Highlights From rstudio::global
R
data science
rstudio::global
, this year’s iteration of the annual RStudio conference, was a few weeks ago. Here were some highlights:
Talks
There were a few talks I really loved:
- Using R to Up Your Experimentation Game, by Shirbi Ish-Shalom. On experimentation, sequential testing, taking big swings, and being statistically rigorous
- Maintaining the House the Tidyverse Built, by Hadley Wickham. On building and maintaining the Tidyverse, and what package maintenance in the real world is like when you have millions of downloads.
- oRganization: How to Make Internal R Packages Part of Your Team, by Emily Riederer. On how using internal packages (like
collegeviner
at CollegeVine!) can improve your R workflow and make teamwork in R dramatically easier, smoother, and more efficient. - Fairness and Data Science: Failures, Factors, and Futures, by Grant Fleming. On model fairness, bias, and evaluation techniques, and why they’re important to get right.
Cool New Things
finetune
, Max Kuhn’s newtune
-adjacent package, is live (albeit a little buggy)! It has some cool new model tuning algorithms, including racing methods withtune_race_anova()
andtune_race_win_loss()
, in addition to my personal favorite:tune_sim_anneal()
for Simulated Annealing! Link to the talk- Major improvements to
shiny
, including some serious caching upgrades that’ll improve performance dramatically! Link to the talk
Other Highlights
- Meeting a bunch of people in the breakout sessions! This year, there were virtual “tables” where you could drag your avatar to “sit down”, and once you were close enough to a table you could hear all of its conversation.