Max kuhn
@topepo.bsky.social
Writing modeling packages at @posit.co (née RStudio). Opinions are my own. https://max-kuhn.org/
My #rstats talk for the R+AI conference, titled "Analyzing #LLM Evaluations," starts at 1:40 EST today.
rconsortium.github.io/RplusAI_webs...
rconsortium.github.io/RplusAI_webs...
November 12, 2025 at 5:38 PM
My #rstats talk for the R+AI conference, titled "Analyzing #LLM Evaluations," starts at 1:40 EST today.
rconsortium.github.io/RplusAI_webs...
rconsortium.github.io/RplusAI_webs...
For some models that are known to tidymodels, you can trick it into just working:
gist.github.com/topepo/0d6f7...
gist.github.com/topepo/0d6f7...
November 5, 2025 at 7:33 PM
For some models that are known to tidymodels, you can trick it into just working:
gist.github.com/topepo/0d6f7...
gist.github.com/topepo/0d6f7...
I just gave a 20m overview of the #TabPFN deep learning model at R/Pharma.
The slides are at: topepo.github.io/2025-r-pharma/
(link to Quarto files on the second slide)
The slides are at: topepo.github.io/2025-r-pharma/
(link to Quarto files on the second slide)
November 5, 2025 at 7:10 PM
I just gave a 20m overview of the #TabPFN deep learning model at R/Pharma.
The slides are at: topepo.github.io/2025-r-pharma/
(link to Quarto files on the second slide)
The slides are at: topepo.github.io/2025-r-pharma/
(link to Quarto files on the second slide)
We had a discussion in my group about this a while back. Use actual random numbers as seeds or common values/sequences, such as 1, 123, 999, 321, etc.
I think to their dismay, I always had 3-4 digit random values, leading to this slide annotation:
I think to their dismay, I always had 3-4 digit random values, leading to this slide annotation:
October 22, 2025 at 9:34 PM
We had a discussion in my group about this a while back. Use actual random numbers as seeds or common values/sequences, such as 1, 123, 999, 321, etc.
I think to their dismay, I always had 3-4 digit random values, leading to this slide annotation:
I think to their dismay, I always had 3-4 digit random values, leading to this slide annotation:
I hesitate to inflict bag envy on this whole site but here they are. Thanks to @visnut.bsky.social and @robjhyndman.com
October 7, 2025 at 1:02 PM
I hesitate to inflict bag envy on this whole site but here they are. Thanks to @visnut.bsky.social and @robjhyndman.com
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!
topepo.github.io/2025_NYR/
Video in about a month.
Great conference!
August 27, 2025 at 11:10 PM
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!
topepo.github.io/2025_NYR/
Video in about a month.
Great conference!
July 16, 2025 at 5:16 PM
I've referenced this paper a few times and never realized what a wild ride the citation shows
February 13, 2025 at 3:38 AM
I've referenced this paper a few times and never realized what a wild ride the citation shows
Abstract submission and registration are now open for the 2025 Nonclinical Biostatistics Conference (June 16-18) At Rutgers University.
I’m conducting an #rstats workshop called “Making Project Work Better with R”.
community.amstat.org/biop/events/...
I’m conducting an #rstats workshop called “Making Project Work Better with R”.
community.amstat.org/biop/events/...
January 13, 2025 at 5:22 PM
Abstract submission and registration are now open for the 2025 Nonclinical Biostatistics Conference (June 16-18) At Rutgers University.
I’m conducting an #rstats workshop called “Making Project Work Better with R”.
community.amstat.org/biop/events/...
I’m conducting an #rstats workshop called “Making Project Work Better with R”.
community.amstat.org/biop/events/...
Someone in the Cosmere just invented stepwise regression
December 7, 2024 at 2:14 PM
Someone in the Cosmere just invented stepwise regression
Tell me you are running reverse dependency checks without telling me that you are running reverse dependency checks
November 25, 2024 at 7:36 PM
Tell me you are running reverse dependency checks without telling me that you are running reverse dependency checks
Now I'm getting this on my reply
November 14, 2024 at 9:21 PM
Now I'm getting this on my reply
Is this the equivalent of a blue checkmark?
November 14, 2024 at 9:08 PM
Is this the equivalent of a blue checkmark?
A new version of the #rstats probably package is on CRAN. A minor update with a bug fix and under-the-hood changes for the upcoming tune version.
But there’s finally a hex logo (thanks to Edgar Ruiz) so we have that going for us. Which is nice.
probably.tidymodels.org
But there’s finally a hex logo (thanks to Edgar Ruiz) so we have that going for us. Which is nice.
probably.tidymodels.org
February 23, 2024 at 12:07 PM
A new version of the #rstats probably package is on CRAN. A minor update with a bug fix and under-the-hood changes for the upcoming tune version.
But there’s finally a hex logo (thanks to Edgar Ruiz) so we have that going for us. Which is nice.
probably.tidymodels.org
But there’s finally a hex logo (thanks to Edgar Ruiz) so we have that going for us. Which is nice.
probably.tidymodels.org
Fantastic talk by Joe Cheng at #positconf2023 on using #shiny with #quartopub.
I've made an #rstats Quarto book demo with a wasm-embedded shiny app to demonstrate how class boundaries change with the complexity of natural splines.
Sources: github.com/topepo/shiny...
I've made an #rstats Quarto book demo with a wasm-embedded shiny app to demonstrate how class boundaries change with the complexity of natural splines.
Sources: github.com/topepo/shiny...
September 26, 2023 at 10:21 PM
Fantastic talk by Joe Cheng at #positconf2023 on using #shiny with #quartopub.
I've made an #rstats Quarto book demo with a wasm-embedded shiny app to demonstrate how class boundaries change with the complexity of natural splines.
Sources: github.com/topepo/shiny...
I've made an #rstats Quarto book demo with a wasm-embedded shiny app to demonstrate how class boundaries change with the complexity of natural splines.
Sources: github.com/topepo/shiny...
They are definitely a hot topic.
The simple methods are very limited (e.g., equal interval widths) and the complex methods all come with handicaps (esp extrapolation).
The image below is from www.tidymodels.org/learn/models...
They are better than nothing but there is probably a long way to go.
The simple methods are very limited (e.g., equal interval widths) and the complex methods all come with handicaps (esp extrapolation).
The image below is from www.tidymodels.org/learn/models...
They are better than nothing but there is probably a long way to go.
September 5, 2023 at 9:50 PM
They are definitely a hot topic.
The simple methods are very limited (e.g., equal interval widths) and the complex methods all come with handicaps (esp extrapolation).
The image below is from www.tidymodels.org/learn/models...
They are better than nothing but there is probably a long way to go.
The simple methods are very limited (e.g., equal interval widths) and the complex methods all come with handicaps (esp extrapolation).
The image below is from www.tidymodels.org/learn/models...
They are better than nothing but there is probably a long way to go.
I'm talking about #rstats conformal inference at #PositConf2023 in Chicago (posit.co/conference/) in about 2 weeks!
September 5, 2023 at 7:52 PM
I'm talking about #rstats conformal inference at #PositConf2023 in Chicago (posit.co/conference/) in about 2 weeks!