kphahn.bsky.social
@kphahn.bsky.social
Reposted
ICYMI: stringr 1.6.0 🧵

This update brings performance gains and new fns for string manipulation in #RStats.

Highlights include: Faster replacements with `str_replace_all()`, 🐫 new case tools to camel, snake, and kebab, case-sensitive str_like(), & more.

Read more: tidyverse.org/blog/2025/11...
January 20, 2026 at 4:10 PM
Reposted
An interesting bit of history:
In 1916, Denmark sold the Danish West Indies to the US. You may know of them as the US Virgin Islands. As part of this sale, the US recognized Danish authority over Greenland.

The text of the complete, ratified agreement is here: www.govinfo.gov/content/pkg/...
January 19, 2026 at 7:06 AM
Reposted
Our guidance regarding performance measures for medical AI models is finally out!

- Stop bashing AUROC, although it does not settle things
- Calibration and clinical utility are key
- Show risk distributions
- Classification statistics (e.g. F1) are improper

www.thelancet.com/journals/lan...
Evaluation of performance measures in predictive artificial intelligence models to support medical decisions: overview and guidance
Numerous measures have been proposed to illustrate the performance of predictive artificial intelligence (AI) models. Selecting appropriate performance measures is essential for predictive AI models i...
www.thelancet.com
December 13, 2025 at 2:04 PM
so much important information: #rstats #quarto #positron etc.
posit.co Posit @posit.co · Nov 14
Have some free time this weekend? How about 2,500 hours?

The recordings from posit::conf(2025) are now online!

Learn about the innovations and insights from conf. We celebrated #RStats and #Python, with a focus on next-gen tools & #AI integration.

See the info here: posit.co/blog/talks-a...
Talk recordings and workshop materials from posit::conf(2025) - Posit
2500+ hours of posit::conf content just landed on YouTube.
posit.co
November 15, 2025 at 7:09 PM
Reposted
Three phrases that capture some of the tidyverse team culture:

Everybody takes out the trash

We suffer so users don’t have to

If it hurts, do it more often
October 3, 2025 at 1:46 AM
Great post by Claire Welsh about some special features of Positron. Thank @juliasilge.com for sharing.
I found this post by Claire Welsh super interesting, outlining how she uses #Positron as someone working in product! 🤩 My own data science skills have been helpful in making product decisions, and I always appreciate learning more about other people who do similar work.

posit.co/blog/positro...
Posit
Dr. Claire Welsh discusses how Positron, Posit's new IDE, can assist product owners in tasks ranging from managing a product backlog to testing new features.
posit.co
September 9, 2025 at 6:04 PM
I have summarized my personal impressions of the migration from RStudio to Positron. After some issues I will now switch to Positron. #rstats #positron #rstudio
kphahn57.medium.com/from-rstudio...
From RStudio to Positron
What I have learned
kphahn57.medium.com
August 25, 2025 at 11:00 AM
Reposted
For the early adopters of acquaint, a package implementing the Model Context Protocol in #rstats: the package has been renamed to mcptools and no longer exports btw's tools by default.

See the package website for more info posit-dev.github.io/mcptools/
Model Context Protocol Server For Your R Sessions
Implements a model context protocol (MCP) server for your R sessions, allowing MCP-compatible apps like Claude Desktop and Claude Code to run R code in your active R sessions in order to better answer...
posit-dev.github.io
July 2, 2025 at 4:52 PM
Some experiments with R, Ollama and medgemma:
Summarize medical transcripts in german.
medium.com/@kphahn57/ol...
Ollama, R and medgemma
Transcript to SOAP in German
medium.com
June 7, 2025 at 9:20 AM
Reposted
Re-proposing the Occam's taser: an automatic electric shock for anyone riding the AI hype train making their models unnecessarily complex
May 27, 2025 at 2:38 PM
Reposted
Introducing the btw package for teaching LLM chat apps about your #RStats package!

Inject "invisible" messages into chats via system prompts and use tool calls to dynamically fetch context when needed.

Check out a dplyr example and learn more in @simonpcouch.com's post! posit.co/blog/custom-...
May 27, 2025 at 4:04 PM
Reposted
On June 13, we will have a workshop on Using LLMs by @hadley.nz
More info: bit.ly/4jmEAJW
Please share!
#RStats #EconSky #AcademicSky
April 19, 2025 at 8:45 AM
Reposted
PROBAST+AI is out! A risk of bias tool for prediction models developed using any kind of analytical approach (stats/ml/ai/ds)

Really proud of this collaboration with many prediction model experts around the world

www.bmj.com/content/388/...
March 24, 2025 at 11:16 AM
Reposted
Dishonest charts are coming from all directions, and they're only going to hit harder. Unless we push back. So, an interactive guide: Defense Against Dishonest Charts flowingdata.com/projects/dis...
Defense Against Dishonest Charts
This is a guide to protect ourselves and to preserve what is good about turning data into visual things.
flowingdata.com
February 13, 2025 at 9:45 PM
Reposted
NEW PAPER

On quantifying uncertainty in your predictions. Excellently led by the group of @richarddriley.bsky.social

www.bmj.com/content/388/...
February 13, 2025 at 11:24 AM