Multivariate matching in observational studies typically has two goals: 1. to construct treated and control groups that have similar distribution of observed covariates and 2. to produce matched pairs or sets that are homogeneous in a few priority variables. This packages implements a network-based method built around a tripartite graph that can simultaneously achieve both goals. A detailed 'RMarkdown' tutorial can be found at < https://github.com/bzhangupenn/match2C/tree/master/tutorial>.