Error-based vs. distribution-based | The running example: a drifting sensor | Detecting the shift with KSWIN | Other distribution-based methods

What is concept drift? | The running example: a credit-approval model | Quick start | The deriva interface | 1. Specify a detector | 2. Fit on a baseline | 3. Advance over new batches | 4. Inspect results | Bridging from tidymodels | Distribution-based detectors | Available methods | Summary

Introduction to FIGS | A regression fit | Variable importance | Visualizing the tree sum | Two-class classification | Missing values | Use inside tidymodels | Bootstrap ensembling | Limitations

The Pima Indians Diabetes Database | Fitting with and without na_method = "mia" | Coverage, not just accuracy | Reading the tree sum | Visualizing the tree sum | When to impute instead | References

The three datasets | Downloading data | Understanding the variables | Managing the cache | Data sources