nipals_pca: Principal component dimensionality reduction using NIPALS

In jonalim/mfBiclust: Pipelines for Bi-Clustering Using Matrix Factorization

Description

Factorizes matrix A as the product of score and loading matrices respectively truncated to k rows and k columns. Uses the Nonlinear Iterative Partial Least Squares algorithm to compute principal components in the presence of missing matrix elements.

Usage

nipals_pca(A, k, cleanParam = 0, verbose = TRUE, ...)

Arguments

A	the matrix to factorize
k	the number of factors to compute
cleanParam	passed to clean()
verbose	report recursive calls and all values of cleanParam
...	Additional parameters will be passed to nipals.

Details

If NIPALS fails, this function will recursively call itself with decreasing values of cleanParam until NIPALS succeeds.

Value

a list containing

m

the data matrix after any cleaning

genericFit

a genericFit-class object

indexRemaining

a list of the row and column indexes remaining after cleaning

jonalim/mfBiclust documentation built on May 4, 2019, 4:13 a.m.