-
Notifications
You must be signed in to change notification settings - Fork 0
/
Copy pathmain.py
182 lines (135 loc) · 4.48 KB
/
main.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
import os
import numpy as np
from tqdm import tqdm
import copy
import pickle
from time import time
from multiprocessing.dummy import Pool as ThreadPool
class genom():
Akc1=0
Akc2=0
Akc3=0
Akc4=0
Akc5=0
Kor1=0
Kor2=0
Kor3=0
Kor4=0
Kor5=0
fitnes=0
low=-0.15
high=0.15
low_k=-0.7
high_k=0.7
low_free=-10
high_free=25
def __init__(self):
self.Akc1=np.random.uniform(self.low, self.high)
self.Akc2=np.random.uniform(self.low, self.high)
self.Akc3=np.random.uniform(self.low, self.high)
self.Akc4=np.random.uniform(self.low, self.high)
self.Akc5=np.random.uniform(self.low, self.high)
self.Akc6=np.random.uniform(self.low_free, self.high_free)
self.Kor1=np.random.uniform(self.low_k, self.high_k)
self.Kor2=np.random.uniform(self.low_k, self.high_k)
self.Kor3=np.random.uniform(self.low_k, self.high_k)
self.Kor4=np.random.uniform(self.low_k, self.high_k)
self.Kor5=np.random.uniform(self.low_k, self.high_k)
def crossover(self,p2):
'''
self.Akc1=(self.Akc1+p2.Akc1)/2
self.Akc2=(self.Akc2+p2.Akc2)/2
self.Akc3=(self.Akc3+p2.Akc3)/2
self.Akc4=(self.Akc4+p2.Akc4)/2
self.Akc5=(self.Akc5+p2.Akc5)/2
self.Akc6=(self.Akc6+p2.Akc6)/2
self.Kor1=(self.Kor1+p2.Kor1)/2
self.Kor2=(self.Kor2+p2.Kor2)/2
self.Kor3=(self.Kor3+p2.Kor3)/2
self.Kor4=(self.Kor4+p2.Kor4)/2
self.Kor5=(self.Kor5+p2.Kor5)/2
'''
self.Akc3=p2.Akc3
self.Akc4=p2.Akc4
self.Akc5=(self.Akc5+p2.Akc5)/2
self.Akc6=(self.Akc6+p2.Akc6)/2
self.Kor3=p2.Kor3
self.Kor4=p2.Kor4
self.Kor5=(self.Kor5+p2.Kor5)/2
def mutate(self, p=0.01):
if np.random.uniform()<p:
self.Akc1=np.random.uniform(self.low, self.high)
if np.random.uniform()<p:
self.Akc2=np.random.uniform(self.low, self.high)
if np.random.uniform()<p:
self.Akc3=np.random.uniform(self.low, self.high)
if np.random.uniform()<p:
self.Akc4=np.random.uniform(self.low, self.high)
if np.random.uniform()<p:
self.Akc5=np.random.uniform(self.low, self.high)
if np.random.uniform()<p:
self.Akc6=np.random.uniform(self.low_free, self.high_free)
if np.random.uniform()<p:
self.Kor1=np.random.uniform(self.low_k, self.high_k)
if np.random.uniform()<p:
self.Kor2=np.random.uniform(self.low_k, self.high_k)
if np.random.uniform()<p:
self.Kor3=np.random.uniform(self.low_k, self.high_k)
if np.random.uniform()<p:
self.Kor4=np.random.uniform(self.low_k, self.high_k)
if np.random.uniform()<p:
self.Kor5=np.random.uniform(self.low_k, self.high_k)
def fit_help(self,percent, t):
temp = percent**4*(1 + 100 * (percent/100)**4/t**2)
self.fitnes=temp
def cal_fitnes(self):
GUI=np.random.choice(np.arange(3))
send=str(GUI) + " " + str(self.Akc1) + " " + str(self.Akc2) + " " + str(self.Akc3) + " " + str(self.Akc4) + " " + str(self.Akc5) + " " + str(self.Akc6) \
+ " " + str(self.Kor1) + " " + str(self.Kor2) + " " + str(self.Kor3) + " " + str(self.Kor4) + " " + str(self.Kor5)
start = time()
percent = os.system("java -jar Simulator.jar " + send)
t = time()-start
self.fit_help(percent, t)
return self.fitnes
def get_best_parents(fitnes):
fitnes = np.array(fitnes)
fit_norma=fitnes/sum(fitnes)
idx1=np.random.choice(np.arange(len(fit_norma)),p=fit_norma)
idx2=np.random.choice(np.arange(len(fit_norma)),p=fit_norma)
return idx1, idx2
def get_worst_parents(fitnes):
temp=1/np.array(fitnes)
p=temp/sum(temp)
return np.random.choice(np.arange(len(p)),p=p)
def mutil_fitnes(genom):
return genom.cal_fitnes()
if __name__ == '__main__':
pool = ThreadPool(4)
n_population=20
n_generation=20000
population=[]
for i in range(n_population):
population.append(genom())
fitnes = pool.map(mutil_fitnes, population)
'''
fitnes=[0]*n_population
for j in tqdm(range(n_population)):
fitnes[j]=population[j].cal_fitnes()
'''
for i in tqdm(range(n_generation)):
childs = []
bad = []
for i in range(4):
idx1, idx2 =get_best_parents(fitnes)
child =copy.deepcopy(population[idx1])
child.crossover(population[idx2])
child.mutate()
childs.append(child)
idx= get_worst_parents(fitnes)
bad.append(idx)
fitnes_childs=pool.map(mutil_fitnes, childs)
for k, child_f,child in zip(bad,fitnes_childs,childs):
fitnes[k]=child_f
population[k]=copy.deepcopy(child)
if i%100==0:
pickle.dump(population, open( "populacija.p", "wb" ) )