get_reduced_classes: Find D-dimensional candidate latent classes from pairwise...

In hillarykoch/CLIMB: Composite LIkelihood eMpirical Bayes

Find D-dimensional candidate latent classes from pairwise latent classes.

Description

This is the second step of the CLIMB procedure. For higher dimensions, this function is fast but requires a lot of memory.

Usage

get_reduced_classes(fits, d, filepath = "lgf.txt")

Arguments

fits	Output from get_pairwise_fits
d	Dimension of data
filepath	Path to write the LEMON graph format fit, to be used internally by CLIMB.
split_into_two	Logical. Defaults to TRUE. Should the problem be split into 2 sub-problems? This will not change the outcome, but setting to TRUE can save memory and time.

Value

M by D matrix of candidate latent classes. Each row is one candidate class.

Author(s)

[e-mail]

Examples

data(sim)
data(fits)

#------------------------------------------------------------
# Not run:
#
# D <- ncol(sim$data)  
# reduced_classes <- get_reduced_classes(fits, D, "lgf.txt")
#------------------------------------------------------------

hillarykoch/CLIMB documentation built on Oct. 24, 2022, 4:27 a.m.