def test_onmi(self):
g = nx.karate_club_graph()
lp_communities = label_propagation(g)
lp2_communities = label_propagation(g)
score = evaluation.overlapping_normalized_mutual_information_MGH(lp2_communities, lp_communities)
self.assertLessEqual(score.score, 1)
self.assertGreaterEqual(score.score, 0)
score = evaluation.overlapping_normalized_mutual_information_LFK(lp2_communities, lp_communities)
self.assertLessEqual(score.score, 1)
self.assertGreaterEqual(score.score, 0)
def test_lp(self):
g = get_string_graph()
coms = algorithms.label_propagation(g)
self.assertEqual(type(coms.communities), list)
if len(coms.communities) > 0:
self.assertEqual(type(coms.communities[0]), list)
self.assertEqual(type(coms.communities[0][0]), str)
def test_variation_of_information(self):
g = nx.karate_club_graph()
lp_communities = label_propagation(g)
louvain_communities = louvain(g)
score = evaluation.variation_of_information(louvain_communities, lp_communities)
self.assertLessEqual(score.score, np.log(g.number_of_nodes()))
self.assertGreaterEqual(score.score, 0)
def test_adjusted_rand(self):
g = nx.karate_club_graph()
lp_communities = label_propagation(g)
louvain_communities = louvain(g)
score = evaluation.adjusted_rand_index(louvain_communities, lp_communities)
self.assertLessEqual(score.score, 1)
self.assertGreaterEqual(score.score, 0)
def test_f1(self):
g = nx.karate_club_graph()
lp_communities = label_propagation(g)
louvain_communities = louvain(g)
score = evaluation.f1(louvain_communities, lp_communities)
self.assertIsInstance(score, evaluation.MatchingResult)
def test_closeness_simple(self):
g = nx.karate_club_graph()
lp_communities = label_propagation(g)
louvain_communities = louvain(g)
score = evaluation.partition_closeness_simple(louvain_communities, lp_communities)
self.assertLessEqual(score.score, 1)
self.assertGreaterEqual(score.score, 0)
def test_nmi(self):
g = nx.karate_club_graph()
louvain_communities = louvain(g)
lp_communities = label_propagation(g)
score = evaluation.normalized_mutual_information(louvain_communities, lp_communities)
self.assertLessEqual(score.score, 1)
self.assertGreaterEqual(score.score, 0)
def test_plot_sim_matrix(self):
g = nx.karate_club_graph()
coms = algorithms.louvain(g)
coms2 = algorithms.label_propagation(g)
viz.plot_sim_matrix([coms, coms2],
evaluation.adjusted_mutual_information)
plt.savefig("cluster.pdf")
os.remove("cluster.pdf")
def test_endntm(self):
G = nx.karate_club_graph()
coms_l = [
algorithms.louvain(G),
algorithms.label_propagation(G),
algorithms.walktrap(G),
]
coms = algorithms.endntm(G, coms_l)
self.assertEqual(type(coms.communities), list)
if len(coms.communities) > 0:
self.assertEqual(type(coms.communities[0]), list)
self.assertEqual(type(coms.communities[0][0]), int)
def test_plot_com_stat(self):
g = nx.karate_club_graph()
coms = algorithms.louvain(g)
coms2 = algorithms.label_propagation(g)
viz.plot_com_stat([coms, coms2], evaluation.size)
plt.savefig("cluster.pdf")
os.remove("cluster.pdf")
viz.plot_com_stat(coms, evaluation.size)
plt.savefig("cluster.pdf")
os.remove("cluster.pdf")
def test_ranking_comp(self):
g = nx.karate_club_graph()
coms = algorithms.louvain(g)
coms2 = algorithms.kclique(g, 2)
coms3 = algorithms.label_propagation(g)
rk = evaluation.ComparisonRanking([coms, coms2, coms3])
rk.rank(evaluation.overlapping_normalized_mutual_information_LFK)
rk.rank(evaluation.overlapping_normalized_mutual_information_MGH)
rk.rank(evaluation.omega)
rnk, _ = rk.topsis()
self.assertEqual(len(rnk), 3)
pc = rk.bonferroni_post_hoc()
self.assertLessEqual(len(pc), 4)
def test_plot_com_properties_relation(self):
g = nx.karate_club_graph()
coms = algorithms.louvain(g)
coms2 = algorithms.label_propagation(g)
viz.plot_com_properties_relation([coms, coms2], evaluation.size,
evaluation.internal_edge_density)
plt.savefig("cluster.pdf")
os.remove("cluster.pdf")
viz.plot_com_properties_relation(coms, evaluation.size,
evaluation.internal_edge_density)
plt.savefig("cluster.pdf")
os.remove("cluster.pdf")
def test_comparison(self):
g = nx.karate_club_graph()
coms = algorithms.louvain(g)
coms2 = algorithms.label_propagation(g)
self.assertIsInstance(coms.normalized_mutual_information(coms2).score, float)
self.assertIsInstance(
coms.overlapping_normalized_mutual_information_LFK(coms2).score, float
)
self.assertIsInstance(
coms.overlapping_normalized_mutual_information_MGH(coms2).score, float
)
self.assertIsInstance(coms.omega(coms2).score, float)
self.assertIsInstance(coms.f1(coms2), evaluation.MatchingResult)
self.assertIsInstance(coms.nf1(coms2).score, float)
self.assertIsInstance(coms.adjusted_mutual_information(coms2).score, float)
self.assertIsInstance(coms.adjusted_rand_index(coms2).score, float)
self.assertIsInstance(coms.variation_of_information(coms2).score, float)
def community_discoverying_algorithms(g):
"""
All Community Discovery algorithms generate as result a NodeClustering object, allowing
also for the generation of a JSON representation of the results. Then evaluate the clusters with fitness
functions (ex. synthetic representation of its min/max/mean/std values ORD communitiy-wise value)
"""
print("Starting computing angel_coms")
angel_coms = algorithms.angel(g.to_undirected(), threshold=0.25)
write_on_file(angel_coms, "communities/angel.json")
draw_community_graph(g, angel_coms, "communities/angel.png")
print("END")
print("Starting computing infomap_coms")
infomap_coms = algorithms.infomap(g.to_undirected())
write_on_file(infomap_coms, "communities/infomap.json")
draw_community_graph(g, infomap_coms, "communities/infomap.png")
print("END")
print("Starting computing louvain_coms")
louvain_coms = algorithms.louvain(g.to_undirected())
write_on_file(louvain_coms, "communities/louvain.json")
draw_community_graph(g, louvain_coms, "communities/louvain.png")
print("END")
print("Starting computing labelpropagation_coms")
labelpropagation_coms = algorithms.label_propagation(g.to_undirected())
write_on_file(labelpropagation_coms, "communities/labelpropagation.json")
draw_community_graph(g, labelpropagation_coms,
"communities/labelpropagation.png")
print("END")
draw_cluster_violin_map(
[angel_coms, infomap_coms, louvain_coms, labelpropagation_coms])
draw_cluster_heatmap(
[angel_coms, infomap_coms, louvain_coms, labelpropagation_coms])
draw_plot_map([angel_coms, infomap_coms], 1)
draw_plot_map([angel_coms, louvain_coms], 2)
draw_plot_map([angel_coms, labelpropagation_coms], 3)
draw_plot_map([infomap_coms, louvain_coms], 4)
draw_plot_map([infomap_coms, labelpropagation_coms], 5)
draw_plot_map([louvain_coms, labelpropagation_coms], 6)
def test_ranking(self):
g = nx.karate_club_graph()
coms = algorithms.louvain(g)
coms2 = algorithms.demon(g, 0.25)
coms3 = algorithms.label_propagation(g)
coms4 = algorithms.angel(g, 0.6)
rk = evaluation.FitnessRanking(g, [coms2, coms, coms3, coms4])
rk.rank(evaluation.fraction_over_median_degree)
rk.rank(evaluation.edges_inside)
rk.rank(evaluation.cut_ratio)
rk.rank(evaluation.erdos_renyi_modularity)
rk.rank(evaluation.newman_girvan_modularity)
rk.rank(evaluation.modularity_density)
rnk, _ = rk.topsis()
self.assertEqual(len(rnk), 4)
pc = rk.bonferroni_post_hoc()
self.assertLessEqual(len(pc), 4)
elif (options.method == 'der'):
communities = algorithms.der(g, **clust_kwargs)
elif (options.method == 'edmot'):
communities = algorithms.edmot(g, **clust_kwargs)
elif (options.method == 'eigenvector'):
communities = algorithms.eigenvector(g, **clust_kwargs)
elif (options.method == 'gdmp2'):
communities = algorithms.gdmp2(g, **clust_kwargs)
elif (options.method == 'greedy_modularity'):
communities = algorithms.greedy_modularity(g,
weight='weight',
**clust_kwargs)
#elif(options.method == 'infomap'):
# communities = algorithms.infomap(g)
elif (options.method == 'label_propagation'):
communities = algorithms.label_propagation(g, **clust_kwargs)
elif (options.method == 'markov_clustering'):
communities = algorithms.markov_clustering(g, **clust_kwargs)
elif (options.method == 'rber_pots'):
communities = algorithms.rber_pots(g, weights='weight', **clust_kwargs)
elif (options.method == 'rb_pots'):
communities = algorithms.rb_pots(g, weights='weight', **clust_kwargs)
elif (options.method == 'significance_communities'):
communities = algorithms.significance_communities(g, **clust_kwargs)
elif (options.method == 'spinglass'):
communities = algorithms.spinglass(g, **clust_kwargs)
elif (options.method == 'surprise_communities'):
communities = algorithms.surprise_communities(g, **clust_kwargs)
elif (options.method == 'walktrap'):
communities = algorithms.walktrap(g, **clust_kwargs)
#elif(options.method == 'sbm_dl'):
ari_markov = adjusted_rand_score(true_communities, markov_communities)
vi_markov = variation_of_information(true_partitions, markov_partitions)
############################### Greedy ###############################
start_time = time.time()
greedy_partition = cd.greedy_modularity(G)
greedy_time = time.time() - start_time
greedy_communities = extract_communities_list(greedy_partition.communities, G)
greedy_partitions = get_partitions(greedy_communities)
nmi_greedy = normalized_mutual_info_score(true_communities, greedy_communities)
ari_greedy = adjusted_rand_score(true_communities, greedy_communities)
vi_greedy = variation_of_information(true_partitions, greedy_partitions)
############################### Label Propagation ###############################
start_time = time.time()
propagation_partition = cd.label_propagation(G)
propagation_time = time.time() - start_time
propagation_communities = extract_communities_list(
propagation_partition.communities, G)
propagation_partitions = get_partitions(propagation_communities)
nmi_propagation = normalized_mutual_info_score(true_communities,
propagation_communities)
ari_propagation = adjusted_rand_score(true_communities,
propagation_communities)
vi_propagation = variation_of_information(true_partitions,
propagation_partitions)
nmi_x = np.arange(8)
ari_x = [x + 0.3 for x in nmi_x]
vi_x = [x + 0.3 for x in ari_x]
walktrap_partition = cd.walktrap(LFR_G)
walktrap_labels = extract_communities_list(walktrap_partition.communities)
nmi_walktrap.append(normalized_mutual_info_score(true_labels, walktrap_labels))
############################### Markov Clustering ###############################
markov_partition = cd.markov_clustering(LFR_G)
markov_labels = extract_communities_list(markov_partition.communities)
nmi_markov.append(normalized_mutual_info_score(true_labels, markov_labels))
############################### Greedy ###############################
greedy_partition = cd.greedy_modularity(LFR_G)
greedy_labels = extract_communities_list(greedy_partition.communities)
nmi_greedy.append(normalized_mutual_info_score(true_labels, greedy_labels))
############################### Label Propagation ###############################
propagation_partition = cd.label_propagation(LFR_G)
propagation_labels = extract_communities_list(propagation_partition.communities)
nmi_propagation.append(normalized_mutual_info_score(true_labels, propagation_labels))
#Plot NMI scores
nmi_graph = plt.gca()
nmi_graph.set_xlim([0, 0.9])
nmi_graph.set_ylim([-0.1, 1])
nmi_graph.plot(np.arange(0.1, 1, 0.1), nmi_infomap, color='#575757',
marker='o', mfc='#f1362b', mec='#f1362b', label="Infomap")
nmi_graph.plot(np.arange(0.1, 1, 0.1), nmi_eigenvector, color='#575757',
marker='o', mfc='#17c436', mec='#17c436', label="Eigenvector")
nmi_graph.plot(np.arange(0.1, 1, 0.1), nmi_leiden, color='#575757',
marker='o', mfc='#d9a000', mec='#d9a000', label="Leiden")
nmi_graph.plot(np.arange(0.1, 1, 0.1), nmi_walktrap, color='#575757',
marker='o', mfc='#532ce9', mec='#532ce9', label="Walktrap")
cs.erdos_renyi_modularity().score))
print("{0:>15s} | {1:.6f}".format(
'Robustness',
cs.normalized_mutual_information(alg(G)).score))
print("{0:>15s} | {1:.1f} sec\n".format('Timing', time() - tic))
return cs
# G = read('toy')
G = read('karate')
# G = read('women')
# G = read('dolphins')
# G = read('got-appearance')
# G = read('diseasome')
# G = read('wars')
# G = read('transport')
# G = read('java')
# G = read('imdb')
# G = read('wikileaks')
info(G)
clusters(G, lambda G: algorithms.girvan_newman(G, level=1))
clusters(G, lambda G: algorithms.label_propagation(G))
cs = clusters(G, lambda G: algorithms.louvain(G))
# clusters(G, lambda G: algorithms.leiden(G))
# clusters(G, lambda G: algorithms.sbm_dl(G))
viz.plot_network_clusters(G, cs, nx.spring_layout(G))
plt.show()
