I'm going to try the problems in #AdventofCode this year - we'll see how far I get in #rstats! Here's an opportunity I took to use 📦 slider by Davis Vaughan and avoid thinking too hard...

Here's my hasty solutions to #AdventOfCode2021 Day 8, in #rstats * Vectorised stringr functions can do a lot for you, especially str_split * Use the test data! I was rushing part 1 and missed a typo, which puts you on a 1 minute timeout * Indexing helps map output to numbers

My Day 14 solutions in #rstats to #AdventofCode * You can use purrr::compose() to repeat a function many times! * tidyr::hoist() enables easy conversion of nested lists to the dataframe structure you want github.com/Zedseayou/adve…

Day 15 solutions in #rstats to #AdventofCode Making use of Thomas Lin Pedersen's great tidygraph📦 to run shortest-path finding. A lot of the work becomes parsing input into a usable format! The test data for the second part also has multiple same-length paths... github.com/Zedseayou/adve…

A colossal Great Plains dust storm. Incredible imagery.

If you use {R6}’s implementation of encapsulated OOP in R, and sometimes you may wish to use more idiomatic functional OOP with such objects. The eponymous function from {R62S3} 📦 generate S3 generics/methods for R6 classes 🔀 xoopr.github.io/R62S3/referenc… #rstats #DataScience

My solution to Day 25 of #AdventofCode in #rstats (just part 1, since I still have 4 stars to get) * lead/lag + coalesce + first/last handle the wraparound well * gganimate 📦 is easy and fun to use! Shows cool patterns as the groups weave and unweave github.com/Zedseayou/adve…

Hunga Tonga-Hunga Ha'apai #volcano is erupting again. Huge shockwave. Tsunami warning. Images courtesy Himawari-8

Fields are basic units of agriculture. 🌾 Digital field boundaries are needed to deliver field-level analytics to farmers. But most of the world lacks field boundary datasets. A thread on filling the data gap with satellites and deep learning, from our latest paper. 🛰 🧵 1/n

A student did this on their exam code, I assume accidentally (?) and it works! In feols you can pipe in data and it will place it correctly even though data isn't the first argument in feols! Laurent Bergé you tricksy genius

I thought I'd try to reconstruct Bryan's discussion with a 13-year-old about the minimum wage based on the below tweet: (Bryan Caplan let me know if anything is inaccurate)

Anecdotally the "general languor without having Levine’s lunchtime-adjacent fillip enter their inbox" hypothesis is checking out

I’m tired of lying about this. I need to share my truth with the world: It never really looked like a duck.