Hi Jamie, Just to chime in with an alternative. I wrote a program that is capable to specify the (G1, G2) blocks in a bipartite network, using a flat prior detailed in a paper. When I started to do the research, graph-tool did not allow such customization. Code: https://github.com/junipertcy/bipartiteSBM Related doc: https://docs.netscied.tw/bipartiteSBM/usage/explore-consistency.html The flat prior was meant to be comparable with graph-tool's `minimize_blockmodel_dl`. If you want a hierarchical partition, you can feed the resulting partition from the `bipartiteSBM` program to the `minimize_nested_blockmodel_dl` function, by setting `b_min` and `b_max` simultaneously it. Graph-tool should build a hierarchy atop. In the meantime, I am curious to know if Tiago has added the function in graph-tool, too! Sincerely, Tzu-Chi -- Sent from: https://nabble.skewed.de/