Am 11.11.21 um 03:03 schrieb Eli Draizen:
> Hi everyone,
>
> I was wondering if it would be possible to provide some more examples of
> how to run a nested mixed membership SBM with edge weights. The new
> version seems to have removed the "overlap=True" option for state_args
> in the minimize_* functions.

Indeed, I will add more examples about this. Could you please open an
issue in the website so I don't forget?

> Is this the correct way to do it now?
>
>     import graph_tool as gta
>     import numpy as np
>     g = .... # build graph
>     e_score = .... #Set edge weights
>     state_args = dict(
>          deg_corr=deg_corr,
>          base_type=gta.inference.overlap_blockmodel.OverlapBlockState,
>          B=2*g.num_edges(), #B_max
>          deg_corr=True,
>          recs=[e_score],
>          rec_types=["real-normal"])
>     state = gta.inference.minimize_nested_blockmodel_dl(
>          g,
>          state_args=state_args,
>          multilevel_mcmc_args=dict(verbose=True))
>     # improve solution with merge-split
>     state = state.copy(bs=state.get_bs() + [np.zeros(1)] * 4, sampling=True)
>
>     for i in range(100):
>          if i%10==0: print(".", end="")
>          ret = state.multiflip_mcmc_sweep(niter=10, beta=np.inf,
>     verbose=True)
This is correct. But note that the "sampling=True" option is no longer
needed.

> I am currently running this for a fully connected bipartite graph with
> 3454 nodes and 55008 edges. I understand it would take longer than the
> non-overlapping version, but do you have any suggestions on how to speed
> it up? The non-overlapping version takes about 15 minutes, while the
> overlapping version is still running after 1 day.
The new version will contain a much faster code for the overlapping case!

But in the mean-time, what you can do is to fit the non-overlapping
model first, and use that as a starting point to the MCMC with overlap.
You do that simply by doing:

        state = state.copy(state_args=dict(overlap=True))

Best,
Tiago




--
Tiago de Paula Peixoto <tiago@skewed.de>
_______________________________________________
graph-tool mailing list -- graph-tool@skewed.de
To unsubscribe send an email to graph-tool-leave@skewed.de