splsDA: splsDA sparse partial least squares discriminant analysis

In zdk123/compPLS: Conduct Partial Least Squares regression for compositional data

splsDA sparse partial least squares discriminant analysis

Description

sparse Partial Least Squares Discriminant Analysis sPLS regression to discriminate classes (via a logistic model) basically this is a wrapper for the splsda function in the caret package, but with default setup for dealing with uneven classes (via the priors option, see details) see caret::splsda for implementation details

Usage

splsDA(x, grouping, eta, K, usePriors = FALSE, ...)

Arguments

x	x with samples in rows, features are columns (not necessarily compositional x)
grouping	a numeric vector or factor with sample classes (length should equal nrow(x))
eta	parameter that adjusts sparsity of the PLS model (between 0 and 1)
K	number of components in the PLS model (default: number of classes - 1)
usePriors	use priors for very biased sample size between groups (ie - put strong penalty on misclassifying small groups)

Details

run this code if you don't need to fit paramaters by cross-validation

Value

a plsda fitted model

See Also

plsDA_main, caret::plsda, caret::splsda

zdk123/compPLS documentation built on April 24, 2022, 2:44 p.m.