Specifying number of blocks in a bipartite SBM
Hi Tiago, Is it possible to specify the number of blocks for both types of nodes when estimating a bipartite SBM? I know that if I specify model.minimize_nested_blockmodel_dl(B_min=G, B_max=G, state_args = {'clabel': model.vp['node_type'], 'pclabel': model.vp['node_type']}) then I will end up with a partition with G blocks. But what I want is a partition in which the type 1 nodes are divided into G1 blocks and the type 2 nodes into G2 blocks, where I specify G1 and G2. Thank you, Jamie -- Sent from: https://nabble.skewed.de/
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/
Am 10.07.20 um 02:10 schrieb jsfog:
Hi Tiago,
Is it possible to specify the number of blocks for both types of nodes when estimating a bipartite SBM? I know that if I specify
model.minimize_nested_blockmodel_dl(B_min=G, B_max=G, state_args = {'clabel': model.vp['node_type'], 'pclabel': model.vp['node_type']})
then I will end up with a partition with G blocks. But what I want is a partition in which the type 1 nodes are divided into G1 blocks and the type 2 nodes into G2 blocks, where I specify G1 and G2.
This is currently not yet implemented. -- Tiago de Paula Peixoto <tiago@skewed.de>
participants (3)
-
jsfog -
Tiago de Paula Peixoto -
Tzu-Chi Yen