samplesToSVD: Finding SVD representation from solutions of matrices P and E

In ragnhildlaursen/SFS: Sampling the set of feasible solutions from a non-negative matrix factorization

Description

Finding SVD representation from solutions of matrices P and E

Usage

samplesToSVD(
  Presults,
  Eresults,
  N,
  Mfit = t(Presults[1:N, ]) %*% Eresults[1:N, ]
)

Arguments

Presults	Matrix of results of P transposed stacked on top of each other. Dimension is (N*results x nrow(P)).
Eresults	Matrix of results of E stacked on top of each other. Dimension is (N*results x ncol(E))
N	The rank of the factorization
Mfit	The initial factorization of P and E to use as a reference for the eigenvectors. Default is the factorization of the first matrix in Presults and Eresults.

Value

The SVD representation of the set of feasible solutions

• P.points - Matrix of P results as SVD solution (results x (N-1)).

• E.points - Matrix of E results as SVD solution (results x (N-1)).

ragnhildlaursen/SFS documentation built on Nov. 15, 2021, 8:28 p.m.