Fit SLFM to the matrices inside a directory

Share:

Description

This function is used to fit a Bayesian sparse latent factor model to a directory of numeric matrices.

Usage

1
2
3
slfm_list(path = ".", recursive = TRUE, a = 2.1, b = 1.1, gamma_a = 1,
  gamma_b = 1, omega_0 = 0.01, omega_1 = 10, sample = 1000,
  burnin = round(0.25 * sample), lag = 1, degenerate = FALSE)

Arguments

path

path to read the matrices from

recursive

if the function should look recursively inside folders

a

prior shape parameter for Gamma distribution

b

prior scale parameter for Gamma distribution

gamma_a

prior parameter for Beta distribution

gamma_b

prior parameter for Beta distribution

omega_0

prior variance of the spike component

omega_1

prior variance of the slab component

sample

sample size after burn-in

burnin

burn-in size

lag

lag for MCMC

degenerate

use the degenerate version of mixture