[graph-tool] Re: overlap/mixed-membership sbm help
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
___
graph-tool mailing list -- graph-tool@skewed.de
To unsubscribe send an email to graph-tool-le...@skewed.de
[graph-tool] Core dump
Hi together, I have an algorithm whose results incl. the graph I save to disk at the end. When I load it again, on some vertices, shortest_distance() crashes python: ``` munmap_chunk(): invalid pointer Aborted (core dumped) ``` Any ideas on how I can debug this? All the property maps are internal and my custom vertex lists are converted to int before pickling and restored using gt.Graph().vertex() afterwards. Thank a lot for ideas! Cheers, Stephan ___ graph-tool mailing list -- graph-tool@skewed.de To unsubscribe send an email to graph-tool-le...@skewed.de
[graph-tool] Re: Core dump
Am 11.11.21 um 10:47 schrieb Monecke, Stephan: Any ideas on how I can debug this? Yes: try to isolate the problem by constructing a minimal, self-contained program that reproduces the crash. -- Tiago de Paula Peixoto ___ graph-tool mailing list -- graph-tool@skewed.de To unsubscribe send an email to graph-tool-le...@skewed.de
[graph-tool] Re: Core dump
I bisected manually until I found the problem: shortest_path() segfaults when the vertex is not present. - - MWE - - import graph_tool.all as gt gt.shortest_path(gt.Graph(), 1, 2) Segmentation fault (core dumped) - - o - - I wrote a ticket for that. Thanks for your help! Von: Tiago de Paula Peixoto Gesendet: Donnerstag, 11. November 2021 11:05 An: graph-tool@skewed.de Betreff: [graph-tool] Re: Core dump Am 11.11.21 um 10:47 schrieb Monecke, Stephan: > Any ideas on how I can debug this? > Yes: try to isolate the problem by constructing a minimal, self-contained program that reproduces the crash. -- Tiago de Paula Peixoto ___ graph-tool mailing list -- graph-tool@skewed.de To unsubscribe send an email to graph-tool-le...@skewed.de ___ graph-tool mailing list -- graph-tool@skewed.de To unsubscribe send an email to graph-tool-le...@skewed.de
