optimizeFactorization: Function to minimize decomposition error

In amalthomas111/SBF: R package to compute Shared Basis Factorization (SBF)

Function to minimize decomposition error

Description

Function iteratively updates u, delta, and V and decreases the decomposition error.

Usage

optimizeFactorization(
  mat_list,
  u,
  d,
  v,
  optimizeV = TRUE,
  initial_exact = FALSE,
  max_iter = 10000,
  tol = 1e-10,
  verbose = FALSE
)

Arguments

mat_list	A list containing numeric matrices
u	A list containing U matrices
d	A list containing delta matrices
v	V matrix
optimizeV	Whether initial V should be update or not. Default TRUE
initial_exact	Whether the initial value of U, Delta, and V gives exact factorization. Default FALSE
max_iter	Maximum number of iterations. In each iteration u, d, and v are updated. Default 1e4
tol	Tolerance value. During the iterations, if the difference between previous best and current best factorization error becomes less than tol, no more iteration is performed. Default tol = 1e-10
verbose	print error at each update. Default FALSE

Value

a list containing optimal U, delta, and V that minimizes the factorization error

Examples

set.seed(1231)
mymat <- createRandomMatrices(n = 3, ncols = 3, nrows = 3)
sbf <- SBF(matrix_list = mymat, orthogonal = FALSE,
           transform_matrix = FALSE)
newU <- updateU(mymat, sbf$delta, sbf$v)
newDelta <- updateDelta(mymat, newU, sbf$v)
newV <- updateV(mymat, newU, newDelta)
opt <- optimizeFactorization(mymat, newU, newDelta, newV, max_iter = 1e4)

amalthomas111/SBF documentation built on Sept. 2, 2022, 11:27 a.m.