Tessera

Måns Thulin

@mansthulin.bsky.social

I work in statistics and AI. Consultant, teacher, researcher. #Rstats user. Book: https://modernstatisticswithr.com Homepage: https://mansthulin.se Co-founder of https://aireview.se

151 Posts253 Follower 104 Folgt

Måns Thulin @mansthulin.bsky.social · 2mo ago

The native pipe has finally become the default in new RStudio installs! 🥳 #Rstats github.com/rstudio/rstu...

Use native pipe (`|>`) instead of magrittr pipe (`%>%`) by default? · Issue #14965 · rstudio/rstudio

Native pipe was introduced in R 4.1, which was released in 2021. I understand it was too early to switch the default at the time of the native pipe support was added in RStudio, but maybe it's rath...

github.com

Dr Di Cook @visnut.bsky.social · 3mo ago

Super excited to get this at my door today! Ursula Laa and my book on exploring high-d data and models is in print! Book website is www.routledge.com/Interactivel... if interested I think my 30% discount code might work for you - msg me. Time to update dicook.github.io/mulgar_book/ too! #rstats

James Balamuta @coatless.bsky.social · 3mo ago

A month later, here is what happened to this work. blog.thecoatlessprofessor.com/posts/extrac... #rstats

Extraction with a Footnote – TheCoatlessProfessor

blog.thecoatlessprofessor.com

James Balamuta@coatless.bsky.social · 4mo ago

Making #rstats portable on #macOS has been the community's Sisyphean boulder since 2009. Now, download a tar.gz, extract, run from: github.com/portable-r/p... Signed, notarized, 4.3.0 - 4.5.3. Blog: blog.thecoatlessprofessor.com/programming/...

A macOS desktop showing two windows side by side. On the left, a Finder window displays the contents of a "portable-r-4.5.3-macos-arm64" folder, with a directory tree expanded under "library" revealing subfolders like base, boot, class, cluster, codetools, compiler, datasets, foreign, graphics, grDevices, grid, and jsonlite (which is expanded to show CITATION, DESCRIPTION, doc, help, and html items). On the right, a Terminal window (zsh shell) shows commands for downloading and extracting Portable R 4.5.3 for macOS ARM64 from GitHub using curl, verifying the R home directory with Rscript, and installing the jsonlite package. The terminal output confirms jsonlite was successfully downloaded (1.1 MB) and installed, with the final line printing {"portable":true} to verify the portable configuration is working.

Måns Thulin @mansthulin.bsky.social · 5mo ago

While we're discussing what we like and dislike in #Rstats, here's why I don't like tidymodels: mansthulin.se/posts/tidymo...

Why I don’t use {tidymodels} – Måns Thulin

mansthulin.se

Dirk Eddelbuettel @eddelbuettel.com · 5mo ago

Josh Goldstein emailed me a nice tip for @rdatatable.bsky.social chaining: if we start a chained `data.table` operation inside a set of parens, we are no longer subject to the 'REPL constraint' and can keep each operation on a line. See ALT text. #rstats Now in the pdf at github.com/eddelbuettel...

Diff of a change the 'Getting Started in R - Tinyverse Edition' manuscript with a nice data.table improvement: inside a pair of ( ) we can place each chained operation on its own line improving readability.

I.e. we have

cw21 <- ( # use '(' to keep ops on separate lines
cw[Time %in% c(0,21)] # i: select rows
[, weight := Weight] # j: mutate
[, Group := factor(Group)]
[, .(Chick,Group,Time,weight)] # j: arrange
[order(Chick,Time)] # i: order
[1:5] # i: subset
)

R Consortium @rconsortium.bsky.social · 5mo ago

R/Medicine CFP is open 🩺🧪 Deadline: March 6 - still time! Submit: Talks, Lightning Talks, Demos, Workshops - Using R + Shiny for health, lab, clinical data First-time speaker? Email for feedback: rmedicine.conference@gmail.com rconsortium.github.io/RMedicine_we... #rstats #datascience

R Medicine 2026 - main graphic - Call for Proposal, deadline March 6

Måns Thulin @mansthulin.bsky.social · 5mo ago

I love RStudio, but I'm flabbergasted by the fact that @posit.co still haven't made |> the default for the Ctrl+Shift+C keyboard shortcut, despite their using it in e.g. the tidyverse documentation and R4DS. #Rstats

Konsta Happonen@konsta.happonen.eu · 6mo ago

So, what's the next #rstats flame war on the schedule? Base |> vs. magrittr %>%?

Måns Thulin @mansthulin.bsky.social · 6mo ago

I've been playing around with {marginaleffects} in some projects lately, and I really like it. Lots of useful stuff in there! If you work with regression models and haven't checked it out already, I strongly recommend that you do so: marginaleffects.com/bonus/get_st... #Rstats #Databs

Quick start – Model to Meaning

marginaleffects.com

Andrew Heiss @andrew.heiss.phd · 8mo ago

Some closing thoughts for my students this semester on LLMs and learning #rstats datavizf25.classes.andrewheiss.com/news/2025-12...

Will you incorporate LLMs and AI prompting into the course in the future?
No.

Why won’t you incorporate LLMs and AI prompting into the course?
These tools are useful for coding (see this for my personal take on this).

However, they’re only useful if you know what you’re doing first. If you skip the learning-the-process-of-writing-code step and just copy/paste output from ChatGPT, you will not learn. You cannot learn. You cannot improve. You will not understand the code.

In that post, it warns that you cannot use it as a beginner:

…to use Databot effectively and safely, you still need the skills of a data scientist: background and domain knowledge, data analysis expertise, and coding ability.

There is no LLM-based shortcut to those skills. You cannot LLM your way into domain knowledge, data analysis expertise, or coding ability.

The only way to gain domain knowledge, data analysis expertise, and coding ability is to struggle. To get errors. To google those errors. To look over the documentation. To copy/paste your own code and adapt it for different purposes. To explore messy datasets. To struggle to clean those datasets. To spend an hour looking for a missing comma.

This isn’t a form of programming hazing, like “I had to walk to school uphill both ways in the snow and now you must too.” It’s the actual process of learning and growing and developing and improving. You’ve gotta struggle.

This Tumblr post puts it well (it’s about art specifically, but it applies to coding and data analysis too):

Contrary to popular belief the biggest beginner’s roadblock to art isn’t even technical skill it’s frustration tolerance, especially in the age of social media. It hurts and the frustration is endless but you must build the frustration tolerance equivalent to a roach’s capacity to survive a nuclear explosion. That’s how you build on the technical skill. Throw that “won’t even start because I’m afraid it won’t be perfect” shit out the window. Just do it. Just start. Good luck. (The original post has disappeared, but here’s a reblog.)

It’s hard, but struggling is the only way to learn anything.

You might not enjoy code as much as Williams does (or I do), but there’s still value in maintaining codings skills as you improve and learn more. You don’t want your skills to atrophy.

As I discuss here, when I do use LLMs for coding-related tasks, I purposely throw as much friction into the process as possible:

To avoid falling into over-reliance on LLM-assisted code help, I add as much friction into my workflow as possible. I only use GitHub Copilot and Claude in the browser, not through the chat sidebar in Positron or Visual Studio Code. I treat the code it generates like random answers from StackOverflow or blog posts and generally rewrite it completely. I disable the inline LLM-based auto complete in text editors. For routine tasks like generating {roxygen2} documentation scaffolding for functions, I use the {chores} package, which requires a bunch of pointing and clicking to use.

Even though I use Positron, I purposely do not use either Positron Assistant or Databot. I have them disabled.

So in the end, for pedagogical reasons, I don’t foresee me incorporating LLMs into this class. I’m pedagogically opposed to it. I’m facing all sorts of external pressure to do it, but I’m resisting.

You’ve got to learn first.

Måns Thulin @mansthulin.bsky.social · 8mo ago

This is so useful. I usually add custom information to the box shown when hovering, using the text geom. An example can be found here: www.modernstatisticswithr.com/eda.html#det... #Rstats #statsky #databs

4 Exploratory data analysis and unsupervised learning | Modern Statistics with R

modernstatisticswithr.com

Brian Romans@clasticdetritus.bsky.social · 8mo ago

recently discovered that with the 'plotly' package in R I can generate an interactive visualization where I can hover over a data point to see some basic info –– very useful for initial stages of data exploration (e.g., quickly/easily learning which data points are outliers) #rstats

screenshot of the bubble that pops up when hovering cursor over a data point in an R plot made with the 'plotly' package

Nicola Rennie @nrennie.bsky.social · 8mo ago

One of the things that has been on my to do list for a very long time, is building a gallery of all of the charts I've made across #TidyTuesday, #30DayChartChallenge, #30DayMapChallenge, and other miscellaneous projects 📊 And it's finally here! Link: nrennie.rbind.io/viz-gallery/ #DataViz #RStats

Data Visualisation Gallery

Gallery of data visualisations created by Nicola Rennie.

nrennie.rbind.io

Daniel Gutierrez @ddgutierrez.bsky.social · 9mo ago

Students often ask: “Is this model good enough?” My reply: “For what?” AUC, precision, F1—none of them matter unless you know what decision you're informing. Always tie metrics to action. #DataScience #MachineLearning #AI #RStats

Claus Wilke @clauswilke.com · 10mo ago

We need to have a conversation about random seeds. Don't use 42. blog.genesmindsmachines.com/p/if-your-ra...

If your random seed is 42 I will come to your office and set your computer on fire🔥

Figuratively. More likely you'll get a stern talking to.

blog.genesmindsmachines.com

Måns Thulin @mansthulin.bsky.social · 10mo ago

Chernoff faces for penguins. 🐧 This made my day.

Michael Friendly@datavisfriendly.bsky.social · 10mo ago

#rstats #dataviz A "line-up" test has been proposed as a human significance test: Can an observer spot a difference that rejects a null hypothesis? Here's glyphs for 20 penguins, representing the main variables with visual features. There are THREE multivariate outliers here. CAN YOU FIND THEM?

Måns Thulin @mansthulin.bsky.social · 10mo ago

A short tutorial on how to create gt cross-tables in one line of code using the #Rstats package ivo.table: mansthulin.se/posts/ivotab...

Måns Thulin @mansthulin.bsky.social · 10mo ago

Happy to hear this. Positron certainly has its merits, but I think RStudio is a much better choice for the quiet majority who use #Rstats for statistical analyses rather than software engineering.

Libby Heeren@libbyheeren.bsky.social · 10mo ago

Is #Positron going to replace #RStudio? 😱 (This is the worry I keep hearing lately 👀) Posit Software Architect @jmcphers.bsky.social says, "there are certainly no plans to get rid of RStudio." Watch the full #DSHangout video with Jonathan here: www.youtube.com/watch?v=1sROVZ83FNg #rstats #databs

Måns Thulin @mansthulin.bsky.social · 11mo ago

This is a great update! I really like how colour palettes are now part of themes and the ink and paper feature in themes.

Thomas Lin Pedersen@thomasp85.com · 11mo ago

I am beyond excited to announce that ggplot2 4.0.0 has just landed on CRAN. It's not every day we have a new major #ggplot2 release but it is a fitting 18 year birthday present for the package. Get an overview of the release in this blog post and be on the lookout for more in-depth posts #rstats

Vincent Arel-Bundock @vincentab.bsky.social · 12mo ago

{tinytable} 0.13.0 for #RStats can now draw interactive tables with sorting, filtering & pages! And the 📦 is still an ultra-simple, powerful, and 0-dep way to create beautiful tables in basically any format: tex, typ, docx, pdf, png, md, etc Gallery & Docs: vincentarelbundock.github.io/tinytable/

Norm Matloff (你有冇諗清楚呀?)@matloff.bsky.social · 12mo ago

#rstats #statistics I've released my new open source book, "Powered by Linear Algebra: the role of matrices and vector space in data science," at matloff.github.io/WackyLinearA.... Turns the classic LA course on its head! Still proves the theorems, but with a deep emphasis on applications.

matloff.github.io

Jonathan Kitt @jonathankitt.bsky.social · last yr.

Wow ... I'd never come across this use of with(), which allows combining the pipe and computing a correlation coefficient. Thanks to @mansthulin.bsky.social for explaining this in his fantastic book www.modernstatisticswithr.com! #Rstats #code #neverknew

Måns Thulin @mansthulin.bsky.social · last yr.

I really enjoyed attending and speaking at R/Medicine this year! I learned a lot. Huge thanks to the organisers! My talk "Bootstrap inference made easy" is now available online: www.youtube.com/watch?v=EeAt... #Rstats #Statsky

Bootstrap inference made easy: p-values and confidence intervals in one line of code

YouTube video by R Consortium

youtube.com

Peter Dalgaard @pdalgd.bsky.social · last yr.

For the #rstats nerds: 405 is "Method not allowed". That is why R 4.0.5 was nicknamed "Shake and Throw", due to this piece of dubious pedagogy: www.gocomics.com/peanuts/1978...

Peanuts by Charles Schulz for July 30, 1978 | GoComics

Read Peanuts—a comic strip by creator Charles Schulz—for today, July 30, 1978, and check out other great comics, too!

gocomics.com

Quite Interesting@qi.com · last yr.

HTTP Error 404 is famously ‘Not Found’. Error 418 is ‘I’m a teapot’, occasionally given in response when you ask a computer to make you a coffee.

zx8754 @zx8754.bsky.social · last yr.

#rstats base ?`$`

$library(dplyr) library(purrr) library(bench) set.seed(42) create_df <- function(rows, cols) { out <- replicate(cols, runif(rows, 1, 100), simplify = FALSE) out <- setNames(out, rep_len(letters, cols)) as.data.frame(out) } results <- press( rows = c(1000, 10000, 100000), cols = c(100, 1000, 10000), { d <- create_df(rows, cols) mark( d$f, d %>% .$f, d %>% pull(f), d %>% pluck("f"), min_iterations = 100 ) } ) ggplot2::autoplot(results)$ bench::mark output plot

Ilya Kashnitsky@ikashnitsky.phd · last yr.

#rstats TIL: if you need to grab one column from a data.frame/tibble as a vector, `purrr::pluck` is roughly an order of magnitude faster than `dplyr::pull`; you only have to quote the column name

Unit: microseconds
expr min lq mean median uq max
purrr::pluck 6.0 7.5 10.29280 8.2 10.8 1827.9
dplyr::pull 56.2 66.9 93.01648 71.3 91.3 9222.4
neval
10000
10000

Måns Thulin @mansthulin.bsky.social · last yr.

This seems like an awfully high RAM usage for a new #Rstats session in RStudio. 🤔 Anyone seen this before?

Screenshot of RStudio, showing that the environment is empty but that 17.12 GiB of RAM is being used.

Måns Thulin @mansthulin.bsky.social · last yr.

The most common question I get when teaching ggplot2: "Can we change the background colour?" Yes, we can - and much more. Here's how to create your own ggplot2 themes: 🔗 mansthulin.se/posts/ggplot... #Rstats #datasky

Creating a custom ggplot2 theme – Måns Thulin

mansthulin.se

Måns Thulin @mansthulin.bsky.social · last yr.

The new release of RStudio prints errors, warnings and messages in different colours. That's a great improvement! I teach introductory #Rstats a lot, and students *always* think messages are errors. Hopefully this will help.

Nicola Rennie @nrennie.bsky.social · last yr.

For the prompt of "Extraterrestrial" on Day29 of the #30DayChartChallenge, I decided to make a chart designed in the style of an extraterrestrial who has never heard of good data visualisation principles! 📊 How many chart crimes can you spot? 🕵️‍♂️ #DataViz #RStats #ggplot2 #Day29

A deliberately chaotic and cluttered chart mocking bad data visualisation practices, featuring overlapping symbols, unreadable labels, clashing colours, and dual y-axes. The design satirises poor visual clarity and overcomplication.

Måns Thulin @mansthulin.bsky.social · last yr.

I created the {boot.pval} #Rstats 📦 aiming to make it easier to run bootstrap tests in R. Here's an example showing how you can use it to implement a correlation test using the bootstrap: mansthulin.se/posts/custom...

Running bootstrap tests using {boot.pval} – Måns Thulin

mansthulin.se