start with an issue for the philosophy.

   • Is the documentation unclear? File an issue.
   • Did you find a bug? File an issue.
   • Do you want to request a feature? File an issue.

Please, don't go through the mechanics of forking / pull requests even for trivial typo changes without filing an issue. File an issue.

Yes! I want more!

edgyr-pi's Issues

Update to RStudio v2022

Last tag: v2022.02.0+443