Exploring correlations in R with corrr and using R with Docker

This time we have two speakers again, Simon Jackson and Tamas Szilagyi.

1. Exploring correlations in R with corrr
   Simon Jackson, https://github.com/drsimonj

Correlations among variables form the basis of many statistical techniques and learning algorithms. Being able to explore correlations can, therefore, help in scenarios like model planning and diagnosis. R has implemented useful correlation methods but rapidly exploring the resulting matrices can be a pain.

For this reason, I wrote corrr, a tidyverse-style package for exploring correlations in R. It lets you work with data frames, instead of matrices, that are easy to manipulate and visualize with new functions or data-frame-centric tools like tidyverse packages dplyr, tidyr, and ggplot2. In this presentation we’ll talk about corrr, how it helps tackle the challenge of exploring correlations in R, features coming in the latest CRAN submission, and its use of the tidyverse philosophy.

As the topic has been changed, I’ll briefly cover where to get material on my previously-planned talk, “R from academia to commercial business”.

2. Using R with Docker Containers
   Tamas Szilagyi, http://tamaszilagyi.com

Docker containers are everywhere nowadays, and there are many use cases for containerizing R code. Encapsulating a shiny app inside a Docker container for example will ensure that your dashboard runs independently of host platform. Or, running data pipelines inside containers to guarantee reproducibility of the results.

Consequently, there are lots of tools available for R users to easily containerise R code and get up and running with Docker. In this talk, I will explain the benefits of using containers based on examples from the R universe, and share a few tips and tricks for efficient Docker based workflows.