forked from Lancephil/Sparrow-Search-Algorithm-Python
-
Notifications
You must be signed in to change notification settings - Fork 0
Expand file tree
/
Copy pathfunction_Ver2.py
More file actions
147 lines (117 loc) · 4.2 KB
/
Copy pathfunction_Ver2.py
File metadata and controls
147 lines (117 loc) · 4.2 KB
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
import numpy as np
import math
import copy
class function:
def __init__(self):
print("function init")
def Parameters(F):
if F=='F1':
ParaValue = [-10,10,1]
elif F=='F2':
ParaValue = [-10, 10, 30]
elif F=='F3':
ParaValue = [-100, 100, 30]
elif F=='F4':
ParaValue = [-100, 100, 30]
elif F=='F5':
ParaValue = [-30, 30, 30]
elif F=='F6':
ParaValue=[-100,100,30]
elif F=='F13':
ParaValue=[-5,5,2]
return ParaValue
def fun(F,X):
if F=='F1':
O=np.sum(X*X)
elif F=='F2':
O=np.sum(np.abs(X))+np.prod(np.abs(X))
elif F=='F3':
O=0
for i in range(len(X)):
O=O+np.square(np.sum(X[0:i+1]))
elif F=='F4':
O=np.max(np.abs(X))
elif F=='F5':
X_len = len(X)
O = np.sum(100 * np.square(X[1:X_len] - np.square(X[0:X_len - 1]))) + np.sum(np.square(X[0:X_len - 1] - 1))
elif F=='F6':
O=np.sum(np.square(np.abs(X+0.5)))
elif F=='F13':
O = 4*(X[0]**2) - 2.1*(X[0]**4) + (X[0]**6)/3 + X[0]*X[1] -4*(X[1]**2) + 4*(X[1]**4)
return O
def Bounds(s,Lb,Ub):
temp=s
for i in range(len(s)):
if temp[i]<Lb[0,i]:
temp[i]=Lb[0,i]
elif temp[i]>Ub[0,i]:
temp[i]=Ub[0,i]
return temp
def SSA(pop,M,c,d,dim,f):
#global fit
P_percent=0.2
pNum=round(pop*P_percent)
lb=c*np.ones((1,dim))
ub=d*np.ones((1,dim))
X=np.zeros((pop,dim))
fit=np.zeros((pop,1))
for i in range(pop):
X[i,:]=lb+(ub-lb)*np.random.rand(1,dim)
fit[i,0]=fun(f,X[i,:])
pFit=fit
pX=X
fMin=np.min(fit[:,0])
bestI=np.argmin(fit[:,0])
bestX=X[bestI,:]
Convergence_curve=np.zeros((1,M))
for t in range(M):
sortIndex=np.argsort(pFit.T)
fmax=np.max(pFit[:,0])
B=np.argmax(pFit[:,0])
worse=X[B,:]
r2=np.random.rand(1)
if r2 < 0.8:
for i in range(pNum):
r1=np.random.rand(1)
X[sortIndex[0,i],:]=pX[sortIndex[0,i],:]*np.exp(-(i)/(r1*M))
X[sortIndex[0,i],:]=Bounds(X[sortIndex[0,i],:],lb,ub)
fit[sortIndex[0,i],0]=fun(f,X[sortIndex[0,i],:])
elif r2 >= 0.8:
for i in range(pNum):
X[sortIndex[0,i],:]=pX[sortIndex[0,i],:]+np.random.rand(1)*np.ones((1,dim))
X[sortIndex[0,i],:]=Bounds(X[sortIndex[0,i],:],lb,ub)
fit[sortIndex[0,i],0]=fun(f,X[sortIndex[0,i],:])
bestII=np.argmin(fit[:,0])
bestXX=X[bestII,:]
for ii in range(pop-pNum):
i=ii+pNum
A=np.floor(np.random.rand(1,dim)*2)*2-1
if i> pop/2:
X[sortIndex[0,i],:]=np.random.rand(1)*np.exp(worse-pX[sortIndex[0,i],:]/np.square(i))
else:
X[sortIndex[0,i],:]=bestXX+np.dot(np.abs(pX[sortIndex[0,i],:]-bestXX),1/(A.T*np.dot(A,A.T)))*np.ones((1,dim))
X[sortIndex[0,i],:]=Bounds(X[sortIndex[0,i],:],lb,ub)
fit[sortIndex[0,i],0]=fun(f,X[sortIndex[0,i],:])
arrc = np.arange(len(sortIndex[0,:]))
#c=np.random.shuffle(arrc)
c=np.random.permutation(arrc)
b=sortIndex[0,c[0:20]]
for j in range(len(b)):
if pFit[sortIndex[0,b[j]],0]>fMin:
X[sortIndex[0,b[j]],:]=bestX+np.random.rand(1,dim)*np.abs(pX[sortIndex[0,b[j]],:]-bestX)
else:
X[sortIndex[0,b[j]],:]=pX[sortIndex[0,b[j]],:]+(2*np.random.rand(1)-1)*np.abs(pX[sortIndex[0,b[j]],:]-worse)/(pFit[sortIndex[0,b[j]]]-fmax+10**(-50))
X[sortIndex[0,b[j]],:]=Bounds(X[sortIndex[0,b[j]],:],lb,ub)
fit[sortIndex[0,b[j]],0]=fun(f,X[sortIndex[0,b[j]]])
for i in range(pop):
if fit[i,0]<pFit[i,0]:
pFit[i,0]=fit[i,0]
pX[i,:]=X[i,:]
if pFit[i,0]<fMin:
fMin=pFit[i,0]
bestX = copy.deepcopy(pX[i, :])
#bestX=pX[i,:]
Convergence_curve[0,t]=fMin
#print(fMin)
#print(bestX)
return fMin,bestX,Convergence_curve