R Weekly 2020-04 Geographic distance, mutable feeling objects, Github Actions

27 Jan 2020

Release Date: 2020-01-27

This week’s release was curated by Jonathan Carroll, with help from the RWeekly team members and contributors.

Highlight

• Geographic distance

• Objects that feel mutable in R
• Deploy your bookdown project to Netlify with Github Actions

Insights

• Geographic distance
• Easily align ggplot2 titles, subtitles, and footnotes
• Create a second legend to bookend the top and bottom of a plot in ggplot2
• Three vectorised simulation approaches to the “Monty Hall Problem” using base R, data.table and tidyverse.
• Shiny dashboards using the googleAnalyticsR package!
• Exploring the Curse of Dimensionality - Part II.
• December 2019: “Top 40” New R Packages
• unpack Your Values in R
• Using unpack to Manage Your R Environment
• Why we wrote wrapr to/unpack
• Objects that feel mutable in R
• Generate synthetic data using R
• RStudio Projects and Working Directories: A Beginner’s Guide
• How to calculate landscape metrics for local landscapes?

• A brief account (via simulation) of the ROC (and its AUC)
• Calculating the Required Sample Size for a Binomial Test in R
• Explanatory Model Analysis with modelStudio
• Factors in R

Resources

• Data Science resources
• Tips and tricks in RStudio and R Markdown
• Descriptive statistics in R

New Packages

📦 Go Live for More New Pkgs 📦

CRAN

• {heddlr} 0.5.0: Bring a functional programming mindset to R Markdown document generation
• {gridtext} 0.1.0: Improved Text Rendering Support for ‘Grid’ Graphics

• {santoku} 0.3.0: A Versatile Cutting Tool
• {round} 0.12-1: Rounding to Decimal Digits
• {progressr} 0.4.0: A Unifying API for Progress Updates
• {fitzRoy} 0.3.1: Easily Scrape and Process AFL Data

GitHub or Bitbucket

• Convenience functions to print a one-time warning or message, or run arbitrary R code only once.

Updated Packages

• {bomrang} 0.7.0: Australian Government Bureau of Meteorology (‘BOM’) Data Client
• {GSODR} 0.2.1: Global Surface Summary of the Day (‘GSOD’) Weather Data Client
• {vctrs} 0.2.2: Vector Helpers
• {tidyr} 1.0.2: Tidy Messy Data
• {rlang} 0.4.3: Functions for Base Types and Core R and ‘Tidyverse’ Features
• {charlatan} 0.4.0: Make Fake Data
• {magick} 2.3: Advanced Graphics and Image-Processing in R
• {gh} 1.1.0: ‘GitHub’ ‘API’
• {dtplyr} 1.0.1; Data Table BackEnd for ‘dplyr’
• {janitor} 1.2.1: Simple Tools for Examining and Cleaning Dirty Data
• {c14bazAAR} 1.2.0: Download and Prepare C14 Dates from Different Source Databases

• {threejs} 0.3.3: Interactive 3D Scatter Plots, Networks and Globes
• {survey} 3.37: Analysis of Complex Survey Samples
• {RcppArmadillo} 0.9.800.4.0: ‘Rcpp’ Integration for the ‘Armadillo’ Templated Linear Algebra Library
• {anytime} 0.3.7: Anything to ‘POSIXct’ or ‘Date’ Converter
• {RPushbullet} 0.3.3: R Interface to the Pushbullet Messaging Service
• {RStudio Connect} 1.8.0

R Internationally

• German blog article: Why and when to create your own R package. Reasons and situations to create your own R package. Advantages of R packages over functions in scripts.

Tutorials

• Embedding Matrix on Tabular Data
• RStudio Projects and Working Directories: A Beginner’s Guide
• A simple example of interacting existing VBA scripts with R/Python.
• How to Remove Outliers in R
• Explainable AI (XAI)… Explained! Or How to whiten any Black Box with LIME
• Deploy your bookdown project to Netlify with Github Actions

R Project Updates

Updates from R Core:

Upcoming Events in 3 Months

Events in 3 Months:

• Thinking about rstudio::conf 2020? See the full conference schedule!

• LondonR & R-Ladies to host ‘watch party’ at RStudio Conference, Jan 29th

• Tidyverse Dev Day, January 31

• Intro to R for Plant Pathologists, Feb 19

• satRday Johannesburg 2020

• satRday Newcastle upon Tyne, April the 4th, Call for speakers for #satRdayNCL closes 31st Jan.

• A list of R conferences and meetings

• This week’s local R-User and applied stats events

More past events at R conferences & meetups.

Datasets

• Airline Safety from Aviation Safety Network

Call for Participation

Your support will keep R Weekly team moving! 💡

Quotes of the Week

Little #rstats tip: While you can easily toggle the display of R code with `code_folding` in #rmarkdown, it's less easy to toggle the *output* of your code. Though there's a (relatively) straightforward way to do this using the beauty of `html_tabsets`. Here's a quick demo! pic.twitter.com/FP7JGsNVaV

— Brendan Cullen (@_bcullen) January 20, 2020

#RStats essential changes for R 4.0.0: since matrices are soon going to have class() of both 'matrix' & 'array', you should really update any logical tests from:

class(X)=='foo'

To:

inherits(X, 'foo')#rstatsmemes pic.twitter.com/hSICMhSh0s

— mikefc (@coolbutuseless) January 25, 2020