Come learn about deep learning with us! The Data Science Learning Community just started reading the free third edition of Deep Learning with Python. Meetings are Tuesday at 3 pm CST. DSLC: dslc.io Book: deeplearningwithpython.io #rstats #pydata

One of my aims at the time was to make sure we can show this relationship in simulated data, inspired by @rmcelreath.bsky.social's approach to statistics. Statistical Rethinking is a life-changing book, this paper wouldn't be the way it is now without it. Very grateful to McElreath!

As the US economy screams ever higher on mostly AI bets, I feel obligated to put a marker down here and emphasize the business value of machine learning in the hope that this future AI winter will miss ML practitioners and researchers. AI winter is coming again when the bubble pops

Hudson Bay polar bears have nowhere to go (no ice of significance) & it's still much warmer than normal. When I started studying the bears here in 1984, they were back on the ice weeks earlier. At 1 kg mass loss/day, the bears would welcome some frigid weather.

“Tens or hundreds of millions of dollars of taxpayer-funded NASA property and laboratories are at risk of either being discarded, mishandled, or out-of-commission for significant time periods.” 🔭🧪 www.gesta-goddard.org/blog/gestas-...

Messy folders haunting your R projects? 👻 This Wednesday Oct 29 at 4:00 PM PDT, I'll lead a workshop on Efficient File Management in R with {fs} hosted by @r-ladies-stl.bsky.social. Let's clean and organize a spooky messy folder together! Register at www.meetup.com/rladies-st-l... #RStats #DataBS

You should read this wonderful little history of the #tidyverse, by @hadley.nz. It reminded me about my early #rstats days as a PhD student (2011 - 2016), where I was constantly trying out the new things Hadley and crew were cooking up. hadley.github.io/25-tidyverse...

A #Slurm user just confirmed that "yay it works. Pretty sick!" Thanks to excellent feedback from several users, it'll soon be even easier to distribute #rstats code via #HPC job schedulers using future.batchtools #parallel #futureverse

AI is powerful, but it's no free lunch - and again, it's no substitute for YOUR expertise.

R and QGIS, name a better combo #databs

“I used R the statistical programming language to analyse each of the 3-hourly netCDFs — a file format for storing multidimensional scientific data — and create a geoJSON file where the data was greater than 35C. These files were then loaded into Qgis and styled….” - @sdbernard.bsky.social #RStats

Data science junkies, get ready! 🚀 "The Test Set" #podcast trailer is here for your viewing pleasure. Tune in July 1st and every Tuesday after for new episodes with hosts @mchow.com, @hadley.nz, and @wesmckinney.com as they welcome thought leaders in #DataScience. Subscribe now: pos.it/thetestset

Bleeding edge update for the #tidyverse purrr package with even more seamless #rstats parallel maps. Introducing our shiniest new adverb: `in_parallel()`. Just wrap your function to take advantage of blazing fast parallel processing via mirai. pak::pak("tidyverse/purrr") purrr.tidyverse.org/dev/

Being able to productionize a ML model is often the goal, however there are many things to keep track of when you do. The orbital package lets you translate your fitted scikit-learn or tidymodels model into SQL that that when run produces predictions. posit.co/blog/databri... #python #rstats

Claude 4 is pretty impressive 🤖

How reliable are LLMs at extracting data from pdfs? Inspired by @simonwillison.net's PyCon talk, I added extracting FEMA's daily operation briefing to my LLM evals suite. Just one model extracted the data from the pdf correctly: Gemini 2.5 Pro Preview. Full results -> kschaul.com/llm-evals/ev...

☕ Coffee and Coding ☕ Do you have an interesting piece of code/work to showcase, an opportunity for collaboration or a code dilemma you would like help with? We would love to hear from you at Coffee and Coding – join the NHS-R Community Slack for more info (postcard.nhsrcommunity.com)! #rstats

Statistical Rethinking with brms, ggplot2, and the tidyverse Second edition by A Solomon Kurz #RStats https://bigbookofr.com/chapters/statistics.html#statistical-rethinking-with-brms-ggplot2-and-the-tidyverse-second-edition

Trying something new: A 🧵 on a topic I find many students struggle with: "why do their 📊 look more professional than my 📊?" It's *lots* of tiny decisions that aren't the defaults in many libraries, so let's break down 1 simple graph by @jburnmurdoch.bsky.social 🔗 www.ft.com/content/73a1...

We're delighted to announce Jonathan McPherson – software architect at Posit – as keynote speaker at posit::conf(2025)! If you're curious about how thoughtful design principles can improve the data science tools you use, you won't want to miss this! Join us Sep 16-18 in Atlanta. pos.it/conf

R+Docker, we use an R pkg project structure (R/, man/, tests/, inst/) plus additional top-level folders like `exec/` for docker-executable scripts, `dev/` for devel/sandbox scripts, `reports/` for one-time reports, and `local/` for gitignored large files. app.R, plumber.R etc go at the top level.

Skip splash screen in RStudio IDE 2024.12+ nanx.me/blog/post/rs... #rstats

So many of my #RStats checks got cleaner when I learned that nrow() always returns a number if you scream it loud enough: ``` nrow(NULL) #> NULL NROW(NULL) #> [1] 0 ```

People just now finding out that much our digital infrastructure runs on COBOL. The SABRE airline/hotel reservation system e.g. runs on virtual 1980s mainframes executing 1970s COBOL. See also many banks. The Old Ones dwell in the dark, undiminished. We cannot kill them without crippling ourselves.

Success in wartime intelligence through statistics! #databs

Illuminated contours of the Gulf of México. #rayshader adventures, an #rstats tale

If you have used #ggplot2 in the last couple of years you owe a great deal to @teunbrand.bsky.social who is behind most of the new features and fixes. Read about his journey to become a part of the ggplot2 core team here:

Baffled by the decisions they made adapting the Wheel of Time to a series

1/ 🌌 New Paper Alert: How can we decode the atmospheres of exoplanets efficiently and reliably? The latest work by my amazing collaborator @timothygebhard.bsky.social introduces Flow Matching Posterior Estimation for atmospheric retrieval. 🚀🧵👇 #AI #MachineLearning #Physics #Astronomy #AcademicSky

Mr. Bond, I have strategically deployed several unlicensed instances of Oracle Database 19c to the infrastructure of every Five Eyes member state. In exactly one hour I shall be notifying Oracle's legal department of these events. I suggest you sit back and relax, Mr. Bond. Enjoy the show.

This statement seems to be intentionally misleading. The question being asked is “did they train on the test set” and the answer is no. #databs

Very proud to announce that @posit.co has joined the opensourcepledge.com. We're committing to spending $2000 / developer / year to support open source software that we use (but don't develop): posit.co/blog/posit-p... #rstats #pydata

Positron in Workbench Positron in Workbench Positron in Workbench Positron in Workbench ...and other cool things. This is a HUGE update for Posit Workbench!! posit.co/blog/rstudio... #Posit #Positron #r #python

#NeurIPS2024 #ML

We are excited to announce posit::conf(2025)! Join fellow data enthusiasts & developers for an unforgettable three days of learning, connection, and inspiration. Learn more and register at posit.co/blog/positco... #datascience #rstats #pydata #quarto #tidyverse #posit

Randy Johnson staring a hole thru Greg Maddux... After getting buzzed by an 86 mph Heater.

I have a theory for why they are so useful. You can think of time as the universal cause of all outcomes. Effects may be mediated by other variables but the passage of time is the root cause of every change.

Statistical Rethinking remains one of my favorite books about stats/probability/bayesian modeling. Here the author has collected his lectures on the topic and made them freely available! #databs

"We're deploying 20TB drives--and the cost of storage has plunged dramatically in the past few decades." Okay, so why hasn't EBS's pricing done much beyond "hold steady" these last few years?

Feeling personally attacked here lol - what do you mean a sql transform of a field gets us as much utility as my cumbersome pipeline that requires bespoke new architecture and managing Python envs?? #databs

I'm writing up a brief history of the tidyverse? What do you want to know about it? #rstats

Wonder how gdpr/ccpa applies (or doesn’t). Does hf respond to “forget me” orders? #databs

My deep learning course at the University of Geneva is available on-line. 1000+ slides, ~20h of screen-casts. Full of examples in PyTorch. fleuret.org/dlc/ And my "Little Book of Deep Learning" is available as a phone-formatted pdf (nearing 700k downloads!) fleuret.org/lbdl/

Some of these recs are unsettling

$ brew uninstall pyenv $ brew install uv $ uv its incredible #databs

Neat to see this. My audiences weren’t terribly impressed with my classifications of dater habitat suitability using random forests ~8 yrs ago lol

Greatest #DataViz hits continue! The iconic (and chilling) "sea ice spiral" is familiar, but #RStats tools like #rayshader and #rayrender let you bring the data to life. By visualizing the actual sea ice, the story becomes tangible, intuitive, and easier to understand. 🌍❄️