discrim

The discrim package provides bindings that enable the parsnip package (part of tidymodels) to fit discriminant analysis and naive Bayes classification models. It supports linear discriminant analysis (LDA), quadratic discriminant analysis (QDA), regularized discriminant analysis (RDA), flexible discriminant analysis (FDA), and naive Bayes models.

The package integrates multiple computational engines (MASS, mda, klaR, sparsediscrim, earth, naivebayes) into a consistent parsnip interface, allowing you to fit different variants of discriminant analysis models using tidymodels workflows. It includes support for regularized versions of LDA and QDA for high-dimensional data, and flexible discriminant analysis that uses MARS features to find nonlinear decision boundaries. All models work seamlessly with tidymodels tools for cross-validation, hyperparameter tuning, and model evaluation.