-
Notifications
You must be signed in to change notification settings - Fork 0
/
learnside.js
182 lines (136 loc) · 4.6 KB
/
learnside.js
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
var inputs , bias , weightsOfInputHidden , result , i ,j ,
inputs = [1,0,1]
var OutPutFinalResult = []
bias = [-0.4,0.2,0.1]
weightsOfInputHidden = [[0.2, -0.3],[0.4,0.1],[-0.5,0.2]]
var weightsOfHiddenToOuput = [-0.3, -0.2]
var resultForHidddentoOutput = []
var errorArray = []
// errorArray[0]=1
var target = 1
var learning = 0.03
var counter = 1
var OutputError = 2
// input to hidden layer
while( OutputError>0.05){
// for(j=0;j<=1;j++){
var oneObjectResult = []
// var oneObjectResult = []
for(i=0;i<=2;i++){
result= inputs[i]*weightsOfInputHidden[i][0]
//console.log('result',result)
// console.log(inputs[2])
oneObjectResult.push(result)
if(i==2){
var hiddenOutPut = oneObjectResult.reduce(function(a, b) { return a + b; }, 0)
hiddenO
utPut = hiddenOutPut + bias[0]
//console.log('hiddenOutPut',hiddenOutPut)
var sigmoidOutput =1/(1+ Math.exp(-hiddenOutPut))
// console.log('sigmoidOutput',sigmoidOutput)
// console.log('OutPutFinalResult',OutPutFinalResult.length)
OutPutFinalResult[0] = sigmoidOutput
}
}
var oneObjectResult = []
// var oneObjectResult = []
for(i=0;i<=2;i++){
result= inputs[i]*weightsOfInputHidden[i][1]
//console.log('result',result)
// console.log(inputs[2])
oneObjectResult.push(result)
if(i==2){
var hiddenOutPut = oneObjectResult.reduce(function(a, b) { return a + b; }, 0)
hiddenOutPut = hiddenOutPut + bias[1]
//console.log('hiddenOutPut',hiddenOutPut)
var sigmoidOutput =1/(1+ Math.exp(-hiddenOutPut))
// console.log('sigmoidOutput',sigmoidOutput)
// console.log('OutPutFinalResult',OutPutFinalResult.length)
OutPutFinalResult[1] = sigmoidOutput
console.log()
}
}
// }
// hidden to output
//console.log('hidden to output')
var resultForHidddentoOutput = []
for(i=0;i<=1;i++){
result= OutPutFinalResult[i]*weightsOfHiddenToOuput[i]
resultForHidddentoOutput.push(result)
if(i==1){
// console.log('check')
//console.log('ARRAY ',resultForHidddentoOutput)
var hiddenOutPut = resultForHidddentoOutput.reduce(function(a, b) { return a + b; }, 0)
hiddenOutPut = hiddenOutPut + bias[2]
//console.log('hiddenOutPut',hiddenOutPut)
var sigmoidOutput =1/(1+ Math.exp(-hiddenOutPut))
//console.log('sigmoidOutput',sigmoidOutput)
OutPutFinalResult[2] = sigmoidOutput
// console.log(OutPutFinalResult[2])
}
}
// error result
//console.log('error result ')
// erorr for output layer
var Oj = OutPutFinalResult[2]
var errorOfOutput = Oj*(1-Oj)*(target-Oj)
errorArray[0] = {type:'Output', value:errorOfOutput}
OutputError = errorArray[0].value
// error of hidden layer
// for(i=0;i<=1;i++){
var Oj= OutPutFinalResult[0]
//console.log(Oj)
// var errorOfOutput = Oj*(1-Oj)*(errorArray[0]*weightsOfHiddenToOuput[i])
var errorOfOutput = Oj*(1-Oj)*(errorArray[0].value) * weightsOfHiddenToOuput[0]
//console.log(errorOfOutput)
errorArray[1] = {type:'Output', value:errorOfOutput}
// }
// for(i=0;i<=1;i++){
var Oj= OutPutFinalResult[1]
//console.log(Oj)
// var errorOfOutput = Oj*(1-Oj)*(errorArray[0]*weightsOfHiddenToOuput[i])
var errorOfOutput = Oj*(1-Oj)*(errorArray[0].value) * weightsOfHiddenToOuput[1]
//console.log(errorOfOutput)
errorArray[2] = {type:'Output', value:errorOfOutput}
// }
//
// //console.log(errorArray)
// update weights
// hiiden layer weights
// weightsOfHiddenToOuput
for(i=0;i<=1;i++){
var daltaWeights = learning * errorArray[0].value * OutPutFinalResult[i]
weightsOfHiddenToOuput[i] = weightsOfHiddenToOuput[i]+daltaWeights
}
//console.log(OutPutFinalResult)
//console.log(weightsOfHiddenToOuput)
// input layer weights
var errorIndex = 1
for(j=0;j<=1;j++){
var oneObjectResult = []
for(i=0;i<=2;i++){
var daltaWeights = learning * errorArray[errorIndex].value * inputs[i]
//console.log('daltaWeights',daltaWeights)
weightsOfInputHidden[i][j] = weightsOfInputHidden[i][j]+daltaWeights
}
errorIndex++
}
//console.log(weightsOfInputHidden)
// calculate bias
var BIASj = learning*errorArray[0].value
bias[2]= bias[2]+BIASj
//console.log(bias[2])
var erorIndexForBias = 1
for(i=0;i<=1;i++){
var BIASj = learning*errorArray[erorIndexForBias].value
bias[i]= bias[i]+BIASj
erorIndexForBias++
}
// console.log(bias[0])
// console.log( 'counter ' + counter + 'error ' +errorArray[0].value)
// console.log(OutPutFinalResult[2])
console.log(OutputError)
// console.log(bias[0])
// console.log(weightsOfHiddenToOuput[1])
counter++
}