D4 and D6 describe the eigenvalue/SVD methods. PCA retains the parent L assignment rather than guessing an immediate statistical subclass.
Package dependencies
Resolved GitHub repositories used by the verified playground build, combining direct and indirect dependencies. This list does not describe internal module dependencies or native system libraries.
SVD-based principal directions and covariance analysis; D6 and L13b remain source-supported. A checked variance probe fails: covariance diag(10,3.5) produces latent [100,12.25] and incorrect explained percentages. The SVD branch squares covariance singular values again. See docs/gams-candidates/forpca/covariance-variances-result.json and docs/gams-sparse-integral.md. The ordinary catalog demo checks direction and an independently formed rank-one reconstruction only; no checked PCA variance driver is published.