Module discriminant 

Discriminant analysis (ISL ch 4.4) — LDA (shared covariance ⇒ linear boundary) and QDA (per-class covariance ⇒ quadratic boundary). Both are Gaussian generative classifiers; the covariance inverse / log-determinant come from linear_algebra::cholesky (no new solver). Kernel-class DenseLinear.

Structs

LdaModel

Linear Discriminant Analysis: one shared within-class covariance. The discriminant is linear, δ_c(x) = xᵀwc + bc.

QdaModel

Quadratic Discriminant Analysis: a separate covariance per class. The discriminant δ_c(x) = −½ ln|Σc| − ½ (x−μc)ᵀ Σc⁻¹ (x−μc) + ln π_c.