-
Notifications
You must be signed in to change notification settings - Fork 0
/
XOR.py
91 lines (88 loc) · 2.87 KB
/
XOR.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
import math
#Sigmoid function
def sigmoid(x):
return 1/(1 + math.exp(-x));
#
in1 = 0;in2 = 0; Target = 0;
#
Wy = 1;Womg = 1;
Walpy = 1;Walpomg = 1;
Wbetay = 1;Wbetaomg = 1;
WAalp = 1;WAbeta=1;
#Feed Forward part
#in_y = in1*Wy; in_omg = in2*Womg;
O_y = sigmoid(in1*Wy); O_omg = sigmoid(in2*Womg);
#
#in_alpy = O_y*Walpy; in_betay = O_y*Wbetay;
#in_alpomg = O_omg*Walpomg; in_betaomg = O_omg*Wbetaomg ;
#
O_alp = sigmoid((O_y*Walpy+O_omg*Walpomg)); O_beta = sigmoid(O_y*Wbetay+O_omg*Wbetaomg);
#in_Aalp = O_alp*WAalp ;in_Abeta = O_beta * WAbeta;
#
O_A = sigmoid((O_alp*WAalp+O_beta * WAbeta));
#
#Error calculation and backpropagation part
E_A = math.pow((Target - O_A),2)/2;
#change in weight edge of A, alpha and A, Beta
dW_Aalp = (Target-O_A)*O_A*WAalp*O_alp;
dW_Abeta = (Target-O_A)*O_A*WAbeta*O_beta;
#Error calculation for alpha , beta , y ,omg
E_alp = E_A*WAalp;
E_beta = E_A*WAbeta;
E_y = O_y*(1-O_y)*((E_alp*Walpy)+(E_beta*Wbetay));
E_omg = O_omg*(1-O_omg)*((E_alp*Walpomg)+(E_beta*Wbetaomg));
#change in weight edge of alpha, y and Beta, Y and alpha, omg and Beta, omg
dW_alpy = E_alp*O_alp*(1-O_alp)*Walpy;
dW_alpomg = E_alp*O_alp*(1-O_alp)*Walpomg;
dW_betay = E_beta*O_beta*(1-O_beta)*Wbetay;
dW_betaomg = E_beta*O_beta*(1-O_beta)*Wbetaomg;
dW_y = E_y*Wy;
dW_omg = E_omg*Womg;
#
print E_A;
print "Welcome to XOR";
print "first layer neurons are y and Omg,\nsecond layer neurons are alpha and beta and A is output\nW denotes weights";
#
while E_A!=0.0000000000:
in1 = int(raw_input("Enter input 1"));
in2 = int(raw_input("Enter input 2"));
Target = int(raw_input("Enter Target"));
#
Wy = Wy + dW_y;
Womg = Womg +dW_omg;
Walpy = Walpy + dW_alpy;
Walpomg = Walpomg + dW_alpomg;
Wbetay = Wbetay + dW_betay;
Wbetaomg = Wbetaomg + dW_betaomg;
WAalp = WAalp + dW_Aalp;
WAbeta= WAbeta + dW_Abeta;
#
#Feed Forward part
#
O_y = sigmoid(in1*Wy); O_omg = sigmoid(in2*Womg);
#
O_alp = sigmoid((O_y*Walpy+O_omg*Walpomg)); O_beta = sigmoid(O_y*Wbetay+O_omg*Wbetaomg);
#
O_A = sigmoid((O_alp*WAalp+O_beta * WAbeta));
#
#Error calculation and backpropagation part
E_A = math.pow((Target - O_A),2)/2;
print E_A;
#change in weight edge of A, alpha and A, Beta
dW_Aalp = (Target-O_A)*O_A*WAalp*O_alp;
dW_Abeta = (Target-O_A)*O_A*WAbeta*O_beta;
#
#Error calculation for alpha , beta , y ,omg
E_alp = E_A*WAalp;
E_beta = E_A*WAbeta;
E_y = O_y*(1-O_y)*((E_alp*Walpy)+(E_beta*Wbetay));
E_omg = O_omg*(1-O_omg)*((E_alp*Walpomg)+(E_beta*Wbetaomg));
#
#change in weight edge of alpha, y and Beta, Y and alpha, omg and Beta, omg
dW_alpy = E_alp*O_alp*(1-O_alp)*Walpy;
dW_alpomg = E_alp*O_alp*(1-O_alp)*Walpomg;
dW_betay = E_beta*O_beta*(1-O_beta)*Wbetay;
dW_betaomg = E_beta*O_beta*(1-O_beta)*Wbetaomg;
dW_y = E_y*Wy;
dW_omg = E_omg*Womg;
print "Work Done";