tensor_decomp: Implements tensor factorization methods, Tucker and...

In KSheu/ksheu.library1: KSheu Library1

Implements tensor factorization methods, Tucker and PARAFAC/canonical polyadic decomp

Description

Tucker: If there is no truncationin one of the modes, then this is the same as the MPCA,mpca. If there is no truncation in all the modes, then this is the same as the HOSVD,hosvd.

Usage

tensor_decomp(matrix, dims, method, xlabs = rownames(matrix),
  ylabs = colnames(matrix)[-1], zlabs = NULL, ranks = A@modes,
  savename = "TEST", plot = F)

Arguments

matrix	matrix that gets coverted to multidimensional array that get converted through as.tensor(arr)
dims	dimensions to fold the input matrix
method	"tucker" or "cp"
plot	plot heamtmaps of eigenvectors
z_labs	labels for the extra dimensions created, rows and colnames are the x and y axis labs
rank	default = 3, number (if CP) or vector of numbers (if tucker) that is <= the ranks of the tensor

Details

Input: Multidimensional array that gets converted to as.tensor Output: Decomposed tensor object

KSheu/ksheu.library1 documentation built on Jan. 28, 2025, 3:26 p.m.