def full_cost(tree_adj, tree, gt_graph, X):
import graph_tool.centrality as gt
from bidict import bidict
vbetween, ebetween = gt.betweenness(gt_graph)
info, prt = augmented_ancestor(tree_adj, X)
root = None
for u, parent in prt.items():
if parent is None:
root = u
break
assert root is not None
Xbar = set(tree_adj.keys()) - X
edge_to_vertex = bidict({})
raw_edge_features = {}
for edge in tree:
tail = edge_tail(edge, prt)[0]
edge_to_vertex[edge] = tail
raw_edge_features[edge] = [ebetween[gt_graph.edge(*edge)],
vbetween[gt_graph.vertex(tail)],
info[tail][1]/len(tree_adj),
info[tail][2]/len(X), [], []]
distances = []
for i in X:
for j in Xbar:
path = actual_tree_path(i, j, prt)
path_len = len(path)
distances.append(path_len)
if j != root:
raw_edge_features[edge_to_vertex[:j]][4].append(path_len)
for edge in path:
raw_edge_features[edge][5].append(path_len)
return raw_edge_features, sum(distances)/len(distances)
def calculate_measures(g, tmp_measures=None, measure_list=['BC', 'T', 'E']):
if tmp_measures is None:
tmp_measures = dict((k, []) for k in measure_list)
tmp_measures['BC'].append(np.mean(gtc.betweenness(g)[0].get_array()))
tmp_measures['T'].append(gtclust.global_clustering(g)[0])
tmp_measures['E'].append(np.mean(gtc.closeness(g,harmonic=True).get_array()))
return tmp_measures
def calculate_betweenness(map_object, original_data_source, filtered_name_string):
betweenness_property_map=betweenness(map_object.graph_object)[0]
betweennes_list=[]
for station_name_string in original_data_source.data["station_names"]:
if station_name_string==filtered_name_string:
betweennes_list.append(-1)
else:
vertex=map_object.station_name_index[station_name_string]
betweennes_list.append(betweenness_property_map[vertex])
return betweennes_list
def calculate_measures(g, tmp_measures=None, measure_list=['BC', 'T', 'E']):
if tmp_measures is None:
tmp_measures = dict((k, []) for k in measure_list)
tmp_measures['BC'].append(np.mean(gtc.betweenness(g)[0].get_array()))
tmp_measures['T'].append(gtclust.global_clustering(g)[0])
tmp_measures['E'].append(
np.mean(gtc.closeness(g, harmonic=True).get_array()))
return tmp_measures
def set_properties(self, subgraph): v_betweenness, e_betweenness = betweenness(subgraph) subgraph.vertex_properties["vertex_betweenness"] = v_betweenness subgraph.edge_properties["edge_betweenness"] = e_betweenness v_closeness = closeness(subgraph) subgraph.vertex_properties["closeness"] = v_closeness l_clustering = local_clustering(subgraph) subgraph.vertex_properties["local_clustering"] = l_clustering bicomp, articulation, nc = label_biconnected_components(subgraph) subgraph.vertex_properties["articulation"] = articulation return subgraph
def plot_centralities(network, title="Centrality measures"):
g = network.graph
comm_size = g.num_vertices()
closeness = centrality.closeness(g).get_array().tolist()
max_eigenval, eigenvec = centrality.eigenvector(g)
# eigenvector = [x/max_eigenval for x in eigenvec.get_array().tolist()] #normalize!
eigenvector = eigenvec.get_array().tolist()
betw, _edges = centrality.betweenness(g, norm=True)
betweenness = betw.get_array().tolist()
P.suptitle(title)
# P.figure()
print "nans", len([x for x in closeness if isnan(x)])
closeness = [0 if isnan(x) else x for x in closeness]
# closeness = [x for x in closeness if not isnan(x)]
closeness = _filtered(closeness, comm_size)
print "closeness", closeness
print "non zeros", len([x for x in closeness if x != 0])
P.subplot(2, 2, 1)
plot_hist(closeness)
P.xlabel("Closeness centrality")
P.ylabel("Number of nodes (total={})".format(len(closeness)))
counts, degrees = stats.vertex_hist(g, "in", float_count=False)
print "counts : ", len(counts), counts
print "degrees: ", len(degrees), degrees
counts = list(counts)
counts.append(0)
P.subplot(2, 2, 2)
P.bar(degrees, counts, align='center', color="#348ABD")
# P.hist(counts, bins=degrees, )
P.xlabel("Degree centrality (in)")
P.ylabel("Number of nodes (total={})".format(sum(counts)))
P.xlim(0, max(degrees))
betweenness = _filtered(betweenness, comm_size)
print "betweenness", betweenness
P.subplot(2, 2, 3)
plot_hist(betweenness)
P.xlabel("Betweenness centrality")
P.ylabel("Number of nodes (total={})".format(len(betweenness)))
eigenvector = _filtered(eigenvector, comm_size)
print "eigenvector", eigenvector
P.subplot(2, 2, 4)
plot_hist(eigenvector)
P.xlabel("Eigenvector centrality")
P.ylabel("Number of nodes (total={})".format(len(eigenvector)))
P.show()
def get_dataframe_all_topolog_metrics(self):
graph = self.get_graph()
eprop_trust = graph.new_edge_property('double')
start_time = time.time()
for e in graph.edges():
v_name_s = graph.vertex_properties['name_proteins'][e.source()]
v_number_s = self.dict_genes[v_name_s]
v_name_t = graph.vertex_properties['name_proteins'][e.target()]
v_number_t = self.dict_genes[v_name_t]
eprop_trust[e] = self.adjacency_matrix[v_number_s, v_number_t]
graph.edge_properties['trust'] = eprop_trust
print('confidence score за :',
'--- %s seconds ---' % (time.time() - start_time))
list_metrics = [
'betweenness', 'pagerank', 'closeness', 'katz', 'hits_authority',
'hits_hub', 'eigenvector', 'eigentrust'
] # 'trust_transitivity'
dict_map = {}
start_time = time.time()
dict_map['betweenness'] = ct.betweenness(graph)[0]
dict_map['pagerank'] = ct.pagerank(graph)
dict_map['closeness'] = ct.closeness(graph)
dict_map['katz'] = ct.katz(graph)
dict_map['hits_authority'] = ct.hits(graph)[1]
dict_map['hits_hub'] = ct.hits(graph)[2]
dict_map['eigenvector'] = ct.eigenvector(graph)[1]
#print('trust_transitivity')
#"dict_map['trust_transitivity'] = ct.trust_transitivity(graph, graph.edge_properties["trust"])
print('все метрики кроме eigentrust за :',
'--- %s seconds ---' % (time.time() - start_time))
start_time = time.time()
dict_map['eigentrust'] = ct.eigentrust(graph,
graph.edge_properties['trust'],
max_iter=10**6)
print('eigentrust за :',
'--- %s seconds ---' % (time.time() - start_time))
start_time = time.time()
dict_metrics = {}
for key in list_metrics:
dict_metrics[key] = []
for v in graph.vertices():
for metric in list_metrics:
dict_metrics[metric].append(dict_map[metric][v])
dataframe_all_topolog_metrics = pd.DataFrame(dict_metrics)
dataframe_all_topolog_metrics.index = graph.vertex_properties[
'name_proteins']
print('получила датафрейм с метриками за :',
'--- %s seconds ---' % (time.time() - start_time))
return dataframe_all_topolog_metrics
def betweenness_influent_nodes(graph):
vertex_prop, edge_prop = gc.betweenness(graph,
weight=graph.edge_properties["weight"])
gd.graph_draw(graph,
vertex_fill_color=vertex_prop,
vertex_size=gd.prop_to_size(vertex_prop, mi=5, ma=15),
edge_pen_width=gd.prop_to_size(edge_prop, mi=0.5, ma=5),
vcmap=matplotlib.cm.gist_heat,
vorder=vertex_prop,
output_size=(1920, 1080),
output=FIGURES_PATH + BETWEENNESS_VIZ)
print("Visualization finished and saved...")
def graph_betweenness_centrality(graph):
nodes, edges = betweenness(graph)
nodes_dataframe = pandas.DataFrame(nodes.get_array())
edges_dataframe = pandas.DataFrame(edges.get_array())
nodes_dataframe.columns = [betweenness_centrality_key]
edges_dataframe.columns = [betweenness_centrality_key]
nodes_dataframe = sort_for_betweenness_centrality(nodes_dataframe)
edges_dataframe = sort_for_betweenness_centrality(edges_dataframe)
return nodes_dataframe, edges_dataframe
def betweenness_correlated_weights(graph,dicProperties,nEdges,nExc): lstWeights = np.zeros(nEdges) rMin = dicProperties["Min"] rMax = dicProperties["Max"] vpropBetw,epropBetw = betweenness(graph) arrBetw = epropBetw.a.copy() arrLogBetw = np.log10(arrBetw) rMaxLogBetw = arrLogBetw.max() rMinLogBetw = arrLogBetw.min() arrLogBetw = -5 + 2 * (arrLogBetw - rMinLogBetw ) / (rMaxLogBetw - rMinLogBetw) arrBetw = np.exp(np.log(10) * arrLogBetw) rMaxBetw = arrBetw.max() rMinBetw = arrBetw.min() lstWeights = np.multiply(arrBetw-rMinBetw,rMax/rMaxBetw) + rMin return lstWeights
def execute(now):
session = store.get_session()
topology = NetworkTopology(session, now)
(node_betweenness, link_betweenness) = gt.betweenness(topology.topology)
closeness = gt.closeness(topology.topology)
for v in topology.nodes:
topology.node_information[v].degree = topology.nodes[v].out_degree()
topology.node_information[v].closeness = closeness[topology.nodes[v]]
topology.node_information[v].betweenness = node_betweenness[topology.nodes[v]]
for l in topology.links:
topology.link_information[l].betweenness = link_betweenness[topology.links[l]]
session.commit()
def build_pivot_generator(N, graph, shared_edges, pivot_strategy, pivot_gen):
"""Return a vertex generator according to the chosen strategy"""
if shared_edges:
non_shared = non_shared_vertices(N, shared_edges)
return iter(lambda: r.choice(non_shared), -1)
if pivot_strategy is PivotStrategy.weighted:
assert pivot_gen, "provide your own generator"
return pivot_gen
if pivot_strategy in [PivotStrategy.uniform, PivotStrategy.no_pivot]:
return None
if pivot_strategy is PivotStrategy.by_degree:
degs = graph.degree_property_map('total').a
weights = np.exp(-degs)/np.sum(np.exp(-degs))
return WeightedRandomGenerator(weights)
if pivot_strategy is PivotStrategy.by_betweenness:
vb, _ = centrality.betweenness(graph)
vb = N*vb.a/5
weights = np.exp(-vb)/np.sum(np.exp(-vb))
return WeightedRandomGenerator(weights)
raise ValueError('check build_pivot_generator call')
def analyze(dir_name):
g = gload.load_graph(dir_name)
with open(os.path.join(dir_name, GRAPH_FILE), 'wb') as f:
pickle.dump(g, f)
sim = galg.simrank(g)
with open(os.path.join(dir_name, SIMRANK_FILE), 'wb') as f:
pickle.dump(sim, f)
prank = galg.psimrank(g)
with open(os.path.join(dir_name, PRANK_FILE), 'wb') as f:
pickle.dump(prank, f)
gr = grdf.calculate_taxonomy2(g)
with open(os.path.join(dir_name, GR_FILE), 'wb') as f:
pickle.dump(gr, f)
vp, ep = centrality.betweenness(analsis_dict['graph'])
with open(os.path.join(dir_name, BET_FILE), 'wb') as f:
pickle.dump(vp, f)
def calculate_cen_b(pre_u, type, pos_type):
# if type == "DeC":
if type == "BetC":
v_be = pre_u.new_vertex_property("float")
v_betweenness = gtc.betweenness(pre_u, vprop=v_be)
pre_u.vertex_properties['betweenness'] = v_be
if type == "CloC":
v_clo = pre_u.new_vertex_property("float")
v_closeness = gtc.closeness(pre_u, vprop=v_clo)
pre_u.vertex_properties['closeness'] = v_clo
# if type == "EigenveC":
# if type == "KatzC":
# if type == "EigentrustC":
if pos_type == "FD":
pos = gt.sfdp_layout(pre_u, p=2.6, K=40, C=1)
if pos_type == "FR":
pos = gt.fruchterman_reingold_layout(pre_u, n_iter=100, r=100, a=10)
nodes = []
convert = []
for v in pre_u.vertices():
s_node = {}
s_node["name"] = v_userid[v]["userid"]
s_node['race'] = v_race[v]["race"]
s_node['cohort'] = v_cohort[v]['cohort']
s_node['gender'] = v_gender[v]['gender']
s_node['c1net'] = v_c1net[v]['c1net']
s_node['c2net'] = v_c2net[v]['c2net']
s_node['c3net'] = v_c3net[v]['c3net']
s_node['c4net'] = v_c4net[v]['c4net']
s_node['c5net'] = v_c5net[v]['c5net']
s_node['c6net'] = v_c6net[v]['c6net']
s_node['c7net'] = v_c7net[v]['c7net']
s_node["out-degree"] = v.out_degree()
s_node["in-degree"] = v.in_degree()
s_node["cx"] = (pos[v][0])
s_node["cy"] = (pos[v][1])
if 'v_be' in locals().keys():
s_node["BetC"] = v_be[v]
if 'v_clo' in locals().keys():
if math.isnan(v_clo[v]):
v_clo[v] = 0
s_node["CloC"] = v_clo[v]
convert.append(v_userid[v]["userid"])
# o_node={}
# o_node[v_userid[v]["userid"]]=s_node
nodes.append(s_node)
all_data = {}
all_data['nodes'] = nodes
links = []
for e in pre_u.edges():
s_edge = {}
s_edge['source'] = convert.index(str(e_source[e]))
s_edge['target'] = convert.index(str(e_target[e]))
s_edge['type'] = e_type[e]['type']
s_edge['year'] = e_year[e]['year']
s_edge['weight'] = len(pre_u.edge(e.target(), e.source(), all_edges=True, add_missing=False)) + \
len(pre_u.edge(e.source(), e.target(),
all_edges=True, add_missing=False))
links.append(s_edge)
all_data['links'] = links
return (all_data)
def _get_edge_centrality(graph):
_, centrality = betweenness(graph, norm=False)
return centrality
def save_centrality(g, node_out_fname, edge_out_fname, weight=None):
"""
:param g: `Graph` instance
:return: None
"""
df = pd.DataFrame()
df['node'] = pd.Series(np.array([int(v) for v in g.vertices()]))
# degree
print('Degree')
num_nodes = len(g.get_vertices())
denom = num_nodes - 1
if g.is_directed():
unnormalized_in_degree = np.array([v.in_degree() for v in g.vertices()])
unnormalized_out_degree = np.array([v.out_degree() for v in g.vertices()])
df['unnormalized_in_degree'] = unnormalized_in_degree
df['unnormalized_out_degree'] = unnormalized_out_degree
df['in_degree'] = unnormalized_in_degree / denom
df['out_degree'] = unnormalized_out_degree / denom
else:
# check whether weighted graph or not
if weight:
unnormalized_degree = np.zeros(num_nodes)
edge_weights = np.array(weight.get_array())
for edge, w in zip(g.get_edges(), edge_weights):
for node in edge[:2]:
unnormalized_degree[node] += w
df['unnormalized_degree'] = unnormalized_degree
df['degree'] = unnormalized_degree / denom
else:
unnormalized_degree = np.array([v.out_degree() for v in g.vertices()])
df['unnormalized_degree'] = unnormalized_degree
df['degree'] = unnormalized_degree / denom
# closeness
print('Closeness')
df['unnormalized_closeness'] = np.array(closeness(g, weight=weight, norm=False).get_array())
df['closeness'] = np.array(closeness(g, weight=weight, norm=True).get_array())
# pageRank
print('PageRank')
df['pagerank'] = np.array(pagerank(g, weight=weight).get_array())
# betweenness
print('Betweenness')
un_node_between, un_edge_between = betweenness(g, weight=weight, norm=False)
node_between, edge_between = betweenness(g, weight=weight, norm=True)
df['unnormalized_betweenness'] = np.array(un_node_between.get_array())
df['betweenness'] = np.array(node_between.get_array())
df.to_csv(node_out_fname, index=False)
# edge
sources = []
targets = []
for e in g.edges():
source, target = list(map(int, [e.source(), e.target()]))
sources.append(source)
targets.append(target)
df = pd.DataFrame()
df['source'] = pd.Series(np.array(sources))
df['target'] = np.array(targets)
# betweenness
df['unnormalized_betweenness'] = np.array(un_edge_between.get_array())
df['betweenness'] = np.array(edge_between.get_array())
df.to_csv(edge_out_fname, index=False)
#!/usr/bin/env python3
# This file is part of Project Ruprecht. See COPYING for license.
import sys
import graph_tool as gt
import graph_tool.centrality as centr
def print_top_e(g, eprops):
names = g.vertex_properties['name']
ee = list(g.edges())
ee.sort(key=lambda a: eprops[a], reverse=True)
for e in ee:
print(names[e.source()], names[e.target()])
g = gt.load_graph_from_csv(sys.argv[1], csv_options={'delimiter': "\t"})
btwv, btwe = centr.betweenness(g)
print_top_e(g, btwe)
def betweenness_centrality(g: Graph):
return centrality.betweenness(g, weight=g.edge_properties['weight'])[0]
def calculate_cen(pre_u, type, pos_type):
# if type == "DeC":
if type == "BetC":
v_be = pre_u.new_vertex_property("float")
v_betweenness = gtc.betweenness(pre_u, vprop=v_be)
pre_u.vertex_properties['betweenness'] = v_be
if type == "CloC":
v_clo = pre_u.new_vertex_property("float")
v_closeness = gtc.closeness(pre_u, vprop=v_clo)
pre_u.vertex_properties['closeness'] = v_clo
# if type == "EigenveC":
# if type == "KatzC":
# if type == "EigentrustC":
if pos_type == "FD":
pos = gt.sfdp_layout(pre_u, p=2.6, K=40, C=1)
if pos_type == "FR":
pos = gt.fruchterman_reingold_layout(pre_u, n_iter=100, r=100, a=10)
nodes = []
convert = []
for v in pre_u.vertices():
s_node = {}
s_node["name"] = v_userid[v]["userid"]
s_node['gender'] = v_gender[v]["gender"]
s_node['schregion'] = v_schregion[v]['schregion']
s_node['school'] = v_school[v]['school']
s_node['major'] = v_major[v]['major']
s_node['marriage'] = v_marriage[v]['marriage']
s_node['grade'] = v_grade[v]['grade']
s_node['liveplace'] = v_liveplace[v]['liveplace']
s_node['height'] = v_height[v]['height']
s_node['weight'] = v_weight[v]['weight']
s_node['scorelevel'] = v_scorelevel[v]['scorelevel']
s_node['birthyear'] = v_birthyear[v]['birthyear']
s_node["out-degree"] = v.out_degree()
s_node["in-degree"] = v.in_degree()
s_node["cx"] = pos[v][0]
s_node["cy"] = pos[v][1]
if 'v_be' in locals().keys():
s_node["BetC"] = v_be[v]
if 'v_clo' in locals().keys():
if math.isnan(v_clo[v]):
v_clo[v] = 0
s_node["CloC"] = v_clo[v]
# print(v_clo[v])
convert.append(v_userid[v]["userid"])
# o_node={}
# o_node[v_userid[v]["userid"]]=s_node
nodes.append(s_node)
all_data = {}
all_data['nodes'] = nodes
links = []
for e in pre_u.edges():
s_edge = {}
s_edge['source'] = convert.index(str(e_source[e]))
s_edge['target'] = convert.index(str(e_target[e]))
s_edge['date'] = e_date[e]['date']
s_edge['dailycount'] = e_dailycount[e]['dailycount']
s_edge['weight'] = len(pre_u.edge(e.target(), e.source(), all_edges=True, add_missing=False)) + \
len(pre_u.edge(e.source(), e.target(),
all_edges=True, add_missing=False))
links.append(s_edge)
all_data['links'] = links
return(all_data)
def _get_vertex_centrality(graph):
centrality, _ = betweenness(graph, weight=graph.ep.weight, norm=False)
return centrality