1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
|
from timeit import Timer
# This is gratefully modeled after the benchmarks found in
# the numpy svn repository. http://svn.scipy.org/svn/numpy/trunk
class Benchmark(object):
"""
Benchmark a method or simple bit of code using different Graph classes.
If the test code is the same for each graph class, then you can set it
during instantiation through the argument test_string.
The argument test_string can also be a tuple of test code and setup code.
The code is entered as a string valid for use with the timeit module.
Example:
>>> b=Benchmark(['Graph','XGraph'])
>>> b['Graph']=('G.add_nodes_from(nlist)','nlist=range(100)')
>>> b.run()
"""
def __init__(self,graph_classes,title='',test_string=None,runs=3,reps=1000):
self.runs = runs
self.reps = reps
self.title = title
self.class_tests = dict((gc,'') for gc in graph_classes)
# set up the test string if it is the same for all classes.
if test_string is not None:
if isinstance(test_string,tuple):
self['all']=test_string
else:
self['all']=(test_string,'')
def __setitem__(self,graph_class,some_strs):
"""
Set a simple bit of code and setup string for the test.
Use this for cases where the code differs from one class to another.
"""
test_str, setup_str = some_strs
if graph_class == 'all':
graph_class = self.class_tests.keys()
elif not isinstance(graph_class,list):
graph_class = [graph_class]
for GC in graph_class:
# setup_string='import networkx as NX\nG=NX.%s.%s()\n'%(GC.lower(),GC) \
# + setup_str
setup_string='import networkx as NX\nG=NX.%s()\n'%(GC,) \
+ setup_str
self.class_tests[GC] = Timer(test_str, setup_string)
def run(self):
"""Run the benchmark for each class and print results."""
column_len = max(len(G) for G in self.class_tests)
print('='*72)
if self.title:
print("%s: %s runs, %s reps"% (self.title,self.runs,self.reps))
print('='*72)
times=[]
for GC,timer in self.class_tests.items():
name = GC.ljust(column_len)
try:
# t=sum(timer.repeat(self.runs,self.reps))/self.runs
t=min(timer.repeat(self.runs,self.reps))
# print "%s: %s" % (name, timer.repeat(self.runs,self.reps))
times.append((t,name))
except Exception as e:
print("%s: Failed to benchmark (%s)." % (name,e))
times.sort()
tmin=times[0][0]
for t,name in times:
print("%s: %5.2f %s" % (name, t/tmin*100.,t))
print('-'*72)
print()
if __name__ == "__main__":
# set up for all routines:
classes=['Graph','MultiGraph','DiGraph','MultiDiGraph']
# classes=['Graph','MultiGraph','DiGraph','MultiDiGraph',
# 'SpecialGraph','SpecialDiGraph','SpecialMultiGraph','SpecialMultiDiGraph']
# classes=['Graph','SpecialGraph']
all_tests=['add_nodes','add_edges','remove_nodes','remove_edges',\
'neighbors','edges','degree','dijkstra','shortest path',\
'subgraph','edgedata_subgraph','laplacian']
# Choose which tests to run
tests=all_tests
# tests=['edges','laplacian']
#tests=all_tests[-1:]
N=100
if 'add_nodes' in tests:
title='Benchmark: Adding nodes'
test_string=('G.add_nodes_from(nlist)','nlist=range(%i)'%N)
b=Benchmark(classes,title,test_string,runs=3,reps=1000)
b.run()
if 'add_edges' in tests:
title='Benchmark: Adding edges'
setup='elist=[(i,i+3) for i in range(%s-3)]\nG.add_nodes_from(range(%i))'%(N,N)
test_string=('G.add_edges_from(elist)',setup)
b=Benchmark(classes,title,test_string,runs=3,reps=1000)
b.run()
if 'remove_nodes' in tests:
title='Benchmark: Adding and Deleting nodes'
setup='nlist=range(%i)'%N
test_string=('G.add_nodes_from(nlist)\nG.remove_nodes_from(nlist)',setup)
b=Benchmark(classes,title,test_string,runs=3,reps=1000)
b.run()
if 'remove_edges' in tests:
title='Benchmark: Adding and Deleting edges'
setup='elist=[(i,i+3) for i in range(%s-3)]'%N
test_string=('G.add_edges_from(elist)\nG.remove_edges_from(elist)',setup)
b=Benchmark(classes,title,test_string,runs=3,reps=1000)
b.run()
if 'neighbors' in tests:
N=500
p=0.3
title='Benchmark: reporting neighbors'
setup='H=NX.binomial_graph(%s,%s)\nfor (u,v) in H.edges_iter():\n G.add_edges_from([(u,v),(v,u)])'%(N,p)
test_string=('for n in G:\n for nbr in G.neighbors(n):\n pass',setup)
b=Benchmark(classes,title,test_string,runs=3,reps=10)
b.run()
if 'edges' in tests:
N=500
p=0.3
title='Benchmark: reporting edges'
setup='H=NX.binomial_graph(%s,%s)\nfor (u,v) in H.edges_iter():\n G.add_edges_from([(u,v),(v,u)])'%(N,p)
test_string=('for n in G:\n for e in G.edges(n):\n pass',setup)
b=Benchmark(classes,title,test_string,runs=3,reps=10)
b.run()
if 'degree' in tests:
N=500
p=0.3
title='Benchmark: reporting degree'
setup='H=NX.binomial_graph(%s,%s)\nfor (u,v) in H.edges_iter():\n G.add_edges_from([(u,v),(v,u)])'%(N,p)
test_string=('for d in G.degree():\n pass',setup)
b=Benchmark(classes,title,test_string,runs=3,reps=10)
b.run()
if 'dijkstra' in tests:
N=500
p=0.3
title='dijkstra single source shortest path'
setup='i=6\nH=NX.binomial_graph(%s,%s)\nfor (u,v) in H.edges_iter():\n G.add_edges_from([(u,v),(v,u)])'%(N,p)
test_string=('p=NX.single_source_dijkstra(G,i)',setup)
b=Benchmark(classes,title,test_string,runs=3,reps=10)
b.run()
if 'shortest path' in tests:
N=500
p=0.3
title='single source shortest path'
setup='i=6\nH=NX.binomial_graph(%s,%s)\nfor (u,v) in H.edges_iter():\n G.add_edges_from([(u,v),(v,u)])'%(N,p)
test_string=('p=NX.single_source_shortest_path(G,i)',setup)
b=Benchmark(classes,title,test_string,runs=3,reps=10)
b.run()
if 'subgraph' in tests:
N=500
p=0.3
title='subgraph method'
setup='nlist=range(100,150)\nH=NX.binomial_graph(%s,%s)\nfor (u,v) in H.edges_iter():\n G.add_edges_from([(u,v),(v,u)])'%(N,p)
test_string=('G.subgraph(nlist)',setup)
b=Benchmark(classes,title,test_string,runs=3,reps=10)
b.run()
if 'edgedata_subgraph' in tests:
N=500
p=0.3
title='subgraph method with edge data present'
setup='nlist=range(100,150)\nH=NX.binomial_graph(%s,%s)\nfor (u,v) in H.edges_iter():\n G.add_edges_from([(u,v,dict(hi=2)),(v,u,dict(hi=2))])'%(N,p)
test_string=('G.subgraph(nlist)',setup)
b=Benchmark(classes,title,test_string,runs=3,reps=10)
b.run()
if 'laplacian' in tests:
N=500
p=0.3
title='creation of laplacian matrix'
setup='H=NX.binomial_graph(%s,%s)\nfor (u,v) in H.edges_iter():\n G.add_edges_from([(u,v),(v,u)])'%(N,p)
test_string=('NX.laplacian_matrix(G)',setup)
b=Benchmark(classes,title,test_string,runs=3,reps=1)
b.run()
|
