def exp2(name, step, max_partition):
gzip_fname = utils.get_gzip_fname(name)
s = network.Network.from_combined(name, gzip_fname)
graph = s.graph
graph_info = GraphInfo(graph)
p = Partitioner(graph)
exp2_impl(name, graph_info, p, max_partition, step)
def exp1(name, step):
exp1_info = {}
gzip_fname = utils.get_gzip_fname(name)
s = network.Network.from_combined(name, gzip_fname)
graph = s.graph
p = Partitioner(s.graph)
cd_s = time.time()
cd_partition = p.community_detection()
exp1_info['community_detection'] = duration(cd_s)
parts = len(cd_partition)
exp1_info['parts'] = parts
metis_s = time.time()
metis_partition = p.metis_partition(parts)
exp1_info['metis'] = duration(metis_s)
random_s = time.time()
rnd_partition = p.random_partition(parts)
exp1_info['rnd'] = duration(random_s)
graph_info = GraphInfo(graph)
partitions = [cd_partition, metis_partition, rnd_partition]
exp1_impl(name, graph_info, p, partitions, step)
return exp1_info
for l in gzip_data:
edges = [int(e) for e in l.split()]
assert len(edges) == 2
G.add_edge(*edges)
with open(pickle_file, "wb") as pickle_data:
pickle.dump(G, pickle_data)
return cls(G)
@classmethod
def from_ego(cls, name, ego_list):
if utils.graph_directness[name]:
G = nx.DiGraph()
else:
G = nx.Graph()
for ego_id in ego_list:
ego_graph = ego.Ego(name, ego_id)
ego_graph.graph_generator()
graph = ego_graph.graph
G.add_nodes_from(graph.nodes())
G.add_edges_from(graph.edges())
return cls(G)
if __name__ == "__main__":
log_helper.init_logger()
name = "facebook"
ego_list = utils.collect_ego_list(name)
gzip_fname = utils.get_gzip_fname(name)
social = Network.from_combined(name, gzip_fname)
g_info = graph_info.GraphInfo(social.graph)
total = 0
for node_id in note_set:
local_set_size = self.get_local_set(partition_list, node_id)
local += len(local_set_size)
total += len(self.graph.neighbors(node_id))
return float(local) / float(total)
@staticmethod
def dump_partition(partition_list):
for d in partition_list:
print(d)
if __name__ == "__main__":
name = "facebook"
gzip_fname = utils.get_gzip_fname("facebook")
s = network.Network.from_combined(name, gzip_fname)
p = Partitioner(s.graph)
cd_partition = p.community_detection()
parts = len(cd_partition)
ego_list = utils.collect_ego_list(name)
print(ego_list)
cd_locality = p.locality_percentage(cd_partition, ego_list)
print(cd_locality)
metis_partition = p.metis_partition(parts)
metis_locality = p.locality_percentage(metis_partition, ego_list)
print(metis_locality)
rnd_partition = p.random_partition(parts)
rnd_locality = p.locality_percentage(rnd_partition, ego_list)
print(rnd_locality)
