def test_degree_p4_weighted(self):
G = nx.path_graph(4)
G[1][2]["weight"] = 4
answer = {1: 2.0, 2: 1.8}
nd = nx.average_degree_connectivity(G, weight="weight")
assert nd == answer
answer = {1: 2.0, 2: 1.5}
nd = nx.average_degree_connectivity(G)
assert nd == answer
D = G.to_directed()
answer = {2: 2.0, 4: 1.8}
nd = nx.average_degree_connectivity(D, weight="weight")
assert nd == answer
answer = {1: 2.0, 2: 1.8}
D = G.to_directed()
nd = nx.average_degree_connectivity(
D, weight="weight", source="in", target="in"
)
assert nd == answer
D = G.to_directed()
nd = nx.average_degree_connectivity(
D, source="in", target="out", weight="weight"
)
assert nd == answer
def test_degree_p4_weighted(self):
G=nx.path_graph(4)
G[1][2]['weight']=4
answer={1:2.0,2:1.8}
nd = nx.average_degree_connectivity(G,weight='weight')
assert_equal(nd,answer)
answer={1:2.0,2:1.5}
nd = nx.average_degree_connectivity(G)
assert_equal(nd,answer)
D=G.to_directed()
answer={2:2.0,4:1.8}
nd = nx.average_degree_connectivity(D,weight='weight')
assert_equal(nd,answer)
answer={1:2.0,2:1.8}
D=G.to_directed()
nd = nx.average_degree_connectivity(D,weight='weight', source='in',
target='in')
assert_equal(nd,answer)
D=G.to_directed()
nd = nx.average_degree_connectivity(D,source='in',target='out',
weight='weight')
assert_equal(nd,answer)
def test_weight_keyword(self):
G=nx.path_graph(4)
G[1][2]['other']=4
answer={1:2.0,2:1.8}
nd = nx.average_degree_connectivity(G,weight='other')
assert_equal(nd,answer)
answer={1:2.0,2:1.5}
nd = nx.average_degree_connectivity(G,weight=None)
assert_equal(nd,answer)
D=G.to_directed()
answer={2:2.0,4:1.8}
nd = nx.average_degree_connectivity(D,weight='other')
assert_equal(nd,answer)
answer={1:2.0,2:1.8}
D=G.to_directed()
nd = nx.average_degree_connectivity(D,weight='other', source='in',
target='in')
assert_equal(nd,answer)
D=G.to_directed()
nd = nx.average_degree_connectivity(D,weight='other',source='in',
target='in')
assert_equal(nd,answer)
def test_in_out_weight(self):
G = nx.DiGraph()
G.add_edge(1, 2, weight=1)
G.add_edge(1, 3, weight=1)
G.add_edge(3, 1, weight=1)
for s, t in permutations(["in", "out", "in+out"], 2):
c = nx.average_degree_connectivity(G, source=s, target=t)
cw = nx.average_degree_connectivity(G, source=s, target=t, weight="weight")
assert c == cw
def test_degree_barrat(self):
G = nx.star_graph(5)
G.add_edges_from([(5, 6), (5, 7), (5, 8), (5, 9)])
G[0][5]["weight"] = 5
nd = nx.average_degree_connectivity(G)[5]
assert nd == 1.8
nd = nx.average_degree_connectivity(G, weight="weight")[5]
assert nd == pytest.approx(3.222222, abs=1e-5)
nd = nx.k_nearest_neighbors(G, weight="weight")[5]
assert nd == pytest.approx(3.222222, abs=1e-5)
def test_degree_barrat(self):
G=nx.star_graph(5)
G.add_edges_from([(5,6),(5,7),(5,8),(5,9)])
G[0][5]['weight']=5
nd = nx.average_degree_connectivity(G)[5]
assert_equal(nd,1.8)
nd = nx.average_degree_connectivity(G,weight='weight')[5]
assert_almost_equal(nd,3.222222,places=5)
nd = nx.k_nearest_neighbors(G,weight='weight')[5]
assert_almost_equal(nd,3.222222,places=5)
def test_in_out_weight(self):
G = nx.DiGraph()
G.add_edge(1, 2, weight=1)
G.add_edge(1, 3, weight=1)
G.add_edge(3, 1, weight=1)
for s, t in permutations(['in', 'out', 'in+out'], 2):
c = nx.average_degree_connectivity(G, source=s, target=t)
cw = nx.average_degree_connectivity(G, source=s, target=t,
weight='weight')
assert_equal(c, cw)
def test_in_out_weight(self):
G = nx.DiGraph()
G.add_edge(1, 2, weight=1)
G.add_edge(1, 3, weight=1)
G.add_edge(3, 1, weight=1)
for s, t in permutations(['in', 'out', 'in+out'], 2):
c = nx.average_degree_connectivity(G, source=s, target=t)
cw = nx.average_degree_connectivity(G, source=s, target=t,
weight='weight')
assert_equal(c, cw)
def test_degree_barrat(self):
G=nx.star_graph(5)
G.add_edges_from([(5,6),(5,7),(5,8),(5,9)])
G[0][5]['weight']=5
nd = nx.average_degree_connectivity(G)[5]
assert_equal(nd,1.8)
nd = nx.average_degree_connectivity(G,weight='weight')[5]
assert_almost_equal(nd,3.222222,places=5)
nd = nx.k_nearest_neighbors(G,weight='weight')[5]
assert_almost_equal(nd,3.222222,places=5)
def all_or_top(graph, n=0):
if n == 0:
description_for_every_node_in_graph(graph)
print("------------------------------------")
d = nx.average_degree_connectivity(graph)
od = collections.OrderedDict(sorted(d.items()))
for key, value in od.items():
print("Degree: " + str(key) + "\t Average degree connectivity: " + str(value))
print("------------------------------------")
else:
bs.print_top_n_by_metric(nx.average_neighbor_degree(graph), "Average neighbor degree", n)
bs.print_top_n_by_metric(nx.average_degree_connectivity(graph), "Average degree Connectivity", n)
def gen_graph_features(G):
#GRAPH STATISTICS
node_connectivity_deg = nx.average_degree_connectivity(G)
graph_connectivity_deg = np.mean(list(node_connectivity_deg))
graph_density = nx.density(G)
graph_node_count = len(G.node)
graph_edge_count = G.number_of_edges()
print("\n\t graph density = " + str(graph_density),
"\n\t graph connectivity degree = " + str(graph_connectivity_deg),
"\n\t node count = " + str(graph_node_count),
"\n\t edge count = " + str(graph_edge_count))
#COMPONENT STATISTICS
#find size (number of nodes) of greatest connected component
Gc = max(nx.connected_component_subgraphs(G), key=len)
max_cc_size = len(Gc)
percent_GC = max_cc_size/graph_node_count
#create a list of connected components sorted by size
cc_list = sorted(nx.connected_component_subgraphs(G), key=len, reverse = True)
#find the major connected components of the map
major_cc_list = []
LCCs_node_count = 0
for cc in cc_list:
if len(cc) >= max_cc_size/2:
major_cc_list.append(cc)
LCCs_node_count += len(cc)
percent_LCCs = LCCs_node_count/graph_node_count
#large connected components (LCCs) statistics
counter = 0
major_cc = Gc
counter += 1
#compute major component stats
node_connectivity_deg = nx.average_degree_connectivity(major_cc)
major_cc_connectivity_deg = np.mean(list(node_connectivity_deg))
major_cc_density = nx.density(major_cc)
LCC_node_count = len(major_cc)
LCC_edge_count = major_cc.number_of_edges()
avg_node_deg = LCC_edge_count/LCC_node_count
normalized_size = LCC_node_count/max_cc_size
return [len(G.node),nx.density(major_cc),percent_GC,major_cc_connectivity_deg]
def knn_pack(graph, *kwargs):
t = dict()
for k in kwargs:
t.__setitem__(k, kwargs[k])
t.__setitem__('asr', nx.degree_assortativity_coefficient(graph))
t.__setitem__('weighted_asr', nx.degree_assortativity_coefficient(graph, weight = 'weight'))
if graph.is_directed():
t.__setitem__('knn', nx.average_degree_connectivity(graph, source = 'out', target = 'in'))
if len(nx.get_edge_attributes(graph, 'weight')):
t.__setitem__('weighted_knn', nx.average_degree_connectivity(graph, source = 'out', target = 'in', weight = 'weight'))
else:
t.__setitem__('knn', nx.average_degree_connectivity(graph))
if len(nx.get_edge_attributes(graph, 'weight')):
t.__setitem__('weighted_knn', nx.average_degree_connectivity(graph, weight = 'weight'))
return(t)
def summary_statistics(self) -> pd.DataFrame:
"""function to return the summary statistics
:return pandas dataframe"""
if self.ppi_file is not None: #ppi file provided
#self.read_ppis()
#self.create_node_dict()
self.compile_files("new_node_list.tsv",
"new_edge_list.tsv") #building new files
self.import_graph("new_edge_list.tsv")
else: #edge and node file provided
self.import_graph(self.edge_list)
network_dict = {
'number_of_nodes':
nx.number_of_nodes(self.graph),
'number_of_edges':
nx.number_of_edges(self.graph),
'density':
nx.density(self.graph),
'average_degree_connectivity':
np.mean(
np.array(
list(nx.average_degree_connectivity(self.graph).values())))
}
network_df = pd.DataFrame(network_dict.values(),
index=network_dict.keys(),
columns=['Values'])
return network_df
def test_single_node(self):
# TODO Is this really the intended behavior for providing a
# single node as the argument `nodes`? Shouldn't the function
# just return the connectivity value itself?
G = nx.trivial_graph()
conn = nx.average_degree_connectivity(G, nodes=0)
assert conn == {0: 0}
def analyse_graph(G):
print(nx.info(G))
n_components = nx.number_connected_components(G)
print("Number of connected components:", n_components)
if n_components > 1:
component_sizes = [
len(c)
for c in sorted(nx.connected_components(G), key=len, reverse=True)
]
print("Connected component sizes:", component_sizes)
lcc_percent = 100 * component_sizes[0] / G.number_of_nodes()
print(f"LCC: {lcc_percent}%")
avg_c = nx.average_clustering(G)
print("Average clustering coefficient:", avg_c)
degree_assortativity = nx.degree_pearson_correlation_coefficient(G)
print("Degree assortativity:", degree_assortativity)
if nx.is_connected(G):
avg_d = nx.average_shortest_path_length(G)
print("Average distance:", avg_d)
else:
avg_distances = [
nx.average_shortest_path_length(C)
for C in (G.subgraph(c).copy() for c in nx.connected_components(G))
]
print("Average distances:", avg_distances)
avg_connectivity = nx.average_degree_connectivity(G)
print("Average degree connectivity:", avg_connectivity)
def draw_graph(nodes, edges, graphs_dir, default_lang='all'):
lang_graph = nx.MultiDiGraph()
lang_graph.add_nodes_from(nodes)
for edge in edges:
if edges[edge] == 0:
lang_graph.add_edge(edge[0], edge[1])
else:
lang_graph.add_edge(edge[0], edge[1], weight=float(edges[edge]), label=str(edges[edge]))
# print graph info in stdout
# degree centrality
print('-----------------\n\n')
print(default_lang)
print(nx.info(lang_graph))
try:
# When ties are associated to some positive aspects such as friendship or collaboration,
# indegree is often interpreted as a form of popularity, and outdegree as gregariousness.
DC = nx.degree_centrality(lang_graph)
max_dc = max(DC.values())
max_dc_list = [item for item in DC.items() if item[1] == max_dc]
except ZeroDivisionError:
max_dc_list = []
# https://ru.wikipedia.org/wiki/%D0%9A%D0%BE%D0%BC%D0%BF%D0%BB%D0%B5%D0%BA%D1%81%D0%BD%D1%8B%D0%B5_%D1%81%D0%B5%D1%82%D0%B8
print('maxdc', str(max_dc_list), sep=': ')
# assortativity coef
AC = nx.degree_assortativity_coefficient(lang_graph)
print('AC', str(AC), sep=': ')
# connectivity
print("Слабо-связный граф: ", nx.is_weakly_connected(lang_graph))
print("количество слабосвязанных компонент: ", nx.number_weakly_connected_components(lang_graph))
print("Сильно-связный граф: ", nx.is_strongly_connected(lang_graph))
print("количество сильносвязанных компонент: ", nx.number_strongly_connected_components(lang_graph))
print("рекурсивные? компоненты: ", nx.number_attracting_components(lang_graph))
print("число вершинной связности: ", nx.node_connectivity(lang_graph))
print("число рёберной связности: ", nx.edge_connectivity(lang_graph))
# other info
print("average degree connectivity: ", nx.average_degree_connectivity(lang_graph))
print("average neighbor degree: ", sorted(nx.average_neighbor_degree(lang_graph).items(),
key=itemgetter(1), reverse=True))
# best for small graphs, and our graphs are pretty small
print("pagerank: ", sorted(nx.pagerank_numpy(lang_graph).items(), key=itemgetter(1), reverse=True))
plt.figure(figsize=(16.0, 9.0), dpi=80)
plt.axis('off')
pos = graphviz_layout(lang_graph)
nx.draw_networkx_edges(lang_graph, pos, alpha=0.5, arrows=True)
nx.draw_networkx(lang_graph, pos, node_size=1000, font_size=12, with_labels=True, node_color='green')
nx.draw_networkx_edge_labels(lang_graph, pos, edges)
# saving file to draw it with dot-graphviz
# changing overall graph view, default is top-bottom
lang_graph.graph['graph'] = {'rankdir': 'LR'}
# marking with blue nodes with maximum degree centrality
for max_dc_node in max_dc_list:
lang_graph.node[max_dc_node[0]]['fontcolor'] = 'blue'
write_dot(lang_graph, os.path.join(graphs_dir, default_lang + '_links.dot'))
# plt.show()
plt.savefig(os.path.join(graphs_dir, 'python_' + default_lang + '_graph.png'), dpi=100)
plt.close()
def summary_statistics(self) -> None:
"""Generates summary statistics for a network."""
molecule_counter = {PROTEIN: 0, RNA: 0, DNA: 0}
for metadata in self.nodes.values():
molecule_counter[metadata[MOLECULE]] += 1
sum_stats = {
'Nodes':
sum(molecule_counter.values()),
'Protein Nodes':
molecule_counter[PROTEIN],
'RNA Nodes':
molecule_counter[RNA],
'DNA Nodes':
molecule_counter[DNA],
'Graph Density':
nx.density(self.graph),
'Average Degree Connectivity':
str(nx.average_degree_connectivity(self.graph))
}
sum_stats.update(
Counter([
attr['rel_type'] for _, _, attr in self.graph.edges(data=True)
]))
self.sum_stats = pd.DataFrame({
'Stat': list(sum_stats.keys()),
'Value': list(sum_stats.values())
})
def computeKnn(graph, knn_file, weight=None):
G = nx.path_graph(4)
G.edge[1][2]['weight'] = 3
print nx.k_nearest_neighbors(G)
knnfs = codecs.open(knn_file, 'w+', encoding='utf-8')
knn = nx.average_degree_connectivity(graph)
print graph, 'knn as follows:'
print knn
sumknn = sum(knn.values())
minknn = min(knn.keys())
maxknn = max(knn.keys())
index = maxknn
currentSum = 0.0
while index >= minknn:
if index in knn.keys():
currentSum = knn[index]
else:
index -= 1
continue
freq = currentSum * 1.0 / sumknn
knnfs.write(str(index) + ',' + str(freq) + '\r\n')
print index, freq
index -= 1
#for (key, value) in knn.items():
# knnfs.write(str(key)+ ',' + str(value) + '\r\n')
knnfs.flush()
knnfs.close()
def avg_degree(G):
'''
Compute the average degree connectivity of graph.
:param G: a networkx graph
:return: avg degree
'''
return nx.average_degree_connectivity(G)
def test_single_node(self):
# TODO Is this really the intended behavior for providing a
# single node as the argument `nodes`? Shouldn't the function
# just return the connectivity value itself?
G = nx.trivial_graph()
conn = nx.average_degree_connectivity(G, nodes=0)
assert_equal(conn, {0: 0})
def analyze_net(G):
print("Analysis:")
avg_k = nx.average_degree_connectivity(G)
print("Average degree:", avg_k)
avg_c = nx.average_clustering(G)
print("Average clustering coefficient:", avg_c)
avg_d = nx.average_shortest_path_length(G)
print("Average distance:", avg_d)
def getresults(graph):
print("Analisando grafo...")
print("Média do grau dos nodos: " + str(nx.average_degree_connectivity(graph)))
print("Coeficiente de clusterização: " + str(nx.average_clustering(graph)))
for g in nx.connected_component_subgraphs(graph):
print("Distância média dos nós: " + str(nx.average_shortest_path_length(g)))
print("Betweenness das arestas: " + str(nx.edge_betweenness(graph)))
print("Betweenness dos nodos: " + str(nx.betweenness_centrality(graph)))
def plotDegreeCorrelationFunction(G):
degreeCorrelation = nx.average_degree_connectivity(G)
dc_sorted = dict(sorted(degreeCorrelation.items()))
plt.title('Degree Correlation Function')
plt.ylabel('Knn(K)')
plt.xlabel('K')
plt.loglog(list(dc_sorted.keys()), list(dc_sorted.values()), '.')
plt.show()
def descriptives(G,grouping = None):
degree = nx.degree_histogram(G)
plt.bar(x = range(len(degree)), height = degree)
plt.savefig('images/degree_hist.png')
plt.close()
neighbor_degree = nx.average_neighbor_degree(G)
dict_to_hist(neighbor_degree,'neighbor_degree')
degree_conn = nx.average_degree_connectivity(G)
dict_to_hist(degree_conn,'degree_conn')
def grow_graphs_using_krongen(graph, gn, recurrence_nbr=1, graph_vis_bool=False, nbr_runs = 1):
"""
grow graph using krongen given orig graph, gname, and # of recurrences
Returns
-------
nth graph --<kpgm>--
"""
import math
from pami import kronfit
from os import environ
import subprocess
tsvGraphName = "/tmp/{}kpgraph.tsv".format(gn)
# tmpGraphName = "/tmp/{}kpgraph.tmp".format(gn)
# if environ['HOME'] == '/home/saguinag':
# args = ("time/bin/linux/krongen", "-i:{}".format(tsvGraphName),"-n0:2", "-m:\"0.9 0.6; 0.6 0.1\"", "-gi:5")
# elif environ['HOME'] == '/Users/saguinag':
# args = ("time/bin/mac/krongen", "-i:{}".format(tsvGraphName),"-n0:2", "-m:\"0.9 0.6; 0.6 0.1\"", "-gi:5")
# else:
# args = ('./kronfit.exe -i:tmp.txt -n0:2 -m:"0.9 0.6; 0.6 0.1" -gi:5')
kp_graphs = []
k = int(math.log(graph.number_of_nodes(),2))+1 # Nbr of Iterations
if 0: print 'k:',k,'n',graph.number_of_nodes()
print " --- Model inference, kronfit learn a Kronecker seed matrix"
P = kronfit(graph) #[[0.9999,0.661],[0.661, 0.01491]]
M = '-m:"{} {}; {} {}"'.format(P[0][0], P[0][1], P[1][0], P[1][1])
if environ['HOME'] == '/home/saguinag':
args = ("time/bin/linux/krongen", "-o:"+tsvGraphName, M, "-i:{}".format(k))
elif environ['HOME'] == '/Users/saguinag':
print tsvGraphName
args = ("bin/macos/krongen", "-o:"+tsvGraphName, M, "-i:{}".format(k))
else:
args = ('./krongen.exe -o:{} '.format(tsvGraphName) +M +'-i:{}'.format(k+1))
for i in range(nbr_runs):
popen = subprocess.Popen(args, stdout=subprocess.PIPE)
popen.wait()
#output = popen.stdout.read()
if os.path.exists(tsvGraphName):
KPG = nx.read_edgelist(tsvGraphName, nodetype=int)
else:
print "!! Error, file is missing"
for u,v in KPG.selfloop_edges():
KPG.remove_edge(u, v)
kp_graphs.append( KPG )
if DBG:
print 'Avg Deg:', nx.average_degree_connectivity(graph)
import phoenix.visadjmatrix as vis
# vis.draw_sns_adjacency_matrix(graph)
vis.draw_sns_graph(graph)
return kp_graphs # returns a list of kp graphs
def runAnalytics(G):
# GRAPH STATISTICS
node_connectivity_deg = nx.average_degree_connectivity(G)
graph_connectivity_deg = np.mean(list(node_connectivity_deg))
graph_density = nx.density(G)
graph_node_count = len(G.node)
graph_edge_count = G.number_of_edges()
print('''
density: {}
connectivity degree: {}
node count: {}
edge count: {}
---------------------------
'''.format(graph_density, graph_connectivity_deg, graph_node_count,
graph_edge_count))
# #COMPONENT STATISTICS
# #find size (number of nodes) of greatest connected component
# Gc = max(nx.connected_component_subgraphs(G), key=len)
# max_cc_size = len(Gc)
# percent_GC = max_cc_size/graph_node_count
# #create a list of connected components sorted by size
# cc_list = sorted(nx.connected_component_subgraphs(G), key=len, reverse = True)
# #find the major connected components of the map
# major_cc_list = []
# LCCs_node_count = 0
# for cc in cc_list:
# if len(cc) >= max_cc_size/2:
# major_cc_list.append(cc)
# LCCs_node_count += len(cc)
# percent_LCCs = LCCs_node_count/graph_node_count
# #large connected components (LCCs) statistics
# counter = 0
# major_cc = Gc
# counter += 1
# #compute major component stats
# node_connectivity_deg = nx.average_degree_connectivity(major_cc)
# major_cc_connectivity_deg = np.mean(list(node_connectivity_deg))
# major_cc_density = nx.density(major_cc)
# LCC_node_count = len(major_cc)
# LCC_edge_count = major_cc.number_of_edges()
# avg_node_deg = LCC_edge_count/LCC_node_count
# normalized_size = LCC_node_count/max_cc_size
# print([len(G.node),nx.density(major_cc),percent_GC,major_cc_connectivity_deg])
# plt.figure()
# nx.draw(G, node_size=5)
return G
def test_degree_p4(self):
G=nx.path_graph(4)
answer={1:2.0,2:1.5}
nd = nx.average_degree_connectivity(G)
assert_equal(nd,answer)
D=G.to_directed()
answer={2:2.0,4:1.5}
nd = nx.average_degree_connectivity(D)
assert_equal(nd,answer)
answer={1:2.0,2:1.5}
D=G.to_directed()
nd = nx.average_degree_connectivity(D, source='in', target='in')
assert_equal(nd,answer)
D=G.to_directed()
nd = nx.average_degree_connectivity(D, source='in', target='in')
assert_equal(nd,answer)
def test_degree_p4(self):
G=nx.path_graph(4)
answer={1:2.0,2:1.5}
nd = nx.average_degree_connectivity(G)
assert_equal(nd,answer)
D=G.to_directed()
answer={2:2.0,4:1.5}
nd = nx.average_degree_connectivity(D)
assert_equal(nd,answer)
answer={1:2.0,2:1.5}
D=G.to_directed()
nd = nx.average_in_degree_connectivity(D)
assert_equal(nd,answer)
D=G.to_directed()
nd = nx.average_out_degree_connectivity(D)
assert_equal(nd,answer)
def average_degree_connectivityPlots(G):
avg_degree_conn = nx.average_degree_connectivity(G)
hist = [i for i in avg_degree_conn.keys()]
values = [i for i in avg_degree_conn.values()]
plt.figure()
plt.xlabel('degree')
plt.ylabel('average connectivity')
plt.plot(hist,values,'ro')
plt.savefig('avg_deg_connectivity')
plt.close()
def assortativity_distribution(graph: nx.Graph) -> None:
assorts = sorted(nx.average_degree_connectivity(graph).items())
assort_x, assort_y = log_binning(dict(assorts), 40)
plt.figure()
plt.scatter(assort_x, assort_y, c='r', marker='s', s=25, label='')
plt.title('Assortativity')
plt.xlabel('k')
plt.ylabel('$<k_{nn}>$')
plt.savefig(os.path.join(common.FIGURES_FOLDER, 'assortativity.png'))
def test_degree_p4(self):
G = nx.path_graph(4)
answer = {1: 2.0, 2: 1.5}
nd = nx.average_degree_connectivity(G)
assert nd == answer
D = G.to_directed()
answer = {2: 2.0, 4: 1.5}
nd = nx.average_degree_connectivity(D)
assert nd == answer
answer = {1: 2.0, 2: 1.5}
D = G.to_directed()
nd = nx.average_degree_connectivity(D, source="in", target="in")
assert nd == answer
D = G.to_directed()
nd = nx.average_degree_connectivity(D, source="in", target="in")
assert nd == answer
def test_degree_p4_weighted(self):
G = nx.path_graph(4)
G[1][2]['weight'] = 4
answer = {1: 2.0, 2: 1.8}
nd = nx.average_degree_connectivity(G, weighted=True)
assert_equal(nd, answer)
D = G.to_directed()
answer = {2: 2.0, 4: 1.8}
nd = nx.average_degree_connectivity(D, weighted=True)
assert_equal(nd, answer)
answer = {1: 2.0, 2: 1.8}
D = G.to_directed()
nd = nx.average_in_degree_connectivity(D, weighted=True)
assert_equal(nd, answer)
D = G.to_directed()
nd = nx.average_out_degree_connectivity(D, weighted=True)
assert_equal(nd, answer)
def test_degree_p4_weighted(self):
G=nx.path_graph(4)
G[1][2]['weight']=4
answer={1:2.0,2:1.8}
nd = nx.average_degree_connectivity(G,weighted=True)
assert_equal(nd,answer)
D=G.to_directed()
answer={2:2.0,4:1.8}
nd = nx.average_degree_connectivity(D,weighted=True)
assert_equal(nd,answer)
answer={1:2.0,2:1.8}
D=G.to_directed()
nd = nx.average_in_degree_connectivity(D,weighted=True)
assert_equal(nd,answer)
D=G.to_directed()
nd = nx.average_out_degree_connectivity(D,weighted=True)
assert_equal(nd,answer)
def graphInfo(graph, weighted=False):
print("Number of Vertices = ", graph.number_of_nodes())
print("Number of Edges = ", graph.number_of_edges())
print("Number of Connected Components = ",
nx.number_connected_components(graph))
if weighted == False:
print("Size of Unweighted Graph = ", graph.size(weight=None))
else:
print("Size of Weighted Graph = ", graph.size(weight="weight"))
averageWeightedDegree = nx.average_degree_connectivity(graph,
weight="weight")
print("Average Weighted Degree = ", averageWeightedDegree)
def compute_metrics(G):
metrics = {}
metrics['e_n_r'] = nx.number_of_edges(G) / nx.number_of_nodes(
G) #Edge to node ratio
metrics['av_clu'] = nx.average_clustering(G) #avg clustering
metrics['av_mdc'] = get_weighted_avg(nx.average_degree_connectivity(G))
metrics['av_deg'] = get_avg_degree(G) #avg_degree
metrics['tran'] = nx.transitivity(G) # transitivity
metrics['den'] = nx.density(G)
metrics['c_cen'] = get_avg(nx.closeness_centrality(G))
metrics['b_cen'] = get_avg(nx.betweenness_centrality(G))
return metrics
def computeKnn(graph, knn_file, weight=None):
G = nx.path_graph(4)
G.edge[1][2]['weight'] = 3
print nx.k_nearest_neighbors(G)
knnfs = codecs.open(knn_file, 'w+', encoding='utf-8')
knn = nx.average_degree_connectivity(graph)
print graph, 'knn as follows:'
print knn
for (key, value) in knn.items():
knnfs.write(str(key) + ',' + str(value) + '\r\n')
knnfs.flush()
knnfs.close()
def run(folder, infilename, title):
ingraph = read_graphml(folder + os.sep + infilename)
try:
#avg_cluster_coeff = nx.average_clustering(ingraph)
#print('average clustering for ' + title + " = " + str(avg_cluster_coeff))
avg_deg_coeff = nx.average_degree_connectivity(ingraph)
print('average degree for ' + title + ' = ' + str(avg_deg_coeff))
except Exception as e:
print (e.__str__())
'''
def randomGraph(G):
n = G.number_of_nodes()
d = nx.average_degree_connectivity(G)
c = sum(i[1] for i in d.items())
p = c / (n - 1)
try:
RG = nx.fast_gnp_random_graph(n, p)
plot(RG)
print 'Global Clustering: {0}\t'.format(str(p)),
l = math.log(RG.number_of_nodes()) / math.log(c)
print 'Average path length : {0}\n'.format(str(l))
except:
'Failed attempt to get connected random graph..Try again!!!'
def randomGraph(G):
n = G.number_of_nodes()
d= nx.average_degree_connectivity(G)
c = sum(i[1] for i in d.items())
p = c/(n-1)
try:
RG = nx.fast_gnp_random_graph(n,p)
plot(RG)
print 'Global Clustering: {0}\t'.format(str(p)),
l = math.log(RG.number_of_nodes())/math.log(c)
print 'Average path length : {0}\n'.format(str(l))
except:
'Failed attempt to get connected random graph..Try again!!!'
def smallworld(G):
n = G.number_of_nodes()
d= nx.average_degree_connectivity(G)
c = sum(i[1] for i in d.items())
c0 = 0.75*(c-2)/(c-1)
beta = random.uniform(0.01,0.1)
try:
SG= nx.connected_watts_strogatz_graph(n,int(c),beta)
plot(SG)
c = ((1-beta)**3)*c0
print 'Global Clustering: {0}\t'.format(str(c)),
print 'Average path length : {0}\n'.format(str(nx.average_shortest_path_length(SG)))
except:
'Failed attempt to get connected small world graph..Try again!!!'
def loi_puissance(self):
distri = nx.degree_histogram(self.graphe)
k_obs=nx.average_degree_connectivity(self.graphe)
tab_deg=self.graphe.degree() #calcul les desgrés de tous les noeuds
list_deg=[]
type(tab_deg)
for key,value in tab_deg.iteritems():
temp = [key,value]
list_deg.append(temp[1])
########################""
f=0
for k in range(1,len(distri)):
f+=(distri[k]-(self.nb* (k**(-GAMMA))))**2
########################
pk=[x/self.nb for x in distri]
gradient, intercept, r_value, p_value, std_err = stats.linregress(distri,pk)
#stats.kstest(k_theo,k_obs)
return f
def test_zero_deg(self):
G=nx.DiGraph()
G.add_edge(1,2)
G.add_edge(1,3)
G.add_edge(1,4)
c = nx.average_degree_connectivity(G)
assert_equal(c,{1:0,3:1})
c = nx.average_degree_connectivity(G, source='in', target='in')
assert_equal(c,{0:0,1:0})
c = nx.average_degree_connectivity(G, source='in', target='out')
assert_equal(c,{0:0,1:3})
c = nx.average_degree_connectivity(G, source='in', target='in+out')
assert_equal(c,{0:0,1:3})
c = nx.average_degree_connectivity(G, source='out', target='out')
assert_equal(c,{0:0,3:0})
c = nx.average_degree_connectivity(G, source='out', target='in')
assert_equal(c,{0:0,3:1})
c = nx.average_degree_connectivity(G, source='out', target='in+out')
assert_equal(c,{0:0,3:1})
def parametres(G,nG):
print "\nParametres disponibles:\n"
###################################
print "\tParametres Globals: \n"
print "\t[1] Betweenness Centralization\n"
print "\t[2] Average path length\n"
print "\t[3] Assortativity degree\n"
print "\t[4] Diameter\n"
print "\t[5] Density\n"
print "\t[6] Cohesion\n"
print "\t[7] Radius\n"
print "\tParametres per cada node de la xarxa: \n"
print "\t[11] Betweenness\n"
print "\t[12] Pagerank\n"
print "\t[13] EigenVectorCentrality\n"
print "\t[14] Average degree connectivity\n"
print "\t[15] Periphery\n"
print "\t[16] Eccentricity\n"
print "\t[17] Center nodes\n"
se = raw_input('Escriu els numeros dels parametres que vulguis amb una coma entre mig:\n')
secom = se.split(',')
for i in range(len(secom)):
if secom[i] == "1":
print "\nBetweenness Centralization:"
print betweenness_centralization(G)
elif secom[i] == "2":
print "\nAverage path length:"
print G.average_path_length()
elif secom[i] == "3":
print "\nAssortativity degree:"
print G.assortativity_degree()
elif secom[i] == "4":
print "\nDiameter:"
print G.diameter()
elif secom[i] == "5":
print "\nDensity:"
print G.density()
elif secom[i] == "6":
print "\nCohesion:"
print G.cohesion()
elif secom[i] == "7":
print("\nRadius:")
print nx.radius(nG)
###################################
elif secom[i] == "11":
print "\nBetweenness:"
print G.betweenness(directed=False, cutoff=16)
elif secom[i] == "12":
print "\nPagerank:"
print G.pagerank()
elif secom[i] == "13":
print "\nEigenVectorCentrality:"
print G.eigenvector_centrality()
elif secom[i] == "14":
print "\nAverage degree connectivity:"
print nx.average_degree_connectivity(nG)
elif secom[i] == "15":
print "\nPeriphery:"
print nx.periphery(nG)
elif secom[i] == "16":
print("\nEccentricity:")
print nx.eccentricity(nG)
elif secom[i] == "17":
print("\nCenter:")
print nx.center(nG)
#####Parametres exclosos#####
#print "Assortativity meu:" # es el mateix que el degree?
#print assortativitymeu(G)
#print("diameter: %d" % nx.diameter(nG))
#print("density: %s" % nx.density(nG))
#print("richclub coefficient: %s" % nx.rich_club_coefficient(nG.to_undirected()))
#print("richclub coefficient: %s" % nx.rich_club_coefficient(nG))
return 2
def average_degree_connectivity(self): return nx.average_degree_connectivity(self.g)
def a_degree_connectivity(G):
return np.average(nx.average_degree_connectivity(G).values())
def avg_degree_connectivity(self):
return nx.average_degree_connectivity(self._graph)
gr = nx.Graph()
#[i for i in itertools.combinations(de, 2) for de in df.topics[:100]]
gr.add_edges_from([i for de in df.topics.dropna()[0:200] for i in itertools.combinations(de,2)])
# <codecell>
gr2 = nx.Graph()
[gr2.add_edge(f[0],t[0]) for f,t in zip(ftdf.fields, ftdf.topics) if f is not NaN and t is not NaN]
gr2.size()
# <codecell>
print('topics network has %s edges and %s nodes'%(gr.number_of_edges(), gr.number_of_nodes()))
nx.average_degree_connectivity(gr)
#nx.draw_networkx(gr)
# <codecell>
gr.remove_node('machine learning')
# <codecell>
deg=nx.degree(gr)
# <codecell>
deg['support vector machine']
deg_sorted = sorted(deg.iteritems(), key=lambda(k,v):(-v,k))
deg_sorted[:50]
def stats(self):
avg_deg_con = nx.average_degree_connectivity(self.graph)
print sorted(avg_deg_con)
def getAverageDegreeOfNeighbours(self):
averageNeighbourDegrees = nx.average_degree_connectivity(self.amazonGraph)
return averageNeighbourDegrees
def parametres(G,nG):
print "\nParametres disponibles:\n"
###################################
print "\tParametres Globals: \n"
print "\t[1] Betweenness Centralization\n" #Aguanta xarxes inconectes
# Es la mesura de betweenes centrality feta en tots els nodes de la xarxa.
#The network betweenness centralization score is calculated based on the betweenness centrality for each individual in the network.
#Betweenness centrality examines the number of times one person lies on the shortest path between two others
#A highly centralized network has a great degree of inequality between individual centrality scores, while an uncentralized network has no inequality between individual centrality scores.
print "\t[2] Average path length\n"
#La longitud de cami mitjana entre dos punts
print "\t[3] Assortativity degree\n"
#positive values of r indicate a correlation between nodes of similar degree,
#while negative values indicate relationships between nodes of different degree.
#This coefficient is basically the correlation between the actual connectivity patterns of the vertices and the pattern expected from the
#disribution of the vertex types.
print "\t[4] Diameter\n"
#Diametre de la xarxa
print "\t[5] Density\n"
#El numero de enllacos respecte al total que hi pot existir a la xarxa
#Si el numero s'apropa a 1 sera una xarxa densa, si es un numero molt petit sera una xarxa escasa o sparse graph.
print "\t[6] Cohesion\n"
#Calcula el numero necesari de vertex per tal de divir una xarxa en dos components.
#The vertex connectivity between two given vertices is the number of vertices
#that have to be removed in order to disconnect the two vertices into two
#separate components. This is also the number of vertex disjoint directed
#paths between the vertices (apart from the source and target vertices of
#course). The vertex connectivity of the graph is the minimal vertex
#connectivity over all vertex pairs.
#
#This method calculates the vertex connectivity of a given vertex pair if both
#the source and target vertices are given. If none of them is given (or they
#are both negative), the overall vertex connectivity is returned.
print "\t[7] Radius\n" #No funciona amb xarxes inconectes
#The radius of a graph is defined as the minimum eccentricity of its vertices
#The eccentricity of a vertex is calculated by measuring the shortest distance from (or to) the vertex, to (or from) all other
#vertices in the graph, and taking the maximum.
print "\tParametres per cada node de la xarxa: \n"
print "\t[11] Betweenness\n"
#Mostra per cada node el seu valor de betweenness, gran = centric.
print "\t[12] Pagerank\n"
#Calcula algorisme de google, es una variant del eigenvector centratily, canviar posicio
print "\t[13] EigenVectorCentrality\n"
#It assigns relative scores to all nodes in the network based on the concept that connections to high-scoring nodes
#contribute more to the score of the node in question than equal connections to low-scoring nodes.
print "\t[14] Average degree connectivity\n"
#The average degree connectivity is the average nearest neighbor degree ofnodes with degree k.
print "\t[15] Periphery\n" #No funciona amb xarxes inconectes
#The periphery is the set of nodes with eccentricity equal to the diameter.
print "\t[16] Eccentricity\n"#No funciona amb xarxes inconectes
#The eccentricity of a node v is the maximum distance from v to all other nodes in G.
print "\t[17] Center nodes\n"#No funciona amb xarxes inconectes
#The center is the set of nodes with eccentricity equal to radius.
print "\t[18] Degree Distribution\n"
#Grau de distribucio del graf, distribucio de probabilitat de connexions en tota la xarxa
#the degree of a node in a network is the number of connections it has to other nodes and the degree distribution is
#the probability distribution of these degrees over the whole network.
print "\t[19] Count the number of Motif\n"
#Troba el numero de motifs, que son agrupacions de nodes, de mida tres o quatre (per defecte 3).
#It is argued that the motif profile (ie. the number of different motifs in the graph)
#is characteristic for different types of networks and network function is related to the motifs in the graph.
#
#S. Wernicke and F. Rasche: FANMOD: a tool for fast network motif detection, Bioinformatics 22(9), 1152--1153, 2006.
#
#Counts the total number of motifs in the graph
#Motifs are small subgraphs of a given structure in a graph.
#This function counts the total number of motifs in a graph without
#assigning isomorphism classes to them.
print "\t[20] Similarity Jaccard\n"
#The Jaccard similarity coefficient of two vertices is the number of their
#common neighbors divided by the number of vertices that are adjacent to
#at least one of them.
se = raw_input('Escriu els numeros dels parametres que vulguis amb una coma entre mig:\n')
secom = se.split(',')
for i in range(len(secom)):
if secom[i] == "1":
print "\nBetweenness Centralization:"
p1 = betweenness_centralization(G)
print p1
param.append("Betweenness Centralization")
param.append(p1)
elif secom[i] == "2":
print "\nAverage path length:"
p2 = G.average_path_length()
print p2
param.append("Average path length")
param.append(p2)
elif secom[i] == "3":
print "\nAssortativity degree:"
p3 = G.assortativity_degree()
print p3
param.append("Assortativity degree")
param.append(p3)
elif secom[i] == "4":
print "\nDiameter:"
p4 = G.diameter()
print p4
param.append("Diameter")
param.append(p4)
elif secom[i] == "5":
print "\nDensity:"
p5 = G.density()
print p5
param.append("Density")
param.append(p5)
elif secom[i] == "6":
print "\nCohesion:"
p6 = G.cohesion()
print p6
param.append("Cohesion")
param.append(p6)
elif secom[i] == "7":
print("\nRadius:")
p7 = nx.radius(nG)
print p7
param.append("Radius")
param.append(p7)
###################################
elif secom[i] == "11":
print "\nBetweenness:"
p11 = G.betweenness(directed=False, cutoff=16)
print p11
param.append("Betweenness")
param.append(p11)
elif secom[i] == "12":
print "\nPagerank:"
p12 = G.pagerank()
print p12
param.append("Pagerank")
param.append(p12)
elif secom[i] == "13":
print "\nEigenVectorCentrality:"
p13 = G.eigenvector_centrality()
print p13
param.append("EigenVectorCentrality")
param.append(p13)
elif secom[i] == "14":
print "\nAverage degree connectivity:"
p14 = nx.average_degree_connectivity(nG)
print p14
param.append("Average degree connectivity")
param.append(p14)
elif secom[i] == "15":
print "\nPeriphery:"
p15 = nx.periphery(nG)
print p15
param.append("Periphery")
param.append(p15)
elif secom[i] == "16":
print("\nEccentricity:")
p16 = nx.eccentricity(nG)
print p16
param.append("Eccentricity")
param.append(p16)
elif secom[i] == "17":
print("\nCenter:")
p17 = nx.center(nG)
print p17
param.append("Center")
param.append(p17)
elif secom[i] == "18":
print("\nDegree Distribution:")
p18 = G.degree_distribution()
print p18
param.append("Degree Distribution")
param.append(p18)
elif secom[i] == "19":
print("\nTotal number of Motif:")
p19 = G.motifs_randesu_no()
print p19
param.append("Total number of Motif")
param.append(p19)
elif secom[i] == "20":
print("\nSimilarity Jaccard:")
p20 = G.similarity_jaccard()
print p20
param.append("Similarity Jaccard")
param.append(p20)
else:
print "Paramatre incorrecte"
return
#####Parametres exclosos#####
#print "Assortativity meu:" # es el mateix que el degree?
#print assortativitymeu(G)
#print("diameter: %d" % nx.diameter(nG))
#print("density: %s" % nx.density(nG))
#print("richclub coefficient: %s" % nx.rich_club_coefficient(nG.to_undirected()))
#print("richclub coefficient: %s" % nx.rich_club_coefficient(nG))
return param
def test_invalid_source(self):
G = nx.DiGraph()
nx.average_degree_connectivity(G, source='bogus')
def get_average_degree_connectivity(self, g):
return nx.average_degree_connectivity(g)
def test_invalid_target(self):
G = nx.DiGraph()
nx.average_degree_connectivity(G, target='bogus')