def test_callable_input(self, adj):
"""Tests if function returns the correct output given a custom method set by the user"""
objective_return = (0, [0])
def custom_method(*args, **kwargs):
"""Mockup of custom-method function fed to ``search``"""
return objective_return
result = subgraph.search(
graph=adj,
nodes=3,
iterations=10,
options={"heuristic": {
"method": custom_method
}})
assert result == objective_return
def test_valid_input(self, graph, monkeypatch, methods):
"""Tests if function returns the correct output under normal conditions. The method is here
monkey patched to return a known result."""
objective_return = (0, [0])
def custom_method(*args, **kwargs):
"""Mockup of custom-method function fed to ``search``"""
return objective_return
with monkeypatch.context() as m:
m.setattr(subgraph, "METHOD_DICT", {methods: custom_method})
result = subgraph.search(
graph=graph,
nodes=3,
iterations=10,
options={"heuristic": {
"method": methods
}})
assert result == objective_return
# .. raw:: html
# :file: ../../examples_gbs/planted.html
#
# .. note::
# The command ``plotly.offline.plot()`` is used to display plots in the documentation. In
# practice, you can simply use ``plot_graph.show()`` to view the graph.
##############################################################################
# A more interesting challenge is to find dense subgraphs of different sizes; it is often
# useful to identify many high-density subgraphs, not just the densest ones. This is the purpose of
# the :func:`~.subgraph.search` function in the :mod:`~.gbs.subgraph` module: to identify
# collections of dense subgraphs for a range of sizes. The output of this function is a
# dictionary whose keys correspond to subgraph sizes within the specified range. The values in
# the dictionary are the top subgraphs of that size and their corresponding density.
dense = subgraph.search(samples, pl_graph, 8, 16, max_count=3) # we look at top 3 densest subgraphs
for k in range(8, 17):
print(dense[k][0]) # print only the densest subgraph of each size
##############################################################################
# From the results of the search we learn that, depending on their size, the densest subgraphs
# belong to different regions of the graph: dense subgraphs of less than ten nodes are contained
# within the planted subgraph, whereas larger dense subgraphs appear outside of the planted
# subgraph. Smaller dense subgraphs can be cliques, characterized by having
# maximum density of 1, while larger subgraphs are less dense. Let's see what the smallest and
# largest subgraphs look like:
densest_8 = plot.graph(pl_graph, dense[8][0][1])
densest_16 = plot.graph(pl_graph, dense[12][0][1])
plotly.offline.plot(densest_8, filename="densest_8.html")
def process_planted(min_size, max_size, max_count, n_samples):
"""Fixture for loading samples from the Planted dataset"""
samples = p_planted[:n_samples]
d = subgraph.search(samples, g_planted, min_size, max_size, max_count)
return d