Lightnews — Scholar-powered news

Max kuhn @topepo.bsky.social · 1d

The kangaroos are nice, but those handles are 💯

1 1

Max kuhn @topepo.bsky.social · 1d

I hesitate to inflict bag envy on this whole site but here they are. Thanks to @visnut.bsky.social and @robjhyndman.com

1 3

Reposted by Max kuhn

Kelly Bodwin @kellybodwin.com · 20d

Shannon's slides are always so unbelievably clear and helpful!!!

github.com/shannonpileg...

I'm having "Ohhhhh that's what that means" moments every 10 seconds here.
#positconf2025

2 15 37

Reposted by Max kuhn

Kevin Baer @kevinbaer.bsky.social · 26d

I'm all in on @topepo.bsky.social and co's new {important} and other variable importance/feature selection tools in tidymodels! #rstats

1 1 7

Reposted by Max kuhn

Ella Kaye @ellakaye.co.uk · Sep 3

Once again, @davisvaughan.bsky.social's extrachecks have saved me from a likely CRAN rejection for an upcoming #RStats package submission. Thanks Davis! github.com/DavisVaughan...

GitHub - DavisVaughan/extrachecks

Contribute to DavisVaughan/extrachecks development by creating an account on GitHub.

github.com

3 9 40

Reposted by Max kuhn

Adam L @lauretig.bsky.social · 28d

Simon Wood, the GOAT of generalized additive models & creator of the mgcv #rstats package, has an Annual Review of Statistics essay on GAMs, available open access #statssky #mlsky

www.annualreviews.org/content/jour...

41 89

Reposted by Max kuhn

Daniel Gutierrez @ddgutierrez.bsky.social · 28d

ML success ≠ Kaggle leaderboard. The real world rewards:
- Clear explanations
- Thoughtful metrics
- Collaboration with domain experts

A 0.01 lift in F1 score won’t save you if no one understands your model.

#DataSciene #MachineLearning #AI #RStats

1 4

Reposted by Max kuhn

Posit @posit.co · Sep 5

Announcing a new blog series on LLMs from @veerle.hypebright.nl!

In Part 2, “Talking to LLMs: From Prompt to Response”, we get hands-on with LLM-powered apps. This guide is for #Python & #RStats users who want to go beyond the basics.

Check it out here: shiny.posit.co/blog/posts/s...

3 6 20

Max kuhn @topepo.bsky.social · Aug 27

The new version of tune (out before posit::conf🤞) has support for both.

My testing so far shows that future(_lappy) and mirai(_map) are very similar in speed-ups most of the time. The times when they differ split in either direction, so I haven’t seen any clear signal.

1 4

Max kuhn @topepo.bsky.social · Aug 27

Slides from my #rstats talk “Measuring LLM Effectiveness” at #dataconfAI with @simonpcouch.com.

topepo.github.io/2025_NYR/

Video in about a month.

Great conference!

1 7 29

Reposted by Max kuhn

The Data Science & AI Conference Presented by Lander Analytics @dataconf.ai · Aug 27

🧠📊 3 days. 2 workshops. 20 talks. 1 amazing community.
#dataconfAI is officially wrapped!

Thanks for showing up with insights, ideas, inspiration, and curiosity. And to all who made it unforgettable—speakers, attendees, sponsors, and volunteers.

See you at the next one! 🚀

3 1

Reposted by Max kuhn

Simon P. Couch @simonpcouch.com · Aug 26

In working on an eval for an experimental tidymodels AI assistant, I realized that today's frontier LLMs know much more about #rstats tidymodels than I thought.

www.simonpcouch.com/blog/2025-08...

A ggplot2 bar plot. On the x axis are two LLMs, Claude Sonnet 4 and Gemini Pro 2.5. Both bars are dodged by three configurations: Predictive, Databot, and `run_r_code()` tool. On the y axis is relative performance, ranging from 0 (baseline model) to 100 (best human modelers). The bar heights for Claude Sonnet 4 are all about the same (around 80) and, for Gemini 2.5 Pro, the height for Predictive, an app designed specifically as a tidymodels assistant, is much lower (around 20).

1 6 22

Max kuhn @topepo.bsky.social · Aug 21

🤫

1 1

Max kuhn @topepo.bsky.social · Aug 20

It's a lot of fun! Everyone gets something out of it.

Plus, @davisvaughan.bsky.social always finds a great barista!

Hadley Wickham @hadley.nz · Aug 20

We still have spots available for tidyverse dev day on Sept 19: www.tidyverse.org/blog/2025/07.... Please come along to contribute to the tidyverse and have a bunch of fun along the way! It's open to all, but is most convenient if you're coming to posit::conf or live near Atlanta #rstats

Tidyverse developer day 2025

Join us in Atlanta for tidyverse developer day on September 19, 2025!

www.tidyverse.org

1 1 7

Reposted by Max kuhn

The Data Science & AI Conference Presented by Lander Analytics @dataconf.ai · Aug 15

Start off The NY Data Science & AI Conference w/ hands-on workshops on Aug 25 in NYC or online:

📊 Machine Learning in R w/ Max Kuhn
🤖 Intro to LLMs/AI w/ Daniel Chen

🎟️ Learn more & register at dataconf.ai/nyc

#RStats #AI #Workshops #databs @topepo.bsky.social @chendaniely.bsky.social

3 7

Reposted by Max kuhn

Emil Hvitfeldt @emilhvitfeldt.bsky.social · Aug 13

Excited to share my newest quarto revealjs plugin: imagemover

Easily reposition and resize images directly in your quarto revealjs slides for a much smoother slidecrafting experience

github.com/EmilHvitfeld...
#quarto

8 51 210

Reposted by Max kuhn

Dr. U @transport-talk.bsky.social · Aug 10

Time to convert this into an LLM powered snippet using {chores} by @simonpcouch.com. #useR2025 #rstats

Dr. U @transport-talk.bsky.social · Jun 8

CreateBranding shiny app is now available here: umair.shinyapps.io/create_brand...

You can now download the palettes, scales and a theme for ggplot2. See a demo here: youtu.be/C7-rhLPrA3o

Try it out and let me know your suggestions for improvements. #rstats

2 5

Reposted by Max kuhn

Kelly Bodwin @kellybodwin.com · Aug 9

Welp, {chores} by @simonpcouch.com is an immediate install for sure.

Basically it's {usethis} plus llm bundled into RStudio/Positron key encoding.

Excited!!! 🧹🧺

#useR2025 #rstats #couchverse?

8 29

Reposted by Max kuhn

Lander Analytics @landeranalytics.com · Aug 5

Don't miss out learning from the best, Max Kuhn! @topepo.bsky.social

#dataBS #Tidymodels #MachineLearning

The Data Science & AI Conference Presented by Lander Analytics @dataconf.ai · Jul 31

📊 Want to level up your R modeling skills? Max Kuhn’s Machine Learning in R workshop is an intro to tidymodels, covering data prep, resampling, tuning & evaluation using real workflows!

📍Aug 25 in NYC or online
🎟️ & info: dataconf.ai/nyc

#RStats #Tidymodels #MachineLearning @topepo.bsky.social

1 2

Reposted by Max kuhn

The Data Science & AI Conference Presented by Lander Analytics @dataconf.ai · Jul 31

📊 Want to level up your R modeling skills? Max Kuhn’s Machine Learning in R workshop is an intro to tidymodels, covering data prep, resampling, tuning & evaluation using real workflows!

📍Aug 25 in NYC or online
🎟️ & info: dataconf.ai/nyc

#RStats #Tidymodels #MachineLearning @topepo.bsky.social

4 8

Max kuhn @topepo.bsky.social · Jul 29

We are super excited to have you join us for the day!

Posit @posit.co · Jul 29

Check out our Modeling & ML with #RStats workshops at posit::conf!

🔢 Intro to ML w/ tidymodels @simonpcouch.com
🏗️ Feature Engineering & Tuning @topepo.bsky.social @emilhvitfeldt.bsky.social
↔️ Causal Inference @malcolmbarrett.malco.io @lucystats.bsky.social

Learn more: pos.it/conf-2025-workshops

4

Max kuhn @topepo.bsky.social · Jul 29

I’m sorry it wasn’t a good experience for you.

I’m surprised by the first two issues; those have both been in positron for a while and work pretty well (for me).

The visual cues are not as isolated as they are in Rstudio. We are aware of that and are working on it. It’s a vscode constraint.

1

Max kuhn @topepo.bsky.social · Jul 28

Positron is definitely visually more than RStudio, and this is a helpful overview.

Posit @posit.co · Jul 28

Take a quick tour of Positron, Posit's next-generation data science IDE, built by the creators of RStudio.

Read the blog to learn more: posit.co/blog/a-quick...

A screenshot of the Positron interface, labeled with its components: "Activity bar," "Primary side bar," "Editor," "Secondary side bar," and "Panel." The text "A quick tour of Positron" is on the left.

1 1 8

Reposted by Max kuhn

Hannah Frick @frick.ws · Jul 25

The call for papers for LatinR 2025 (online) is now open! You can present in English, Spanish, or Portuguese 🗣️ #RStats latinr.org/en/blog/en/2...

Call for papers – LatinR 2024

latinr.org

5 4

Max kuhn @topepo.bsky.social · Jul 25

We've released 4 new chapters of Applied Machine Learning for Tabular Data.

Includes: Bayesian optimization, feature selection, model comparisons, classification metrics, calibration, #rstats computing sections, and more

blog.aml4td.org/posts/2025-0...

Part 3 is Finished, Part 4 Started – Applied Predictive Modeling Blog

blog.aml4td.org

2 9 50