It was such a pleasure to speak today at #PositConf2025 about how BADLY 😱 it went when I tried to learn Python ~10 years ago, and how tooling in the ecosystem has changed (gotten better!) in the intervening decade. You can check out my slides at: juliasilge.github.io/get-unstuck-...

It’s been a lot of fun working on this project and connecting marketing methodology with my political science roots.

There are a lot of introductions to Python, but this one is mine. It includes a walkthrough of using uv for environment management, Polars for data wrangling, seaborn.objects for visualization, and scikit-learn for modeling. #rstats #python #pydata

Headed back home after a whirlwind visit! Logan is a gorgeous city and USU is doing really neat work with teaching data analytics

PyData Northern Utah is partnering with HackUSU for a special in-person meetup: An introduction to Python dashboards using polars, seaborn.objects, and Quarto: www.meetup.com/pydata-north...

For our first PyData Northern Utah meetup of 2025, we are continuing to look at a data wrangling tools with @healthandstats.bsky.social as guide. For anyone in and around Cache Valley, join our Northern Utah chapter at www.meetup.com/pydata-north...

This may be Python ignorance on my part, but I'm wondering if there is something under the hood that makes Positron not play well with Jupyter notebooks? @coatless.bsky.social have you seen anything like this? github.com/posit-dev/po...

Lots of folks are discovering the relevance of decision theory to the practical problem of analyzing an A/B. A quick🧵

Woohoo, our paper is live! This haiku summarizes it best: Raw model results? Stop! Hard to understand! Use {marginaleffects}

No trick, all treats - posit::conf(2024) talks are now on YouTube! 🍬 Over a thousand people gathered in Seattle and online to dive into all things open-source data science. With 100+ talks, there's a lot to explore! Check out the playlist: www.youtube.com/playlist?lis... #RStats #Python

For anyone in and around Utah State University, come join a new chapter of PyData and learn about data wrangling with Polars next week! #python #databs

This administrative attack on the nonprofit sector fits with a pattern of authoritarian restrictions on civil society that's been going on for the past decade+ @suparnac.bsky.social and I (and others) have done a bunch of research on this from an intl/comparative perspective #nonprofitsky #polisky 🧵

My first screencast in quite a while! Take a first look at how to use #Positron, the new data science IDE I have been working on, for data analysis with #rstats, using this week's #TidyTuesday dataset on orca encounters 🐳 youtu.be/5BojM5EciPs

We've got a brand new, baby website for Positron! Take a look if you are interested in getting started, and please let us know how it goes: positron.posit.co

Just discovered the Causal Quartet by @lucystats.bsky.social, @travisgerke.bsky.social, and @malcolmbarrett.malco.io: four datasets that have the same unadjusted casaul relationship between treatment/outcome but different true effects. This is great for teaching DAG-based causal inference! #rstats

George Stagg just released a new #quarto extension that allows for an embedded full feature code editor inside of Quarto slide deck! It supports both #rstats via #webr and #python via #pyodide. Announcement: mastodon.social/@[email protected]... Extension: github.com/r-wasm/quart...

President Venn Diagram xkcd.com/2962

The International Society for Bayesian Analysis tells me Statistical Rethinking has won the 2024 DeGroot Prize for its contributions to "statistical inference, decision theory and statistical applications". This is huge honor especially given the previous winners who have influenced me so much.

@tjmahr.com Did you see the new way to install {cmdstanr} and other Stan friends with {renv} now? No more custom repo entry is needed—it just uses R-Universe now discourse.mc-stan.org/t/announceme...

New blog post! Read about Posit's new Positron editor, see some of the neat new features it has, and check out the settings and extensions I use. It includes a bonus workaround for connecting to a remote server with SSH! #rstats

#positron IDE beta has been updated since last week! Two nice changes: 1. Quarto Extension (not the CLI) is now bundled; and, 2. Prelim support for #polars in data explorer A full list of changes can be viewed using GitHub's compare url with tags. github.com/posit-dev/po...

Some dueling long-haul (~30min+) videos exploring the #Positron IDE for data wrangling and visualization. Both videos look at the new data explorer, plotting capabilities, projects, and graph support. #rstats: www.youtube.com/watch?v=-jhT... #Python: www.youtube.com/watch?v=izaZ...

Curious about Positron, @posit.co's new IDE? I try and summarize how it appears to bring together the best of RStudio and VS Code. #rstats #pydata #positron occasionaldivergences.com/posts/positr...

~3 minute-long video looking at #positron's #rstats debugging capabilities. In short: browser() and traceback() are supported with a nice explorer. Positron engages the debug mode with red outline. Care is needed when engaging debugger. No script breakpoints yet (?) youtu.be/p_4ZS-nnQ2Q

HELLO POSITRON #rstats Read about it: www.infoworld.com/article/3715... Download it: github.com/posit-dev/po...

It’s happening! It’s happening! www.infoworld.com/article/3715...

ICYMI StanCon '24 Sept 10-12 at the Oxford School of Witchcraft and Probabilistic Programming, now accepting proposals for contributed talks and tutorials. Hope to see Stan stans there #stats

Goodhart’s Law xkcd.com/2899

@mjskay.com has an official paper out for the ggdist package in R: pubmed.ncbi.nlm.nih.gov/37883271/ ggdist provides powerful tools for visualizing uncertainty. the paper has examples, lessons about package development, and can help you better understand grammar of graphics. #StatSky #rstats

Quarto 1.4 has been released 🎉 The release brings new formats for dashboards and Typst, a new manuscript project type, a cross-reference overhaul, Shiny for Python support, and a ton of other updates. Learn more on the Quarto blog: quarto.org/docs/blog/po...

There’s a good chance that I get to teach a course in causal inference in my new gig. What are the must-haves for a modern causal inference course? Especially for an audience that is primarily trained in ML techniques?

I'm excited to announce that I've accepted a position at Utah State starting this Fall. I'll be mentoring students in the Analytics Solutions Center in addition to the usual professorial responsibilities, and I couldn't be happier about it.

At least we didn’t have to give up being president of Harvard.

Oh no! The Posit Primers ({learnr} tutorials for basic tidyverse #rstats skills) have been discontinued and replaced with Posit Recipes, which are helpful, but fulfill a different need. I used these for all my R classes 😬😭 Fortunately the old primers are all on GitHub github.com/rstudio-educ...

Happy New Year, everyone!

Mike Godwin, author of Godwin's law: Go ahead and compare Trump to Hitler. www.washingtonpost.com/opinions/202...

Updated {macrtools} R package to 0.0.3! 🔧 - 🔄 macOS support extended to Sonoma (14.0.0). - ✨ Checks for Xcode app IDE presence. - 🔧 Improved error messages. - 📚 Docs refined Learn more at: github.com/coatless-mac... #rstats #macos #compiledcode

Cool kid custom handle time.

Happy Four Seasons Day to all who celebrate.

Creating reproducible environments for R workflows that include some Python might be niche, but it's my niche. Maybe it'll be useful for you as well. occasionaldivergences.com/posts/rep-env/

One of the core advantages of the Bayesian framework in industry experimental settings is that we can actually use what we learn from one experiment as prior information in follow-on experiments rather than pretending we forget everything at each iteration.

"This was a terrible tragedy, but sometimes these things just happen and there’s nothing anyone can do to stop them."

It’s happening! Quarto Dashboards are in the v1.4 pre-release! quarto.org/docs/dashboa...

🎯 Materials from our workshop on deploying ML models: posit-conf-2023.github.io/vetiver/ 🎯 Slides for my talk today on reliable maintenance of ML models: juliasilge.github.io/ml-maintenan...

It took me a few episodes, but I’m starting to really like Discovery. Yes, I know I’m late to the modern Star Trek party.

I told a colleague that logistic regression is AI and they got mad at me, so I made a chart. Find yourself. I am "Tinder is AI". (I posted this on that other site and a friend complained "y u no skeet this too!?!!?" so here you go)

Phew the fall 2023 version of my GSU #rstats dataviz class is up and live, now with shiny new catalog numbers (PMAP 8551 and 4551 (now for undergrads too!)) I haven't taught this as a fall class since 2018 so it'll be nice to have 15 weeks instead of 7 All content is CC-licensed, so have at it!

Does dreaming of real-world science conversations using Star Trek technobabble mean anything? I’m assuming there’s no such thing as an “ablative distribution.”