-
Notifications
You must be signed in to change notification settings - Fork 0
/
test_ser_tree_bicriteria.py
214 lines (198 loc) · 10.3 KB
/
test_ser_tree_bicriteria.py
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
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
# Copyright 2023 Janos Czentye
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at:
#
# http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
import math
import pathlib
import time
import networkx as nx
import pandas as pd
from slambuc.alg.app import NAME
from slambuc.alg.tree.serial.bicriteria import biheuristic_btree_partitioning, bifptas_ltree_partitioning, \
bifptas_dual_ltree_partitioning
from slambuc.alg.tree.serial.pseudo import pseudo_btree_partitioning, pseudo_ltree_partitioning
from slambuc.alg.util import recalculate_partitioning
from slambuc.misc.random import get_random_tree
from slambuc.misc.util import evaluate_ser_tree_partitioning
def run_validation_test(tree: nx.DiGraph, M: int, L: int, root: int = 1, cp_end: int = None, delay: int = 10):
print(" Run pseudo BTree algorithm ".center(80, '#'))
b_part, b_opt_cost, b_opt_lat = pseudo_btree_partitioning(tree, root, M, L, cp_end, delay)
evaluate_ser_tree_partitioning(tree, b_part, b_opt_cost, b_opt_lat, root, cp_end, M, L, delay)
print(" Run bi-PTAS BTree algorithm ".center(80, '#'))
bc_part, bc_opt_cost, bc_opt_lat = biheuristic_btree_partitioning(tree, root, M, L, cp_end, delay)
evaluate_ser_tree_partitioning(tree, bc_part, bc_opt_cost, bc_opt_lat, root, cp_end, M, L, delay)
print(" Run pseudo LTree algorithm ".center(80, '#'))
l_part, l_opt_cost, l_opt_lat = pseudo_ltree_partitioning(tree, root, M, L, cp_end, delay)
evaluate_ser_tree_partitioning(tree, l_part, l_opt_cost, l_opt_lat, root, cp_end, M, L, delay)
print(" Run bi-PTAS LTree algorithm ".center(80, '#'))
lc_part, lc_opt_cost, lc_opt_lat = bifptas_ltree_partitioning(tree, root, M, L, cp_end, delay)
evaluate_ser_tree_partitioning(tree, lc_part, lc_opt_cost, lc_opt_lat, root, cp_end, M, L, delay)
def test_ser_bicriteria_tree_partitioning():
tree = nx.read_gml(pathlib.Path(__file__).parent / "data/graph_test_tree_ser.gml", destringizer=int)
tree.graph[NAME] += "-ser_bic"
params = dict(tree=tree,
root=1,
cp_end=10,
M=6,
# L=math.inf,
L=430,
delay=10)
run_validation_test(**params)
def test_random_bicriteria_tree_partitioning(n: int = 10):
tree = get_random_tree(n)
tree.graph[NAME] += "-ser_bic"
params = dict(tree=tree,
root=1,
cp_end=n,
M=6,
L=math.inf,
# L = 430,
delay=10)
run_validation_test(**params)
def get_accuracy_stats(n: int = 10, M: int = 6, L: int = math.inf, Epsilon: float = 0.0, Lambda: float = 0.0,
stop_failed: bool = False):
tree = get_random_tree(n)
tree.graph[NAME] += "-ser_bic"
params = dict(tree=tree,
root=1,
cp_end=n,
M=M,
# L=math.inf,
L=L,
delay=10,
bidirectional=False)
bic_params = params.copy()
bic_params.update(Epsilon=Epsilon, Lambda=Lambda)
#
# Btree exact
_s = time.perf_counter()
btree_part, btree_opt_cost, btree_opt_lat = pseudo_btree_partitioning(**params)
btree_time = time.perf_counter() - _s
_s = time.perf_counter()
# Bi-Btree reference
b_part, b_opt_cost, b_opt_lat = biheuristic_btree_partitioning(**params)
bc_exact_time = time.perf_counter() - _s
_s = time.perf_counter()
# Bi-Btree
bc_part, bc_opt_cost, bc_opt_lat = biheuristic_btree_partitioning(**bic_params)
bc_bic_time = time.perf_counter() - _s
#
# Ltree exact
_s = time.perf_counter()
ltree_part, ltree_opt_cost, ltree_opt_lat = pseudo_ltree_partitioning(**params)
ltree_time = time.perf_counter() - _s
# Bi-Ltree reference
_s = time.perf_counter()
l_part, l_opt_cost, l_opt_lat = bifptas_ltree_partitioning(**params)
lc_exact_time = time.perf_counter() - _s
_s = time.perf_counter()
# Bi-Ltree
lc_part, lc_opt_cost, lc_opt_lat = bifptas_ltree_partitioning(**bic_params)
lc_bic_time = time.perf_counter() - _s
#
# Bi-Ltree dual reference
_s = time.perf_counter()
dl_part, dl_opt_cost, dl_opt_lat = bifptas_dual_ltree_partitioning(**params)
dlc_exact_time = time.perf_counter() - _s
_s = time.perf_counter()
# Bi-Ltree dual
dlc_part, dlc_opt_cost, dlc_opt_lat = bifptas_dual_ltree_partitioning(**bic_params)
dlc_bic_time = time.perf_counter() - _s
#
print(" Results ".center(80, '#'))
print(f"Btree: {b_part}")
print(f"Bi-Btree: {bc_part}")
print(f"Ltree: {l_part}")
print(f"Bi-Ltree: {lc_part}")
print(f"DLtree: {l_part}")
print(f"Bi-DLtree: {lc_part}")
if btree_part:
print("#### Btree alg execution:")
print(f"Btree pseudo exact cost: {btree_opt_cost:>5}, lat: {btree_opt_lat:>5} "
f"with time: {btree_time * 1000:>3.4f} ms")
print(f"Bi-Btree ref. exact cost: {b_opt_cost:>5}, lat: {b_opt_lat:>5} "
f"with time: {bc_exact_time * 1000:>3.4f} ms")
if bc_part:
print(f"Bi-Btree bicriteria cost: {bc_opt_cost:>5}, lat: {bc_opt_lat:>5} "
f"with time: {bc_bic_time * 1000:>3.4f} ms")
bc_sum_cost, bc_sum_lat = recalculate_partitioning(tree, bc_part, root=params['root'], cp_end=params['cp_end'],
delay=params['delay'])
b_cost_err = (bc_sum_cost - btree_opt_cost) / btree_opt_cost
b_lat_err = (bc_sum_lat - L) / L
print(f">>>> Errors: {b_cost_err = :.4f} (<= {Epsilon=}) [{bc_sum_cost}/{btree_opt_cost=}], "
f" {b_lat_err = :.4f} (<= {Lambda=}) [{bc_sum_lat}/{L=}]")
else:
b_cost_err, b_lat_err = 0, 0
if ltree_part:
print("#### Ltree alg execution:")
print(f"Ltree pseudo exact cost: {ltree_opt_cost:>5}, lat: {ltree_opt_lat:>5}, "
f"with time: {ltree_time * 1000:>3.4f} ms")
if lc_part:
print(f"Bi-Ltree ref. exact cost: {l_opt_cost:>5}, lat: {lc_opt_lat:>5}, "
f"with time: {lc_exact_time * 1000:>3.4f} ms")
print(f"Bi-Ltree bicriteria cost: {lc_opt_cost:>5}, lat: {lc_opt_lat:>5}, "
f"with time: {lc_bic_time * 1000:>3.4f} ms")
lc_sum_cost, lc_sum_lat = recalculate_partitioning(tree, lc_part, root=params['root'], cp_end=params['cp_end'],
delay=params['delay'])
l_cost_err = (lc_sum_cost - ltree_opt_cost) / ltree_opt_cost
l_lat_err = (lc_sum_lat - L) / L
print(f">>>> Errors: {l_cost_err = :.4f} (<= {Epsilon=}) [{lc_sum_cost}/{ltree_opt_cost=}], "
f" {l_lat_err = :.4f} (<= {Lambda=}) [{lc_sum_lat}/{L=}]")
else:
l_cost_err, l_lat_err = 0, 0
if dlc_part:
print(f"Bi-Ltree dual ref. cost: {dl_opt_cost:>5}, lat: {dlc_opt_lat:>5}, "
f"with time: {dlc_exact_time * 1000:>3.4f} ms")
print(f"Bi-Ltree dual bic. cost: {dlc_opt_cost:>5}, lat: {dlc_opt_lat:>5}, "
f"with time: {dlc_bic_time * 1000:>3.4f} ms")
dlc_sum_cost, dlc_sum_lat = recalculate_partitioning(tree, dlc_part, root=params['root'],
cp_end=params['cp_end'], delay=params['delay'])
dl_cost_err = (dlc_sum_cost - ltree_opt_cost) / ltree_opt_cost
dl_lat_err = (dlc_sum_lat - L) / L
print(f">>>> Errors: {dl_cost_err = :.4f} (<= {Epsilon=}) [{dlc_sum_cost}/{ltree_opt_cost=}], "
f" {dl_lat_err = :.4f} (<= {Lambda=}) [{dlc_sum_lat}/{L=}]")
else:
dl_cost_err, dl_lat_err = 0, 0
if stop_failed:
assert all((b_cost_err <= Epsilon if math.isfinite(b_cost_err) else True,
l_cost_err <= Epsilon if math.isfinite(l_cost_err) else True,
dl_cost_err <= Epsilon if math.isfinite(dl_cost_err) else True,
b_lat_err <= Lambda if math.isfinite(b_lat_err) else True,
dl_lat_err <= Lambda if math.isfinite(dl_lat_err) else True,
l_lat_err <= Lambda if math.isfinite(l_lat_err) else True))
return ((btree_time * 1000, bc_exact_time * 1000, bc_bic_time * 1000, b_cost_err, b_lat_err), # Btree stats
(ltree_time * 1000, lc_exact_time * 1000, lc_bic_time * 1000, l_cost_err, l_lat_err),
(ltree_time * 1000, dlc_exact_time * 1000, dlc_bic_time * 1000, dl_cost_err, dl_lat_err)) # Ltree stats
def test_accuracy():
print(get_accuracy_stats())
def stress_test(iteration: int = 100, n: int = 10, M: int = 6, L: int = math.inf, Epsilon: float = 0.5,
Lambda: float = 0.5, stop_failed: bool = False):
b_stat, l_stat, dl_stat = zip(*[get_accuracy_stats(n, M, L, Epsilon, Lambda, stop_failed=stop_failed)
for _ in range(iteration)])
b_df = pd.DataFrame(b_stat, columns=['Btree', 'BiBtree_exact_time', 'BiBtree_time', 'B_omega', 'B_lambda'])
l_df = pd.DataFrame(l_stat, columns=['Ltree', 'BiLtree_exact_time', 'BiLtree_time', 'L_omega', 'L_lambda'])
dl_df = pd.DataFrame(dl_stat, columns=['Ltree', 'BiDLtree_exact_time', 'BiDLtree_time',
'Dual_L_omega', 'Dual_L_lambda'])
pd.set_option('display.expand_frame_repr', False)
print(" Btree stat ".center(80, '#'))
print(b_df.describe().transpose())
print(" Ltree stat ".center(80, '#'))
print(l_df.describe().transpose())
print(" Dual Ltree stat ".center(80, '#'))
print(dl_df.describe().transpose())
if __name__ == '__main__':
# test_ser_bicriteria_tree_partitioning()
# test_random_bicriteria_tree_partitioning()
# test_accuracy(n=10, M=6, L=450, Epsilon=0.5, Lambda=0.2, stop_failed=True)
# test_accuracy(n=10, M=6, L=math.inf, Epsilon=0.5, Lambda=0.2, stop_failed=True)
stress_test(iteration=100, n=20, M=20, L=855, Epsilon=0, Lambda=0, stop_failed=True)
# stress_test(iteration=100, n=40, M=30, L=855, Epsilon=0.5, Lambda=0.2, stop_failed=True)