def test_attribute_mixing_dict_multigraph(self):
d = nx.attribute_mixing_dict(self.M, "fish")
d_result = {
"one": {"one": 4},
"two": {"two": 2},
}
assert d == d_result
def get_assortativity(network):
assortativity_dict = nx.attribute_mixing_dict(network,
'sex',
normalized=True)
if 'female' in assortativity_dict:
f2f = assortativity_dict['female']['female'] if 'female' in assortativity_dict['female']\
else 0
f2f = round(f2f, 3)
f2m = assortativity_dict['female']['male'] if 'male' in assortativity_dict['female']\
else 0
f2m = round(f2m, 3)
else:
f2f = 0
f2m = 0
if 'male' in assortativity_dict:
m2m = assortativity_dict['male']['male'] if 'male' in assortativity_dict['male']\
else 0
m2m = round(m2m, 3)
m2f = assortativity_dict['male']['female'] if 'female' in assortativity_dict['male']\
else 0
m2f = round(m2f, 3)
else:
m2m = 0
m2f = 0
return (f2f, m2m, f2m, m2f)
def test_attribute_mixing_dict_undirected(self):
d=nx.attribute_mixing_dict(self.G,'fish')
d_result={'one':{'one':2,'red':1},
'two':{'two':2,'blue':1},
'red':{'one':1},
'blue':{'two':1}
}
assert_equal(d,d_result)
def test_attribute_mixing_dict_undirected(self):
d = nx.attribute_mixing_dict(self.G, "fish")
d_result = {
"one": {"one": 2, "red": 1},
"two": {"two": 2, "blue": 1},
"red": {"one": 1},
"blue": {"two": 1},
}
assert d == d_result
def test_attribute_mixing_dict_multigraph(self):
d = nx.attribute_mixing_dict(self.M, 'fish')
d_result = {
'one': {
'one': 4
},
'two': {
'two': 2
},
}
assert d == d_result
def plotMixPattern(self, attribute, attribute_value):
mix_dict = nx.attribute_mixing_dict(self.graph, attribute)
norm_mix = self.mixingDictingToProb(mix_dict)
# O atributo abaixo define o valor de atributo que desejamos analisar o padrão de mixagem
# por exemplo, mixagem para o partido PT ou para o sexo Masculino
D = norm_mix[attribute_value]
plt.bar(range(len(D)), list(D.values()), align='center')
plt.xticks(range(len(D)), list(D.keys()), rotation='vertical')
plt.show()
def atribute_mixing(graph, probability):
d = nx.attribute_mixing_dict(graph, attribute, normalized=probability)
od = collections.OrderedDict(sorted(d.items()))
for key, value in d.items():
print("------------------------------------")
print("Atribute1: " + str(key))
print("------------------------------------")
ovalue = collections.OrderedDict(sorted(value.items()))
for k, v in value.items():
print("Atribute2: " + str(k), end='\t')
if probability == True:
print("Probability: " + str(v))
else:
print("Count: " + str(v))
def attribute_info():
c_list, m_list, p_list = util.get_lists()
cmp_list = util.append_arrays(c_list, m_list, p_list)
interaction_types = ['mentions', 'replies', 'retweets']
for interaction_type in interaction_types:
edge_list = util.get_edge_list(interaction_type)
g = create_graph_edge_weights(edge_list)
cmp_g = create_graph_subset(g, cmp_list)
add_types(cmp_g)
print('{} Assortativity: '.format(interaction_type),
nx.attribute_assortativity_coefficient(cmp_g, 'type'))
print('{} Mixing: '.format(interaction_type),
nx.attribute_mixing_dict(cmp_g, 'type', normalized=True))
def test_attribute_mixing_dict_directed(self):
d = nx.attribute_mixing_dict(self.D, 'fish')
d_result = {
'one': {
'one': 1,
'red': 1
},
'two': {
'two': 1,
'blue': 1
},
'red': {},
'blue': {}
}
assert d == d_result
def attribute_assortativity_coefficient(G, nodal_attribute):
mixing_dict = nx.attribute_mixing_dict(G, nodal_attribute)
total_sum_of_degree = float(2 * G.size())
sum_e_ii = 0
for row in mixing_dict:
column = row
sum_e_ii += mixing_dict[row][column] / total_sum_of_degree
sum_e_ij = 0
for row in mixing_dict:
sum_e_ij += sum(
np.asarray(mixing_dict[row].values()) / total_sum_of_degree
)**2 # (\sum_i e_ij)**2 as a_i == b_i for undirected graph.
return (sum_e_ii - sum_e_ij) / (1 - sum_e_ij)
def avrg_runner(link_pred_obj: LinkPrediction, count: int, mu: int,
basedir: str) -> np.array:
"""
Runs AVRG on the input graph and returns the combined adjacency matrix
"""
extract_type = 'mu_random'
mu = 5
clustering = 'leiden'
link_pred.set_method(method=f'AVRG_link_{clustering}_{mu}')
train_g: nx.Graph = nx.from_scipy_sparse_matrix(link_pred_obj.adj_train,
create_using=nx.Graph)
train_g.add_edges_from(
link_pred_obj.val_edges.tolist()) # add the validation edges too
nx.set_node_attributes(train_g,
name='value',
values=link_pred_obj.vals_dict)
list_of_list_clusters = get_clustering(g=train_g,
outdir=basedir,
clustering=clustering,
use_pickle=False,
filename='na',
write_pickle=False)
root = create_tree(list_of_list_clusters) if isinstance(
list_of_list_clusters, list) else list_of_list_clusters
train_lmg = nx_to_lmg(nx_g=train_g)
extractor = AVRGLinkExtractor(g=train_lmg,
attr_name=att_name,
clustering=clustering,
mu=mu,
extract_type=extract_type,
root=root)
avrg_link = extractor.extract()
mix_dict = nx.attribute_mixing_dict(train_g, 'value')
gen = AttributedEnsureAllNodesGenerator(grammar=avrg_link,
attr_name='value',
mixing_dict=mix_dict,
use_fancy_rewiring=True)
gen_graphs = gen.generate(10)
return gen_graphs
def test_attribute_mixing_dict_undirected(self):
d = nx.attribute_mixing_dict(self.G, 'fish')
d_result = {
'one': {
'one': 2,
'red': 1
},
'two': {
'two': 2,
'blue': 1
},
'red': {
'one': 1
},
'blue': {
'two': 1
}
}
assert_equal(d, d_result)
def test_attribute_mixing_dict_multigraph(self):
d=nx.attribute_mixing_dict(self.M,'fish')
d_result={'one':{'one':4},
'two':{'two':2},
}
assert_equal(d,d_result)
