def test_pool_filtered(self):
g = nx.karate_club_graph()
# Louvain
resolution = ensemble.Parameter(name="resolution",
start=0.1,
end=1,
step=0.1)
# randomize = ensemble.BoolParameter(name="randomize", value=False)
louvain_conf = [resolution]
# Demon
epsilon = ensemble.Parameter(name="epsilon",
start=0.1,
end=1,
step=0.1)
demon_conf = [epsilon]
methods = [algorithms.louvain, algorithms.demon]
for communities, scoring in \
ensemble.pool_grid_filter(g, methods, [louvain_conf, demon_conf],
quality_score=evaluation.erdos_renyi_modularity,
aggregate=max):
self.assertIsInstance(communities.communities, list)
self.assertIsInstance(scoring.score, float)
def test_pool(self):
g = nx.karate_club_graph()
# Louvain
resolution = ensemble.Parameter(name="resolution", start=0.1, end=1, step=0.1)
randomize = ensemble.BoolParameter(name="randomize")
louvain_conf = [resolution, randomize]
# Angel
threshold = ensemble.Parameter(name="threshold", start=0.1)
angel_conf = [threshold]
methods = [algorithms.louvain, algorithms.angel]
for communities in ensemble.pool(g, methods, [louvain_conf, angel_conf]):
self.assertIsInstance(communities.communities, list)
def test_grid(self):
g = nx.karate_club_graph()
resolution = ensemble.Parameter(name="resolution",
start=0.1,
end=1,
step=0.1)
for communities in ensemble.grid_execution(graph=g,
method=algorithms.louvain,
parameters=[resolution]):
self.assertIsInstance(communities.communities, list)
def test_pool(self):
g = nx.karate_club_graph()
# Louvain
resolution = ensemble.Parameter(name="resolution",
start=0.1,
end=1,
step=0.1)
# randomize = ensemble.BoolParameter(name="randomize")
louvain_conf = [resolution]
# Demon
epsilon = ensemble.Parameter(name="epsilon", start=0.1)
demon_conf = [epsilon]
methods = [algorithms.louvain, algorithms.demon]
for communities in ensemble.pool(g, methods,
[louvain_conf, demon_conf]):
self.assertIsInstance(communities.communities, list)
def test_pool_filtered(self):
g = nx.karate_club_graph()
# Louvain
resolution = ensemble.Parameter(name="resolution", start=0.1, end=1, step=0.1)
randomize = ensemble.BoolParameter(name="randomize", value=False)
louvain_conf = [resolution, randomize]
# Angel
threshold = ensemble.Parameter(name="threshold", start=0.1, end=1, step=0.1)
angel_conf = [threshold]
methods = [algorithms.louvain, algorithms.angel]
for communities, scoring in \
ensemble.pool_grid_filter(g, methods, [louvain_conf, angel_conf],
quality_score=evaluation.erdos_renyi_modularity,
aggregate=max):
self.assertIsInstance(communities.communities, list)
self.assertIsInstance(scoring, float)
def test_grid_search(self):
g = nx.karate_club_graph()
resolution = ensemble.Parameter(name="resolution", start=0.1, end=1, step=0.1)
randomize = ensemble.BoolParameter(name="randomize")
communities, scoring = ensemble.grid_search(graph=g, method=algorithms.louvain,
parameters=[resolution, randomize],
quality_score=evaluation.erdos_renyi_modularity,
aggregate=max)
self.assertIsInstance(communities.communities, list)
self.assertIsInstance(scoring, float)
def polling_and_optimisation():
threshold = ensemble.Parameter(name="threshold",
start=0.25,
end=1.0,
step=0.25) # numeric range
angel_conf = [threshold]
resolution = ensemble.Parameter(name="resolution",
start=0.1,
end=1,
step=0.1) # numeric range
louvain_conf = [resolution]
methods = [algorithms.angel, algorithms.louvain]
for coms, scoring in ensemble.pool_grid_filter(
g,
methods, [louvain_conf, angel_conf],
quality_score=evaluation.erdos_renyi_modularity,
aggregate=max):
print("%s\nCommunities:\n %s \nConfiguration: %s \nScoring: %s\n" %
(coms.method_name, coms.communities, coms.method_parameters,
scoring))
###########################################################################
### Testing Methods
###########################################################################
# graphs = [school_g]
# names = ['school']
# references = [school_gt]
results_scoring['lpam_python_cm'] = getResults(
clustering_method=lpam_python_cm,
graphs=graphs,
names=names,
references=references,
parameters_list = [ [ensemble.Parameter(name="k", start=max(1,len(r.communities)-3), end=len(r.communities)+1, step=1), \
ensemble.Parameter(name="threshold", start=0.25, end=0.75, step=0.05),\
ensemble.Parameter(name="seed", start=0, end=10, step=1)
] for r in references ]
)
# right implementaion of amplified commute distance
results_scoring['lpam_python_amp'] = getResults(
clustering_method=lpam_python_amp,
graphs=graphs,
names=names,
references=references,
parameters_list = [ [ensemble.Parameter(name="k", start=max(1,len(r.communities)-3), end=len(r.communities)+1, step=1), \
ensemble.Parameter(name="threshold", start=0.25, end=0.75, step=0.05),\
ensemble.Parameter(name="seed", start=0, end=10, step=1)
] for r in references ]
)
