he/him - writing statistical software @posit_pbc (née RStudio)🥑 on the other sites @ simonpcouch
Chicago, IL, U.S.A.
Joined September 2019
- Tweets769
- Following328
- Followers2.1K
- Likes4.7K
Pinned Tweet
Some news: I've just open-sourced the draft of a book I'm working on about how #rstats tidymodels users can make their code run faster without sacrificing predictive performance!
simonpcouch.com/blog/2024-10…
Simon Couch retweeted
Released a cauldron of R code today 🎃👻
Introducing Kolmogorov-Arnold Networks for Time Series in R!
It's an R package called {kantime}, a super experimental port of Nixtla's {neuralforecast} from Python to {modeltime} in R
ALT frankiethull/kantime Bindings for Nixtla's neuralforecast Library in R, specifically KANs, to use with {modeltime} package. Nixtla's KAN is called from R via {reticulate}, which is then ported into {parsnip} and bridged to {modeltime}.
Simon Couch retweeted
Monitoring Models in Production is essential to ensure accuracy, detect drift, and maintain fairness over time. For more, check out our recent blog by Myles Mitchell on "Vetiver: Monitoring Models in Production".
#DataScience #MLOps #Rstats
jumpingrivers.com/blog/vetiv…
This is tomorrow! #rstats folks, come through!
I'll be giving a free virtual talk on speeding up your
#rstats tidymodels code at R/Pharma 2024 this Tuesday and would love to see you there! I'll also be announcing a new project.👀
Register here: rinpharma.com/
ALT A title slide reading "From hours to minutes: Accelerating your tidymodels code." Beside the text is a hex wall containing stickers for core packages like parsnip, recipes, and tune.
I'll be giving a free virtual talk on speeding up your
#rstats tidymodels code at R/Pharma 2024 this Tuesday and would love to see you there! I'll also be announcing a new project.👀
Register here: rinpharma.com/
ALT A title slide reading "From hours to minutes: Accelerating your tidymodels code." Beside the text is a hex wall containing stickers for core packages like parsnip, recipes, and tune.
The #rstats tidymodels team is hard at work implementing postprocessing! There will be changes in almost all of our core packages as well as an entirely new package included in this set of releases. We value your opinion—let us know what you think: tidyverse.org/blog/2024/10/p…
ALT Two ggplot2 dotplots, side-by-side. Both show the true value of an outcome versus the predicted value from a machine learning model. On the left, the "plain" model tends to predict is lower than it actually is, and increasingly so as the true value grows larger. The behavior of the "postprocessed" model is corrected.
Simon Couch retweeted
📢 Tomorrow is the official release of @bbiinnyyuu and my book, Veridical Data Science, with @mitpress. I'm so excited to for physical copies to be out in the world!
Get your copy from mitpress.mit.edu/97802620491…
Simon Couch retweeted
Updated the documentation for {maize} 🌽
String kernels are the newest addition which were a bit tricky to plant 👨🏻🌾
And I'm liking how this table turned out, thanks to @gt_package & gtUtils 🙌🏻
Simon Couch retweeted
Announcing Shiny Assistant, an AI-powered tool that can help you build Shiny applications!
shiny.posit.co/blog/posts/sh…
Postprocessing is coming to tidymodels! We've got a set of changes for updating model predictions coming to the next version of our #rstats packages.
Let us know what you think: tidyverse.org/blog/2024/10/p…
Simon Couch retweeted
{maize} works with {stacks} too! 🌽🥞
{stacks} is part of the tidymodels ecosystem for combining many models into a new ensemble model
Here's an ensemble example of three SVMs with different types of kernels 🌽🌽🌽
The dopest parts about {maize}? It jives well with the tidymodels ecosystem!
Here, I'm answering a typical question keeping everyone up at night, "what does conformal inference quantile regression prediction intervals look like for a laplacian kernel support vector machine" 🤓
ALT Code snippet of setting up a laplace kernel svm spec, svm wkflow fitting, conformal quantile fit, and create conformal predictions
Simon Couch retweeted
Today marks the 10 year anniversary of my first lintr commit (github.com/r-lib/lintr/commi…)
Hard to believe it was 10 years ago
Apparently originally I called it styleR, long before the current styler package came into being 😄
A new release of #rstats broom made it to CRAN yesterday! v1.0.7 includes changes to tidiers for objects from survival, boot, car, and (base) stats.
Read more: broom.tidymodels.org/news/in…
The tidymodels team recently released a set of features for assessing model fairness. I got to drop by the Cascadia #rstats Conf this summer to share about our process in putting that toolkit together, and a recording of the talk is now live!
youtube.com/watch?v=1wKt-574…
Simon Couch retweeted
cpp11 0.5.0 is on CRAN 🥳
The main thing to note is that we've removed all non-API R calls, so if you use cpp11 you should no longer get any NOTEs when submitting to CRAN! #rstats
cpp11.r-lib.org/news/index.h…
Simon Couch retweeted
Let me take this opportunity to shill the infer #RStats package
infer.tidymodels.org/article…
read the vignette which explains why charts like the one below can be ignored nowadays (we have very good computers these days)
Simon Couch retweeted
This R repo wouldn't be possible without {modeltime} and {tidymodels} 🙌🏻
It shows how easy conformal prediction is in R
A nice read if you already have @predict_addict's book and Tidy Modeling with R by @topepos & @juliasilge
R fans of conformal prediction, there are some repos translating code for my book into R (courtesy of @frankiethull)
I am also going to post a few libraries in R in comments.
#confoRmalprediction
Simon Couch retweeted
arXiv -> alphaXiv
Students at Stanford have built alphaXiv, an open discussion forum for arXiv papers. @askalphaxiv
You can post questions and comments directly on top of any arXiv paper by changing arXiv to alphaXiv in any URL!