Intro to Workboots: Make Prediction Intervals from Tidymodel Workflows
summary
workboots generates prediction intervals for tidymodel workflows by bootstrap resampling, fitting many models (demo uses ~2,000 resamples) to any model type such as XGBoost and returning a nested tibble of predictions that summarise to median and lower/upper intervals via predict_boots. It also estimates variable importance distributions with vi_boots. Requirements: residuals roughly normal and nonreal‑time use (run time ~5–20 minutes). Demos use Palmer penguins and Ames housing and the method produces usable intervals in the examples.
AI-written summary · about
chapters (4)
related talks
16mIntro to Workboots: Make Prediction Intervals from Tidymodel Workflows
22mConformal Inference with Tidymodels
57mGood Practices for Applied Machine Learning
27mI have no clue what my model is doing...
19mUnderstanding, Generating, and Evaluating Prediction Intervals
21mTidypredict with recipes, turn workflow to SQL, spark, duckdb and beyond
28mIID Got You Down? Resample Time Series Like A Pro
37mCreating correct and capable classifiers