Configure DCM Initialization and Constants

Generates the necessary Nimble constants, inits, and monitors specifically tailored for Diagnostic Classification Models (DCMs).

Usage

configure_dcm(info, X, priors = NULL)

info: Graph structural properties from get_graph_info.
X: A numeric matrix representing the observational participant data.
priors: Optional list of prior specifications. Can be provided as common pairs (e.g., list(beta = c(mean, std), theta = c(mean, std), lambda = c(mean, std))) or individual parameter arrays (e.g., list(beta_mean = c(...), beta_std = c(...), theta_mean = matrix(...), ...)). If NULL, default priors (mean 0, std 2) are generated. Passing a standard deviation of 0.0001 or similar effectively acts as a point distribution, enabling the use of pgdcm as a scoring-only model when parameter means are supplied from a previous calibration.

A list with constants, inits, monitors, and data.