Fig. 4From: DAGBagM: learning directed acyclic graphs of mixed variables with an application to identify protein biomarkers for treatment response in ovarian cancerRun time comparison: the Hill Climbing (HC) algorithm implemented in DAGBagM (orange) versus HC implemented in bnlearn (blue) versus mDAG (purple) across a collection of settings with different combinations of number of nodes and number of edges. Note that mDAG was only run for up to \(1512\) nodes, as beyond that the computational time increases drasticallyBack to article page