My last R posts: How conformalization helps weak models, fast conformal prediction with jackknife+ (and no refitting), and sklearn in R

This post is mainly (but not only) a test, because I had a broken xml feed for my R posts, and I wanted to see if it was fixed.

It’s about my last R posts from june and july, which are:

• Using scikit-learn models in R easily with the tisthemachinelearner R package
• How conformalization helps weak models
• Fast Conformal Prediction for Some Machine Learning Models (jackknife+ and no refitting)

Using scikit-learn models in R easily with the tisthemachinelearner R package

This post is about the tisthemachinelearner R package, that allows to use scikit-learn models in R. It is a wrapper around the tisthemachinelearner Python package. Prediction intervals can be computed using either split conformal prediction, surrogate methods or the bootstrap.

Read: https://thierrymoudiki.github.io/blog/2026/06/21/r/tisthemllearner

In this post, we compare split conformal prediction across several predictive models, using R package mlS3.

Read: https://thierrymoudiki.github.io/blog/2026/06/07/r/conformalization-helps-weak-models

Fast Conformal Prediction for Some Machine Learning Models (jackknife+ and no refitting)

It’s surprisingly fast to obtain conformal jackknife+ prediction intervals for Machine Learning models of the form \(\hat{y} = Sy\) (including Ordinary Least Squares, Ridge Regression, Random Vector Functional Link Networks, Kernel Ridge Regression, smoothing splines, and local polynomial regression). No refitting involved, just Linear Algebra. Read https://www.researchgate.net/publication/408161842_Fast_Conformal_Prediction_for_Some_Machine_Learning_Models_via_Closed-Form_Jackknife.