progress_plot: Visualize edge detection performance in presence of prior...

In robertness/bninfo: Queries and information theoritic operations on Bayesian networks.

Description

In network structure inference, when some edges are known to exist a priori, it is not useful to treat those edges as one would treat newly detected edges when evaluating performance. Given a gold standard network, this function visualizes detection of new edges given model averaging results. Prior edges are solid black lines, newly detected edges are solid green lines, undetected edges are dashed black lines.

Usage

progress_plot(model_averaging, ref_net, prior_edges, threshold = NULL)

progress_plot_list(model_averaging, ref_net, prior_edges, threshold = NULL)

Arguments

model_averaging	an object of class bn.strength
ref_net	the gold standard network structure
prior_edges	matrix of a priori known edges
threshold	the cut off for edge inclusion, defaults to NULL where the threshold attribute of the model_averaging argument is used.

robertness/bninfo documentation built on May 27, 2019, 10:32 a.m.