def learn_parameters(G):
tr_list = [0, 5, 10]
h_list = [1, 10, 20, 50, 100]
d_list = [0, 5, 10]
p_list = [100, 10000, 10000000000000000]
repeat = 10
results = []
for tr in tr_list:
for h in h_list:
for d in d_list:
for p in p_list:
temp = []
for i in range(repeat):
cdr = cd.CommunityDetector(G, tr, h, d, p)
cdr.run()
temp.append(cdr.compute_accuracy(False))
results.append((sum(temp) / float(repeat), {
"tr": tr,
"h": h,
"d": d,
"p": p
}))
sorted_results = sorted(results)[-10:]
pprint.pprint(sorted_results)
def main():
G = gg.Graph()
set1 = G.create_nodes(200, {"clustId": 1}, 0)
set2 = G.create_nodes(400, {"clustId": 2}, 0)
set3 = G.create_nodes(50, {"clustId": 3}, 0)
set4 = G.create_nodes(50, {"clustId": 3}, 0)
for i in range(10):
G.create_random_edges(2500, [set1, set2, set3, set4],
[[0.29, 0.01, 0, 0], [0, 0.5, 0, 0],
[0, 0, 0.10, 0], [0, 0, 0, 0.10]], [
gg.UniformDistrib(),
gg.UniformDistrib(),
gg.UniformDistrib(),
gg.UniformDistrib()
], i)
for i in range(10, 15):
G.create_random_edges(
500, [set1, set2, set3, set4],
[[0.1, 0, 0.4, 0], [0, 0.1, 0, 0.4], [0, 0, 0, 0], [0, 0, 0, 0]], [
gg.UniformDistrib(),
gg.UniformDistrib(),
gg.UniformDistrib(),
gg.UniformDistrib()
], i)
for i in range(15, 25):
G.create_random_edges(
500, [set1, set2, set3, set4],
[[0.2, 0, 0, 0], [0, 0.2, 0, 0], [0, 0, 0, 0.6], [0, 0, 0, 0]], [
gg.UniformDistrib(),
gg.UniformDistrib(),
gg.UniformDistrib(),
gg.UniformDistrib()
], i)
# set5 = G.create_nodes(50, {"clustId":5}, 2)
# G.create_random_edges( 200,
# [set5],
# [[1]],
# [gg.UniformDistrib()],
# 2)
# G.create_random_edges( 200,
# [set5],
# [[1]],
# [gg.UniformDistrib()],
# 3)
# G.create_random_edges( 100,
# [set5],
# [[1]],
# [gg.UniformDistrib()],
# 4)
# learn_parameters(G)
# cdr = cd.CommunityDetector(G, sorted_results[-1][1]["tr"], sorted_results[-1][1]["h"], sorted_results[-1][1]["d"], sorted_results[-1][1]["p"])
cdr = cd.CommunityDetector(G, 10, 100, 5, 10000)
cdr.run(False, False)
# G.create_random_edges(200, [set1, set2], [[0,1],[0,0]], [gg.UniformDistrib(), gg.UniformDistrib()],4)
# G.create_random_edges(200, [set1, set2, set3, set4], [[0,0,0,0.5],[0,0,0.5,0],[0,0,0,0],[0,0,0,0]], [gg.UniformDistrib(), gg.UniformDistrib(), gg.UniformDistrib(), gg.UniformDistrib()],5)
# G.create_random_edges(400, [set1, set2], [[0.49,0.02],[0.,0.49]], [gg.UniformDistrib(), gg.UniformDistrib()],10)
# G.plot_sequence()
# G = gg.Graph(True)
# set1 = G.create_nodes(100)
# set2 = G.create_nodes(80)
# G.create_random_edges(500, [set1, set2], [[0.49,0.01],[0.01,0.49]], [gg.UniformDistrib(), gg.UniformDistrib()])
# G.plot_sequence()
# tree = st.SegmentTree()
# elements = [(0,5),(0,5),(0,10),(0,7),(5,10),(15,20),(12,st.SegmentTree.infinite)]
# for i in range(len(elements)):
# tree.insert(i, elements[i])
# # query = [0,3,5,7,11,12,20,98]
# # for q in query:
# # print "query=",q," - result=",tree.query(q)
# # pdb.set_trace()
# to_delete = range(len(elements))
# random.shuffle(to_delete)
# print to_delete
# for d in to_delete:
# tree.delete(d, elements[d])
# print "query=",elements[d][1]," - result=",tree.query(elements[d][1])
# pdb.set_trace()
def main(factor):
G = gg.Graph()
set1 = G.create_nodes(200, {"clustId": 1}, 0)
set2 = G.create_nodes(200, {"clustId": 2}, 0)
set3 = G.create_nodes(200, {"clustId": 2}, 0)
set4 = G.create_nodes(50, {"clustId": 4}, 0)
set5 = G.create_nodes(50, {"clustId": 5}, 0)
for i in range(4):
G.create_random_edges(
500 * factor, [set1, set2, set3, set4, set5],
[[0.303, 0.005, 0.005, 0.001, 0.001],
[0, 0.17, 0.24, 0.001, 0.001], [0, 0, 0.17, 0.001, 0.001],
[0, 0, 0, 0.05, 0.001], [0, 0, 0, 0, 0.05]], [
gg.UniformDistrib(),
gg.UniformDistrib(),
gg.UniformDistrib(),
gg.UniformDistrib(),
gg.UniformDistrib()
], i)
for i in range(4, 7):
G.create_random_edges(
40 * factor, [set1, set2, set3, set4, set5],
[[0, 0, 0, 0, 0], [0, 0, 0, 0, 0], [0, 0, 0, 0, 0],
[0, 0, 0, 0, 1], [0, 0, 0, 0, 0]], [
gg.UniformDistrib(),
gg.UniformDistrib(),
gg.UniformDistrib(),
gg.UniformDistrib(),
gg.UniformDistrib()
], i)
for i in range(7, 9):
G.delete_random_edges(
200 * factor, [set1, set2, set3, set4, set5],
[[0, 0, 0, 0, 0], [0, 0, 1, 0, 0], [0, 0, 0, 0, 0],
[0, 0, 0, 0, 0], [0, 0, 0, 0, 0]], [
gg.UniformDistrib(),
gg.UniformDistrib(),
gg.UniformDistrib(),
gg.UniformDistrib(),
gg.UniformDistrib()
], i)
G.delete_random_edges(0, [set1, set2, set3, set4, set5],
[[0, 0, 0, 0, 0], [0, 0, 1, 0, 0], [0, 0, 0, 0, 0],
[0, 0, 0, 0, 0], [0, 0, 0, 0, 0]], [
gg.UniformDistrib(),
gg.UniformDistrib(),
gg.UniformDistrib(),
gg.UniformDistrib(),
gg.UniformDistrib()
], 9)
# for i in range(9,14):
# G.delete_random_edges( 40*factor,
# [set1, set2, set3, set4, set5],
# [[0 ,0 ,0 ,0 ,0 ],
# [0 ,0 ,1 ,0 ,0 ],
# [0 ,0 ,0 ,0 ,0 ],
# [0 ,0 ,0 ,0 ,0 ],
# [0 ,0 ,0 ,0 ,0 ]],
# [gg.UniformDistrib(), gg.UniformDistrib(), gg.UniformDistrib(), gg.UniformDistrib(), gg.UniformDistrib()],
# i)
cdr = cd.CommunityDetector(G, 10, 100, 5, 10000)
cdr.run(True, True, False, 0, 4)
G.update_nodes(set5, {"clustId": 4})
cdr.run(True, True, False, 4, 7)
G.update_nodes(set3, {"clustId": 3})
cdr.run(True, True, False, 7, 12)