Am 16.04.21 um 13:10 schrieb Filippo Valle:
Hello,
I am running /minimize_nested_blockmodel_dl/ on a certain network and
obtaining some partition, good.
Then I add some nodes to the network and I do another run of
/minimize_nested_blockmodel_dl/and obtain another partition.
The question I want to answer is: *does adding those nodes help me
finding a better partition?*
I am going to compare NestedBlockState.entropy() of the two run, but I
am not sure this is correct.
How should I take into account the fact that the networks are slightly
different?
This comparison is not meaningful. From the point of view of model
selection, the description length can only be used to compare identical
networks.
Before you conduct your analysis, you should have a clear criterion of
what you consider a "better" partition.
Best,
Tiago
--
Tiago de Paula Peixoto <ti...@skewed.de>
_______________________________________________
graph-tool mailing list
graph-tool@skewed.de
https://lists.skewed.de/mailman/listinfo/graph-tool