Mara Averick

Top Tweets of 2018

Sat, 29 Dec 2018 00:00:00 +0000

Top two tweets for each month in 2018, according to Twitter analytics. January Such a 🆒 viz ⚒: "Intro to gghighlight: Highlight ggplot lines & points w/ predicates" by @yutannihilation https://t.co/SPVeKw0UdN #rstats #dataviz pic.twitter.com/VsSpXUZn88 — Mara Averick (@dataandme) January 28, 2018 ICYMI, .@StephdeSilva's tips for getting to know your data (also makes for pretty solid relationship advice) 🤓💔https://t.co/LUZe6tpPNq pic.twitter.com/KEdIot3IGo — Mara Averick (@dataandme) January 21, 2018 February Now on CRAN 👇

reprex-ing with {datapasta} 🍝

Tue, 30 Oct 2018 00:00:00 +0000

Scenario 1: from spreadsheet to reprex So, you’ve been asked to make a reprex and you want to include a bit of data that you have in a spreadsheet.1 Meet {datapasta}, a package by Miles McBain that can make your life a whole lot easier. Once you’ve installed datapasta, you simply copy a selected number of rows from your spreadsheet (remember, this is a minimal reproducible example), and click the Paste as tribble option from the DATAPASTA section of the Addins dropdown.

🎨 Palettes - Archer poster edition

Thu, 19 Jul 2018 00:00:00 +0000

Jo-fai Chow’s rPlotter is described as “[w]rapper functions that make plotting in R a lot easier for beginners.” I came upon it in its best and highest form, though, as “another” colour-palette generator for R.1 To be specific, I was drawn in by Palette Tarantino, which you can catch the code for in Chow’s post, Towards (Yet) Another R Colour Palette Generator. rPlotter-generated colour palettes for: Reservoir Dogs, Pulp Fiction, Kill Bill, and Django Unchained

reprex-cellence 👑

Sun, 24 Jun 2018 00:00:00 +0000

In honour of the triumphant return of reprex to CRAN, let’s revisit what I refer to as Jenny Bryan’s keys to reprex-cellence.1 The three keys are as follows: code that actually runs code that I don’t have to run code that can be easily run2 Our example True to the ésprit de reprex, let’s look at a real world example to unpack why these attributes are so useful.

Hand-crafted curation: meta-#SoDS18

Tue, 05 Jun 2018 00:00:00 +0000

😎 Summer of Data Science (#SoDS18) is upon us, and (thanks to Renée (aka @BecomingDataSci)) there is already some great guidance out there. I’m a big fan of “mini” projects — and, unlike Highlander, I’m pretty sure there can only be one isn’t the name of the SoDS game. So, I wanted to share some ideas I have for potential micro-projects… 🙌 do something by hand I know, I know — this borders on heresy.

Ceci n'est pas un conf, 🎬✌️

Wed, 30 May 2018 00:00:00 +0000

This year’s rOpenSci unconf was held in Seattle, and (after the great time I had at last year’s event), I was thrilled to be able to go back for more! All the fun, with like 40% less mind-crippling impostor-syndrome anxiety.1 😬 Once again, Sean Kross did a great job of rounding up the projects, complete with emoji annotations. 💥 Interested in what we worked on at #runconf18? You can find the full list of projects here: https://t.

Uncounted casualties

Mon, 28 May 2018 00:00:00 +0000

There is a relationship between war (and death) and counting things.1 Some of the best known data visualizations pertain to the casualties of war (e.g. Florence Nightingale’s coxcomb chart, Diagram of the Causes of Mortality, 1858). When I came upon Lt. Colonel Albert G. Love’s 1931 “War Casualties”2 my curiosity was piqued. Focused on “the World War” (at that date there had only been one), Colonel Love describes his task in the preface:

NBA Horserace 🏇

Sat, 07 Apr 2018 00:00:00 +0000

Inspired by Laura Ellis (aka @LittleMissData’s) post, Data Storytelling, I decided to take Flourish for a spin to look at NBA-team win percentages over the course of the season thus far. 🏀 As advertised, Flourish is a tool for “powerful, beautiful, easy data visualisation” that lets you “quickly turn your spreadsheets into stunning online charts, maps and interactive stories.” Clearly someone on the Flourish team and I are on the same wavelength when it comes to carnival games and data visualization, since the animated rank or relative-score chart that caught my eye turned out to be named the Horserace template.

Sourcetree fetch rebase

Sun, 25 Feb 2018 00:00:00 +0000

Ever the acolyte of Jenny Bryan when it comes to all things workflow, I try to follow Happy Git and GitHub for the useR as much as possible. I’ve never been especially good at Git, per se, but the shell met my needs for a long time…a long time, that is, before I started working with a team that really uses GitHub for version control. So, I decided to give Sourcetree a whirl.

ggraph all grown up 🎈🦒

Fri, 23 Feb 2018 00:00:00 +0000

Thomas Lin Pedersen (known aliases: @thomasp85, data imaginist) has quite an array of cool packages, all of which you should check out. However, in celebration its CRANniversary 🎈 I thought ggraph deserved a roundup post of its own.1 Dispatches from a proud papa 👨 ggraph: A grammar of graphics for relational data The repo from whence it sprang… More cool R viz-ness from @thomasp85 "ggraph" https://t.co/ncCKDHlQAx #rstats #dataviz #ggplot2 pic.

Hammertime ✖️ tibbletime

Fri, 05 Jan 2018 00:00:00 +0000

A very silly bit of R fun, featuring 📦: tibbletime, tidyquant, and magick packages (as well as an awesome gif of MC Hammer by someone named bert). The code itself (see RPub here) is a combination of Tidy time series Analysis with tibbletime by Matt Dancho, and How to Add a Logo to ggplot by Magick by Daniel P. Hadley (in which you’ll learn how to make a very awesome Vincent-Vega version of an mtcars scatter-plot, below).

Git guides

Sat, 30 Dec 2017 00:00:00 +0000

A completely selfish collection of resources (in addition to the GitHub Guides, which are excellent) to turn to when I inevitably get myself into Git trouble… Git pretty by Justin Hileman "Git pretty" by @bobthecow 🙌 flowchart to follow so @jimhester_ doesn't have to 🚒 rescue my 🔥ed repos!https://t.co/NFjIxOqZ03 #git #github pic.twitter.com/nQhhznmsog — Mara Averick (@dataandme) December 20, 2017 Excuse me, do you have a moment to talk about version control?

2017 - A 🐦❤️ year in review

Wed, 20 Dec 2017 00:00:00 +0000

I have no words of wisdom when it comes to summing up 2017. 🙊 So, I guess I’ll let some algorithm do the talking.1 🤖💬 Behold 🙌: two Top Tweets for each month (according to Twitter analytics), some more R-related than others. January 🎊 RStudio IDE Easy Tricks You Might’ve Missed by Sean Lopp These are indeed handy! "RStudio IDE Easy Tricks You Might've Missed" by @lopp_sean https://t.co/JFlwhkyCzL via @rstudio #rstats pic.

Docker-izing your work in R 🐳

Mon, 20 Nov 2017 00:00:00 +0000

Updated: 2018-08-19 I am decidedly a novice when it comes to any and all things Docker (which, incidentally, is why I’ve been reading these posts). But, since it’s helpful for me to have all of these resources together in one place, I thought I’d share the love around.1 R Docker turorial for reproducible research. via rOpenSci Labs This is actually a series of 6 lessons put together at the rOpenSci unconf in 2016.

R-blog 📝 tips from an inveterate tweeter thereof

Fri, 10 Nov 2017 00:00:00 +0000

Preemptive peccavi This post is selfish. The following items are tips, only insofar as they are to my liking. What’s more, the material isn’t even original. I wrote these up in response to a thread on the RStudio Community site.1 So, consider yourselves warned. ☠️ 1. Put a bird on it. 🐦 Actually, not just a bird (damn my dedication to TV show references), I’d recommend having your twitter handle (if you’re on there) visible in writing.

leaRning out loud: 📽️ slides from EARL Boston

Sun, 05 Nov 2017 00:00:00 +0000

🎉 I gave a talk at EARL Boston, and somehow managed not to die! 😵 Couldn’t make it? Here are the slides from the aforementioned, non-fatal event for your enjoyment. They likely won’t make too much sense on their own, and I plan on doing a proper write-up at some point. But, until that time comes… I’ve also put a high(er)-resolution version of these same slides (with working links 🔗) up at: https://bit.

Refind 🔎♻️

Fri, 27 Oct 2017 00:00:00 +0000

It’s not much in the way of a post, but (as tweeted earlier 👇), I’ve been digging refind of late.1 .@refindcom is pretty dope- all the 🔗s I tweet get stashed in my profile there, so, 👉👀 https://t.co/zfHl3cs31a — Mara Averick (@dataandme) October 11, 2017 Yes, it’s just another bookmarking tool— or discovery engine, or whatever it is the kids are calling things these days… The point is, it takes all the links from my tweets 🐦, and puts them in a much more browsable format.

Papyrus 📜

Sat, 21 Oct 2017 00:00:00 +0000

I came upon a post by Andrew Heiss, Working with R, Cairo graphics, custom fonts, and ggplot, and couldn’t help but notice that the featured font was one I’d seen before… I thought I could forget about it— move on with my life. Maybe I could just load the tidyverse, recreate Andrew’s plot, and be done with the whole thing. library(tidyverse) # Create sample data set.seed(1234) # This makes R run the same random draw df <- data_frame(x = rnorm(100), y = rnorm(100)) # Create plot p <- ggplot(df, aes(x = x, y = y)) + geom_point() + labs(title = "This is a title", subtitle = "This is a subtitle") + annotate("text", x = 0, y = 0, label = "This is some text", family = "Papyrus", color = "darkred", size = 8) + theme_light(base_family = "Comic Sans MS") p But no— that was too easy.

Updating your version of Hugo for blogdown on Netlify

Fri, 20 Oct 2017 00:00:00 +0000

Fair warning ⚠️ All of the information that follows is indeed in blogdown: Creating Websites with R Markdown by Yihui Xie, Amber Thomas, and Alison Presmanes Hill. But, since this is the second time I got a bit lost in this process, I’m writing it up for my own (and anyone else’s incidental) edification.1 Plus, just look at all the failed deploys that happen when I mess it up…it renders my Netlify deploy panel hideous!

Goodreads 👍📚 Part 2: rvesting descriptions

Tue, 17 Oct 2017 00:00:00 +0000

Preamble Much of the code herein has been adapted from the excellent post “Where my girls at? Scraping Goodreads and using ML to estimate how many children’s books have a cetnral female character” by Giora Simchoni, which I highly recommend. I’ll also make reference to my earlier post using my Goodreads data, should you want to check that out. 😱 Why aren’t you using the API?! Calm down! Yes, I know there is a perfectly lovely Goodreads API– and, by “perfectly lovely” I mean for getting pretty much any data other than book descriptions (which, as you will soon discover, are the objects of my desire).

🙈 mousey methods

Sat, 14 Oct 2017 00:00:00 +0000

Why show us the wrong way? I know, this seems like a perverse manner of going about things, but these are ways of getting information that will help others to help you. This means that there’s a decent chance you’re new to this whole endeavor, and may need some stop-gap, familiar methods before you begin to feel comfortable with best practices. There are good reasons to steer clear of your 💽 mouse.

Keeping up with blogdown 🌟

Sat, 07 Oct 2017 00:00:00 +0000

Updated: 2019-03-05 Pro tip! Alison Presmanes Hill has been writing a series of posts, each a different “Spoonful of Hugo.” If you’re building your blog on Hugo, these will make your life much easier in the long run, and are not to be missed! As of today, the topics covered are: Archetypes The netlify.toml file Page bundles Troubleshooting your build Back to our regularly scheduled programming… I’ve loved blogdown 📦 for so long, that I was kind of shocked 😱 when I saw Yinui Xie’s post, Announcing blogdown: Create Websites with R Markdown, on the RStudio blog just last month!

Custom snippets in RStudio: ⚡ faster tweet chunks for all

Sat, 30 Sep 2017 00:00:00 +0000

✂️ Snippets in RStudio As Sean Lopp describes in his super helpful post, RStudio IDE Easy Tricks You Might’ve Missed ⭐: Code snippets are a shortcut to insert common boilerplate code. You might also refer to them as “text macros”1— either way, they’re useful for speeding up the process of doing a thing in RStudio that you do frequently. In my case, this includes inserting “tweet chunks” in blogdown posts using a Hugo shortcode.

😸 purrr-ty posts

Tue, 26 Sep 2017 00:00:00 +0000

Updated: 📆 2018-10-06. 🎉🐱 purrr-tiest cheat sheet Purrr royal decree (ok, I’ll stop with the 🐱 puns now), the purrr 📦 now has its very own official RStudio cheat sheet: Apply Functions Cheat Sheet The purrr package makes it easy to work with lists and functions. This cheatsheet will remind you how to manipulate lists with purrr as well as how to apply functions iteratively to each element of a list or vector.

StackHack: an alternate route to 💯 Stack Overflow points

Sat, 23 Sep 2017 00:00:00 +0000

Over on the new RStudio Community site1, a useful thread arose re. asking questions ❓ on this new site vs. Stack Overflow: Choosing between this site and StackOverflow for posting a question. I won’t bother trying to recap the nuances of the discussion here (since you can go read it directly; see 🔗 above), but I wanted to stash a bit of info here from one of my responses that might be helpful to new coders and/or users of Stack Overflow (SO).

RegEx resources

Thu, 21 Sep 2017 00:00:00 +0000

Updated: 📆 2018-04-06. Regular Expressions, RegEx, regexps– call them what you like, but no matter how you slice it, they’re useful af. After all, data spelunking is all about patterns, and that’s precisely what regexps are for: (wo)manhandling patterns in strings. Good Sirs Wickham and Grolemund, in their masterful tome, r4ds, describe them thusly: They [regexps] take a little while to get your head around, but once you understand them, you’ll find them extremely useful.

R workflow fun

Mon, 04 Sep 2017 00:00:00 +0000

I was excited to see Declutter your R workflow with tidy tools1, a preprint by Zev Ross, Hadley Wickham, and David Robinson, among the many excellent papers in the Practical Data Science for Stats PeerJ collection.2 Workflows, whether data-analysis-related or not, are never a one-size-fits-all matter of course. Among other things, the workflows here differ in scope (heck, one of them doesn’t even use R). That said, there’s lots to be learned from the processes, mistakes, and hard-wrought wisdom 🦉 of others.

tidyeval resource roundup

Sat, 26 Aug 2017 00:00:00 +0000

Updated: 2018-10-06 Though, naturally, the Honorable Professor Doctor Sir Lord General Hadley Wickham’s Programming with dplyr article is the ne plus ultra of getting oneself aquainted with the latest and greatest in tidyeval programming recipes, quoting, and quasiquotation; there are some other resources out there. After posting a link to Edwin Thoen’s ⭐️ Tidy evaluation, most common actions, I was asked if there were other tidyeval tutorials I knew of.

Onboarding with rOpenSci: Tandem tales of package review

Wed, 23 Aug 2017 00:00:00 +0000

On of the ways in which rOpenSci does such a stellar job of package “quality control” is through its open, peer-review onboarding process. I had the chance to take part in this by reviewing Nicholas Tierney’s awesome visdat package 📦. In what I’m told is an rOpenSci first, Nick and I both wrote up our experiences for the rOpenSci blog — giving you a peek into the reviewer and reviewee process.

Stringr Visual Cheat Sheet

Thu, 17 Aug 2017 00:00:00 +0000

So, kinda the whole point of this “making-a-blog” endeavor (aside from getting to play with Yihui Xie’s super awesome blogdown 📦) was to give a more permanent home to some of the resources I’ve found and/or tweeted. So, without further ado… stringr «aide-mémoire» This visual guide (or cheat sheet, if you prefer) was made by Lise Vaudor (@LVaudor on twitter), and originally shared in her blog post, Manipuler des strings avec R.

Visual ⁉️: A process in progress

Tue, 15 Aug 2017 00:00:00 +0000

I’ve always been a fan of marginalia.1 Likewise, I’ve been known to use a post-it or two (or five) from time to time. Why? Sometimes it’s just easier to communicate when you keep your visual options open.2 Such was the case the other day, when I had a question about how best to approach a problem in GitHub. (I’m not going to explain it in words, because it’s hard, and I’m lazy, and that’s how I ended up here in the first place!

👍📚 Part 1: rgoodreads 📦

Fri, 11 Aug 2017 00:00:00 +0000

🤓 What is “normal” reading? There are very few things considered “normal” (in the “ordinary” sense of the word) about my reading habits. Let’s just say that, when I saw #BookLoversDay was trending on Twitter, my first thought was: it me! However, it also served as a reminder of my struggles regarding reading and “normalcy” of the Gaussian distribution variety.1 My beef is not with books themselves, but rather the limitations of rating them on Goodreads using a restrictive 5-star scale.

Boston EARL Keynote

Thu, 10 Aug 2017 00:00:00 +0000

The rumors are true 😱 (ok, maybe there weren’t really any rumors), I’ll be speaking at the “Enterprise Applications of the R Language”, better known as EARL, conference in Boston. Technically, it’s a keynote (🗝📝). But, since I’ve never even given a regular note, I can’t really say whether or not there will be more 🔑s involved than usual.1 Actual real adult data scientist, David Robinson (aka @drob), will also be keynoting,2 which means you should totally go.

GIF-ted: from video to GIF in Photoshop

Tue, 25 Jul 2017 00:00:00 +0000

When I found out that Amelia McNamara and Aran Lunzer were about to release their extremely awesome, interactive essay on histograms, I was quick to offer my advice in my true area of expertise: making GIFs out of video for the internet. Amelia found my infinite wisdom to be helpful, and suggested that I do a blog post about it.1 And so, here we are… Big picture: I read a cool post by Christopher J.

Ceci n'est pas un conf.

Wed, 21 Jun 2017 00:00:00 +0000

Meta lesson learned

This is not the post I originally intended to write about the rOpenSci Unconf (and, I promise, this isn’t just a lame attempt to take the Magritte joke a step further). Rather, what follows is wisdom gleaned both at the unconf, and in in my (failed) attempts to write the aforementioned original post capturing my unconf experience.¹