def test_best_partition(self):
partition = agglomerate.best_partition(self.g6, opt.delta,
opt.tolerance, 1, self.g6)
gret1, zret1 = agglomerate.induced_graph(self.label_dict1, self.g6,
self.g6)
gret2, zret2 = agglomerate.induced_graph(partition, self.g6, self.g6)
GM = nx.isomorphism.GraphMatcher(gret1, gret2)
assert GM.is_isomorphic()
def test_generate_dendogram(self): partition_list = agglomerate.generate_dendogram(self.g6, opt.delta, opt.tolerance, 1,nx.Graph()) # a e # | | # b---e---c => lpa => e---e---e => induced graph => e # | | # d e assert len(partition_list) == 1 print(partition_list[0]) gret1,zret1 = agglomerate.induced_graph(self.label_dict1,self.g6,self.g6) gret2,zret2 = agglomerate.induced_graph(partition_list[0],self.g6,self.g6) # partition list should be such that all nodes have the same label and # should therefore induce the same graph as label_dict1 GM = nx.isomorphism.GraphMatcher(gret1,gret2) assert GM.is_isomorphic()
def test_induced_graph(self):
gret,zret = agglomerate.induced_graph(self.label_dict1,self.g0,self.g1) # if all the nodes belong to the same community
assert len(gret.nodes()) == 1 # there will only be one node in the induced graph
assert len(zret.nodes()) == 1 # likewise for the induced zgraph
GM = nx.isomorphism.GraphMatcher(gret,zret)
assert GM.is_isomorphic() # if graph and zgraph are equal, the induced graph is the same
gret,zret = agglomerate.induced_graph(self.label_dict3,self.g1,self.g5)
# graph: b---b---a induced graph: 2
# | \ | => 1 [ b------a ] 3
# a---a
GM = nx.isomorphism.GraphMatcher(gret,self.g3)
assert GM.is_isomorphic()
# zgraph: b---b---a induced zgraph: 1
# | => 1 [ b------a ] 2
# a---a
GM = nx.isomorphism.GraphMatcher(zret,self.g4)
assert GM.is_isomorphic()
def test_generate_dendogram(self):
partition_list = agglomerate.generate_dendogram(
self.g6, opt.delta, opt.tolerance, 1, nx.Graph())
# a e
# | |
# b---e---c => lpa => e---e---e => induced graph => e
# | |
# d e
assert len(partition_list) == 1
print(partition_list[0])
gret1, zret1 = agglomerate.induced_graph(self.label_dict1, self.g6,
self.g6)
gret2, zret2 = agglomerate.induced_graph(partition_list[0], self.g6,
self.g6)
# partition list should be such that all nodes have the same label and
# should therefore induce the same graph as label_dict1
GM = nx.isomorphism.GraphMatcher(gret1, gret2)
assert GM.is_isomorphic()
def test_induced_graph(self):
gret, zret = agglomerate.induced_graph(
self.label_dict1, self.g0,
self.g1) # if all the nodes belong to the same community
assert len(gret.nodes()
) == 1 # there will only be one node in the induced graph
assert len(zret.nodes()) == 1 # likewise for the induced zgraph
GM = nx.isomorphism.GraphMatcher(gret, zret)
assert GM.is_isomorphic(
) # if graph and zgraph are equal, the induced graph is the same
gret, zret = agglomerate.induced_graph(self.label_dict3, self.g1,
self.g5)
# graph: b---b---a induced graph: 2
# | \ | => 1 [ b------a ] 3
# a---a
GM = nx.isomorphism.GraphMatcher(gret, self.g3)
assert GM.is_isomorphic()
# zgraph: b---b---a induced zgraph: 1
# | => 1 [ b------a ] 2
# a---a
GM = nx.isomorphism.GraphMatcher(zret, self.g4)
assert GM.is_isomorphic()
def test_best_partition(self): partition = agglomerate.best_partition(self.g6, opt.delta, opt.tolerance, 1,self.g6) gret1,zret1 = agglomerate.induced_graph(self.label_dict1,self.g6,self.g6) gret2,zret2 = agglomerate.induced_graph(partition,self.g6,self.g6) GM = nx.isomorphism.GraphMatcher(gret1,gret2) assert GM.is_isomorphic()
