genpls: Generalized Partial Least Squares with Deflation +...

In bbuchsbaum/genpca: Generalized Principal Component Analysis

Generalized Partial Least Squares with Deflation + Gram-Schmidt, Adaptive Rank by Default

Description

Forms embedded data using the supplied constraints and performs a multi-component NIPALS algorithm with deflation and Gram–Schmidt orthogonalisation. Constraint ranks are chosen adaptively by default.

Usage

 genpls(
  X, Y,
  Mx = NULL, Ax = NULL,
  My = NULL, Ay = NULL,
  ncomp = 2,
  preproc_x = multivarious::pass(),
  preproc_y = multivarious::pass(),
  rank_Mx = NULL, rank_Ax = NULL,
  rank_My = NULL, rank_Ay = NULL,
  var_threshold = 0.99,
  max_k = 200,
  maxiter = 200, tol = 1e-9,
  verbose = FALSE,
  ...
 )

Arguments

X	numeric matrix of predictors
Y	numeric matrix of responses
Mx, Ax, My, Ay	optional row/column constraint matrices
ncomp	number of latent factors to extract
preproc_x, preproc_y	pre-processing objects from multivarious
rank_Mx, rank_Ax, rank_My, rank_Ay	ranks for partial eigen expansion
var_threshold	fraction of variance to capture when ranks are not set
max_k	cap on eigen expansion size
maxiter, tol	iteration control parameters
verbose	print progress messages
...	additional arguments stored in the result

Value

A cross_projector object of class c("genpls","cross_projector","projector").

bbuchsbaum/genpca documentation built on July 16, 2025, 11:03 p.m.