Skip to contents

Build Model Configuration

build_model_config.Rd

The primary entry point for bridging the structural igraph topology and the observational data frame into a compiled Nimble configuration list. Automatically detects the exact model geometry to invoke the appropriate compiler.

Usage

build_model_config(graph, dataframe, priors = NULL)

Arguments

graph

An igraph object representing the conceptual architecture.

dataframe

A data.frame of raw responses matching graph Tasks.

priors

Optional list of prior specifications. See configure_dcm or configure_sem for details.

Value

A configuration list encompassing Nimble requirements. The list contains:

  • constants: A list of graph constants (e.g., node counts, adjacency matrices, priors).

  • inits: A list containing initial values for MCMC sampling (e.g., beta_root, theta, lambda).

  • monitors: A character vector of node names to monitor during MCMC.

  • data: A list containing the aligned observational response matrix X.

  • code_file: The resolved absolute path to the underlying Nimble model generator script.

  • model_object: A character string denoting the name of the function to invoke (e.g., "DiBelloBN").

  • graph: The fully verified and topologically sorted igraph object.

  • type: A character string denoting the detected architecture (e.g., "DCM", "SEM").