R/propose_stay.R

In cia: Learn and Apply Directed Acyclic Graphs for Causal Inference

#' StayStill proposal.
#' 
#' @param partitioned_nodes Labelled partition.
#' 
#' @noRd
StayStill <- function(partitioned_nodes) {
  
  current_nbd <- CalculateStayStillNeighbourhood(partitioned_nodes)
  
  proposed <- ProposeStayStill(partitioned_nodes)
  partitioned_nodes <- proposed$partitioned_nodes
  rescore_nodes <- proposed$rescore_nodes
  
  new_nbd <- CalculateStayStillNeighbourhood(partitioned_nodes)
  
  return(list(
    state = partitioned_nodes, 
    current_nbd = current_nbd, 
    new_nbd = new_nbd,
    rescore_nodes = rescore_nodes))
}

#' Propose that the partition stays still.
#' 
#' dag <- UniformlySampleDAG(c('A', 'B', 'C', 'D', 'E', 'F'))
#' partitioned_nodes <- GetPartitionedNodesFromAdjacencyMatrix(dag)
#' ProposeStayStill(partitioned_nodes)
#' 
#' @param partitioned_nodes A labelled partition.
#' 
#' @return A proposed labelled partition.
#' 
#' @noRd
ProposeStayStill <- function(partitioned_nodes) {
  return(list(partitioned_nodes = partitioned_nodes, 
              rescore_nodes = NULL))
}

#' Calculate neighbourhood for staying still.
#' 
#' @param partitioned_nodes A labelled partition.
#' 
#' @noRd
CalculateStayStillNeighbourhood <- function(partitioned_nodes) {
  return(1)
}