simulate_count_data: Simulate Count Data from Poisson NMF Model

In fastTopics: Fast Algorithms for Fitting Topic Models and Non-Negative Matrix Factorizations to Count Data

Simulate Count Data from Poisson NMF Model

Description

Simulate a counts matrix X such that X[i,j] is Poisson with rate (mean) Y[i,j], where Y = tcrossprod(L,F), L is an n x k loadings (“activations”) matrix, and F is an m x k factors (“basis vectors”) matrix. The entries of matrix L are drawn uniformly at random between zero and lmax, and the entries of matrix F are drawn uniformly at random between 0 and fmax.

Usage

simulate_count_data(n, m, k, fmax = 1, lmax = 1, sparse = FALSE)

Arguments

n	Number of rows in simulated count matrix. The number of rows should be at least 2.
m	Number of columns in simulated count matrix. The number of columns should be at least 2.
k	Number of factors, or “topics”, used to determine Poisson rates. The number of topics should be 1 or more.
fmax	Factors are drawn uniformly at random between zero and fmax.
lmax	Loadings are drawn uniformly at random between zero and lmax.
sparse	If sparse = TRUE, convert the counts matrix to a sparse matrix in compressed, column-oriented format; see sparseMatrix.

Details

Note that only minimal argument checking is performed. This function is mainly used to simulate small data sets for the examples and package tests.

Value

The return value is a list containing the counts matrix X and the factorization, F and L, used to generate the counts.

fastTopics documentation built on Sept. 11, 2024, 5:14 p.m.